• Summary
• Downloads
• Explanatory Notes
• Related Information
• Past Releases

EXPLANATORY NOTES


INTRODUCTION

1 This publication presents the results of a study of the effects of taxation and government expenditure on the distribution of income among private households in Australia in 2003-04. Previous studies were conducted in relation to 1984, 1988-89, 1993-94 and 1998-99. The approach taken is only one of several ways of undertaking such a study.

CHANGES IN THIS ISSUE

2 There are several significant changes between the 2003-04 study and the previous studies. There have been changes in the data sources used in the study, and in the study methodology. Consequently the results of this study are not directly comparable with the results of previous studies. The main changes that have impacted on this study are outlined in this section.

Changes in Household Expenditure Survey 2003-04

3 The Household Expenditure Survey (HES), which is one of the major data sources used in the study, underwent a number of major changes in 2003-04. The survey was integrated with the 2003-04 Survey of Income and Housing (SIH). This led to the redefinition of a number of income items so that they were aligned with the corresponding SIH data items, resulting in some loss of comparability between 1998-99 and 2003-04 HES data for private income and taxes on income.

4 The main data items affected were:

• income from wages and salaries no longer includes income from bonuses (including leave loading); in 2003-04 these bonuses averaged $13.55 per week per household
• income no longer includes the value of selected goods and services provided free or at reduced cost by employers to employees for their own private use or goods and services obtained from a person's own unincorporated business (in the 1998-99 HES the value of incidental items such as food and motor vehicle fuel were included); in 2003-04 the value of these goods and services averaged $1.30 per week per household
• income no longer includes the income of children aged under 15 years; in 2003-04 this income averaged $0.30 per week per household
• the integration also resulted in a change to income tax liability estimates (referred to as taxes on income in this study), which are now derived only using data items available in the SIH
• the survey included an expanded range of questions to collect details about income - in particular, information was collected about expected income in the current financial year from own unincorporated business and investments, and therefore estimates for these components of income no longer depend on data collected with respect to the previous financial year.

Changes in the study of the effects of government benefits and taxes on household income:

6 The main changes in this study were:

• the methodology for estimating and allocating taxes on production has changed, with more taxes being allocated to households. 1998-99 data for this component have been reestimated on the 2003-04 basis. For more information see Appendix 4
• some government expenditure transaction types that were included in social transfers in kind allocated to households in 1998-99 have been treated differently in this study. These relate to subsidy expenses (since they are treated as part of taxes on production), and capital transfer expenses (since there is no obvious way to measure how the resulting benefits accrue to households, and since they do not necessarily accrue to households within the same period); in 2003-04 these transactions averaged $0.61 per week per household for education, $0.49 for health and $1.19 for social security and welfare
• government revenue from sales of goods and services has been offset against expenses before allocation. In the 1998-99 study only the estimated revenue under the higher education contribution scheme (HECS) was offset against expenses, and the estimated revenue from HECS payments understated the actual revenue. If the 1998-99 method for treating government revenue had been applied in 2003-04, there would have been an extra $9.02 per week per household allocated for education, an extra $7.31 for health and $1.12 for social security and welfare
• a revised methodology for allocating government expenditure on the child care benefit was introduced using more sophisticated modelling techniques
• recording of transactions in Government Finance Statistics (GFS) (which are used to identify the amount of government expenditure to be allocated in the study) is likely to be more accurate than in 1998-99. Accrual accounting was introduced into GFS in 1998-99 and there are a number of problems with the data for that year
• some data sources used in allocating health expenditure have changed which may have impacted on the distribution of the allocation.

• adoption of revised terminology, consistent with the 1993 System of National Accounts and latest government finance classifications. Direct benefits are now referred to as Social assistance benefits in cash, Indirect benefits as Social transfers in kind, Direct taxes as Taxes on income and Indirect taxes as Taxes on production
• revised classifications for present data on social assistance benefits in cash and taxes on production
• inclusion of the items equivalised private, disposable and final income, which provide additional indicators of a household's relative wellbeing when compared to other households of different size and composition
• new tables presenting data by equivalised private, disposable and final income quintile, net worth quintile, contribution of government pensions and allowances to gross household income, states and territories
• tables showing comparisons of data with previous studies are no longer included. However, Appendices 4 and 5 compare 2003-04 data with the 1998-99 study as well as explain the changes in the methodology between the two studies.

9 While income is usually received by individuals, it is normally shared between partners in a couple relationship and with dependent children. To a lesser extent, it may be shared with other children, other relatives and possibly other people living in the same household, for example through the provision of free or cheap accommodation. This is particularly likely to be the case for children other than dependants and other relatives with low levels of income of their own. Even when there is no transfer of income between members of a household, nor provision of free or cheap accommodation, members are still likely to benefit from the economies of scale that arise from the sharing of dwellings.

