The Politics of Envy
by Helen Hughes

To read the correspondence regarding this article, please click here

Contrary to widespread belief,the real value of low incomes in Australia has risen while the proportion of people living below the poverty line has fallen.

For a rational economist, poverty is intolerable, particularly in high income countries. So is the concentration of wealth and power in the hands of a few in a corrupt society. But equality of incomes, in contrast to equality of opportunity, is neither desirable nor feasible. Millions of people have been impoverished, imprisoned and killed in the name of equality in the past century. The distribution of income across society only matters if it is not in keeping with a nation's un-derstanding of its social values.

According to a Newspoll survey in January 2000, 64% of respondents strongly agreed, and 19% partly agreed, that 'the poor are getting poorer'.1 This perception is wrong. Since the 1970s, the proportion of people in poverty has fallen and the real value of low incomes has risen more than that of middle and high incomes. The rich have not benefited at the expense of the poor. The politics of envy are unfortunately sowing disharmony by distorting the facts about poverty and income distribution in Australia.

Measuring poverty

Poverty has an absolute and relative context. The concern in Australia, however, is not with absolute minimum incomes on which people can survive, but with minimum incomes on which people can 'live decently' 2 -that is, with relative poverty. Ronald Henderson laid the foundations of research on poverty in Australia in the 1960s. A Commission of Inquiry into Poverty in 1975 adopted a Henderson Poverty Line as the demarcation income level for poverty in the Australian context.3 Any measure of poverty is necessarily subjective and imperfect, and hence subject to continuing debate. Some consider the Henderson Poverty Line too generous and so likely to encourage welfare dependency, while for others it is not high enough to enable people to 'live decently'.

The Henderson Poverty Line has, nevertheless, become accepted as a measure of poverty in Australia. It is updated by per capita disposable household income to incorporate inflation and reflect rising standards of living.4 The updates have raised it for a couple with two children from $62.70 per week in the September quarter of 1973 to $467 per week in the December quarter of 2000 (see table 1 overleaf ).

With bipartisan political commitment to adjusting welfare payments (including unemployment, single parent, old age and disabled pensions, housing and other benefits) to meet the needs of persons and families in poverty, welfare payments have been raised close to, or above, the Henderson Poverty Line levels (see table 1).

Depending on welfare or on low wages means living below standards that most Australians enjoy, but many single people and families manage: children attend school, but help out with after school and holiday jobs. Independent young people are able to make ends meet by living in group households. As children grow older, mothers can work, and some low income families raise their living standards appreciably.

We know that some Australians are, nevertheless, not 'living decently'. Charitable organisations assist some persons and households that need help to feed and clothe themselves. Some people are living in the streets.5

The social aspects of poverty are more insidious than mere penury. The incidence of illness, alcohol addiction and smoking is higher in low income families than in medium to high income families. Heavy drinking often leads to domestic violence. Gambling addiction appears to be at least as high, if not higher, in low income families than for Australians as a whole. Drug addiction also appears to be more prevalent than in higher income families and can lead to crime. Family violence and break-up is a danger, with dysfunctional families making life difficult for children and young adults.

The patterns of behaviour that cause individuals and families to fall below levels of income represented by welfare assistance have been summed up as behavioural poverty.6 This is not to suggest that only people on low incomes behave badly. Rather, bad behaviour that at higher incomes levels can go unnoticed can spell disaster more quickly and permanently for low income households.

Misleading statistics

Recent analyses of poverty in Australia are based on the Australian Bureau of Statistics (ABS) survey data on incomes and expenditures.7 The ABS data suggest that in 1999-2000 the income of 14% of couples with children, and 24% of single parents, fell below $499 and $299 a week respectively.8 If these figures were correct, they would mean that a high proportion of potential recipients do not access the welfare payments to which they are entitled.

