Abstract
In this chapter recent Australian experience of poverty is explored with new measurements of poverty using the deprivation weighted index described in Chapter 4.1 The new measurements concern the most recent years for which unit record data are available, 1981–82, 1985–86 and 1989–90. The main tools for these measurements are the poverty indexes (the head-count index, the income gap index and the deprivation weighted poverty index) and the method of interpreting these indexes, described in Chapter 4. The indexes use equivalent disposable income as the criteria for ranking groups in poverty. Jenkins and Lambert (1993b) present a dominance test, which, if it succeeds, means the direction of change is robust to the choice of poverty indexes of the poverty gap class, although the magnitudes of the differences may vary. The first section outlines the main assumptions and definitions used in applying the poverty indexes and the data from which the measurements are made. In the second section results are presented for 1981–82, 1985–86 and 1989–90 concentrating on trends over the three years. In the third section results are presented for 1989–90, the latest year for which there is data, concentrating on differences between groups. In the final section of the chapter, two key assumptions used to calculate poverty lines, the relativities between groups (equivalence scales) and the method of updating are varied and the effects on poverty measurement noted.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Reference
While the limitations of the head-count index of poverty had been known for some time and particularly since Sen’s (1976) paper, the more sophisticated indexes which have been described in Chapter 4, were not used on Australian data until the end of the 1980s. Kakwani (1986) reported the value of the Sen and Kakwani index in his book but his exposition relied primarily on simple measures of poverty. More widespread application required three things to happen. First data needed to be supplied in unit record form, second until the development of decomposable indexes the indexes were severely limited in the applicability to within-population analysis, and finally a straightforward means of interpreting the values of the indexes needed to be developed. Advantage is taken of the confluence of these three events to make the new measures of poverty contained in this chapter.
However as noted in footnote 7 of Chapter 2, recent work by Saunders (1992) and Whiteford and Kennedy (1993b) suggests that the effect of non-cash income varies across countries.
The choice of 0.5 for a has no easy intuitive meaning but the effect of its choice may be illustrated with a simple example. If there were just two poor families each with the same number of members but one with income 90 per cent of the poverty line and the other with income of 10 per cent of the poverty line. When a is 05 an increase of a given amount (of income) for the poorer poor family will reduce poverty by the same amount as an increase of three times as much in the income of the richer poor family.
Percentage changes over time are quoted as the change between the two values as a percentage of the average value over the relevant period, thus between 1981–82 and 1985–86 poverty rises by 2.45 points which is 21 per cent of its average value over the period, 11.83 per cent.
The definition of working used throughout the discussion of results emanating from the analysis of the unit record tapes in this Chapter, in the previous and subsequent chapters is any work during the calendar year of the survey either full-time or part-time. Thus working income units may well include many who were unemployed or under-employed for much of the year.
The exclusions detailed in the Appendix 1 indicate that the population for analysis makes up about 80 per cent of the resident population of Australia. The main exclusions likely to bias the demographic structure are the exclusion of income units of single persons under the age of 21. These exclusions will underestimate the representation of single persons in the population analysed.
However with a slightly different normalisation the average gap could readily be expressed as the average dollar shortfall of those who are poor. The gap measure used here is equation 4.17 with a set equal to one. Multiplying the value of the gap index by nz/q will produce the average dollar shortfall of those who are poor. For z of $9,877 for couples with no dependents and not working the average dollar shortfall is $3,821.
There was little divergence between movements in per capita household disposable income and average weekly earnings over the period 1973–74 to 1986–87, but between 1986–87 and 1988–89 per capita household disposable income rose faster than average weekly earnings before again moving at much the same rate for the period 1988–89 to 1992–93. Between 1973–74 and 1992–93 per capita household disposable income rose by a factor of 5.68. Over the same period average weekly earnings rose by 5.24. The benchmark poverty line which was 56.8 per cent of average weekly earnings in 1973–74 was therefore 61.6 per cent of average earnings in 1992–93.
The difference between the head-counts calculated using the detailed and simplified Henderson scales in 1985–86 and 1989–90 warrants some comment. More detailed examination of the results for 1985–86 reveals that two-thirds of the increase in poverty calculated using the simple Henderson scales occurs in the group single non-working persons. The detailed Henderson scales allow for age and sex differences and so will be sensitive to changes in the age and sex profile of this group. The ageing of the population over the period of study reduced the relative poverty line for this group in the detailed scales relative to that used in the simplified scales (ageing has both age and sex effects since it also reduces the masculinity of the group). The head-count measure of poverty is also particularly sensitive to the level of benefits and pensions. Consequently the detailed scales, employing an effectively lower poverty line measured lower poverty in 1985–86 and 1989–90.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Johnson, D.T. (1996). New Estimates of Poverty in Australia. In: Poverty, Inequality and Social Welfare in Australia. Contributions to Economics. Physica, Heidelberg. https://doi.org/10.1007/978-3-662-12729-2_6
Download citation
DOI: https://doi.org/10.1007/978-3-662-12729-2_6
Publisher Name: Physica, Heidelberg
Print ISBN: 978-3-7908-0942-8
Online ISBN: 978-3-662-12729-2
eBook Packages: Springer Book Archive