Testing Fannie Mae's Tattered Balance Sheet
By Peter Eavis
Senior Columnist
03/25/2003 08:09 AM EST
thestreet.com
To listen to Fannie Mae (FNM:NYSE) , you'd think that last year's multibillion-dollar erosion of shareholder wealth hardly mattered. But it does, and Fannie's beaten-up stock won't recover till the company's management is more forthcoming about how the loss came about.
By the end of last year, shareholder equity -- the accumulated wealth available to owners of the giant government-sponsored mortgage lender -- had plunged some $4.5 billion from its first-quarter high, a drop of 22%. The picture is even worse if you exclude a $4 billion equity gain created out of thin air by an extraordinary shift of assets.
This sharp decline did receive coverage in the press last fall, but Fannie and its Wall Street backers rejected the concerns. They noted that the equity declines in question aren't reflected in net income and pointed out that other key measures of capital remained strong; accounting rules, not poor financial management, were to blame for the hole in equity. Overall, the bull claim is that the decline in equity cannot be treated as a true economic loss. A Fannie spokeswoman reiterated those arguments when asked for comment.
When you look more closely, however, the debate over how the equity evaporated is far from being resolved. For instance, there is no hard evidence that accounting rules have led to a major misrepresentation of Fannie's balance sheet. And it appears that Fannie left itself vulnerable to a sharp drop in interest rates last year. Moreover, the response it then made to reconfigure its balance sheet appears to have "locked in" at least part of the big loss that showed up in equity. In other words, shareholder wealth that was once there will not be easily recouped if interest rates move up or stabilize.
In stark contrast, Freddie Mac (FRE:NYSE) , which buys and guarantees mortgages just like Fannie, showed no drop in equity last year. In fact, Freddie's equity jumped 60% to $24.6 billion in 2002. While Freddie may do its accounts slightly differently from Fannie, it is unlikely that this substantially skews a comparison of each company's equity. Yet Freddie's stock trades at only 1.5 times its equity, or book, value, compared with Fannie's 4 times.
Monday, Fannie's stock fell $1.50, or 2.2%, to $66.12.
Big Challenge
So what actually happened last year? The bear thesis is that Fannie's managers were brutally caught out by a sharp drop in market interest rates that took place from March to September last year. Declines in rates are a big challenge to Fannie and Freddie. While lower interest rates bolster house prices and mortgage volumes, they also lead to quicker prepayment of the mortgages that lenders hold as assets on their balance sheets. The problem for Fannie and Freddie is that they have borrowed money to buy those mortgages. Therefore, during times of widespread mortgage prepayment, the price of that debt also needs to come down if profit margins are to be maintained. To get the price of their debt down quickly, Fannie and Freddie need to have the ability to cancel outstanding debt and reissue it at lower rates.
But how does this trace back to equity? Well, Fannie and Freddie effectively build flexibility into their liabilities partly by using derivatives to "hedge" them. And each quarter, the change in the value of these derivatives is factored into the equity value, mainly through a line called "other comprehensive income," which doesn't affect net income.
Fannie's Eroding Equity
Shareholders' equity at Fannie Mae, with and without a
$4.1 billion one-off gain
* The adjusted equity totals exclude a one-off increase of $4.1 billion that was created in Q3 by shifting assets in the balance sheet.
Source: Freddie Mac, Detox
The mark-to-market loss on Fannie's "cash flow hedging" ballooned last year, going from $4.8 billion in the first quarter to $16.3 billion by the end of the year. In Fannie's other comprehensive income presentation, that fourth-quarter loss is offset by net unrealized gains on securities of $4.5 billion, but roughly $4.1 billion of that is due to a reclassification of assets in Fannie's balance sheet that took place in the third quarter.
The lion's share of the losses in the cash-flow hedging was probably caused by a drop in the value of derivatives, called pay-fixed swaps, that require Fannie to pay a fixed-rate interest stream to a counterparty while receiving a floating rate of interest in return. When rates drop, these types of swaps become less valuable. Indeed, in the third quarter, the loss on cash flow hedges was $6.9 billion, though Fannie doesn't break out in detail how that loss was arrived at.
