Poverty in Australia

Consultant's Report

Problems of Measuring Poverty

THE standard measure used for estimating the number of persons in poverty in Australia is the Henderson Poverty Line, first calculated by Professor Ronald Henderson who chaired the National Poverty Inquiry which reported in the mid-1970s.

The measure used by Professor Henderson was derived from New York based studies of the consumption patterns of American poor persons living in cities.  The relative cost of living for various US poor families of different size was used by Professor Henderson as basic data and translated using equivalence scales (as a basis for determining whether families of different composition and size had the same income) to Australian data.

The Melbourne Institute of Applied Economic and Social Research periodically updates the Henderson measures of poverty in line with changes in wages and prices.  Two concepts of poverty were highlighted by Professor Henderson, one adjusted and the other not adjusted for housing costs.

The housing costs adjustment recognises that homeowners, even though they may have little income, are in a superior financial position to non-homeowners required to pay rent.  Since many pensioners own their own house outright, such an adjustment is absolutely essential to assessing the number of Australians in poverty.

As the need for adjustments for the cost of housing demonstrates, caution is needed if policy recommendations are based on abstract measures of numbers of persons in poverty.  A summary measure (particularly one based on experience in another country in an earlier decade) can not accurately assess the widely differing financial and economic needs of individuals.

Housing is not the only area where there can be significant differences in the circumstances of individuals with the same measured income.  Health and community services is another important case where needs and costs may be different.  There are also regional differences in the cost of living that need to be reckoned with.

Indeed whilst pensioners forced to rent accommodation in the private markets in the major capitals (especially Sydney and Melbourne) face high rentals for low quality accommodation, that does not mean that all non-homeowners will face housing problems.  Renters in rural or some regions of the country (or those in subsidised public housing) could for example be in a more fortunate position.

One important item commonly overlooked in assessments of the adequacy of low incomes in Australia is the imputed value of the comprehensive range of services provided by the Commonwealth, state and local governments to the holders of the Pensioner Health Benefit (PHB) card.

The PHB card provides benefits with a value which varies from state to state and from locality to locality.  The value has been estimated in the range of an additional $10 to $15 per week even following the introduction of the Medicare system which extended free medical services to the whole population.  Major benefits offered under the PHB card include free pharmaceuticals, transport concessions, other concessions such as for rates and community services and comprehensive access to the health system.

These and other reasons mean that measured or reported income alone does not provide an accurate assessment of the needs of the person or family.  Users of crude measurements of poverty based on aggregate data need to be aware of these major limitations.  At best poverty lines are merely guides to potential social welfare problem areas.

This is not to deny that there are poverty and areas of unmet social welfare need in this country.  But as the comparisons of trends over time show, poverty is a relative concept the dimensions of which must be evaluated in terms of specific details and comparisons with the income and well-being of the less poor segment of the population.

Specific groups, such as non-homeowners required to rent in the private rental market, will clearly have greater needs than others.  But to consider all pensioners and beneficiaries as being groups in need is to ignore the fact that some may be comfortably off even with little other income than the pension.  Many of our pensioner homeowners are likely to fall into that category of person.

Substantial improvements have occurred in social welfare benefits recently.  These include increase in the pension relative to average weekly earnings (pensions have increased at a faster rate than wages) and the introduction of the government's family assistance package (providing additional payments of up to $28 per week tax free to low income families with children).

The implementation of these changes will have reduced the numbers affected by and the incidence of poverty in Australia.  These benefits have been substantial (especially after allowing for the tax that would have been paid on additional wages or property income to provide benefits of the same value).

One final aspect of the measurement of poverty is also relevant.  Measuring the income of certain groups in society, especially the self employed and farmers, is a difficult exercise.  Whilst there are many poor farmers and self employed persons, these groups are over-represented (compared to their relative importance in the total population) in the numbers of persons shown to be in poverty.  Problems of measurement of incomes have certainly contributed to this phenomenon.

Measuring the income of this segment of the population is difficult, especially measuring the imputed income accruing to the business and allowing for cash transactions which are prevalent in the self employed segment of the economy.

Trends in Welfare Assistance

Reviewing the total numbers of persons in receipt of social security and veterans affairs pensions highlights the comprehensive range of income support provided to the Australian population.  This support comes from the taxes paid by the working population and by business and other taxpayers.

