1
Child Welfare and Living Standards in
Australian Households:
Some New Evidence+
Ma. Rebecca Valenzuela
Department of Economics Monash University
Abstract The measurement of the costs of children is an immensely significant and important exercise in a whole range of economic and social policy areas. In the economic literature, a conventional approach to estimating these costs is through the analysis of micro unit expenditure data within the context of a utility framework. This approach yields child cost estimates (otherwise known as equivalence scales) that allow one to make direct comparisons between households of different sizes and composition. Observed differences in the scale values across households and over time bears important implications for the welfare of children in alternative economic and social settings. This study employs the equivalence scale approach to update previous estimates of costs of children in Australia. A new methodology is applied to the 1984, 1988-89, 1993-94 and 1998-99 Australian Household Expenditure Survey to calculate equivalence scales and examine changes in the spending patterns of families over time. From these results, the paper also draws out important policy implications for children welfare and living standards. Among other things, it is shown that the advent of children results in a substantial reallocation of expenditures towards “non-adult” goods, and that families need to increase their income by some 20 percent if the pre-children living standards are to be maintained. Further, the results show that the estimated scales are stable across a wide range of income levels; however, the ratios are shown to decline over time indicating a possible decline in children’s general welfare levels over the years. +Note: Preliminary draft. Please do not quote without permission.
2
1. Introduction
The question of “How much does it cost to raise a child in Australia today?” has been the
subject of social and economic research for many years now. For families with children and
for prospective parents, a realistic idea of the costs involved in child-rearing would be a very
useful tool for planning and budgeting purposes. Interest in this issue is equally important for
many organizations and the government. Child welfare agencies greatly benefit from these
information, in the same way that family courts consider them as a critical element in resolving
claims for child support. Likewise, the federal government considers these costs as essential
elements for such critical matters as assessing the progressiveness and effectiveness of tax and
social security systems, analyzing the poverty and income distribution issues, and comparing
living standards issues among the citizenry. It is necessary and important for the government to
examine the nature and level of these costs as they provide valuable insights into the welfare
and well-being of children in our society.
For purposes of planning and policy, analyzing costs of children and the associated welfare
concerns needs to be taken in a longer-term context. Have the costs of raising children
increased over time? Does having children exert more financial pressure on families today than
they did in the past? If so, by how much more? What are the coping mechanisms of families
today in meeting the costs of children, and how do they compare with those in earlier periods?
Have government policies been able to successfully assist families during periods when the
financial demands of child-rearing is high? These are all very important questions that lie at the
heart of economic and social governance. Unfortunately, research efforts to undertake
intertemporal analysis of the household welfare and living standards, particularly in regards to
how any changes in welfare levels affect the well-being of children in Australia, have been
virtually limited, if not conspicuously non-existent. Much of this is due to data limitations
and/or methodological constraints. Recent advances in data accessibility, and econometric and
computing technology have driven international researches to investigate welfare issues in an
intertemporal setting. These same research developments provide the impetus for this study to
investigate the same for the case of Australia.
3
In this paper, we take advantage of the rich information from the ABS’s expenditure survey
data sets, and of the recent advances in econometric and computing technology in an attempt
to fill this gap in the economic and social policy research. This study employs the equivalence
scale approach to update previous estimates of costs of children in Australia and examine
changes in the spending patterns of families over time. Results from this study bears many
important implications for policy and will go a long way in stepping up research efforts
towards a more wholistic look at household and child welfare issues in Australia.
The paper is structured as follows: Section 1 introduces the topic and discusses its aims and
significance. Section 2 describes the two most common ways of estimating the cost of children
and reviews some relevant recent work. Section 3 describes the demand system model used
and outlines the estimation methodology. Section 4 discusses the data and results from the
empirical application and a final section concludes.
2. Alternative Approaches and Previous Work
In the economic and social science literature, there are two competing approaches to the
estimation of child costs: a) the budget standards approach, and, b) the expenditure behaviour
approach. The budget standards or “baskets-of-goods” approach is characterised by the
creation of an ‘ideal’ basket of goods and services for certain model families – ideal according
to some expert opinion. Nutritionists and medical experts are among those who are relied on to
provide some minimum diet requirements for particular family types. Other experts who may
be involved in the process are educators, psychologists and similar social scientists. From the
prescribed basket of goods, the cost of a child is derived from that component of the budget
that is attributable to children. As such, budget standard estimates answer the question “What
should be the cost of a child?”.
An alternative procedure to estimate child costs is through the analysis of household
expenditures. Cost of child estimates based on this approach are based on information
gathered from asking thousands of households how much they actually spend on specific
4
commodity items and inferring child costs from this large set of quantitative information. This
approach is thus also popularly known as the “large-scale survey” method. The underlying
assumption here is that the standard of living of a household is largely determined by the
household’s expenditure behaviour and its demographic characteristics. The details of the
statistical linkages involved are captured in what are called “demand equations” which are then
estimated using information from the expenditure records. The resulting parameters are then
used to estimate the cost of children (also know in the literature as equivalence scales) for
different family types. Such estimates show how much parents actually spend on their
children, even though the amount spent might be considered inadequate or excessive by some
other standards. As such they answer the question, “What is the cost of a child?”
Which one to use – depends on what the purpose of the child estimation exercise. The basket
of goods approach is more appropriate for determining a desirable level of expenditure for
children and can guide parents/policymakers on some minimum provision as in some budget
for foster care & other residential arrangements. In contrast, the budget demand approach is
provides useful information on the actual level of expenditures accruing to children at a
particular point in time, and could be used to assess current welfare levels, determining actual
capacities of families to provide for children, and can also be used to identify areas/situations
where the government can intervene policy wise.