10 Household characteristics, including household income, are therefore the main information required for analysing income distribution. In this publication, the income distribution measures are all calculated with respect to households as most of the relevant characteristics of persons relate to their household circumstances. Such measures are sometimes known as household weighted estimates.

11 Income refers to regular and recurring cash receipts from employment, investments and transfers from government, private institutions and other households. A set of income concepts have been used in this study to describe the effects of different types of government benefits and taxes.

Private Income

12 The starting point is private income, which is the most restricted concept of income used in the study. It comprises all regular cash payments received excluding social assistance benefits in cash. Sources from which private income may be received include:

• wages and salaries (whether from an employer or own incorporated business)
• profit/loss from own unincorporated business (including partnerships)
• investment income (interest, rent, dividends, royalties)
• private cash transfers (e.g. superannuation, regular workers' compensation, income from annuities and child support, and other transfers from other households).

• income in kind including employee benefits such as the provision of a house or a car and employer contributions to pension and superannuation funds - however, income in kind provided as part of a negotiated salary sacrifice arrangement is regarded as cash or "near cash" income and included within the scope of income presented in this publication; it is estimated that about two-thirds of salary sacrificed income is included in the 2003-04 estimates of gross income used in this study
• capital transfers such as inheritances and legacies, maturity payments on life insurance policies, lump sum compensation for injuries or other damage
• capital gains and losses.

Disposable Income

15 Disposable income is derived by deducting estimates of taxes on income from gross income. Disposable income better represents the economic resources available to meet the needs of households.

Disposable income plus social transfers in kind

16 The value of government social transfers in kind for education, health, housing and social security and welfare is added to disposable income to give disposable income plus social transfers in kind.

Final Income

17 Final income is the most extensive concept of income used in the study. Final income is equal to household disposable income plus social transfers in kind, less taxes on production.

18 The following diagram illustrates these different income concepts.

20 The equivalence scale has been used to adjust private, disposable and final income for differing household sizes and composition. For more information on equivalised income refer to Appendix 2.

Lowest income decile

21 While equivalised income generally provides a useful indicator of economic wellbeing, there are some circumstances which present particular difficulties. Some households report extremely low and even negative income in the survey, which places them well below the safety net of income support provided by social security pensions and allowances. Households may underreport their incomes in the survey at all income levels, including low income households. However, households can correctly report low levels of income if they incur losses in their unincorporated business or have negative returns from their other investments.

22 Studies of income and expenditure reported in the past HES surveys have shown that such households in the lowest income decile and with negative gross incomes tend to have expenditure levels that are comparable to those of households with higher income levels (and slightly above the average expenditures recorded for the fifth income decile). This suggests that these households have access to economic resources, such as wealth, or that the instance of low or negative income is temporary, perhaps reflecting business or investment start up. Other households in the lowest income decile in past surveys had average incomes at about the level of the single pension rate, were predominantly single person households, and their principal source of income was largely government pensions and allowances. However, on average, these households also had expenditures above the average of the households in the second income decile, which is not inconsistent with the use of assets to maintain a higher standard of living than implied by their incomes alone.

23 It can therefore be reasonably concluded that many of the households included in the lowest income decile are unlikely to be suffering extremely low levels of economic wellbeing. Income distribution analysis may lead to inappropriate conclusions if such households are used as the basis for assessing low levels of economic wellbeing. For this reason, tables showing statistics classified by equivalised income quintile include a supplementary category comprising the second and third income deciles, which can be used as an alternative to the lowest income quintile (for an explanation of quintiles and deciles, see Appendix 1).

24 With the 2003-04 HES, analysis of households in the lowest income decile can be improved through direct observation of the expenditure and net worth of these households. An examination of these low income households is presented in Appendix 4 of Household Wealth and Wealth Distribution, Australia, 2003-04 (cat.no. 6554.0).

MAJOR DATA SOURCES

25 The three major data sources used in this study are the 2003-04 ABS Household Expenditure Survey (HES), ABS Government Finance Statistics, and Input-Output tables from the Australian System of National Accounts (ASNA).

HOUSEHOLD EXPENDITURE SURVEY

26 The 2003-04 HES collected detailed information about the expenditure, income, assets, liabilities and household characteristics of private dwellings throughout Australia. The sample consisted of approximately 7,000 households, which were enumerated from July 2003 to July 2004. The summary of the results from the survey was published in Household Expenditure Survey, Australia: Summary of Results, 2003-04 (cat.no. 6530.0).

