APPENDIX 5 COMPARISON OF INCOME, BENEFITS AND TAXES FOR 1998-99 AND 2003-04 STUDIES
This appendix compares the results of the 2003-04 study of the distributional impact of government taxation and expenditure on the income of private households, with those of the previous study which was conducted in respect of 1998-99. Differences between the two studies that may impact on the comparison are also outlined.
In previous issues of this publication, data from the latest study were compared with data from each of the previous studies (conducted in respect of 1984, 1988-89 and 1993-94). This was done by conducting a separate comparative study utilising similar methodology to that used in the 1984 study. However this became increasingly difficult with each study because of the changes that had occurred in both the methodology and data sources. In particular, the implementation of accrual accounting in government finance statistics in 1998-99 meant that it was not possible to compile comparable estimates of social transfers in kind. Further changes have been made in both methodology and data sources in the 2003-04 study and it is no longer possible to compile data for the comparative study.
COMPARABILITY OF 1998-99 AND 2003-04 DATA
The 1998-99 data presented in this appendix differ from those originally published in the 1998-99 issue of this publication. Some revisions were made to the data to correct errors detected after the original release. These revisions were incorporated in the 1998-99 confidentialised unit record file that was re-released in 2002. Changes have also been made to the methodology to make the 1998-99 study results more comparable with the 2003-04 study. The main changes are:
For comparison purposes, the results for 1998-99 included in this appendix have been presented with:
- estimates of average weekly household income in 1998-99 were revised upwards for private income, income from Veterans' Affairs pensions and income from other government pensions and allowances, totalling $10
- estimates of social assistance benefits in cash for 1998-99 have been recompiled using the 2003-04 classification
- estimates of taxes on production for 1998-99 have been recompiled using the revised methodology outlined in Appendix 4, resulting in the total 1998-99 estimate for the selected taxes on production increasing by $43 per week per household.
However, significant differences remain between the two studies and they should be compared with caution. For more information see paragraphs 3 to 6 of the Explanatory Notes.
- the revised terminology used in the 2003-04 study applied (for more information see paragraph 7 of the Explanatory Notes)
- the application of equivalence scales to produce estimates of equivalised private income and equivalised final income for comparison with 2003-04 data
- the figures adjusted to 2003-04 dollars using changes in the Consumer Price Index.
COMPARISON OF INCOME ESTIMATES
Table A5.1 shows that average weekly private household income has increased by 9% in real terms between 1998-99 and 2003-04. However, the items are not directly comparable because of improvements in the estimation of business and investment income in 2003-04, and the redefinition of a number of income items as a result of the integration of the Household Expenditure Survey and the Survey of Income and Housing in 2003-04.
A5.1 Income, benefits and taxes, Comparison between 1998-99 and 2003-04 studies, All households
Income, benefits and taxes - Average weekly value ($)
|Private income |
|Social assistance benefits in cash(b) |
|Age pension |
|Disability support pension |
|Veterans' Affairs pension |
|Family tax benefit(c) |
|Parenting payment |
|Unemployment and student allowances |
|Other government pensions and allowances |
|Total social assistance benefits in cash |
|Gross income |
|Taxes on income |
|Disposable income |
|Selected social transfers in kind |
|Education benefits |
|School education |
|Tertiary education |
|Other education benefits |
|Total education benefits |
|Health benefits |
|Acute care institutions |
|Community health services |
|Other health benefits |
|Total health benefits |
|Housing benefits |
|Social security and welfare benefits |
|Total selected social transfers in kind |
|Disposable income plus social transfers in kind |
|Selected taxes on production(d) |
|Alcoholic beverages |
|Tobacco products |
|Motor vehicle fuel |
|Ownership of dwellings |
|Food and non-alcoholic beverages |
|Meals out and fast food |
|Motor vehicle purchases |
|Clothing and footwear |
|Other taxes on production |
|Total selected taxes on production |
|GST component in total selected taxes on production |
|Final income |
|Total benefits allocated |
|Total taxes allocated |
|Net benefits allocated |
|Equivalised private income |
|Equivalised disposable income |
|Equivalised final income |
|* estimate has a relative standard error of 25% to 50% and should be used with caution |
|** estimate has a relative standard error greater than 50% and is considered too unreliable for general use |
|. . not applicable |
|(a) In 2003-04 dollars, adjusted using changes in the Consumer Price Index (multiplied by 1.17816). |
|(b) Excludes overseas pensions. |
|(c) Income for 1998-99 refers to Family Allowance payments. |
|(d) Includes GST component. |
Social assistance benefits in cash
Social assistance benefits in cash are comparable between the two studies. Average social assistance benefits in cash paid to households increased 4% in real terms from 1998-99 to 2003-04. Unemployment and student allowances decreased by 29% whereas disability support pensions increased by 13%. The average amount of family tax benefit received by households in 2003-04 was up 40% on the amount of family allowance received in 1998-99.
