Editor's note: On Monday night, TSC's Peter Eavis received the 2005 Gerald Loeb Award for Commentary, recognizing his outstanding work uncovering and explaining accounting problems at Fannie Mae. Judges in the country's top annual awards for business journalism said Eavis "spelled out the accounting troubles at Fannie Mae long before any findings of wrongdoing were raised and dug into Fannie's financial statements finding numerous areas of concern, even going so far as to tell federal regulators where to look." Below, in chronological order, are five key columns Eavis wrote on Fannie last year.

Jan. 30, 2004: "Freddie Flap Highlights Fannie Flaw"

It became sorely apparent Thursday that there is no way that Fannie Mae ( FNM) could survive the ordeal that its sibling Freddie Mac ( FRE) is going through.

Freddie has fessed up to a range of accounting missteps over the past six months but still hasn't released financial statements for any period in 2003. Because of the failure to provide 2003 numbers, Freddie's regulator told the company Thursday that it must hold capital that is 30% in excess of its statutory minimum capital requirement. Freddie almost certainly already has enough capital on its books to meet the extra 30% requirement.

But if Fannie were forced to do the same, it would be in dire straits, because it wouldn't have enough capital. In fact, it would have to issue at least $7 billion in stock to pad out its capital number, which, at current levels, looks precariously low. Defenders of Fannie would argue that Fannie isn't going to find itself in trouble with its regulator, the Office of Federal Housing Enterprise Oversight, because it hasn't got into any trouble over its accounting. But in the midst of the Freddie revelations, OFHEO decided it might be a good idea to conduct a probe of Fannie's accounts to see whether it was also up to no good.

The investigation has only just begun. There are plenty of areas where Fannie's accounting looks suspect , so it would not be a total surprise if OFHEO found unsatisfactory financials at Fannie. If it did, it would be a wise regulatory move to instantly demand that Fannie raise a huge slug of new capital to see it through the turmoil, simply because the giant mortgage buyer is much more weakly capitalized than Freddie.

Capital Idea

How do we sort our way through the numbers to show that Fannie would be reeling if OFHEO forced it to do what Freddie did? Let's start with what capital actually is. At financial institutions, capital measurements involve subtracting liabilities from assets to see how much "cushion" against shocks the balance sheet has. For government-sponsored entities such as Fannie and Freddie, there are a number of complex capital yardsticks. One key measurement OFHEO looks at is called the minimum capital requirement. This is a statutory requirement arrived at by mandating that capital must be equivalent to at least 2.5% of assets on Fannie's or Freddie's books, plus 0.45% of the mortgages they have guaranteed but don't own.

In order to pass the minimum capital requirement test, Fannie and Freddie's so-called core capital has to be in excess of the minimum capital requirement. And core capital is basically assets minus liabilities, but excluding gains or losses on certain types of derivatives, which are financial instruments used to hedge against interest rate moves.

So how do Fannie and Freddie's numbers stack up when it comes to these capital measures?

As of Nov. 30, Freddie had a minimum capital requirement of $25.1 billion, but its core capital was well in excess of that at $33.2 billion. If we increase the minimum capital for Freddie by 30%, as OFHEO demands, the new hurdle that Freddie must meet is $32.6 billion. In other words, its core capital is over $500 million in excess of the harsher requirement.

In contrast, Fannie fares very badly under this exercise.

At the end of last year, Fannie said I had an estimated minimum capital requirement of $31.5 billion, exceeded by its core capital of $34.4 billion. However, if we add 30% to the minimum requirement, we get $41 billion. That is $6.6 billion over Fannie's core capital.

Clearly, if Fannie finds itself in Freddie's shoes, the stock would be hammered, as investors would fear a large dilutive equity issuance to boost capital. Of course, any whiff of accounting chicanery at Fannie would already have sent the stock skidding, but fear of dilution would add to the slide.

Derivative Dealings

Indeed, if both companies are found to have accounting issues, the regulators would almost certainly consider introducing new capital rules. If, as is sensible, the authorities started to require that the companies include net derivatives gains or losses in their capital reporting, Fannie would be in the ICU in minutes. Freddie might be OK, though.

