The Coming GAAP Gap for Pensions
How closely will plan trustees follow GASB rules when they bill employers?
The final round of deliberations is underway. Sometime next year, the Governmental Accounting Standards Board will be releasing major new standards for pension accounting and financial reporting. These will become the generally accepted accounting principles (GAAP) for governmental pensions and ultimately for other retirement benefits. Until the final rules are released, we won't know exactly how sweeping the GASB's changes will be, but it's pretty evident to most observers that public officials will feel a seismic shift:
- Unfunded pension fund liabilities will be reportable on the balance sheets of the employers, and pension expenses will be shown as costs regardless of the level actually contributed.
- New disclosure rules will require pension funds and employers to display the assumptions trustees approve for projecting investment returns by asset class.
- For certain plans with unfunded liabilities, the discount rate used for accounting purposes will be lower, with a blended rate that includes a municipal borrowing rate for some portion of the liabilities under certain circumstances. (A lower number will result in higher liabilities and pension expense, because the discount rate is an exponential divisor for converting future costs into today's dollars.)
- Amortization periods used for unfunded liabilities will be shorter on a going-forward basis, often using the current employees' average remaining service period (which is about 12-15 years in most plans instead of the 30 years previously set as the maximum allowed amortization period). This avoids kicking the cost can to the next generation.
- Future investment gains and losses will be recognized over five years if GASB adopts its current proposal.
Based on some of the professional dialogue and industry chatter, what we don't know at this point is:
- Which pension plans will be compelled to use the lower "blended" discount rate and exactly what formula will be used. There are still unanswered questions and concerns about the technical methodology GASB proposed.
- How exactly the GASB will treat existing unfunded liabilities in its new scheme. The rules for future experience-based changes are clear, but the transition provisions in the exposure draft were pretty vague. With $700 billion or more of outstanding unfunded pension liabilities to pay off and almost $2 trillion of OPEB deficits, this transition issue is a big deal.
Even more uncertain is how public employers will use this new information. Already there is considerable discussion of a "divorce" between accounting and funding policies. Whereas the "annual required contribution" (ARC) under earlier accounting standards has traditionally been used for calculating employer contributions, GASB has announced its intent that funding policies need not follow accounting rules hereafter. This has led to considerable consternation and speculation about whether pension trustees will go their own way in making funding (contribution) policies. Here's what to watch for:
Discount rate. The discount rate is how we convert future liabilities to today's dollars to reflect the time value of money. GASB's exposure draft pronounced that the board favors continued use of the expected rate of return on investments where assets exist, but has leaned toward using a municipal bond interest rate where liabilities exceed resources. The resulting blended discount rate would be lower for some employers, and there is good reason to believe that many will be reluctant to apply the blended rate to calculate required employer contributions. For many public employers, the result would be "sticker shock" with budgetary impacts beyond the fiscal capacity of many governments to absorb without resorting to layoffs, pay freezes and other cutbacks.
On the other hand, the GASB's proposed disclosure requirements will compel pension trustees to publish their expected rate of return on a publicly reported asset allocation and projections of investment returns by asset class (stocks, bonds, real estate, international, private equity, etc.). That will make it more difficult for trustees to fudge the numbers to make the contributions come out where politicians desire, as I discussed in last month's column. It strikes me as implausible that pension trustees will use two separate expected rates of return — one for accounting and one for funding. The key issue here will be the blended discount rate for those with substantial unfunded liabilities, and we won't know until the rules are final just how many governments will actually be so affected. The "depletion" model suggested by GASB in its exposure draft would leave a lot of wiggle room to circumvent the blended rate (and opportunities for abuse, as my public comments to the board suggested). If GASB's blended rate produces a pension cost that makes the employer gag, I doubt it will be used for funding policy, and the trustees will be on their own to figure out what's fair. If we're not careful, the result could be what I'd call "anarchy" in pension funding policy.
