Life expectancy considerations for pension policy, by ethnicity, New Zealand

Deaths are still high at young and middle ages within Māori and Pacific groups. This analysis shows the scale of the challenge to improve health outcomes and reduce deaths at young and middle ages, but also shows that, over time, more people of all ethnic groups can expect to receive the public pension for longer.

This post is based on a note originally prepared for the Expert Advisory Group to the Retirement Commissioner in December 2020 to explain a way of thinking about reasonable expectations for lifespans for New Zealanders and how that varies across the population, specifically by ethnicity. It first reprises some basics about life expectancy measures, then assesses available data on longevity by ethnic groups.

Lifespans for New Zealand over-65s

Lifespan, or age at death, varies across the population.  Life expectancy, as commonly used, doesn’t tell us what the expected lifespan might be for a person, or the average of the population.

Life expectancy at birth is not a good measure of how long people are expected to live while receiving NZ Super, the public pension in New Zealand. First, life expectancy at age 65 is higher than life expectancy at birth because the group of 65-year-olds is relatively healthier than the group was at birth. Second, it’s important to understand which life expectancy measure is used (at any age).

Period life expectancy is often used in analysis of the health of groups of people as it measures average mortality between populations at a point in time.  It is calculated as the average length of life left at a given age, assuming people experience population average death rates as they were at that time.  In reality, death rates change every year, so period life expectancy is an entirely hypothetical indicator of how long people might live.

Cohort life expectancy is a better measure of potential lifespan because it uses information on how death rates are expected to change over time, so as people go through their lives.  It is the average length of life left at a given age for a group of people born in the same year, based on expected future average death rates for the population.

Using cohort life expectancy, these charts show the range of expected lifespans for NZ population cohorts born in different years who survived to age 65. This is the best way of looking at how long people might live while receiving NZ Super.

The charts show a number of things we know about longevity, including (1) women, on average, live longer than men and (2) lifespans are getting longer over time.  The charts also show that these “truths” are outcomes of a statistical distribution, so are not always true for everybody.

The black columns also show how the life expectancy measure (even when using cohort life expectancy) is not the best or only measure of how long, typically, people live.  It’s an average of a skewed distribution and is always on the low side.  Other more insightful measures are shown on the charts.

Variation by ethnicity

New Zealand’s statistical agency, StatsNZ, only publishes cohort life expectancy estimates for the total New Zealand population but it does produce period life tables for different ethnic groups.  However, there are problems using period life tables for comparing mortality by ethnicity:

  • Comparing period life expectancy between New Zealand ethnic groups gives an indicator of relative mortality levels, not of differences in current average lifespans (because period life expectancy is a hypothetical construct).
  • The latest period data, 2012-14, is old.
  • Historic trends in mortality comparisons between ethnic groups are difficult to analyse because of different definitions of ethnicity on death registration, birth registration and census forms during the 1980s and early 1990s.

Nevertheless, the data point to an undeniable mortality disadvantage for Māori and Pacific ethnicities:

  • While non-Māori period life expectancy has steadily increased, Māori period life expectancy increased more rapidly up until around 1980, after which it was stable until increasing again from the late 1990s*.
  • The gap between Māori and non-Māori period life expectancy has fluctuated, but over the long term narrowed. For the period 2012-14, period life expectancy at age 65 for Māori men was 15.4 years, and women 17.5 years. The non-Māori figures were 19.1 and 21.6 years, implying a gap for that period of 3.8 years for men and 4.1 years for women**.

As shown earlier, period life expectancy figures show disparities in levels of population health but are not good measures of how long people are living or might live in future. We should not focus too much on the single measure of the gap between period life expectancies if we are asking about differences in likely lifespans between groups.

Another way to see differences in lifespan is to look at the distribution of actual deaths by age.

The first chart here shows the estimated number of deaths by age from 2019 to 2038 in Māori and Pacific groups. Note these are the estimated number of actual deaths, not deaths from a cohort of 100,000 lives at age 65 as shown in previous charts (because we don’t have that information).  The number of deaths increases proportionately as the size of the population grows.

The ageing of deaths is very marked. Fewer people are expected to die aged 45-69, and many more at ages 70 and above.

From the same data, the second chart shows the estimated proportion of deaths in broad age groups by ethnicity for 2019 and 2038. The expected trend is for deaths to increasingly occur over age 80 for all ethnicities – this is the second “truth” referred to previously. It is also true that deaths are still occurring disproportionately at young and middle ages within Māori and Pacific groups. This shows (1) the scale of the challenge to improve health outcomes and reduce deaths at young and middle ages, and (2) more people of all ethnic groups can expect to receive NZ Super for longer.

* Ministry of Health; Blakely et al. (2007)

**StatsNZ Complete New Zealand Period Life Tables: 2012–14, median estimates

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.