If we include the derivatives losses at Fannie, we arrive at a capital number that is equal to shareholders' equity, as measured under generally accepted accounting principles. At the end of last year, Fannie's shareholders' equity was $22.4 billion, a mere faction of its $1 trillion in assets. The real shocker is that Fannie would have to almost double its shareholders' equity for it to meet the $41 billion minimum capital requirement (increased by 30%). That would mean issuing over $20 billion in stock, which is over a fourth of its market value.

Some defenders of Fannie might think that using shareholders' equity is unfair, because they contend that the vast majority of derivatives losses included in that capital measure will be made back in the future. Not so. First, most of the net $12 billion in net derivatives losses are likely to be unrecoverable . Second, the fact that Fannie refuses to give a number for unrecoverable losses, despite congressional pressure , only reinforces the notion that they are scandalously huge.

It might sound drastic to make Fannie issue $20 billion of stock. But the company really ought to do this anyway, because its equity is way too low for the risky manner in which it runs its business and for the size of its asset base. Issuing that much stock may hurt a good number of investors, but failing to do so only risks undermining the entire American housing market at a later date.

To see how badly Fannie has been managed, we need only look at Freddie's shareholders' equity. At the end of 2002 (the latest number available), it was $31.3 billion, but that figure has probably grown through the addition of retained earnings through 2003. That $31 billion number is just short of the $32.6 billion minimum requirement (increased by 30%). And with 2003's profits, Freddie's shareholders' equity, which includes all derivatives gains or losses, could exceed the tougher minimum requirement. Quite simply, Freddie hasn't been taking billions of dollars in unrecoverable losses on its derivatives.

This is not an argument to invest in Freddie. Any company that can't report financials for over a year should not be touched. But it doesn't say much for Fannie if even scandal-plagued Freddie can put it to shame so easily. Just consider the nightmare if Fannie's books are found to be cooked by OFHEO.

April 1, 2004. "A New Reason to Fret About Fannie"

An unusual remark from Fannie Mae's ( FNM) regulator Wednesday suggests that the nation's second-largest financial institution could soon become the latest company to face accusations of accounting irregularities.

The Office of Federal Housing Enterprise Oversight, which regulates government-sponsored Fannie and its smaller rival Freddie Mac ( FRE), said that a special accounting review that it is conducting of Fannie's books "may result in a restatement of prior period results and a revision of the respective capital calculations." It is extremely rare for a regulator to raise the possibility of accounting adjustments at a company if there are no grounds for suspecting restatements will have to be made.

In a prepared statement, Fannie commented: "The language in the letter with respect to the OFHEO special examination straightforwardly states the fact that it is ongoing. As previously announced, OFHEO has recently retained a national accounting firm to supplement the agency's efforts and obviously has not reached any conclusions. We look forward to continuing to work with OFHEO on the examination."

Last year, Freddie got into trouble for abusing its accounting to understate earnings. The scandal would have been far worse if Freddie had been overstating past numbers, because the company would've been fooling outsiders into believing that it was stronger than it really was.

OFHEO has not said whether Fannie has been overstating or understating earnings, but there are several ways in which Fannie may have been giving its reported numbers an artificial boost . One area of accounting that has received a lot of attention is the valuation of a portfolio of troubled bonds. Fannie may have overstated earnings and capital in 2003 by failing to take the appropriate amount of losses on these bonds.

The bonds in question are backed by loans made to purchase mobile homes. The original lender, Conseco Finance, went through a bankruptcy restructuring in 2003. As a result of the restructuring, the mobile home-backed bonds that Conseco sold have a lesser claim on the cash flows from the underlying loans. This has resulted in their credit rating being reduced. At the end of last year, Fannie had $8 billion of mobile home bonds on its books, 70% of which were originally serviced by Conseco.

Detox first zeroed in on these troubled bonds over a year ago and took some more close looks at the portfolio in the fall. It concluded that Fannie's writedowns on the bonds -- just over $200 million at the end of last year -- weren't sufficient. Using a similar writedown taken by another institution on similar bonds, this column calculated that a writedown of around $1 billion might be in order. But a loss of that size would have been extremely embarrassing for Fannie at a time when Freddie was getting heat for its accounting and when its critics in Congress and the White House were formulating legislation to tighten regulation of Freddie and Fannie.

