Determinants of employment and labour force participation: A cohort analysis of Indigenous and non-Indigenous Australians, 1986-96

Author/editor: Gray, MC, Hunter, B
Year published: 1999
Issue no.: 186

Abstract

Data from the 1986, 1991 and 1996 Censuses is used to conduct a cohort analysis of the probability of employment and participating in the labour force for Indigenous and non-Indigenous males and females. Single-year age cohorts are used in the first ever longitudinal regression analysis of Indigenous labour force status. The other contribution of this paper is to analyse labour market processes at an aggregate level by using census information on the entire Australian population (separately for Indigenous and other Australians). By distinguishing aggregate and micro-labour market processes the analysis details what happens to the population, on average, as the level of educational attainment increases.

This paper demonstrates that factors which are related to the probability of employment and participating in the labour force for the Indigenous population differ from those for the non-Indigenous population. At an aggregate level, it is found that the increasing educational attainment of Indigenous Australians which occurred between 1986 and 1996 has not resulted in the anticipated improvements in employment levels. In contrast, employment was found to be positively related to increases in educational attainment in the non-Indigenous population. Marital status is found to be the most important determinant of the probability of employment and participating in the labour force for Indigenous people, especially females. This may reflect the large financial disincentives to work facing many married Indigenous females.

Longitudinal analysis of the probability of employment and participating in the labour force, 1986, 1991 and 1996 Censuses

The cohort nature of the data constructed for this study allows the first longitudinal regression analysis of the determinants of labour force status for Indigenous and non-Indigenous Australians. The advantage of this approach is that statistical techniques can be used to control for unobservable differences between the Indigenous and non-Indigenous populations such as ability and schooling quality, as well as discrimination and other attitudes.

Probability of non-Community Development Employment Projects scheme employment

  • After taking into account cohort specific factors, it is found that region of residence has no effect upon the probability of employment of Indigenous males and females. In contrast, region of residence is an important determinant of the probability of employment of non-Indigenous males and females.
  • Difficulty in speaking English plays only a minor role in affecting the probability of employment for Indigenous Australians, while for non-Indigenous Australians it is a significant factor in reducing the probability of employment.
  • An increase in the proportion of a cohort with a university degree has no statistically significant effect upon the probability of employment for Indigenous cohorts. In contrast, for non-Indigenous cohorts, an increase in the proportion with a degree plays an important role in increasing the probability of employment.
  • The role of diploma level qualifications in increasing the probability of employment is somewhat surprising. For Indigenous males and females, and non-Indigenous males, diploma level qualifications appear to have no effect upon the probability of employment. For non-Indigenous females, an increase in the proportion of a cohort having a diploma level qualification is associated with a significantly decreased probability of employment. One reason for this apparently anomalous result may be that diploma level study is an intermediary step for many women who eventually upgrade their qualification to a degree.
  • For Indigenous males, the regression estimates show that being married is associated with a lower probability of being employed as compared to being single. In contrast, for non-Indigenous males being married or widowed, divorced or separated is associated with a significantly increased probability of being employed. Being married or widowed, divorced or separated is associated with a lower probability of being employed for Indigenous and non-Indigenous females.

Probability of participating in the labour force

  • A striking finding for Indigenous males is that none of the explanatory variables are significantly related to the proportion of a cohort which is participating in the labour force and therefore other factors must explain the decision to participate in the labour force. For Indigenous females, having a diploma level qualification and marital status are found to be significant determinants.
  • The probability of participating in the labour force for non-Indigenous Australians is found to be related to a wide range of factors including region of residence, level of educational attainment and marital status.

Prospects for achieving statistical equality in employment and labour force participation

The estimates allow an analysis of the effects of achieving statistical equality in educational attainment and region of residence between Indigenous and non-Indigenous Australians on employment and labour force participation. It is found that moving Indigenous people to the buoyant job markets in major urban areas would have only a limited effect on narrowing the gap between Indigenous and non-Indigenous employment rates. Similarly, increasing the post-secondary educational attainment of Indigenous Australians to that of non-Indigenous Australians would not resolve Indigenous labour market disadvantage. Other difficult to measure factors, such as unobserved heterogeneity (discrimination and lack of worker motivation), are likely to be as important as the number of jobs available in an area.

Policy discussion

It is often asserted that one of the keys to improving Indigenous labour market outcomes is increasing the level of formal education of the entire population, especially younger age cohorts. The analysis outlined in this paper casts some doubt upon this presumption. This is not to suggest that increases in educational attainment which have been achieved by Indigenous cohorts are not without benefit as evidenced by the effect they have on incomes. However, formal education needs to be combined with a series of other policies aimed at combating Indigenous labour market disadvantage. The results clearly demonstrate that a large part of Indigenous employment disadvantage is not simply due to lack of educational attainment and the level of labour demand in the regions where Indigenous Australians live but is also due to unobserved differences such as schooling quality, assimilation, discrimination and other attitudes.

ISBN: 0 7315 2621 X

ISSN:1036 1774

Updated:  16 June 2009/Responsible Officer:  Centre Director/Page Contact:  CASS Marketing & Communications