3 Conceptual Framework & Estimation Methodology
The main interest in this study is the estimation of cost of children (hereon referred to as
equivalence scales) through the econometric analysis of household expenditure data. A
framework of analysis is first defined. Let the economic environment consists of H households
and n private goods with per-unit prices p=(p1, p2, …pn). It is implied that all households face
the same prices. Household h with demographic characteristics δδ h has total expenditure xh and
preferences represented by a utility function (( ))hhh qUu δδ== , assumed continuous, increasing
and quasi-concave in consumption (( ))hnhh qqq ,...,1== . Here, (( ))hhh qUu δδ== , is assumed to be
the household's utility level as measured relative to a demographic reference unit.
5
Household maximisation of utility subject to a total expenditure constraint implies Marshallian
demand functions for each good, expressed as functions of p, xh, and δδ h The achieved utility
level uh can be summarised by the indirect utility function
(( ))hhh xpVu δδ== ,
which is non-decreasing in p, increasing in xh and homogenous of degree zero in p and xh
Moreover, using standard duality results, demands can also be interpreted as choices which
minimise the expenditure required to achieve some reference utility level uh This is
equivalent to saying that preferences for household h can also be represented using a consumer
cost function
(( ))hh puC δδ,,
which is increasing in uh and p, and linearly homogenous and concave in p. With this model, an equivalence scale sh is defined using the consumer cost function as
(( ))(( ))rh
hhh puC
puCs
δδδδ
==,,
,,
where u=ur is the utility level for the reference household type (with characteristics c). An
equivalence scale thus shows the relative cost of maintaining household h with composition
δδ h at the same utility level u=ur enjoyed by the reference household r with composition δδ r.
Deaton and Muellbauer (1980) offer an interesting parallelism between equivalence scales and
price indices by saying that equivalence scales are to welfare comparisons between households
of different characteristics what cost-of-living indices are to welfare comparisons for a given
household facing different prices.
6
There are two general types of econometric equivalence scales. Some equivalence scales are
calculated based on an observable variable selected as a ``proxy'' for household welfare or
utility.Another type is derived from the direct specification of a utility function. The various
models under this general classification are comprehensively reviewed in Valenzuela (1997).
The equivalence scales or estimates of child costs in this paper were estimated using the
‘complete demand systems’ based scale model. Estimation of this type of scales commences
with the specification of preferences which are done using cost or indirect utility functions
rather than direct utility functions. The demand functions are then estimated using household
budget data sets followed by the computation of the scales using the estimated parameters. One
specific procedure of estimating scales in this context is the Price Scaling (PS) technique1. It
replaces the original cost function cR of the reference household R (a childless adult couple)
by:
(( )) (( )) (( ))nR ppucPmPucx ,.....,,,,, 10 ηη==ηη≡≡ (1)
where x is aggregate household expenditure, P is the price vector, η is the vector containing the
age/sex distribution of children in the household, u is the utility variable and m0(.) is the
general equivalence scale. In specifying m0 directly in terms of prices and household
composition, PS does not require the complex estimation of the commodity specific mis which
characterises the Barten model.
Applying the Price Scaling demographic technique requires the prior specification of the cost
function, cR, of the reference household. We choose the following functional form2.
1 See Ranjan Ray, “Measuring the Cost of Children: An Alternative Approach,” Journal of Public Economics 22 (1983): 89-102. 2 This was introduced by James Banks, Rinchard Blundell, and Arthur Lewbel, “Quadratic Engel Curves and Consumer Demand”, The Review of Economic and Statistics, VolLXXIX No. 4 Nov 1997, pp527-39.
7
(( )) (( )) (( ))(( ))Puc
PubPaPu
−−++==
1,log Rc (2)
where a (.) is homogeneous of degree one, and b(.), c(.) are homogeneous of degree zero in
prices, P. Choice of appropriate functional forms for a(.), b(.), c(.) yields, in budget share form
wi, the following rank three Price Scaled demographic demand system3.
[[ ]] [[ ]] ∑∑∏∏====
ββ−−λλ ββ++λλ++ββ++αα==n
jij
n
kkiiii
kkpw1
2
1
logloglog jRR p x x (3)
i = 1,.....,n
where xR is the ‘per equivalent’ real expenditure of the household, i.e. its real expenditure
deflated by the equivalence scale, (( )) ,2
1 10 ds
s
D
da nnm ∑∑ ∑∑
== ==
θθ++==ηη ds where na is the number of
adults, θds is the scale parameter corresponding to a child in age group d and sex group s, and
nds is the number of children in that age/sex group. Since prices are fixed in a single cross
section, we can without loss of generality choose pi = 1. The estimating form for eqn. (3)
becomes:
2
00
loglog
λλ++
ββ++αα== mx
mxw iiii (4)
i = 1,.....n
where ∑∑ ∑∑ ∑∑ ==αα==λλ==ββ 1,0 iii .
Note that eqn. (4), which is a rank 3 demand system, specialises to the rank 2 form of
Working-Leser if λi = 0. Note, further, that the general equivalence scale m0 can be directly
calculated on budget data from the demographic parameter (θd) estimates of eqn. (4).
8
4 Data and Results
The data used in this study are derived from the 1984, 1988-89, 1993-94 and 1998-99
Household Expenditure Survey (HES) conducted by the Australian Bureau of Statistics (ABS).
These are the latest body of data conducted of a series of surveys designed to obtain details of
expenditure, income and a wide range of demographic characteristics of Australian private
households on a nationwide basis. The public-use tapes contain a total of 4492 (1984 HES),
7225 (1988-89 HES), 8390 (1993-94 HES) and 6893 (1998-99 HES) households representing
more than 5 million Australian households from all over the country for each year the surveys
were conducted.