In addition, taking a large writedown in 2003 could have prevented Fannie from hitting an earnings number that allowed senior executives to receive a lucrative grant of stock options.

Monday, an American Banker article, citing several sources familiar with the matter, said that OFHEO examiners had focused on the mobile home portfolio. The article said: "The sources said the agency has serious concerns that Fannie may not be showing enough of a loss." OFHEO is doing nothing to dispel the American Banker story by saying Tuesday that its examination may result in accounting restatements at Fannie.

If OFHEO does find that Fannie has underestimated losses on its Conseco bonds, it will be much harder for the company to stop legislation that aims to give a new regulator greater powers.

April 6, 2004 "Fannie Flap Points to Options Grants"

When a regulator raises questions about a company's accounting, the stock usually gets crushed.

Mysteriously, this hasn't happened to Fannie Mae ( FNM) in the wake of last week's flap over its books. But with all the worries swirling around the mortgage giant, bullish investors could be showing a bit too much confidence in management.

Last Thursday, the Office for Federal Housing Enterprise Oversight (OFHEO) surprised Wall Street by saying Fannie "may not have applied the proper accounting guidance" when writing down the value of certain assets. Though the remarks were highly unusual, the market has since shrugged them off: On Monday, Fannie's stock closed at $75.05 -- which is 1% above Wednesday's closing price.

Even so, it's worth noting that Wall Street often ignores the first whiff of scandal (recall Enron). And when it comes to Fannie, there is enough material in the public domain to suggest that OFHEO is right to have big doubts about the company's bookkeeping.

Fannie, which throws its weight around in Washington with its lobbying efforts, is unlikely to win this battle. Fannie's numbers strongly suggest that the company wanted to avoid taking losses on certain assets. The motive may have been to ensure that 2003's per-share earnings exceeded $6.46, the level at which senior execs, including CEO Franklin Raines, get to cash in a lush options grant. The company may also have feared a big writedown in assets would strengthen the hand of politicians who want to strengthen regulation of Fannie and also Freddie Mac ( FRE), which was at the center of an accounting scandal last year for understating its earnings.

At the same time, OFHEO, criticized in the past for being ineffectual and uninquisitive, is showing that it is not afraid of Fannie. Washington chatter says that the battle between Fannie and OFHEO has gotten so highly charged that it could result in Raines losing his job if Fannie is seen to be at fault. If OFHEO doesn't come up with any real dirt on Fannie after its review, OFHEO Chief Armando Falcon would be the one resigning.

Investors must care because proof of accounting missteps could trigger a drawn-out process of accounting restatement at Fannie similar to the one that is occurring at Freddie, which has yet to release 2003 financials. Fannie, which holds one dollar of capital for every $45 of assets, is much more weakly capitalized than other financial institutions, which means it is much more vulnerable to shocks. If Fannie had to hold more capital during a restatement process, it would have to raise as much as $15 billion to get leverage to a reassuring level. A capital repair job of that size could involve issuing billions of dollars in stock, which would dilute existing shareholders. And regulators would almost certainly impose limits on future growth.

So what is OFHEO looking at? Its review is giving Fannie's books a broad sweep, but the agency appears to have found something it doesn't like about how Fannie has valued bonds that are backed by mobile home loans, as well as other assets. Detox looked closely at this issue last fall and zeroed in on mobile home loan bonds issued by a company called Conseco Finance, which had gone into bankruptcy protection.

This column argued that Fannie's writedowns on its Conseco bonds were way too small, judging by an impairment charge on similar bonds taken in late September by the Federal Home Loan Bank of New York. This column calculated that, going by the size of FHLB of New York's writedown, Fannie needed to reduce the value of mobile home bonds issued by Conseco by $1.15 billion. In the third quarter of last year, Fannie had written down its mobile home bonds by only $150 million, with $100 million having taken place in 2003.

