Our column of Dec. 15, 2008, ("Tick, tick, tick goes the P-bomb," page 15) described two pending disasters. The first was terrible financial results faced by companies with defined-benefit pension and other post-employment benefit plans because of what the market's recent free fall did to their fund assets. The second disaster was the forthcoming unfaithful representations of those results in employers' financial statements. Turns out we were right, which is too bad for everyone. Lest anyone think otherwise, we take no pleasure in what we're writing about.

In that column, we looked at several companies and suggested what scale of numbers could possibly appear in their 2008 reports. Because our estimates were based on reported 2007 numbers, they were not right on target, but they seem to be in the neighborhood. The real numbers are even uglier than we thought, but we're yet to see anyone in the press pick up on this story, so we're back again with more bad news.

A few decades back, there was talk about exploding neutron bombs over enemy territory, unleashing a deadly force that would kill every living thing while leaving structures intact. In a sense, GAAP for pensions is an accounting neutron bomb. Even though huge losses have destroyed vast amounts of many companies' real wealth, their financial statements are still standing and in many cases virtually untouched by the economic turmoil.

This is no laughing matter, because the gross distortions in these ostensibly undamaged financial statements will lead some individual investors totally astray, while the lack of public accountability will lull managers and regulators into doing nothing to fix fundamental economic and accounting problems.


To explain, SFAS 87 and its one-baby-step-at-time cousin, SFAS 158, created and perpetuated convoluted smoothing techniques that buffer employers' financial statements against shocks from unexpected real changes in values of pension and OPEB assets and liabilities. Of course, this buffering helps managers escape the slings and arrows of criticism.

The good news is that we've seen evidence that suggests that the reported numbers don't fool the capital markets, because these companies' stock and bond prices reflect real results. (Chalk another one up for capital market efficiency.) Nevertheless, the disconnect between what is happening and what is reported should give everyone a big headache.


We set out to quantify the differences between GAAP results and the unfiltered truth in order to get a handle on the magnitude of the distortions and other covered-up truths. We dissected some prominent companies' 2008 10-Ks for the following items related to pension and OPEB plans:

* Reported cost: The imaginary GAAP number that subtracts expected asset returns from service and interest cost and then adds and subtracts various adjustments derived from long-ago events, including plan amendments and deferred gains and losses.

* Real cost: Service cost plus interest, plus or minus actuarial adjustments to the liability, plus or minus actual return on the assets.

* Reported income: Pretax GAAP income from continuing operations.

* Adjusted income: Pretax GAAP income from continuing operations adjusted by adding back the reported cost and subtracting the real cost. (We assumed any capitalized costs carried forward from 2007, such as cost of goods sold, were offset by capitalized costs from 2008 pushed into the future.)

* Debt/equity ratio: Total reported liabilities divided by total reported equity.

* Adjusted debt/equity ratio: The sum of total reported liabilities plus the value of the pension fund, all divided by total reported equity (doing so overcomes the offsetting of pension/OPEB assets and liabilities under SFAS 158).

We compiled results for a non-random sample of eight of the top 25 companies identified by Fortune.


Three of the eight firms we looked at had notable differences between their reported and adjusted incomes.

General Electric reported annual cost of $2.4 billion, but we estimate its real cost was $22.3 billion (ouch!). Adjusting for this nearly $20 billion understatement turns the reported profit of $19.8 billion into a $0.1 billion loss. For some reason, the company's press releases have not mentioned this disparity.

IBM (which froze its plan several years ago) reported a net gain of $0.2 billion from pensions and OPEB. Our analysis computes a real cost of $22.9 billion, which would convert its reported profit of $16.7 billion into a $6.4 billion loss.

AT&T reported annual cost of $0.3 billion, but its estimated real cost is $25.5 billion. Adjusting for this error wipes out the company's reported profit of $19.9 billion and turns it into a $5.3 billion loss.

Some may try to label these understated reported costs and overstated reported incomes as "favorable" for shareholders in the sense that they prevent the markets from getting the bad news.

That is complete and utter nonsense.

Like we were forced to do, the markets do the best they can with published information, knowing full well that there is more to the story. Leaving the markets gasping for more data only adds to their uncertainty, leading to more risk, and more discounted stock prices. Only a naive observer would conclude that deliberate earnings misstatements could be favorable for anyone.


The next two companies are seeking federal assistance for obvious reasons; the probability that these efforts won't work is evidenced by our findings.

General Motors reported $2.7 billion of annual cost, but our real amount is $16.3 billion. Adjusting for this difference turns the company's operating loss of $29.4 billion into an unimaginable $43 billion loss. The debt-to-equity ratio isn't meaningful because equity is negative, but we observed that reported liabilities totaled an oppressive $176 billion, while the adjusted amount jumps to $280 billion. Kaboom!

Ford reported an $0.8 billion pension/OPEB gain for 2008, but the real cost appears to be $4.9 billion, which is enough to push its reported loss of $14.4 billion up to $20.1 billion.

Its debt-to-equity ratio is not meaningful because of its deficit, but its reported liabilities of $291 billion climb to $347 billion when the asset offsetting is undone. Kaboom! (Incidentally, Ford's real cost is relatively low because of a $7.9 billion actuarial gain on its OPEB liability.)


With offsetting in place, IBM has a reported debt/equity ratio of 7.0, which would be bad enough if it were real. Unwinding the offsetting of fund assets against the liability raises the ratio all the way to 12.7, which is tantamount to saying that equity is about 7.3 percent of the total assets' book value.


Three other companies are in better shape, but GAAP accounting for pensions nonetheless produces material errors.

Exxon reported annual cost of $2.1 billion, but we figure it's more like $11.1 billion. Adjusting for that $9 billion difference reduces the reported profit of $81.8 billion down to a still-solid $72.8 billion. The debt-to-equity ratio bumps up from 1.0 to 1.2. Although the errors are not as significant as the earlier ones, a $9 billion deviation in expense and an $18.3 billion understatement of liabilities don't support a conclusion that the accounting principles are working well.

Chevron reported $0.9 billion annual cost but the real cost is $4 billion, converting its reported income of $43 billion to $39.9 billion. Its debt-to-equity ratio is unique in our sample because it is less than one; correcting for the GAAP distortion barely moves it from 0.86 to 0.90.

HP reported annual cost of only $0.1 billion, but the real cost is $0.9 billion, which is enough to nudge its reported income of $10.5 billion down by 8 percent to $9.7 billion. The debt-to-equity ratio is bumped up from 1.9 to 2.4 when the pension assets are considered.


We wish we weren't writing this particular column. If we can figure out better numbers from looking in the footnotes, they should be in the financial statements, instead of the GAAP garbage.

We chide our friends at FASB for tolerating this a-bomb-inable accounting. While we laud them for their progress in 2006, we want to beat up on them for not doing anything visible since then to ensure that these bad numbers will no longer be reported.

Of course, doing so will also force managers to deal with the mess they've created, instead of hiding from the truth behind the flimsy protection of GAAP. Their auditors and lawyers might tell them that's all they need to do, but economics tells a completely different story.

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.

(c) 2009 Accounting Today and SourceMedia, Inc. All Rights Reserved.

http://www.webcpa.com/ http://www.sourcemedia.com/

Register or login for access to this item and much more

All Accounting Today content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access