An analysis of Social Security data reveals the following major trends:

1. Social Security and Veterans Affairs pensioners and beneficiaries have increased as a percentage of the population over the last two decades, despite recent government action to restrict the numbers of pensioners (by reintroducing the assets test and the income test for those aged 70 or more).  Relevant data is:

   Pensioners and Beneficiaries expressed as percentage
   of Total Population

   1969	1978	1983	1987
   8.7%	16.6%	21.3%	20.2%

   
   

2. For assessing the impact of welfare outlays on the tax burden, the relationship between the growth in the number of government income support recipients and the growth in the labour force is a key variable.  In 1987, income support recipients represented 42.5 per cent of the labour force compared with 20.3 per cent in 1969.  That change virtually doubles the tax required to be collected from each member of the workforce to pay the required income support payments.

   Pensioners and Beneficiaries expressed as percentage
   of Total Labour Force

   1969	1978	1983	1987
   20.3%	37.2%	47.1%	42.5%

   
   

   The data shows that a peak was reached in 1983 before the Labor party introduced measures to reduce the number of age pensioners.  The fall from 47.1 per cent of the labour force in 1983 to 42.5 per cent in 1987 reflects declining levels of unemployment over that period and the impact of the specific policy decisions affecting age pensioners.

3. The data presented above permits an approximate estimate to be made of the impact of increasing numbers of income support recipients on personal tax burdens.  The average pension and unemployment benefit payment has varied around an average of 20 per cent of Average Weekly Earnings (AWE) (the maximum benefit is currently around 25 per cent of AWE).  Because of the large number of relatively low income wage earners, average wage levels also represent approximately 80 per cent of AWE (the precise figure may be even lower and closer to 75 per cent).  Using these figures, income support payments for approximately 42 per cent of the labour force necessitate average tax collections from each worker equal to 8.5 per cent of AWE or 10.6 per cent of average wages.

4. The absolute numbers of pensioners and beneficiaries has also increased significantly over the years as shown below.  In some parts of Australia upwards of 20 per cent and as high as 40 per cent of the population are recipients of pensions and benefits from the Commonwealth government.

   Total Numbers of Pensioners and Beneficiaries

   Recipients	1969	1978	1983	1987
   Age	710,527	1,264,778	1,417,218	1,346,925
   Service	76,202	123,955	341,181	375,112
   Other	284,865	927,142	1,520,310	1,560,059
   Total	1,071,514	2315,875	3,278,709	3,282,096

   
   

   This data shows a large growth hi the numbers (included under the category of other) of the pensioner and beneficiary population who were not aged (i.e. not age or service pensioners).  In 1969, 27 per cent of the income support recipients came from groups of work force age.  By 1987, that figure had increased to 47 per cent of the total pensioner population.

5. The increased importance of social welfare outlays over the past two decades is highlighted by the increasing share of PAYE personal income tax collections required to finance social welfare payments as broadly defined in the Budget papers.  Trends are shown below:

   Social Welfare Outlays as Proportion of Personal
   Income Tax Collections

   	1968/69 $m	1978/79 $m	1982/83 $m	1986/87 $m
   Total Receipts	6,228	25,567	44,345	72,184
   of which Net PAYE	2,379	12,804	18,840	28,136
   Social Welfare as % of PAYE	47.4	63.7	74.6	73.0

   
   

   Net PAYE are not of course the only source of tax collections, but represents the tax contributions of the bulk of the population.  The table shows that the largest part of PAYE collections are necessary to finance social welfare outlays even though social welfare outlays are only 28 per cent of total Commonwealth outlays.

Reasons for Increase in Income Support Recipients

The causes of the increase in numbers on pension and benefit have been analysed over the years by a number of researchers including recently by the Social Security Review chaired by Professor Bettina Cass and earlier by the Social Welfare Policy Secretariat.

The reasons presented for the growth in numbers include:-

• demographic changes;
• increased unemployment;  and
• government policy decisions.

During the period of the 1970s and early 1980s, discretionary policy changes included the abolition of the means test for pensioners aged 75 or more (later extended briefly to those aged 70 or more), relaxation of the pensions income test, abolition of the assets test and the introduction of new benefits especially the supporting parents benefit.  These changes all resulted in a substantial increase in numbers of pensioners and beneficiaries.

