By Frank Grossman, FCIA
“Circumstantial evidence is a very tricky thing,” answered Holmes thoughtfully. “It may seem to point very straight to one thing, but if you shift your own point of view a little, you may find it pointing in an equally uncompromising manner to something entirely different.”
The Boscombe Valley Mystery by Sir Arthur Conan Doyle
A seemingly inexorable trend of mortality improvement has emerged over the past century. Declining rates of population mortality in the developed western world have been variously attributed to several causes, including: i) improved dietary standards; ii) better public health institutions and programs (e.g., childhood immunization); and iii) better public infrastructure (e.g., proper treatment of waste). However, whether the beneficial influence of these diverse factors has substantively run its course remains an open question.
In principle, the several risks of out-of-sample extrapolation are widely acknowledged. Yet, how well does this projection technique serve contemporary actuarial practice?
Trend extrapolation, long familiar to pricing actuaries, appears to have gained greater currency even among valuation actuaries. Is reflecting expected mortality improvement in one’s work as easy as the rote reduction of mortality rates by one or two per cent each year over the entire projection period?
There seems to be some risk in taking such an approach – without digging deeper to examine the dynamics of life expectancy. Hence the need to consider how likely the drivers of past mortality improvement may be sustained into the future, and – critically – what new drivers of change may emerge in their stead.
Our daily bread
The growth of agribusiness has led to the advent of monoculture farming on a massive scale in pursuit of the twin financial goals of greater yield and more profit. Many varieties of plants have been hybridized to grow more readily and to enable their produce to better withstand the vicissitudes of processing and long-distance transportation. A case in point of a vegetable that is not grown commercially is the coreless Nantes variety of carrots, which is much prized by backyard gardeners but too fragile to be harvested mechanically.
Nowadays those living in northern climes can easily dispense with traditional winter root crops and opt for leaf lettuce all year; and tart currants and gooseberries that may be hard to find at the grocery store have been widely supplanted by all-season celluloid strawberries from California. The obvious result has been ever more food at lower prices – recognizing that recent price jumps, variously attributed to supply chain issues during the pandemic, are a reversal of the historical trend. A concomitant outcome, however, is that many people now eat a narrower range of foodstuffs that are grown ever farther away. The result has been less “diet diversity” and a growing risk of relying on a shrinking group of suppliers.
Another factor to consider is that humans have consumed genetically modified (GM) plants, or livestock that have been fed GM grain, for more than a generation without apparent ill effect. The improved genetics of these new varieties deliver many wholly desirable qualities, such as enhanced yield, hardiness, visual appeal and flavour. But genetic modification is also undertaken to make plants resistant to pesticides and herbicides, and it’s the effect of the unseen residue of these deadly compounds on our produce that is often overlooked. This is aside from the presence of microplastics in our food chain – small enough to even cross the blood-brain barrier – which brings new meaning to an old phrase, “You are what you eat.”
Our public health
Much has been written about the looming threat posed to baby boomers’ quality of life by the obesity epidemic. Many factors contributing to the onset of this affliction are lifestyle-based and rooted in the pursuit of convenience: poor diet (e.g., overreliance on processed food) and lack of physical activity. In this light, Type 2 diabetes is an example of a disease that may yet assume a different mortality dynamic going forward than in days past.
The indiscriminate use of antibiotics – by the medical profession as well as agribusiness – has diminished their effectiveness, and it risks turning back the clock on the treatment of infections. Additionally, a sizeable backlog of medical diagnostic work and surgeries accumulated during the pandemic, which seems likely to persist. Today roughly one in five Canadians does not have a primary care physician. This reflects a general decline in the capacity of our public health care system, let alone an affront to the equal access requirement of the Canada Health Act.
There are also many questions relating to the direct and indirect effects of COVID-19 that are as yet without definitive answers in the wake of the pandemic. For example, how prevalent are the cases of “silent suffering” attributed to long-COVID? Will there be a lasting impact on the mental health of Canadians due to the intermittent periods of enforced lockdown isolation, particularly among the younger generation? What to make of the significant number of so-called “excess deaths” above pre-pandemic levels – after accounting for the deaths officially attributed to COVID-19 – and their possible link to the vaccination programs?
Our environment
Habitat destruction has led to the extinction of untold species of flora and fauna. Might environmental degradation similarly translate into higher human mortality rates too? Our need for clean air and water is paramount – though usually taken for granted – yet climate change is changing the rules of the game. This year’s record-shattering wildfires in northern Canada have affected the air quality of southern communities, plainly demonstrating that what happens “over the horizon” really can matter. Heat stress, particularly as wet-bulb temperatures approach 35°C, is an emerging peril for young and old. Combine hot summer temperatures with humidity and poor air quality, and the result is a veritable triple threat to human health.
The actuarial angle
A perennial challenge when evaluating a future state scenario is to harness one’s actuarial imagination while at the same time avoiding Chicken Little alarmism. The focus should be on a richly detailed qualitative scenario – one that delivers a coherent storyline – all in support of the quantitative analysis. Actuaries have traditionally not been afraid to challenge conventional wisdom.
The key is to continue to use rigorous methods that weigh the relevance of historical data and assess both the uncertainty of current conditions and the outlook for change. In so doing, it is very important to periodically challenge one’s own assumptions – if only to arrive at the proper degree of actuarial objectivity and to minimize bias.
The shadow of uncertainty alone is not a sufficient excuse for inaction but rather an opportunity to revisit one’s margins for parameter misestimation and model misspecification.
When thinking about future mortality, there are several factors that could slow its rate of improvement that are surely worth considering. The point being that one actuary’s established fact may simply be another actuary’s placeholder or working assumption. A couple of paragraphs after the excerpt cited at the beginning of this article, Sherlock Holmes remarks, “There is nothing more deceptive than an obvious fact.” Therein lies the wisdom of shifting one’s vantage point to obtain an alternate view of future mortality – if only for a moment or two.
This article reflects the opinion of the author and does not represent an official statement of the CIA.
The following comments were shared by readers:
Joe Nunes: Frank is asking the right questions. Simply projecting the future from past patterns is surely not a robust way to anticipate how the world will unfold. For many years, before COVID 19 was a thing, I have been ‘predicting’ that mortality improvement will slow and maybe even reverse. My Generation X is not as healthy as the folks born in the 20s, 30s, and 40s in terms of lifestyle and diet. It is those earlier generations on which we base our longevity expectations and the roots of their longevity likely doesn’t apply to many of us earlier in the curve.
My view doesn’t worry me personally, my future will be what it will be. But when we are doing the math to properly price and reserve insurance and annuities we need to be continuously refining our assumptions about the future.
