Were the GSEs Out-of-Control?

I must admit, I've been a little puzzled by some of the reactions to the GSE bailout. Yes, the taxpayers will have to pay for the GSE bailout. No, it's not fair. But some people seem to think that the GSEs were out-of-control, mortgage-slinging cowboys whose gross excesses we're now picking up the tab for.

For example, Megan McArdle writes:

Because they are government sponsored, the government let them get away with practices that would never fly in the private market.

And Professor Stephen Bainbridge also writes:

The implicit government guarantee—which now has become explicit—thus has had serious adverse consequences. It has encouraged Fannie and Freddie management to take risks that firms subject to market competition and lacking a government safety net would never take. In particular, Freddie and Fannie have increasingly held on to mortgages rather than reselling them as asset backed securities, which exposes them to much higher repayment and other risks.

Professor Bainbridge seems to think that "firms subject to market competition" would never do something as risky as keeping mortgage-backed securities on their balance sheets. This is, of course, profoundly stupid (though not at all surprising, as law professors tend to be the worst breed of market-worshipping amateur economists). All the financial houses kept MBS on their balance sheets. Bear Stearns, anyone? The idea that "market discipline" would have kept the GSEs in line is truly laughable. The market provided absolutely no discipline during the housing boom—why do you think we're in this mess to begin with?

Both McArdle and Professor Bainbridge seem to think that the GSEs were out-of-control, and taking wild risks. The GSEs weren't buying or securitizing subprime mortgages though. In fact, the only thing that kept the GSEs from running full speed into the subprime market was government regulation—specifically, the conforming loan limits in their charters.

I think the confusion stems from a misunderstanding of the externality that the GSEs' implicit government guarantee creates. The implicit government guarantee creates an externality because the public bears some of the cost of the GSEs' exposure to interest rate risk. Any mortgage portfolio is subject to interest rate risk. Investors use derivatives such as interest rate swaps to hedge their interest rate risk. If an investor has too much unhedged interest rate risk, creditors will demand that the investor pay a higher interest rate to compensate the creditors for their interest rate exposure.

Like the big financial houses, the GSEs use an imperfect dynamic hedging stategy, which essentially hedges short-term interest rate risk but leaves long-term interest unhedged (it's a lot more complicated in reality, but that's the idea). Creditors should demand a higher interest rate as compensation for their exposure to the unhedged long-term interest rate risk, but because the government implicitly guarantees the GSEs' debt, creditors lend money to the GSEs at artificially low rates. Of course, someone has to bear the cost of the GSEs' exposure to unhedged interest rate risk, and that someone is the U.S. taxpayer.

That's (more or less) how the GSEs' implicit government guarantee creates an externality, not by inducing the GSEs to take wild risks and act like cowboys. The GSEs only take on "excessive risk" in the sense that they take on more risk than is justified by their low borrowing costs. Before the collapse of the mortgage market, the GSEs were actually taking on less risk than other banks—it was still too much risk relative to their capital base and the low borrowing costs, but it was low risk nonetheless.