Earlier this month, the Department of the Treasury released its long-awaited housing finance reform plan, which focused mainly on the two government-sponsored enterprises, Fannie Mae and Freddie Mac. The feds took both GSEs over in 2008, when rising defaults jeopardized their viability. The blueprint's release date was propitious, eleven years almost to the date after the controversial bailout.
But was the controversy justified? As the GSEs' long-overdue exit from government "conservatorship" becomes a more realistic possibility, some market participants with a vested interest are telling us that Uncle Sam actually made money by rescuing the mortgage giants – to the tune of $110 billion. Since putting taxpayer resources into failing firms is such good business, perhaps we should stop worrying and learn to love bailouts.
If you think that's a good idea, I have a "liar loan" to sell you. No, the bailouts were not a sound investment. They weren't really an investment at all in the usual sense, but rather a political imposition on taxpayers in the heat of the meltdown. And they definitely weren't sound. Whoever makes that claim is committing two economic fallacies at once: ignoring alternative uses of scarce capital and neglecting the high risk that rescuing Fannie and Freddie involved at the time.
Let's start with the opportunity cost of bailout funds. The Treasury has so far spent around $190 billion to keep Fannie and Freddie in the black. It has collected around $300 billion in dividends from the two GSEs. That gives a cash accounting profit of $110 billion, for a cumulative rate of return of 57 percent, or 5.7 percent each year that the GSEs have been in conservatorship.
Impressed? Don't be. Since September 1, 2008, five days before the Fannie-Freddie bailout, the aggregate stock price of America's largest listed companies – as captured by the S&P 500 index – has increased by 146 percent. That figure understates the returns to shareholders, since many S&P 500 firms hand out a share of their annual profits in dividends, which are not captured by the stock price.
Had the Treasury invested the GSE bailout money in an index tracker fund in September 2008, it would have $437 billion today. Better still, had it not used taxpayer money at all to rescue Fannie and Freddie, then those resources could have found a better use elsewhere, including as private investment into American enterprise. It is neither the job nor a special talent of government to allocate capital in the economy.
But even an analysis that takes into account opportunity cost is misleading, because it fails to consider the true risk of the GSE bailout. As MIT's Deborah Lucas pointed out in a recent paper:
The press typically reports bailout costs on an ex post cash basis despite the problems with that approach . . . In their most recent update dated September 27, 2018, [ProPublica] report[ed] a total net government "profit" of $97 billion . . . in 2012 former president Barack Obama claimed that, "We got back every dime used to rescue the banks."
Yet, as Lucas explains, ex post cash accounting "fails to recognize that at the point in time when assistance is committed there is no assurance about how much will be recovered." All investments are risky, but bailouts are particularly risky because they typically happen under conditions of great financial stress and uncertainty.
When the Treasury took over Fannie and Freddie, it signed a big check with only soft limits on how much capital it might disburse to keep the GSEs in the black. Indeed, the government's commitment of taxpayer funds increased in the years after the bailout. On the other hand, even Treasury officials had little concrete idea of the scale of bad loans they might have to write off, and of when (if at all) the two mortgage giants would return to profitability. If buying GSE stock really was such a one-way bet, why were private stockholders scrambling to sell off their stakes in 2008?
Bearing in mind the great uncertainty surrounding the government's true exposure and future inflows at the time of the bailout, Lucas uses fair-value accounting methods from the Congressional Budget Office to come up with a more realistic estimate of the net value of the bailout guarantee: minus $311 billion. Even attempts to estimate this value using the prosperous time before the crash as the starting point find a net cost to the Treasury: minus $8 billion using the methods that Lucas reports as most reliable.
Either of these approaches are justifiable. But no economist worth their salt would be happy with the assertion that, because in hindsight the inflows from the GSEs have exceeded the Treasury's outlays, the bailout was a good idea. It would be like suggesting that playing the lottery is (statistically) advisable just because you won the prize last time. Furthermore, no-one should forget that, because Fannie and Freddie guarantee 44 percent of single-family mortgages and have ten cents of capital for every $100 of assets, the U.S. taxpayer remains very much exposed to potential losses.
The Fannie-Freddie bailout of 2008 was a costly undertaking, ushering in 11 years of uninterrupted taxpayer support of home lending, and undermining financial stability. No amount of post hoc rationalization can change that. But a more competitive guarantor market that radically shrinks Fannie and Freddie's role in housing finance, and with it the U.S. taxpayer's exposure, might prevent another costly bailout in the future.