GASB's ineffective public pension reporting standards set to take effect
Public pension plans across the United States face a remarkable funding crisis. A study by economist Andrew Biggs and published by State Budget Solutions found that the average public pension plan was just 41 percent funded. That is, plans had assets to pay less than half of their liabilities owed. In an effort to increase the transparency of how these systems report their assets and liabilities, the Governmental Accounting Standards Board (GASB), which sets public sector financial reporting standards, issued updated guidelines for public pensions.
GASB Statements 67 and 68, first issued in June 2012 following a six-year research and comment process, include many changes. The new standards take effect in the upcoming fiscal year, yet they remain widely misunderstood, as these new standards fall short of bringing about true transparency or real change.
Revised standards for discounting pension liabilities and the elimination of Annual Required Contributions from the measurement of pension obligations will have the biggest impact. Plans will no longer smooth their reported assets over a three to five year period, but will instead immediately incorporate them. Plans will also all begin calculating liabilities under the Entry Age Normal Cost Method. Finally, all plans will be required to list their pension liabilities on the face of their financial statements. In summary, the GASB changes affect accounting reporting only and do not link accounting reporting of pension liabilities with pension funding.
GASB's revised standards are more a slightly modified continuation of a disaster than a sharp turn towards transparency and sound management. Despite attention in recent months, the near impossibility of accessing timely, comprehensive public sector financial information means that the changes will be unnoticeable for another several years. They continue to ignore basic economic truths, may encourage risky investment practices and are riddled with possible unintended consequences. They will also reduce comparability between plans.
Importantly, none of GASB's standards dictate plan funding itself. Governments remain free to do so as responsibly or, in many cases, as irresponsibly as they choose. GASB only sets the standards, adopted voluntarily, for how the public sector reports their financial figures. Getting these standards right remains vital, though, as the figures presented accordingly are the public's only glimpse into the inner workings of their government.
Confusion over the timing of GASB's standards taking effect highlights the incredible lack of up-to-date public pension financial information. GASB 67, for example, is effective "for financial statements for fiscal years beginning after June 15, 2013." In most states, that means that the standards will kick in for fiscal year 2014, which ends June 30, 2014. The first round of financial reports incorporating these new standards will not be published for several more months. As of April 2013, for example, Illinois had yet to publish its Comprehensive Annual Financial Report for the fiscal year that ended June 30, 2012.
The second set of revisions, GASB 68, is not made effective for an entire year after the first. It applies to fiscal years beginning after June 15, 2014. What all of this means is that despite the talk of pension systems facing immediate and drastic changes this June, none will show up in financial reports until as late as Spring 2015.
Discount Rate Disconnect
Certainly the most discussed change in standards is an adjustment to how plans discount their liabilities. Prior standards have allowed plans to discount liabilities based on the expected long-term yield of the plan's assets, typically around 8 percent. Critics have long argued that the assumed investment returns used by public plans are unrealistically high and out of touch with reality. From this perspective, investment assumptions of 7 to 8 percent have played into a vicious cycle in which plans radically understate their unfunded liabilities leading to lower than necessary employer and employee contributions.
The new method uses a blended discount rate. Now, plans will project forward the length of time that existing assets along with expected contributions and investment returns will be enough to pay yearly benefits. Liabilities in these years will be discounted in the same manner that they are today. Liabilities that exist outside the years in which they can be covered with existing assets and expected gains will be discounted at the rate of a high-grade municipal bond.
Economist Andrew Biggs described a hypothetical in which a pension system owes benefits of $1 million per year for the next 50 years, and holds $10 million in assets assumed to achieve an 8 percent annual return:
Under current pension accounting practices, the plan discounts all future benefit liabilities by the 8 percent interest rate projected for plan assets. In contrast, economists would discount these liabilities at a risk-adjusted interest rate, generally ranging between 4 percent and 6 percent.
