Defining poverty

Most commentators will tell you that “poverty” should be defined and measured relative to the living standards of specific societies. This means that “poverty” in Africa is very different from “poverty” in Australia. To be poor in Africa means you are starving: to be poor in Australia means you cannot afford to eat out at a restaurant.

There is an obvious problem, however, in defining “poverty” in this way, for it swiftly becomes indistinguishable from “inequality”. Conceptually, the two terms are quite distinct: poverty means not having enough; inequality means not having as much as somebody else does. But once “poverty” is defined relatively, this distinction gets horribly blurred. People who cannot afford to eat out at a restaurant, for example, clearly have a lower living standard than most of their contemporaries, but does this mean they do not have enough money to live on? Doesn’t there come a point where the general level of affluence in a society rises to such an extent that the language of poverty no longer applies to those at the bottom?

Logically there are two basic strategies for combating poverty. One, the “capitalist” strategy, is to allow economic growth to improve everybody’s living standards (raising all boats). The other, the “socialist” strategy, is to take resources away from more prosperous people and transfer them to those with less. The disturbing thing about a relativistic definition of poverty is that it rules out the first strategy a priori, for no matter how successful we might be in improving everybody’s living standards, the number of “poor” people will remain exactly the same for as long as relative income shares remain the same.

This is why socialist academics and left-wing activists have universally adopted a relativistic definition of poverty, for it leads directly to the income redistribution policies they favour. Once locked into such a definition, anyone who wants to reduce poverty is obliged to sign up to a socialist political agenda.

Defining “poverty” relatively also plays havoc with international poverty comparisons. The United States, for example, has a higher average living standard than most European countries, but it is also a lot less equal in terms of income distribution. Because incomes are more spread out in the US, “relative poverty” researchers will tell you “poverty” is a bigger problem there than in Europe (for there are more people on relatively low incomes). But all this is really saying is that the US is more unequal than Europe - the income distribution is more spread out, so there is a longer “tail” which gets defined as “poverty”.

If the US economy grows at a much faster rate than that of Europe, and all Americans improve their living standards as a result, your average poverty researcher is unlikely to be impressed, for as the standard of living rises, so the definition of “poverty” has to be revised upwards, which means researchers will identify just as many “poor” people as before. If, on the other hand, the Europeans compress the gap between top and bottom, poverty professionals will be gushing in their plaudits even if the economy is stagnating. This is because the number of people in the “poverty tail” will have fallen, even though everyone is falling further behind the Americans.

These paradoxes also arise in poverty debates in Australia. Wedded to relativistic definitions of “poverty”, our welfare organisations and social policy intellectuals insist that “poverty” in this country is widespread. According to the Australian Council of Social Service (ACOSS), for example, nearly a quarter of us are living in poverty, and a recent Senate poverty inquiry (which became an uncritical mouthpiece for ACOSS and other members of the welfare lobby) suggested that as many as 3.5 million Australians were poor.

These poverty professionals also insist that, despite a surge in average living standards, “poverty” in Australia is getting worse. What they really mean, however, is that income inequality is increasing. Average disposable incomes in Australia rose by 14 per cent in just eight years from 1995 to 2003, but those at the bottom of the distribution (where most people are on welfare benefits and very few earn any money of their own) improved their average incomes by “only” 12 per cent. Despite the fact that everyone has been getting much better off, the poverty lobby thinks things have been getting worse, for their focus is on the inequality question (how much one group has in comparison with another) rather than on the poverty question (whether people at the bottom are improving their living standards).

The capitalist strategy for combating poverty obviously works. It is remarkable that our capitalist economy can raise the living standards of the least advantaged people by one-eighth in real terms in just eight years. Poverty professionals are generally scathing about the so-called “trickle-down” theory of growth, which holds that increased prosperity accrues first to the richest people but later spreads throughout a society, but the figures suggest rich and poor alike have benefited hugely from Australia’s booming capitalist economy. Yet still our poverty activists and intellectuals insist that poverty is getting worse.

The Alice-in-Wonderland quality of claims like these has been heightened by the refusal of welfare groups and social policy academics to take seriously the severe problems in the income statistics they are using.