The estimation procedure used for the estimation of the cost of children here required the use
of information from households composed of related persons with one or two adults and at
most three children only. This resulted in eight household types. Adults are all persons aged
17 or older and children refer to all those aged 16 or younger. Households not belonging to any
of these types are excluded. Information from some 300 households from each year of data
was also discarded because of reported negative expenditures on certain items4. These
observations were not consistent with the economic model set up for this purpose.
The number and characteristics of household types from the three data set are given in Table 1.
Of the 5337 households considered from the 1998-99 HES, 36 percent were of the type (2,0)
where the first number in the bracket refers to the number of adults, and the second number
refers to the number of children. Further, 27 percent were of the type (1,0). Effectively, 62
percent of the total households in the sample were households without children. These
children-free households were mostly headed by persons in the older age groups (average age
of household head is 48 years for couples and 53 years for singles) and are inferred to have
3 See Geoffrey Lancaster and Ranjan Ray, “Comparison of Alternative Models of Household Equivalence Scales; The Australian Evidence on Unit Record Data”, Economic Record, 74 (1998): pp. 1-14 for details of derivation.
9
children who are already moved out and are financially independent. A number of then are
retired couples or individuals. In contrast, the 37 percent of sample households that have
children had household heads aged between 25 and 40.
The table further shows that two-adult households have higher weekly incomes compared to
one-adult households, and households with children have higher incomes than those without.
Large variances associated with these averages indicate that the absolute differences in the
reported incomes are not significant among households with two adults and among households
with one adult, whether they have children or not. As may be expected, however, there are
significant differences in income levels between one-adult and two-adult households. Reported
levels of total expenditures were, on average, consistently higher than reported total income but
the large variances indicate no significant differences in the values.
The households covered in the 1984 and 1988-89 HES were very similarly distributed, though
income and expenditure levels were progressively lower in these earlier years. In 1984, single
adult households with one child were younger and poorer compared to those in similar
situations in the later years.
The analysis in this paper is restricted to goods and services that are for private consumption.
There are11 broad commodity groups as follows:
1. Housing includes expenses incurred for the payment of rent, mortgage, property rates,
house and contents insurance as well as housing repairs and maintenance.
2. Fuel and power includes all expenses towards electricity, gas and other fuels.
3. Food includes all expenses towards bakery products, flour and other cereals, meat and fish,
dairy products, fruits and vegetables, miscellaneous food (jam, jellies, coffee, tea), non
alcoholic beverages, meals out and take away food.
4. Alcohol and tobacco refers to all expenses towards the purchase of cigarettes and all types
of alcoholic beverages.
4 For example, of the households which reported negative expenditures in the 1988-89 dataset, 72 percent of the negative expenditures were on transport while 27 percent on recreation and entertainment.
10
5. Clothing and footwear includes all expenses towards the purchase of clothing and
footwear for men, women and children, clothing accessories (e.g. ties, gloves,
handkerchiefs) as well as clothing and footwear services (e.g. drycleaning and shoe
repairs).
6. Household furnishings and equipment includes all expenses towards furniture and floor
coverings, blankets and rugs, household linen and furnishings, household appliances,
glassware, tableware, household utensils and cleaning agents. This category also includes
expenditures incurred for the operation of the household such as gardening services,
housekeeping, childcare and the repair and maintenance of household durables.
7. Medical and health care covers items such as accident and health insurance premiums,
practitioner’s fees, prescriptions, medicines, pharmaceutical products, hospital and other
health charges.
8. Transport refers to all expenses made for the purchase of motor vehicles, petrol and fuels,
vehicle registration and insurance, vehicle servicing and repairs, driver’s licenses, driving
lessons, subscriptions to motor organisations, vehicle hire, as well as public transport fees.
9. Recreation and entertainment includes expenses for the purchase of television and other
audiovisual equipment, books, newspapers and other printed materials, recreational
equipment (cameras, musical instrument, toys), gambling, entertainment and recreational
services. Holiday expenses as well as those incurred for animal pets are also included in
this category.
10. Personal care pertains to expenses towards toiletries, cosmetics, hair dressing and beauty
services.
11. Others includes expenses for miscellaneous goods (watches, jewellery, stationery), interest
payments on selected credit services, education fees, life insurance and other miscellaneous
services.
11
Chart 1. Distribution of Expenses for a Typical (2,2) Family, 1998-99 HES.
Housing31%
Others7%
Recreation & Entertainment
10%
Alcohol & Tobacco
3%Food19%
Fuel3%
Household furnishing and
Equipment4%
Medical and Health
4%
Personal Care2%
Transport13%
Clothing and Footwear
4%
Detailed expenditure information from the ABS household surveys reveal interesting spending
patterns of Australian families. For the typical household composed of two adults and two
children, nearly 50% of the household budget is taken up by Food and Housing. If we
through in expenditures on Transportation, these three items make up 61 percent of the total
expenditures for a typical (2,2) family (See Chart 1). Recreation & Entertainment (including
holidays and travel) take up 10% of the budget, while Household Goods & Services, Clothing
and Footwear, and Medical and Health Care take up 4 per cent each. Household has a wide
variety of miscellaneous expenditures (lumped in Others at 7%), and this is significant amount
as it includes all the education/school tuition fees in this category.
Some variations across household types are worth noting. The budget share of Housing for two
adult, zero children families is substantially lower than the rest of the household types (24
percent share as against over 30 percent share for all other household types). To a large extent,
this is driven by the large numbers of older couples who would fully own the house they live
in, and so do not register a spending much on rent or mortgage. It is also noted that budget
share of Housing for single parent families is somewhat than for two-parent families, with
12
single parent with 1 child paying an average of 35 percent of their disposable income to rent or
mortgage expenses. Budget share for food tends to increase with the number of children in the
family, regardless the number of adults, while the budget share of Alcohol & Tobacco tends to
decrease with the number of children present, as does the budget share for Medical and Health
care. Recreation and Entertainment expenses appears to have a larger budget share for
childless households. On the other hand, the budget share of Transport is only about 12% for
single parent families compared to the 14% share of Transport in two-parent families. The
budget share for Transport is level for single adult families, while it is seen to decline slightly
with increasing numbers of children in two-adult families.
