3.5  The age structure of the working age population

Within the working age population, both employment and productivity vary by age. All else equal, a change in the age structure of the working age population implies a change in average participation and productivity levels, the and terms in Equation 1. Some commentators imply that these charges will be significant (OECD 1998).

The relationship between the age structure of the working age population and average participation levels across the working-age population can be expressed with an equation of the form

(5)     , ,

where is average hours worked by people in age group , and is the proportion of the working age population in age group . A similar equation can be used for productivity, though in this case the ’s measure hours worked by people in age group as a proportion of hours worked by the whole working age population.

Figure 7 shows the proportion of age groups 20-24 to 60-64 in employment, an approximation of the ’s. Estimates are shown for New Zealand and for the OECD as a whole. Obtaining analogous estimates for productivity is more difficult, since all of the usual measures such as wages or time in employment are flawed in important ways. The standard assumption, however, is that the productivity rises and then declines with age (Stephenson and Scobie 2002: 9-10).

Figure 7 – Employment ratios by age, New Zealand and OECD, 2001

Source: Calculated from data from the OECD Labour force data online database.

Note: Employment ratios are defined here as the number employed full time or part time divided by the population in the age group. All results are for males and females combined.

Are ongoing age structural changes sufficiently large to produce a significant effect on participation or productivity? Table 4 presents data on the age-structural changes. The projected age structures for 2050 are noticeably different from those of 1950: for instance, the proportion of the New Zealand working age population aged 60-64 increases from 7.5% in 1950 to 11.6% in 2050. Changes in other age groups are, however, smaller, and the mean age of the New Zealand working age population rises by only 3.3 years over the century between 1950 and 2050.

Source: Calculated from data from the UN Population Division’s World Population Prospects online database.

*The high figure for 1950 is not a typological error. It probably reflects the effects of the fertility decline from the late 19^th Century to the early 20^th Century, which mean that older cohorts in 1950 were unusually large, compared with younger cohorts.

As Figure 8 shows, changes in the mean age of the working age population are relatively small across the OECD, and much smaller than changes in the mean age of the entire population. Similarly, cross-country differences in the mean age are far smaller for the working age population than for the total population. The reason is that most variation across time and between countries in the overall age structure of the population is due to variation at the youngest and oldest ages.

The expression for the mean age of the working age population is where is the mean age of people in age group .[6] This has exactly the same form as Equation 5, the equation for average participation, with the ’s replacing the ’s. The fact that the age structural changes do not have much effect on the mean age therefore suggests that they would not have much effect on the participation rate either. This is indeed the case. If, for instance, the OECD age specific employment ratios shown in Figure 7 are applied to the age structure of the New Zealand population in 1950, the overall ratio is 69.2%. If the same age specific ratios are applied to the age structure in 2050, the overall ratio is 67.2%. The peak occurs in 2000, when the overall ratio reaches 69.9%. The changes are small. Performing the same calculations for other countries yields similar results. In the absence of good estimates of age-specific productivity rates, no such calculations are possible for productivity. But because the relationship between overall rates, age-specific rates, and age structure is essentially the same, there is no reason to believe that the effect of changing age structures on productivity would be any more striking.

Figure 8 – Estimates and projection of the mean age of the working age and total population, OECD countries and regions

(i) Working age population

(ii) Total population

Source: Calculated from data from the UN Population Division’s World Population Prospects online database.

The overall conclusion is that the effects of changes in the age structure of the working age population on participation and productivity are likely to be small. There is therefore little scope for trends in the age of the working age population to advantage or disadvantage New Zealand relative to other OECD countries.