Under the GASB's proposed method, the pension would calculate how long it expects its assets to last. In this case, the pension's $1 million in assets earning 8 percent annual returns could pay full benefits from 2010 through 2025. Benefits during this funded period would be discounted at an 8 percent interest rate. Benefits from 2026 through 2050, which are unfunded, would be discounted by using an interest rate derived from an indexed portfolio of high-quality municipal bonds. Total plan liabilities would equal the sum of these two calculations.
GASB's new blended discount rates might seem like a step towards acknowledging reality. Indeed, it is expected by many that the new system will lead to many states reporting far greater liabilities than they have in recent years. Researchers from Boston College ran the numbers for 120 public plans in the Public Plans Database and found that in fiscal year 2010, the aggregate funded ratio fell from 76 percent under the old standards to 57 percent under the new blended discount rate.
Instead, GASB flubbed a tremendous opportunity to make progress towards responsible pension funding. Many economists have challenged GASB's fundamental premise that liabilities should be discounted based on the performance of a plan's assets. They should instead be discounted according to the level of risk in the liabilities themselves, as public pension benefits are largely guaranteed and thus carry very little risk of nonpayment. See Andrew Biggs' assertion above that an economist would discount liabilities at a risk-adjusted interest rate. GASB's new standards continue this mistaken connection.
This change in the method for discounting liabilities also exacerbates GASB's "burning cash" problem. Economist Robert Novy-Marx has identified this situation under current GASB standards where perverse incentives create an unintended option for reducing plan liabilities: literally destroying money. This can happen because, as he writes,
The marginal value of a dollar's worth of an asset, as measured by the GASB, can be negative. By destroying a dollar's worth of T-bills or other cash equivalents, a manager decreases a plan's assets but increases the remaining assets' expected returns. These higher asset returns decrease the plan's liability, as recognized by the GASB, and this decrease can more than offset the loss of assets.
Novy-Marx describes hypothetical pension Plan A and B. Plans A and B are identical except that B holds an additional $10,000 in treasury securities. On the surface, B is clearly as a better funding position by $10,000. Yet, under current GASB standards, Plan A would be fully funded while Plan B would have to list a $3,000 unfunded liability. The new standards will make the problem even worse, bringing Plan B's listed unfunded liability to $7,000. "The proposed amendments to the GASB's standards, designed to ‘improve the decision-usefulness of information in employer financial reports' would thus make Plan B, which is self-evidently $10,000 better funded than Plan A, report that it is an additional $4,000 less well funded."
No More Annual Required Contributions
Annual Required Contributions currently provide observers with one of the surest indications of a government's willingness to fund its pension promises. These contributions represent the percent of payroll that a plan sponsor must pay each year to amortize their unfunded liabilities over a maximum thirty-year period. Every public plan sponsor makes them and they can be found in existing financial reports.
Under the new standards, Annual Required Contributions will no longer exist. Plans will instead choose between using an actuarially determined contribution or a statutory contribution. The ability to compare plans will be reduced because the guidelines for actuarially determined contributions will no longer be uniform across plans. Further, since plans that choose to use an actuarially determined contribution will have to report underlying assumptions while those that choose a statutory contribution will not, many will simply choose the latter.
According to Boston College researchers, "relying on statutory rates raises potential concerns - they may not be set to adequately reflect a plan's funding needs and their static nature makes it more difficult for a plan's funding strategy to respond to changing conditions." This can already be seen today. Virginia, for example, sets a "statutory required contribution" that is 10 percent less than its Annual Required Contribution. Using this trick, the state is able to show that it is meeting 100 percent of its required obligation.
The systematic underfunding of public pension plans is sure to continue as long as GASB's skewed standards remain in effect. The fact of the matter is the private sector already relies on Generally Accepted Accounting Principles (GAAP) that maximize transparency and accessibility. The next time GASB embarks on a six-year journey to revamp some of its standards, it need only look to GAAP for inspiration. Barring that drastic of a change, more piecemeal solutions include restoring the connection between plan funding and financial reporting previously held maintained by Annual Required Contributions and discounting liabilities at a single rate reflective of plan risks.