Relationship Between Inflation and Bd

8/8/2019 Relationship Between Inflation and Bd

1/12

The Relationship between Inflation and the Budget Deficit in TurkeyAuthor(s): Kivilcim MetinSource: Journal of Business & Economic Statistics, Vol. 16, No. 4 (Oct., 1998), pp. 412-422Published by: American Statistical AssociationStable URL: http://www.jstor.org/stable/1392610Accessed: 25/11/2010 00:07

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=astata .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

American Statistical Association is collaborating with JSTOR to digitize, preserve and extend access to Journalof Business & Economic Statistics.
http://www.jstor.org/action/showPublisher?publisherCode=astatahttp://www.jstor.org/stable/1392610?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=astatahttp://www.jstor.org/action/showPublisher?publisherCode=astatahttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/1392610?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=astata


2/12

T h e Relationship etween Inf la t ion a n d th e

B u d g e t D e f i c i t in T u r k e y

Kivilcim METINDepartment f Economics, Bilkent University, nkara, Turkey [email protected])

This articleanalyzes

heempirical elationship

etween nflation and thebudget

deficit fortheTurkish conomy by a multivariate ointegration nalysis. A single-equation model shows that

the scaled budget deficit (as well as income growth and debt monetization) ignificantly ffectsinflation n Turkey. The conditional model of inflation s constant, nd t encompasses previouslyestimated model.

KEY WORDS: Cointegration; ncompassing; xogeneity; Turkish nflation.

An extensive iterature as examined he relationship e-tween the budget deficit and nflation. At a theoretical evel,Sargent and Wallace 1981) showed hat under certain con-ditions, f the time paths of government pending and taxesare exogenous, bond-financed eficits are nonsustainable,and the central bank should eventually monetize he deficit.

This will increase the money supply and inflation n thelong run. These findings have subsequently een general-ized for the open economy case and for alternative ormsof financing see Scarth 1987; Langdana 990).

The empirical relationship between the deficit and in-flation in developed countries has been studied in detail(see Hamburger nd Zwick 1981; Dwyer 1982; Hein 1983;Ahking and Miller 1985; King and Plosser 1985; Protopa-padakis and Siegel 1987; Burdekin and Wohar 1990; Ho1990). Empirical studies of developing countries ncludethose of Dornbush and Fisher (1981), Bhalla (1981), Sid-diqui (1989), Choudhary nd Parai (1991), Buiter and Pa-tel (1992), Dogas (1992), Sowa (1994), Hondroyiannis ndPapapetrou 1994), and Metin (1995). These studies did notyield conclusive esults on the relationship etween he bud-get deficit and nflation, ither n the short run or in the longrun. Specifically, Hamburger nd Zwick (1981) found thatgrowth n Federal Reserve debt holdings exerted a signif-icant inflationary mpact on the U.S. economy over 1961-1982, yet a growth n nonmonetized debt had a negativeshort-run ffect on inflation. Ahking and Miller 1985) mod-eled deficits, money growth, and inflation over 1950-1980as a trivariate utoregressive rocess. They found govern-ment deficits to be inflationary n the 1950s and 1970s butnot in the 1960s. Using a rational-expectations acro model

of Peruvian nflation, Choudhary nd Parai (1991) foundthat budget deficits, as well as the growth rate of moneysupply, have significant mpacts on inflation. Similarly, Do-gas (1992) found that the public deficit affects inflation nGreece. Hondroyiannis nd Papapetrou 1994) also founda relationship between the Greek government budget andprice level. Using an error-correction odel, Sowa (1994)found that inflation n Ghana s influenced more by outputvolatility han by monetary actors, both n the long run andin the short run.

For Turkey, Metin (1995) analyzed nflation using a gen-eral framework f sectoral relationships nd found that fis-

cal expansion was a determining actor for inflation. Theexcess demand or money affected nflation positively, butonly in the short run. On the other hand, imported nfla-tion, the excess demand or goods, and the excess demandfor assets in the capital markets had little or no effect oninflation. A key policy implication of Metin (1995) is that

Turkish nflation could be reduced rapidly by eliminatingthe budget deficit.The aforementioned eneral iterature nfluences he cur-

rent study, which builds directly on Metin (1995). The largepublic-sector udget deficits and he relatively high nflationin Turkey during the last four decades have sparked de-bate on their consequences or the Turkish conomy. Themain question s whether bond-financed eficits are infla-tionary or whether only monetized deficits are inflationary.To answer this question, his article investigates he rela-tionship between Turkish nflation and budget deficits over1950-1987. Although he government hifted from mone-tizing the deficit to bond financing n the mid-1980s, theshort annual sample on Treasury bonds precluded ortingout the effects of this alternative means of deficit inancing.Therefore, have used Metin's (1995) dataset or analyz-ing the relationship etween nflation and the public-sectorbudget deficit, considering closed-economy ublic-financeapproach. The closed-economy ssumption may appear e-strictive, but Metin (1995) showed he lack of external ef-fects in the determination f Turkish nflation. The empiri-cal analysis herein s of general nterest because many otherdeveloping ountries have experienced udget and nflationdifficulties imilar o those in Turkey.

Section 1 presents a historical background o the Turk-

ish economy or 1950-1987, and Section 2 develops a the-oretical framework based on the public-finance pproach.Section 3 tests for budget deficits and inflation being coin-tegrated and finds that they are). Although weak exogene-ity does not appear alid, a parsimonious onditional modelis still developed Sec. 4). This model is empirically on-stant, whereas the corresponding marginal model is not,thus showing super exogeneity for dynamics parameters.

? 1998 American Statistical AssociationJournal of Business & Economic Statistics

October 1998, Vol. 16, No. 4

412


3/12

Metin: The Relationship Between Inflation nd the Budget Deficit n Turkey 413

Additionally, he new conditional model encompasses hemodel of Metin (1995).