27 Previous Household Expenditure Surveys were conducted in 1974-75, 1975-66, 1984, 1988-89, 1993-94 and 1998-99.

28 Information reported in the HES is used as the basis for modelling the effects of various government benefits and taxes on household income. The survey provided details on the composition of households and the characteristics of their members, the level and sources of their income, and the patterns of their expenditure. Household income data were used to provide measures of private income and social assistance benefits in cash from the government; income, personal and household characteristics and taxation criteria for 2003-04 were used to calculate taxes on income paid; characteristics of household members and their expenditure patterns were used to identify recipients of social transfers in kind from government; and expenditure data were used to calculate taxes on production paid.

29 The Household Expenditure Survey and Survey of Income and Housing User Guide, Australia, 2003-04 (cat. no. 6503.0), describes the definitions, concepts, methodology and estimation procedures used in the HES and the SIH.

Survey scope and coverage

30 The survey collects information by personal interview from usual residents of private dwellings in urban and rural areas of Australia, covering about 98 per cent of the people living in Australia. Private dwellings are houses, flats, home units, caravans, garages, tents and other structures that are used as places of residence at the time of interview. Long-stay caravan parks are also included. These are distinct from non-private dwellings which include hotels, boarding schools, boarding houses and institutions. Residents of non-private dwellings are excluded.

31 The survey also excludes:

• households which contain members of non-Australian defence forces stationed in Australia
• households which contain diplomatic personnel of overseas governments
• households in collection districts defined as very remote or Indigenous Communities - this has only a minor impact on aggregate estimates except in the Northern Territory where such households account for about 23% of the population.

Final sample

33 The final sample on which estimates were based, is composed of households for which all necessary information is available. The information may have been wholly provided at the interview (fully-responding) or may have been completed through imputation for partially responding households. Of the selected dwellings, there were 9,753 households in scope of the survey, of which 6,957 (71%) were included as part of the final HES estimates. The final sample consists of those 6,957 households, comprising 13,748 persons aged 15 years and over.

Reliability of estimates

34 The estimates provided in the survey are subject to two types of error, non-sampling and sampling error.

35 Non-sampling error can occur in any collection, whether the estimates are derived from a sample or from a complete collection such as a census. Sources of non-sampling error include non-response, errors in reporting by respondents or recording of answers by interviewers, and errors in coding and processing the data. Non-sampling errors are difficult to quantify in any collection. However, every effort is made to reduce non-sampling error to a minimum by careful design and testing of the questionnaire, training of interviewers and data entry staff, and extensive editing and quality control procedures at all stages of data processing.

36 The estimates are based on a sample of possible observations and are subject to sampling variability. The estimates may therefore differ from the figures that would have been produced if information had been collected for all households. A measure of the sampling error for a given estimate is provided by the standard error, which may be expressed as a percentage of the estimate (relative standard error). Further information on sampling variability is given in Appendix 3.

Underestimation of some expenditure

37 The average expenditure on tobacco recorded by households in the sample is well below the level which would be expected from estimates of apparent consumption of this item i.e. recorded Australian production plus imports less exports. Reported expenditure on gambling is also well below the expected level. No adjustment has been made to any of the reported expenditure data.

GOVERNMENT FINANCE STATISTICS

38 The ABS regularly produces summaries of government revenues and expenses. These government finance statistics (GFS) provide Commonwealth, state or territory and local government taxation revenues classified by type of tax and expenditures classified by purpose and type of economic transaction. The Government Purpose Classification (GPC) identifies the functional areas to which expenses relate (e.g. health, housing and welfare) while the Economic Transactions Framework (ETF) identifies the type of transaction. For example, direct cash payments to households are distinguished from expenses relating to the payment of administrative staff and from expenses on building construction. It is from the combination of these classifications that direct and indirect expenses in various programs were identified.

39 Estimates of total government expenses (for Commonwealth, state or territory and local government) used for social transfers in kind, and to compare the results of the allocation of social assistance benefits in cash, were specially tabulated and reflect 2003-04 data at the time of release of 2004-05 GFS. Taxation information, used to assess the results of tax imputation methods, was obtained from the 2004-05 issue of Taxation Revenue (cat. no. 5506.0).

INPUT-OUTPUT TABLES

40 Input-Output tables form part of the ASNA and provide a means of undertaking detailed analysis of the process of production, the use of goods and services (products) and of the income generated in that production. They show, for the economy as a whole and for groups of products, the total resources in terms of domestic output and imports, and the uses of goods and services in terms of intermediate consumption, final consumption, gross capital formation and exports.

41 The estimation of the incidence of taxes on production to households is based on extensive use of these Input-Output tables. Australian National Accounts: Input-Output Tables, 2001-02 (cat. no. 5209.0.55.001) includes the supply-use tables with detailed explanatory notes on the data sources, content and methods of construction used.

