According to a recent article in InsuranceNewsNet, The Life Mortality Improvement Subgroup of the Society of Actuaries’ (SoA) Mortality and Longevity Oversight Advisory Council (MLOAC) is seeking to switch from general population data to insured population data as an input into its mortality modelling effort.
The reason cited in the article is because, despite the similarities in the data at age cohorts lower than 80, “insured population mortality improvement sharply drops” after that.
A change in approach by an actuarial body is hardly new; as the understanding of mortality evolves, as must the models. Indeed, the UK’s Continuous Mortality Investigation made the most significant changes to its model in many years just this year, the result of a lengthy feedback process among participants. Pension schemes and insurers in Britain all use the CMI model as a fundamental input into their own approaches.
But, if the SoA does make the change, how much of an impact will this have on life expectancy (LE) forecasts in the life settlement market?
Little, because of the approach that some in the space already take.
“Many life settlement LE providers already rely heavily on their own proprietary, insured-population tables tailored to the life settlement cohort. They are not likely to change underwriting practices because of any changes to the general population mortality tables,” said Liam Bodemeaid, Actuarial Consultant at Actuarial Risk Management (ARM).
The life settlement cohort is, in the grand scheme of insured populations, small. In 2023, 3,181 transactions closed the secondary market, compared with 9.5 million new life insurance policies issues in the USA in the same year.
That is not comparing apples to apples, as the latter includes all age cohorts. But still, it illustrates the small size of the life settlement market when compared to the overall space. Additionally, mortality nuances permeate the life settlement group which are a root cause of the aforementioned tailored approach that some life expectancy providers take – with a notable trend emerging.
“There are important distinctions between the general and life settlement populations. Life settlement insureds tend to come from higher socio-economic groups, they are typically wealthier, more educated, and have greater access to quality healthcare,” said Bodemeaid.
“And the longevity gap between high and low-income groups has only widened in recent years, with evidence suggesting improved mortality outcomes among life settlement cohorts.”
The nature of underwriting insureds in the life settlement market – in the sense that life expectancy predictions are very specific to the individual – also means that comorbidities have a significant impact on any given insured’s projection. According to Bodemeaid, this reinforces a reason why a LE provider in the market is likely to be influenced less by a change in a broader dataset.
“In life settlements, mortality must be assessed on an impairment-specific basis. Medical advancements progress unevenly; some impairments see breakthrough treatments sooner than others. So, applying blanket adjustments risks introducing distortions into underwriting,” he said.
There is another reason why life expectancy data for the life settlement cohort differs from the general population. In primary life insurance, selection occurs after policy issuance when insureds need to prove they are of an acceptable health at underwriting. This means that mortality is lower for an insured who recently took out a policy compared to someone else of the same gender, age and smoking status who took out their policy several years beforehand, generally. Not so in life settlements.
“The inverse is true in life settlements, in that insureds generally need to prove impairment to receive their settlement amounts. Therefore, the structure of the mortality tables between these two population subsets needs to be inherently different and is another reason why LE underwriters use bespoke life settlement mortality tables to underwrite on the secondary and tertiary markets,” said Bodemeaid.
But despite the comparatively smaller cohort of insureds that the life settlement market deals with, one interesting thought experiment could be in whether pension plan sponsors and life insurers could use the data that the life settlement market has to tweak their mortality modelling even further.
That is because the life settlement market is regularly underwriting American Seniors, whereas the life insurance industry is not. And not every policy that comes to market is purchased by an investor or warehouser, so LE underwriters have much more data than the 3,000 or so policies that are sold on the secondary market each year – especially since these insureds are re-underwritten as and when their policies transact in the tertiary life settlement market.
Still, if there was to be a ‘mortality table’ for the life settlement market specifically, it could well show additional gains over the coming years thanks to a list of emerging trends.
AI-driven medical innovation – the acceleration of drug discovery, diagnosis, and personalised treatment pathways – is just one. The targeting of the biology of ageing and the increasing take-up of weight loss and metabolic drugs are two others.
“The life settlement population, by virtue of wealth and education, has more access to cutting-edge diagnostics, and generally follows more proactive and preventive health regimes. And as sequencing costs drop and datasets grow, therapies tailored to an individual’s genetic profile will become increasingly common and high-net-worth individuals are more likely to benefit early from this next wave of healthcare personalisation,” said Bodemeaid.
“While these developments may not change actuarial assumptions overnight, they reinforce the need for underwriting frameworks to evolve with the times. They also present a long-term risk to mortality assumptions, particularly in life settlement portfolios with an older, affluent, and health-engaged insured base.”