GAAP for pensions: Sanctioned fraud must go

OK, it's the busy season for professors. As we have done in the past, we're bringing back a previous column or two for an encore so that we can take a breather from the tyranny of the biweekly deadline while we collect our wits after our busy season.

We chose this column, which appeared in the May 19-June 1, 2003, issue, because of its timeliness. Recent headlines and cover stories have described how the private pension system is in distress - Ford, General Motors and United Airlines, for instance, are all in deep trouble.

In addition, we've been called on by a couple of financial analysts to help them spot the lurking risks hidden in the footnotes for pensions and other benefits. One company that we examined was showing a net pension asset of $5 billion on its balance sheet, but digging through the note produced the staggering result that its gross debt was approaching $100 billion, while its earmarked assets were coming up nearly $15 billion short.

There can be no doubt that the Frankenstein monster (SFAS 87) patched together by the Financial Accounting Standards Board of the 1980s to escape political heat is once again wreaking havoc in the village. It's time to pull the plug and do the right thing - replace the convoluted bad accounting with straightforward reporting of the debt and the assets, and the resulting income effects.

Enough is enough.

Straight frauds like Enron, Adelphia and WorldCom aren't the only examples in which adverse consequences follow bad accounting practices. In those cases, management set out to deceive by blatantly breaking the bounds of generally accepted accounting principles.

Ironically, a bigger fraud has been foisted on the capital markets by managers who haven't lied or even been aggressive, but have simply followed the GAAP path, step by step. We're talking about SFAS 87 on defined-benefit pensions, and our prior criticism of this standard has been greatly bolstered by a paper called, "Did Pension Plan Accounting Contribute to a Stock Market Bubble?" It was written by Julia Coronado and Steven Sharp, both with the Federal Reserve Board, which recently released their findings.

The authors conclude that what they characterize as "opaque" pension accounting contributed to the stock markets' irrational exuberance during the late 1990s, saying that "despite the fact that the accounting accruals that arise from pension plans are sometimes a very misleading measure of the underlying value of net pension obligations, we find that the market seems to have focused largely on those accruals." In other words, they're saying that even though it is common knowledge that pension accounting is horrid, their research shows that the capital markets seem to have relied on it.

In addition to its obvious challenges to assumptions of market efficiency, this conclusion also condemns FASB and its constituents for retreating to the footnotes to disclose sensitive but useful information. We have never gotten over this lame excuse in SFAS 87: "The board acknowledges that the delayed recognition included in this statement results in excluding the most current and most relevant information from the employer's statement of financial position. That information is, however, included in the disclosures required." (Par. 104) Of course, if the useful information isn't on the balance sheet, it isn't fully included on the income statement either.

We thus join with Coronado and Sharp in holding FASB's feet to the fire for failing to live up to its own claim in Concepts Statement 5 that "the most useful information about assets, liabilities, revenues, expenses and other items of financial statements and their measures should be recognized in the financial statements." (Par. 9)

Before we comment further, some readers may need more background on GAAP for defined-benefit pensions. Like stock options, pensions are highly risky deferred compensation. In exchange for current work, the company promises to pay employees virtually unknown benefits during the unknown duration of their retirement years.

Because of the risk, federal law (ERISA) requires employers to set aside assets to pay this insane debt. Of course, the value accumulated depends on the amount invested and the returns. The wild markets of the 1990s lulled managers into believing that they had ample assets on hand, especially by assuming that the unprecedented returns would continue indefinitely into the future. Obviously, the transmogrification of the seemingly unstoppable bull market into a ravaging bear has crashed those values and wild expectations about future returns. At the same time, declining interest rates have pumped pension liabilities to record highs, thus producing many underfunded plans.

SFAS 87 is not up to today's economic climate because it is an incredible paragon of artificial smoothing of volatile changes in asset values, actuarial estimates and amended benefit formulas.

Without going into great detail, SFAS 87 covers up the real volatility in several ways. One offsets the fund's return (an investing outcome) against the periodic pension expense (an operating outcome), thus exaggerating the reported operating income. More significantly, the amount deducted from the expense is the expected return, based on past assumptions about the market instead of actual observations.

SBC Communications, for example, deducted $3.4 billion of expected return in 2002, when it actually had a $3.4 billion dollar loss. This $6.8 billion swing helped produce a reported income of $5.6 billion. (The company's other post-retirement benefit plans produced another $1.5 billion income overstatement.) Note that SBC's management reported these false numbers as mandated by GAAP.

Another compulsory smoothing technique defers actuarial changes in the liability, subject to corridor amortization in the future. For SBC, this provision delayed recognition of a $2.5 billion loss for pensions and another $4.9 billion for post-retirement benefits. Putting all these numbers together, SBC's reported income is exaggerated by a staggering $15.7 billion. Without smoothing, its $5.6 billion of reported earnings would be a $10.1 billion loss. We're not picking on this company - it was unfortunate enough to have its financial statements within arm's length when we were writing.

We're certainly not accusing SBC management of any crime; after all, they stuck to the GAAP path. However, we fault them and the auditors and the investment community and the media and academia for not questioning an acceptable practice that destroys the credibility and utility of reported income and financial position.

So, what does this scandal have to do with the objective of financial reporting? Quite a lot, actually. According to FASB's Concepts Statement 1, financial reporting is to "provide information that is useful to present and potential investors and creditors and other users in making rational investment, credit and similar decisions." (Par. 34)

How can we disagree with such a virtuous objective? Simply because it sets the bar too low by permitting the useful information to be sequestered in the footnotes. Coronado and Sharpe describe the consequences: "Complicated distortions embedded in the bottom-line figures that are emphasized in financial statements and press releases can distort security market prices substantially, even if the underlying details disclosed in the footnotes to financial statements can be used by experts to more accurately measure value."

This condemnation of practice is an empirical observation based on hard data, not merely a casual opinion.

The higher bar that we would like to raise is simple: Managers should report useful information to enhance capital market efficiency. Despite their flaws, U.S. capital markets are highly efficient, but not completely so. Thus, we think that they will be more efficient and stable if managers present more complete information.

What this Federal Reserve study reveals is that simply delivering information in obscure footnotes isn't enough. When we frame the objective as the pursuit of efficiency, FASB's and management's jobs are not done when data is banished to obscure footnotes instead of being made readily consumable in some other form, preferably the statements themselves.

Because of these flaws, we are pleased to note that FASB Chairman Robert Herz has described SFAS 87 as "a Rube Goldberg device that doesn't even work." However, we no longer expect FASB to provide the only cure. All managers have the ability to voluntarily improve their financial reporting. Whether the improved information results from a new standard or a voluntary corporate initiative, investors and other users will benefit, and so will the managers themselves. As Quality Financial Reporting shows, providing more informative financial reports lowers investor risk, and that translates into a lower cost of capital and higher stock prices.

Although we advocate voluntary disclosure, GAAP for pensions and other benefits must be reformed for the sake of the credibility of all financial statement items.

Paul B.W. Miller is a professor at the University of Colorado at Colorado Springs, and Paul R. Bahnson is a professor at Boise State University. The authors' views are not necessarily those of their institutions. Reach them at paulandpaul@qfr.biz.

For reprint and licensing requests for this article, click here.
MORE FROM ACCOUNTING TODAY