Hello. In this session, I'm gonna be talking about longevity risk, give you an update on what's been happening in terms of UK mortality, UK longevity, and also looking at how you can hedge that risk and in particular a new solution that we've been working on.
I thought I'd start with some nice headlines that we've had over the last few months in terms of longevity and mortality. All a bit depressing, [chuckles] in terms of the impact that AI could have on the human race, and also, more realistically perhaps, the impact that climate change has been having on UK mortality.
So yeah, start with a suitably depressing theme, but thankfully it's not all bad news in reality.
So we'll have a look at what's actually been happening in terms of UK mortality on the next slide.
So here we look at mortality rates for males aged sixty-five, going back to the year two thousand. So this is the rate of deaths amongst that population, and we can see that over the start of the decade following the millennium, we had really significant reductions in mortality rates.
So this was people were living longer than they previously had before, and there was a couple of big drivers for this.
The first of all was statins, so people taking these drugs, which helped to reduce the incidence of cardiovascular disease, and also a fairly massive reduction in smoking, which had started years before. So if you go back to the nineteen seventies, fifty percent of the UK population or adult population were smokers, and now it's about ten percent. So those two drivers really pushed down mortality rates over that period, so that's all good news.
And then for the next sort of six or seven years or so, we saw that mortality rates were relatively flat. So the impact of the statins and the reduction in smoking had sort of come through and population mortality was fairly stable.
And then obviously year two thousand we had COVID and we saw there, as you see on the right-hand side of the chart, you can see that mortality rates sort of picked up at that point. They increased, not massively, they increased by about seven percent during COVID at its worst.
But now we've seen mortality rates come back down and in particular, if you look at the green bar at the end, you can see that mortality rates are now lower than they were at the start of COVID.
So you can see on that final stat on the chart there, mortality rates are currently six percent below the average over the last ten years. So that's really good news in terms of UK population mortality.
If we look then at what that means for life expectancy, so this is obviously how long you expect somebody to live for from their current age. So here we look at again, sixty-five-year-olds, both males and females, and we look at the life expectancy back from about two thousand and eight up until today.
And you can see that at the start of the chart, life expectancy was relatively stable and had been for a while. So from two thousand and eight to sort of two thousand and fifteen, life expectancies for females were just shy of twenty-six years, and for males it was around twenty-three and a half years. So fairly high, fairly stable life expectancies.
And then you can obviously see in the chart that those life expectancies start to fall from two thousand and fifteen onwards.
And that wasn't because people were dying more quickly. As we saw in the chart on the previous page, mortality rates was fairly stable during that period. It's more due to the fact that the actuaries realized that they'd over extrapolated the sort of trends that they'd seen in falling mortality rates and had sort of extrapolated that too far into the future and come up with these life expectancies which in retrospect seemed a bit too long lived.
So between two thousand and fifteen and twenty twenty-three, they started to reduce the life expectancy assumptions. So life expectancy for females, instead of being sort of close to twenty-six years, fell... Life expectancy at age sixty-five, I should say, fell from twenty-six years to around twenty-four and a half years.
So a decent reduction, one and a half years over that sort of seven or eight year period.
And then obviously with COVID happening, that had a negative impact on life expectancy and it also meant that the actuaries sort of put their models to some extent on hold because it wasn't clear what would happen because the mortality data that they had in the years twenty twenty, twenty twenty-one, et cetera, wasn't really something that was going to be repeated.
And the models rely very much on historical data, so the models sort of went on hold almost for that period.
But now, as a result of improvements in mortality coming through again, we're starting to see a slight uptick in life expectancy. So we can see it here on the chart obviously from twenty twenty-three onwards. For females, it's a very modest increase of a couple of months, and then for males it's more like five months.
So still a fairly modest increase, but meaningful and one that for a typical pension scheme might increase the liability value by one to one and a half percent depending on the makeup of the membership.
What could the future drivers be that might push these life expectancy values even higher?
And there's obviously some fairly big tailwinds, the main one being the fact that the NHS is under quite a lot of strain at the moment and we hear some awful statistics in terms of waiting times et cetera and people in corridors.