METHODS

UNIT OF ANALYSIS

42 The household is the basic unit of analysis in this publication. A household consists of one or more persons, at least one of whom is at least 15 years of age, usually resident in the same private dwelling. The persons in a household may or may not be related.

43 The household is adopted as the basic unit of analysis because it is assumed that sharing of the use of goods and services occurs at this level. If smaller units, say persons, are adopted, then it is difficult to know how to attribute to individual household members the use of shared items such as food, accommodation and household goods. Intra-household transfers are excluded. For example, if one member of the household were to pay board to another member of the same household then this is not considered as an increase in the amount of income or housing costs of the household.

BENEFITS AND TAXES ALLOCATED

44 The aim of the study has been to allocate only those benefits and taxes relevant to households. No attempt has been made to allocate the whole of government expenditure and revenue. Those government expenses and revenues which are allocated and those that are not allocated in the study are illustrated in the following graphs.

46 Many social transfers in kind were not allocated because:

• there was no clear conceptual basis for allocation
• they related to segments of the population not covered by HES
• target groups could not be identified within HES data
• expenditure on target groups could not be isolated in GFS data.

SOCIAL ASSISTANCE BENEFITS IN CASH

48 Social assistance benefits in cash were defined as selected payments in cash by Commonwealth, state or territory and local government to Australian residents and cover:

• age pension
• disability support pension
• Veterans' Affairs pension
• family tax benefit
• parenting payment
• unemployment and student allowances such as newstart allowance, youth allowance, sickness allowance, mature age allowance, abstudy/austudy allowances
• other government pensions and allowances such as carer payment, carer allowance, widow allowance, special benefit, wife pension, partner allowance etc.

50 National accounts figures for 2003-04 show that social assistance benefits paid in cash to Australian households were $72,438 million. However, this figure includes some health benefits, which for practical reasons are allocated as health related social transfers in kind (see below). The social assistance benefits in cash recorded in GFS figures that most closely correspond to the estimates provided from the HES are those relating to social security and welfare and education. The expenses on social assistance in cash in these areas amounted to $68,740 million. Of this amount, the study allocated $55,003 million to households. The discrepancy between expenses reported in GFS and the amount allocated is due to:

• scope exclusions in the HES. The HES does not cover the whole population, and in particular, excludes residents of special dwellings. Many residents of special dwellings, such as nursing homes, are recipients of these benefits
• cash benefits that are not covered by income questions in the HES. These benefits comprise irregular or one-off cash payments such as crisis or disaster payments
• the reconciliation credit element of family tax benefit which was not treated as a 2003-04 receipt in HES
• under-reporting of government benefits and pensions by HES respondents.

52 Except for government expenditure on housing (see details following), social transfers in kind were based on the cost to government of the provision of those services. More specifically, the total value of social transfers in kind was defined as Commonwealth, state or territory and local government expenses, net of intra-government transfers, minus personal benefit payments paid in cash minus government revenue from the sale of goods and services. In the case of health benefits, however, some benefits paid in cash which were not collected as personal benefits in the HES are allocated together with the health social transfers in kind.

Education

53 Social transfers in kind were allocated for school education, tertiary education and other education benefits. School education includes benefits from pre-school education, primary and secondary education, student transportation, special education and education n.e.c. Tertiary education includes benefits from university education, technical and further education, and tertiary education n.e.c.

School education

54 Government expenses relating to pre-school education were allocated to households containing children aged 3, 4 or 5 years. An average benefit per child attending pre-school in each state and territory was derived by dividing GFS expenses in each state and territory by the number of children attending pre-school in that state or territory as measured by the 2002 Child Care Survey (cat. no. 4402.0). The 2002 figures were adjusted to represent the HES reference period using the change in the number of 3 to 5 year olds over this period according to Population by Age and Sex, Australian States and Territories (cat. no. 3201.0). The number of children attending pre-school in each household was imputed according to pre-school participation rates. Pre-school participation rates were separately derived for 3, 4 and 5 year olds by dividing the number of children attending pre-school (largely as measured by the Child Care Survey) by the estimated population of 3, 4 and 5 year olds in that state or territory. The benefit received by households was the (imputed) number of children attending pre-school multiplied by the average pre-school benefit for their state or territory of residence. Of $485 million available for allocation, $460 million was allocated for pre-school benefits. Underallocation occurred because the number of 3 to 5 year olds reported in the HES was less than the number reported in the Child Care Survey.