As trustees begin to re-evaluate their liabilities and consider the accounting treatments of certain streams of costs, I also believe it will be necessary for them to re-evaluate the way they think about investment asset allocations and the resultant projections of income. Even if public pension funds have a perpetual life, their liability streams do not. Today's retirees will not live forever, so it seems unreasonable that tomorrow's taxpayers should bear the financial downside of investing a portfolio today in risky assets to fund those liabilities in the same way a plan invests to meet future obligations for today's rookie employees whose life expectancy is 50 or 60 years. These kinds of segmented asset-liability analytics should be applied more rigorously in the new world of GASB pension disclosures, to form a more appropriately composed asset allocation plan. If the conceptual goal of a pension amortization schedule is to assure intergenerational equity by pre-funding the liabilities before workers retire, then it seems logical that the asset allocation modeling for that group should look more like a 529 college savings funds' declining-risk glide path than a collegiate endowment fund or Alaska's permanent fund (which both have a perpetual investment horizon and no liabilities to match). Otherwise, trustees push the consequences of investment risk to the next generation in violation of the very principle of intergenerational equity that we're seeking to promote.
Amortization periods. The area most likely to see deviation of funding policies from financial reporting practices will be the amortization of today's unfunded liabilities. At issue here is the concept of intergenerational equity. Presently a large number of public pension plans and the vast majority of OPEB retiree medical benefits plans use amortization schedules of 25 to 30 years as permitted under current GASB standards. That practice virtually guarantees that the succeeding generation of taxpayers will bear the financial and opportunity costs of today's public servants' retirement benefits, which is dead wrong in principle. If GASB tightens up the amortizations of today's outstanding liabilities immediately to the average remaining service period of incumbent workers, that would have the impact of refinancing a house from a 30-year mortgage to a 15 year mortgage — required payments would spike up significantly.
We don't know yet where GASB plans to take the accounting profession in the transition period. As a compromise for the purposes of standardizing the transition, I personally suggested that they consider a maximum 20-year closed amortization period that would drop to 19 years the second year, 18 in the third and so forth, until the remaining amortization matches the average remaining service period. That would enable employers to migrate incrementally to the new standard during this decade as baby-boomer retirements surge. But it would avoid impossibly high hurdles in the initial years as the weak economy stifles state and local government revenues. For many employers, this would likely result in a hefty initial increase in pension costs — which would still be painful but probably manageable for trustees and employers seeking to align their contribution rates with their accounting policies. It's a simple but workable solution to a complex transition problem.
Once GASB finalizes its standard on this issue, I'm hoping that the public finance professional associations can quickly rise to the occasion and provide some guidance to fill in any gaps that GAAP leaves behind. That will be especially important if the new accounting standards produce contribution calculations that are so volatile or unreachably high for most employers that most employers simply can't get there from here. Anarchy in state and municipal retirement funding policies will further erode investor and public confidence in the entire system and financial analysis would be impaired by GASB's actions, not improved.
Trustees in the hot seat. One thing is for sure: pension trustees and their paid professionals will find themselves explaining and defending their plans' funding policies if they use assumptions that differ markedly form GASB's final accounting standards. Those who adopt reasonable strategies to make the transition to a more defensible contribution policy should fare well enough. But those who blithely ignore the concepts of intergenerational equity embodied in GASB's forthcoming standards will do so at their own peril. Current plan participants, and certainly the retirees, would have a strong case that trustees who short-sheet the contributions reportable under accounting standards (for the sake of budgetary convenience) will have failed to discharge their fiduciary responsibilities. And younger citizens as a class of prospective taxpayers may have a cause to take to the courts or the ballot box to assert their rights if the trustees impose increasing cost burdens on them by virtue of lazy amortization schedules, over-aggressive asset allocations and unrealistic investment projections. These could become key issues in fiduciary law and pension politics in the near future.
We're still at least six months away from GASB's pronouncement, so I don't want to jump the gun. But as readers can see, the challenges facing plan trustees and public managers will likely be more manageable if everybody is prepared for those conversations and the professional associations are timely with much-needed guidance. Otherwise, the funding-policy anarchy scenario becomes more likely. That would be the worst of all worlds.