An increase in UK death statistics is concerning for longevity models, so relied upon in the pensions industry
A pessimist sees the difficulty in every opportunity; an optimist sees the opportunity in every difficulty. Put me down as a pessimist. And yet, when it comes to the implications of increasing life expectancy, I fear my brethren are guilty of the glass-half-full perspective.
Sure, for advisers working with clients, life expectancy is a literally unpredictable element that complicates investing for or in retirement. The duration of necessary cashflows is unknown because life expectancy is unknown. Since longevity insurance is unpopular with clients, averages provide a rule of thumb but no more for advisers constructing life plans.
Yet, for DB schemes and the government, the cost of meeting their pension promises depends, among other things – investment returns, inflation, regulation – on how long members live.
Longevity is a critical variable in discounting the present value of future cashflows. Indeed, a one-year difference in life expectancy today would have five times the effect on a pension scheme’s liabilities than it did in 1990 due to the impact of lower for longer interest rates, according to Club Vita.
Nonetheless, our professional lives risk blinding us to the public good. Public Health England’s recently published Health Profile for England reminds us how lucky we are, as well as highlighting new challenges.
In 1841, when the UK’s first actuarial life table was published, the average newborn girl was not expected to see her 43rd birthday. A newborn girl today can expect to live past her 83rd birthday. Baby boys’ life expectancy over the same period has risen from just 40 to 79 .
More time on this planet doing things that give us pleasure (and pain) – it is hard to think of anything more profound. Life is precious and the most basic measure of a society’s progress is its success in extending it. The gains of rising life expectancy are, in this sense, priceless.
So I was disturbed by the Office for National Statistics’ (ONS) provisional death data for the week ending 12 January 2018, which cite 15,050 deaths against the normal recent level of 13,170 for the second week in January. That is 1,900 extra deaths – 14 per cent up on normal.
Nor is this a one-week blip. The same figures for the middle two weeks of December projected more than 2,000 extra deaths compared with the normal trend (an average of the five years 2011-2015). As the UK population ages and grows, one expects the annual death numbers to rise – but not at this rate.
Death rates in 2017 were around 5 per cent above the recent average. Indeed, life expectancy at birth has flatlined since 2011.
As Professor Danny Dorling puts it, “For the first time in well over a century the health of people in England and Wales as measured by the most basic feature – life – has stopped improving.”
Life expectancy for a woman in the UK is now lower than in more than half the countries of the European Union – the wealthier nations with which the UK would like to compare itself such as France, Germany, Italy, but also Greece, Portugal and Spain. Life expectancy rose by a whole year in Japan, already the nation with the longest, in the same period in which it flatlined in the UK, 2011-15.
The impact of these past five years of higher than expected death rates on future projections is dramatic. The ONS now estimates that, by 2041, life expectancy at birth for women will be 86.2 and for men 83.4. Both figures are almost a whole year lower than projected in 2014.
So what has happened?
Flu epidemics are often cited as being worse than in previous periods, whether because of mutations in the virus or greater gaps in health and social care provision, as budgets have been reduced in the austerity years.
The initial spike in mortalities was among older women living alone, a group disproportionately reliant on such services as social care home visits, and the availability and quality of nursing homes.
A change in modelling technique might also be relevant. In reaction to a history of actuarial science underestimating future life expectancy rises, the ONS has moved to a dynamic projection that is more sensitive to recent data.
If the flatlining in life expectancy rises turns out just to be a five-year blip rather than a long-term trend, the model will eventually recalibrate. In what are very long-term projections, statistical blips are, of course, eminently plausible.
But, in all honesty, no-one knows for sure.
What we do know is that the ONS is still projecting that life expectancy will continue to rise across the next 25 years. This is a good thing. Even when people living longer gives advisers, governments and pension schemes a headache.
Gregg McClymont is head of retirement at Aberdeen Standard Investments