Long before the global pandemic, jurisdictions across Canada were embarking on pension solvency reform, with some moving from strict 100 per cent solvency funding requirements to going-concern plus models, which incorporate a funding cushion.
Low interest rates, which make solvency funding very expensive, have been a key driver pushing the going-concern plus model, says Dean Newell, vice-president of Actuarial Solutions Inc., noting the pandemic has likely solidified the low interest rate trend. So what steps are governments taking in the various jurisdictions across Canada?
In 2016, Quebec came first, introducing the new model for a funding cushion, which it called a stabilization provision. The cushion considers two measures: the percentage of assets allocated to variable income securities and the ratio between the duration of plan assets and liabilities.
“From my perspective, I think Quebec has achieved probably the most appropriate or the best balance because their approach both reflects the mix of fixed income versus more risky assets in terms of determining the level of the [stabilization provision] and it also takes into account the extent to which the funded position of the pension plan is protected against changes in interest rates,” says Gavin Benjamin, senior director for retirement at Willis Towers Watson.
In determining how to calculate a funding cushion, it’s key to find a balance between simplicity and a more perfect reflection of the level of risk in the plan, adds Benjamin, noting other jurisdictions didn’t reflect interest rate risk because they likely thought it was adding in more complexity than was actually required. “From my perspective, it’s an important element of risk protection in terms of protection against interest rate changes. [The] benefits of reflecting the interest rate risk overweighs the additional complexities associated with reflecting that element.”
In 2018, Ontario moved to an enhanced going-concern model, taking a different approach than Quebec. The province’s provision for adverse deviation calculation considers if the fund is open or closed, as well as the percentage of non-fixed income assets. Its rules also throw a discount rate into the mix. In particular, if a plan’s discount rate exceeds the benchmark discount rate, an additional amount will be added to the funding cushion requirement.
Overall, Newell says he’s a fan of Ontario’s approach to the PfAD because, although not perfect, it’s simple and practical. However, some argue the approach is overly simplistic because it doesn’t account for alternative asset classes that may be fixed income-like without actually qualifying as fixed income.
“But I think that’s a problem for the very large pension plans that have the more alternative asset mixes available,” Newell says. “I think if you’re looking at the smaller single employer pension plans that we see a lot of, under $100 million, many of them are not necessarily investing heavily into the alternative investments.”
One issue with the Ontario approach is it distinguishes between closed and open plans, Benjamin notes. “From my perspective, whether a DB provision is closed or open is not a direct indicator of the risk embedded in the plan. And so, if Ontario were to review the approach for calculating the PfAD, in my view that’s something that should be looked at.”
He also highlights issues with the rules surrounding the benchmark discount rate. “The intention of this adjustment is to provide protection against a situation in which the actuary might be too aggressive when selecting the discount rate assumptions. But from my perspective, the fundamental problem with the benchmark discount rate is that it’s determined based on the formula that’s contained in the Ontario Pension Benefits Act Regulations and that formula is based on Government of Canada bond yields. Whenever you compare something as complex as a discount rate assumption to a formula that doesn’t allow for any element of judgment, there will be situations in which the benchmark discount rate comparison is not appropriate.”
The current environment is a good example of the problem because the Government of Canada bond yields are very low and adjustments to the PfAD are occurring in cases where the discount rate assumption selected by an actuary is appropriate and not overly aggressive, Benjamin notes.
When British Columbia implemented a funding cushion for DB plans registered in the province starting on Dec. 31, 2019, it went a fundamentally different route than Quebec and Ontario. B.C.’s PfAD has a five per cent minimum and is generally equal to the monthly long-term benchmark Government of Canada bond yield as of a certain valuation date, multiplied by five. Further, pension plans with a minimum 70 per cent fixed income allocation can see their funded cushions reduced.
Greg Heise, a partner at George & Bell Consulting Inc., was on the actuarial working group for the province when determining which course to take for the PfAD. He notes the PfAd was only intended to address contribution volatility. “When you put it in that perspective, if the PfAD in B.C. is there to only address contribution volatility, it’s viewed quite differently and we came at it from a very different perspective than perhaps some of the other provinces did.”
The actuarial working group debated various topics at length, he says. “We looked at obviously pretty rudimentary issues: stocks versus bonds and those kinds of strategies, but also leveraging, illiquids, alternatives, all of those kinds of asset classes. And essentially, once we agreed upon the fact that we did not want to restrict or funnel plan sponsors down the road to a particular investment philosophy, it became pretty clear what to do, which is essentially come up with more of an agnostic PfAD — agnostic when it comes to investment philosophy.”
In planning when developing the approach, the PfAD was closer to 10 per cent, notes Heise, and because bond yields are so low today, PfADs are closer to five, which was the minimum set. While the group did discuss interest rates dropping and potentially entering negative territory, it didn’t expect them to go as low as quickly as they did. “Other than the absolute magnitude of the PfAD, we haven’t had any issues, at least none that the working group has dealt with.”
Further, he says many plans that are on a path to winding up may have already moved to 100 per cent long bonds or matching bond strategies and now may need to potentially contribute more than they did under the previous regime that required 100 per cent solvency funding and going concern. “While now, you have to fund at least to 105 per cent going concern for a plan like that and, if you’ve got a matching bond strategy, that interest rate with the five per cent PfAD is likely requiring you to fund at a higher level . . . than 85 per cent solvency.”
In practice, this means some pension plans may be required to overfund their plans, says Heise. And the concept of a solvency reserve account, which allows a plan to access money if it’s overfunded at a certain point, doesn’t exist for PfAD contributions. “There will be some plans that have some challenges and might be grappling with some trapped capital.”
Other jurisdictions are starting to move toward a going-concern plus model. For instance, Nova Scotia implemented a PfAD in April 2020, introducing an approach that’s most similar to Ontario, except for the distinction between closed and open plans. In particular, the N.S. PfAD is calculated by combining a fixed five per cent with a percentage based on the plan’s non-fixed income target asset allocation. It also distinguishes between funds with and without solvency funding requirements. In July 2020, New Brunswick proposed a similar approach, inviting feedback from the industry.
Missing across the board
Across jurisdictions, one item that hasn’t been addressed at all is the use of leverage in pension plan investments, says Neil Lamb, vice-president and director of relationship management at TD Asset Management, noting that, the lower interest rates go, the more plans will use leverage because it’s so much harder to meet funding targets with low bond yields. “I really think that legislators, ideally — but if not that, regulators — should talk about . . . how to interpret the PfAD rules in a world where you’re using leverage, because people can take on risk or they can hedge risk with leverage and I don’t think the two should be treated the same.”
In Benjamin’s view, the approach for calculating the PfAD will likely evolve over time as the industry applies the funding cushion and learns lessons. Further, as plans start using new investment products, it will likely lead to some refining of these approaches. “But on the other hand, it is fundamentally difficult to prescribe an approach that is going to appropriately capture the risk embedded in a certain investment class in all situations. I think things will evolve, but there will always be that element of balancing getting to a more perfect answer with simplicity.”
Instead of a prescriptive approach, Lamb suggests that legislators take a principles-based approach that considers how investments are lower or higher risk in a pension plan’s context. “I think that’s ideal and [there’s] some challenges with how that gets enforced and regulated, but if we can trust plan administrators to disclose correctly the risks that they think they’re taking then that is somewhere that we can eventually get to.”