34
-omparative Saving Behavior -f Rural and Urban Households n the Philippines meo M. Bautista -3rioB. Lamberte printed from irnal of Philippine Development 17, No. 2, 1990 S -- . S, 0 "un * E R rint No. 233

-omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Embed Size (px)

Citation preview

Page 1: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

-omparative Saving Behavior -fRural and Urban Households n the Philippines

meo M Bautista -3rioB Lamberte

printed from irnal of Philippine Development

17 No 2 1990

S -- S 0un

E R

rint No 233

11 6

PJournal ofPhilippineDevelopment Number Thirty One Volume XV II No 2 1990

COMPARATIVE SAVING BEHAVIOR OF RURAL AND URBAN HOUSEHOLDS IN THE PHILIPPINES

Romeo M Bautista and Mario B Lamberte

Marginal saving rates for Philippine households are estimashyted distinguishing between rural and urban households byregion and by income group At a given income level rural households generally save more than urban households both on average and at the margin This contrasts with the higher average saving rates for urban households in the various regions attributable to their higher incomes The estimated marginal saving rates for rural households in many regions are found to be higher than those for their urban counterparts In the context of agriculture-baseddevelopment faster growth of rural incomes need not result in lower aggregate savings

I INTRODUCTION

The aggregate saving rate the fraction of national income that is not spent on current consumption has long been widely regarded as a key factor in economic growth1 In the dynamic Harrod-Domar model the saving rate and the incremental capital-output ratio jointshyly determine the growth rate of the economy The critical role of saving in capital accumulation and economic development is also recognized in the two-gap and classical growth models Even in the

Paper presented during the ACPC-PIDS sponsored seminar-workshop on Rural Savings and Informal Credit in the Context of Financial Liberalization held on June 7-8 1990 at the G Licaros Auditorium 18th Floor Multi-Storey Building Central Bank of the Philippines A Mabini Street Manila

The authors are Research Fellow at the International Food Policy Researrch Instishytute (IFPRI) in Washington DC and Vice-President of the Philippine Institute for Development Studies (PIDS) respectively The authors benefited from comments on an earlier draft of this paper by Dale Adams Michael Lipton and John Mellor as well as from participants in seminars at IFPRI PIDS and Ohio State University (OSU) The funding support from the United States Agency for International Development (USAID) throughthe NEDA-Training and Development Issues (TDI) Project and the research assistance of Maricn Alicando are also gratefully acknowledged

1 Past surveys discussing the role of saving in economic development and its detershyminants include Snyder (1974) and Gersovitz (1988)

150 JOURNAL OF PHILIPPINE DEVELOPMENT

neoclassical growth model in which savings do not influence economshyic growth in the steady state a higher saving rate isassociated with more rapid growth of the economy in its movement towards long-runequiitbrium As noted by Gersovitz (1988) this transitional path(which can take time) is more meaningful than alternative steadystates in representing the evolution of developing economies

There are of course other determinants of economic growthTechnological progress institutional development domestic policiesand the external economic environment have also been emphasizedin the development literature To these additional influences on ecoshynomic growth can be attributed the lack of a simple correlationsometimes observed between savings and growth in developing cotnshytries (LDCs) Only when these explanatory variables are simultashyneously taken into account would one be able to assess empiricallytheir separate effects on growth Although Arthur Lewiss famousdictum that raising the saving rate is the central problem in economshyic development can be disputed (Deaton 1989 39) few woulddoubt that economic growth cannot be long sustained under condishytions of declining saving rates

Savings including both domestic and foreign finance the(physical and human) capital formatin needed to increase outputand this isof particular importance to typically capital-scarce LDCsApart from it direct contribution to output growth capital accumushylation also makes pcssble the employment of complementary proshyduction inputs in abundant supply - for example unskilled labor inmost developing countries shy and serves as a vehicle for the adoptionof improved technologies embodied in new investments

While there have been brief periods of significant inflow ofexternal financial resources to some LDCs in the past foreign savingscannot be expected to provide a sustainable basis for financingdomestic investment Raising the national saving rate is particularlyessential to developing countries with a heavy debt-service burdenand limited capacity to obtain loans in foreign capital markts Indeed macroeconomic adjustment programs oriented to theresumption of long-run growth invariably emphasize the need to expand domestic savings

Household saving is usually the largest component of domesticsavings in developing countries especially the lower-income predoshyminantly agricultural LDCs This contrasts with the much greaterimportance of corporate saving in developed countries The abilitywillingness and opportunity of households to save over time cantherefore significantly influence the rate and sustainability of capitalaccumulation and economic growth in developing countries

151 BAUTISTA amp LAMBERTE SAVING BEHAVIOn

There has recently been a growing recognition of the pivotalrole of agriculture and the rural sector in promoting a more rapid and equitable economic growth for low-income LDCs 2 Advocates of an agriculture-based development strategy (ADS) emphasize the expanshysion of public investments in the rural sector and the removal ofpolicy-induced price biases against agricultural products These are expected to raise farm productivity agricultural output and ruralincome which in turn will generate increased intermediate demand for nonagricultural inputs and more importantly will stimulate conshysumption demand for food and labor-intensive industrial goods and services generating employment and income multiplier effects on therural regional and national economies In the short to medium termrural household income is expected to increase at a faster rate comshypared to urban household income

How will the adoption of ADS and tne associated shift in ruralshyurban income distribution affect aggregate household savings In any given year it is commonly observed that the average saving rateof rural households is lower sometimes very much lower than that of urban households 3 If fixed saving rates for the rura and urban sectors are assumed the rising real income of rural households relashytive to urban households in a growth process that is agriculturallydriven will likely result in a lower aggregate saving rate calling intoquestion its sustainability Such an assumption on household saving behavior however is questionable

A positive relationship between the saving rate and incomein developing countries at least within certain ranges of income levels has been obtained in past empirical studies using household survey data (eg Bhalla [1980] for India) or cross-country nationalincome accounts (eg Moore 1981) for Asian countries An observed lower average saving rate for rural households (relative to urban households) may then be explained simply by their lower averageincome However rural household incomes can increase rapidly inthe course of agriculture-based development which may prevent a decline in the aggregate saving rate or even raise it Furthermore the improvement of investment opportunities in the rural areas associashy

2 See for example Adelman (1984) and Bautista (1988) An early statement ofan agriculture-based development strategy can be found in Mellor (1976)3 Lipton (1977 247) has aptly pointed out that apart from the significant undershyestimation of agricultural saving embodied directly in investment some rural saving isdrained off by price twists to finance socially low-yielding urban investment and that

farmers would have more incentive to save and to embody their savings in farm investmentif its returns were not artificially depressed by policies turning the terms of trade againstagriculture

152 JOURNAL OF PHILIPPINE DEVELOPMENT

ted with the ADS provides an additional stimulus to increased savingsby rural households

Empirical evidence on the relative size of the marginal savingrates for rural and urban houeholds is thin and contradictory In a study using 1963-72 survey data on Souith Korean householdsfarmers were found to have been considerably more thrifty with their marginal propensity to consume being almost half that of urban consumers (Lluch et al 1977 99) Based on a similar analyticalframework estimates of the marginal propensity to save (MPS) for Mexico show averages of 011 for rural households compard to 025 for urban households in Chile the values tend to be around 030with slightly lower levels for comparable urban households (p 241)Using data from household surveys in Bangladesh for each year from 197677 to 1978-79 the MPS out of transitory income isestimated to be consistently and significantly higher among rural households (Chowdhury 1987) On the other hand the findings from an Indian study using time series data indicate a higher aggregate MPS for the urban sector (Gupta 1970) To be sure comparability of saving rate estimates from independent studies for different countries by diffeshyrent investigators is severely impaired by differences in data and measures used and by the varying analytical approaches and estimashytion techniques adopted in deriving the estirrates4

In this paper we investigate the comparative saving behavior of rural and urban households in the Philippines using FIES (FamilyIncome and Expenditure Survey) data for 1985 the latest year for which such data are available Apart from the contribution to empishyrical knowledge in an area previously not given systematic attention in the development literature some policy interest attaches in the Philippine context to an examination of the differential saving behashyvior of rural and urban households considering the recent shift towards more rural-oriented development policies as expressed in the governments Medium-Term Philippine Development Plan 1987shy1992

Section II addresses some theoretical and empirical issues in savings behavior especially as they relate to existing conditions in developing countries and in particular the Philippines The discusshysion leads to the specification of household saving functions distinshyguished by region by location (rural and urban) and by income class In Section III we describe and evaluate the data base for the

4 For instance the use of an extended linear expenditure system and of sophisshyticated econometric estimation procedures in the studies on Korea Mexico and Chile cited above contrasts sharply with the simple linear regression of income on saving employed in the studies on Bangladesh and India

153 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

study indicating also some relevant characteristics of sample houseshyholds Section IV presents and analyzes the estimated saving equashytions The findings of the study are summarized and their implicashytions for development policy and stratey are briefly discussed in the concluding section

I1ANALYTICAL CONSIDERATIONS

In a fundamental sense ones saving isa sacrifice of current conshysumption that will allow for an increase in future consumptionSaving is a means not only of reducing fluctuations in income and smoothing consumption over time but also of earning interest Given the intertemporal nature of the saving process it is the lifetime income of the individual and not just the currect income as impliedin the Keynesian consumption function that should influence current saving This life-cycle model also suggests a higher rate of saving during certain periods eg in preretirement years in order to provide for consumption in old age The age of the individual would therefore be an appropriate determinant of savings Additionally existing investment opportunities (or more precisely the return to investments) and the cost of borrowing can also significantly affect saving behavior

That current saving depends on lifetime resources is also an important implication of Friedmans (1957) permanent income hypothesis Current income is viewed as the sum of permanent and transitory incomes the former reflecting the individuals lifetime earnings In its extreme form the permanent income approach postushylates equality between an individuals current consumption and pershymanent income implying that the MPS is zero out of permanentincome and one out of transitory income Existing empirical studies do not oear out this strict version of the permanent income hypotheshysis but they provide support to the view that the marginal saving rate is higher out of transitory income than out of permanent income

Quantitative information on actual choices about savings is typically available at the household rather than individual level at least among developing countries Such data are often providedthrough countrywide income and expenditure surveys and less freshyquently through special locationally more focused surveys on changes in family assets and liabilities Both sources of data are subjet to error Household income particularly the nonmonetized component is often observed to be underreported with significantdiferences in the degree of underestimation amng surveys done in the same country for different years (Berry 1985) If nonmonetized

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 2: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

