Copyright_Sebnem Ragiboglu_123RF

One of the many consequences of the coronavirus pandemic was an unprecedented interest in the life expectancy of Canadians.

While the increased focus on life expectancy seemed to wane once the worst of the pandemic had passed, a recent report by the Canadian Institute of Actuaries on mortality improvements has brought the topic of life expectancy — and its impact on defined benefit pension plan sponsors — back into the spotlight.

Read: More work to be done by pension plan sponsors on factoring climate into mortality, morbidity assumptions: CIA

Pension plan mortality assumption

One of the most important assumptions that actuaries make when valuing pension plan liabilities and costs is the mortality assumption. The mortality assumption is important because it affects how long pension plan members are assumed to live and, as a result, for how long they are expected to collect their pensions once they retire.

The mortality assumption is made up of the base mortality table, which contains the assumed annual rates of mortality of the pension plan members at a particular point in time; and the mortality improvement scale, which contains the assumed decreases in mortality rates over time, associated with increases in life expectancy.

The CIA’s report focused on the mortality improvement scale and doesn’t address the base mortality table assumption. It focused on pre-pandemic data (up to 2019) and didn’t include data from 2020 and 2021, as the pandemic’s impact would have led to distortions that don’t reflect long-term mortality trends.

The report emphasized that it’s extremely difficult to predict long-term mortality improvement rates, as they’re subject to a lot of uncertainty.

Read: Could pandemic-related mortality rates impact DB pension plans?

It also recommended a new best estimate mortality improvement scale with rates that vary by sex at birth, age and calendar year. The short-term improvement rates in the new improvement scale reflect recent mortality improvement experience.

Mortality improvement rates in the new scale trend over a 30-year period to an assumed long-term mortality improvement rate of 1.3 per cent per year. This long-term improvement rate was selected from a reasonable range of between one per cent to 1.9 per cent, which highlights the uncertainty regarding future changes to life expectancies. The long-term improvement rate is higher than the long-term rates currently being used by Canadian pension actuaries but is in line with the rates used in a number of other countries.

Adoption of the new mortality improvement scale would increase the life expectancy of pension plan members compared to current assumptions being used by pension actuaries and would therefore increase pension liabilities and costs. The magnitude of the increase in liabilities will vary depending on the characteristics and circumstances of each plan, but the increase is expected to be in the range of one per cent to five per cent for many plans.

Next steps

This month, the CIA issued guidance indicating that both the new mortality improvement scale and the mortality improvement assumptions currently used by actuaries are appropriate to value pension plans. However, it will still need to be determined whether the prescribed mortality assumption used to calculate commuted values payable to former DB plan members will be revised to adopt the new improvement scale, which would increase commuted values. Also, pension regulators and auditors have yet to indicate their views on the new improvement scale.

Read: Calculation for commuted pension values confusing to some plan members: actuary

While it’s still too early to know the prevalence and speed of adoption of the new mortality improvement scale, it’s important for pension plan sponsors to measure the impact of adopting the new improvement scale on pension liabilities and costs.

If they haven’t done so recently, plan sponsors can review and, if appropriate, update the base mortality table assumption currently being used to value their pension plan. In addition to more accurate pension liabilities, updating the plan’s base mortality table assumption will reduce the impact of the CIA publishing new base mortality tables.

And, given the uncertainty regarding future changes to life expectancies, it’s important for plan sponsors to stress test the effects of different mortality outcomes and assess the sponsor’s pension risk management strategy in relation to this unknown.

Read: How commuted-value calculation changes will impact lump-sum DB pension payments