1. HISTORICAL ACKGROUND

This section presents a brief economic history of Turkey,focusing on inflation and budget inancing.

From he 1950s until 1980, the Turkish overnment on-

sistently ollowed a policy of import ubstitution, with pro-hibitions on imports of commodities. State economic en-terprises SEE's) were established o produce agriculturalcommodities, several manufactured oods, and minerals.In the late 1950s, the Turkish economy experienced se-vere balance-of-payment ifficulties nd rising nflation. Ef-forts to control nflation onsisted argely of price controls.Private-sector irms responded ither by shutting down orby selling on the black market. SEE's, however, old at of-ficial prices and experienced osses. As inflation ncreased,these losses reached enormous amounts. The losses wereautomatically inanced by the credits extended by the Cen-tral Bank to the SEE's, resulting n high money growth seeAktan 1964; Okyar 1965; Fry 1972, 1980; Krueger 1974,1995; Onis and Riedel 1993).

In 1958, Turkey mplemented a fairly typical Interna-tional Monetary Fund (IMF)-supported tabilization pro-gram, which improved he foreign-exchange ituation anddrastically reduced nflation. The most important ompo-nent of the program was an increase n the prices of SEEgoods, a component hat was featured prominently n the1970 and 1980 reforms as well. Raising hose prices n 1958resulted n an immediate and once-and-for-all ncrease nthe price level, after which the reduced rate of expansionof Central Bank credits reduced nflation. Although nfla-

tion dropped rom 25% n 1958 to less than 5% n 1959, realgross domestic product which had been declining) startedgrowing mmediately due to the greater availability f im-ports.

Turkey was among he more rapidly growing developingcountries during most of the 1960s, with an annual nfla-tion rate of 5%-10%. The nominal exchange rate was keptconstant after the 1958 devaluation. nvestment pendingincreased and was financed mainly by foreign aid. In thelate 1960s, foreign aid did not increase, but the rate of in-vestment spending was maintained. n addition, ome dif-ficulties appeared n obtaining mports, creating visible re-straints on economic activity and growth.

Although nflation was rising at the time, the main reasonfor the 1970 devaluation was foreign-exchange ifficulties.After the devaluation, xport earnings ncreased sharply,and Turkish workers n Germany and other western Eu-ropean countries started remitting a significant amount offoreign exchange. Because there was no mechanism ead-ily at hand for the Central Bank to sterilize these inflows,the money supply expanded apidly and nflation ncreased,reaching an annual rate of 25% by 1973. In the early andthe mid-1970s, the problem of the growing public-sectordeficit also arose from the expenditure ide. In particular,large salary increases were granted o civil servants, andsubstantial ncreases in transfer payments were made to

SEE's, which had financial deficits due to both increasedwage costs and a rise in the rate of investment y the SEE's(see Onis and Riedel 1993). The growth of governmentspending during a boom in the mid-1970s ed to rising bud-get deficits, for which the Central Bank provided a majorpart of the financing. The public sector borrowing equire-ment (PSBR) was 4.3% of gross national product GNP) n1973, more than doubling o 10.7% n 1979.

Inflation eached about 100% n 1980, apparently ed bymonetization of the public-sector deficit. Policy changesin the early 1980s were designed to shift Turkey's growthstrategy away from import substitution nd toward greaterintegration with the international market. The 1980 stabi-lization program attempted o deal with inflation by cre-ating greater efficiency in operating he SEE's, restrain-ing the growth of public expenditure, educing subsidies,and attempting o improve revenue collection. Under thegovernment's iberalization program, he financial perfor-mance of SEE's improved ubstantially. Unlike their per-formance during he previous decades, SEE's appeared ohave contributed

ositivelyto the financial position of the

central government n the 1980s. The government's estric-tive stance could not be fully maintained, however. ThePSBR remained at about 6% of GNP during he first halfof the 1980s and rose to 8.3% in 1987, the highest since1980. Contributing actors included slow growth of rev-enues, a strong ncrease n budget ransfers o loss-makingSEE's, higher than planned wage and salary raises in thepublic sector, and an election. After 1980, policy reformscontinued. Although nflation ell to approximately 5% n1982, it started ising again and continued o be a problemthroughout he 1980s.

2. THE ECONOMIC RAMEWORKThis section summarizes he theoretical model underlying

the empirical analysis. In a closed economy, t is assumedthat all public debt takes the form of noninterest-bearingmoney. The public sector budget dentity s then

G - T = AH (1)

orG-T AH

= (2)py PY 'where G is public-sector xpenditures, T is public-sector

revenues, Y is real income, P is the price level, and H isbase money. In a steady-state rowing economy, t followsthat

A(H*) =(H*) Q4H zP LY

AH- H* (Ap + Ay), (3)

pYwhere A is the difference operator; H*, Ap, and Ay arescaled base money (H/PY), inflation, and the growth rateof real ncome, respectively; nd variables n lower case arein logarithms. t is assumed hat the long-run ncome elas-ticity of the demand or money s unity. Then the simplified


4/12

414 Journal of Business & Economic Statistics, October 1998

2.0

2.1

Figure !.Consumer Price Index and Base Money: p = , h =

1/.4

PY1

.78

- = c +.2B - (5)

-4.91950 1955 1969" 1965 197 1975 1989 985 1999

Figure 1. Consumer Price Index and Base Money: , h =

budget constraint s

A(H*) _B-

H* (Ap + Ay). (4)

Solving (4) for Ap, I obtain he following relation:

Ap = c + 1B - Vb2Ay, (5)

where B is the scaled budget deficit (G - T)/H, c is theconstant term (interpretable s the inertial nflation rate),and V) and V2 are slope coefficients associated with thescaled deficit and ncome growth. Here, V1 and Q2 are equalcoefficients with an opposite sign [see Phelps 1973), Anandand van Wijnbergen 1989), and Rodrik (1990) for theoryand empirical nalysis]. The remainder f this article empir-ically analyzes he relationship etween the budget deficit,inflation, base money, and real income growth.

