The changing relationship between work and death risk

As we get older and our risk of dying increases, we are less likely to work.  How has that changed over time, and what could that mean for pensions or superannuation?

Retirement is changing. People are working longer and seeking more flexible work at older ages.  It is still the case that there is a lower proportion of the population participating in the labour force – whether employed or self-employed – at older ages than in middle age.

New Zealand has seen a marked increase in this measure at older ages. It ranks 7th in the OECD for men’s average age of exit from the labour force and equal 5th for women (2009-14). New Zealand is also a long-living country, with life expectancy at age 65 around the OECD average for women and top quartile for men.

In many countries, more working at older ages and increasing life expectancy are reasons for raising the age at which the state pension is payable.  A counter-argument to increasing state pension age is often to do with health status or ability to work at older ages.  It is therefore important to examine the changing relationship between how much work is actually done and measures of health at older ages.

One, albeit extreme, measure of health status is the one-year death rate at a given age. This is the probability that the average person of a given age will die in the next year. Death rates increase by age. The higher the one-year death rate is, the shorter the retirement will be for someone who starts retirement at that age. Lower one-year death rates mean longer average lifespans.

The charts below show the changing nature of the relationships between labour force participation rate and the one-year death rate for New Zealand over the last thirty years. The data points represent five-year age

nz-labour-participation-vs-death-rate-men-oct16The lines have moved to the left as death rates have fallen in each age group. The lines have also moved up at all ages for women, and for men over age 59, as labour force participation has increased. As we knew, New Zealanders are working more at older ages, and average lifespans are longer.

For men, the 1986 line is above that for 2016. This means that for any given death rate, men used to work more than they do now. Whether that means the rate of men working should increase to achieve the same relationship relative to 1986’s death rate depends on your view of 1986 being a good baseline.

There is no magic in the choice of 1986 as a base year, although at that point the age of eligibility for New Zealand Superannuation (NZS) was age 60. The increase to the current age of 65 was phased in from 1992 to 2001. The increase in age of eligibility has been cited as a reason why labour participation at older ages has increased.

For women, the two lines for 1986 and 2016 overlap. The 2016 line is almost vertical, squashed up against the left. The scope to improve death rates is limited, while there is still scope to increase participation rates from current levels for the over 65s. This suggests we have to consider a new relationship between working and death rates in future.

The 65-69 year-old age group is highlighted by circles on the plots. For men, the labour participation rate for this age group is just slightly higher in 2016 than it was for the age group 60-64 years in 1986. But the death rate for age group 65-69 has fallen dramatically, and is now less than even the 60-64 age group in 1986. For women, the same comparison for labour participation rate is even more marked, and the death rate has followed the same pattern but with less of a fall as it was low to start with.

For both men and women, the current age 70+ group works only slightly less than the 65-69 age group thirty years ago. The death rate for the current age 70+ group is less than that of the 65-69 age group thirty years ago.

The charts suggest that on both labour participation and one-year death rate measures, age 65 is at least the new 60, and age 70 is the new 65.

There is a lot about increasing eligibility age not shown by this analysis, including the critical questioning of the need to increase eligibility age if NZS can continue to be afforded in future. However, the charts illustrate an argument for increasing age of eligibility for NZS as a result of labour participation and health improvements already seen. The charts are a summary of how health and working have both changed to create new relationships, but the analysis cautions against relying too much on future mortality improvements alone as a rationale for future increases. Finally, the charts remind us of those who have a mortality disadvantage at the crucial ages of 65 and over as men still have higher death rates compared to women.


Thanks for the inspiration for this post to Paul Johnson of the Institute for Fiscal Studies. Paul included a similar analysis for men in the UK in his excellent presentation to the Institute and Faculty of Actuaries’ International Mortality and Longevity Symposium, London, September 2016.

The shape of the graph is very similar to that for New Zealand although New Zealand has higher labour force participation and lower mortality rates.  While New Zealand is keeping its age of eligibility for the public pension at age 65, the UK is on the road to increase its state pension age to 66 by 2020 and 67 by 2028.


The New Zealand charts in this post were prepared by the author using Statistics New Zealand labour force participation rates issued August 2016 and cohort life tables consistent with 2014-base data.

Leave a Reply