This summer, the Canadian Institute of Actuaries (CIA) released new pension mortality tables. The tables revealed longer life expectancies than were previously provided by commonly used tables. Some of the industry reaction has focused on the longevity risk faced by DB pension plans.
Longer life expectancies for DB pension plans are, in my view, primarily an issue of cost rather than risk. If the issue were primarily about risk, it could be dealt with by purchasing annuities or some other hedging vehicle. But the issue isn’t primarily about risk and cannot be hedged away.
Much of what is described as longevity risk isn’t risk at all. Risk has to do with the possibility of future events that are different than expected. Yes, we are living longer than our ancestors did, but this has long been expected. Pension actuaries have been building projected improvements into their mortality tables for years. All that the actuaries got wrong was the extent of the mortality improvement.
Longevity risk has to do only with mortality improvements that are different than expected. The new CIA tables are said to increase liabilities for typical DB pension plans by 5% to 10%. This is compared to projections of the prior 1994 table. And, frankly, a one-time reset of 5% to 10% after almost two decades doesn’t suggest that longevity risk is especially large. (Contrast this with the constant volatility of solvency results over the same period!)
There is good reason to expect longevity risk to continue being comparatively small. The maximum recorded lifespans have changed only slowly, from age 103 in 1798 to age 122 today. Most of the advances in life expectancy have come from reductions in the risk of “premature” death rather than increases in maximum lifespans. This suggests to me that future life expectancy increases (both expected and unexpected) will very likely come up against the law of diminishing returns.
Even if longevity risk is comparatively small, is there any harm to hedging it? Maybe. First, you should expect any financial institution to include a spread in the price of its hedging products. This spread is an extra cost that has to be evaluated against the size of the risk you are hedging. Second, and most important, hedging longevity risk can be a serious distraction from the focus we should be putting on our pension plans’ much bigger issues, such as resolving our funding challenges, using our risk budgets better and making sure the cost of our plan designs are sustainable.
Even if you decide to hedge your longevity risk, you will run into two problems. One, you won’t be able to escape the cost of expected longevity increases, as this will be built into the cost of the hedge you buy. And, two, you will only be able to hedge part of the risk of unexpected longevity increases, as you can’t hedge the risk imbedded in your current service cost.
The projection scales of pension mortality tables cause the cost of DB pension plans to increase slowly over time (all other things being equal). This has been going on for years under the old table and will continue under the new table.
On top of this, DB pension plans will have to cope with the one-time cost of changing to the new table. For most DB pension plans, this will impact funding requirements. It might push some to the tipping point where design changes are required. Hedging longevity risk won’t change this. We should not allow the hedge sellers to confuse us on this point and let longevity risk divert our attention from the more significant pension issues.
Calvin Jordan is CEO of the Nova Scotia Health Employees’ Pension Plan. The views expressed are those of the author and not necessarily those of Benefits Canada.