The misunderstanding of life expectancies

by David Knox | Mercer | 18 April 2018

While this article is targeted at financial advisers, it is food for thought about your own longevity and the need to consider funds lasting longer than previously thought. While it is based on Australian statistics they will be similar to New Zealand ones:

As the baby boomers approach or enjoy retirement, a fundamental question they all face is: "how long does my money have to last?" Or, to put it somewhat more bluntly, "when am I going to die?"

Of course, this is a very difficult question, and no-one knows the exact answer. As a result, we often express the answer in terms of life expectancy. That is, we quote the average life expectancy for a person of a certain age and gender is X years. But, often this answer is misunderstood or poorly used.

To understand this, let's go back in time.

When a current baby boomer retiree was born, shortly after World War II, life expectancy at birth was 66.1 years for males and 70.6 years for females.¹ But, if the baby born shortly after WWII had known they were going to reach age 65, then life expectancy was 77.3 years for males and 79.4 years for females.² In other words, if a person knew they were going to survive to age 65, their life expectancy was much higher.

The same logic applies today. The latest Australian Life Tables show that the life expectancy at birth is 80.1 years for a baby boy and 84.3 years for a baby girl³ - about 14 years longer than the post-WWII figure.

Hence, we often read that life expectancy for the Australian population is about 82 or 83 years. But this is a very misleading statistic for those involved in the retirement planning industry.

As was the case with the previous stats, a baby boomer who has now survived to age 65 and is preparing for retirement will have a longer life expectancy. Indeed their life expectancy is actually 84.2 years for males and 87.1 years for females.⁴

However, even these numbers are misleading as they are based on current mortality rates, not the future rates. And one thing is for sure - average mortality rates will continue to decline with the resulting increase in life expectancies. This has happened for the last 100 years and there is no reason to expect such improvements to disappear.

In fact, the Australian Government Actuary makes allowance for such an outcome by looking at a cohort of lives rather than a cross section at any particular date. This measure takes into account the improvements that could be expected over the future lifetime of an individual. This approach increases the life expectancy for a 65-year old to 86.0 years for males and 88.6 years for females, based on continuation of the mortality improvement experienced over the last 25 years.⁵ This realistic cohort approach adds slightly less than two years to the life expectancy of the 65-year old based on the 2011 census data.

If we extend this cohort life expectancy approach for another seven years (i.e. bring it forward to 2018), the average life expectancy for a current 65-year old becomes 87.1 years for males and 89.3 years for females.⁶

We also need to recognise that life expectancies vary across socioeconomic groups. In short, those in the highest socioeconomic quintile (the top 20%) have a longer life expectancy than those in the lowest quintile. AIHW data suggests that the difference between the median age of death for those in the highest and lowest quintiles is 4 years for males and 3 years for females.

As many clients of financial advisers will be in the highest quintile, it's appropriate to reflect this fact in the life expectancies used in retirement planning. This means that for many clients of financial advisers, a realistic average life expectancy is 89 years for males and 91 years for females.

Of course, life expectancies reflect averages. This means that about half the retirees will actually live beyond these ages. Indeed, about one quarter will live an extra five years or more - that is, beyond 94 for males and 96 for females.

Therefore, as we consider the likely life expectancies for many clients, we should not be using any number in the 80s. A figure closer to 95 is both more realistic as well as providing a little buffer in case the individual lives longer than the average.

To summarise, Figure 1 shows the components of life expectancy for a typical 65-year old person in 2018.

Figure 1:  Likely life expectancy for today's retirees

Retirees.jpg

Source: Mercer

ENDNOTES

1. Australian Life Tables, 1946-48.

2. ibid.

3. Australian Life Tables, 2010-12.

4. ibid.

5. ibid.

6. Australian Institute of Health and Welfare, Mortality Over Region and Time books, Socioeconomic group 2011-2015.