Dialogue: How much people spend better gauge of poverty

Poverty studies have been barking up the wrong tree and diverting attention from the more serious underlying problem of welfare dependency, writes DAVID GREEN*.

To reach valid conclusions about poverty, such as the poor are getting poorer or there are more people below the poverty line, is no simple matter. The choice of poverty measure might well determine the result obtained.

Poverty researchers in New Zealand have generally used cash income as the measure of living standards. Occasional attempts to calculate a broader measure of material standards of living have been made.

Typically, poverty studies based on an income measure try to determine the number of people below a poverty line, defined as a percentage of median income. It is now widely acknowledged among recognised researchers that such measures of poverty are deficient. Any such benchmark is inevitably arbitrary.

If, say, 50 per cent of the median income is chosen, why not choose 40 per cent or 60 per cent? Moreover, such a technique almost guarantees that some proportion of the population will always be "in poverty", regardless of increases in their overall living standards.

Martin Wolf, the associate editor of the Financial Times in London, has highlighted some of the strange implications of this approach to measuring poverty.

"If, for example, the incomes of those with relatively low standards of living rise by a smaller proportion than the overall average, numbers in relative poverty increase.

"If, in contrast, the decline in incomes of those on well-below-average incomes is smaller than average, relative poverty declines.

"In the first situation poverty increases, even though everybody is better off. In the second, poverty falls, even though everybody is worse off. This seems crazy."

Annual income is also an imperfect measure of living standards. People's income often fluctuates from year to year, and some are able to draw on savings when their income is low.

Many people move from a low-income decile to a higher one over time - the population "in poverty" is not static. Cash income does not measure the value of owner-occupied housing, and some income is not reported.

For these reasons, what people actually spend is a better measure of their material welfare than their income. The income measure overstates poverty.

Eurostat, the European statistical agency, says that household expenditure "better reflects the availability of both declared and undeclared resources of low-income groups" and is a more satisfactory indicator of permanent income than income declared at any one time.

There can be large differences in results between the income and expenditure measures of poverty.

New Zealand researchers have found that only 45.6 per cent of households in the lowest-income decile are also in the lowest-expenditure decile, and that 10 per cent of households in the bottom-income decile are in the top three expenditure deciles.

Even expenditure does not capture living standards fully. It leaves out goods and services in kind that are provided free to the consumer, such as health and education services.

Not so long ago it was common for anti-poverty campaigners to insist on the importance of the so-called social wage, and trade unions made raising the social wage one of their central demands.

British estimates for 1994-95 showed that low-income people received benefits in kind (mainly health and education) worth on average about $NZ12,000 a year. Leaving out the value of these benefits ignores over 47 per cent of their final income.

A consumption measure, which includes goods and services provided in kind, is the most comprehensive and satisfactory measure of living standards.

No studies of poverty based on a consumption measure have been done in NZ in recent years. The approach developed by the authors of the 1990 Planning Council report, Who Gets What? The Distribution of Income and Wealth in New Zealand, and the Department of Statistics in The Fiscal Impact on Income Distribution 1987/88, come closest to that measure, although the allocation of some Government spending and taxes to households in such studies is inevitably speculative.

Poverty is a serious social problem and its alleviation should be a major goal of Governments, ranking above efforts to engineer income distribution.

Poverty alleviation is essentially dependent on policies for wealth creation and economic growth. Policies of redistribution can have only limited impact, particularly in today's open international environment.

There is also a risk that a narrow focus on poverty, reflected in poverty studies based on income and even in more soundly based studies, diverts attention from the more serious underlying problem - that of welfare dependency and the incentives it creates, not least to weaken personal and family responsibility.

Those who treat poverty as always and everywhere the result of sheer misfortune or a lack of cash miss important dimensions of the problem.

The key issue is how society can best assist its least fortunate members without creating welfare dependency.

* David Green is the director of Civitas (the Institute for the Study of Civil Society) in London. He is the author of Poverty and Benefit Dependency, published this month by the Business Roundtable.