55 Government expenses relating to primary and secondary education and student transportation were allocated to households containing primary and secondary school students. An average benefit, for both education and transportation, was calculated for six student types: government primary, Catholic primary, other non-government primary, government secondary, Catholic secondary and other non-government secondary. Data on average expenditure for government school children was obtained from the Ministerial Council on Education, Employment and Youth Affairs' (MCEETYA) National Schools Statistics Collection, and average expenditure per student type for all non-government school students was obtained from the Department of Education, Science and Training.

56 Numbers of full-time equivalent students in August 2003 and August 2004 were obtained from Schools, Australia (cat. no. 4221.0). These were averaged to obtain 2003-04 estimates and aggregate expenditure was calculated. This was compared with GFS expenses on primary and secondary education and an adjustment factor was calculated and applied to average expenditure by student type. This ensured that average student benefits reflected GFS expenses. Households were allocated benefits according to the reported number of members who attended schools of each type. Of $23,999 million available, $24,652 million was allocated. Overallocation of benefits occurred because the number of school students reported in the 2003-04 HES exceeded the estimates of school students provided in Schools, Australia.

Tertiary education

57 Government expenses relating to university education were allocated to higher education students. Average benefits were derived by deducting government revenue from the sale of university education services (which includes payments under the Higher Education Contributions Scheme (HECS)) from GFS expenses and dividing net expenses by benchmark enrolment data from the 2003 and 2004 issues of Education and Work, Australia (cat. no. 6227.0). Part-time students were assumed to receive half the benefits of full-time students. Benefits were allocated to households according to the number of members who reported themselves as attending higher education. Of the $5,169 million available for allocation, $5,086 million was allocated. Underallocation of benefits occurred because HES numbers of higher education students, which exclude students living in student residences, were less than benchmark estimates of student numbers.

58 Government expenses relating to technical and further education were allocated to Technical and Further Education (TAFE) students. Average benefits were derived by dividing GFS expenses by the estimated number of TAFE students from the HES. Part-time students were assumed to receive half the benefits of full-time students. Benefits were allocated to households according to the number of members who reported themselves as attending TAFE. Of the $3,415 million available for allocation, the entire amount ($3,415 million) was allocated.

59 Government expenses relating to tertiary education n.e.c. were allocated to all persons who reported that they attended a tertiary institution either full or part-time. An average benefit was derived by dividing GFS expenses by benchmark enrolment data for higher education students and estimated number of TAFE students from the HES. The same benefit was allocated to all student types regardless of institution type and full-time or part-time status. Benefits were allocated to households according to the number of members who reported themselves as tertiary students. Of the $52 million available for allocation, the entire amount ($52 million) was allocated.

Other education benefits

60 Government expenses relating to special and other education were allocated to all pre-school, primary and secondary education students. An average benefit was derived for each state and territory by dividing GFS expenses in each state and territory by the reported number of pre-school students based on the 2002 Child Care Survey (cat. no. 4402.0) and the number of primary and secondary students from the 2003 and 2004 issues of Schools, Australia (cat. no. 4221.0). The 2002 Child Care Survey figures were adjusted to represent the HES reference period using the change in the number of 3 to 5 year olds over this period according to Population by Age and Sex, Australian States and Territories (cat. no. 3201.0). An equal average benefit was allocated to each student and household benefits were the sum of household members' benefits. Of $1,476 million available, $1,517 million was allocated. Overallocation of benefits occurred because the number of school students reported in the 2003-04 HES exceeded the estimates of school students provided in Schools, Australia.

Health

61 Health benefits were allocated for acute care institutions, community health services, pharmaceuticals and other health benefits. Other health benefits cover public health services, health research and health administration n.e.c.

62 These benefits were allocated to households according to an insurance premium approach. Instead of allocating benefits according to actual use of health services (which implies that benefits increase with ill health), members of the HES population were allocated benefits according to the average utilisation rates for their age, sex and state or territory of residence groups.

Acute care institutions

63 Government expenses relating to acute care institutions were allocated to all persons according to hospital bed utilisation rates (average number of days in hospital per person) for their age, sex and state or territory of residence group. Hospital utilisation was used as an indicator of the use of all institutional services and benefits. The utilisation rates were calculated using patient days obtained from Australian Hospital Statistics, 2002-03 (cat. no. 8906.0) and 2003 and 2004 resident population estimates from Population by Age and Sex, Australian States and Territories (cat. no. 3201.0).

64 The benefit allocated to households was the sum of each member's utilisation rate multiplied by the average benefit per hospital bed day in their state or territory of residence. The average benefit per hospital bed day was derived by dividing GFS expenses per state or territory by the number of days spent in hospital by the state or territory population from Australian Hospital Statistics, 2002-03 (cat. no. 8906.0). Of $19,813 million available for allocation, $18,352 million was allocated. Underallocation of benefits occurred because the HES excludes residents of special dwellings.

