Cohort life expectancy – the best measure of average lifespan

Statistics NZ produces two measures of life expectancy. Both period and cohort life expectancy measures apply death numbers to the population to give death rates, from which life tables can be derived. This article compares and contrasts the different measures.

## Period life expectancy

Period life expectancy is the average length of life remaining at a given age, assuming people experience the age-specific death rates of a specific period from the given age onwards. For example, life expectancy at birth for the period 2012–14 is based on death rates for that period, and takes no account of changes in death rates after 2012–14. Period life expectancy is therefore a hypothetical construct. Although period life expectancy is a timely and useful summary of relative mortality levels between populations (eg males and females, Māori and non-Māori, different areas of New Zealand, different countries, different time periods), it will typically underestimate expected lifespans (assuming death rates continue to decline).

## Cohort life expectancy

Cohort life expectancy is the average length of life remaining at a given age, experienced by people born in the same year. For example, life expectancy at birth for people born in 1900 is based on death rates experienced by those people at each age throughout their life (ie death rates for age 0 in 1900, for age 1 in 1901, for age 2 in 1902, and so on). Cohort life expectancy measures the mortality experience of a particular cohort (eg all people born in 1900) by following that cohort from birth to death (ie until the death of the last survivor). Since cohort life expectancy is based on the death rates of the same group of individuals over their lifetime (allowing for people leaving and joining the cohort through migration), it presents a true historical record of mortality experienced by a birth cohort. The question ‘What lifespan should I expect?’ will therefore be correctly answered by using the cohort life expectancy of a specific cohort.

## Example of 1961 birth cohort

The 1960–62 period life tables indicate the period life expectancy (at birth) of a newborn girl as 73.8 years (see figure 1). This was derived from the death rates of the specific period 1960–62, which are based solely on those who died during that period.

The 1961 cohort life table indicates the cohort life expectancy (at birth) of a girl born in 1961 as 84.8 years – 11 years more than the period life table figure. In fact, most (more than 90 percent) of the 1961 birth cohort are still alive as of early 2016, so death rates at remaining ages (above mid-50s) need to be projected to calculate the cohort life expectancy. This makes cohort life expectancy more difficult to calculate, as it needs an extensive historical series of death rates and/or plausible projections of future death rates at each age.

## How long will New Zealanders live?

The cohort life tables answer some fundamental questions about the longevity of New Zealanders. For males and females born in the late 1870s, life expectancy at birth was about 51 and 55 years, respectively. Life expectancy has progressively increased since, except for males born around the early 1890s and late 1910s who were most affected by deaths during the two World Wars. (The period life tables do not include war deaths, which largely occurred overseas.) For males and females born in the mid-2010s, life expectancy is projected to be around 90 and 93 years, respectively.

## New Zealanders reaching age 65 can now expect to live into their late 80s

For those reaching age 65 in 2016 (born in 1951), the cohort life tables indicate a life expectancy of 86.2 years for males and 88.9 years for females (see figure 2). That is, a further 20–22 and 23–25 years, on average, for males and females, respectively. These tend to be longer than most people expect, because their expectations of lifespan are generally based on the age at death of their parents, or of other family and friends, or of those currently dying.

#### Figure 2

Life expectancy at age 65 is also much higher now than it was at birth (age 0). For those born in 1951, the cohort life tables indicate a life expectancy at birth of 77.5 years for males and 82.5 years for females. Having survived the first 65 years of life, their expected lifespan is now much longer, at 86.2 years for males and 88.9 years for females.

The average lifespan beyond age 65 is a useful statistic to be aware of, especially for those thinking about work and retirement planning. An individual’s lifespan will, however, depend on their behaviours and lifestyle, as well as genetic traits like sex and ethnicity.

For those born after 1951, the average life expectancy at any age will be even longer than those born in 1951. However, the life expectancy of more recent birth cohorts is more uncertain. For example, some of those born since 1990 can expect to live into the 22nd century. Future death rates at each age are naturally uncertain over such a long period of time, despite the relatively sustained decreases experienced over the last century.

## More than 10 percent of those born today will become centenarians

Life tables also present an array of survival probabilities at each age. Less than 1 percent of those born in the late 19th century reached age 100 (see figure 3). For those born in the mid-20th century, about 2 percent of males and 6 percent of females are likely to reach age 100. And for those born in the mid-2010s, perhaps 11 percent (or 1 in 9) of males and 17 percent (or 1 in 6) of females will reach age 100. The greater uncertainty in the survival probabilities for the more recent birth cohorts is shown by the shaded fan in figure 3 – there is an estimated 50 percent chance the survival probability will be within this range.