Unfortunately, the ABS has not organised its surveys according to Henderson Poverty Line sources or welfare payments levels. Nor are the household income and expenditure data upon which these figures are based consistent. After tax household income estimates are substantially below expenditure estimates, suggesting that survey respondents, as in other countries, do not report 'black' incomes. Tax collection data also indicate that surveys of expenditure estimates are themselves too low because they do not accord with spending on alcohol, tobacco, drugs and gambling, as reflected in taxes.

Both ABS household income and expenditure survey estimates are below aggregate national income and expenditure estimates, although these too are known not to include much of the 'black' economy. Christopher Bajada estimated that informal income could equal 15% of GDP.9 This would be close to 20% of reported private income.10

The extent of poverty

The 1975 Commission of Inquiry into Poverty surveyed and attempted to quantify the number of people and households living below Henderson's poverty line in 1972-73. It came up with a range of figures, depending on household and housing characteristics. For adult income units the proportion in poverty was 10.2% before housing was taken into account and 6.7% after housing was taken into account. Home ownership, access to public housing and the availability of reasonable rental accommodation were important factors in mitigating poverty. For all Australians, 8.2% were deemed poor before housing and 6.4% after housing was taken into account.

Using ABS income survey data discussed above in its STINMOD model, the National Centre for Social and Economic Modelling has argued that poverty had increased by 1995, putting the proportion of all persons in poverty at 11.8% before housing and 9.8% after housing.11 Anthony King, using the same model, came to 14.4% of all persons in poverty before housing and 11.2% after housing in 1996. 12 These results exaggerate the extent of poverty in Australia. They do not mesh with other evidence. Despite all the care that has undoubtedly been taken, the STINMOD model appears to exaggerate the income data bias against low income groups.

The best estimate

Australians living in poverty include those who are disabled or disoriented. Single elderly people remain particularly at risk. So do some young people. Long term welfare recipients become welfare dependent and susceptible to the problems of behavioural poverty. Welfare dependency and poverty is high among some indigenous Australians. Some migrants do not acquire sufficient English to be able to move into the labour force and progress within it. Some poverty is inherited, but behavioural poverty is not only a danger for the poor. Becoming the head of a single parent household has pushed some women from middle to higher income groups below the poverty line: from shirt sleeves to shirt sleeves in three generations can lead to poverty in the fourth.

Apart from small indigenous communities in remote Australia and on the fringes of some country towns, most poor people live in cities where they are concentrated in old centres and on the outskirts. Low incomes are closely linked to poor education, low workforce participation, and high unemployment, particularly for women. Leafy suburbs, in contrast, typically enjoy dual household incomes and high youth employment.13

But the situation is not entirely bleak. The Henderson Poverty Line is now 35% higher in real income terms than in 1972-73. Living standards have improved substantially for old age pensioners who formed a high proportion of poor households in the early 1970s. Pensions have risen appreciably and a smaller proportion of the aged are dependent on pensions.14 Other welfare payments have been greatly extended and increased to meet Henderson Poverty Line levels. Non-cash payments for education and health (social wages), have increased considerably.15 The 'black' economy has grown markedly, adding to the low incomes that are reported.

Some people in low skilled jobs struggle to make ends meet, but increased employment flexibility, part-time and casual jobs have benefited many low income households. Mother and father can split the time at work and at home so that children can be nurtured. Since 1982 family payments and rent assistance have been extended to low income working families with children. Tax incidence on low income families was reduced in June 2000.

The principal cause of poverty is now unemployment. Numbers seeking work have risen since the 1970s, peaking at nearly 13% in 1993, and after falling to 5.9% in June 2000, were again 7% in April 2001. To this must be added another 2 or 3 percentage points to account for unemployed workers only marginally attached to the labour force and those no longer looking for work, so that the total out of work adds up to 9 to 10%. Unemployment and disability payments have lifted the unemployed above the poverty line, but not having a regular job remains a major cause of poverty.