But why would Fannie want to risk accounting scrutiny by failing to write down these bonds? Perhaps it feared giving its enemies in Congress any ammunition. But Fannie execs may also have feared that a large writedown could deprive them of a generous options grant they would receive if Fannie's core business earnings were $6.46 per share or higher in 2003. Taking a big writedown in the fourth quarter could have left earnings uncomfortably close to the $6.46 mark.

Here is the math. In 2003, Fannie made $7.29 per share in core earnings, which is 83 cents above $6.46. Where would earnings be if Fannie had taken a $1.15 billion impairment charge?

First, Fannie had already taken $150 million in writedowns on Conseco bonds by the fourth quarter, which leaves $1 billion. If we tax-effect that sum at Fannie's tax rate of 26%, we get $740 million. With 974 million shares outstanding at the end of last year, $740 million works out at 76 cents in per-share earnings. Subtract that from the reported core earnings of $7.29 and we get $6.53 -- a mere 7 cents above the $6.46 target.

Fannie has yet to release its proxy statement for 2003, which should detail last year's compensation. But in light of the facts, OFHEO may also want to take a good look at that document.

Sept. 2, 2004: "Fannie Probe Turns to Derivatives"

Did mortgage giant Fannie Mae ( FNM) cook its books in order to exclude large derivatives losses from its income statement?

Finding an answer to that question is one of the main aims of a probe being conducted by Fannie's regulator, the Office of Federal Housing Enterprise Oversight, or OFHEO. And according to a person familiar with the investigation, one of the reasons that OFHEO has sent subpoenas to Fannie is to obtain information that would help it decide whether the company misapplied generally accepted accounting rules to keep losses on derivatives out of earnings.

As this column has often reported , Fannie has suffered gargantuan losses on derivatives under CEO Franklin Raines. The losses, which ballooned to nearly $17 billion in 2003, appeared to be the result of a risky interest rate hedging strategy that backfired badly. The derivatives losses in question are amortized into earnings over time. The OFHEO investigation aims to determine whether Fannie was amortizing too few losses into its income statement and thus artificially boosting earnings, according to the person familiar with the probe, which began in February and is expected to end by the end of this month.

Fannie has strived hard to give Wall Street earnings that grow at a strong and steady rate. If it was able to do that only because it was abusing its accounts to keep derivatives losses out of earnings, Fannie's reputation would be hurt badly and CEO Raines would come under fire.

Fannie didn't comment. Its stock dropped 60 cents to $73.85 Wednesday.

OFHEO declined to comment on any aspect of its investigation, which the regulator launched soon after its probe of accounting missteps at Freddie Mac ( FRE), Fannie's rival mortgage buyer. Having seen some of the tricks Freddie played, OFHEO said right from the start that its probe of Fannie would look at how earnings were calculated. Back in December, OFHEO said that one of the things it would do in its Fannie probe was to "focus on transactions that significantly accelerate or defer the pattern of income recognition."

With that being one of its main aims, it made perfect sense for OFHEO to look at how Fannie's billions of dollars of derivatives losses were treated. In theory, the accounting rule that covers derivatives -- known as FAS 133 -- could be abused to keep a portion of derivatives losses out of earnings. In a November 2003 letter from OFHEO director Armando Falcon to Raines obtained by TheStreet.com, OFHEO asked for documents pertaining to a range of subjects, but specifically requested documents "related to FAS 133."

Preparations for FAS 133 were started under former Fannie chief James Johnson, now a private equity investor and adviser to John Kerry. Johnson didn't return a call seeking comment.

However, Fannie may be doing all it can to avoid scrutiny of its derivatives accounting. At a congressional hearing in July, OFHEO director Armando Falcon has said that Fannie's cooperation with the probe has been "spotty." According to an AP report, he also said, "We've had some instances where deadlines have been missed without explanation, submissions in response to requests for information that weren't complete -- even though there were assertions that they were complete."

The lack of cooperation may have been the reason for the issuance of subpoenas, a development reported by The Wall Street Journal on Aug. 20. According to the person familiar with the investigation, the subpoenas are aimed at getting documents that would show whether certain accounting treatments were implemented as part of willful attempts to artificially smooth earnings.

Of course, while the derivatives losses don't all go into earnings when they occur, they have all been reflected in the balance sheet -- in the equity number. However, to the bemusement of many investors, Fannie has asked market participants to look at equity numbers that leave out the FAS 133-related derivatives losses when assessing the strength of the company's capital.