Community health services

65 Government expenses relating to community health services were allocated to all persons according to the doctor visit rate for their age, sex and state or territory of residence. Doctor visits were used as an indicator of utilisation for all non-institutional benefits and services such as dentists, specialists, maternal and infant centres, chiropractors, pathology services and domiciliary care. Utilisation rates for doctors were calculated using data on professional attendances obtained from the Medicare Australia website (www.medicareaustralia.gov.au) and resident population estimates from Population by Age and Sex, Australian States and Territories (cat. no 3201.0).

66 The benefit allocated to households was the sum of each member's utilisation rate multiplied by the average benefit per doctor visit in their state or territory of residence. An average benefit per doctors visit was derived by dividing GFS expenses per state or territory by the number of doctor visits made by the state or territory population (from Medicare Australia). Of $16,206 million available for allocation, $15,743 million was allocated. Underallocation of benefits occurred because the HES excludes residents of special dwellings.

67 In previous studies data on doctor visits were obtained from the National Health Survey. Use of Medicare data to allocate social transfers in kind relating to community health services may impact on the comparability of the distribution with previous studies.

Pharmaceuticals

68 Government expenses relating to pharmaceuticals, medical aids and appliances were allocated to all persons according to their eligibility for pharmaceutical concessions as well as usage of prescribed medicines for their age, sex and state or territory of residence group. In 2003-04, concessional benefits were available to holders of pensioner concession cards, health care cards, Commonwealth seniors health cards and Department of Veterans' Affairs Gold, Orange or White cards. Expenses relating to pharmaceuticals, medical aids and appliances were divided between those who were eligible for concessions and those who were not, in proportion to the cost to government of concessions provided by the Department of Health and Ageing. Utilisation rates were calculated using data on numbers of prescriptions obtained from the Department of Health and Ageing and resident population estimates from Population by Age and Sex, Australian States and Territories (cat. no 3201.0). Estimates of concession card holders were adjusted to account for persons holding more than one card.

69 Household benefits were the sum of each household member's utilisation rate multiplied by the average benefit per prescribed medicine according to their eligibility for concessions. Average benefits per prescribed medicine for those who were eligible for concessions and those who were not, were derived by dividing GFS expenses by total prescribed medicine utilisation for the two groups. For persons receiving concessions, total prescribed medicine utilisation was the product of benchmark numbers of holders of each type of concession card (obtained from annual reports of the Department of Families, Community Services and Indigenous Affairs and the Department of Veterans' Affairs) multiplied by the average utilisation rate for those eligible for concessions (derived by applying utilisation rates calculated using Department of Health and Ageing data and resident population estimates to persons who reported holding cards in the HES). For others, total prescribed medicine utilisation was the product of the estimated resident population (minus those who are holders of concession cards) multiplied by the average utilisation rates. Benefits were adjusted according to state and territory differences in expenses. Of the $6,397 million available for allocation, $5,880 million was allocated. Underallocation of benefits occurred because the HES excludes residents of special dwellings.

Other health benefits

70 Government expenses relating to public health, health research and health administration n.e.c. were allocated to all persons. An average benefit was derived by dividing GFS expenses per state and territory by the estimated resident population, from Population by Age and Sex, Australian States and Territories (cat. no. 3201.0). Benefits per household were equal to the number of members multiplied by the average benefit. Of the $5,532 million available for allocation, $5,405 million was allocated. Underallocation of benefits occurred because the HES excludes residents of special dwellings.

Housing

71 Government expenses relating to housing largely involves building new houses for rent or at a subsidised cost. These expenses were not allocated amongst HES households because it is difficult to identify likely future recipients of the benefits. Payments of the First Home Owners Grant were not allocated since they are regarded as a capital expense.

72 Instead, benefits were allocated to households in government rental accommodation according to the value of their rent subsidy. The value of their rent subsidy was taken to be the difference between the rent paid by the household and the estimated value of private market rent according to the state or territory, region, type of dwelling and number of bedrooms. Median market rents for private dwellings were obtained from the 2001 Census and the rents were adjusted to December 2003 prices according to the percentage change in the Consumer Price Index (CPI). In total, $1,406 million was allocated.

Social security and welfare

73 Government expenses relating to social security and welfare programs, other than direct cash payments (see Social assistance benefits in cash described previously) and payments for Child Care Benefit (CCB), were allocated to persons who received social security and welfare cash benefits. An adjustment was made to GFS expenses to exclude government expenditure on residential aged care amounting to $4,433 million. Average social transfers in kind for different types of benefit recipients were calculated by dividing indirect GFS expenses by the number of recipients. The number of recipients was based on data in Yearbook, Australia, 2007 (cat. no. 1301.0), adjusted using HES data to avoid double counting of persons receiving multiple benefits within each benefit type. Household benefits were the sum of household members' benefits. Of $13,089 million available for allocation, $12,336 million was allocated. Underallocation of benefits occurred because of HES population exclusions and under-reporting of government cash benefits by HES respondents.