3. THE DATA, UNIT-ROOT TESTS, ANDCOINTEGRATION ANALYSIS

This section tests for unit roots in the series of in-terest (Sec. 3.1) and for cointegration between the series(Sec. 3.2).

6.3-

1.8.

1959 1955 1969 1965 1979 1975 1989 1985 1999

Figure 2. Revenues and Taxes: g = , t = --.

.12

-.041950 1955 1960 1965 1970 1975 1980 1985 1990

Figure 3. The Growth Rate of Real Income: Ay =

3.1 The Data and Unit-Root Tests

The data used are annual over 1950-1987. Budget expen-ditures

(G)and

budgetrevenues

(T)are from the

budgetand final accounts, respectively Turkish ira (TL) Billion].The general budget deficit (G - T) is the primary deficit,which excludes nterest payments TL Billion). The budgetdeficit does not include the SEE's deficit. Because reliablestatistics about SEE's deficits are available only after thesecond half of the 1970s, the general budget deficit s there-fore used as a proxy or the total deficit. The price evel (P)is the consumer price index with base year 1980, Y is realGNP (TL 1980 Billion), and H is base money. The compo-nents of base money are currency n circulation, ault cash,legal reserves, and Central Bank sight deposits TL Billion).The Appendix describes he data n greater detail.

Figures 1--4 show (h, p), (t, g), Ay, and (Ap, B), respec-tively. Visually, all series appear t least I(1); he augmentedDickey-Fuller (1981) (ADF) test statistics n Table 1 sup-port the graphical xplanation. and h appear (2) (Fig. 1),and h* is I(1). Government xpenditures g) and revenues(t) also seem to be 1(2) (Fig. 2), but the scaled deficit B isclearly I(1). Ay is I(0) and, from its plot, looks like a

.64r-

.56

.48

-.9.

1959 1955 1969 1965 1979 1975 1989 1985 1999

Figure 4. Inflation nd the Rescaled Budget Deficit: p =B =- -


5/12

Metin: The Relationship Between Inflation and the Budget Deficit in Turkey 415

Table 1. Augmented Dickey-Fuller Test Statistics

Variable

Null order g t B p h y h*

I(1) -1.15 -.89 -1.03 -.28 -.44 -1.39 -1.11(10) (1) (2) (1) (1) (5) (3)

1(2) -2.14 -2.20 -8.07** -2.78 -3.68 -8.51** -5.04**(3) (3) (1) (0) (0) (0) (2)

I(3) -5.95** -6.68** -7.26** -9.02**(2) (2) (0) (0)

NOTE: For a given variable nd null order, wo values are reported. he first ow s the t value, which s the ADF tatistic, nd the second row s the longest ignificant ag with ignificant value.Five ags are allowed n each variable's ADF regression, ut twelve ags are allowed or g and t. All regressions nclude constant erm and a trend. The sample s 1954-1987 (T = 34) if thevariables re in their og levels (except B), 1955-1987 (T = 33) if hey are in first differences, nd 1956-1987 (T = 32) ifvariables re in second differences. he critical alues are from MacKinnon(1991, table 1). Here and elsewhere n this article, ** and * denote rejection t the 1% and 5% critical alues.

stationary eteroscedastic eries (Fig. 3). Figure 4 capturesthe essence of the cointegration nalysis: Both Ap and thescaled budget deficit B share the same upward rend overtime.

3.2 System Cointegration Analysis

This subsection tests for cointegration among the se-ries (Ap, h*, B, Ay). I test for cointegration n a first-ordervector autoregression VAR), using the multivariate oin-tegration procedure of Johansen 1988) and Johansen andJuselius 1990). The VAR ncludes a constant erm, a trend,and an impulse dummy i1980). The impulse dummy rep-resents the structural hange in the Turkish economy thattook place n 1980. The constant and 1980 enter he systemunrestrictedly. he trend s restricted o lie in the cointegra-tion space because a quadratic eterministic rend n levelsof economic variables s not usually a sensible long-runoutcome (see Doornik and Hendry 1994). The cointegra-tion results are quite sensitive o the lag length of the VAR.

Our choice of one lag is based on the Schwarz and Hannan-Quinn criteria, both of which pointed to a single lag. Theestimation period s 1952-1987.

Table 2 summarizes he cointegration esults. It includesthe eigenvalues, he max and trace statistics, he standard-ized estimated feedback coefficients a and cointegratingvector p', and statistics for testing restrictions on a. Thecointegration est statistics are corrected or sample size(see Reimers 1992), and they suggest three cointegratingvectors. The residual misspecification ests appear atisfac-tory. None of the equations xhibits autocorrelation, nd heequations or B and Ay have nonnormal esiduals.

Because I find hree stationary elations, need to identifythe estimated ointegrating ectors before I interpret hem.Assuming hat Ay is trend tationary, he second row of thep3 s an inflation elation, and the third cointegrating ectoris including ust Ap and B, I test the identification f allcointegrating ectors. The expected 3' matrix will be

1 0 0 0

'= 01.*0 ,0 ? 10,J

and implementing hose identification estrictions eads tothe restricted orm /' and a matrix eported n Table 3. Thelikelihood ratio test statistic suggests that all three cointe-

grating vectors are identified X2(2) = 1.1559 .5611] (seeJohansen 1991, theorem 5.1).