Tables 2a-2d present the estimates of commodity-specific scales. A two-adult household with
no children is chosen to be the reference household for which the equivalence scale sih is set to
1.0. The Table 2d, scale value of 1.41 for Housing of a (2,1) household means that this type of
household needs a housing budget 41 percent more than the typical 2 adult, no children
household to maintain the same standard of living as the latter. Similarly, the Fuel and power
scale of 0.71 for the (1,0) type household implies that a childless couple would require
substantially less (in per capita terms) in Fuel expenditures than a single member householder
to be on comparable standards of living – a clear demonstration of the economies of scale
advantage for multiperson households. The equivalent 1993- 94 scale values for these
commodities and household types are 1.38 and 0.68 respectively (in Table 2c) and can be
similarly interpreted. The equivalent scale values for 1988-89 are 1.49 for Housing and 0.67
for Fuel and power (in Table 2b), and for 1984 are 1.56 for Housing and 0.62 for Fuel and
Power (in Table 2a) and can be similarly interpreted.
For most expenditure items, the commodity scales increase with the increase with the increase
in the number of children. These increases are observed to occur at a decreasing rate indicating
economies of scale for additional children. After the first child, there exists strong economies
of scale for additional children, particularly for expenditures towards Housing, Fuel & Power,
Household furnishings and Transport. For two-adult households in 1998-99, the decline in the
Housing scale after the first child is unexpected. One interpretation is that this could reflect the
growing practice of many young families to get children to share a room, rather than provide
13
for a room per child. The scales in the three previous years clearly show that economies of
scale in housing were achieved with every additional child.
There is some indication that budget shares for Alcohol and Tobacco tend to increase with the
number children, particularly for the earlier years. Also, the scales for Medical and health care,
Recreation and entertainment, and Others exhibit no defined trend for one-adult households. A
more thorough investigation of expenditure patterns of households may be required for us to
provide definitive explanations for such deviations but one possibility is that the presence of
children in the household tends to influence expenses away from ‘adult goods’ under which
alcohol, tobacco and other miscellaneous goods are classified.
Has there been a significant change in the scale relativities over time? Information from
Tables 3 and 4, which compare some estimates for the four survey years provide some
answers. In Table 3, commodity-specific scale estimates are presented in such a way that
comparisons over time is facilitated. There are noticeable changes in the scales over time. For
two-adult households with children, the scales for Housing, Fuel & power, Alcohol & tobacco,
Medical & health care, Transport and Recreation and Entertainment have decreased
consistently in the last between 1984 and 1994, only to go back up again the last few years.
Such decrease in the scale values indicates that the cost requirements for the additional family
member (to maintain standard of living) are not as much as it used to be. In other words,
between 1984 and 1994, the cost of having children was less than before! The 1998-1999
scales however show a reversal of trend which then indicates that costs of having children have
increased since. The scale for Food remained more or less the same over time, implying then
that the implied affordability of having children in more recent years. This analysis however
holds only for a two-parent family. For single parent households, the typical trend of the scale
values for most commodities is a decline between 1984 and 1994, and then an increase
thereafter.
It is further noted here that the largest differences in the scale estimates occurred in the one-
adult, three-children household groups. Since the number of households in this group is
14
relatively small, and the standard errors of the estimated scales are relatively high, these
differences may reflect sampling error.
The general scales are presented in Table 4. Because these scales depend on some reference
income xr, they are computed for three income levels5. First noted is that the estimated general
scales are stable over different reference income levels. Also, for two-adult households, the
1993-94 scales are less than both the 1984 and 1988-89 scales. The 1988-89 and 1993-94
estimates are generally quite similar. Further, the conclusion by Binh and Whiteford (1990)
that “there is strong evidence of economies of scale in the second child but adding the third
child increased these households’ needs considerably” no longer holds for the later data sets.
The 1993-94 and 1998-99 scales, in particular, suggest equal cost requirements for the 2nd and
3rd children.
6 Conclusion
This paper introduces an improved maximum likelihood estimation procedure for estimating
equivalences scales using the QUAIDS under the demand system approach. Cost estimates of
children over the last 15 years were presented, in the form of both commodity specific and
general scales, and updates estimates for Australia.
In general, the results show that to be able to maintain the living standards of the household
before the addition of children, the family budget in 1998-99 will have to be increased by about
26 per cent for the first child for two parent families, and by about 23 percent for single parent
families. The additional budget requirements for the 2nd and 3rd children will still be positive
but not as much as that of the first child. Initial estimates show that larger adjustments are
required by two- parent families compared to single-parent families, where the opposite was
true for previous years.
The commodity specific scales show that the budget requirements vary across the various
commodity groups. For the first child, there is a significantly high budget requirement (41 per
15
cent) to meet Housing needs of the child while in terms of Food, the budget needs to be
adjusted by around 25 percent (on average). The results show that single parent households
needs more assistance to meet Housing adjustment needs for than two-parent families. After
the 1st child, large gains in economies of scale were observed for Housing, Fuel & Power,
Household Goods and Transport items. Also, the findings seem to suggest that the presence of
children induces adult family members to consume less of the “adult goods” such as Alcohol &
Tobacco.
References: Australian Bureau of Statistics, Information Paper on Household Expenditure Survey of
Australia, various years, AGPS Canberra.
