Slowdown in life expectancy increases for Canadians raises retirement planning and funding questions
By Paul Grant, FIA, ACIA
On September 20, 2018, the Chief Actuary at the Office of the Superintendent of Financial Institutions released the latest mortality experience factsheet for Old Age Security (OAS) recipients in Canada. This factsheet showed that the pace of observed increases in life expectancy for OAS beneficiaries has slowed over the past 15 years. In this article, we review the potential impact of the slowdown on employer-sponsored pension plans, as well as Canadians’ retirement planning needs, and the possible impact if longevity begins to accelerate again.
Figure 1. Average annual decrease in mortality rates by age group
OAS beneficiary life expectancies have been rising over the past 15 years, and they continue to rise. However, the pace of those increases has been slowing. The average life expectancy of a 65-year-old male in the OAS population increased by 2.3 months in the five-year period from 2003 to 2007, by 2.3 months in the period from 2008 to 2012, but by only 0.9 months in the period from 2013 to 2017. Figure 1 shows decreasing mortality improvements for older Canadians over approximately the past 15 years.
What has caused the slowdown?
Similar trends have been observed in the British and America populations. Actuaries in Canada and abroad had anticipated a slowdown in mortality improvements from the high levels seen in the early 2000s. Rapid gains in retiree life expectancy in the 20th century were explained by reduced prevalence of smoking and significant improvements in cardiovascular disease prevention and treatment. It was unlikely that these trends would be sustained in the future.
Furthermore, the data showed a “golden cohort” of people born between 1925 and 1945 who seemed to enjoy more rapid improvements in life expectancy during their lifetimes than generations born before or after them. It has been suggested that this effect arose from their avoidance of major conflict, healthier early childhood diets, and the expansion of public health systems after World War II, amongst other factors. It has been noticed that this “golden cohort” effect was wearing off as this generation aged and that it would ultimately fall to zero as this group disappears from the population.
However, the recent slowdown is more dramatic than anticipated. The OAS report does not get into the possible reasons. However, the slowdown in life expectancy improvements in many western countries has led to speculation that it may be correlated with lower real income growth and rising insecurity following the global financial crisis in 2008-9. The opioid epidemic, increased prevalence of the flu, dementia, reduced ability to sustain and repeat the positive effects of prior cardiovascular intervention, and other factors have also been cited.
What does it mean for employersponsored pension plans?
OAS recipients are a broader population than the population in employer-sponsored pension plans. OAS recipients include those who were out of the workforce during their working years as well as those who were self-employed or marginally employed. A strong correlation exists between life expectancy and affluence, so OAS trends may not be reproduced in workplace pension plan populations. The smaller size of the employer-sponsored plan population and more diffuse data sources also make the OAS results difficult to conclusively replicate for employed populations.
Most employer-sponsored defined benefit pension plans are funded on assumptions that already assume a significant future reduction in mortality improvement rates. If this reduction turns out to be happening faster than expected, there could be a slight improvement in funding levels. However, barring a reduction in life expectancy, this impact would be small compared to the effects of investment returns, contribution rates, and interest rates.
It is also possible that longevity increases could accelerate again, which could have a larger impact on pension plans. Medical interventions such as immunotherapy, genetic therapy, nanotechnology, or biological drugs, as well as medical improvements from “Big Data” and artificial intelligence could potentially cause life expectancy increases to accelerate once again.
More speculatively, some theorists have suggested a future where aging itself can be halted or reversed. The lifespan of certain species of worms has already been extended more than fivefold using genetic techniques in the laboratory. This idea may seem far-fetched, but the pace of change could rapidly increase in our lifetimes. The impact could go far beyond the impact from people giving up smoking in the 20th century. Whether the technology is developed, on what timeline, and how quickly it becomes available to the masses, will be critical to the impact on pensions.
On the more pessimistic side, the impact of climate change, resistance to antibiotics, opioids, obesity, and rising inequality could be dramatic in the other direction. The important point is that the future will not be a repeat of the past. Things could change in ways we have never experienced and most of us cannot even imagine. Even a return of life expectancy improvements to the levels seen in the early 2000s could reduce funding levels by at least 10% and put many fully funded defined benefit plans back into deficit. It should be noted that a 10% impact on funding from longevity expectation changes is hardly unprecedented – many plans have seen improvements to that extent in the last decade.
Those relying on private savings or employersponsored capital accumulation plans for retirement income face even greater uncertainty. Although macro trends will be an influence, variation in life expectancy at an individual level is almost impossible to predict. There is evidence that Canadians underestimate their exposure to longevity risk. Refer to our recent News & Views article on Canadians’ retirement risk perceptions and strategies, which noted that most respondents to a recent survey underestimated their personal life expectancies, undervalued life annuities, and did not understand the cumulative impact of inflation.
Employees need to plan for their retirement needs and consider the risk of outliving their funds. Employers can support their employees through access to guidance, education, advice and trusted capital decumulation strategies, including annuitization.
Mortality improvements have been slowing in the national population, but why and what this means for employer-sponsored pension plans, particularly over the long run, is unknown. While there is no short-term pressure to change funding assumptions, each defined benefit pension plan sponsor should be considering its sensitivity to longevity risk and its ability to bear it, together with the potential implications for itself and its members. The long-term future is highly uncertain, and longevity risk could still be material. Plan data analysis, future scenario modelling, and exploring risk mitigation or transfer strategies could be well advised at the current time. It is better to do this analysis now, from the current position of funding strength for many plans, than later. There is no guarantee that the better news for pension plans over recent years in investment returns, interest rates, and possibly mortality experience will continue.
For individuals financing their own retirement or relying on employer-sponsored capital accumulation plans, the risks are even greater. Employers should consider what support they can provide in educating employees about longevity risk and possible decumulation strategies.