11 6

PJournal ofPhilippineDevelopment Number Thirty One Volume XV II No 2 1990

COMPARATIVE SAVING BEHAVIOR OF RURAL AND URBAN HOUSEHOLDS IN THE PHILIPPINES

Romeo M Bautista and Mario B Lamberte

Marginal saving rates for Philippine households are estimashyted distinguishing between rural and urban households byregion and by income group At a given income level rural households generally save more than urban households both on average and at the margin This contrasts with the higher average saving rates for urban households in the various regions attributable to their higher incomes The estimated marginal saving rates for rural households in many regions are found to be higher than those for their urban counterparts In the context of agriculture-baseddevelopment faster growth of rural incomes need not result in lower aggregate savings

I INTRODUCTION

The aggregate saving rate the fraction of national income that is not spent on current consumption has long been widely regarded as a key factor in economic growth1 In the dynamic Harrod-Domar model the saving rate and the incremental capital-output ratio jointshyly determine the growth rate of the economy The critical role of saving in capital accumulation and economic development is also recognized in the two-gap and classical growth models Even in the

Paper presented during the ACPC-PIDS sponsored seminar-workshop on Rural Savings and Informal Credit in the Context of Financial Liberalization held on June 7-8 1990 at the G Licaros Auditorium 18th Floor Multi-Storey Building Central Bank of the Philippines A Mabini Street Manila

The authors are Research Fellow at the International Food Policy Researrch Instishytute (IFPRI) in Washington DC and Vice-President of the Philippine Institute for Development Studies (PIDS) respectively The authors benefited from comments on an earlier draft of this paper by Dale Adams Michael Lipton and John Mellor as well as from participants in seminars at IFPRI PIDS and Ohio State University (OSU) The funding support from the United States Agency for International Development (USAID) throughthe NEDA-Training and Development Issues (TDI) Project and the research assistance of Maricn Alicando are also gratefully acknowledged

1 Past surveys discussing the role of saving in economic development and its detershyminants include Snyder (1974) and Gersovitz (1988)

150 JOURNAL OF PHILIPPINE DEVELOPMENT

neoclassical growth model in which savings do not influence economshyic growth in the steady state a higher saving rate isassociated with more rapid growth of the economy in its movement towards long-runequiitbrium As noted by Gersovitz (1988) this transitional path(which can take time) is more meaningful than alternative steadystates in representing the evolution of developing economies

There are of course other determinants of economic growthTechnological progress institutional development domestic policiesand the external economic environment have also been emphasizedin the development literature To these additional influences on ecoshynomic growth can be attributed the lack of a simple correlationsometimes observed between savings and growth in developing cotnshytries (LDCs) Only when these explanatory variables are simultashyneously taken into account would one be able to assess empiricallytheir separate effects on growth Although Arthur Lewiss famousdictum that raising the saving rate is the central problem in economshyic development can be disputed (Deaton 1989 39) few woulddoubt that economic growth cannot be long sustained under condishytions of declining saving rates

Savings including both domestic and foreign finance the(physical and human) capital formatin needed to increase outputand this isof particular importance to typically capital-scarce LDCsApart from it direct contribution to output growth capital accumushylation also makes pcssble the employment of complementary proshyduction inputs in abundant supply - for example unskilled labor inmost developing countries shy and serves as a vehicle for the adoptionof improved technologies embodied in new investments

While there have been brief periods of significant inflow ofexternal financial resources to some LDCs in the past foreign savingscannot be expected to provide a sustainable basis for financingdomestic investment Raising the national saving rate is particularlyessential to developing countries with a heavy debt-service burdenand limited capacity to obtain loans in foreign capital markts Indeed macroeconomic adjustment programs oriented to theresumption of long-run growth invariably emphasize the need to expand domestic savings

Household saving is usually the largest component of domesticsavings in developing countries especially the lower-income predoshyminantly agricultural LDCs This contrasts with the much greaterimportance of corporate saving in developed countries The abilitywillingness and opportunity of households to save over time cantherefore significantly influence the rate and sustainability of capitalaccumulation and economic growth in developing countries

151 BAUTISTA amp LAMBERTE SAVING BEHAVIOn

There has recently been a growing recognition of the pivotalrole of agriculture and the rural sector in promoting a more rapid and equitable economic growth for low-income LDCs 2 Advocates of an agriculture-based development strategy (ADS) emphasize the expanshysion of public investments in the rural sector and the removal ofpolicy-induced price biases against agricultural products These are expected to raise farm productivity agricultural output and ruralincome which in turn will generate increased intermediate demand for nonagricultural inputs and more importantly will stimulate conshysumption demand for food and labor-intensive industrial goods and services generating employment and income multiplier effects on therural regional and national economies In the short to medium termrural household income is expected to increase at a faster rate comshypared to urban household income

How will the adoption of ADS and tne associated shift in ruralshyurban income distribution affect aggregate household savings In any given year it is commonly observed that the average saving rateof rural households is lower sometimes very much lower than that of urban households 3 If fixed saving rates for the rura and urban sectors are assumed the rising real income of rural households relashytive to urban households in a growth process that is agriculturallydriven will likely result in a lower aggregate saving rate calling intoquestion its sustainability Such an assumption on household saving behavior however is questionable

A positive relationship between the saving rate and incomein developing countries at least within certain ranges of income levels has been obtained in past empirical studies using household survey data (eg Bhalla [1980] for India) or cross-country nationalincome accounts (eg Moore 1981) for Asian countries An observed lower average saving rate for rural households (relative to urban households) may then be explained simply by their lower averageincome However rural household incomes can increase rapidly inthe course of agriculture-based development which may prevent a decline in the aggregate saving rate or even raise it Furthermore the improvement of investment opportunities in the rural areas associashy

2 See for example Adelman (1984) and Bautista (1988) An early statement ofan agriculture-based development strategy can be found in Mellor (1976)3 Lipton (1977 247) has aptly pointed out that apart from the significant undershyestimation of agricultural saving embodied directly in investment some rural saving isdrained off by price twists to finance socially low-yielding urban investment and that

farmers would have more incentive to save and to embody their savings in farm investmentif its returns were not artificially depressed by policies turning the terms of trade againstagriculture

152 JOURNAL OF PHILIPPINE DEVELOPMENT

ted with the ADS provides an additional stimulus to increased savingsby rural households

Empirical evidence on the relative size of the marginal savingrates for rural and urban houeholds is thin and contradictory In a study using 1963-72 survey data on Souith Korean householdsfarmers were found to have been considerably more thrifty with their marginal propensity to consume being almost half that of urban consumers (Lluch et al 1977 99) Based on a similar analyticalframework estimates of the marginal propensity to save (MPS) for Mexico show averages of 011 for rural households compard to 025 for urban households in Chile the values tend to be around 030with slightly lower levels for comparable urban households (p 241)Using data from household surveys in Bangladesh for each year from 197677 to 1978-79 the MPS out of transitory income isestimated to be consistently and significantly higher among rural households (Chowdhury 1987) On the other hand the findings from an Indian study using time series data indicate a higher aggregate MPS for the urban sector (Gupta 1970) To be sure comparability of saving rate estimates from independent studies for different countries by diffeshyrent investigators is severely impaired by differences in data and measures used and by the varying analytical approaches and estimashytion techniques adopted in deriving the estirrates4

In this paper we investigate the comparative saving behavior of rural and urban households in the Philippines using FIES (FamilyIncome and Expenditure Survey) data for 1985 the latest year for which such data are available Apart from the contribution to empishyrical knowledge in an area previously not given systematic attention in the development literature some policy interest attaches in the Philippine context to an examination of the differential saving behashyvior of rural and urban households considering the recent shift towards more rural-oriented development policies as expressed in the governments Medium-Term Philippine Development Plan 1987shy1992

Section II addresses some theoretical and empirical issues in savings behavior especially as they relate to existing conditions in developing countries and in particular the Philippines The discusshysion leads to the specification of household saving functions distinshyguished by region by location (rural and urban) and by income class In Section III we describe and evaluate the data base for the

4 For instance the use of an extended linear expenditure system and of sophisshyticated econometric estimation procedures in the studies on Korea Mexico and Chile cited above contrasts sharply with the simple linear regression of income on saving employed in the studies on Bangladesh and India

153 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

study indicating also some relevant characteristics of sample houseshyholds Section IV presents and analyzes the estimated saving equashytions The findings of the study are summarized and their implicashytions for development policy and stratey are briefly discussed in the concluding section

I1ANALYTICAL CONSIDERATIONS

In a fundamental sense ones saving isa sacrifice of current conshysumption that will allow for an increase in future consumptionSaving is a means not only of reducing fluctuations in income and smoothing consumption over time but also of earning interest Given the intertemporal nature of the saving process it is the lifetime income of the individual and not just the currect income as impliedin the Keynesian consumption function that should influence current saving This life-cycle model also suggests a higher rate of saving during certain periods eg in preretirement years in order to provide for consumption in old age The age of the individual would therefore be an appropriate determinant of savings Additionally existing investment opportunities (or more precisely the return to investments) and the cost of borrowing can also significantly affect saving behavior

That current saving depends on lifetime resources is also an important implication of Friedmans (1957) permanent income hypothesis Current income is viewed as the sum of permanent and transitory incomes the former reflecting the individuals lifetime earnings In its extreme form the permanent income approach postushylates equality between an individuals current consumption and pershymanent income implying that the MPS is zero out of permanentincome and one out of transitory income Existing empirical studies do not oear out this strict version of the permanent income hypotheshysis but they provide support to the view that the marginal saving rate is higher out of transitory income than out of permanent income

Quantitative information on actual choices about savings is typically available at the household rather than individual level at least among developing countries Such data are often providedthrough countrywide income and expenditure surveys and less freshyquently through special locationally more focused surveys on changes in family assets and liabilities Both sources of data are subjet to error Household income particularly the nonmonetized component is often observed to be underreported with significantdiferences in the degree of underestimation amng surveys done in the same country for different years (Berry 1985) If nonmonetized

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 3: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

150 JOURNAL OF PHILIPPINE DEVELOPMENT

neoclassical growth model in which savings do not influence economshyic growth in the steady state a higher saving rate isassociated with more rapid growth of the economy in its movement towards long-runequiitbrium As noted by Gersovitz (1988) this transitional path(which can take time) is more meaningful than alternative steadystates in representing the evolution of developing economies

There are of course other determinants of economic growthTechnological progress institutional development domestic policiesand the external economic environment have also been emphasizedin the development literature To these additional influences on ecoshynomic growth can be attributed the lack of a simple correlationsometimes observed between savings and growth in developing cotnshytries (LDCs) Only when these explanatory variables are simultashyneously taken into account would one be able to assess empiricallytheir separate effects on growth Although Arthur Lewiss famousdictum that raising the saving rate is the central problem in economshyic development can be disputed (Deaton 1989 39) few woulddoubt that economic growth cannot be long sustained under condishytions of declining saving rates

Savings including both domestic and foreign finance the(physical and human) capital formatin needed to increase outputand this isof particular importance to typically capital-scarce LDCsApart from it direct contribution to output growth capital accumushylation also makes pcssble the employment of complementary proshyduction inputs in abundant supply - for example unskilled labor inmost developing countries shy and serves as a vehicle for the adoptionof improved technologies embodied in new investments

