The Fallacy of Banks Needing 'Skin in the Game' for Mortgages

By now, anyone who follows financial news has heard the theory that banks originated so many bad mortgages during the housing bubble, in part, because they had no skin in the game. Since they could securitize the loans -- that is, bundle them up and sell them to investors as bonds -- there was no reason why the banks should care whether the loans would go bad or not. So the solution sought by last summer's financial regulation bill was to require banks to keep some of portion of the mortgages they originate. Then, the theory goes, they'll be more careful, since they'll loose if those loans are bad too. Unfortunately, this seemingly simple idea could cause more harm than good.

At a House Committee on Oversight & Government Reform subcommittee hearing on Wednesday afternoon, the new risk retention requirements were addressed by a panel of expert witnesses. A few believed the risk retention requirements would help for the basic reason already explained, and a few thought they would hurt. One witness, financial industry analyst and consultant Joshua Rosner, provided a particularly perspicuous explanation (.pdf) of why more skin in the game won't just turn out to be ineffective, but will actually be harmful:

On the surface, this appears to make sense. If a lender or securitizer knows he will have to drink the poison in the chalice he offers to others, then he would be more careful. If, however, because of his belief in his modeling prowess, his own systemic importance, or his financial strength, the pourer believes that he has an enhanced immunity to poison or that he will be first to receive an antidote, then perhaps he may ignore the disincentives to poison others. More likely, as we saw in the past crisis, the same firms that poisoned their investing customers failed to recognize the power of the poison. After all, the banks that were in most dire need for direct government support were so precisely because they had ingested large quantities of the poison they had sold to others. Often, as we witnessed with Bear Stearns and Merrill Lynch, these firms didn't have the operational controls, available information or resulting ability to fully model their exposures. To force them to increase concentrations of these held securities will only increase their risks.

The reality is that the banks who issued bad mortgages actually had very, very large amounts of skin in the game. That, in fact, is what triggered the financial crisis. When that skin proved toxic, it nearly killed them all. The problem wasn't that they didn't care about issuing bad loans: it was that they didn't realize that the loans were bad. They really believed that the housing market would continue to go up and that most borrowers would be able to refinance their way out of trouble.

Instead, Rosner suggests enhanced transparency as the solution. He said that if investors better understood what they were buying, then they would not have bought so much garbage. They often didn't perform detailed analysis, but instead relied on the rating agencies to say whether the assets were safe or not.

Rosner isn't a mouthpiece for the big banks. Later in his testimony, he lamented the fact that the summer's financial regulation bill didn't crack down on "too big to fail" institutions by essentially forcing them to either shrink themselves or ramp up their risk management in such a way that their systemic risk disappeared.

The reality is quite simple. If there was no market for bad loans, then banks would not originate them. If investors have the data and time they need to analyze the loans that banks securitize, then they will not buy a deal that will turn toxic. Thus, there would be no market for the bad loans, and banks would not originate them. Skin in the game is not necessary to avoid bad loans, just better transparency for investors.