Pension Discount Rate 101 - What's going on & What do you think?
Nov. 3, 2017
Earlier this week, the Department of Labor & Industries filed the formal paperwork to once again consider a reduction in the discount rate for pension reserves in the Washington workers' compensation system.
Pension Discount What?
Self-insurers have been hearing about the Pension Discount Rate from us for several years because self-insured employers who are funding pensions are uniquely sensitive to changes in the discount rate the Department chooses.
Since this is at its heart a financial and actuarial, rather than claims management, issue, workers' comp eyes tend to glaze over when the Pension Discount Rate comes up. It's not the sexiest topic in the industry.
But heading into 2018, talk of the Pension Discount Rate could heat up. This blog is an effort to explain why, and what's going on, and ask your opinion on the matter.
When Labor & Industries calculates the pre-funded reserve on a new total permanent disability pension, for either the State Fund or a self-insured employer, it calculates the future value of the annuity taking into account an assumed rate of return on the invested funds.
Because, according to the time value of money, the future value of invested funds is higher than the present value of those funds, the pension reserve is calculated assuming the value of the annuity will grow over its lifetime at a given rate, and so less money is required in the pension reserve up front. The liability is smaller, in other words, than it would be if a lesser, or no, rate of return was assumed of the funds that are reserved to cover it.
That assumed given rate of return each year on the invested funds covering the pension reserves is the Pension Discount Rate.
The move from 6.5 percent to 4.5 percent
Historically, the Pension Discount Rate has been at 6.5 percent, for decades. And for decades, outside auditors, inside actuaries, and others at L&I have raised the concern that their pension fund investments are yielding nothing like 6.5 percent.
In other words, the assumed rate of return is too high. Stated differently, the pension liabilities stated on the books are too small.
Because the money *has* to be there to fund the entitlement for the long tail of the pension, the assets in the pension reserve are invested primarily in what the finance community considers to be risk free instruments like 20-year treasury notes backed by the federal government.
And the yield on these notes, particularly new ones that have to be purchased to replace matured ones, is nowhere near six or more percent. It's under three percent.
The issue was brought to a head in 2012 by the Department, consulting with an ad hoc group of business and labor leaders, who compared the existing 6.5 percent discount rate to the roughly 2.5 percent "risk free" discount rate sought by L&I actuaries, and split the middle, setting a target discount rate of 4.5 percent to hit through gradual reductions over a ten year period through 2022.
What happens when the Pension Discount Rate drops?
Why so gradual? If it's a matter of accurately stating the size of the system's pension liabilities, and ensuring they are adequately funded, why not make the drop all at once? Here's the thing:
The lower the assumed future investment yield, the higher the present day liability. The higher the present day liability, the more funding is necessary today to cover it.
For the State Fund, that means the ratio of assets over liabilities that L&I calls its contingency reserve goes down when the discount rate goes down, because the size of the liabilities goes up.
For every one-tenth of a percent drop in the discount rate, there is a $30-50 million dollar drop in the size of the Accident Fund contingency reserve. That is an issue for State Fund insured employers, because the contingency reserve helps offset premium rate increases and stabilize premiums. It can also create political problems for the Department, because a significant restatement of liabilities at a much lower assumed investment yield could, in some years, show the pension reserve as insolvent and inadequately funded.
For self-insured employers, every one-tenth of a percent drop in the discount rate hits in two ways -- it increases the present size of the pension liabilities in the Second Injury Fund, and it increases the present size of individual, cash funded pension liabilities.
Both impacts mean more money up front, through both the Second Injury Fund assessment for both the socialized and experience-rated pensions in that fund, and in an assessment to "true up" the pre-funding of the cash funded pension annuity.
So the one-tenth of a percent drop in discount rate adds about $6-8 million to the liabilities in the Second Injury Fund, and about $1 million to the cash funded pensions. The vast majority of self-insured pensions, by the way, some 80-85 percent a year over the last few years, is funded by the Second Injury Fund.
Accordingly, to minimize volatility in the contingency reserve, and to mitigate against high, unexpected assessments to self-insurers, the Department has gradually stepped down the rate by one-tenth of a percent over the last few years. It is currently at 6.2 percent.
But then 2017 happened
Then 2017 happened. Given its mix of investments in other funds, and given the incredible returns the market has provided this year, the Department's combined accident and medical contingency reserve has skyrocketed by $1.6 billion, to pre-recession highs.