Bustling Hedge Row
The picture was very different at Freddie. Last year, it also made a big loss on cash-flow hedges, including pay-fixed swaps, of $12 billion. But, critically, it had also bought insurance using types of derivatives that rise in value in a falling interest rate environment. Those effectively offset the cash-flow derivatives loss. Buying that protection can be expensive, and that cost goes a long way toward explaining why Freddie has slimmer margins than Fannie.
Some on Wall Street believe that Freddie has more gains than Fannie simply because it classifies its mortgages in a different way. But these analysts seem not to realize that the gains Freddie does show on its mortgages would not occur had it not hedged the prepayment options embedded in its mortgages.
Next, Fannie's defenders say: Why should investors care about derivatives losses that affect equity but not net income? Well, just as there is more than one way to skin a cat, there is more than one way to value a company. Earnings aren't everything. Equity is what's left over after assets are subtracted from liabilities. And if derivatives losses at Fannie weren't flowing through comprehensive income, they would effectively be increasing liabilities. And investors have always run from companies that allow the liability side of their balance sheet to grow faster than the asset side -- even when earnings are rising.
Freddie's Rising Equity
Shareholders' equity at Freddie Mac
Source: Freddie Mac, Detox
Fannie's boosters often respond that if all of the company's balance sheet were marked to market -- and not just its derivatives -- a sizable part of the equity hole would be recovered. They claim that would happen in a falling rate environment because the fair value of the mortgage-backed securities would be higher.
But there are two flaws in this line of thought. First, liabilities would also increase in fair value, possibly offsetting most of the gain in assets and leaving equity more or less unchanged. This is effectively what happened in Fannie's 2001 fair-value balance sheet, released last spring in its annual financial statement for 2001. Second, if the fair-value picture is so much better, why doesn't Fannie just produce a fair-value balance sheet every quarter to clear this all up? Despite the stock-crunching brouhaha of the past six months, Fannie hasn't seen fit to release such a balance sheet, leaving investors to wait for its 2002 annual statement, which should be out next month.
Duration
The strongest argument in Fannie's favor is that its derivative losses will be recouped as interest rates adjust. But this idea, too, is weak. To understand why, investors need to grasp the important concept of duration and something called the duration gap. Duration estimates how sensitive a bond's price is to a change in interest rates. For example, if a bond has a duration of seven years, a 1% drop in interest rates will raise its value by 7%, while a 1% rise in interest rates will lower its price by 7%.
A duration gap of an entire portfolio estimates how the values of assets and liabilities (and any derivatives used to hedge them) change relative to shifts in interest rates. A duration gap of minus one year, for instance, means liabilities would rise 1% more than assets if interest rates were to fall by 1%.
In August 2002, the duration gap was minus 14 months. That's just another way of saying that Fannie's risk managers had arranged their balance sheet so that its liabilities would reprice upwards much faster than its assets if interest rates fell. They made a conscious choice to take that stance, meaning they consciously opened themselves up to risk that faster-rising liabilities would crush equity, which is of course what happened.
But doesn't a large negative duration gap mean Fannie is poised to recoup its "lost" equity if interest rates rise? Yes, but Fannie took substantial measures to decrease its duration gap in the second half of last year. And a smaller negative duration gap means a less positive impact on equity if rates rise. But why did Fannie act to close its gap, and thus lose the chance to reap gains when rates rose, when it was telling the market that the gap was not a danger signal?
The reason has to be that it was a danger signal, at least to other people. A duration gap of minus 14 months means that another big drop in interest rates could have pushed Fannie's real equity -- just over $12 billion at year-end, excluding the fancy asset shifting -- into single figures. If that had happened, investors, regulators and anti-Fannie politicians would have been bringing up the "I-word" -- insolvency. In the past, Fannie had allowed interest rate moves to deal with outlying duration gaps, but in the fall of last year it was just too big. The lack of hedging allowed it to get that big. And that reluctance to buy insurance shows that Fannie is a big risk taker.
When management fesses up to that and pledges to be more like its brother Freddie, earnings estimates may come down, but it may also be time to buy Fannie. |
