A previous post showed how a slowdown in mortality improvements in some countries has caused researchers to look for the possible causes, with “austerity” as a leading suspect. This post shows New Zealand isn’t suffering from such a negative trend.
What “slowdown in mortality improvement” means: A mortality rate measures the probability of death. A standardised mortality rate (SMR) measures the death rate for an age group, keeping numbers at each age constant so a trend over time is only the result of changes in mortality rather than movements in population numbers. If an SMR gets smaller then mortality is improving. That’s a good thing as it means fewer deaths are occurring in the age group and average lifespans are getting longer.
Generally, mortality has consistently improved in most countries, at all ages, over centuries –we are living longer. The rate at which mortality has improved over the last few decades has been fast enough for the increase in life expectancy to seem surprising. However recently, in some countries, the rate of mortality improvement has been lower than in previous years. This means that life expectancy is still increasing, but at a slower pace than previously.
The countries where this seems to be happening include the UK, Canada, the US, Portugal and Ireland, but not Spain, and trends are unclear in Germany, Belgium and Finland.
What’s happened in New Zealand? Mortality improvement has followed a pattern common to many developed countries. New Zealand mortality improved at a modest pace in the late sixties and early seventies, then the improvement accelerated through the eighties and even more so through the nineties. But crucially, when the England & Wales* trendline flattens markedly after 2010, there is only a faint slowdown for New Zealand men, and none discernible for New Zealand women.
The New Zealand experience is shown in the table below and in this chart (PDF). 
| Average annual improvement in standardised mortality rate, ages 50-89, New Zealand | ||
| Years | Female | Male |
| 1966-1975 | 1.2% | 0.6% |
| 1975-1990 | 1.5% | 1.6% |
| 1990-2008 | 2.0% | 2.8% |
| 2008-2015 | 2.0% | 2.5% |
Source: http://www.longlifepensions.com using Statistics NZ cohort life tables 1876-2015.
See below** for details.
Mortality rates change year to year, and it’s important to look for long-term trends. The improvement rates shown in the table depend to some extent on the start and end dates of each period. These have been chosen to fit, by eye, inflection rates in the chart but are also similar to the England & Wales analysis. Conveniently each period covers two New Zealand government terms, National then Labour, except the last which was all under a National-led government.
The chart shows a remarkably linear long-term downward trend in mortality rate so that the story of rise and (very slight) fall in the rate of change is truly second-order. The relentless improvement in mortality is a more important trend than the possible slight slowdown in the rate of mortality improvement – we are still living longer.
There is another faint sign of a possible blip in recent years. Between 1966 and 2015, there have been only 11 years of mortality worsening (that is, an increase in SMR for this age group) for New Zealand females and nine years for males. Most of these were in the first half of the period we are looking at here: both males and females have had only four such years after 1990. However, for both groups two of those years of mortality worsening occurred since 2010.
Summary: Mortality improvement still has momentum in New Zealand, with only a slight slowdown for men. The more dramatic slowdowns seen in other countries don’t seem to be happening in New Zealand yet, but it’s worth watching to see if the recent trend in male mortality rates signals a change.
* Most detailed analysis of the UK uses data from England & Wales.
UPDATE June 2018: the Office for National Statistics in the UK finds statistical evidence for a “breakpoint” around 2010 when the trendline mortality improvements slowed from the preceding long-term mortality declines, for different age groups, both male and female, in England and Wales.
** More details on calculations for this post: Standard population is New Zealand Estimated Resident Population 2015. Simple linear trend fitted to Δmx. SMR age group of ages 50-89 and standardised population of 2015 were chosen to be consistent with the actuarial analysis of mortality data for England & Wales referenced here, subject to data availability.