While there have been brief periods of significant inflow ofexternal financial resources to some LDCs in the past foreign savingscannot be expected to provide a sustainable basis for financingdomestic investment Raising the national saving rate is particularlyessential to developing countries with a heavy debt-service burdenand limited capacity to obtain loans in foreign capital markts Indeed macroeconomic adjustment programs oriented to theresumption of long-run growth invariably emphasize the need to expand domestic savings

Household saving is usually the largest component of domesticsavings in developing countries especially the lower-income predoshyminantly agricultural LDCs This contrasts with the much greaterimportance of corporate saving in developed countries The abilitywillingness and opportunity of households to save over time cantherefore significantly influence the rate and sustainability of capitalaccumulation and economic growth in developing countries

151 BAUTISTA amp LAMBERTE SAVING BEHAVIOn

There has recently been a growing recognition of the pivotalrole of agriculture and the rural sector in promoting a more rapid and equitable economic growth for low-income LDCs 2 Advocates of an agriculture-based development strategy (ADS) emphasize the expanshysion of public investments in the rural sector and the removal ofpolicy-induced price biases against agricultural products These are expected to raise farm productivity agricultural output and ruralincome which in turn will generate increased intermediate demand for nonagricultural inputs and more importantly will stimulate conshysumption demand for food and labor-intensive industrial goods and services generating employment and income multiplier effects on therural regional and national economies In the short to medium termrural household income is expected to increase at a faster rate comshypared to urban household income

How will the adoption of ADS and tne associated shift in ruralshyurban income distribution affect aggregate household savings In any given year it is commonly observed that the average saving rateof rural households is lower sometimes very much lower than that of urban households 3 If fixed saving rates for the rura and urban sectors are assumed the rising real income of rural households relashytive to urban households in a growth process that is agriculturallydriven will likely result in a lower aggregate saving rate calling intoquestion its sustainability Such an assumption on household saving behavior however is questionable

A positive relationship between the saving rate and incomein developing countries at least within certain ranges of income levels has been obtained in past empirical studies using household survey data (eg Bhalla [1980] for India) or cross-country nationalincome accounts (eg Moore 1981) for Asian countries An observed lower average saving rate for rural households (relative to urban households) may then be explained simply by their lower averageincome However rural household incomes can increase rapidly inthe course of agriculture-based development which may prevent a decline in the aggregate saving rate or even raise it Furthermore the improvement of investment opportunities in the rural areas associashy

2 See for example Adelman (1984) and Bautista (1988) An early statement ofan agriculture-based development strategy can be found in Mellor (1976)3 Lipton (1977 247) has aptly pointed out that apart from the significant undershyestimation of agricultural saving embodied directly in investment some rural saving isdrained off by price twists to finance socially low-yielding urban investment and that

farmers would have more incentive to save and to embody their savings in farm investmentif its returns were not artificially depressed by policies turning the terms of trade againstagriculture

152 JOURNAL OF PHILIPPINE DEVELOPMENT

ted with the ADS provides an additional stimulus to increased savingsby rural households

Empirical evidence on the relative size of the marginal savingrates for rural and urban houeholds is thin and contradictory In a study using 1963-72 survey data on Souith Korean householdsfarmers were found to have been considerably more thrifty with their marginal propensity to consume being almost half that of urban consumers (Lluch et al 1977 99) Based on a similar analyticalframework estimates of the marginal propensity to save (MPS) for Mexico show averages of 011 for rural households compard to 025 for urban households in Chile the values tend to be around 030with slightly lower levels for comparable urban households (p 241)Using data from household surveys in Bangladesh for each year from 197677 to 1978-79 the MPS out of transitory income isestimated to be consistently and significantly higher among rural households (Chowdhury 1987) On the other hand the findings from an Indian study using time series data indicate a higher aggregate MPS for the urban sector (Gupta 1970) To be sure comparability of saving rate estimates from independent studies for different countries by diffeshyrent investigators is severely impaired by differences in data and measures used and by the varying analytical approaches and estimashytion techniques adopted in deriving the estirrates4

In this paper we investigate the comparative saving behavior of rural and urban households in the Philippines using FIES (FamilyIncome and Expenditure Survey) data for 1985 the latest year for which such data are available Apart from the contribution to empishyrical knowledge in an area previously not given systematic attention in the development literature some policy interest attaches in the Philippine context to an examination of the differential saving behashyvior of rural and urban households considering the recent shift towards more rural-oriented development policies as expressed in the governments Medium-Term Philippine Development Plan 1987shy1992

Section II addresses some theoretical and empirical issues in savings behavior especially as they relate to existing conditions in developing countries and in particular the Philippines The discusshysion leads to the specification of household saving functions distinshyguished by region by location (rural and urban) and by income class In Section III we describe and evaluate the data base for the

4 For instance the use of an extended linear expenditure system and of sophisshyticated econometric estimation procedures in the studies on Korea Mexico and Chile cited above contrasts sharply with the simple linear regression of income on saving employed in the studies on Bangladesh and India

153 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

study indicating also some relevant characteristics of sample houseshyholds Section IV presents and analyzes the estimated saving equashytions The findings of the study are summarized and their implicashytions for development policy and stratey are briefly discussed in the concluding section

I1ANALYTICAL CONSIDERATIONS

In a fundamental sense ones saving isa sacrifice of current conshysumption that will allow for an increase in future consumptionSaving is a means not only of reducing fluctuations in income and smoothing consumption over time but also of earning interest Given the intertemporal nature of the saving process it is the lifetime income of the individual and not just the currect income as impliedin the Keynesian consumption function that should influence current saving This life-cycle model also suggests a higher rate of saving during certain periods eg in preretirement years in order to provide for consumption in old age The age of the individual would therefore be an appropriate determinant of savings Additionally existing investment opportunities (or more precisely the return to investments) and the cost of borrowing can also significantly affect saving behavior

That current saving depends on lifetime resources is also an important implication of Friedmans (1957) permanent income hypothesis Current income is viewed as the sum of permanent and transitory incomes the former reflecting the individuals lifetime earnings In its extreme form the permanent income approach postushylates equality between an individuals current consumption and pershymanent income implying that the MPS is zero out of permanentincome and one out of transitory income Existing empirical studies do not oear out this strict version of the permanent income hypotheshysis but they provide support to the view that the marginal saving rate is higher out of transitory income than out of permanent income

Quantitative information on actual choices about savings is typically available at the household rather than individual level at least among developing countries Such data are often providedthrough countrywide income and expenditure surveys and less freshyquently through special locationally more focused surveys on changes in family assets and liabilities Both sources of data are subjet to error Household income particularly the nonmonetized component is often observed to be underreported with significantdiferences in the degree of underestimation amng surveys done in the same country for different years (Berry 1985) If nonmonetized

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 4: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

151 BAUTISTA amp LAMBERTE SAVING BEHAVIOn

There has recently been a growing recognition of the pivotalrole of agriculture and the rural sector in promoting a more rapid and equitable economic growth for low-income LDCs 2 Advocates of an agriculture-based development strategy (ADS) emphasize the expanshysion of public investments in the rural sector and the removal ofpolicy-induced price biases against agricultural products These are expected to raise farm productivity agricultural output and ruralincome which in turn will generate increased intermediate demand for nonagricultural inputs and more importantly will stimulate conshysumption demand for food and labor-intensive industrial goods and services generating employment and income multiplier effects on therural regional and national economies In the short to medium termrural household income is expected to increase at a faster rate comshypared to urban household income

How will the adoption of ADS and tne associated shift in ruralshyurban income distribution affect aggregate household savings In any given year it is commonly observed that the average saving rateof rural households is lower sometimes very much lower than that of urban households 3 If fixed saving rates for the rura and urban sectors are assumed the rising real income of rural households relashytive to urban households in a growth process that is agriculturallydriven will likely result in a lower aggregate saving rate calling intoquestion its sustainability Such an assumption on household saving behavior however is questionable

A positive relationship between the saving rate and incomein developing countries at least within certain ranges of income levels has been obtained in past empirical studies using household survey data (eg Bhalla [1980] for India) or cross-country nationalincome accounts (eg Moore 1981) for Asian countries An observed lower average saving rate for rural households (relative to urban households) may then be explained simply by their lower averageincome However rural household incomes can increase rapidly inthe course of agriculture-based development which may prevent a decline in the aggregate saving rate or even raise it Furthermore the improvement of investment opportunities in the rural areas associashy

2 See for example Adelman (1984) and Bautista (1988) An early statement ofan agriculture-based development strategy can be found in Mellor (1976)3 Lipton (1977 247) has aptly pointed out that apart from the significant undershyestimation of agricultural saving embodied directly in investment some rural saving isdrained off by price twists to finance socially low-yielding urban investment and that

farmers would have more incentive to save and to embody their savings in farm investmentif its returns were not artificially depressed by policies turning the terms of trade againstagriculture

152 JOURNAL OF PHILIPPINE DEVELOPMENT

ted with the ADS provides an additional stimulus to increased savingsby rural households

Empirical evidence on the relative size of the marginal savingrates for rural and urban houeholds is thin and contradictory In a study using 1963-72 survey data on Souith Korean householdsfarmers were found to have been considerably more thrifty with their marginal propensity to consume being almost half that of urban consumers (Lluch et al 1977 99) Based on a similar analyticalframework estimates of the marginal propensity to save (MPS) for Mexico show averages of 011 for rural households compard to 025 for urban households in Chile the values tend to be around 030with slightly lower levels for comparable urban households (p 241)Using data from household surveys in Bangladesh for each year from 197677 to 1978-79 the MPS out of transitory income isestimated to be consistently and significantly higher among rural households (Chowdhury 1987) On the other hand the findings from an Indian study using time series data indicate a higher aggregate MPS for the urban sector (Gupta 1970) To be sure comparability of saving rate estimates from independent studies for different countries by diffeshyrent investigators is severely impaired by differences in data and measures used and by the varying analytical approaches and estimashytion techniques adopted in deriving the estirrates4

In this paper we investigate the comparative saving behavior of rural and urban households in the Philippines using FIES (FamilyIncome and Expenditure Survey) data for 1985 the latest year for which such data are available Apart from the contribution to empishyrical knowledge in an area previously not given systematic attention in the development literature some policy interest attaches in the Philippine context to an examination of the differential saving behashyvior of rural and urban households considering the recent shift towards more rural-oriented development policies as expressed in the governments Medium-Term Philippine Development Plan 1987shy1992

Section II addresses some theoretical and empirical issues in savings behavior especially as they relate to existing conditions in developing countries and in particular the Philippines The discusshysion leads to the specification of household saving functions distinshyguished by region by location (rural and urban) and by income class In Section III we describe and evaluate the data base for the