From the standardized 3' eigenvectors, he first cointe-grating vector s the growth ate of real ncome. The secondone is an inflation elation:

Ap=

.58B + .35h*. (6)The public sector deficit B enters with a positive coefficient(.58), and scaled base money h* also has a positive coef-

Table . A Cointegration nalysis of {Ay, Ap, B, h*}

Eigenvalues .739 .662 .445 .085

Hypotheses r = 0 r < 1 r < 2 r < 3

Max statistic 40.4 32.5 17.7 2.795% critical value 27.1 21.0 14.1 3.8

Trace statistic 93.2 52.9 20.4 2.795% critical value 47.2 29.7 15.4 3.8

Standardized igenvectors P'Variable Ay Ap B h* Trend

1 .188 -.124 -.088 .0006-1.222 1 -2.515 -.769 .0128-1.042 1.745 1 .009 -.0191

.125 .611 -.443 1 .0052

Standardized djustment oefficients a

Ay -1.200 .097 .054 .006Ap -.692 -.129 -.201 .042B .079 .337 -.185 .071h* -.016 .076 -.037 -.115

Weak exogeneity test statistics

Variable Ay p B h*

X2(5) 2.41 12.963 12.506 38.848p value [.4911] [.0047]** [.0058]** [.000]**

Diagnostic tatistics

Variable Ay p B h*

Normality X2(2) 11.35** .61 7.93* .24ARCH 1 F(1, 25) 1.14 .58 .25 1.61AR 1-2 F(2, 25) .86 1.29 1.27 1.44

NOTE: r is the hypothesized umber f cointegrating ectors. The critical alues or he cointe-gration ests are from Osterwald-Lenum 1992). The Jarque-Bera 1980) normality est statistichas a X2 distribution ith 2 df under he null of normal rrors. ARCH F(dfl, df2) refers o thetest for ARCH rrors, ntroduced y Engle (1982). The AR1 F(dfl, df2) is the test for residualautocorrelation.


6/12


Table 3. A Restricted-Form ointegration nalysis

Standardized igenvectors '

Variable Ay Ap B h* Trend

1.000 0.000 0.000 0.000 0.0000.000 1.000 -0.585 -0.349 0.0000.000 1.148 1.000 0.000 -0.012

Standardized djustment oefficients

Ay -1.348 -.024 -.038

Ap -.348 -.487 -.085B -.157 .747 -.611h* -.067 .162 -.134

Weak exogeneity est statistics

Variable Ay Ap B

X2(6) 74.151 16.136 23.989p value [.000] [.000] [.000]

ficient (.35). The third stationary elationship s betweeninflation and the scaled budget deficit.

The standardized coefficients how that the main effectof the first cointegrating ector s on Ay. From the secondcolumn of a, feedback of the second cointegrating ectoron both B and Ap is .75 and -.49, respectively. The thirdcointegrating ector primarily ffects the scaled deficit B.Weak exogeneity for 3 can be tested using the Johansen(1992a,b) procedure. The results suggest that Ap, B, andh* cannot be assumed weakly exogenous or 0, but Ay canbe (see Table 2). Weak exogeneity of the variables s alsotested ointly with identification estriction nd rejected orAp, B, and Ay (see Table 3).

For inference, onditional models should have regressorsthat are weakly exogenous; ee Engle, Hendry, nd Richard(1983). In the context of cointegration, weak exogeneitymeans that nference about he cointegrating ector can beperformed n the conditional model without oss of infor-mation relative to a system analysis. Even lacking weakexogeneity, single-equation modeling can proceed, treat-ing the system-based stimated ointegration oefficients sgiven; see Juselius (1992). Section 4 develops such a con-ditional model and examines ts properties.

4. SINGLE-EQUATION ODELING

This section develops a parsimonious, onditional, ingle-equation model for inflation, n which inflation depends on

the scaled budget deficit, the real growth rate of income,and scaled base money. Section 4.1 develops a parsimoniousconditional model from a general autoregressive istributedlag and shows the constancy f this conditional model. Sec-tion 4.2 estimates some marginal quations and tests theirconstancy. Finally, Section 4.3 compares he model esti-mated by Metin (1995) with the conditional model devel-oped n this article, using the standard ncompassing rame-work.

4.1 Single-Equation Analysis and the Constancy of aConditional Model

Because weak exogeneity does not appear valid (except

for Ay), Juselius's (1992) approach s used for single-equation modeling. Recalling he cointegration nalysis nthe previous Section 3.2, a single inflation quation s con-structed. The inflation model includes he error-correctionterms ECM's) obtained rom the earlier ointegration nal-ysis. The first ECM (CI2) s constructed sing Equation 6),and the second ECM (CI3) s obtained rom the third row ofthe f' matrix given n Table 3. Then he general ECM model

involves A2p, AB, Ay (because t is stationary), Ah*, theirlags, and the lagged ECM's. Here, single-equation model-ing starts with an unrestricted ourth-order utoregressivedistributed ag (ADL) in the (log) levels of the variables,written as an error-correction odel:

k-2 k-2 k-2

A2pt = E liABt-Bi -E 2iAYt-i + 5E 3iAh;-ii=O i=O i=O

k-2

+ 5E04iA2pt-i + 5CI2t-1i=o

+ 6CI3t-1 + c + ut, (7)where k = 4 and c represents he constant erm, trend, andimpulse dummies 1980 and d55. The model suffered roma major outlier n 1955 that was not explained by the vari-ables in the information et and did not correspond o anyprevious historical events. Thus, I created a dummy d55)to pick this up. This equation s a reparameterization f theADL model and is in I(0) space. Furthermore, his equa-tion obviates he need for weak exogeneity with respect othe cointegrating stimates rom the Johansen-system ro-cedure.

Equation 7) is fitted over 1954-1986. Estimation esults

and diagnostic statistics are reported n Table 4, column2. The diagnostic statistics test against several alternativehypotheses-residual autocorrelation DW and AR), skew-ness and excess kurtosis (normality), utoregressive on-ditional heteroscedasticity ARCH), and heteroscedasticity(RESET). The estimated ECM model embodies he sensiblelong-run olution n (6) and has good diagnostic tatistics.The RESET test suggested a possible nonlinearity n themodel, however, perhaps because many of the disequilibriaare likely to interact.