Banks, J, R. Blundell and A Lewbel, Quadratice Engle Curves and Consumer Demand, Review of Economics and Statistics, Vol LXXIX Nov 1997 No.4 pp527-539.
Binh, T.N. and P. Whiteford (1990), “Household Equivalence Scales: New Australian Estimates from the 1984 Household Expenditure Survey”, The Economic Record, 66, 221-234.
Blundell, R. and A. Lewbel (1991), “Demographic Variables in Empirical Microeconomics: The Information Content of Equivalence Scales”, Journal of Econometrics, 50: 49-68.
Bradbury, B. (1994), “Measuring the Cost of Children”, Australian Economic Papers, June, 120-138.
Browning, M. (1992), “Children and Household Economic Behaviour”, Journal of Economic Literature, 30, 1434-1475.
Deaton, A. and J. Muellbauer (1986), “On Measuring Child Costs: With Applications to Poor Countries”, Journal of Political Economy, 94: 720-745.
Nelson, J. (1993), “Household Equivalence Scales: Theory vs Policy?”, Journal of Labour Economics, 11(3): 471-493.
Griffiths, W.E. and Valenzuela, M.R. (1998), “Missing Data from Infrequency of Purchase: Bayesian Estimation of a Linear Expenditure System”, Advances in Econometrics, vol.13, pp 47-74.
Griffiths, W.E. and M.R.Valenzuela (2001), “A New Procedure for Estimating Cost of Children: A New Procedure Applied to Australian Data”, Journal of Quantitative Economics.
5 Levels were made comparable to those used by Binh and Whiteford (1990) to facilitate comparison.
16
Griffiths and Valenzuela (2002) “A Gibbs Sampler for Estimating SURs with Cross Set Restrictions on the Parameters”, contributed paper at the Joint Statistical Meetings 2002, Aug 11-15, 2002, New York.
Kakwani, N. (1980), Income Inequality and Poverty: Methods of Estimation and Policy Applications, New York, Oxford University Press.
Lancaster, G. and R. Ray (1998),“Comparison of Alternative Models of Household Equivalence Scales: The Australian Evidence on Unit Record Data”, The Economic Record, 74, (1998), 1-14.
Lancaster, G. and Ray, R. and Valenzuela, M.R, (1999), “A Cross-Country Study of Household Poverty and Inequality on Unit Record Household Budget Data”, Economic Development and Cultural Change (July 1999).
Pollak, R.A. and T.J. Wales (1979), “Welfare Comparisons and Equivalence Scales”, American Economic Review, 68: 348-359.
Pollak, R.A. and T.J. Wales (1992), Demand System Specification and Estimation, New York and Oxford: Oxford University Press.
SHAZAM User’s Reference Manual Version 9 (2001)McGraw Hill, Canada.
Valenzuela, M.R. (2002) “The Cost of Children and Living Standards in Australia”, in Australian Economy and Society: Education, Work and Welfare, J Kelly and M Evans (eds), in press.
Valenzuela, M.R (1999) “Cost of Children in Australia”, Family Matters, Winter 1999, Issue
No. 53. pp. 71-76. Reprinted in Valenzuela, M.R. (2000) in “Guide to the cost of children in Australia”. Australian Institute of Family Studies, March.
Valenzuela, M.R. (1997), “Alternative Approaches to the Estimation of Household Equivalence Scales, Unpublished PhD dissertation.
Valenzuela, M.R. (1996), “Engel Scales for Australia, the Philippines and Thailand: A Comparative Analysis”, Australian Economic Review, 114: 189-198.
17
Table 1. Sample Characteristics
1984 HES Household Type (no. of adults, no. of children) (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Sample Size 777 79 82 38 1272 406 643 275
Age of HH Head 53.33 26.16 30.80 30.58 49.41 32.07 33.16 36.48
Average Weekly 234.76 195.39 242.68 264.13 435.60 513.66 497.32 553.55 HH Income (195.11) (106.82) (141.89) (143.89) (293.19) (289.94) (254.02) (352.12)
Average Weekly 241.71 246.26 297.71 330.44 445.11 557.37 556.30 615.21 HH Expenditure (200.67) (163.98) (248.58) (244.02) (307.26) (292.10) (304.67) (460.34)
1988-89 HES Household Type (no. of adults, no. of children)
(1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Sample Size 1372 132 103 42 2074 532 889 388
Age of HH Head 52.68 33.52 30.28 28.76 48.38 32.68 33.84 35.12
Average Weekly 306.73 274.54 315.38 313.64 595.93 697.00 767.83 720.57 HH Income (246.53) (172.51) (166.09) (159.76) (417.50) (579.69) (493.50) (378.47)
Average Weekly 255.04 281.20 315.15 310.44 461.62 555.94 603.90 623.19 HH Expenditure (194.60) (162.12) (153.57) (142.05) (285.24) (285.82) (348.75) (321.88)
Note: Standard errors are in parentheses
18
Table 1. Sample Characteristics (continuation)
1993-94 HES Household Type (no. of adults, no. of children) (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Sample Size 1702 192 149 60 2509 690 845 425
Age of HH Head 53.03 34.08 33.69 30.58 48.12 33.1 35.46 35.04
Average Weekly 281.89 339.99 377.83 346.83 547.18 612.54 634.78 632.85 HH Income (227.89) (162.32) (151.71) (100.60) (345.90) (333.63) (349.07) (365.70)
Average Weekly 319.23 404.52 468.86 440.3 593.52 695.37 739.92 762.18 HH Expenditure (232.00) (218.53) (278.42) (253.24) (347.72) (345.04) (370.39) (390.11)
1998-99 HES Household Type (no. of adults, no. of children)
(1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Sample Size 1420 174 136 57 1947 561 745 291
Age of HH Head 54.21 36.39 36.03 35.18 50.85 37.85 38.70 37.82
Average Weekly 312.90 377.39 419.39 384.98 607.37 679.92 704.61 702.46 HH Income (202.82) (144.46) (135.02) (89.53) (307.85) (296.93) (310.67) (325.47)
Average Weekly 354.35 449.02 520.43 488.73 658.81 771.86 821.31 846.02 HH Expenditure
(257.98) (243.01) (309.60) (281.60) (386.66) (383.68) (411.87) (433.80)
Note: Standard errors are in parentheses
19
Proportion of Expenditures for Sample Households, 1998-99 Survey
Commodity Groups (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Housing 0.31 0.35 0.33 0.31 0.24 0.30 0.30 0.30
Fuel 0.04 0.04 0.04 0.04 0.03 0.03 0.03 0.03
Food 0.17 0.17 0.21 0.23 0.18 0.17 0.18 0.19
Alcohol & Tobacco 0.04 0.04 0.03 0.03 0.04 0.04 0.03 0.03
Clothing and Footwear 0.02 0.04 0.03 0.02 0.03 0.04 0.04 0.04 Household furnishing and Equipment 0.04 0.04 0.03 0.04 0.05 0.05 0.04 0.04
Medical and Health 0.05 0.02 0.02 0.03 0.05 0.04 0.04 0.04
Transport 0.11 0.10 0.11 0.12 0.14 0.14 0.13 0.12
Recreation & Entertainment 0.10 0.09 0.09 0.07 0.11 0.09 0.10 0.11
Personal Care 0.02 0.01 0.02 0.02 0.02 0.01 0.02 0.01
Others 0.06 0.06 0.06 0.06 0.06 0.07 0.07 0.08
20
Table 2a. Estimates of Commodity-Specific Scales, Australia 1984.