On balance, taking into account the higher levels of unemployment and the emergence of some low skilled, low paid jobs, but also the positive trends in the reduction of poverty among old people and the increases in welfare payments and social wages, the proportions of people living in poverty have almost certainly fallen since the Commission of Inquiry into Poverty's findings in 1975 of 8.2% in poverty before housing and 6.2% after housing. Correcting for the underestimation of household income, I estimate that no more than 5% of the population-that is, well under one million people-are now living below the poverty line. The proportion is of course much higher for groups particularly at risk. But the level of poverty in Australia is well below estimates based on ABS survey income data that understate low incomes.

Income distribution and the politics of envy

International comparisons consistently place Australia at the top of living standards in spite of its mere 20th ranking in world per capita GDP in purchasing power parity terms ( and 27th place at official exchange rates).16 Standards of living have been rising steadily. Climate, space and cultural openness help to create reasonably comfortable lifestyles for most people, even those earning modest incomes. Women's careers and earnings have prospered. People live longer, have time for several careers and enjoy much more recreation than their parents or grandparents. This does not mean that most Australians would not like to have higher incomes. Continuing GDP growth-that is, the size of the national cake-remains as important as the size of the slices the various income groups enjoy. Political interest groups, however, have sought to deflect the debate from the policies that are needed to stimulate economic growth to claims that growth worsens income distribution.

A democratic society determines the income distribution it wants by voting for policies it desires. An equal distribution, with each quintile having the same proportion of income (that is a Gini coeffidient of 0.0)17, is clearly nonsensical. People have different capacities and diffferent choices between income and leisure, partiulcarly at different times in their lives. Income differentials are essential to high productivity and growth, just as income mobility is vital for a dynamic and innovative high growth society. Entrepreneurs- risk takers-are better paid than university professors. But the wealthy do not take away from the well being of middle and lower income groups. As The Economist recently pointed out,18 wealth creates many jobs and some of the newly rich use their wealth for public benefit. Provided mean spirited envy is held at bay, everyone gains.

Income differentials

It is widely agreed that low incomes have risen more rapidly than middle and high incomes in Australia from the 1970s, with accelerating rises from the mid-1990s.19 Peter Saunders, in the 2001 Year Book of Australia, notes that income inequality declined between 1915 and 1969, with much of the reduction taking place after the 1930s depression. Saunders also agrees that there was further improvement to 1981. 20 Trends in the 1980s and 1990s are in dispute.

The data for income distribution among Australian families show a Gini coefficient increase from 0.33 in 1968-69 to 0.39 in 1997-98. Saunders hence deduces that income distribution worsened in the 1980s and 1990s. But the extent of inequality is exaggerated by several percentage points by the non-reportage of 'black' income. Correcting for a 20% underestimation of total private income would raise the relative level of the bottom quintile of households more than that of the top quintile, as unreported earnings typically add up to a greater proportion of income in the bottom quintiles. It seems highly likely that 'black' income has grown in the 1980s and 1990s so that the failure to include it has increased the income data bias against low incomes

Saunders has also presented the distribution of the weekly income of full-time workers. The Gini coefficient based on full-time weekly earnings fell from 0.35 in 1986 to 0.34 in 1997-98 (for equivalent disposable income of all income units). Margins of error in such calculations are inevitable, but this strongly implies stability. Weekly earnings, in any case, are a poor indicator of income distribution and should not be used as a proxy for it. They do not take into account tax, welfare or social wages. They do not indicate income mobility, notably from low earnings for young people to later peak earnings. Differences in weekly earnings must be expected to rise as Australia becomes a 'clever' or 'knowledge' country and continues to grow. Hopefully the proportion of unskilled workers will continue to decline.

Expenditure differentials

Data on reported incomes are unreliable; in particular, because those earning lower reported incomes often spend a lot more than they claim to be earning. Expenditure differentials are therefore a better indication of well being. Economists Garry Barrett, Thomas Crossley and Christopher Worswick use household consumption rather than household income to study distribution. They find lower Gini coefficients and less narrowing for 1975 to 1993 (the latest data available) than Saunders.