The general ECM can be simplified. Modeling generalto specific, a parsimonious model of inflation s obtained(Table 4, col. 3):

A2pt = + .2487 + .002153trend + .3762i1980[.1912] [.00142] [.0479]

+.3357d55 - .3729A2pt_1 + .2031ABt[.0583] [.1864] [.1451]

-.704Ayt + .5179Ayt_2 + .5045Ah_2[.3809] [.3128] [.2884]

-.1772 CI2t-1 - .1062 CI3t_1 (8)[.1110] [.0561],

where R2 = .89, & = .0476, DW = 1.60, AR(2, 20) = 1.77,ARCH: F(1,20) = .13, Normality: X2(2) = 1.85, andRESET: F(1, 21) = 4.57.


7/12

Metin: The Relationship Between Inflation and the Budget Deficit in Turkey 417

Table 4. The Conditional nd Marginal Models

Dependent variable

A2p A2p AB Ay Ah*

Estimation method

OLS OLS RLS RLS RLS

Sample 1954-1986 1954-1986 1954-1986 1954-1987 1954-1986

Constant .097(0.659) .249(0.191) .038(0.026) .057(0.013)Trend .0019(0.0019) .0021(0.0014)il 980 .340(0.159) .376(0.048) .351(73.44) -.067(157.94) -.324(170.72)d55 .350(0.078) .336(0.058)A2Pt-1 -.359(0.282) -.373(0.186)A2Pt-2 .0532(0.201)ABt .165(0.278) .203(0.145)ABt-1 .140(0.591) -1.04(0.185)ABt-2 .079(0.355) -.862(0.326)ABt-3 -1.100(0.603)A Bt4 -.680(0.316)ABt-5 -.437(0.314)Ayt -.739(0.567) -.704(0.381)Ayt-1 -.010(0.580)AYt-2 .435(0.494) .518(0.313)Ayt-3 -.392(0.223)AYt-4Ayt-5 .358(0.154)Ahb* -.183(0.230)Aht,_ .139(0.271)

Ah_2 .495(0.450) .505(0.288)

Ah_-3Ah_-4Ah*_5 .400(0.115)

CI2t-1 -.086(0.104) -.177(0.111)CI3t- -.138(0.361) -.106(0.056)

R2 .9032 .8937 .5896 .4347 .4680& .0533 .0476 .0967 .0385 .0674F, df 9.3352(16, 16) 18.494(10, 22) 6.226(6, 26) 7.433(3, 29)DW 1.62 1.60 2.01 2.55 1.68

Normality X2 .2379 1.848 6.130* 1.009 .145

AR1-2 F, df 1.9397(2, 14) 1.77(2, 20) .168(2, 24) 1.580(2, 27) .925(2, 29)ARCH 1 F, df .205(1, 14) .1306(1, 20) .561(1, 24) .869(1, 27) .634(1, 29)RESET F, df 4.747(1, 15)* 4.571(1, 21)* .027(1, 25) .000(1, 28) 1.194(1, 30)

NOTE: The diagnostic hecks for residual utocorrelation AR 1-2F est with he degrees of freedom hown) onfirm he choice of relevant ag, residual eteroscedaticity f the ARCH orm ARCH1 F test) suggested by Engle 1982). RESET-F s a regression pecification est. It ests the null of correct pecification f the original model against he alternative hat powers of the dependentvariable re present.

White (1980) estimated standard rrors are in parenthe-ses. A2p depends on its own first ag and the current caledpublic-sector deficit. It is also influenced by real incomegrowth, ts second lag, and the lagged monetization f theeconomy. The time trend and dummies have an impacton inflation. Equation 8) suggests a positive relationshipbetween inflation and an appropriately caled deficit. TheECM's explain the behavior of inflation by revealing rela-tively rapid reactions. This model closely matches he the-ory model and appears statistically satisfactory rom thediagnostic ests except for the RESET F.

Parameter onstancy s also an important tatistical prop-erty. To examine constancy, ecursive east squares s usedbecause sequences of constancy tests yield tools for in-vestigating constancy rom the corresponding ne-step n-novations. From the sequence of innovations, Chow testscan be constructed or parameter onstancy distributed sF(1, t - k - 1) on the null]. Graphs provide a convenientway of portraying vidence about constancy. igure 5 shows

the recursively stimated oefficients f variables n (8) andplus or minus wice their recursively stimated tandard r-rors. Coefficients ary only slightly relative o their ex antestandard rrors. Figure 5 also records one-step residualsand corresponding alculated equation standard rrors orconditional nflation quation with 0 ? 2 estimated tandarderrors. The equation standard rror varies little. Figure 5finally plots the breakpoint Chow (1960) statistic for theinflation quation, which remains constant over the sampleperiod considered.

4.2 Nonconstancy of Marginal Models

Nonconstancy of the marginal models is related to theconcept of super exogeneity, which mplies that the param-eters of the conditional model remain constant, ven whilethose of the marginal model change (i.e., the Lucas critiquedoes not hold). This subsection stimates marginal modelsfor Ay, AB, and Ah*. Because of the results n Section 3.2,the parameters f interest here include ust the parameters


8/12


Constant = Trend = A2LP.1=+ 2S.E.= ...... + 2S.E. = ..... 2SE =.

.9 - . 8 - .6

"994 . ..3.- , . .....

................... ,

9.004- .3

4.3 .S..........':. ......... ....... .' . .. . ................. -. e99- -.6

-.3 .-.912 -91975 1980 1985 1990 1975 1980 1985 1998 1975 1980 1985 1990

ABt = Ayt = Ayt-2 =.+ 2S.E. =.. + 2SE. =... + 2SE.=

8 1

.6 - , .4 .8

.4-

.69 .-

.2 - --.4

-.2-.8 ..2

.6 ... ?.

