This page created:19 April 1999

Last modified: 19 April 1999

Maintained by:

Review of:

Living Decently: Material Wellbeing in Australia

Peter Travers & Sue Richardson

Oxford University Press, Oxford, 1993

ISBN 0 19 553360 7

x + 246

Is poverty in Australia getting worse ? Much contemporary discussion suggests that it is. Concern has been raised over increased inequality, the 'disappearing middle class', and the rise of 'urban ghettos' in Australia. Travers and Richardson survey most of these issues, coming to fairly optimistic conclusions. They conclude (p 200)

'The outcome of this approach is a picture of Australian society that is quite positive. We ahve found little evidence, for instance, to suggest that Disraeli's portrayal of Britain in 1845 as 'Two Nations' is an apt description of Australia today. Nor do we find evidence of a more modest division of society into 'insiders' and 'outsiders''

However, Travers and Richardson suggest, the achievement of this positive outcome in Australia has depended on preconditions that are now under threat: full employment, a 'good enough' social welfare system, very high levels of home ownership, and high quality government provision of specific goods and services.

In this relatively short work, Travers and Richardson cover issues including social mobility, trends in inequality over time, international comparisons and the relationship between material well-being and human well-being. The discussion of these issues is useful and well-balanced. The basically optimistic conclusions drawn by Travers and Richardson seem justified, especially when it is borne in mind that their analysis stops in 1987, before the recession and the impact of labour market 'reform'. As they concede, it seems unlikely that an analysis undertaken today would be quite so favourable.

In this short review, I will concentrate on the issues where the book makes a conceptual break with the majority of the previous literature, namely the use of 'full income' concepts derived from Becker's theory of time allocation and the treatment of the distinction between relative and absolute poverty.

The most notable innovation in the book is the use of the concept of full income. This is a monetary measure incorporating after-tax earned income, pensions and benefits, the value of leisure time, imputed rent from home ownership, the service flow from consumer durables, and the net benefit from indirect taxes, government provision of goods and services and transfers between households. The key finding is that correlatiosn between the different components of full income are so low that (p37) inequalities in the different spheres of material well-being tend to offset each other (emphasis in original).

This proposition is correct in its own terms. But is it sufficiently surprising to justify the emphasis it is given? In examining this point is useful to divide full income into three classes. The first consists of after-tax earned income, pensions and benefits and the value of leisure time. Travers and Richardson compute the value of leisure time by assuming that all individuals have 50 hours per week to divide between work and leisure. For employed people, those with home duties and the retired (who total over 90 per cent of the sample), leisure time is valued at either the actual wage or the wage imputed from an earnings function. Leisure time is valued at zero for the unemployed and at 25 per cent of average earnings for those over 65. In effect, this procedure means that instead of examining income inequality, we are examining inequality in wages or earning capacity (the unemployed may be fitted into this picture if the value of their benefits is treated as the wage for a 50-hour week). In a neoclassical world where everyone is on their labour supply curve, earned incomes will be more unequal than wages if and only if the supply curve of labour slopes upwards. The same will be true in a Keynesian world, provided in addition that low-wage workers are more vulnerable to involuntary unemployment.

The second component of full income consists of imputed rent and service flows from assets. On a life-cycle perspective, we would expect these flows to be highest in the retirement period and to offset the inequality in earned income over the life-cycle. Within any age cohort, however, this component of income should be highly correlated with wages or earning capacity. Finally, it is encouraging, but not surprising that government expenditure and inter-household transfers are modestly equalising.

The analysis of full income undertaken by Travers and Richardson is theoretically sound. From a life-cycle perspective, unless financial bequests are important, the only relevant issue is inequality in earning capacity. But the result that different components of full income tend to offset each other is not so much an empirical finding about the nature of income distribution in Australia in 1987 as a necessary consequence of the theoretical framework. The life-cycle model predicts that earned income in any given year will be more unequal than earning capacity. This prediction is supported by the data. Travers and Richardson find that the Gini coefficient for full income, at 0.23, is smaller than the coefficient for any of the components of full income. But does a Gini coefficient of 0.23 represent a little wage inequality or a lot ? In the absence of more detailed international and cross-section evidence, it is hard to say much.

The treatment of poverty measurement by Travers and Richardson is a very promising contribution to the resolution of a perplexing problem. Most people would say that poverty is a problem in Australia, but not to the extent that it is in India, or that it was in Australia in the 1930s or the 1890s. Neither relative nor absolute measures of poverty seem capable of bearing out these intuitions. As Travers and Richardson rightly say, relative measures of poverty simply equate poverty and inequality. In the absence of any clear long-term trend in inequality, there is no reason to suppose that there is any less relative poverty today than a hundred years ago. On the other hand, virtually everyone in Australia now has an income greater, in real terms, than the forty-two shillings a week Mr. Justice Higgins regarded as sufficient to keep a working class family of four in 'frugal comfort' in 1907. Thus, the use of an absolute poverty measure in a context of growing real incomes leads rapidly to the conclusion that poverty has been abolished.

The approach favoured by Travers and Richardson combines direct judgements about the specific requirements for 'living decently' with distributional measures based on the full income measure. They find consistent evidence that households in the bottom decile of the full income distribution are much more likely than others to be deprived on such measures as ownership of a car and ability to raise $1500 in an emergency. Most such households could therefore be regarded as poor. Travers and Richardson do not, however, favour the use of a 'poverty line' based on full income. Rather, they argue, poverty is a 'thick' ethical concept that cannot usefully be reduced to a position on a univariate scale. Much more remains to be said on this topic.

Travers and Richardson raise more questions than they answer. But, with the complex problems of poverty and inequality this will be true of any significant contribution. Living Decently is an important book that should be read by everyone concerned with these vital issues.

Go back to John Quiggin's home page

Read more book reviews by John Quiggin