Edward L. Glaeser

What’s big, risky, and losing billions?

Mortgage giants Fannie Mae and Freddie Mac need to go away — but slowly

May 20, 2010

THIS WEEK, the Senate rejected a $400 billion cap on the taxpayer bailout of the Federal Home Loan Mortgage Corporation and the Federal National Mortgage Association, better known as Freddie Mac and Fannie Mae. The decision may ensure that the two firms’ collapse will be the most costly event of the economic downturn. Their old model — private companies with an implicit public guarantee — created behemoths that gambled trillions of dollars and lost billions in taxpayers’ money.

This old system was unacceptable, but what shall we put in its place?

Let’s start with the basic question: Why should Fannie and Freddie exist at all? Supporters made two standard arguments. The simpler one is that Freddie and Fannie increase homeownership by making borrowing cheaper. It is surely true that if the federal government guarantees mortgages against default without charging enough for that guarantee — which is what happened — then this underpriced guarantee is yet another implicit subsidy for home borrowing and homeownership.

But a policy isn’t wise just because it promotes homeownership. There are more transparent ways of doing so, such as a direct annual homeownership tax credit. Besides, how much promotion does homeownership need? The home mortgage-interest deduction already encourages people to overborrow to bet on the housing market.

The more complicated argument for Fannie and Freddie is that, by guaranteeing mortgages and packaging them into securities, the two firms make a secondary mortgage market possible. The secondary market is important because it allows banks that write mortgages to share their risk with investors. Indeed, the current crisis might have been worse, not better, had banks been holding all of the nation’s mortgages in their own portfolios. Americans are lucky that their mortgage losses were borne by investors worldwide — not just by banks that were ultimately backstopped by the US government.

Defenders of Freddie and Fannie have argued that, without the two firms as intermediaries, banks would keep their good mortgages in their own portfolios and sell only the worst loans to secondary investors. Investors would wise up and avoid the securities, and the secondary market would break down — making new mortgages more expensive.

But is this argument right? In fact, there are several securitized debt markets that exist without any federal guarantee. Securities backed by credit-card debt and corporate obligations, for instance, trade in a global market. Even jumbo mortgages were securitized without help from Fannie and Freddie. If anything, there was too much of a market for subprime securities without a federal guarantee.

But despite my skepticism about the mission of Fannie and Freddie, we won’t and shouldn’t get rid of them overnight. They still hold vast portfolios. The housing market is still fragile. We should reform them and gradually reduce their role in the markets, moving slowly to avoid panic and dislocation.

One option is to reconfigure Freddie and Fannie as a purely public agency — one that is slow and risk-averse. For-profit entities are good at taking risks and making money, neither of which is wise if the public is ultimately paying the bill. The reformed agency would be forbidden to innovate or to hold loans in their own portfolios. Its job would be to guarantee, bundle, and securitize standard, moderate-sized mortgages to borrowers with good credit scores and sizable down payments.

The new entity’s primary obligation would be to avoid costing taxpayers — not to promote homeownership or any other ill-defined social aims. The entity should be isolated from legislators pushing broader objectives, and should charge fees hefty enough to cover its own costs under almost any circumstances.

Some governmental entities err by doing too much — as Fannie and Freddie did before the fall — while others are terrified of overreaching and do very little. This mortgage-selling entity should be a model of conservative inaction.

If the new entity’s fees are sufficiently high, and its due diligence is sufficiently onerous, then banks and borrowers will bypass it altogether. That’s the dream. We’ve seen what happens when government-guaranteed entities aggressively dominate the mortgage market. Let’s see what happens when the government gradually prices itself out of business.

Edward L. Glaeser, a professor of economics at Harvard, is director of the Rappaport Institute for Greater Boston.