At the end of June, there were $8.5 billion of unrecoverable derivatives losses in Fannie's equity that had yet to be amortized into earnings. If, as makes sense, that sum were subtracted from the capital numbers Fannie asks investors to concentrate on, the company would look weakly capitalized. Since Fannie was betting on a drop in bond prices at the end of the second quarter, the recent rally in bonds may have created more derivatives losses, and the company's true capital base may be even weaker than it was at the end of the second quarter.

OFHEO's ongoing probe has scored a couple of successes. It led to Fannie changing its accounting for some impaired bonds, an issue this column was first to flag. Fannie ended up having to book heavier losses on the bonds.

Also, it would appear that OFHEO is leaning on Fannie to make some changes in its corporate structure to improve risk management. Last week, Fannie said it was going to "move transactional risk management responsibilities to its business units while strengthening the integration of its financial risk assessment capabilities at the corporate level." Why the changes? Well, in the press release containing the news on the organization change, Fannie CFO Timothy Howard implied the impetus for changes in risk management were coming from OFHEO.

Howard said that "because OFHEO's proposed corporate governance rule addresses issues related to the organization of risk management activities, the company anticipates making further changes to its risk-management structure once the final rule is in place."

Falcon's term at OFHEO is due to come to an end at the beginning of October, but an OFHEO spokeswoman said Falcon wants to ensure a smooth changeover and, if necessary, he would stay through the transition to the new director.

A Bush victory in November would be a disaster for Fannie, which has resisted reforms proposed by the White House to tighten regulation of Fannie and Freddie. The two companies have been told by the White House that a second Bush administration's stance would be tougher, according to a person familiar with relations between the administration and the government-sponsored entities, or GSEs.

However, Fannie management could find it hard to resist reform calls even under a possible Kerry presidency if OFHEO finds it did keep derivatives losses out of earnings.

One of the big corporate tricks of the '90s was to find ways to manipulate accounting to defer or bury losses. There would be few supporters left in Congress if Fannie were discovered to have had engaged in Enronesque practices and ethics.

Nov. 16, 2004. "Fannie Mae: Day Late, Many Dollars Short"

Mortgage giant Fannie Mae ( FNM), with its government-sponsored status and its perfect credit rating, has long been seen as one of the bluest blue chips in the stock market.

But Monday, Washington-based Fannie became one of market's biggest delinquents, by failing to file results for its third quarter with the Securities and Exchange Commission on time and by announcing that it may have to book a staggering $9 billion of losses after its regulator questioned the way it accounts for financial instruments called derivatives.

A charge of $9 billion would rank among the biggest ever outside of the telecom industry meltdown and would almost certainly lead to the departure of Fannie's management team, led by CEO Franklin Raines. It also would give the White House and certain forces in Congress the ammunition they need to press forward with far-reaching reforms of Fannie Mae and its rival Freddie Mac ( FRE). Both companies have been hit by accounting scandals, and both operate under special advantages granted by congressional charter.

In September, Fannie's regulator, the Office of Federal Housing Enterprise Oversight, or OFHEO, published a report that argued that Fannie had not properly applied a derivatives accounting rule known as FAS 133. As a result, OFHEO contended, Fannie's earnings and a key measure of capital may have been substantially overstated.

Monday, Fannie said that if the SEC finds that Fannie didn't apply FAS 133 correctly, the company will have to record a $9 billion after-tax cumulative loss in earnings. By definition, that sum would also have to come out of capital, almost certainly pushing Fannie's capital below the level required by regulators and by congressional statute.

Detox estimates Fannie would be undercapitalized to the tune of $11 billion if Fannie has to book a $9 billion charge.

That $9 billion net loss figure is large enough, but even that number doesn't adequately show the full losses that Fannie may have kept out of earnings by allegedly misapplying FAS 133. It is important to understand that the $9 billion is a net figure, made up of after-tax derivative losses of $13.5 billion and after-tax gains of $4.5 billion. The before-tax number for those losses is actually $20 billion. If it is found that FAS 133 was incorrectly applied at Fannie, the talk will soon turn to motive. Clearly, with derivatives losses of that scale, Fannie execs may have deliberately decided to misapply FAS 133 to keep its stock up, and keep regulators and its political opponents off its back.

