GASB's Robert Attmore Discusses New Pension Rules

Courtesy of Pew Center on the States
Robert Attmore

The Governmental Accounting Standards Board is poised to approve new regulations that will have a major impact on the way states calculate their pension liabilities. The changes are meant to make it easier to compare the health of pension funds from one state to the next. They also would have the side effect of making states' long-term finances appear in worse shape than their current balance sheets indicate.

One set of changes relates to the accounting methods states use to calculate their pension obligations. Under GASB's current reporting standards, governments can choose from six different accounting methods. The new rules would have states all use the same method. In addition, the orientation of these rules would change significantly. The current system takes a "funding-based" approach that focuses on how much states pay into the pension fund each year. The new system would take an "accounting-based" approach that employs a longer-term view to ensure that the overall costs of providing benefits are accounted for.

Other proposed changes relate to how pension funds calculate investment returns. The new rules would flatten the effects of boom-and-bust cycles by smoothing investment gains and losses over a standard five-year period. In addition, governments would no longer be able to use historical rates of return to determine their long-term liabilities if they haven't set aside enough money to pay retirees. Instead, they'll have to use a combination of a historical rate of return and a lower rate pegged to the municipal bond market-which will make their long-term liabilities appear larger.

Perhaps the most important change is that all of this information would have to appear in the body of states' balance sheets. So policy makers, lawmakers, investors and reporters — or anyone who simply wants to know about a state's fiscal health — would get a clearer view. If GASB approves the new standards, they will be released for public comment in early July and are likely to be implemented within a year or two.

GASB was established in 1984 by a group of national, state and local government organizations and the Financial Accounting Foundation as a way to establish uniform accounting and financial reporting standards tailored specifically to state and local governments, now known as Generally Accepted Accounting Principles (GAAP). While it is an independent organization that lacks any ability to enforce the standards it sets, GASB's requirements are taken as gospel by auditors, and bond rating agencies pay attention to whether a government conforms to GAAP when issuing opinions about the soundness of potential investments. Some states also have laws requiring them to comply with GAAP.

GASB Chairman Robert Attmore gave a preview of the proposed changes last week at a Pew Center on the States conference on public sector retirement costs. Stateline caught up with Attmore to find out more about how the new rules will impact states. Below is an edited transcript of the interview.

What is the history behind GASB's current standards for pensions and other post-employment benefits?

The pension standards were adopted back in the mid-1990s. Before then, there were no pension-specific reporting requirements for governments. Those statements were based on a funding approach for pension obligations. I don't in any way want to be critical of those decisions even though we're going to suggest improvements to them shortly. It was a major step forward. It provided lots of useful information about the size of the obligations that governments were incurring — the deferred compensation that doesn't get paid until people retire.

The OPEB standards were issued in 2004 and implemented on a staged basis over a three-year period so that now all governments are reporting their OPEB obligations. OPEB is "other post-employment benefits." The biggest segment is retiree health benefits, which are in some cases extensive and expensive. They are generally not pre-funded the way. Pension obligations are. For pension obligations, governments are setting aside money in pension trusts to pay for those benefits in the future. Until very recently, very few governments set money aside to pay for OPEB obligations in the future. They were all on a pay-as-you-go basis.

Why is GASB revisiting the current pension standards?

We actually began this process as part of our strategic planning. We periodically revisit existing standards to see whether or not they are meeting the objectives that were set forth and whether they're providing the decision-useful information people are looking for.

We began that process in 2006 and our staff did research on how the existing pension standards were being implemented, what kind of information was being communicated, whether it was meeting the needs. In 2008, our staff presented the board with a research report that suggested that some significant improvements could be made. The board evaluated that report and concurred.

In 2009, we issued an Invitation to Comment, which is a very early-stage document where we present alternatives of different ways to approach pension accounting and reporting. We got a lot of public feedback on that, held some hearings, and a year later in 2010 we issued a Preliminary Views document on employer reporting for pensions. We got a lot of feedback in public hearings and comments and have now completed our re-deliberation based on all the feedback.

We are ready to issue two Exposure Drafts, one for reporting about the pension plans themselves, which are separately run, and one for employers or sponsors of pension plans — the governments themselves. Those two Exposure Drafts will propose significant changes to the ways that governments are currently doing their accounting and reporting.

What feedback have you heard about the approaches GASB is considering?

 Views are across the spectrum. On one end of the spectrum are people who think current pension accounting is fine: It's not broken, don't fix it. On the other extreme are people who think it's fundamentally flawed and requires complete revision. What we're going to propose is not throwing out the overall system but making some significant improvements and modifications to it. We'll wind up somewhere in between those extremes.

What are the most significant changes that you're making? And what is the impact likely to be?

Switching from a funding-based approach to an accounting-based approach is one big deal. We think it's important to get the accounting correct so that the costs are allocated appropriately to the periods when employees were providing services, not when the benefits are paid sometime in the future.

The discount rate is a hot topic. Some people think the only way to discount this liability is using a risk-free rate. Others think the current approach of using the expected rate of return is just fine and we should leave it there.

We've decided, again, somewhere in the middle. To the extent that there are assets set aside in a trust that will earn long-term investment returns, we think that is the appropriate discount factor. But if there aren't enough assets set aside to pay the obligations, then we think a risk-free municipal bond index or a borrowing rate is the more appropriate way to go. If you have some assets set aside but not enough to cover the benefits, we'll have them calculate a blended rate of return, using the rate of return to the extent that there are assets, and then using the municipal bond index rate for the rest.

Many expenses would get recognized immediately, things such as retroactive benefit increases where they change the benefits for people who have already retired or are inactive. If there are expenses that are going to be deferred and recognized or amortized over a long period of time, they come in a couple different types. If there are actuarial assumption changes and demographic changes and things like that, we're suggesting that they be recognized over a period of the average remaining service life of active employees. To the extent that those apply to people who are already retired, they would be recognized immediately. But for active employees, it would be recognized over their remaining service lives.

When investment returns differ from expected returns, those differences would be amortized over a five-year period. There would be a smoothing of investment gains and losses, but over a fixed period of five years, which is different than the current requirement. There is no specific period right now.

States suffered sticker shock when GASB came out with the OPEB standards in 2004. Is that going to happen again?

The numbers will be different. The appearance looking at a government balance sheet will be that the government is in a weaker financial position because that unfunded liability has not previously been on the balance sheet.

It has been disclosed on the notes but not on the balance sheet. Adding a large liability to the balance sheet will make the government's net position lower, and make them appear to be weaker. The economic reality is that nothing has changed; it's the presentation that has changed. We took information that was previously in the notes and put it on the face of the financial statement and therefore it appears to be a negative impact. But the rating agencies have been aware of that and have been factoring in those obligations as they do their credit rating anyway, so it's not going to change much from their perspective.

On the whole, will the new standards be simpler or more complex?

It will be easier for users of financial statements to understand what's going on. There will be greater transparency, greater consistency, and more comparability between different pension systems. From a preparer's perspective, it depends on what they are doing now whether it is easier or simpler. It clearly will be less complex because there will be fewer options to choose from. The more options, the more complex things get.