Demographic trends are also contributing to the growth in numbers, but is not the most important consideration.  The Australian population is continuing to age, a trend common to all other developed countries.  But at this stage of our history, Australia still remains a relatively young country with the median age of the population some 5 years below that in other countries such as the United States.  The greatest impact of ageing will be felt, on present projections, in twenty years tune around the year 2010.

The growth in unemployment which commenced around 1974 and peaked in the early 1980s added substantially to the numbers of unemployment beneficiaries and sole parent pensioners.  There have been other contributing factors including increased marriage break-up resulting in the growth in numbers of sole parents.

Indeed, the growth in numbers of sole parent and widow pensioners since the supporting parents benefit was introduced in 1973 has been particularly large.  Numbers receiving a pension have increased fourfold from 80,000 to 331,000 persons.  (The total population increased by only 22 per cent over this period).

Without examining each pension and benefit separately, reference can be made to similar adverse trends in take-up of other benefits such as invalid pension.  The numbers of invalid pensioners have increased from 1.8 per cent to 2.3 per cent of the eligible age group over the ten years 1977 to 1987.

In summary, the growth in the total number of persons receiving pensions and benefits does not necessarily reflect a massive growth in inequality in the community.  Discretionary policy decisions of government have played a large role in increasing dependency on Commonwealth income support.

Incentives to Stay on Welfare

A social security system which provides a comprehensive safety net, as in Australia, presents major problems for ensuring that pensioners and beneficiaries are encouraged to seek employment and/or make best use of their available financial resources.

One defect in the present system is that the combined impact of the pensions income test and personal income tax (combined with other income tests in the welfare system) creates major poverty traps.  These reduce incentives to seek and take employment when offered.  Alternatively, poverty traps provide strong incentives not to declare income to social security and the tax office (when PAYE tax has not been withheld).

Some examples illustrate the dimensions of the poverty traps.  The unemployment benefit income test allows a free area of $30 per week, subjects the next $40 per week to a 50 per cent reduction and then reduces benefit by 100 per cent of income in excess of $70 per week.  In addition income tax is payable on that income when total taxable income (including unemployment benefit) exceeds the tax threshold at marginal rates of 25.25 per cent or 30.25 per cent (including the medicare levy).

Unemployed persons living in housing commission homes are subject to a further reduction in income of at least 20 per cent of their income because subsidised rents charged are linked automatically to income.  Furthermore assistance provided to children is also subject to an income test in relevant ranges of income.  The combined impact of all these overlapping systems can be tax rates in excess of the additional income received from employment.  There can be no doubting the potential impact on incentives to seek employment.

Similar traps apply also to sole parents who have in addition to face substantial child care costs in seeking employment.  Effective tax and pension income test rates are however lower for sole parents than for unemployment beneficiaries, because the sole parent income test exempts the first $40 of income and reduces the pension by 50 per cent of any income in excess of that amount.  Poverty traps can still generate combined income test and tax rates approaching 100 per cent of income received.

Age and service pensioners are another group where financial incentives encourage the dissipation of their assets (to avoid the harsh impact of the assets test in some circumstances) or over-investment in exempt assets such as the owner occupied home.  Pensioners scared of losing their valuable PHB card in many instances do not invest so as to maximise their income.

The effective pension income test and income tax rates on pensioners generally fall in the range of 62 per cent to 67 per cent, significantly higher than for the highest income groups in the community but lower than for some other pensioner groups.

One major cause for concern about poverty traps is the discouragement of savings for retirement by persons concerned about losing or not getting a pension.  Guaranteeing everyone an adequate age pension as part of a social security safety net may consequently reduce wealth creation through retirement savings that otherwise would have eventuated.

Policy Implications -- General

Complex policy issues are raised by recommendations to further redistribute income or wealth to groups assessed to be in poverty or most in need.  Whilst there is universal agreement about the need to ensure that the poor in society are fully protected by our social security and welfare systems, this does not necessarily require substantial additional social welfare outlays.

In particular, there is an urgent need to ensure that the welfare system provides financial signals and incentives consistent with wealth creation in this country (and the development of new industries).  If the social security system or high taxes to finance that system discourage individuals from saving or seeking to get ahead by gaining additional income, the whole community will suffer in the long run.

