As noted above, trends in the various components of participation in Australia have resulted in solid overall growth in hours worked per head of the population aged over 15. This, in turn, has been supportive of economic growth and rising incomes per head. These trends are detailed below.
The first component in participation is the labour force participation rate, which measures the proportion of the population over 15 who are either working, or available to work.
The labour force participation rate has risen modestly, from around 60 per cent in the mid 1960s to around 64 per cent at present. This gradual upward trend masks quite different trends among men and women. Female participation rose dramatically from below 40 per cent in 1966 to over 55 per cent in 2003. At the same time, male participation fell markedly from 84 per cent in 1966 to around 72 per cent in 2003 (Chart 2).
Chart 2: Participation rates by gender, Australia
Source: Australian Bureau of Statistics, Labour Force, Australia, Preliminary (Cat. No. 6202.0, March 2003).
Broadly similar trends in overall participation rates by gender are apparent across the OECD (Table 2). The breadth and strength of these trends suggests that they are mainly driven by economic opportunities and social factors. However, policy settings also can have an influence.
Table 2: International participation rate changes(a)
- This table shows participation rates for the OECD definition - the population between the ages of 15 and 64. The participation rates for Australia in this table are therefore higher than other references to participation in this statement.
Source: OECD (2002c).
Consistent with the general trend, Australian participation rates for men aged between 55 and 59 (that is, the proportion of men in this age group who are working or available for work) fell over the last 20 years. But Australia's rates fell more than in other key economies. For example, from 1981 to 2002, the Australian participation rate for this group fell 9 percentage points; whereas, from a nearly identical starting point, the US participation rate fell only 5 percentage points.
While Australian participation rates for women aged between 55 and 59 increased significantly, these increases were consistent with the trend across most of the OECD countries. So, while the Australian female participation rate almost doubled for this age group, Australia's OECD ranking remained stable for the group.
Low and declining participation rates (by international standards) for Australian males aged between 55 and 59 suggest specific policies or conditions in Australia over the last 20 years may have discouraged participation by older men more than in other countries. Accelerating structural adjustment in the 1980s may have displaced some workers with lower skill levels, as also happened in the United States. But to a greater degree than in the United States, displaced Australian workers may have had difficulty finding alternative employment due to a less flexible labour market, and a mismatch between their skills and those required for new jobs, with insufficient incentives to bridge the gap. Some of these workers may have become discouraged in the search for employment and hence left the labour market.
Related to this, the numbers of people in this age group receiving the Disability Support Pension, which is similar in value to the age pension, has grown markedly.1 Some older workers may have effectively retired early on this pension (or other welfare payments) rather than continue to look unsuccessfully for work. With increasing numbers of people in these age groups and increasing proportions of them receiving welfare payments, the resultant falling rate of participation, if sustained, could have significant economic effects in the decades ahead.
OECD research shows that it is more difficult to reverse retirement decisions, once taken, than it is to encourage people still in employment to delay retirement (OECD 2003a). Those most weakly 'attached' to the labour force tend to be more likely to initiate early retirement before age 65. OECD evidence shows that workers who did not complete secondary education, sole parents and the moderately disabled are other groups similarly at risk of withdrawing from participation.
In Australia, there are 2.8 million people under the age of 65 on income support. This is over 20 per cent of all working-age Australians. Parenting Payments are paid to sole parents supporting a child or the partner of a person who is unemployed and supporting a child. Parenting Payments and the Disability Support Pension have more recipients than there are for the Newstart Allowance, which is the main unemployment payment. The Newstart allowance has an activity test, while the Disability Support Pension and Parenting Payments do not.
Comparisons across the OECD countries show problems of underutilisation of labour resources through low participation are ubiquitous. In almost all countries, low participation detracts more from the potential labour force than excessive unemployment (OECD 2003a).2
European studies also indicate that, across all age groups, withdrawal from participation is seldom reversed and leads to a high risk of persistent low standards of living (OECD 2003a). So there are social as well as economic reasons for concern about falling participation rates among those of working age. It is also unlikely that withdrawees have made sufficient financial provision for permanent withdrawal from the labour force. In equity terms, the falling participation problem is similar to that of long-term unemployment.
The IGR noted that recent participation trends combined with an ageing population would reduce the overall participation rate markedly over time, from around 64 per cent in 2003-04 to slightly over 55 per cent in 2041-42. This decline could become significant from the latter part of this decade, other factors unchanged. The IGR emphasised, however, that such projections make little allowance for change over time in behaviour with respect to labour force participation.
For example, if participation rates for each age and gender cohort rose towards the top fifth of the OECD by 2020-21 and remained there, instead of at recent levels, then the overall participation rate in Australia would fall to around 60 per cent rather than 55 per cent. This would lead to a GDP per capita level by 2041-42 that was around 9 per cent higher than projected in the IGR.
The second component of participation is the unemployment rate, the proportion of people participating in the labour force but unable to find work.
After exceeding 10 per cent during the early 1990s, the unemployment rate has fallen to around 6 per cent in 2003 (Chart 3).
Chart 3: Unemployment rate, Australia
Source: Australian Bureau of Statistics, Labour Force, Australia, Preliminary (Cat. No. 6202.0, March 2003).