4 For instance the use of an extended linear expenditure system and of sophisshyticated econometric estimation procedures in the studies on Korea Mexico and Chile cited above contrasts sharply with the simple linear regression of income on saving employed in the studies on Bangladesh and India

153 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

study indicating also some relevant characteristics of sample houseshyholds Section IV presents and analyzes the estimated saving equashytions The findings of the study are summarized and their implicashytions for development policy and stratey are briefly discussed in the concluding section

I1ANALYTICAL CONSIDERATIONS

In a fundamental sense ones saving isa sacrifice of current conshysumption that will allow for an increase in future consumptionSaving is a means not only of reducing fluctuations in income and smoothing consumption over time but also of earning interest Given the intertemporal nature of the saving process it is the lifetime income of the individual and not just the currect income as impliedin the Keynesian consumption function that should influence current saving This life-cycle model also suggests a higher rate of saving during certain periods eg in preretirement years in order to provide for consumption in old age The age of the individual would therefore be an appropriate determinant of savings Additionally existing investment opportunities (or more precisely the return to investments) and the cost of borrowing can also significantly affect saving behavior

That current saving depends on lifetime resources is also an important implication of Friedmans (1957) permanent income hypothesis Current income is viewed as the sum of permanent and transitory incomes the former reflecting the individuals lifetime earnings In its extreme form the permanent income approach postushylates equality between an individuals current consumption and pershymanent income implying that the MPS is zero out of permanentincome and one out of transitory income Existing empirical studies do not oear out this strict version of the permanent income hypotheshysis but they provide support to the view that the marginal saving rate is higher out of transitory income than out of permanent income

Quantitative information on actual choices about savings is typically available at the household rather than individual level at least among developing countries Such data are often providedthrough countrywide income and expenditure surveys and less freshyquently through special locationally more focused surveys on changes in family assets and liabilities Both sources of data are subjet to error Household income particularly the nonmonetized component is often observed to be underreported with significantdiferences in the degree of underestimation amng surveys done in the same country for different years (Berry 1985) If nonmonetized

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 5: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

152 JOURNAL OF PHILIPPINE DEVELOPMENT

ted with the ADS provides an additional stimulus to increased savingsby rural households

Empirical evidence on the relative size of the marginal savingrates for rural and urban houeholds is thin and contradictory In a study using 1963-72 survey data on Souith Korean householdsfarmers were found to have been considerably more thrifty with their marginal propensity to consume being almost half that of urban consumers (Lluch et al 1977 99) Based on a similar analyticalframework estimates of the marginal propensity to save (MPS) for Mexico show averages of 011 for rural households compard to 025 for urban households in Chile the values tend to be around 030with slightly lower levels for comparable urban households (p 241)Using data from household surveys in Bangladesh for each year from 197677 to 1978-79 the MPS out of transitory income isestimated to be consistently and significantly higher among rural households (Chowdhury 1987) On the other hand the findings from an Indian study using time series data indicate a higher aggregate MPS for the urban sector (Gupta 1970) To be sure comparability of saving rate estimates from independent studies for different countries by diffeshyrent investigators is severely impaired by differences in data and measures used and by the varying analytical approaches and estimashytion techniques adopted in deriving the estirrates4

In this paper we investigate the comparative saving behavior of rural and urban households in the Philippines using FIES (FamilyIncome and Expenditure Survey) data for 1985 the latest year for which such data are available Apart from the contribution to empishyrical knowledge in an area previously not given systematic attention in the development literature some policy interest attaches in the Philippine context to an examination of the differential saving behashyvior of rural and urban households considering the recent shift towards more rural-oriented development policies as expressed in the governments Medium-Term Philippine Development Plan 1987shy1992

Section II addresses some theoretical and empirical issues in savings behavior especially as they relate to existing conditions in developing countries and in particular the Philippines The discusshysion leads to the specification of household saving functions distinshyguished by region by location (rural and urban) and by income class In Section III we describe and evaluate the data base for the

4 For instance the use of an extended linear expenditure system and of sophisshyticated econometric estimation procedures in the studies on Korea Mexico and Chile cited above contrasts sharply with the simple linear regression of income on saving employed in the studies on Bangladesh and India

153 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

study indicating also some relevant characteristics of sample houseshyholds Section IV presents and analyzes the estimated saving equashytions The findings of the study are summarized and their implicashytions for development policy and stratey are briefly discussed in the concluding section

I1ANALYTICAL CONSIDERATIONS

In a fundamental sense ones saving isa sacrifice of current conshysumption that will allow for an increase in future consumptionSaving is a means not only of reducing fluctuations in income and smoothing consumption over time but also of earning interest Given the intertemporal nature of the saving process it is the lifetime income of the individual and not just the currect income as impliedin the Keynesian consumption function that should influence current saving This life-cycle model also suggests a higher rate of saving during certain periods eg in preretirement years in order to provide for consumption in old age The age of the individual would therefore be an appropriate determinant of savings Additionally existing investment opportunities (or more precisely the return to investments) and the cost of borrowing can also significantly affect saving behavior

That current saving depends on lifetime resources is also an important implication of Friedmans (1957) permanent income hypothesis Current income is viewed as the sum of permanent and transitory incomes the former reflecting the individuals lifetime earnings In its extreme form the permanent income approach postushylates equality between an individuals current consumption and pershymanent income implying that the MPS is zero out of permanentincome and one out of transitory income Existing empirical studies do not oear out this strict version of the permanent income hypotheshysis but they provide support to the view that the marginal saving rate is higher out of transitory income than out of permanent income

Quantitative information on actual choices about savings is typically available at the household rather than individual level at least among developing countries Such data are often providedthrough countrywide income and expenditure surveys and less freshyquently through special locationally more focused surveys on changes in family assets and liabilities Both sources of data are subjet to error Household income particularly the nonmonetized component is often observed to be underreported with significantdiferences in the degree of underestimation amng surveys done in the same country for different years (Berry 1985) If nonmonetized

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 6: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

153 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

study indicating also some relevant characteristics of sample houseshyholds Section IV presents and analyzes the estimated saving equashytions The findings of the study are summarized and their implicashytions for development policy and stratey are briefly discussed in the concluding section

I1ANALYTICAL CONSIDERATIONS

In a fundamental sense ones saving isa sacrifice of current conshysumption that will allow for an increase in future consumptionSaving is a means not only of reducing fluctuations in income and smoothing consumption over time but also of earning interest Given the intertemporal nature of the saving process it is the lifetime income of the individual and not just the currect income as impliedin the Keynesian consumption function that should influence current saving This life-cycle model also suggests a higher rate of saving during certain periods eg in preretirement years in order to provide for consumption in old age The age of the individual would therefore be an appropriate determinant of savings Additionally existing investment opportunities (or more precisely the return to investments) and the cost of borrowing can also significantly affect saving behavior

That current saving depends on lifetime resources is also an important implication of Friedmans (1957) permanent income hypothesis Current income is viewed as the sum of permanent and transitory incomes the former reflecting the individuals lifetime earnings In its extreme form the permanent income approach postushylates equality between an individuals current consumption and pershymanent income implying that the MPS is zero out of permanentincome and one out of transitory income Existing empirical studies do not oear out this strict version of the permanent income hypotheshysis but they provide support to the view that the marginal saving rate is higher out of transitory income than out of permanent income

Quantitative information on actual choices about savings is typically available at the household rather than individual level at least among developing countries Such data are often providedthrough countrywide income and expenditure surveys and less freshyquently through special locationally more focused surveys on changes in family assets and liabilities Both sources of data are subjet to error Household income particularly the nonmonetized component is often observed to be underreported with significantdiferences in the degree of underestimation amng surveys done in the same country for different years (Berry 1985) If nonmonetized

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 7: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

154 JOURNAL OF PHILIPPINE DEVELOPMENT

consumption is also understated the saving rate (but not necessarilythe level of savings) will be overestimated On the other hand household assets and liabilities tend to be incompletely enumerated (data on cash and jewelry are especially difficult to obtain) and nonshyfinancial investment is frequently not properly valued

Given the limitations of the data used some analysts have simply examined the determinants of household savings on a variableshyby-variable basis (eg Alamgir 1976) instead of using a systematicapproach to the modeling of saving behavior While income is geneshyrally acknowledged as the principal influence on saving there are conceptual andor data-related ditficulties in a developing country context in representing the income variable by current income or by some measure of lifetime savings In this study estimation of houseshyhold saving functions makes use of the two alternative measures of the income variable

The above representation of intertemporal decisionmakingabout individual saving needs modification if the household is to be the unit of analysis as required by data availability One complicating factor is the possible influence of family size on savings Other thingsthe same the higher the proportion of household members that consume more than they produce the lower will be the household saving (Left 1969) Family size as such would not be the relevant explanatory variable among rura families particularly even children can contribute significantly to the households production and income Some measure of dependency reflecting the unemployshyment of household members would be more appropriate

The strong retirement motive for saving in the life-cycle hyposhythesis can also be called into question in the context of the strongfamily ties that characterize many LDC households Especially in the rural areas where the extended family system is more prevalent there is a sense of obligation to care for the older economically inactive household members This also reduces the need for the youngermembers to save since their future consumption is expected to be provided for (at least partly) and weakens the expected relationship between age (of household head) and savings

Many households in developing countries are observed to have a low saving rate or even a negative one at low income levels This is sometimes interpreted to indicate problems of survey data reliability Alternatively however the low or negative saving rate may reflect rational household responses to current transitorily low incomes or to the high consumption needs of the poor since current consumpshytion is more likely to influence survival and efficiency-at-work at low levels of consumption (Gersovitz 1988 410)

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 8: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

1-05 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

As pointed out above there is someempirical evidence of risingsaving rates as income level increases As a prominent exampleBhalla (1980) has estimated a nonlinear saving function for rural households in India in which the average propensity to save is zero at the subsistence level and increases at an accelerating rate in the lowshyincome range followed by a deceleration and eventual tapering off to an asymptotic value However as in the formulation of other nonshylinear saving functions (eg quadratic or semilogarithmic) the process or mechanism that leads to the nonlinear relation between the saving rate and income level is not spelled out Due to deficienshycies in the data used the view has been expressed recencly that hypotheses about behavioral nonlinearities in savings cannot be dsentangled from problems in measuring the variables (Gersovitz1988 411) Alternatively one can estimate a linear saving function for (homogeneous) households differentiated by income group and then compare their estimated saving propensities This is the apshyproach adopted in the present study