This indicates that low income households have been able to smooth out income fluctuations. They maintain their expenditure in bad times by borrowing and with help from relatives. Income tax changes have benefited low income recipients. While there was a slight decline in consumption equality between 1975 and 1993, the bottom 10% experienced 'relatively minor deterioration in their consumption shares.21

Income and expenditure distribution is also affected by economic cycles. Sustained growth and falling unemployment in the second half of the 1990s almost certainly made household expenditure more equal, offsetting the minor rise in expenditure inequality found to 1993. Distribution of consumption since the 1970s thus seems to have been stable.

Policy conclusions

A strong community social welfare sector exists in Australia with a vested interest in exaggerating the extent of poverty and income inequality. Some 10,000 to 11,000 welfare organisations exist. Fifty large ones dominate. More than 50% of the income of charities is contributed by the Government. The combined annual expenditure of charities was nearly $4.8 billion in 1993-94. They attracted $400 million in tax concessions. The industry employed about 100,000 workers and relied on many thousands of volunteers.22 The Australian Council of Social Services (ACOSS) thus has a large constituency. It is joined in advocacy by many of its members. In their pleas for support, the focus has sometimes been on atypical cases at the cost of verisimilitude. 'Crying wolf ' is unlikely to be helpful in the long run. With the growth of very large social service organisations, moreover, transparency and service quality are needed to ensure that charitable organisations do not replicate the bureaucratic behaviour of the government agencies they have replaced. Above all, poverty needs to be reduced so that their activities are not needed on today's large scale.

The principal concern for policy is to increase further employment. This would reduce poverty directly and erode behavioural poverty. Apart from requiring the continuation of sound macroeconomic policies, this needs further microeconomic reforms, notably in labour markets, to remove current obstacles to employment. Inflexible industrial relations continue to be the main barrier to creating more jobs. Reversing the gains achieved through part-time and contract work particularly affects low income earners. The unreasonable administration of unfair dismissal and equal opportunity principles plagues small business, which employs nearly 40% of the labour force. Inept tax administration and regulation also discourages small businesses. State payroll taxes continue to favour capital against jobs in spite of the introduction of the GST.

On the labour supply side, attention has been turned to reducing dependence on welfare and encouraging workers to find jobs by reducing taxes on low incomes and proposing the introduction of tax credits.23 Work for the Dole and funding for study instead of unemployment increase experience of work and willingness to enter the labour force.

The quality of education and training are key, notably at the primary, secondary and post-secondary levels. Much could be done to endow children from discouraging backgrounds with equal opportunities by supporting before and after school homework and tutoring programmes, helping children from low income areas to access sport and cultural recreation, and by scholarships to stimulate achievement. The continuing expansion of apprenticeship and other training schemes, with more entry for girls into technical trades, will increase labour force participation and raise lifetime earnings.

Worldwide experience indicates that if basic economic and social policies are not conducive to high employment, merely throwing money at the problem will not work. Welfare payments may enable people to 'live decently', but on a continuing basis are counterproductive by leading to welfare dependency.

Endnotes

1 Institute of Public Affairs, Facts (April 2001), 7.

2 P. Travers and S. Richardson, Living Decently: Material Well Being in Australia (Melbourne: Oxford University Press, 1993).

3 Commonwealth of Australia, Commission of Inquiry into Poverty, Poverty in Australia: First Main Report, 2 vols (Canberra: The Government Printer of Australia, 1975).

4 J. Humphries, 'Reforming Wages and Welfare Policy: Six Advantages of a Negative Income Tax', Policy 17: 1 (Autumn 2001), 19-22, suggests that the Henderson Poverty Line should only be adjusted for inflation to prevent welfare dependency and to ensure the eradication of poverty in Australia. But poverty can also be eradicated if the proportion of people under the more generous Henderson Poverty Line continues to be reduced. This contrasts with taking a fixed share of income as a poverty line; that means that some people will always be poor!