Commodity Specific Scales Household Type (no. of adults, no. of children)
Commodity Type (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Housing 0.73 1.18 1.23 1.17 1.00 1.56 1.58 1.60 (0.04) (0.08) (0.09) (0.19) (0.00) (0.09) (0.08) (0.10)
Fuel & Power 0.62 1.06 1.08 1.28 1.00 1.31 1.42 1.59 (0.02) (0.05) (0.06) (0.11) (0.00) (0.04) (0.04) (0.05)
Food 0.51 0.75 0.95 1.15 1.00 1.23 1.45 1.65 (0.01) (0.04) (0.05) (0.10) (0.00) (0.03) (0.03) (0.04)
Alcohol & Tobacco 0.50 0.67 0.79 0.67 1.00 1.23 1.18 1.28 (0.03) (0.05) (0.06) (0.07) (0.00) (0.07) (0.06) (0.05)
Clothing & Footwear 0.44 0.87 1.26 1.21 1.00 1.28 1.43 1.71 (0.05) (0.15) (0.15) (0.30) (0.00) (0.11) (0.11) (0.15)
Household Furnishings 0.53 0.85 0.99 1.04 1.00 1.26 1.35 1.45 & Equipment (0.04) (0.08) (0.11) (0.20) (0.00) (0.14) (0.09) (0.11)
Medical & Health Care 0.49 0.45 0.57 0.65 1.00 1.30 1.38 1.49
(0.04) (0.08) (0.11) (0.14) (0.00) (0.08) (0.06) (0.07)
Transport 0.49 0.68 0.83 1.07 1.00 1.36 1.29 1.38 (0.04) (0.07) (0.09) (0.23) (0.00) (0.08) (0.08) (0.13)
Recreation 0.53 0.70 0.79 0.80 1.00 1.08 1.28 1.36 & Entertainment (0.04) (0.09) (0.07) (0.21) (0.00) (0.10) (0.11) (0.14)
Personal Care 0.58 0.72 1.12 1.08 1.00 1.20 1.24 1.31
(0.04) (0.09) (0.15) (0.14) (0.00) (0.09) (0.08) (0.09)
Others 0.58 0.82 0.91 1.17 1.00 1.76 1.83 2.01 (0.06) (0.16) (0.11) (0.15) (0.00) (0.14) (0.19) (0.27)
Note: The estimated standard errors are in parentheses.
21
Table 2b. Estimates of Commodity-Specific Scales, Australia 1988-89.
Commodity Specific Scales
Household Type (no. of adults, no. of children) Commodity Type (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Housing 0.82 1.03 1.15 1.28 1.00 1.49 1.52 1.65
(0.04) (0.08) (0.09) (0.19) (0.00) (0.09) (0.08) (0.10)
Fuel & Power 0.67 0.92 1.06 1.11 1.00 1.21 1.34 1.44 (0.02) (0.05) (0.06) (0.11) (0.00) (0.04) (0.04) (0.05)
Food 0.53 0.73 0.94 1.06 1.00 1.24 1.42 1.58 (0.01) (0.04) (0.05) (0.10) (0.00) (0.03) (0.03) (0.04)
Alcohol & Tobacco 0.57 0.46 0.39 0.34 1.00 0.95 0.86 0.76 (0.03) (0.05) (0.06) (0.07) (0.00) (0.07) (0.06) (0.05)
Clothing & Footwear 0.53 0.91 0.92 1.40 1.00 1.28 1.40 1.64 (0.05) (0.15) (0.15) (0.30) (0.00) (0.11) (0.11) (0.15)
Household Furnishings 0.55 0.66 0.77 0.81 1.00 1.45 1.15 1.32 & Equipment (0.04) (0.08) (0.11) (0.20) (0.00) (0.14) (0.09) (0.11)
Medical & Health Care 0.54 0.47 0.68 0.51 1.00 1.26 1.28 1.31
(0.04) (0.08) (0.11) (0.14) (0.00) (0.08) (0.06) (0.07)
Transport 0.52 0.57 0.62 0.78 1.00 1.02 1.19 1.37 (0.04) (0.07) (0.09) (0.23) (0.00) (0.08) (0.08) (0.13)
Recreation 0.54 0.58 0.51 0.82 1.00 1.03 1.28 1.37 & Entertainment (0.04) (0.09) (0.07) (0.21) (0.00) (0.10) (0.11) (0.14)
Personal Care 0.54 0.78 0.97 0.73 1.00 1.19 1.29 1.17
(0.04) (0.09) (0.15) (0.14) (0.00) (0.09) (0.08) (0.09)
Others 0.57 1.02 0.89 0.81 1.00 1.39 1.79 2.08 (0.06) (0.16) (0.11) (0.15) (0.00) (0.14) (0.19) (0.27)
Note: The estimated standard errors are in parentheses.