Social transfers in kind
There was a 7% increase in social transfers in kind across all households between 1998-99 and 2003-04. However, there were a number of changes in the methodology and data sources used in the estimation of social transfers in kind in 2003-04 which impact on comparability. Some government expenditure transaction types that were allocated in the 1998-99 study have not been included in social transfers in kind in 2003-04. These relate to:
In addition, in the 2003-04 study 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 there had been no change in the expenditure transaction types included, or in the treatment of government revenue from sales of goods and services, the 2003-04 estimates for average weekly household social transfers in kind relating to education would have been $10 higher ($97), those relating to health would have been $8 higher ($121) and those relating to social security and welfare would have been $2 higher ($37).
- subsidy expenses (since they are allocated as part of taxes on production)
- 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 within the same period)
There were some other changes (outlined in paragraph 6 of the Explanatory Notes) that cannot be quantified.
Taxes on income
There was a 1% real increase in taxes on income between 1998-99 and 2003-04. Estimates of income tax were modelled in both the studies based on income data collected in the Household Expenditure Survey and the relevant taxation legislation. The Medicare levy surcharge, telephone allowance and the medical expenses rebate components of the tax model were imputed in the 1998-99 study, but there was insufficient information available in 2003-04 to impute them.
Taxes on production
Appendix 4 outlines the new methodology introduced in 2003-04 to calculate taxes on production. This methodology has also been applied to 1998-99 to produce the estimates of taxes on production shown in table A5.1. While the data shown in table A5.1 show similar amounts estimated for taxes on production, it is estimated that the unallocated amount of taxes due to underestimation of expenditure in the HES increased from about $6 billion (in 2003-04 dollars) in 1998-99 to about $15 billion in 2003-04, or about $22 per week per household. The effect was partially offset by an increase in taxes on production included in Household Final Consumption Expenditure (HFCE) due to the introduction of The New Tax System (TNTS) in 2000.
Final income in 2003-04 cannot be directly compared with the 1998-99 figure since all the comparability issues outlined previously impact on this aggregate.
COMPARISON OF INCOME DISTRIBUTION
A5.2 Distribution of household income, benefits and taxes, by Equivalised private income quintile
Equivalised private income quintile
|Income share |
|Private income |
|Social assistance benefits in cash or kind |
|Total taxes |
|Final income |
|Equivalised private income |
|Equivalised final income |
|- nil or rounded to zero (including null cells) |
Table A5.2 compares households income shares for quintile groups between 1998-99 and 2003-04. The quintile groups are formed by ranking households according to their equivalised private income and dividing them into five equal groups.
Private income was more equally spread across the equivalised private income quintiles in 2003-04 than in 1998-99. The lowest three quintiles received greater shares of private income in 2003-04 and the highest two quintiles received lesser shares. The difference in quintile shares shown in private income were reduced by government benefits and taxes so that final income is more evenly distributed than private income in both years.
Final income follows a similar pattern to private income, with final income in 2003-04 being slightly more evenly distributed across the quintiles than in 1998-99. If the level of unallocated taxes on production in 2003-04 had been similar to that estimated for 1998-99, it is likely that final income estimates would have been more evenly distributed.
The share of government benefits going to the lowest and highest quintiles has increased since 1998-99 while the shares going to the remaining quintiles has fallen. Taxes are more equally distributed in 2003-04, with the share of taxes paid by the lower four quintiles increasing and the share paid by the highest quintile decreasing.
Part of the change in the distribution of income is due to the effects of changes over time in household size and composition. When adjustments are made for household size and composition, the change in the spread of both private and final income is reduced. The share of equivalised final income going to the lowest quintile is 4% higher than in 1998-99 (compared with 7% for final income) and the share of equivalised final income going to the highest quintile is similar to 1998-99 (compared with a 2% fall for final income).