5 The Salvation Army estimated that there were about 100,000 homeless people in Australia, quoting gambling addiction as an important cause of homelessness. The Australian (15 June 2001).

6 Lucy Sullivan distinguishes between behavioural versus financial poverty, arguing that the income levels represented by welfare payments would be more than adequate to live on if they were managed well. L. Sullivan, Behavioural Poverty (Sydney: The Centre for Independent Studies, 2000), 4-8.

7 Australian Bureau of Statistics, Income Distribution, (ABS 6523.0), annual to 1999-2000, is based on an annual Survey of Income and Housing Costs, and Household Expenditure, Australia: The Effects of Government Benefits and Taxes on Household Income ABS 6537.0) quinquennial to 1998-9, is based on quinquennial household expenditure surveys.

8 Income Distribution, 1999-2000,12.

9 C. Bajada, 'Estimates of the underground economy in Australia', Economic Record 75: 231 (2000), 369-384.

10 G. S. Dorrance, 'Estimates of income and expenditure distribution in Australia', mimeographed (July 2001).

11 National Centre for Social and Economic Modelling, 'Poverty in Australia', Income Distribution Report 4 (Canberra: University of Canberra, 1996).

12 A. King, 'Income Poverty Since the Early 1970s', in ed. R. Fincher and J. Nieuwenhuysen, Australian Poverty: Then and Now (Melbourne: Melbourne University Press, 1998).

13 R. G. Gregory and B. Hunter, 'The Macro Economy and the Growth of Ghettos and Urban Poverty in Australia', Discussion Paper 325 (Canberra: Centre for Economic Policy Research, The Australian National University, 1995).

14 National Centre for Social and Economic Modelling, 'Changes in the Incomes of the Aged', Income Distribution Report 9 (Canberra: University of Canberra, 1998).

15 National Centre for Social and Economic Modelling, 'Impact of the Social Wage', Income Distribution Report 2 (Canberra: University of Canberra, 1995).

16 World Bank, World Development Indicators 2001 (Washington DC, 2001).

17 The Gini coefficient is a summary measure of income distribution, ranging from 0.0 for absolutely equal distribution to 0 .45 and upward for relatively unequal distribution. It has to be used with great caution, because different patterns of income distribution can have similar Gini coefficients. Data have wide margins of error and measurement differences over time and between countries may create distortions. It is particularly important to note that poverty is not a function of income distribution. Poverty could be eliminated even though income distribution widened.

18 The Economist (16 June 2001).

19 Chamber of Commerce and Industry of Western Australia, In Support of Free Enterprise (Perth: 2001).

20 P. Saunders, 'Household Income and Its distribution', in D. Trewin, Australian Statistician, 2001 Yearbook of Australia, (Canberra: Commonwealth of Australia, 2001).

21 G. F .Barrett, T.F. Crossley and C. Worswick, 'Consumption and Income Inequality in Australia', The Economic Record 76: 233 (2000), 116-138.

22 Commonwealth of Australia, Industry Commission, Charitable Organizations in Australia, Report No 45 (1995).

23 T. Abbott, 'The Welfare Revolution',The Age (14 June 2001).

Box

a B. Bradbury and M. Jantti, Child Poverty Across Industrialized Nations, Innocenti Occasional Papers, Economic and Social Policy Series (Florence: UNICEF, 1999).

b 17 June 2000.

c World Bank, World Development Indicators 2001 (Washington, D.C.: 2001).

d National Centre for Social and Economic Modelling, Income Distribution Report 10 (Canberra: University of Canberra, 1999).

Helen Hughes is Emeritus Professor of Economics,The Australian National University,and Senior Fellow at The Centre for Independent Studies.

Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)