3  Labour force

The ageing of New Zealand’s population will have profound effects on the size and composition of the labour force. The ramifications for the economy are varied and depend on how both labour supply and labour demand respond to the age-structural changes in the working population.

3.1  Supply-side issues

The growth rate of New Zealand’s working age population is projected to decline and become negative by the year 2041 (Statistics New Zealand 2001c). At the same time its composition will be changing with the working age population getting older, the proportion of Maori and Pacific Islanders in the working age population increasing, and labour supply becoming increasingly concentrated in “young” regions such as Auckland (Koopman-Boyden 1993).

Source: Statistics New Zealand Household Labour Force Survey (Statistics New Zealand INFOS Database).

The shrinking of the working age population raises the prospect of an economy-wide reduction in total labour supply by the middle of this century. However the actual size of any reduction will depend critically on future labour force participation rates. Over the past fifty years there has been a slight increase in total labour force participation[7] (see Figure 2). A long term decline in male participation rates has been offset by an increase in female participation rates[8], however while female rates may continue to rise it is reasonable to expect that they will reach a plateau and no longer be the source of a growing labour force.

If we decompose female labour force participation rates by age we find that there is a significant reduction in labour force participation rates amongst women over the age of fifty. This is in contrast to male participation rates that decline rapidly for workers over the age of 60. There is evidence to suggest that this difference in age-specific participation rates is a cohort effect. In 1986 the decline of female labour force participation appeared to begin at the ages of 40 to 44, whilst in 1992 the change looks to have occurred between the ages of 45 and 49. Overall, between 1987 and 1996 there was a 12.3 percent increase in female labour force participation rates for women between the ages of 50 to 54 (Dixon 1996). These increases may well be due to cohort specific participation rates and what we are seeing in the data is cohorts of women, with higher lifetime participation rates, displacing cohorts with lower lifetime participation rates. Subsequently we may find that the trend of increasing labour force participation amongst women will cease when older cohorts reach retirement age.

In the absence of female labour force participation driving up overall labour participation rates, there is another potential source for growth in the labour supply. Increases in longevity indicate that there is potential for an overall increase in labour force participation rates amongst older workers as better health extends individual’s productive working lives. This is in contrast however, to current trends in labour force participation amongst older workers in New Zealand. The average age of retirement has declined over the last fifty years with the average age of retirement for men falling from 64.8 in 1950, to 62 in 1995 and the average age of retirement for women has falling from 61.2 to 58.6 (Auer and Fortuny 2000).

A key question for enquiry then is: What affects the decision to retire? One contributing factor is the disincentive to work from the implicit tax rate that workers face around the age of retirement. In the case of countries with public pension schemes the decision to delay retirement carries the cost of foregone public benefits or higher tax rates, and the cost of foregone income from a private pension scheme (for which an extra periods contribution may make little or no difference to the level of entitlement post-retirement). This sort of implicit tax rate in New Zealand was, when the superannuation surcharge applied, estimated to be, on average, twenty-eight percent of gross income (Blondal and Scarpetta 1998). Although the surcharge no longer applies a superannuitant would still face higher tax rates if he or she chose to work after the age of 65.

The decision to retire is also related to health, wealth, and education (Haider and Loughran 2001) (Quinn 1998). In a study of older workers in the Engineering, Printing, and Manufacturing Union in New Zealand respondents were asked what factors they felt were important in the decision to retire. The largest response (95.9 percent) was in the category of “Your health” and the second largest was “financial security” (McGregor 2000). Education was not explicitly considered; however the strong correlation between education and wealth and income suggests that it is also a factor in the retirement decision.

Equally, uncertainty is an important aspect of the decision to retire (Lumsdaine 1995). Individuals are apt to delay retirement in expectation of retiring when conditions are more favourable; uncertainty of when such a time will be keeps people in the labour force. In the study of older workers cited above, 84.4 percent of respondents considered “the increasing cost of living” as an important factor in the decision to delay retirement. Although the statement is not an explicit reference to uncertainty the implication is that workers are concerned about potential future increases in living costs reducing their living standards in retirement.

The role that uncertainty has in the decision to retire is particularly pertinent for New Zealand, where retirement income policy has sought to create certainty in order to allow people to plan more easily for retirement. In general one would expect that greater certainty of retirement income would tend to lower labour force participation rates amongst older workers.[9]

There have been few studies of the retirement decision in New Zealand. The debate surrounding the economics of population ageing may be significantly enhanced by further research on this topic. In particular studies carried out at a microeconomic level or qualitative surveys would particularly enhance our understanding of the retirement decision, as they are more able to capture the heterogeneity of older workers than quantitative studies. The “Employment of the Older Worker Survey”(McGregor 2000) was a useful addition to the literature in this respect, and microeconometric analyses carried out by the OECD (Burkhauser, et al. 1998) also appear quite successful in explaining the factors that affect retirement decisions in a way that respects the heterogeneous nature of the elderly as a demographic group.

Another issue that deserves fuller investigation within the New Zealand context is whether or not worker productivity declines with age. Analysing changes to the size of the working age population represents a fairly simplistic view of labour supply when what is really in question is the future supply of labour inputs, not just the number of workers in the economy. The median age of people in New Zealand is expected to rise to 45.5 in 2051 from 34 in 1999 (Statistics New Zealand 2000). We can be fairly certain that there are some functions that decline with age, not least of which are cognitive functions such as the ability to learn new skills, but older workers often compensate for these things through, for example, lower rates of absenteeism. (Campbell 1993) (Auer and Fortuny 2000)[10]. If a negative relationship between age and productivity exists there is at least the potential for a decline in output per worker.

Any such decline could be expected to affect different sectors quite differently, depending on the age structure of their workforce. Service sectors employing largely young workers might for example experience little change. However, here as in many aspects of population ageing, there will probably be a series of dynamic responses and interactions. As the overall age structure changes, the relative supplies of labour of different ages and skill levels will change, potentially leading to changes in relative wage rates. Firms could be expected to adjust the age mix of their workforce in response to these price signals, so that the present age structures of employment in particular industries or sectors may well evolve.

In the future we may expect that a larger proportion of the workforce, particularly younger workers, will be Maori or Pacific Islanders and this may also affect productivity. Currently Maori and Pacific Islanders are proportionately under-represented in tertiary education. Were such trends to continue there is a strong likelihood of a decline in productivity from a lack of advanced skills in the economy.

Source: Organisation for Economic Co-operation and Development 2000, p.72

Furthermore Maori and Pacific Islanders are disproportionately represented in unemployment statistics and have lower rates of labour force participation than the general population (see Table 1). It is too difficult to say whether such outcomes will persist. However they do raise important questions about the future size of the labour force.

Given it is expected that there will be major regional differences in population age structure over the next fifty years, some regions will see their working age population decrease more rapidly than others. Although one would expect that the market would respond to this with migration out of regions with relatively high labour supply, there is potential for substantial labour shortages in those regions with ageing populations, at least in the medium term. This too is an area for further enquiry.[11]