There are no easy answers to dealing with current social welfare problems.  Very difficult conceptual questions have to be considered.  For example, the higher the level of income guaranteed to persons in retirement or in the case of unemployment, invalidity and similar untoward events during working life, the lower will be the financial incentives for private savings or insurance to protect against risks.

Lower savings means additional consumption which will reduce aggregate resources available to the community for investment purposes.  Higher levels of consumption will also divert investments to areas such as housing, cars and personal consumption from more productive alternatives.

Not every one accepts the point that a generous social security system can reduce aggregate savings, but Australia is certainly in need of greater savings effort and investment to deal with our massive foreign debt.  Increasing taxes to finance additional social welfare outlays would certainly involve a large risk with our future.

This submission leads to the conclusion that it is by no means clear that inequality in Australia would be reduced if further resources were devoted to direct social security outlays under present arrangements.  Present problem areas include the poverty traps, adverse financial incentives for saving, the efficient use of assets in the system and the very large tax burden already resulting from present outlays.  The higher that taxes and income test rates are, the greater the adverse incentives in the system.

Policy Implications for Various Groups

This submission can not cover all of the problems faced by the present social security and welfare systems.  But if it is agreed, as we believe it is, that addressing inequalities and concentrating welfare support on groups most in need should be given highest priority, the key question to be answered is whether increased social welfare outlays will assist the achievement of those objectives.

For the aged, the approach needed is to ensure that policies are directed towards encouraging the most efficient use of available assets to generate retirement incomes.  This requires addressing the incentives to over-invest in housing and/or dissipate wealth before retirement.  Changes in the present assets test and reducing the combined impact of taxation and the pensions income test are appropriate policy responses, not additional outlays.

Similarly continued participation in the work force for as long as possible should be encouraged, to cope with the impact of Australia's ageing population on the available labour supply in the future.  The United States has for example introduced measures to raise the retirement age in that country.  Yet no similar action has been taken in Australia even though the retirement age for females (at age 60) is five years younger than that for men despite the fact that women live up to eight years longer on average than men.

Sole parents are the group shown by research studies based on poverty lines to be most in need.  The recently introduced family allowance supplement may have changed these results considerably because of the major benefits offered under that package.  However, additional funding may not be the best solution to deal with the financial positions of sole parents.

The best solution is one that would discourage continued reliance on social security benefits, the adequacy of which will in any event be placed under increasing pressure by our ageing population and taxpayer hostility to tax increases.  Looking to the causes of sole parenthood suggests a better approach than increasing outlays.

The Hawke Government is now attempting to place part of the financial burden for supporting sole parents on the former spouse.  This will reduce financial pressures for irretrievable marriage break-up merely to obtain the sole parent pension.  But the new rules will have no impact whatsoever on incentives for marriage break-up, for example in the families of the unemployed where compulsory maintenance payments would not be sought under the new rules.

Separation can thus still generate additional income for an unemployed couple with children.  (This results from the higher rate of pension paid to a sole parent compared to the wife of an unemployment beneficiary).

Dependence on the state for support can also be reduced by retraining and providing incentives for part and full tune work.  But while the income test and the formulae for determining rents of subsidised public housing continue to apply, sole parents will continue to have little financial incentive to get ahead compared with remaining on the pension.

The preferred policy both for the community and the individual in the long run may accordingly be to provide less rather than more financial assistance to sole parents.  Concentrating assistance only on emergency situations or on cases where young children are involved and child care expenses are highest could provide a viable alternative to present arrangements.  The important message is that sole parents pension rules should recognise that the community can not afford to give continuing income support to individuals merely because they have a child.

Indeed the recent government action to deprive sole parents of their pension when their youngest child turns 16 (compared with the previous age 25 for student children) represents belated recognition of this point.

The above and similar suggestions may impose short run hardships for individuals which would need to be dealt with by special benefits and other short term action depending upon the extent of transitional problems.

One final word.  Generating job opportunities is the best means of reducing the demands on the social security safety net.  Making the social security system and its incentives contribute to the needed structural changes in investment and the economy would help encourage employment growth.  The most effective way of reducing inequality is to ensure that employment opportunities are maximised.