In many developing countries where capital markets are signifishycantly fragmented (McKinnon 1973) households face different investment opportunities and costs of borrowing that could lead to differences in i-vrginal saving rates Lower-income households tend to be more vulnerable to capital market imperfections and this is attributable in large measure to their weak information base and to their inability to meet collateral requirements In the Philippinesthe differing locations of households in geographic areas separatedby wide spaces and having different consumption and production patterns are likely to imply varying rates of return to investments Access to credit likewise differs influenced in part by regionalvariations in the effectiveness of financial intermediation In addishytion insofar as taste is concerned (pjre time preferences in conshysumption) some ethnic classes in certain regions are traditionally known either for their frugal or spendthritt ways It would be approshypriate therefore to distinguish household saving functions by geoshygraphic regions

Within a given region it is necessary for purposes of this study to distinguish rural and urban households in their saving behavior Apart from likely differences in demographic characteristics social practices and educational background rural households are of course much more engaged in agriculture and face greater income variability relative to urban households il issometimes asserted that the consumption pressures of the demonstration effect are weaker in the rural areas where the scope for conspicuous consumption is more limited so that rural families will save more compared with

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 9: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

156 JOURNAL OF PHILIPPINE DEVELOPMENT

their urban counterparts Contrary to this the findings of a studyfor Taiwan indicate that increased awareness and ownership of modern consumption goods led to higher savings and that this relationship held within income groups (Freedman 1970 31)

It has been argued (although contrary currents of argumentalso exist) that there are more immediate outlets for investment in the farm than in the cities and that if interest paid and the rate of return in the rural sector are higher than those available to the urban population then farmers marginal propensity to save will be higher Also based on the permanent income hypothesis the relashytive instability of farm income would imply greater saving out of current income for rural households than for urban households

Some of the above considerations relate not to the rural-urban distinction which is locational but to the differentiation of houseshyholds by source of income Accordingly it would be useful to disshytinguish also between farm and nonfarm households especially since the proportions of farm (nonfarm) households in urban (rural) areas are not insignificant (see below)

Ill THE DATA BASE AND PROFILE OF SAMPLE HOUSEHOLDS

As indicated above the primary source of data for this study is the Family Income and Expenditure Survey (FIES) for 1985 The FIES is supposed to be conducted by the National Census and Statistics Office every five years beginning 1961 However after the third survey in 1971 it was only in 1985 that the next FIES was undertaken While previous surveys were carried out through only one round of interviews the 1985 FIES entailed two visits by enumerators (in July 1985 and January 1986) which obtained information for the first and second halves of the year Extensive reinterviews were also conducted subsequently to follow up on seemingly questionable survey responses Additionally some qualitashytive improvements were implemented in the 1985 FIES in dealingwth nonsampling errors and in the inclusion of noncash income and expenditures It isalso worth noting that the 1985 FIES and the earlier surveys differ in the number of regions into which the samplehouseholds are classified and in the regional grouping of provinceswhich precludes direct comparability of regional data from those

5 surveys

5 For an extended discussion of the relia )ility of 1985 FIES data and a comparisonwith other sources of savings data see Oshima 1988) and Lamberte and Bautista (1989 Ch IV)

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 10: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

157 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

The sample consists of 17495 households They were selected using a stratified two-stage cluster sampling design with villages or barangays (classified as either urban or rural) as the primary sampling units and households within each sample barangay as the secondary sampling units The sampling fraction was typically 1400 for urban areas and 1600 for rural areas with special sampling fractions applied to relatively small areas A total of 16971 sample households were successfully interviewed in the two visits

The distribution of sample households by region and by urbanshyrural classification is given in Table 1 (See Annex I for the names of regions and the provinces and chartered cities comprising each region and Annex II for the classification of rural and urban houseshyholds in the 1985 FIES) Since this study also examines the savingbehavior of farm and nonfarm households the sample households in each region are also classified in the table according to whether their income is derived mainly from agricultural or nonagricultural activishyties

Rural households constitute some 53 percent of the total sample households It is worth noting that 42 percent of rural houseshyholds are classified as nonfarm households While the proportionvaries by region Jranging from 25 to 54 percent) this indicates subshystantial nonagricultural activities in the rural areas As regards urban households a not insignificant proportion (11 percent) is engagedin farm activities also varying by region (from 7 to 32 percent) The lack of substantial correspondence between rural and farm households and between urban and nonfarm households suggests the usefulness of estimating saving functions separately for these four household categories

The mean values of household income savings and saving rate by region location and main source of income are shown in Table 2 The large differences in average household income across regions - especially those relating to the National Capital Region (NCR) the highest income region - reflect the past uneven developshyment of the Philippine economy and the geographic concentration of income growth Region Ill which is located close to the NCR ranks second The average household income in Region VIll which is the most depressed in the country is less than 30 percent of that in the NCR Within each region a wide disparity between rural and urban average household incomes can be observed with urban households consistently showing higher values in all the twelve regions A similar disparity is found if households are classified according to their main source of income In particular the average

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 11: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

158 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 1 DISTRIBUTION OF SAMPLE HOUSEHOLDS BY LOCATION AND BY MAIN OCCUPATION

Rural Urban

Region Farm Nonfarm

Sub total Farm Nonfarm

Sub total

Total

I

II

III

IV

381 (462) 410

(664) 366

(436)

593 (490)

444 (538) 207

(336) 474

(564)

618 (510)

825 (1000) 617

(1000) 840

(1000) 1211 (1000)

36 (87) 73

(316) 62

(71) 160

(139)

379 (913) 158

(684) 809

(929)

988 (861)

415 (1000) 231

(1000) 871

(1000) 1148 (1000)

1240

848

1711

2359

V

VI

VII

VIII

IX

X

XI

XII

NCR

438 (582) 654

(680)

387 (489)

436 (693)

411 (693)

374

(613) 481

(746) 359

(664)

-

314 (418) 307

(320)

404 (511)

193 (307)

182 (307)

236

(387) 164

(254) 182

(336)

-

752 (1000) 961

(1000)

791 (1000)

629 (1000)

593 (1000)

610

(1000) 645

(1000) 541

(1000)

-

50 (157) 89

(161)

70 (131)

56 (218)

36 (188)

74

(226) 107

(218) 40

(186)

-

268 (843) 465

(839)

463 (869)

201 (782)

156 (812)

254

(774) 383

(782) 175

(814)

2404

318 (1000) 554

(1000)

533 (1000)

257 (1000)

192 (1000)

328

(1000) 490

(1000) 215

(1000)

2404

1070

1515

1324

886

785

938

1135

756

2404

(1000) (1000)

TOTAL 5290 (587)

3725 (417)

9015 (1000)

853 (107)

7103 (893)

7956 (1000)

16971

Note Figures in parentheses are percentages of total rural or total urban households See Annex I for the names of regions and provinces chartered cities in each region

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 12: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

159 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 2 MEAN VALUES OF HOUSEHOLD INCOME (Y) SAVINGS (S)

AND SAVING RATE (SY) BY REGION LOCATION AND MAIN SOURCE OF INCOME

RegionHouseholds

1 All Households Rural Urban Farm Nonfarm

II All Households Rural Urban Farm Noniarm

II All Houeholds Rural Urban Farm Nonfarm

IV All Households Rural Urban Farm Nonfarm

V All Households Rural Urban Farm Nonfarm

VI All Households Rural Urban Farm Nonfarm

Y

(P)

32194 27490 41544 26780 40818

29286 24950 408P7 18305 31087 40439 28750 51712 18162 37346

31478 24073 39289 15764 34903

21506 17089 31953 13924 27374 26699 19027 40007 15530 31131

S

(P)

6732 5689 8805 5722 8720

6324 5513 8490 4160 4459 6664 3158

10045 1998 6909

5255 3967 6614 1472 6441

2619 1510 5242 1496 4037 4465 2537 7809 1899 5521

SlY

()

209 207 212 204 214

216 221 208 227 144 165 110 194 110 185

167 165 168 93

184

122 88

164 107 148 167 133 195 122 172

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 13: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

160 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 2 (continued)

RegionHouseholds Y S SlY

(P) (P) (

VII All Households 21758 4902 225 Rural 15465 3076 199 Urban 31096 7612 244 Farm 11261 2016 179 Nonfarm 25488 6391 251

VIII All Households 18666 2811 151 Rural 15568 1842 99 Urban 26249 5182 197 Farm 11916 1247 105 Nonfarm 26884 5845 217

IX All Households 24094 5226 217 Rural 22426 5128 229 Urban 29254 5529 189 Farm 15013 2274 152 Nonfarm 29192 6516 223

X All Households 27787 6265 226 Rural 21729 3535 163 Urban 39054 11342 290 Farm 26598 6620 249 Nonfarm 29107 3834 132

X 1 All Households 29210 6004 206 Rural 21831 3416 156 Urban 38924 9409 242 Farm 23195 3376 146 Nonfarm 32569 7519 231

XII All Households 25940 -3660 -141 Rural 21216 2499 118 Urban 37829 -6583 -174 Farm 17411 3442 198 Nonfarm 32434 -2818 -87

NCR All Households 64449 12790 198

Source Calculated from basic data in the 1985 Family Income and Expenditures Survey

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 14: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

161 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

income of farm households is about one-half that of nonfarm houseshyholds in almost all the regions

All regions except one show positive average household savings As indicated in Table 4 below the definition of savings used in this study includes expenditures on equipment consumer durables and education The housrthold saving rates vary considerably across regions ranging from -14 to 23 percent The relatively high saving rates in Regions I and II are not surprising considering that the population there (mostly the Ilocanos) is traditionally known for thriftiness The other regions that show relatively higher saving rates are Region VII where the second premier city of the country is located and Regions IX X and XI all from the Mindanao area which are heavily populated by migrant families There is a consishyderable difference in the saving rates between urban and rural houseshyholds within the same region In particular the formers saving rates are higher than those of the latter in nine regions

The differences in saving rates between farm and nonfarm households within the same region are also quite substantial The saving rates of nonfarm households in nine regions are higher than farm households This sit jation occurs in almost all regions where the saving rates of urban households are found to be higher than those of rural households

Table 3 compares the average saving rates between rural and urban households belonging to the same income bracket Two imporshytant observations should be noted First urban households dissave if their annual income falls below V20000 while rural households dissave if their annual income is below P15000 The differential cut-off income for dissaving between rural and urban households could be due to the higher cost of living in the urban areas Second for the same income class the saving rates of rural households are higher than those of urban households in all but one income class6

This would seem to suggest that the lower saving rates observed across regions for rural households can be attributed at least in part to their lower incomes compared to urban households

IV ESTIMATION RESULTS

Our preliminary regressions indicate the lack of significance and sometimes theoretically incorrect signs of the estimated coshyefficients of the following explanatory variables age of household

6 Lipton (1977 Ch 10) cites similar evidence of higher rural saving rates in Pakistan end India

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 15: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

162 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 3 HOUSEHOLD SAVING RATES BY INCOME CLASS AND LOCATION

Income class Philippines Urban Rural (in pesos) () () )