........ .1.2

Ah 2 012 = - Cl3' = -

6 93-

......... .'9 .

.3 - -.1 /-.2......

-.4-.3 -.3-

-.6 -.6 -.41975 1980 1985 1990 1975 1980 1985 1990 1975 1980 1985 1990

(a)

.1

h - Resi2StepW - CHOW+22SE .E.=..2 .. .. 1%crit= ......

-.1 -." ".. ---

-.15 9


9/12


ReslStep= N4 CHOWs=_ 1%. crit=......S2 S. E ......1.5-

18 1.2

?0 .9 .. .. .. 1.2

. .. ... ..... .

. ..6.-.09

........................

-.18

-.27 01975 1980 1985 1990 1975 1980 .985 1990

(a) (b)

ReslStep= N4 CHOWs= I_.% cr it=+ 2-S. E.=.

.12- 1 -

.08-

..... ........... ........ . ..8

.04.6

1 4/ .4

..1 ...........................

1975 1980 1985 1990 1975 1980 1985 1990

(C) (d)

ReslStep=_ N4 CHOWs= IV. cit= ......+ 2*S. E.-=.18 1- -

0..86

-.06 ...4

-.18 .2 ... ....1975 1980 1985 1990 1975 1980 t985 1990

(e) (f)

Figure 6. (a) One-Step Residuals From a Marginal Model for AB With 0 ? 2 Estimated Standard Errors; (b) Breakpoint Chow Statistics for aMarginal Model of AB Normalized by Their One-off 1% Critical Values; (c) One-Step Residuals From a Marginal Model for Ay With 0 f 2 EstimatedStandard Errors; d) Breakpoint Chow Statistics for a Marginal Model of Ay, Normalized by Their One-off 1% Critical Values; (e) One-Step ResidualsFrom a Marginal Model for Ah* With 0 f 2 Estimated Standard Errors; (f) Breakpoint Chow Statistics for a Marginal Model of Ah*, Normalized byTheir One-off 1% Critical Values.

for dynamics n the conditional model. For each marginalvariable, we began with fifth-order utoregression includ-ing a constant, rend, and i1980) and applied a sequential

reduction procedure. The results are reported n Table 4,columns 4-6. For AB all lags matter. The residuals arenonnormal. igure 6, (a) and (b), graphs he one-step resid-


10/12


Table . Encompassing Test Statistics or Equation 8)and Metin's 1995) Equation 9)

Null hypothesis

Equation 8) Metin 1995)

Statistic Distribution Distribution

Cox N(0, 1) -2.75 N(0, 1) -7.39Ericsson N(0, 1) 1.86 N(0, 1) 3.95

Sargan X2(6) 6.79 X2(9) 14.71F F(6, 15) 1.19 F(9, 15) 2.63& .0487% .0601%

NOTE: T = 1954-1986.

uals and the sequence of breakpoint Chow statistics, whichshow considerable nonconstancy, with possible breaks in1977 and 1984.

For Ay, the third and fifth lags matter. Statistically, hemodel appears well specified with no rejections rom thediagnostic ests available. Figure 6, (c) and (d), plots the re-cursively estimated quation tandard rrors and the break-point Chow statistics. The marginal model of Ay appearsconstant.

For Ah*, only the fifth lag matters. The equation s sta-tistically satisfactory, nd t appears onstant Fig. 6, (e) and(f)]. Because he conditional model for A2p is constant andthe marginal model of AB is nonconstant, AB (at least)appears uper exogenous or the dynamic parameters n theinflation quation.

4.3 Encompassing Implications f the Conditional Model

A congruent model should encompass previous empir-ical findings explaining he same dependent variable seeHendry and Richard 1982, 1989; Mizon and Richard 1986).Consider wo rival

explanations,enoted Ml and M2. The

question was whether M2 can explain features of the datathat Ml cannot. This can be a test of Ml, with M2 provid-ing an alternative o see whether M2 captures any specificinformation ot embodied n Ml (see Doornik and Hendry1994, p. 237). Several variants f encompassing ave beenproposed-variance (Cox 1961), parameter Hendry 1983),reduced-form Ericsson 1983), exogeneity (Hendry 1988),and forecast (Chong and Hendry 1986). In this subsectionwe compare Equation 8) with an inflation equation esti-mated by Metin (1995), using such encompassing ests. Themodel from Metin (1995) is

Apt= - .064 +

1.111Bt-

3.901A((G-T)/Y)t

[.039] [.135] [.670]

+ 1.663Apv, + .229AECM-Mt[.362] [.099]

- .272(ECM-UIP)t/2 + .074ECM-PPPt_1[.093] [.044]

+ .257d55t - .234Ayt, (9)[.020] [.166]

where R2 = .8973,& = .0601, DW = 2.072, AR(2,26)= .55, ARCH: F(1,26) = 2.77, normality: X2(2) = 1.33,and RESET: F(1, 27) = 3.74. In the work of Metin (1995),ECM represents ectoral excess demands, where ECM-M,

ECM-PPP, nd ECM-UIP were derived rom the monetarysector, rom purchasing ower parity, and from uncoveredinterest-rate arity, and d55 is a dummy variable, whichpicks up a major outlier n 1955. Finally Ap, is consumerprice index (CPI) nflation or industrial ountries. Table 5reports he encompassing est results. As shown n Table 5,Equation 8) variance dominates Equation 9) (.00487 vs..0601). None of the encompassing ests reject (8), and allreject (9); the new model encompasses he old one. (Notethat APtl was added o (8) to calculate he encompassingtests.)