22
Table 2c. Estimates of Commodity-Specific Scales, Australia 1993-94.
Commodity Specific Scales
Commodity Type Household Type (no. of adults, no. of children) (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Housing 0.82 1.28 1.19 1.30 1.00 1.38 1.31 1.27 (0.04) (0.10) (0.10) (0.14) (0.00)- (0.08) (0.07) (0.09)
Fuel & Power 0.68 0.94 1.12 1.28 1.00 1.23 1.34 1.36 (0.03) (0.06) (0.09) (0.16) (0.00) (0.06) (0.05) (0.06)
Food 0.52 0.74 0.96 0.99 1.00 1.22 1.44 1.62 (0.03) (0.06) (0.08) (0.11) (0.00) (0.06) (0.07) (0.09)
Alcohol & Tobacco 0.56 0.56 0.48 0.58 1.00 0.97 0.82 0.82 (0.06) (0.10) (0.10) (0.15) - (0.11) (0.09) (0.12)
Clothing & Footwear 0.47 0.79 1.14 1.11 1.00 1.11 1.56 1.62 (0.11) (0.24) (0.36) (0.48) (0.00) (0.24) (0.33) (0.45)
Household Furnishings 0.48 0.79 0.88 0.86 1.00 1.23 1.25 1.15 & Equipment (0.06) (0.14) (0.18) (0.31) (0.0) (0.17) (0.16) (0.17)
Medical & Health Care 0.50 0.52 0.66 0.51 1.00 1.10 1.27 1.23
(0.06) (0.12) (0.16) (0.22) (0.00) (0.15) (0.14) (0.16)
Transport 0.53 0.65 0.73 0.81 1.00 1.21 1.13 1.22 (0.08) (0.20) (0.17) (0.34) (0.00) (0.18) (0.17) (0.21)
Recreation 0.51 0.58 0.74 0.66 1.00 0.95 1.02 1.23 & Entertainment (0.07) (0.14) (0.18) (0.29) (0.00) (0.14) (0.14) (0.20)
Personal Care 0.51 0.71 0.73 0.54 1.00 0.96 1.15 1.19
(0.07) (0.17) (0.16) (0.18) (0.00) (0.15) (0.15) (0.19)
Others 0.55 0.76 1.51 0.90 1.00 1.30 1.55 1.92 (0.10) (0.17) (0.82) (0.36) (0.00) (0.25) (0.25) (0.35)
Note: The estimated standard errors are in parentheses.
23
Table 2d. Estimates of Commodity-Specific Scales, Australia 1998-1999.
Commodity Specific Scales Commodity Type Household Type (no. of adults, no. of children)
(1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Housing 0.84 1.31 1.21 1.30 1.00 1.41 1.31 1.31 (0.03) (0.09) (0.10) (0.12) (0.00) (0.05) (0.05) (0.06)
Fuel & Power 0.71 0.96 1.15 1.31 1.00 1.25 1.37 1.39 (0.02) (0.06) (0.09) (0.15) (0.00) (0.06) (0.05) (0.06)
Food 0.57 0.83 1.10 1.29 1.00 1.29 1.50 1.72 (0.03) (0.06) (0.08) (0.11) (0.00) (0.06) (0.07) (0.09)
Alcohol & Tobacco 0.56 0.56 0.48 0.58 1.00 0.97 0.82 0.82 (0.04) (0.08) (0.10) (0.13) (0.00) (0.11) (0.10) (0.12)
Clothing & Footwear 0.52 0.81 1.16 1.22 1.00 1.17 1.59 1.64 (0.08) (0.24) (0.33) (0.44) (0.00) (0.20) (0.33) (0.45)
Household Furnishings 0.53 0.81 0.90 1.03 1.00 1.35 1.28 1.17 & Equipment (0.05) (0.14) (0.18) (0.31) (0.00) (0.17) (0.16) (0.17)
Medical & Health Care 0.51 0.62 0.67 0.64 1.02 1.32 1.30 1.25
(0.06) (0.12) (0.15) (0.22) (0.00) (0.15) (0.14) (0.16)
Transport 0.55 0.68 0.77 0.85 1.00 1.29 1.15 1.22 (0.05) (0.16) (0.17) (0.30) (0.00) (0.14) (0.16) (0.20)
Recreation 0.52 0.59 0.75 0.67 1.02 0.97 1.04 1.25 & Entertainment (0.07) (0.14) (0.18) (0.25) (0.00) (0.12) (0.13) (0.18)
Personal Care 0.52 0.82 0.73 0.55 1.02 0.98 1.17 1.21
(0.07) (0.16) (0.16) (0.18) (0.00)) (0.15) (0.15) (0.16)
Others 0.55 0.78 1.51 0.92 1.00 1.33 1.58 1.96 (0.10) (0.14) (0.30) (0.29) (0.00) (0.18) (0.22) (0.30)
Note: The estimated standard errors are in parentheses.
