Those who follow pension finance closely know that there's been a big discussion about whether to show public pension liabilities on the employers' financial statements, and if so, then how to value the liabilities. Traditionally, pension liabilities have been disclosed in the footnotes rather than on the employers' books. The unfunded liabilities are massive, on the order of magnitude of $1 trillion for pension funds and another $1½ to $2 trillion for retiree medical plans (so-called OPEB for "other post-employment benefits"). In California, for example, these unfunded liabilities are equal to the entire outstanding bonded debt of the state and its subdivisions. And the numbers could be even larger. Pension funds have traditionally calculated the present value of the liability for future pension payments using the expected investment return from their portfolio assets. A new school of financial economists has argued that this is too optimistic. Their focus is on the "market value of liabilities" (MVL). They claim that the discount rate should be the "risk-free" rate like a government bond, which would make the liabilities look much larger than they do today.
GASB is seriously considering a middle way that reflects a pragmatic and reasoned approach. Their tentative view as posted on their website is that they would use the expected investment return for assets in the pension portfolio for those assets, but only for those assets. Any liabilities that cannot be paid from the pension fund assets so invested would henceforth be valued using a much lower discount rate that reflects the employers' borrowing costs. The discount rate they mention on their website is a tax-free high-quality municipal bond index rate, which will be controversial as I shall explain below. That rate today would be something around 5 percent and not the 8 percent presently used on average for the portfolio rate. As a result, pension liabilities would be much higher under the new math. (Remember that the discount rate is raised exponentially in the denominator of a stream of future values, so a larger discount rate reduces the projected liability).
GASB's rationale is that the pension fund is the "primary obligor" for the pension liabilities, but the employer is the "secondary obligor" for those future liabilities for which the pension fund lacks sufficient assets. Thus, the discounting at the employer's borrowing rate makes sense for that (secondary) component of the liabilities to be shown on the employer's balance sheet. After all, the pension funds can do little to change that portion of the liabilities through actions of their own — they don't control the employer contributions. So the pension funds really don't have a primary voice in that part of the governmental accounting discussion; it's really an employers' issue. (In fact, GASB cannot require pension funds to use this new methodology in their actuarial practices; it primarily provides standards for how the costs are recorded by employers. However, some states require that contributions be made in accordance with GAAP, and most employers presently follow that practice.)
For the record, I believe GASB would be wiser to use a taxable municipal bond index rate for the discounting calculation. That would reflect current reality, since state and local governments are prohibited by federal law from ever issuing tax-exempt bonds to finance pension obligations anyway. Ever since 1986, all pension obligation bonds must be issued as taxable. An imputed taxable municipal bond rate today would be around 6 percent, which would still raise the pension liabilities, but far less painfully for the employers under this new accounting convention. Ironically, it would reduce the actuarial liabilities for OPEB plans that are presently unfunded, so the long-run budgetary impact for many employers would be neutralized.
Finally, here's the real zinger: GASB has decided to address the issue of intergenerational equity by correcting a long-standing problem in how long these liabilities can be paid off. The accounting technique is called amortization. Presently, GASB allows pension funds to amortize (pay down) their unfunded liabilities over a period as long as 30 years — even though most employees now on the payroll will retire long before then, and many retirees will be deceased! The corporate accounting board (FASB) requires such liabilities to be amortized over the average remaining service lives of the employees, and GASB is reportedly moving in that direction as well. Most public employers have a workforce with an average remaining service life of 12 to 15 years, reflecting career spans of 25 to 30 years on average plus high levels of seniority in aging Baby Boom workforces. So this rule alone could potentially cut the amortization period in half, which almost doubles the annual cost for this aspect of the liabilities.
Mathematically, it doesn't take a genius to figure out what all this means for public employers. Annual pension budget costs will increase significantly if these rules become effective (probably 2013 at the earliest, judging from the project timetable and past implementation practices). First, the unfunded actuarial accrued liabilities (UAAL) for pensions will increase by 30-33 percent for most employers by cutting the discount rate. Then, the amortization period for most employers will be reduced materially. The cumulative result will be an increase in annual costs to amortize the unfunded liabilities (which already doubled in the past recession), by 150-200 percent above current levels.
Imagine you bought a house five years ago with a 30-year fixed mortgage and were transferred involuntarily to keep your job; you had to move to a more expensive city where the same house now costs twice as much and were then told that the only available mortgage is for 15 years. Then imagine what would happen if the value of your new house fell in a slumping property market so that your home equity is eroded. That's essentially the situation facing today's state and local governments in their pension plans. They now realize they need a smaller house.
Note that normal costs for current service of current employees would not be significantly affected. We're mainly talking about the costs to pay for services previously rendered. But that often represents half of the pension bill, and in some plans with shrunken or older work forces and large retiree rolls, it's the lion's share of the costs already. So the impact will vary from employer to employer and state to state. But the direction is unmistakably uniform: Costs are going up.
Pension costs are heading higher anyway because of the investment losses the plans' portfolios sustained in 2008. Most employers will begin to feel those impacts in 2011 and 2012. Where "actuarial smoothing" has been used too liberally, the true budget impact has been deferred to 2013. The GASB changes would redouble or even triple those oncoming cost increases.
Those who believe that we've just been "kicking the can down the street" for years will rejoice that GASB is finally addressing some major deficiencies in the current accounting model. Those who find themselves cutting services, payrolls and salaries for another decade to pay these bills will probably be less cheerful. Those who get laid off or furloughed endlessly to pay for irrevocable pension bills incurred by elderly workers will be completely humorless. If anything brings to light the unsustainable nature of past pension promises, this will be it, regardless of which side you're on.
The MVL crowd of financial economists will still complain that their views were not adopted, but that becomes a largely academic point now. The financial impact of GASB's new approach has far broader significance in cleaning up the books, encouraging sound funding practices, and establishing genuine intergenerational equity than the MVL crowd ever dreamed. I'd rather see GASB take the approach it has chosen, than to adopt a theoretical approach that creates its own separate set of problems including potential overfunding and pension raids by public employees and retirees in the future.
GASB's due-process agenda will take many, many months to complete this project. Their preliminary views are just that — preliminary — and they may change course in the exposure draft of their final standards in 2011. But those who are interested in commenting on these ideas should get a copy of their official Preliminary Views document in June when it's released, and submit comments or attend an October public hearing.
Public managers who now see that the light at the other end of this tunnel is an oncoming train will be wise to begin preparing their elected officials and employee groups for an inevitable cost increase that will require further retrenchments — even when the economy begins to produce additional revenues. Smart managers will have their actuaries calculate the numbers this year using the tentative GASB formulas, to see how large the fiscal impact could be. Other than the double-dips of 1982 and 1937, this may be the first economic expansion in American history in which the payrolls of state and local governments nationwide are cut while the economy is growing.