But there are some, I think some sort of green shoots and some prospects that could increase life expectancy fairly meaningfully over the coming decade or so.
I guess the first one of these is the anti-obesity medicines, the GLP-1s. Obviously they can... when they're taken, they lead to many more health benefits beyond just weight reduction. The impact on the pensioner population, it's probably fair to say, is quite small at present, but I think that is something that could grow significantly.
And then in terms of cancer treatments, early detection, improved treatments are all having an impact.
And then artificial intelligence is the final one as well. It's not all about killer robots. It could really sort of speed up drug discovery and quicker diagnosis.
So there's a few factors there. I wouldn't bet against life expectancy increasing in the future fairly dramatically, 'cause I think there's a lot of science pushing in that direction.
And then sort of more practically, what might an increase in life expectancy mean for a pension scheme?
Well, here we've got a sort of simple example of a scheme. If we start on the left-hand side, it's 95% funded on a low dependency basis, and it's running on indefinitely.
And given where they are, their funding level and their objective, this scheme could invest 60% in corporate bonds and 40% in gilts, and it would expect on some reasonable assumptions to be able to meet all of its future cashflow requirements.
So then we switch to the right-hand side.
In this case, we have the same scheme, but their scheme actuary has increased the life expectancy assumption of the membership by one year.
That would push the liability value up by about 3%. So as a result, the funding level would come down to 92%.
And if the scheme remained invested in its 60% corporate bonds, 40% gilt portfolio, it would run out of money at some point because its pensions are now going on for longer than thought.
And in this case, the scheme would need to increase the allocation to corporate bonds from 60 to 85% if it wanted to meet all future expected pension payments.
It's only a fairly modest additional investment return of 35 basis points, but when you've already de-risked your scheme quite significantly, the 60% corporate bond, 40% gilt portfolio is fairly low risk.
Having to find 35 basis points means that you have to increase the allocation to your risky assets, your corporate bonds in this case, fairly significantly.
So it can have a fairly meaningful and immediate impact. People often think of longevity risk as being a bit of a slow burner. But if your scheme actuary changes their life expectancy assumptions, then the impact is immediate.
So what can you do about that longevity risk?
Well, for a start, you can hedge it, and that's what we're gonna talk about now.
So here we just have a really, really simple example of an individual pensioner. They're receiving £10,000 per annum, and for this particular individual, we think they're going to live for another fifteen years. So they have a fifteen-year life expectancy, and that's shown in the top chart by the sort of blue-green bars.
But we don't really know, obviously, how long they're gonna live. They could live less than that. They could live a lot longer.
But as trustees, what you would do is you would put aside assets such that you can afford to meet those pension payments for the next fifteen years.
And then if they live fifteen years, perfect, you've met all their cash flows quite happily.
But obviously, as a scheme, you're financially exposed to the risk that they might live longer than you expected, and the scheme would then run out of money and would have to find some additional contributions.
So that's without a longevity hedge in place.
If you put a longevity hedge in place, you can remove that risk.
So what you do effectively is you find a willing counterparty who wants to take on the longevity risk, and in all cases, that's a reinsurance company.
They have lots of mortality risk, so that's the opposite of longevity risk. And so for them, being able to take on some longevity risk is actually very beneficial because it offsets the mortality risk they've got.
And then you agree with them to lock into this fifteen-year life expectancy assumption, and so you pay cash flows for those fifteen years.
But in addition, you pay a small amount on top of that to the reinsurer as well, the risk fee.
So rather than having this cash flow stream that you have to pay for an unknown period of time, is it five years, fifteen years, twenty years?
With the longevity swap in place, you lock into a series of cash flows which are fixed for fifteen years. They're slightly higher than what you think you might have to pay under a best estimate scenario, but that's the cost effectively of passing the longevity risk across to the reinsurer.
Because the pension scheme now, with this contract in place, pays these fixed cash flows for fifteen years to the reinsurance company, and in return, the reinsurer pays the pensions for as long as the individual survives.
