The Deception at the Heart of Colorado’s Public Pension Crisis
The proposed rescue of Colorado PERA is just one part of the much larger crisis facing most public sector defined benefit pension plans in the United States.
At the heart of this crisis lies what might charitably be called a financial economics muddle, or, less charitably, a deliberate deception.
The key question is the rate at which a pension plan’s liabilities to pay future benefits to its beneficiaries should be discounted to their present value. This can then be compared to the value of the plan’s assets, to determine the percent of future liabilities which have been funded. For example, the largest sub-plan in PERA, the Schools Division, was, per the latest actuarial report, only 56% funded.
Note that in terms of financial economics, this discount rate issue is a totally separate from the rate of return that the pension plan expects to earn on its investments in the future.
For example, when a state or company issues fixed rate bonds, it offers to pay investors a stream of specific interest and principal payments over a defined period of time. These promised future payments are liabilities of the state or company. To decide what they should pay for these bonds, investors assess the likelihood that the issuer will default in the future, and, if it does, how much of their investment will be lost. Put differently, investors assess the credit risk of the issuer, and decide how much they need to be paid to bear it.
Bond ratings represent credit risk assessments by companies like Standard and Poor’s, Moody’s and Fitch. The lower the bond rating, the higher the estimated credit risk.
Investors must also estimate the future rate of inflation, so that the real value of the payments they expect to receive from the issuer is not eroded by price level changes.
Having made these two assessments, investors decide on the rate of return they require to buy the bonds and bear these risks. They then use this rate to discount the promised future bond payments to their present value, which is the maximum price they will offer to the issuer to buy its bonds. The higher the required rate of return, the lower the present value of the future payments, and thus the lower the price an investor will offer.
In the case of PERA, pension beneficiaries are the economic equivalent of bondholders, in that they both hold the risky liabilities of an issuer — in this case, PERA’s promise to make future pension payments. However, beneficiaries don’t have the option of deciding for themselves the appropriate rate at which to discount these liabilities to their present value. Instead, PERA sets it for them. Today that rate is 7.25%.
And that has enormous implications.
Consider how some other defined benefit pension plans calculate the present value of their liabilities.
Perhaps the most conservative pension plan in the world is run by the Bank of England. It discounts its liabilities using the yield on long term bonds issued by the government of the United Kingdom, whose return is regularly adjusted to account for changes in the rate of inflation (technically, these are called index-linked gilts). The BoE also invests its pension plan assets in index-linked gilts whose maturity is matched to the maturity of its pension plan’s liabilities. This pension plan is extremely expensive to run, but it is also intellectually honest.
Now consider defined benefit pension plans run by US corporations. By law, the rate that they must use to discount their future liabilities is linked to the average yield on long-term corporate bonds. Note that use of this rate implies that pension payment obligations are as risky as payments on a company’s bonds. Note also that the rate at which future pension payments are discounted to their present value has nothing to do with the rate of return the pension plan expects to earn on its portfolio, which includes investments in multiple asset classes.
Today the long-term nominal (not inflation protected) yield on BBB rated corporate bonds is about 4.50%.
Note that this is higher than the yield on US government bonds of comparable maturity, because there is a possibility that the company will go out of business, and no longer be able to make all the payments promised to its pension plan beneficiaries
If this happens, the US Pension Benefit Guarantee Corporation would take over the plan and impose its maximum cap on the annual payment it will make to any beneficiary. Especially for employees expecting high pensions, transfer of their plan to PGBC usually results in a significant reduction in their benefits.
And now we get to the crux of the matter.
If, as corporate pension plans are required to by law, PERA believed that the riskiness of its promised pension payments was equal to the riskiness of bonds issued by the state of Colorado (i.e., if it adopted the approach used by private sector pension plans), then it should have used the yield on the state’s AA rated long term general obligation bonds (currently around 4.00%) to discount its liabilities to their present value.
To be sure, the Government Accounting Standards Board mandates that public sector pension funds use their expected investment return to discount forecast future benefit payment to their present value. While this helps to minimize the reported size of the pension liability, in financial economic terms it makes no sense, unless, as we shall see below, another assumption is true. But that said, there is nothing stopping PERA from reporting in its financial statement notes the size of its pension liability based on a 4.00% discount rate.
Or maybe there is.
Because 4.00% is considerably less than the 7.25% discount rate PERA uses, the lower discount rate would substantially increase the reported size of PERA’s liabilities, and thus dramatically reduce its funded ratio, which would make the scale of the current crisis much, much worse.
But PERA has not taken this approach. Instead, it has discounted its future liabilities using a rate of 7.25%. Unless GASB intended to disregard financial economics and mandated use of the higher discount rate for no other reason than to mislead taxpayers and other users of public pension fund accounting statements (which would be fraud), then use of this higher rate must imply that PERA believes there is a significant risk to its beneficiaries that their promised pension benefits will not be paid in full (e.g., due to poor investment results, revisions to actuarial assumptions, or other causes).
How much risk?
The current yield on long-term corporate bonds rated single B (well below investment grade) is about 6.25%. Thus PERA’s use of a 7.25% discount rate suggests that its pension promises are actually risker than future payments on single B rated bonds.
Deeply hidden at the heart of the PERA crisis is the ugly truth that somebody has been very badly deceived.
If PERA’s pension liabilities are really no riskier than the state of Colorado’s general obligation bonds, then it is the taxpayers who have been fooled, because PERA’s true liabilities are far larger than what has been acknowledged, and in order to meet them either future taxes will have to be far higher or severe cuts will have to be made in other areas of government spending.
On the other hand, if PERA’s liabilities really are riskier to hold than a portfolio of single B rated bonds, then it is the plan beneficiaries who have been fooled, as the probability they will receive their full pension benefits is far lower than most of them currently believe.
In actual fact, the structure of the bailout plan proposed by PERA — which includes both higher taxpayer contributions and cuts in future benefits — suggests that both Colorado taxpayers and plan beneficiaries have been played for fools in the past, and apparently will continue to be played for fools in the future.
Until that changes, our growing public pension crisis will only intensify.