5. CONCLUDING EMARKS

This article examines he relationship etween he public-sector deficit and inflation. System cointegration nalysissuggests three stationary elationships. Although weak ex-ogeneity does not hold for variables oncerned except Ay),one is still able to develop a conditional model for inflation.In that model, an increase n the scaled budget deficit mme-diately ncreases nflation. Real income growth has a nega-

tive immediate ffect and positive second-lag effect on in-flation. Monetization f the deficit also affects nflation at asecond ag. These dynamics re consistent with nstitutionaland general knowledge of the economy. The conditionalmodel of inflation s constant over the sample period, eventhough several significant tructural reaks occurred dur-ing the period. Breaks included three devaluations, truc-tural stabilization, and economic liberalization programs.As further vidence of its specification, he new conditionalmodel of inflation encompasses he inflation equation ofMetin (1995). The major inding rom the new equation sthat budget deficits as well as real income growth and debtmonetization) ignificantly ffect inflation n Turkey.

ACKNOWLEDGMENTS

I am indebted o Neil Ericsson, David Hendry, and thereferees or helpful comments. Ebru Voyvoda has providedvaluable research assistance.

APPENDIX: DATA

This appendix describes the data, lists the definitionsused, and gives their units and sources. The sample periodis 1950-1987.

G, T: The budget expenditure G) and the revenue T) arethe general budget expenditures nd revenues rom the bud-get and final accounts, respectively TL Billion). Ministryof Finance and Custom General Directorate of Account-ing, Statistical Year Book of Turkey 1990, State Instituteof Statistics Prime Ministry Republic of Turkey, Table No.367, page 471.G - T: The general budget deficit s the general budget ex-penditure minus the general budget revenue-that is, theprimary deficit, which excludes nterest payments TL Bil-lion). The budget deficit does not include he SEE's deficit.Because reliable tatistics about SEE's deficits are availableonly after the second half of the 1970s, the general budgetdeficit s therefore used as a proxy for the total deficit.


11/12


P: Price level is the CPI. The base year is 1980 (IMF In-ternational Financial Statistics, everal ssues).Y: Y is nominal GNP, divided by the GNP deflator TLBillion). Nominal GNP is obtained rom IMF InternationalFinancial Statistics, everal ssues.H: H is base money. The components of base money arecurrency n circulation, ault cash, legal reserves, and Cen-tral Bank sight deposits TL Billion). Reserve money s ob-tained from the database f the Central Bank of Turkey.

[Received June 1995. Revised April 1998.]

REFERENCES

Ahking, F. W., and Miller, S. M. (1985), "The Relationship etween Gov-ernment Deficits, Money Growth and Inflation," ournal of Macroeco-nomics, 7, 447-467.

Aktan, R. (1964), "Analysis nd Assessment f the Economic Effects, Pub-lic Law 480," The First Programme of Turkey, The State Planning Or-ganization Publication, Ankara.

Anand, R., and van Wijnbergen, . (1989), "Inflation nd he Financing fGovernment xpenditure: n Introductory nalysis With an Applica-tion to

Turkey,"The World Bank Economic Review, 3, 17-38.

Bhalla, S. S. (1981), "The Transmission f Inflation nto DevelopingEconomies," in World Inflation and the Developing Countries, eds.W. R. Cline and Associates, Washington, DC: Brookings Institution,pp. 52-101.

Buiter, W. H., and Patel, U. R. (1992), "Debt, Deficits, and Inflation: AnApplication o the Public Finances of India," ournal of Public Eco-nomics, 47, 171-205.

Burdekin, R. C. K., and Wohar, . M. (1990), "Deficit Monetisation, utputand Inflation n the United States, 1923-1982," Journal of EconomicStudies, 17, 50-63.

Chong, Y. Y., and Hendry, D. F. (1986), "Econometric valuation f LinearMacro-Economic Models," Review of Economic Studies, 53, 671-690.

Choudhary, .A. S., and Parai, A. K. (1991), "Budget Deficit and nflation:The Peruvian Experience," pplied Economics, 3, 1117-1121.

Chow, G. C. (1960), "Tests of Equality Between Sets of Coefficients nTwo Linear Regressions," Econometrica, 28, 591-605.

Cox, D. R. (1961), "Tests f Separate amilies f Hypotheses," n Proceed-ings of the Fourth Berkeley Symposium on Mathematical Statistics andProbability Vol. 1), ed. J. Neyman, Berkeley: University f CaliforniaPress, pp. 105-123.

Dickey, D. A., and Fuller, W. A. (1981), "Likelihood Ratio Statistics orAutoregressive ime Series With a Unit Root," Econometrica, 9, 1057-1072.

Dogas, D. (1992), "Market ower n a Non-monetarist nflation Model orGreece," Applied Economics, 24, 367-378.

Doornik, J. A., and Hendry, D. F. (1994), PcGive Professional 8.0: An Inter-active Econometric Modelling System, London, International ThompsonPublishing.

Dornbusch, R., and Fisher, S. (1981), "Budget Deficits and Inflation," nDevelopment in an Inflationary World, eds. M. J. Flanders and A. Razin,

New York: Academic Press.Dwyer, G. P. (1982), "Inflation nd Government eficits," Economic n-

quiry, 20, 315-329.Engle, R. F. (1982), "Autoregressive onditional Heteroscedasticity With

Estimates f the Variance f United Kingdom nflations," conometrica,50, 987-1007.

Engle, R. F., Hendry, D. F., and Richard, .-F. 1983), "Exogeneity," cono-metrica, 51, 277-304.

Ericsson, N. R. (1983), "Asymptotic roperties f Instrumental ariablesStatistics for Testing Non-nested Hypotheses," Review of EconomicStudies, 50, 287-304.

Fry, M. J. (1972), Finance and Development Planning in Turkey, Leiden:E. J. Brill.- (1980), "Money, nterest, nflation nd Growth n Turkey," ournal

of Monetary Economics, 6, 535-545.Hamburger, M. J., and Zwick, B. (1981), "Deficits, Money and Inflation,"

Journal of Monetary Economics, 7, 141-150.