Under 2000 -32 -78 -21 2000 - 3999 -36 -86 -30 4000 - 5999 -22 -22 -22 6000 - 7999 -14 -17 -14 8000 - 9999 -8 -12 -7

10000 - 14999 -2 -7 -1 15000 - 19999 3 -2 4 20000 - 29999 5 2 7 30000 - 39999 10 8 12 40000 - 59999 13 11 16 60000 - 99999 17 15 23

100000 - 249999 24 33 24 250000 - 499999 3332 24 500000 and over 65 64 80

Total 13 16 10 Total no of households 95663 36024 59639

Source Lamberte and Lim (1987) Note Saving isdefined as total income minus total expenditures

head entered in quadratic form educational attainment of houseshyhold head distinguishing among five education categories and wifes employment status (employed or unemployed) In most cases these variables are highly correlated with household income so that their separate effects on savings cannot be disentagled Accordingly these variables have been excluded in subsequent reshygressions

Table 5 presents the estimation results for all households in each of the 13 regions distinguished in the 985 FIES based on the Keynesian saving function in which the income variable is measured as current household income (after taxes) From 48 to 95 percent of the variance in regional household savings is explained in the various estimated equations Without exception the coeffishycient estimates for the income variable are highly sigificant (at the 1 percent leve) They range widely from 0334 (Region II) to 0775 (Region X) indicating that an aggregate saving rate for Philippine households would mask large differences in regional

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 16: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

163 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Table 4 DEFINITION OF VARIABLES

Notation Definition

Y Household disposable income in pesos ( = household current income less taxes)

YP household permanent income in pesos YT Household transitory income in pesos

S Household savings in pesos ( = Y less total) household expenditures net of expenses on durable furnitures equipment and education)

LOC Location (1 for urban household 0 for rural household)

D Main occupation of household head (1 for nanfarm household 0 for farm household)

DR Dependency ratio ( = number of unemployed household members divided by household size)

saving propensities It is worth noting that the MPS estimate for the National Capital Region (Metro Manila) is the second lowest among the 13 regions in either of the two specifications As expecshyted the coefficient estimates for DR (dependency ratio) are negativebut in some cases they are not significantly different from zero 7

Distinguishing between rural and urban households in each region the estimated saving equations in Table 6 also based on the current income model show some significant differences in MPSestimates for the two household classes Rural households in Reshygions I and IX have markedly higher coefficient estimates for the income variable compared to urban households while the oppositeis true for Regions IV VII X XI and XlI It is not possible thereshyfore to make a generalized inference on the relative saving propenshysities of rural and urban households in the Philippines

The MPS estimates for rural households which range from 0322 (Region VIl) to 0735 (Region IX) in the specification without

7 As John Mellor has suggested in a private communication the composition not just the number of unemployed household members (the numerator in the dependencyratio) would be a relevant factor Because education expenditure counts as saving morechildren of school-going age should imply higher saving - whichcounterbalances the exshypected negative effect of DR

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 17: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

164 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 5 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

R2 R 2Region Conit Y Const Y DR

I -7353 0438 070 -6389 0438 -1418 070 (-186) (534) (-63) (534) (-10)

II -3460 0334 048 -2074 0335 -2233 048 (-70) (279) (-20) (280) (-15)

III -13289 0498 066 -12411 0493 -1272 066 (-277) (580) (-103) (580) (-08)

IV -9387 0465 066 -8008 0466 -2104 066 (-299) (682) (-107) (682) (-20)

V -5966 0339 054 -4194 0400 -2733 054 (-168) (355) (-50) (356) 1(-23)

VI -10419 0558 077 -8562 0557 -2934 07 (-246) (718) (-78) (717) (-18)

VII -5019 0456 072 -2525 0458 -4100 073 (-175) (588) (-39) (593) (-43)

VIII -5052 0421 060 -4066 0422 -1569 060 (-158) (362) (-57) 363) (-15)

IX -11600 0698 086 -9697 0699 -2744 087 (-299) (708) (-85) (710) (-18)

X -15266 0775 095 -13612 0775 -2497 095 (-327) (1281) (-93) (1281) (-12)

XI -13664 0673 086 -10788 0673 -4244 086 (-307) (824) (-81) (825) (-23)

XII -10954 0563 075 -8944 0563 -2853 075 (-229) (471) (-59) (472) (-14)

NCR -12352 0380 073 -7194 0380 -7354 074 (-162) (821) (-28) (820) (-21)

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of determination

Significant at the one percent level Significant at the five percent level

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 18: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

165 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

the dummy variable for main source of income may seem quitehigh Nearly all these estimates however are even lower than the estimated MPS of 0728 obtained in a recent study of rural savingsin the Philippines (Rodriguez and Meyer 1988) using a different data set (based on a special survey involving 980 rural households in six provinces)

Also from Table 6 a significant negative influence of the dependency ratio is seen for rdral households in most regions 8 in the case of urban households only one region (V I) shows a statisticshyally significant coefficient etimate for DR A possible explanation is that compared to urban hcuseholds there are greater opportunitiesfor farm work among the very young and very old m bers of rural households but less opportunity for spending the addiional income That the marginal saving rates differ by source of income for rural and urban households in several regions is indicated by the statisticshyal significance of the coefficient estimates for the interaction term DY They are mostly positive for rural households implying that their MPS out of nonagricultural income is higher than that out of agricultural income 9

Tables 7 and 8 contain the estimated saving equations using the permanent (lifetime) income specification As discussed in Annex Ill the estimation of permanent income is based on the hypotheshytical earning capacity of households determined from the estimated relationship for each region between household income and various indicators of the stock of human physical and financial assets Transhysitory income is derived residually after subtracting permanent inshycome from disposable income The coefficient estimates for both permanent and transitory incomes are seen to be statistically signishyficant almost always at the one percent level It is also remarkable that the estimated MPS out of transitory income is higher than hat out of permanent income in all but one of the estimated equationsHigher values of the adjusted coefficient of determination are shownin Tables 7 and 8 compared to those in Tables 5 and 6 indicatingthat a larger proportion of the variance in savings is explained byconsidering the separate influences of the permanent and transitory

8 The coefficient estimate for the dependery ratio has also been found in theRodriguez-Meyer study to be negative it is not statistically significant in the current income specification of the saving equation but significant at the 5 percent leve based on the per manent income model

9 imilarly Bhalla (1980) obtained generally lower MPS estimates for Indian ruralhouseholds out of agricultural income relative to other income sources Based on time-series national income accounts data Burkner (1981) found a significant positive relationshipbetween household savings and the ratio of industrial to agricultural income for the Philipshypines but a negative relationship for Thailand

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 19: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Table 6 ESTIMATED SAVING EQUATIONS CURRENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const Y DR R2 Const Y DR D DY R2

Region I Rural -8352 0561 -2024 876 -2707 8317 -1884 -7486 0273 078

(-79) (512) bull (-14) (-21) (82) (-13) (-70) (68)

Urban -4579 0358 -2100 070 -640 8204 -2168 -4112 0155 078 (-24) (308) (--08) (-01) (09) (-08) (-18) (02)

Region I1 Rural -2498 0364 -1686 055 -4107 0443 -1667 2582 -0125 057

(-30) (277) (-14) (-45) (209) (-14) (29) (-46) o c Urban -2686 0331 -3815 043 -4896 0432 -3932 2441 -0103 043 z

(-08) (132) t-08) (-10) (30) (-09) (06) (-07) -

Region III -Rural -7489 0475 -4346 052 -4192 0354 -3483 -6292 0169 054

(-6 4) (304) (-28) (32) (106) (-23)- (-53)- (45) F

Urban -17782 0518 1507 070 -9879 0321 1586 -8264 0201 070 -(-85) (445) (-05) (-2 2) (3 0) (16) (-20) (18) m

Region IV m Rual -2821 0362 -2944 049 -3369 0434 -319 198 -0076 050 m

(-39) (341) (-30) (-40) (163) (-C1) (02) (-26) 0

Urban -13330 0514 -323 0 73 -5889 0313 -6 -8652 -0210 873

(-103) (555) (-02) (-25) (36) f-00) (-40) (-24) A

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 20: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Table 6 (continued)

RegionHousehold

Region V-gtRural

Urban

Region VI Rural

Urban

Region VIIlt Rural

Urban

Region VIII Rural

Urban

Region IX Rural

Const

-3033 (-44)

-6986 (-32)

-4462 (-65)

-15732 (-54)

-1312 (-29) -3564

(-24)

-3375 (-56)

-5664 (-28)

-8403 (-27)

y

0417 (300)

0413 (193)

0429 (412)

0597 480)

0322 (3653 0518

(425)

0400 (284)

0444 (199)

0735 (166)

DR

-3928 (-41)

-1528 (-05)

-1908 (-19)

-506 (-01)

-975 (-15) -7740

(-37)

-1596 (-19)

-1227 (-04)

-4271 (-33)

2

855

054

064

081

063

077

056

061

091

Const

-286 (-04)

1205 (035)

-3732 (-52)

-6711 (-19)

-823 (-14) -2242

(08)

-2572 (-37)

-2352 (-03)

-2973 (-27)

0238 (75)

0300 (03)

0404 (232)

0476 (124)

0238 (76)

0313

(19)

0336 (102)

0252 (18)

0506 (166)

R

-3764 (-41)

-1784 (-06)

-1864 (-19)

--84 (-02)

-984 (-15) 07111

(-34)

-1487 (-18)

-1002

0

-4424 (-66)

-8767 (-29)

-2062 (-31)

11219 (-43)

-58 (-01) -6977

(-26)

-1504 (-22)

-3863 (-04)(-14)

-4549 -10463 (-33) (-123)

DV

0232 (60)

0399 (35)

0048 (22) deg

0138 (34)

0086 (26)

0212

(13)

0085 (23)

0199 (14)

0273 (85)

R2

C --I

057 gt

055

064

081

~ m

T

0n

ltlt

M

m

063

078

056

061

093

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 21: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Table 6 (continuedl

RegionHousehold Const 2y DR R2

Const Y DR D DY R

Urban -7150 0450 -672 053 -3855 0299 -440 -4127 0168 033 (-23) 144) (-02) (-10) (26) (-01) (-14) (-14)

lgion X Rural -3635 0415 -2752 065 -2819 0393 -2803 -2413 0860 065

(-44) (334) (-24) (-33) (229) (-32)- (-32)- (23) Urban -18685 8796 -1615 -096 -105 0157 -1414 -21065 0644 097

(-57) (022) (-03) (-00) (10) (-03) (59) (42)

Region X I Rural -1459 0323 -3220 050 -1770 0369 -3582 103 -0068 052

(-21) (256) (-34) (-24) (197) (-38) (02) (-26) 9-Urban -16649 0711 -2432 089 -7134 0495 -2727 -10536 0223 089 0