Australia's unemployment rate has generally been between the rates experienced in the United States and the Euro area over recent years (Chart 4). The Euro area is characterised by high unemployment, particularly compared to the more flexible labour markets of the United States, United Kingdom and New Zealand. The recent economic slowdown in the United States has raised US unemployment rates to Australian levels for the first time in 20 years.
Chart 4: Standardised unemployment rate, Australia, United States
and Euro area
Source: OECD (2003b).
While the unemployment experience in Australia has, in general, been better than in many other developed countries, there have been other countries that have experienced lower levels of unemployment over a consistent period of time, most notably the United States.
Countries choose the level of unemployment they are willing to tolerate. Policy actions influence unemployment outcomes. For example, if minimum wages are high compared to the median wage, businesses will be less willing to employ some lower skilled workers.3 Employment protection legislation which makes it harder to dismiss employees also makes it riskier and more expensive for employers to hire new employees, and contributes to the existence of unemployment. Tax and welfare policies also can interact to increase unemployment if benefit payments are not work-tested or time-limited, or are generous compared to after-tax incomes from employment (OECD 1999). While these issues are seen as contentious by some in Australia, they are part of an increasing consensus about the causes of persistent unemployment. These issues are taken up in more detail later in this statement.
The key remaining element that contributes to overall participation is the average number of hours which people work.
Average hours worked per employee in Australia has fallen only slightly over the past two decades, after falling sharply between the mid 1960s and early 1980s (Chart 5). Since mid 2001 average working hours have fallen further, apparently due to full-time employees working fewer average hours. In recent months, the average number of hours worked has begun to climb again.
Chart 5: Average hours worked per employee, trend, Australia
Source: Australian Bureau of Statistics, Modeller's Database (Cat. No. 1364.0.15.003, December 2002), unpublished data, and Labour Force Statistics, Australia (Cat. No. 6203.0, March 2003).
While Australia's average hours of work fell by around 10 per cent over the last four decades, most other OECD countries have recorded larger falls. Average hours fell by over 30 per cent in Germany and Norway and by around 20 per cent in the United Kingdom and Japan. An important exception is the United States, which has around the same average hours of work today as four decades ago.
The fall in average hours worked in Australia almost entirely reflects the increase in the number of people in part-time employment, as average hours worked by part-time and full-time employees both increased over the past two decades. In the mid 1960s, around one in ten employees worked part time. By the start of 2003, over one in four worked part time. Australia's overall rate of part-time employment (27 per cent) is the second highest in the OECD, after the Netherlands (33 per cent). Japan (25 per cent), Switzerland (25 per cent), New Zealand (23 per cent), and the United Kingdom (23 per cent) also have high rates of part-time work.
The rise in part-time employment in Australia parallels rapidly increasing labour force participation by women. Availability of part-time work is likely to have facilitated higher participation by women, and by formerly non-working parents in both single-parent families and dual-income families. The tendency for young people to have longer periods in education, and for workers to have periods of further education and retraining over their working lives, are also assisted by the availability of part-time work.
While the strong growth of part-time employment might be taken as evidence of welcome improvement in labour market flexibility, it also bears another interpretation. It may, in part, be the means by which employers escape rigidities in the regulation of full-time employment. Surveys suggest significant proportions of part-time workers wish to work more hours, and there is some evidence that full-time workers performing both paid and unpaid overtime may wish to work fewer hours. These results suggest a labour market still too inflexible to allow part-time and full-time workers and employers to arrive at mutually advantageous arrangements.
It is impossible to prescribe what tomorrow's participation performance ought be. Participation outcomes (that is, hours worked per head by the working-aged population) are a product of individual and family choices in the context of labour market options and policy influences.
It is worth reiterating, though, that European experience suggests falling participation rates, like long-term unemployment, bear mostly on the least affluent, are difficult to reverse, and can lead to persistent poverty and social disengagement.
Demographic projections show more of the labour force will be concentrated in the older age brackets within which participation in Australia has become relatively low by international standards. It could be that business, in response to changing demographic structures, might change its employment practices and make it more attractive for this group to actively participate. However, it would appear that these low participation rates are at least in part driven by insufficiently flexible and competitive markets and unintentional interactions among policies, rather than changing worker preferences.
Of the three contributors to participation outcomes, it is very difficult to predict the future trend in average working hours. Tomorrow's more productive workers might choose lower average working hours. But for the other two contributors to participation outcomes, it seems likely Australia would be a more fair society if participation rates were higher and the unemployment rate lower than today.
1 Between 1990 and 2000, the number of male Disability Support Pension recipients aged 50-59 increased by over 60 per cent. In the same period, the number of men in this age group grew by only around 40 per cent. Over the longer term, male Disability Support Pension recipients have increased by over 400 per cent since 1972, while the male population increased by only 45 per cent. (Department of Family and Community Services 2002; Department of Family and Community Services 2001).
2 The OECD illustrates the relative magnitudes by assuming any unemployment over 5 per cent is 'excessive', and that low participation countries could have participation rates (for each of three age brackets and both genders) as high as the third highest experienced in the OECD.
3 Australia has the second highest minimum wage in the OECD, at 57.9 per cent of the median wage. Only France, at 60.8 per cent of the median wage, has a higher minimum wage. In contrast, the minimum wage in New Zealand is 46.3 per cent of the median wage, 41.7 per cent in the United Kingdom and 36.4 per cent in the United States. See Keese and Puymoyen (2001).