Hein, S. E. (1983), "Discussion," in The Economic Consequences of Gov-ernment Deficits, ed. L. H. Mayer, Boston: Kluwer Nijhoff, pp. 75-85.

Hendry, D. F. (1983), "Comment," conometric eviews, , 111-114.- (1988), "The Encompassing mplications f Feedback Versus Feed-forward Mechanism n Econometrics," xford Economic Papers, 40,132-149.

Hendry, D. F., and Richard, .-F. 1982), "On he Formulation f EmpiricalModels n Dynamic Econometrics," ournal f Econometrics, 0, 3-33.

- 0(1989), Recent Developments n the Theory of Encompassing,"in Contributions to Operations Research and Economics: The TwentiethAnniversary f CORE, ds. B. Cornet and H. Tulkens, Cambridge, MA:MIT Press, pp. 393-440.

Ho, L. S. (1990), "Government eficit Financing nd Stabilisation," our-nal of Economic Studies, 17, 34-44.

Hondroyiannis, ., and Papapetrou, . (1994), "Cointegration, ausalityand Government udget-Inflation elationship n Greece," pplied Eco-nomic Letters, 1, 204-206.

Jarque, C. M., and Bera, A. K. (1980), "Efficient Tests for Normality,Homoscedasticity nd Serial Independence f Regression Residuals,"Economic Letters, 6, 255-259.

Johansen, . (1988), "Statistical nalysis of Cointegration ectors," our-nal of Economic Dynamics and Control, 12, 231-254.

- (1991), "Estimation nd Hypothesis Testing of Cointegration ec-

tors in Gaussian Vector Autoregressive Models," Econometrica, 9,1551-1580.-0(1992a), "Cointegration n Partial Systems and the Efficiency of

Single Equation Analysis," Journal of Econometrics, 52, 389-402.- (1992b), "Testing Weak Exogeneity nd he Order f Cointegration

in UK Money Demand Data," Journal of Policy Modelling, 14, 313-334.

Johansen, ., and Juselius, K. (1990), "Maximum ikelihood Estimationand Inference n Cointegration-With Applications o the Demand orMoney," Oxford Bulletin of Economics and Statistics, 52, 169-210.

Juselius, K. (1992), "Domestic nd Foreign Effects on Prices n an OpenEconomy: The Case of Denmark," Journal of Policy Modelling, 14, 401-428.

King, R. G., and Plosser, C. I. (1985), "Money Deficits and Inflation,"in Understanding Monetary Regimes (Carnegie-Rochester ConferenceSeries on Public Policy, 22), pp. 147-196.

Krueger, A. 0. (1974), Foreign Trade Regimes and Economic Development:Turkey, ew York: National Bureau of Economic Research.- (1995), "Partial Adjustment nd Growth n the 1980s in Turkey,"

in Reform, Recovery, and Growth, eds. R. Dornbush and S. Edward,Chicago: University f Chicago Press, pp. 33-368.

Langdana, F. K. (1990), Sustaining Budget Deficits in Open Economies,London.

MacKinnon, . G. (1991), "Critical Values for Cointegration ests," nLong-run Economic Relationship: Readings in Cointegration, eds. R. F.Engle and C. W. J. Granger, Oxford, U.K.: Oxford University Press, pp.267-276.

Metin, K. (1995), "An Integrated Analysis of Turkish nflation," OxfordBulletin of Economics and Statistics, 57, 513-533.

Mizon, G. E., and Richard, .-F. 1986), "The Encompassing rinciple ndits Application o Testing Non-nested Hypotheses," conometrica, 4,657-678.

Okyar, O. (1965), "The Concept of Etatism," conomic ournal, 75, 99-111.

Onis, Z., and Riedel, J. (1993), Economic Crisis and Long-Term Growthin Turkey The Comparative Macro Economic Series), Washington, C:The World Bank.

Osterwald-Lenum, . (1992), "A Note With Quantiles of the Asymp-totic Distribution f the Maximum Likelihood Cointegration ank TestStatistics," Oxford Bulletin of Economics and Statistics, 54, 461-472.

Phelps, E. (1973), "Inflation nd the Theory of Public Finance," wedishJournal of Economics, 75, 67-87.

Protopapadakis, . A., and Siegel, J. J. (1987), "Are Money Growth andInflation Related o Government eficits? Evidence From Ten Industri-alized Economies," International Economic Journal, 3, 79-96.

Reimers, H.-E. (1992), "Comparisons f Tests or Multivariate ointegra-tion," Statistical Papers, 33, 335-359.


12/12


Rodrik, D. (1990), "Premature iberalization, ncomplete Stabilization:The Ozal Decade in Turkey," n Lessons of Economic Stabilization andits Aftermath, ds. M. Bruno, F. Stanley, E. Helpman, nd N. Livitian,Cambridge, MA: MIT Press, pp. 323-353.

Sargent, T., and Wallace, N. (1981), "Some Unpleasant MonetaristArithmetic," Quarterly Review, Federal Reserve Bank of Minneapolis,5, 1-18.

Scarth, W. M. (1987), "Can Economic Growth Make Monetarist Arith-

metic Pleasant?" Southern Economic Journal, 53, 1028-1036.

Siddiqui, A. (1989), "The Causal Relation Between Money and Inflationin a Developing Economy," International Economic Journal, 3, 79-96.

Sowa, N. K. (1994), "Fiscal Deficits, Output Growth nd Inflation Targetsin Ghana," World Development, 22, 1105-1117.

White, H. (1980), "A Heteroskedasticity-consistent ovariance Matrix Es-timator and a Direct Test for Heteroskedasticity," conometrica, 8,817-838.

Documents

Relationship Between Inflation and Bd