The UK’s Continuous Mortality Investigation (CMI) carries out research into mortality and morbidity experience using data from pension schemes and insurers. The CMI released its base series of projections (known as the ‘92’ series, after the year its data came from) in 1999 and on a regular basis since.
The CMI is an offshoot of the Institute and Faculty of Actuaries and it released its latest set of projections in the middle of July (CMI_22). These projections are based on data from 2022, and unlike the previous two years, this time the committee decided to incorporate Covid-19 data into their projections, says Cobus Daneel, chair of the CMI’s Mortality Projections Committee.
Initially, the CMI decided to ignore Covid-19 and took the view that 2020 was an outlier and given that the model relies on extrapolating a projection from current mortality trends, including data from that year would distort the results. This view was repeated in 2021 but this year the committee decided to – partially – incorporate Covid-19-era data in its projections.
So, what changed the CMI’s mind?
“Despite a good start to 2022 with fatality rates falling close to 2019 levels we saw a rebound half way through that year and mortality rates ended up still being about 6% higher than in 2019. The CMI Mortality Projections Committee needed to decide if 2022 was again an outlier like 2020 and 2021, and what we concluded was that 2022 is likely to be somewhat indicative of the post pandemic trend.” said Daneel.
Daneel stresses that 2022 is just one data point, therefore the CMI decided to opt for a compromise and added a 25% weighting to the 2022 data when making its future life expectancy projections.
“The CMI concluded that 2022 is indicative for three reasons: much lower week-to-week volatility in mortality – unlike the peaks and troughs we saw during Covid. Secondly it was clear that Covid wasn’t the only driver behind the high mortality levels, and finally there is a growing belief that Covid may fade away but it will be with us for a while.”
The 25% weighting translates into a slightly more than half a year’s reduction in life expectancy for both males and females in the UK and it will shave roughly 2% off the average pension scheme’s liabilities.
“Or to put it another way, it’s the biggest year-to-year model change the CMI has made since it started with the current methodology in 2009,” says Daneel.
Daneel says at this stage it is difficult to say whether the non-Covid-19 deaths are secondary effects of the pandemic, particularly given the different impact that lockdown had on various age cohorts. Younger generations were more affected than older people and the committee chair says that this is due to socioeconomic reasons and will likely have a negative impact on that cohort’s life expectancy.
Daneel, however, says that evidence for a slowing down in life expectancy improvements, or even a fall, among younger people has been mounting since before the pandemic even struck.
The actuary says that the decline in UK mortality since the 1970s has been mainly driven by improvements in the treatments of heart and cardiovascular diseases, but this had already begun to stall around 2010.
“So there has been a slowdown in the main driver of the improvement in life expectancy since 2010 or 2011, but more since around 2017 we have seen that slowdown become an increase in mortality among younger age groups, which is a bit concerning.”
Daneel says there is no clear explanation at present for this trend but that it is most likely linked to rising obesity rates and changing lifestyles.
“The outlook in the actuarial community on life expectancy probably tends to be slightly less optimistic these days,” he says.
It’s not all bad news. Daneel points to countries in Scandinavia, where mortality rates have reverted to pre-pandemic levels, as proof there is scope for improvement in the UK. Likewise, he cites potential breakthroughs in the treatment for obesity and neurological diseases such as Alzheimer’s.
And while he says that the short-term outlook for life expectancy in the UK is bleak, this also needs to be put in the context of life expectancy being on an upward trend for a long period and there is still potential for future improvements.
“If you take a long-term view of history you will see that mortality rates have been decreasing for a very long time; we’re talking about centuries here. And five years, or even a decade of pausing or slightly slowing down, doesn’t mean that that trend has come to an end. We haven’t reached peak life expectancy yet.”
Daneel may be optimistic over the potential for life expectancy to renew its upward trend but the data from 2023 so far has not been encouraging, with excess mortality still around the 2022 levels. According to the chair, the CMI only placed a 25% weighting on the 2022 data because it was unsure whether the impact was short-lived. Two years of excess mortality back-to-back suggest the opposite may be true and further reductions in life expectancy projections could be on the way.
“So far, we have only placed a 25% weight on the 2022 data but at this stage we fully expect the next version of the CMI model to reduce life expectancies further. The question is just how much will it be reduced?
The experience in 2023 is pretty bad and it looks like mortality rates will end up at about 6% above 2019 levels. This is similar to 2022. So, we are still not seeing any improvements, and this provides extra evidence that the new normal has changed and that we may not get back to the pre-pandemic trend anytime soon.”