Indeed, there is every reason to believe that any misapplication of FAS 133 was not a mistake. In its report on Fannie's accounting, OFHEO included a quote from Fannie accounting chief Jonathan Boyles that suggests the company knew it wasn't applying FAS 133 correctly. In an interview in August as part of OFHEO's probe of Fannie's accounting, OFHEO says Boyles said the following: "We have several known departures from GAAP in our adoption of FAS 133. We have cleared those with our auditors." (GAAP stands for generally accepted accounting principles, which are the accounting rules that the SEC expects public companies to follow.)

Monday, Fannie didn't say when it would file its third-quarter results, but said it failed to do so on time -- Monday was the deadline -- because it was "advised by its independent auditor that it is unable to complete its review" of Fannie's third-quarter results. Last week, this column predicted that Fannie's auditor, KPMG, would be unlikely to give its approval of the numbers.

Fannie said that it believes that its applications of FAS 133 and FAS 91 -- another rule OFHEO thinks Fannie incorrectly applied -- are "consistent with" GAAP. Fannie said that it has made its case to the SEC on both rules, with KPMG concurrence. However, Fannie also said that it will "modify its accounting, if necessary, to comply with the SEC's views."

Clearly, Fannie hopes the SEC will come down on its side. But, as Detox argued last week, the chances of that happening are slim. The SEC understands FAS 133 extremely well and is fully aware of the ways in which companies try to misapply it to make their earnings look better than they really are.

Some Fannie supporters have argued that the SEC may just tell Fannie to change its FAS 133 accounting, but allow it to avoid a costly and damaging restatement of past results. But if $9 billion of losses have been left out of earnings, the sum is so large that anyone in the SEC would have a very hard taking the "go-and-sin-no-more" line.

The nub of the issue is whether Fannie should have given nearly all its derivatives a special treatment that allows companies to keep losses on them out of earnings in the period they occurred. Instead, they accumulate on the balance sheet -- in a special equity account -- where they are amortized into earnings over time. For derivatives to qualify for this special treatment to stay out of earnings, companies have to show very carefully why they qualify -- and this requires careful measurement of the derivatives' value and extensive documentation.

If a company can prove that a derivative possesses "hedge effectiveness" -- i.e., changes in its value closely offset the change of a value of the asset or liability being hedged -- any losses (or gains) stay out of earnings in the period they occur and go onto the balance sheet.

OFHEO charges that Fannie has failed to show why its derivatives qualify for hedge effectiveness in the vast majority of cases. FANNIETOX

The SEC has made a point of looking out for ways in which companies abuse FAS 133 to obtain hedge effectiveness for their derivatives. In a speech last December, an official from the SEC's Office of the Chief Accountant said that the SEC had "recently observed situations of 'sloppy' documentation and aggressive interpretations of Statement 133's hedge accounting guidance." OFHEO alleges that this is what happened at Fannie.

Fannie lobbied hard to prevent the implementation of FAS 133. It forced some changes, but the rule that became effective in 2001 was still not to Fannie's liking. It is therefore believed that Fannie, which has never been afraid to use its huge influence in Washington, never really tried to implement a rule it didn't like.

The key concern for the market is how much capital Fannie now has to raise. In the midst of its accounting problems, Fannie agreed with OFHEO in September that it would increase its minimum capital requirement by 30%. That means that Fannie's new capital requirement is around $41 billion. At the end of September, Fannie likely had so-called core capital of $38.5 billion. Subtracting $9 billion of net losses would take that down to $29.5 billion, which is $11.5 billion below the minimum requirement.

If you liked this article you might like

Selloffs Suggest a Looming Credit Crunch

Selloffs Suggest a Looming Credit Crunch

GM Lending Arm Pumped Up

GM Lending Arm Pumped Up

Berkshire's In for a Bruising

Berkshire's In for a Bruising

Fannie Mae's Trust Fund Troubles

Fannie Mae's Trust Fund Troubles