Can We Prevent Future Crises?

Between 1837 and the Great Depression, there were nine financial panics, occurring not just regularly but according to the same script. “We had big bank runs with people withdrawing from their checking accounts over and over again,” says Gary Gorton, Frederick Frank Class of 1954 Professor of Finance. It wasn’t until the creation of deposit insurance that these types of crises finally came to an end.

After the 1930s, there was a 50-year “quiet period,” in which the U.S. financial system proved remarkably stable. The reason, says Gorton, was a web of regulations that discouraged banks from taking excessive risks.

Now, Gorton says, we’ve had the first of a new type of financial crisis, sparked by the largely unregulated shadow banking system. Gorton, one of the foremost experts on the 2008 financial crisis, says that the likelihood of another panic is increased by a general lack of understanding of newer—and evolving—debt models such as mortgage-backed securities, plus Washington’s inability to make the kinds of changes that would be effective in mitigating risk. “There’s this paradox,” he said, “that once one of these things happens, the populist anger towards bankers prevents any intelligent discussion. And by the time things calm down, nobody cares about it any more.”

Gorton has proposed a new type of test for financial institutions modeled on the stress tests run in the first half of 2009. The new tests would be engineered to assess how well banks are handling risk, as well as how stable they’d be in the face of a number of hypothetical scenarios. “The stress tests were one of the really important things to come out of the crisis,” he said, “and this is a way of taking that further and building on what we actually did.”

TRANSCRIPT

Q: Can regulators keep up with changes in the financial system?

Gary Gorton: From 1934 till about the mid-’80s was a very quiet period. And banks in the United States had basically—they were little monopolies. There was no branching across state lines. Other institutions couldn’t offer checking accounts. So they had an incentive to internalize risk management and not do anything to jeopardize being a bank. In the mid-’80s, following the high inflation of the late ’70s, money market mutual funds appeared and took huge amounts of business away from them; the junk bonds appeared, took huge amounts of business away from them. And so the system began to evolve, and part of their response was securitization. So, as a regulator, you can’t be in the business of being in an arms race with the financial system because you’ll lose. I think it’s better to try to design a system that is such that the innovation is out in the open, and banks want to show it to you. And that’s a system which is difficult to design. You want to give banks incentives to be banks. So, it’s not a popular thing to say, but they’re going to have to have little bits of monopoly power.

A good example is Canada. Canada hasn’t had a financial crisis in—I don’t know—150 years. And Canada has a small number of very large banks. You can imagine that happening here—people would go nuts. But they’re in a kind of informal club with the central bank, and the implicit agreement is you behave, and you can be in the club. It’s very valuable to be in the club. You misbehave; you’re out of the club. So again, it’s a system where entry is not so easy. And so, you want to not lose your franchise of being a bank. And that’s a system we know works, but it’s not a popular thing to say, “We need these firms that have monopoly power.”

Q: Five years after the crisis, where do you see the greatest risk for the next one?

Gorton: If you think about U.S. history in 1837, 1857, 1873, 1884, 1890, 1893, 1907, 1914, and the Great Depression, it was the same thing happening over and over again. We had big bank runs on people withdrawing from their checking accounts over and over and over again. And until you got deposit insurance, there was just sort of no end in sight. So, now we’ve had the first one with this other kind of bank debt. We haven’t really solved that. So I expect to see it over and over and over again until, one way or another, we hit upon a solution. This is the history of all market economies. The arrangements that prevented this from happening again—were not—they were accidental almost. So we’re not at the point where you can say, “Look, here’s the problem, and here’s the solution.” Part of it is this paradox that, once one of these things happens, the populist anger towards bankers prevents any intelligent discussion, and by the time things calm down, nobody cares about it anymore. We’re on to the next issue, whatever it is.

Q: You’ve proposed a new way to test the stability of the financial system. How would it work?

Gorton: The best way to think of it is with a bank stress test that the Fed did on banks. So we would propose asking banks on a regular basis about 200 questions of the following sort: “Do these computations, and tell me the answers: If house prices went down by 5%, 10%, 15%, 20%, what’s the dollar change in the value of your firm, including derivatives.” You do this for a wide variety of prices, exchange rates, interest rates; and you can also ask about: “If your largest three counterparties failed, what would happen?” So you get these sensitivities to various scenarios that you propose. You don’t ask them, “What would you do if…” It’s not, “What would you do?” It’s, “What would happen right now if this happened?” So this we would propose be done on a regular basis, and this could be aggregated, and this would allow you to answer these kinds of questions: Is risk building up? Is it being dispersed? Is it being concentrated and so on? Now it would also be a way for economists to kind of think about the world.

Macroeconomics is driven completely by national income accounting. If you had these risk measurements, we’d start to think in terms of risk. So, this is basically something which is doable. I spent a long time at several large institutions—Goldman Sachs, Citibank, for example—and asking them, “Could you compute these numbers?” They’re very different. They’re different kinds of firms. They’re different global reaches, and so on. But we know from the stress test that it’s possible to get started doing this. So, the stress test was one of the really important things that came out of the crisis, and this is a way of taking that further and building on what we actually did.

Q: How did economics contribute to the crisis? Can these shortcomings be alleviated?

Gorton: Economics is a subject which depends very heavily on abstraction, which is important. Reality is very complicated, so you need to abstract from reality in models; that’s what models are for. So, models are sort of accounting systems for ideas, so to speak. I think part of the problem is that, since we didn’t experience a financial crisis since the Great Depression, there was this implicit view that one was just never going to happen. And if it wasn’t going to happen, there was no reason to look back at history, because that was irrelevant; and there was no reason to really study institutions, because you didn’t really need to bother with that. And abstraction became abstraction, not from reality, but from other models. So a lot of economic theory is—it’s like modern art. It’s kind of a conversation with itself, and they kind of lost sight of reality.

So, for example, graduate programs in economics don’t require economic history anymore. A lot of top programs don’t even have economic historians. And most business schools have faculties that don’t really understand all the institutions. They couldn’t really teach about it even if they wanted to, because they just don’t know about it. So there’s a lot…linking to all this reality—historical and institutions currently—is something that economists are going to have to do.