Back in 2002, a team at the Centre for Independent Studies highlighted some obvious flaws in the ABS income survey data. We pointed out that many of those appearing in the bottom income decile (the core of the “poor”) claim to have impossibly low (sometimes zero or even negative) incomes, far below the minimum guaranteed them by the income support system. We also asked why, on average, they are spending more than twice what they claim to be receiving. And we flagged the fact the income people say they receive from welfare benefits accounts for only 70 per cent of the value of welfare benefits that are actually paid out. We concluded that many of those at the bottom of the distribution of reported incomes are getting more than they say in survey answers, and the poverty statistics are therefore hopelessly unreliable.

We were fiercely criticised by the welfare lobby for exposing these problems, but the ABS itself later confirmed that its income data are indeed so inaccurate at the bottom end that they should not be used to make poverty estimates.

Yet the welfare groups and academics press on as if nothing had happened. The Senate Committee’s poverty estimates, for example, totally disregarded the problems in the income data despite being alerted to them by the CIS submission. Similarly, a recent report by the St Vincent de Paul Society claims that nearly a quarter of the population is living in households with a weekly income below $400, but this takes the income data at face value as if the ABS had never issued its warnings. And my namesake at the Social Policy Research Centre (“the other Peter Saunders”) has just produced a book with the stated aim of “restoring credibility to Australian poverty research” in which he devotes just two sentences out of 155 pages to the problems in the income data. Like the three monkeys, the Senate Committee, the Vinnies researchers and the “other” Peter Saunders apparently hear, see and speak nothing about this core problem. Presumably they hope we will all just forget about it.

Given the confusion around the concept of poverty, and the inadequacy of the income data, I sometimes wonder whether it is worth persevering with the notion of “poverty” applied to the Australian context. However, there clearly are some people who live lives of material (and often cultural and spiritual) deprivation in this country, for whom the term may still be appropriate. But if we insist on retaining it, two points need to be made very clear.

The first is that the number of “poor” people is nowhere near the sorts of figures bandied around by the welfare lobby - I would estimate no more than 5 per cent of the population, probably less. Research on various indicators of “hardship” (such as being unable to afford heating at home, or having to pawn or sell something to raise cash) suggests around 3 per cent of the population are experiencing difficulties like these. Longitudinal research on low incomes (less than half the median) finds fewer than 4 per cent remain under this level for three successive years (for most people on low incomes are in transit to something better). For these and other reasons, I think the problem of serious deprivation is relatively small, and this makes it manageable.

The second point is that “poverty” or “hardship” is not simply, or even primarily, a matter of low incomes. Research on “hardship” has found that the great majority of households on the lowest incomes report no hardship, and that half of the households that do report hardship are on incomes above the conventional half-median “poverty line”. The only explanation is that most people on low incomes manage to “get by”, and that those who suffer do so because of the way they handle their money and organise their lives. Although the poverty lobbyists refuse to recognise it, there is a significant “behavioural” component to what they call “poverty”.

This is borne out by looking at the level of welfare benefits paid to people with no other source of income. For many years the Melbourne Institute has been updating the (extremely generous) “Henderson poverty line” and comparing it with the income available from welfare benefits. With the sole exception of a single person on unemployment allowance, every type of household on every type of benefit is entitled to an income that puts it above the Henderson line. If there is a deprivation problem, therefore, it is not one that is going to be solved by increasing welfare benefits.

So what should be done? Where possible, we should aim to get people into paid work, for everyone agrees that a full-time wage is the best guarantee against “poverty”. This means enabling people to work (for example, by expanding child care, providing training opportunities, and so on) as well as requiring them to do so (by tightening up the welfare eligibility rules). It also means stepping up intensive assistance for the “hard cases”, just as the Americans have done. What we must not do is what the welfare lobby keeps demanding - increase welfare spending.

One of the greatest ironies of the poverty debate in this country is that the activists and academics who keep insisting that poverty is a huge and growing problem also argue for more welfare spending to solve it. But if welfare was the solution, the “poverty” of which they complain would have disappeared long ago, for government spending on income support has grown from 3 per cent to 8 per cent of GDP in the last 40 years. Whatever else we take from this debate, let’s all agree to stop pursuing policies that have patently failed to work.