(61) (630) (06) (-19) (60) (-07) (-36) (27) zRegion Xl1

Rural -69 0332 -6328 044 -549 0443 -7269 0127 047 0-16 (-01) 199) (-49) (-05) 159) (-58) (-00) (-00) M

Urban -21298 0634 5641 083 -12402 0625 3032 -8977 0017 084 -shy(-50) (321) (10) (-22) (52) (05) (-22) (01) M

z m

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R denotes adjusted coefficient of 0 ltm

determination mSifnificant at the one percent level 0

Significant at the five percent level z -1

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 22: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

169 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Table 7 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

ALL HOUSEHOLDS BY REGION

t-values R2 denotes adjusted coefficient of determination

Region Const YP YT R2

I -2625 0290 0494 073

(48) (197) (540)

II -39 0218 0388 050 (-01) (105) (274)

III -7336 0346 0546 068 (-100) (214) (560)

IV -3592 0281 0527 070 (-77) (216) (704)

V -2560 0242 0472 058 (-52) (125) (362) deg

V I -3230 0291 0622 081 (-57) (177) (786)

VII -2475 0339 0511 074 (-67) (255) (565)

Vill -2846 0303 0479 062 (-64) (151)deg (346)

IX -7106 0512 0737 088 (-115) (226) (715)

X -8960 0548 0803 096 (-147) (331) (1385)

X I -7506 0462 0722 998 (129) (261) (856)

XlI -1804 0211 0652 082 (-27) (94) (578)

NCR -3865 0263 0401 075 (-36) (228) (816)

Notes See Table 4 for definitions of symbols Numbers in parentheses are

Significant at the one percent level

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 23: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Table 8 ESTIMATED SAVING EQUATIONS PERMANENT INCOME MODEL

RURAL AND URBAN HOUSEHOLDS BY REGION

RegionHousehold Const YP YT R2 Const YI 1 I) DY DVT R2

Region I Rural -5693 0414 0618 078 -1854 0218 0361 -4380 0212 0278 07

(-92) (210) (500) (-16) (42) (88) (-30) (37) (64) Urban -887 0235 0409 073 -2170 0205 0243 1567 0026 0167 073

(-08) (111) (307) (-06) (09) (09) (04) (01) (06)

Region I1 Rural -2213 0310 0386 056 -3282 0360 0468 1210 -0059 -0143 057

(-34) (130) (246) (-35) (93) (200) (09) (-12) (-46) 0L

Urban -250 0214 0389 046 -6949 0413 0435 8690 -0230 -0047 046 c m

(-01) (51) (130) (-14) (24)-- (29)- (15) (-13) (-03) zgt Region III 0-

Rural -2635 0202 0566 058 -2312 0184 0428 -210 0012 0176 059 -0n (-30) (70) (333) (-18) (36) (116) (-01) (02) (42)

Urban -11374 0414 0556 070 -9070 0328 0320 -2165 0083 0240 071 1

(-89) (186) (416) (-16) (22) (29) (-04) (05) (22) z Region IV m

m Rural -1235 0215

(-24)(110) 0416

(110) 052

(392) deg -2459 0279

(-25) 0461 (56)

1808 (167)

-0080 (14)

-0055 052 (-14) (-18)

lt ri

Urban -6372 0330 0567 075 -5028 0279 0332 -1478 0052 0239 076 K (-77) (176) (567) (--20) (27) (37) (-06) (05) (27)

mZ -shy4

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 24: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Table 8 (continued)

RegionHousehold Const yp YT R2 Const P T D DYP YT Rc

Region VRural -2712

(-55) 0247 (93)

0473 (300)

057 -531 (-07)

0086 (17)

0276 (83) deg -2704

(-25) 0188 (30)

0246 (65)

059

Cn -

gt

Urban -2423 0240 0483 058 3764 -0154 0139 -6306 0405 0355 059 a)

0

Region VI (-17) (62) (198) (10) (-10) (11) (-16) (25) (28) m

-c Rural

Urban

-3423 (-81)

-5232

0313 (168)

0330

0481 (394)

0651

066

084

-3512 (-52)

-5265

0322 (86)

0422

0421 (230)

0491

-84 (-01)

-1241

-0012 (-03)

-0078

0108 (44)-

0174

066

084

z o

z

Region V hRural

Urban

Region VIII

(-35)

-1732 (-60)

-4143

(-49)

(112)

0311 (205)

0378 (169)

(513)

0327 (301)

0572 (410)

063

079

(-18)

-500 (-10)

-1275 (-04)

(51)

0167 (41) deg

0269 (15)

(120)

0262 (81)

0308 (18)

(-04)

-912 (-14)

-3401

(-11)

(-09)

0153 (34)

0117 (06)

(40)

0062 (18)

0266 (15)

064

079

lt

Rural

Urban

Region IX

-2429 (-57)

-2974

(-24)

0274 (113)

0312

(79)

0457 (276)

0507

(189)

058

063

-1695 (-25)

-2825

(-10)

0199 (41)

0243

(16)

0373 (110)

0258

(16)

150 (01)

262

(01)

0052 (09)

0061

(04)

0106 (27)

0256

(16)

059

063

Rural -8410 (-145)

0604 (276)

0761 (759)

092 -5085 (-50)

0443 (75)-

0514 (158)

-9780 (-63)

0291 (44)

0270 (79)

093

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 25: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

Table 8 (continued)

RegionHousehold Const YP YT R2 Conit YP YT D DYP DYT R2

Urban

Region X Rural

Urban

Region X I Rural

Urban

-4414 0340 (-26) (69)

-3811 0338 (-78) 169)

-12655 0616 (-91) (259)

-696 0188 (-12) (79)

-11964 0549 (-100) (221)

0507 136)

0459 (296)

0821

(962)

0369 (255)

0749 (631)

055

066

097

053

090

-2212 (-07)

-2699 (-42)

-162

(-01)

-404 (-05)

-8748 (-27)

0198 0362 (13) (28)

0289 0449 (97) (202)

0106 0186 (06) (12)

0180 0410 (51) (205)

0486 0499 (49) (57)

-2356 (-06)

-3077 (-29)

-14424

(-38)

-1781 (-14)

-3778 (-11)

0151 (09)

0113 (27)

0529

(32)

0055 (11) 0064 (06)

0156 (11)

0029 (09)

0636

(42)

-0089 (-31) 0256 (29)

054

056

097

5

090 C-M

z

Region XII Rural

Urban

-202 0127 (03) (43)

-8792 0406 (-52) (107)

0417 (218)

0699

(343)

048

086

-1782 0221 (-14) (32)

-9366 0595

(-23) (43)

0464 157)

0656

(50)

1735 (10)

-549 (-01)

-0106 (-13)

-0186 (-13)

-0080 (-21)

0046 (04)

048

086

0 -i

-0a

z

Notes See Table 4 for definitions of symbols Numbers in parentheses are t-values R2 denotes adjusted coefficient of lt

determination

Significant at the one percent level Significant at the five percent level

mr 0

zm

-4

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 26: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

173 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

components of current income The importance of lifetime resourcesrather than just current income in the determination of household savings (both rural and urban) in the Philippines is therefore indicated

From Table 7 the regional MPS for all households rangesfrom 0218 to 0548 out of YP and from 0388 to 0803 out of YT - estimates that differ substantially from the values of zero and onerespectively as postulated in the strict version of the permanentincome hypothesis The latter result also emerges when a distinction is made between rural and urban households (Table 8)

It is evident from Table 8 that there are some significantdifferences from the MPS out of YP and YT estimated for rural and urban households in each region Again it is not possible to generalshyize on the relative magnitudes of the saving propensities between the two household classes We note that rural households show higherMPS out of permanent income in four regions (I II V and IX) and out of transitory inccme in three regions (I Ill and IX) It also appears from the significant coefficient estimates for DYP and DYT in several equations that given the households location (rural or urban) the marginal saving rates differ by main source of income such that many of them have a positive sign again indicating higherMPS values for nonfarm vis-a-vis farm households

A final set of regressions distinguishes three income groups (firstsecond and third terciles) among rural and urban households in each region Table 9 summarizes the resulting MPS estimate from the pershymanent income model10 A striking observation is the markedlyhigher MPS out of either permanent or transitory income for the higher income group This is especially true among rural households which show comparable average MPS estimates for the low- and middle-income groups that are only about one-half for the highshyincome group In the case of urban households the MPS first increases sharply from the smallest value for the low-income group to an intermediate value for the middle-income group before climbing to the largest value for the high-income group

A possible explanation for this comparative saving behavior ofrural and urban households would be as follows As indicated above there is a wide differential between the average incomes of rural and urban households in the Philippines Because middle-income rural households also have generally low income levels (compared not only to the high-income rural household group but also to middle-income urban households) they presumably face conditions inimical to

10 For a full presentation of the regression results (involving 144 estimated savingequations in all) see Lamberte and Bautista (1989)

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 27: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

174 JOURNAL OF PHILIPPINE DEVELOPMENT

Table 9 REGIOAL AVERAGE VALUES OF ESTIMATED MAPS OUT OF

PERMANENT AND TRANSITORY INCOME BY RURAL AND URBAN HOUSEHOLDS AND BY INCOME GROUP

Low income Middle income High income All income groups

MPS out of PR RuraL 016 015 034 022 Urban 011 027 038 025

MPS out of YT Rural 024 027 053 035 Urban 014 030 056 033

Note Each entry represents the simple average of MPS estimates for the 12 regions based on the regression of Son YP and YT

saving to a similar extent as do low-income households (eg relatingto investment opportunities coping with borrowing costs and conshysumption needs)

The average MPS for low-income rural households out of either permanent or transitory income is seen in Table 9 to be higher reshylative to their urban counterparts While the average MPS estimates are higher for urban households in the other income categories theyare not significantly different (based on the two-tailed t-test) from those for rural households except in the middle-income group andthen only out of permanent income (027 versus 015) Likewise theMPS estimates for rural and urban households averaged across allregions and income groups are not significantly different It seemsreasonable to infer from all this considering the much lower averageincome of rural households relative to urban households in each ofthe three income categories that in general rural households in the Philippines have not only a higher average saving rate (as shown earlier) but also a higher marginal saving rate than urban householos at the same income level

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 28: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

175 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

V CONCLUSIONS

The empirical findings of this study can be briefly summarized and some inferences can be made from them as follows

(1) Income is the most important determinant of household saving in the Philippines This result is true for all the altershynative measures of current income and its permanent and transitory components to the inclusion or exclusion of other explanatory variables and for all differences in the classification of households used (by region by rural-urban and by income class)

(2) Lifetime factors as represented in the permanent income measure by certain household characteristics have a signishyficant influence on household savings The findings also bear out the hypothesis that the marginal propensity to save out of transitory income is higher than out of permashynent income

(3) Marginal saving rates vary widely among households in different regions between rural and urban households and among different income classes The aggregate saving rate is therefore subject to change as income is redistributed across different household classes