74 Government expenditure on CCB was allocated to households using a revised methodology. In previous studies, government assistance for child care was allocated to households with children under 12, according to household income and the probability that the children were attending eligible child care. The probability of a child attending care was the sum of the ratios of the number of children attending long day care, family day care, occasional care and outside school hours care to total numbers of children in these categories according to age and whether the children attend school as reported in the 2002 Child Care Survey (cat. no. 4402.0). These probabilities were applied to the HES sample, summed for each household member and then multiplied by the probability of children attending child care according to household labour force status. This probability was then multiplied by the rate of assistance provided according to their income and number of children.

75 In the 2003-04 study a regression method was used in allocating CCB to households. It involved fitting a prediction model to the 2002 Child Care Survey (cat. no. 4402.0) dataset using items that are common to both the Child Care Survey and the HES as predictors. Two models were used - firstly a logistic model to determine whether or not formal child care would be used, and secondly a Poisson model to determine the count of hours of formal child care, if used. The models were developed at the child level, but family and household influences on the use of child care were accounted for by including family and household composition type variables in the models.

76 The variables included in the logistic model to estimate whether or not a child would attend formal child care were:

• age of child (under 5 years / 5 years and over)
• state or territory
• part of state or territory (capital city / balance of state)
• mean age of parents (5 year age groups from 25 to 55)
• labour force status of parents (at least one parent not employed / employed parent(s))
• principal source of income of parent(s) (at least one parent with own unincorporated business as principal source / other)
• income unit type (couple / lone parent)
• household type (standard one family / other)
• family type (according to whether or not there are children aged 15 and over or other relatives present)
• joint income of parents

• age of child (under 1 year / 1 to 5 years / 6 to 10 years / 11 years)
• state or territory
• part of state or territory (capital city / balance of state)
• number of children aged 0 to 11 in the household
• hours worked by the parent who works least (as a log term)

79 CCB was allocated to children who lived in households that had expenditure on formal child care (excluding preschool) or for whom the model predicted that they would attend child care (i.e. the '1/0' item referred to above was '1'). Other children received no allocation. If the allocation had been restricted to households that had expenditure on formal child care, households who pay nothing for child care because their child care costs are completely refunded would not have had a chance of receiving an allocation. The 2003-04 method of allocation produces a far lumpier allocation than the 1998-99 approach which allocates benefits from child care assistance to all households with children under 12.

80 The amount of CCB allocated to each child was based on the probability of the child attending child care, multiplied by the number of hours of care predicted by the model, multiplied by a factor to account for differential rates of CCB at various income ranges. For those children in households with expenditure on formal child care the probability of using child care was set to 1, regardless of the probability allocated by the model. For other households the probability output from the model was used.

81 The administrative component of CCB was allocated equally among all children who received a CCB allocation.

82 The allocations were summed to the household level. Of the $1,768 million spent on CCB ($1,388 million in direct payments and $380 million in administrative costs), all was allocated.

TAXES ON INCOME

83 Taxes on income is the sum of personal income tax plus the Medicare levy for all members of the household.

84 Estimates of income tax were modelled, rather than collected from respondents, for a number of reasons. Firstly, changes in income, family or other circumstances of the respondent, which are not described in the survey, may affect full year income tax assessments. Secondly, income tax assessments are only made after the end of the financial year, and therefore are not yet available at the time that current income is collected from respondents. Thirdly, the income tax assessment of respondents may be affected by certain expenditures which they make, such as donations to charities, or other particular circumstances which are not captured in the survey. Finally, the HES provides sufficient relevant information to allow a relatively comprehensive model to be constructed.

85 Taxes on income were imputed according to the following steps:

• for each individual, non-taxable components were deducted from reported gross income to give taxable income
• an approximate adjustment was made for deductions such as union dues and other work-related expenses
• tax payable was imputed from taxable income using 2003-04 marginal tax rates
• tax offsets were calculated according to household characteristics and eligibility criteria for age pensioners, beneficiaries, low income earners, dependent spouses, sole parents, residential zones, and franked dividend imputation credits
• total tax offsets were subtracted from gross tax to give final tax
• the Medicare levy, calculated using 2003-04 tax rules, was added to final tax
• individual final tax plus Medicare levy was aggregated for households.