24
Table 3 Estimates of Commodity-Specific Scales
Commodity-Specific Scales (Sih) Household Type (no. of adults, no. of children)
Commodity Type Year (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3) Housing 1984 0.73 1.18 1.23 1.17 1.00 1.56 1.58 1.60
1988-89 0.82 1.03 1.15 1.28 1.00 1.49 1.52 1.65 1993-94 0.82 1.28 1.19 1.30 1.00 1.38 1.31 1.27 1998-99 0.84 1.31 1.21 1.30 1.00 1.41 1.31 1.31
Fuel & Power 1984 0.62 1.06 1.08 1.28 1.00 1.31 1.42 1.59 1988-89 0.67 0.92 1.06 1.11 1.00 1.21 1.34 1.44 1993-94 0.68 0.94 1.12 1.28 1.00 1.23 1.34 1.36 1998-99 0.71 0.96 1.15 1.31 1.00 1.25 1.37 1.39
Food 1984 0.51 0.75 0.95 1.15 1.00 1.23 1.45 1.65 1988-89 0.53 0.73 0.94 1.06 1.00 1.24 1.42 1.58 1993-94 0.52 0.74 0.96 0.99 1.00 1.22 1.44 1.62 1998-99 0.57 0.83 1.10 1.29 1.00 1.29 1.50 1.72
Alcohol & Tobacco 1984 0.50 0.67 0.79 0.67 1.00 1.23 1.18 1.28 1988-89 0.57 0.46 0.39 0.34 1.00 0.95 0.86 0.76 1993-94 0.56 0.56 0.48 0.58 1.00 0.97 0.82 0.82 1998-99 0.56 0.56 0.48 0.58 1.00 0.97 0.82 0.82
Clothing & Footwear 1984 0.44 0.87 1.26 1.21 1.00 1.28 1.43 1.71 1988-89 0.53 0.91 0.92 1.40 1.00 1.28 1.40 1.64 1993-94 0.47 0.79 1.14 1.11 1.00 1.11 1.56 1.62 1998-99 0.52 0.81 1.16 1.22 1.00 1.17 1.59 1.64
Household Furnishings 1984 0.53 0.85 0.99 1.04 1.00 1.26 1.35 1.45 & Equipment 1988-89 0.55 0.66 0.77 0.81 1.00 1.45 1.15 1.32
1993-94 0.48 0.79 0.88 0.86 1.00 1.23 1.25 1.15 1998-99 0.53 0.81 0.90 1.03 1.00 1.35 1.28 1.17
Medical & Health Care 1984 0.49 0.45 0.57 0.65 1.00 1.30 1.38 1.49 1988-89 0.54 0.47 0.68 0.51 1.00 1.26 1.28 1.31 1993-94 0.50 0.52 0.66 0.51 1.00 1.10 1.27 1.23 1998-99 0.51 0.62 0.67 0.64 1.02 1.32 1.30 1.25
Transport 1984 0.49 0.68 0.83 1.07 1.00 1.36 1.29 1.38 1988-89 0.52 0.57 0.62 0.78 1.00 1.02 1.19 1.37 1993-94 0.53 0.65 0.73 0.81 1.00 1.21 1.13 1.22 1998-99 0.55 0.68 0.77 0.85 1.00 1.29 1.15 1.22
Recreation 1984 0.53 0.70 0.79 0.80 1.00 1.08 1.28 1.36 & Entertainment 1988-89 0.54 0.58 0.51 0.82 1.00 1.03 1.28 1.37
1993-94 0.51 0.58 0.74 0.66 1.00 0.95 1.02 1.23 1998-99 0.52 0.59 0.75 0.67 1.02 0.97 1.04 1.25
Personal Care 1984 0.58 0.72 1.12 1.08 1.00 1.20 1.24 1.31 1988-89 0.54 0.78 0.97 0.73 1.00 1.19 1.29 1.17 1993-94 0.51 0.71 0.73 0.54 1.00 0.96 1.15 1.19 1998-99 0.52 0.82 0.73 0.55 1.02 0.98 1.17 1.21
Others 1984 0.58 0.82 0.91 1.17 1.00 1.76 1.83 2.01 1988-89 0.57 1.02 0.89 0.81 1.00 1.39 1.79 2.08 1993-94 0.55 0.76 1.51 0.90 1.00 1.30 1.55 1.92 1998-99 0.55 0.78 1.51 0.92 1.00 1.33 1.58 1.96
25
Table 4 Estimates of General Scales
General Scales (Sh)
Reference Income Year Household Type (no. of adults, no. of children) (1,0) (1,1) (1,2) (1,3) (2,0) (2,1) (2,2) (2,3)
Low Income 1984 0.58 0.70 0.77 0.88 1.00 1.17 1.26 1.40
($350 p.w.) 1988-89 0.58 0.72 0.81 0.92 1.00 1.23 1.32 1.45 1993-94 0.56 0.78 0.91 0.91 1.00 1.18 1.25 1.33 1998-99 0.57 0.83 0.93 0.95 1.00 1.26 1.31 1.42
Medium Income 1984 0.58 0.73 0.82 0.94 1.00 1.27 1.36 1.49 ($600 p.w.) 1988-89 0.58 0.72 0.80 0.91 1.00 1.23 1.33 1.46
1993-94 0.56 0.77 0.92 0.90 1.00 1.18 1.25 1.34 1998-99 0.57 0.83 0.93 0.95 1.00 1.25 1.31 1.41
High Income 1984 0.57 0.76 0.86 0.99 1.00 1.36 1.44 1.58 ($1000 p.w.) 1988-89 0.58 0.72 0.79 0.91 1.00 1.23 1.33 1.47
1993-94 0.56 0.77 0.92 0.89 1.00 1.18 1.25 1.34 1998-99 0.57 0.80 0.93 0.94 1.00 1.24 1.29 1.38