(4) The marginal propensity to save of households in the Metro Manila area is estimated to be lower than in any of the countrys other twelve regions except one This would seem to suggest that a reversal of the past pattern of regional income growth biased toward Metro Manila can possitively affect the aggregate saving rate

(5) It is difficult to generalize about the relative size of the marginal saving rates between rural and urban households This contracts with the invariably higher average savingrates observed for urban households in the various regionsattributable to their higher incomes relative to rural houseshyholds However the estimated marginal saving rates for rural households in many regions are higher than their urban counterparts indicating a substantial scope for increased savings with rising rural income in the context of agriculture based development

(6) By main source of income nonfarm households in the rural areas tend to savemore in the margin than farm

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 29: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

176 JOURNAL OF PHILIPPINE DEVELOPMENT

(agricultural) households An increasing share of nonshyagriculture in rural income over time as the intersectoral linkage effects of agricultural growth work themselves out can then lead to a higher marginal saving rate among rural households

(7) At a given income level rural households generally save more or dissave less than urban households both on average and at the margin Moreover the marginal saving rate of rural households increases more rapidly as they move up from the low- and middle-income groups to the high-income group compared to their urban counterparts Under these conditions even if the average income and savings of urban households were initially higher faster income growth among rural households will not necessarily result in a lower aggregate saving rate

In sum the empirical evidence presented in this paper does not provide support to the notion that the adoption of an agricultureshybased development strategy and the associated shift in rural-urban income distribution inevitably entail some sacrifice in domestic savings and thereby in capital formation and the sustainability of economic growth Indeed the observed large potential for expandedrural savings is likely to be realized as the income prospects and inshyvestment opportunities are improved for rural households by inshycreased public investment in the rural areas and by reduced agrishycultural price distortions Nothing definite can be predicted of course until one does a detailed general equilibrium analysis takinginto systematic account the various factors affecting aggregate savings

As a final point the macroeconomic benefits of savings in a developing country context as described above do not enter in the calculation of individuals or households acting in isolation so that aggregate private savings are likely to be lower than is socially desishyrable (Sen 1967) To deal with this externality what is needed is not additional public sector saving but a subsidy to saving presushymably in the form of a higher marginal return (Deaton 1988 41)In fact there is an antisavings bias in many LDC government market interventions that repress financial intermediation keep interest rates low and reduce investment opportunities (McKinnon 1973) Such policy-induced sources of undersaving in the Philippines as discussed in Tan (1981) and Lamberte and Lim (1987) need to be addressed first before any ambitious government program of savings mobiliaation can be rationalized

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 30: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

177 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Annex I NAMES OF REGIONS AND PROVINCES

AND CHARTERED CITIES IN EACH REGION

Region

I

II

III

Name

Ilocos

Cagayan Valley

Central Luzon

IV Southern Tagalog

V

VI

VII

VIII

IX

Bicol

Western Visayas

Central Visayas

Eastern Visayas

Western Mindanao

X Northern Mindanao

XI Southern Mindanao

XII Central Mindanao

NCR National Capital Region

Provinceschartered cities

Abra Benguet Ilocas Norte Ilocos Sur La Union Mountain Province Pangasinan Batanes Cagayan Ifugao Isabel Kalinga-Apayao Nueva Vizcaya Quirino Bataan Bulacan Nueva Ecija Pampanga Tarlac Zambales Angeles City Olongapo City Batangas Cavite Laguna Marinduque Occidental Mindoro Oriental Mindoro Palawan Quezon Rizal Romblon Aurora Albay Camarines Norte Camarines Sur Catanduanes Masbate Sorsogon Aklan Antique Capiz Iloilo Negros Occidental Iloilo City Bacolod City Bohol Cebu Negros Oriental Siquijor Cebu City Eastern Samar Northern Samar Western Samar Leyte Southern Leyte Basilan Sulu Tawitawi Zamboanga del Norte Zamboanga del Sur Zamboanga City Agusan del Norte Agusan del Sur Bukidshynon Camiguin Misamis Occidental Misamis Oriental Surigao del Norte Butuan City Cagayan de Oro City Davao del Norte Davao del Sur Davao Oriental South Cotabato Surigao del Sur Davao City Lanao del Norte Lanao del Sur Maguinshydanao North Cotaato Sultan Kudarat Iligan City Manila Pasig Quezon City Caloocan City Pasay City Makati Other Metro

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 31: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

178 JOURNAL OF PHILIPPINE DEVELOPMENT

Annex II CLASSIFICATION OF RURAL AND URBAN HOUSEHOLDS

There is always an element of arbitrariness in differentiating between rural and urban areas which determines the classification of households into rural and urban In the 1985 FIES urban areas are defined to consist of shy

1 In their entirety all cities and municipalities having a population denshysity of at least 1OO persons per square kilometer

2 Poblaciones or central districts of municipalities and cities which have a population density ot at least 500 persons per square kilometer

3 Poblaciones or central districts (not included in 1 and 2) regardless of the population size which have the following

i street pattern ie network of streets in either parallel or right angle orientation

ii at least six establishments (commercial manufacturing recreational andor personal services)

iii at least three of the following (1) a town hall church or chapel with religious services at least

once a month (2) a public plaza park or cemetery (3) a public market or building where trading activities are carried

on at least once a week (4) a public building like a school hospital puericulture and

health center or library 4 Barangays having at least 1000 inhabitants which meet the conditions

set forth in 3 above and where the occupation of the inhabitants is predominantly nonfarming or nonfishing

All areas not falling under any of the above classifications are considered rural

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 32: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

179 BAUTISTAamp LAMBERTE SAVING BEHAVIOR

Annex III ESTIMATING PERMANENT INCOME

We follow Bhalla (1980) and Hyun et al (1979) among othersin using an earnings function on which to base the estimation of permanent income for-each household The procedure basically involves regressing disposable income on various indicators of earning capacity of the household and the predicted value is taken as the measure of permanent income The earning capacity of a house hold is assumed to be related to its stock of human capital and physical and financial assets

The available human capital of a household may be gauged in terms oi its educational background occupational status and the proportion of members employed We therefore included the following explanatory variables in the regression equation for household earning capacity educational attainment ofhousehold head represented by the five education categories distinguished in the 1985 FIES main source of income classified into fifteen occupationalcategories and the dependency ratio which is expected to have a negative effect

FIES data do not include ownership of physical and financial assets What we have done isropresent these stock variables by proxies for flow variables that can be identified with the asset values Available data on purchases of consumer durables are used to represent the ownership of physical assets As to financial assets the following income flow data are used as proxies inheritance received during the year pension and retirement benefits workmens compensation social security benefits dividends from investments and profits from the sale of stocks The assumption is ihat the higher the income from such sources the larger isthe value of financial assets held by the household

The estimated earnings equations for the thirteen regions are deemed geneshyrally satisfactory based on the expected signs of the coefficients t-values and adjusted coefficients of determination Copies of the regression results are ava~lshyable upon request from the authors

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 33: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

180 JOURNAL OF PHILIPPINE DEVELOPMENT

REFERENCES

Adelman I Beyond Export-Led Growth World Development 12 No 9 (1984)

Alamgir M Rural Savings and Investment in Developing Countries Some Conceptual and Empirical Issues Bangladesh Development Studies 4 No 1 (1976)

8autista R M Agriculture-Based Development Strategy Issues and Policy Framework Paper presented at a seminar of the Development Policy Gi-up Resource Systems Institute East-West Center Honolulu on July 20 1988

Berry A On Trends in the Gap Between Rich and Poor in Less Developed Guntries Why We Know amp) Little Review of Income and Wealth 51 No 4 (1995)

Bhalla SS The Measurement of Permanent Income and Its Application to Savings Behavior journalofPolitical Economy 88 No 4 (1980)

Burkner H P Savings Behavior and Savings Mobilization in Developing Coushytries Bochum Studienverlag Brockmeyer 1981

Chowdhury N Household Savings Behavior in Bangladesh Issues and Evishydence Bangladesh Development Studies 15 No 3 (1987)

Deaton A Saving in Developing Countries Paper presented at the First Annual World Bank Conference on Development Economics Washingshyton DC on April 27-28 1988

Freedman D S The Role of the Consumption of Modern Durables in Ecoshynomic Development Economic Development and Cultural Change 19 No 1 (1970)

Friedman M A Theory of the Consumption Function Princeton Princeton University Press 1957

Gerovit M Saving and Development Handbook of Development Econoshymics Edited by H Chenery and NT Srinivasan Amsterdam Elsevier 1988

Gupta K L On Some Determinants of Rural and Urban Saving Behavior Economic Record 46 (1970)

Hyun K N D W Adams and L J Hushak Rural Household Savings Behashyvior in South Korea American journal of Agricultural Economics 61 (1979)

Lamberte M B and R M Bautista Comparative Saving Behavior of Rural and Urban Households The Philippines 1985 Final report to the Training and Development Issues (TDI) Project National Economic and Development Authority Manila 1989

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981

Page 34: -omparative Rural and Urban Households n the …pdf.usaid.gov/pdf_docs/PNABK627.pdfin the development literature. ... have been brief periods of significant inflow of ... Philippine

181 BAUTISTA amp LAMBERTE SAVING BEHAVIOR

Lamberte MBand J irm Rural Financial Markets A Review of the Literature PIDS StaffPaper Series No 87-02 Makati Philippine Institute for Development Studies 1987 Leff N Dependency Rates and Saving Rates American Economic Review

59 No 5 (1969) Lipton M Why Poor People Stay Poor Urban Bias in World Development

Cambridge Harvard University Press 1977 Lluch C A A Powell and R A Williams Patterns in Household Demand

andSoving Washington DC World Bank 1977 McKinnon R I Money and Capital In Economic Development Washington

DC Brookings Institution 1973 Mellor J W The New Economics of Growth A Strategy for India and the

Developing World Ithaca Cornell University Press 1976 Moore B J Domestic Savings in Selected Developing Asian Countries

Economic Staff Paper No 2 Manila Asian Development Bank 1987 Oshima H Personal Savings in the 1985 Philippine Family Income and Expenshy

diture Survey Ouezon City University of the Philippines School of Economics 1988

Rodriguez J A and R L Meyer The Analysis ct Saving Behavior The Caseof Rural Households in the Philippines PIDS Working Paper Series No88-20 Makati Philippine Institute for Development Studies 1988

Sen A Isolation Assurance and the Social Rate of Discount QuarterlyJournalof Economics 81 No 1 (1967)

Snyder DW Econometric Studies of Household Saving Behavior in DevelopingCountries A Survey Journalof Development Studies 10 No 2 (1974)

Tan E The Structure and Growth of the Philippine Financial Market and theBehavior of its Major Components PIDS Working Paper Series No 81-06 Makati Philippine Institute for Development Studies 1981