87 In total, the HES population was calculated to have paid $85,719 million in taxes on income. GFS figures for 2003-04, however, show revenue from income tax levied on individuals to be $102,622 million. The main reasons for the underestimation of taxes on income in this study are:

• the calculation of tax liability on regular cash income only. Taxes such as capital gains tax were not calculated because the HES did not collect the relevant information
• insufficient data to model some components
• scope exclusions in the HES
• understatement of income in the 2003-04 HES.

89 The methodology used to calculate taxes on production in 2003-04 differs substantially from the method used to calculate indirect taxes in the 1998-99 study. Further information on the differences between these methods is provided in Appendix 4.

90 In allocating the taxes on production, it was assumed that industries will pass the burden of the taxes on production they pay to the purchasing industries and/or final consumers through higher prices. Also, the burden of the tax will be passed from one industry to another until the total burden of the tax is passed on to a final demand sector, one of which is the household sector.

91 The amount of taxes on production paid by HES households was calculated as follows:

• the estimation of the incidence of taxes on production to households is based on the extensive use of Input-Output tables from within the ASNA. The Input-Output tables present a comprehensive picture of the supply and use of goods and services in the economy and the income generated from production. It records the flows of products from one industry to another and to final demand for consumption. The 2001-02 Input-Output tables compiled in terms of 109 commodity classifications (IOCC) were used to calculate a tax rate for each of these commodity classifications:
• household expenditure is classified in the HES according to the Household Expenditure Classification (HEC). Approximately 600 HEC codes were mapped to 109 IOCC codes
• the expenditure on each HEC code was multiplied by the relevant tax rates to estimate the total final incidence of taxes on production on household consumption expenditure for each household.

• for owner occupiers, taxes on ownership of dwellings were taken to be equal to expenditure on local government rates and land tax
• for private renters, the proportion of rent constituting taxes on production was estimated, based on the amount of rates paid by owner occupiers.

• HES excludes some of the population
• household expenditures were, to a degree, understated, particularly for highly taxed items such as tobacco and gambling
• the tax rates derived from the Input-Output information refer to the 2001-02 financial year. In some cases, the tax rates used in this study will be higher than those in existence in 2003-04 and in other cases, they will be lower.

PRODUCTS AND SERVICES

SPECIAL DATA SERVICES

95 This publication provides a summary of the results of the study of the effects of government benefits and taxes on household income. The ABS offers specialist consultancy services to assist clients with more complex statistical information needs. Subject to confidentiality and standard error constraints, values for the data items included in this publication can be provided for HES population subgroups as requested. HES population subgroups can be defined according to the HES data items listed in Appendix 4 of Household Expenditure Survey and Survey of Income and Housing User Guide, Australia, 2003-04 (cat. no. 6503.0). All specialist consultancy services attract a service charge and clients will be provided with a quote before information is supplied. For further information, contact ABS information consultants on 1300 135 070.

UNIT RECORD FILE

96 For clients who wish to undertake more detailed analysis of the survey data, a confidentialised unit record file (CURF) will be made available in the near future. Both the basic and expanded versions of the 2003-04 HES CURF have been revised by appending the study estimates to each household record. Clients who have already purchased the 2003-04 Household Expenditure Survey and Survey of Income and Housing - Confidentialised Unit Record Files, 2003-04 (cat. no 6540.0) will receive the revised CURF free of charge. A full range of up-to-date information about the availability of ABS CURFs and about applying for access to CURFs is available via the ABS web site <https://www.abs.gov.au> (see Services We Provide, Confidentialised Unit Record Files (CURFs)). Inquiries to the ABS CURF Management Unit should email: curf.management@abs.gov.au, or telephone (02) 6252 5853.

RELATED PUBLICATIONS

97 Users may wish to refer to the following ABS products which relate to government benefits, taxes and household income:

• Household Income and Income Distribution, Australia, 2003-04 (cat.no. 6523.0)
• Household Income and Income Distribution, Australia, Detailed Tables, 2003-04 (cat. no. 6523.0.55.001)
• Household Expenditure Survey, Australia: Summary of Results, 2003-04 (cat.no. 6530.0)
• Household Expenditure Survey, Australia: Detailed Expenditure Items, 2003-04 (cat. no. 6535.0.55.001)
• Household Expenditure Survey and Survey of Income and Housing User Guide, Australia, 2003-04 (cat. no. 6503.0)
• Household Wealth and Wealth Distribution, Australia, 2003-04 (cat.no. 6554.0)
• Information Paper: Experimental Estimates of Personal Income for Small Areas, Taxation and Income Support Data, 1995-96 to 2000-01 (cat. no. 6524.0)
• Labour Force, Australia (cat. no. 6202.0)