Most people wonder how long they have left on this world, and what they’ll do with that time, but few people are as preoccupied with mortality as the professionals employed to assess risks for pension schemes and insurance companies. And it’s fair to say that these actuaries don’t quite come at it from the same point of view as the rest of the population; your early demise might just be their idea of an early Christmas present.
Evidence of this follows on from the latest report from the Continuous Mortality Investigation of the Institute and Faculty of Actuaries. Based on new assumptions, it suggested that future improvements in death rates will be significantly less than expected, even last year. Some reports have linked that change to a sustained spike in winter deaths. In particular, life expectancy for a 65-year-old man fell by 1.3%, which means nearly four months less of life. For a 65-year-old woman it was 2% lower – or nearly six months less. Not a huge shift, you might think, but it has got the pensions actuaries who work with defined benefit (DB) pension schemes very excited indeed.
The reasoning is as follows. If people’s lives are expected to be shorter, then they won’t be drawing pensions for as long. This means that the value of these pensions – that is, the money that needs to be set aside to pay them – falls. This is big news for DB schemes, which are committed to paying out a certain amount to retirees. The other kind of workplace pension, defined contribution or DC schemes, only pay out from the amount you have managed to accrue over the years. Your death is of less interest to them, as longevity doesn’t directly affect contribution rates.
In fact, the estimated impact on UK DB pension schemes is somewhere between 2% and 3% of their liabilities, equivalent to tens of billions of pounds, which is why pensions actuaries are so excited. But it doesn’t stop there.
If DB pension schemes benefit from this fall in projected longevity, then so do the employers that support such schemes. This is because employers have to make up the difference if there isn’t enough money in a pension scheme to pay the benefits – and a fall in life expectancy makes this less likely. If the employer doesn’t have to pay so much into the pension scheme, then it can pay higher dividends to shareholders and is less likely to be driven into bankruptcy through pension costs, so employees can sleep more soundly.
And it’s not just pension schemes that could benefit: individuals might be better off too. Although contributions to DC schemes wouldn’t change, shorter life expectancy means that retirement funds don’t need to last as long. It also means that and it should be cheaper to buy an annuity, an insurance policy that gives you a fixed annual payment once you retire for as long as you live.
So the fact that life expectancy has fallen is nothing but positive: for individuals, pension schemes, companies, the economy, for everyone. Sort of.
The weaknesses in this argument should be clear. You’d have to be extemely optimistic to view the shortening of your life expectancy as a good thing. Being less likely to die broke is positive, I suppose, but one can take the suggestion to “always look on the bright side of life” too far.
And although the benefits are clear, it’s not really a game changer for DB pension schemes. True, the numbers are significant in the context of longevity risk. But DB pension scheme deficits – the differences between the assets pension schemes have and values of the benefits they need to pay out – can (and do) change dramatically over the course of a year. That can be due to changes in both long-term interest rates and the value of schemes’ investments.
Consider the aggregate deficit of UK pension schemes on the basis used by the Pension Protection Fund. In April 2016, this was £188.7 billion; in August 2016 it had ballooned to £413.1 billion; and in January 2017 it had shrunk back to a “mere” £196.5 billion. Suddenly, a few tens of billions doesn’t seem that material in the grand scheme of things.
More important, though, is what these mortality figures represent. When calculating this sort of life expectancy – known as cohort life expectancy – we need to make assumptions about how mortality rates will change in the future. These assumptions are derived using past data to help to project future changes. The data that most influences future longevity assumptions is often relatively recent.
In other words, assumptions as to what might happen to mortality in 30 years’ time might largely be driven by what has happened in the last five years or so. This means that small changes in mortality rates now can mean huge changes in what we expect longevity to be in the long run.
The approaches used to derive mortality projections are not badly structured, but there is always a huge degree of uncertainty over what future mortality will be for the population. This is an important point, because it is average life expectancy that we are talking about – and very few people are average.
To put it another way, there remains much more uncertainty over how long each individual will live, whatever the projection of an average might tell us. You may well have a life expectancy of 24.1 years, as the CMI assumes you would have as a 65-year-old woman; but you could drop dead tomorrow. Or you could live another 50 years and get a telegram. In this context, how you live your life can also have an impact on how long you live – but one look at Keith Richards shows that even this is not definitive.
Paul Sweeting does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond the academic appointment above.