The average life expectancy of a newborn human in 1900 was 32 years, and according to Our World in Data, by 2021, that more than doubled, to 71 years.
So, what are the chances then that the next 100 years or so brings about a similar jump?
Pretty much zero, according to Implausibility Of Radical Life Extension In Humans In The Twenty-First Century, a new article published in Nature Aging from longevity experts S. Jay Olshansky, Bradley J. Wilcox, Lloyd Demetrius and Hiram Beltran-Sanchez.
The quartet’s research used demographic survivorship metrics from national vital statistics in Australia, France, Hong Kong, Italy, Japan, South Korea, Spain, Sweden, Switzerland and the US from 1990 to 2019 and found that improvements in life expectancy had slowed down since 1990, ultimately concluding that under the best of conditions, only about 15% of females and 5% of males could survive to 100 years of age in the 21st century.
The impact of the findings will be felt across health systems, urban planning and retirement planning the world over, and certainly, for companies in the longevity and morality markets, as well; but for Olshansky, the cat has been out of the bag for a long time.
“We made this prediction – that life expectancy had to slow down – in 1990. It’s not about curing this and that and AI – unless it influences aging, it’s not going to work. Aging is currently immutable.”
Longevity risk is prevalent in many industries, ranging from the niche, such as life ILS and life settlements, to the broader, such as reverse mortgages and defined benefit pensions.
In life settlements and reverse mortgages in particular, the findings of the research may be felt less keenly.
“Life settlement fund managers are obviously concerned about right tail risk, and rightly so. That still exists at the individual level – after all, those who are most likely to make it to 100 are those in their late 90s, so a life settlement fund with these policies still has this risk,” said Olshansky.
“And there’s also a legitimate concern about right tail risk in the Hong Kong reverse mortgage market. 12-13% of Hong Kong females already live to 100 – that’s way higher than anywhere else in the world. Reverse mortgage providers can be exposed here,” he added.
But at the life insurance and population levels, the impact could be profound.
It is generally accepted that life insurers do account for potential improvements in mortality in their actuarial models, but if they have been taking too conservative an approach, then making adjustments could mean that life insurance becomes more accessible to the broader population.
In the US, for example, industry group Life Insurance Marketing and Research Association (LIMRA) and Life Happens publish The Insurance Barometer, an annual study that tracks the perceptions, attitudes, and behaviours of adult consumers in the United States, with a particular focus on life insurance. This year’s edition reports that while around half of Americans have life insurance, 42% say they need more and the main reason given for not purchasing life insurance is that it’s too expensive.
Will this new research mean that life insurers can revisit their mortality modelling effort to identify where they might have been too conservative on risk – and perhaps, bring down pricing?
“I’m guessing that’s exactly what’s going to happen. It’s obviously more nuanced than that – they will need to do a little more homework to make sure the dynamics apply directly to the populations they are underwriting, as there will be variations. But I do think that many of them will look at their existing pricing and modelling assumptions because they are probably going to be wrong, and the longer the forecasts go to the future, the more wrong they will be,” said Olshansky.
It’s a similar situation in the pension risk transfer (PRT) market. These deals can involve thousands of retirees: Just recently, RAC secured a deal with Aviva to insure the pension liabilities of approximately 19,000 people. General population mortality data is also applicable to this industry, and according to David Blake, Professor of Finance & Director of Pensions Institute at Bayes Business School, getting this right is no easy task.
“Getting the trend increase in life expectancy right is the key job of the pricing actuaries appointed by life insurers. If they underestimate the trend, then their pricing will be too low, and this will eventually eat into reserves. If they do the opposite, they will be overpricing. It is a fine line,” he said.
If the models are indeed overcompensating for increasing life expectancy, however, the likelihood that there would be any kind of pause in bulk purchase annuity transaction activity whilst insurers review and tweak their pricing is low.
“Pricing changes every day in the light of new information. If the new information points to a significant slowdown or even a flattening of the trend, then prices will respond immediately. Each insurer will look at what its competitors are doing to confirm that it has made the right decision – the prices won’t jump, they will change slowly. But there will be no pause in PRT activity,” said Blake.
Only time will tell in terms of whether life insurers around the world adjust their pricing, whether that be for life insurance policies, pension risk transfer deals or annuities. After all, life expectancy is not the only determinant of price. But according to Olshansky, they have begun to take action.
“We’re already seeing actuaries at insurers scrambling in terms of how to adjust their approach to future mortality,” he said.
“Because if insurers have been building in a 1-2% mortality improvement each year into their models, they have been mispricing their book for the last three decades.”
