The 3 Digit Age

There’s much I love about being a life actuary. It’s a great blend of skillsets requiring mathematical acuity, humility, long-term perspectives. There’s constant intellectual challenge, continuous learning, diverse career paths, medium to high earnings possibilities, and last but certainly not least; I love the pride of belonging to a self-regulated profession which emphasizes maintaining public trust in our work as a whole. That last one I didn’t care for much in my twenties. Now in my forties, it’s a top feature not a bug.

I’ve joked with friends and colleagues alike. I don’t think I want to live to age 100. But what are my chances for reaching age 100? Many think the probability is increasing. Others think it’s just a numbers game with growing population it’s inevitable to have more centenarians. I’ll give my two cents with some data below. But first, the disclaimer…

“No man should know too much about his own destiny... If I know too much about my own future, I could endanger my own existence” -Doc Emmitt Brown.

Now, for the parts I don’t like about being a life actuary. Obviously the exam process is a drag. I had to put “real life” on ice for 7 years while I became credentialed. But the purpose of this post is the other thing that eats away at my soul being a life actuary. I spend most weeks at work either reviewing mortality results or researching a new life table with varying mortality rates and slopes. I think coroner would be a worst career path, but life actuary studying death trends is a close second for me given my terror thinking about my own collective future ending.

There has been a lot of discussion lately around life expectancies falling in the United States due to the lingering impact of Covid and increased risk due to poor economic conditions, increased illicit drug use, increased anxiety and depression.? This decrease in US Population life expectation are statistics from birth and even small changes in mortality for young and middle ages will have a dramatic impact on the calculated life expectation.

For focusing on retirement and late age life expectation, one must start from a later age to properly compare/contract probabilities for reaching a particular age.? And also the actuary should consider the appropriate mortality table in estimating the survival curve and mortality expectations. For example, https://mort.soa.org/ has over 220 mortality tables for Insured Lives Mortality in the United States.

I recently reviewed the 100,000 unique insured lives with the goal of estimating the probability of reaching age 100. There are over 52,000 mortality events from 2001 through to the end of 2023, but this ranges across all ages. Of the remaining 48,000 Insureds who remain alive as of 12/31/2023, roughly 700 are over age 100.? This metric of 700 for every 100,000 is quite high compared to published population statistics.? Japan has the highest proportion of centenarians as a country at roughly 74 per 100,000.? The United States does not make the top 10 list and is estimated to be roughly 29 centenarians per 100,000 people.

The reason for this healthier longevity is selection bias based on starting age and the health/wealth status.? We’re not dealing with overall population, but a cohort of lives who actively bought expensive life insurance coverage 15-20 years ago, and the premiums on these high-face policies have continued to be paid despite the steep increase in price at later ages.

Now, let’s narrow the scope a little to only the cohort of lives who have had the chance to reach age 100.? This means filtering for insured lives who were born prior to 1924.? Today is the 9th of September 2024. Anybody born 12/31/1923 and prior has now had the chance to reach this illustrious three-digit age, with enough wiggle room to sort out any incurred but not yet reported noise.?

Using mortality data available at my present company, I have identified over 9,000 insured lives with date of birth prior to 1924.? As the above graph shows, the average age of death for this cohort is age 93.? The graph has a nicely shaped normal distribution. The standard deviation is +/- 5 years.? That means there’s a 68% chance to live between age 88 and 98.? There is a 95% chance to live between age 83 and 103 based on this healthy cohort of 7,000 insured lives born prior to 1924. The centenarian rate for this cohort was 7.6%, higher than one would expect using a population-based mortality table.

Life settlement mortality does not follow population mortality patterns.? This set of lives is the high net-worth sub-cohort of the life insured cohort.? All data above are purely death and exposure statistics.? There is no underwriting or life expectancy calculations.? Pure facts and statistics.

I look forward to discussing this and anything else mortality related.? Start with a joke from “Back to the Future” 1, 2, or 3 if you want to catch my attention.

#lifeactuary #3DigitAge #lifesettlements