The postcode approach is best in analysing life expectancy

Alan Collins

or Subscribe to Feed

Estimating life expectancy is an important part of an actuary’s job. Last month’s Office of National Statistics (ONS) report on the issue of life expectancy certainly brought a real focus to this important aspect of our role and, as a man who lives in one of the lowest ranked areas for longevity in the UK, it also cast a bit of an unwelcome shadow over my day.

Confronted by headlines such as ‘Scotland the Grave’ and ‘Increase in North-South Life Expectancy Divide’, the Scottish media highlighted how the recently published ONS survey showed how the average UK man will live until he is 77.9, compared with only 75.4 years in Scotland. The comparable figures for women are 82.0 and 80.1 respectively.

Somewhat worryingly for me, the average male in Glasgow will die aged 71.1 years. Unsurprisingly, given the health issues that continue to plague many parts of this city, this is the lowest for any area in UK. This is in sharp contrast to Kensington and Chelsea where the average man can expect to live for 84.4 years, exposing a staggering gap of over 13 years in life expectancy between two regions of the same country.

There are important lessons in the ONS study for actuaries, as well as for sponsoring employers and trustees of defined benefit schemes. For defined benefit arrangements, it is the scheme (and ultimately the sponsoring employer) who is exposed to the risk of how long each member lives. The longer each member lives, the longer a pension will need to be provided for and hence the cost of providing the pension increases.

The study reinforces the need to consider and manage the risks associated with life expectancy on a scheme-by-scheme basis. For each additional year of life expectancy, the reserves required – and the ultimate cost of the scheme – increase by around three per cent. So taking the extremes above, the reserving requirements could vary by up to 40 per cent!

However, we need to be careful on drawing conclusions from this study on two fronts. Firstly, members of pension schemes tend to live a lot longer than those with no pension provision – which gives me some personal comfort in relation to the above statistics. This is borne out of many studies on life expectancy by insurance companies and by analysing data from self-administered pension schemes. Currently, most pension schemes assume that current pensioners will live into their mid-late eighties and that future pensioners will live into their nineties.

Secondly, in my view – and this is where views in the actuarial profession differ – it is not geography but socio-demographic factors that matter where life expectancy is concerned. If geography alone was a significant factor, why would the gap in life expectancy between neighbouring areas such as Glasgow and East Dunbartonshire be over seven years, whereas the gap between Glasgow and Manchester is less than three years and only around four years between Glasgow and areas in London? This point was summed up by Duncan McNeil, Labour MSP for Greenock and Inverclyde, who said: “someone in my community can expect to live around 10 years less than someone else who lives just minutes along the road in a better-off area.”

This is why analysis at a postcode level is so important, where life expectancy is considered on a street by street basis. This is the most effective and accurate current method for assessing life expectancy for most pension schemes. I would urge trustees of schemes and sponsoring employers to ensure this area is given appropriate attention and that a postcode analysis is carried out at least every three years to coincide with the formal valuation of a scheme’s funding level. The only reason for departing from this is if your scheme is large enough to conduct its own mortality study. However, this would only apply to schemes with thousands or even tens of thousands of pensioner members.

Before I get completely morose about the ONS report and the implied implications on a Glaswegian male like myself, I can take some comfort that there are other factors at play in determining what ultimately accounts for the number of innings we are likely to be on this earth. It is important that these are also accounted for on a wider scale for those of us who manage pension schemes to ensure we have the appropriate funding levels in place.

Alan Collins

Post by Alan Collins

Head of Trustee Advisory Services at Spence he provides actuarial, funding and investment advice to trustees and sponsors of ongoing defined benefit schemes.