That huge increase in the contingency reserve provides enough room for the Department to drop right away to a 4.5 percent discount rate without taking the Accident Fund underwater.
Indeed, while dropping the rate to 4.5 percent would cause a $735 million increase in pension liabilities, the Accident Fund contingency reserve is currently $1.28 billion, and would end up at $545 million in reserve.
But what about self-insureds?
If the Department did take that drop -- and it desperately wants to act now, while the iron is hot -- it would increase the size of Second Injury Fund pension liabilities by $111 million, and cash funded self-insured pensions by $17 million.
In other words, it would entirely destabilize next year the already high Second Injury Fund assessment, and it would present a new $17 million bill in the aggregate to all self-insureds who are cash funding a pension.
Pay me now or pay me later?
Or is it? There is a pay-me-now or pay-me-later aspect of all this that bears mention. For cash funded pensions, each year they go through an actuarial review called "experting." The experting process recalculates the annuity on each pension taking into account mortality, disability, marital status, and investment income.
All things equal, in years where the investment income does meet or exceed the discount rate, the self-insurer actually gets a rebate because, for that year, it has overfunded the pension.
And similarly, in the Second Injury Fund, no additional load on the assessment would be added; the assessment rather would reflect the cash necessary to pay the forecasted benefits in the coming year.
However, in years the Department's pension investments do not reach the discount rate, the self-insurer gets a bill for the difference on cash funded pensions, and sees a spike in the Second Injury Fund assessment. For example, when the 2017 Second Injury Fund assessment almost doubled, that was in part because $30 million of the $70 million assessment was due to (a) the drop in discount rate from 6.3 to 6.2 ($6 million) and a shortfall in actual versus anticipated yield ($24 million).
So, assuming the actuaries are right, that we are now in an era where we cannot reasonably expect to see annual returns on risk-free pension investments at anywhere above 4.5 percent, an argument exists that the self-insurer is either going to pay now or pay later. Pay now, up front, the new and estimated true size of the liability, or pay for it slowly, through an annual load on the Second Injury Fund or in individualized assessments on each cash funded pension.
You see, there's some theoretical appeal to ripping the band-aid off all at once, and then actually being in a position to receive rebates, rather than assessments, in years where investments beat a much more conservative discount rate.
But that's a very large and sticky band-aid, and there is a big sore underneath it, considering the comparatively astronomical frequency of pension awards in the Washington system, given its mix of incentives, return-to-work challenges, and restrictions on claim settlement.
So for that reason, WSIA has generally taken the position with respect to the Department that the very slow, incremental drops in discount rate are necessary to avoid utter destabilization of the Second Injury Fund and cash funded pensions.
Although there isn't necessarily disagreement that a discount rate north of 4.5 percent is probably too high, we've instead taken an institutional position somewhat reminiscent of the famous prayer of the young St. Augustine: "Lord, make me virtuous... but not yet."
But as noted above, now in 2017, where the room exists, the Department very much wants to move aggressively on the discount rate, maybe all the way to 4.5 percent, while remaining sensitive to the up-front impacts on the self-insured community.
A separate self-insurance discount rate?
So one concept -- details TBD -- has emerged. Right now, because it just is the way it is by law, self-insured and state fund pensions are comingled in a pension reserve account, and computed on the same basis. But there's nothing in the laws of the universe that dictate that be so. There is already precedent for holding some self-insured pensions in a distinct account (e.g., the Second Injury Fund). And there is evidence in other states that self-insured workers' comp liabilities are discounted at higher rates than public/private insured liabilities.
What if the Legislature were to allow separate accounting of state fund pension liabilities and self-insured pension liabilities, under a separate discount rate for each?
Such a move, if authorized, could allow the Department to book State Fund liabilities however it chooses, while leaving more or less the status quo, pay-as-you-go, system in place for self-insurers, allowing the self-insured community the opportunity to provide annual input on where, and how quickly, the anticipated rate of return on its pensions should go.
We're certainly open and curious about the idea, and wonder if it is implementable and what safeguards, if any, are needed to protect the self-insured community and its workers from any unforeseen ill effects of separate discounting and discounting at higher than normal, truing up the difference on a year-to-year basis, as occurs now.
What do you think?