RDP 9804: Stylised Facts of the Australian Labour Market 1. Introduction

During the 1990s, the unemployment rate has fallen from a peak of over 11 per cent to a low of just over 8 per cent by late 1997. Nevertheless, unemployment remains a major problem in the Australian economy.

To gain some understanding of the trends in unemployment, this paper examines some stylised facts of developments in the labour market over the past 30 years. Our purpose is primarily an accounting one: in general, we do not propose causal explanations for the relationships that are presented. Rather, we examine the movements in the aggregate unemployment rate by considering the unemployment performances of different groups as well as trends in employment growth and changes in participation rates. In doing so, we use recent data to build on similar work by Chapman (1990), Gregory (1991), Fahrer and Heath (1992), Norris and Wooden (1996), Borland (1997a), and Loundes (1997).

The aggregate trend in the unemployment rate conceals quite diverse experiences among the various components of the labour force. Consequently, in Section 2 we break down the aggregate unemployment rate by gender, employment status, skill, age, duration and industry. To provide an alternative overview of labour market developments, we also examine the likelihood with which people move in and out of employment, and in and out of the labour force.

In Section 3, we compare the labour market experiences of the two most recent recoveries, namely those following the recessions of 1982–83 and 1991. We show how the trends in the unemployment rate in the two recoveries reflect the differing trends in employment growth and participation rates. We also examine which sectors of the labour market have fared relatively better or worse in the two recoveries. In general, we find that the 1980s recovery was more evenly spread across all sectors of the labour market, whereas in the 1990s recovery, different sectors of the labour market have had markedly contrasting experiences.

In summary, our paper highlights the following points:

• The significant rise in the aggregate unemployment rate over the past 30 years is largely accounted for by the rise in the unemployment rate for full-time male workers, which in turn reflects slower employment growth in the traditionally male-dominated industries. The falling participation rate for full-time males has in fact lessened the rise in the measured unemployment rate.
• The rapid expansion in the number of part-time jobs (both male and female) has been more than met by an expansion in the supply of people willing to work part-time.
• The unemployment rate for unskilled workers is more than double that for skilled workers. While employment growth has been much stronger for skilled workers, the supply of skilled workers has increased roughly commensurately.
• A comparison of the 1980s and 1990s recoveries reveals significantly slower growth in male full-time employment in the 1990s. On the other hand, growth in male part-time employment has been stronger in the 1990s. The net result has been a smaller decline in the male unemployment rate in the 1990s than in the 1980s.
• There have been similar trends in the female unemployment rate in both the 1980s and 1990s recoveries. However, these reflect quite different forces: in the 1980s, the participation rate was rising rapidly but employment growth was also strong, whereas in the 1990s, the participation rate has been relatively flat but employment growth has also been more subdued.