Saturday, September 26, 2009

Toward a New Philosophy of Regulation

In the early stages of the financial crisis, I recall having a conversation with a Citigroup banker. I was waxing indignant about the failure of regulation leading up to the collapse. Her shrugging response: It's a losing game for regulators anyway; Wall Street will always stay a step ahead of them.

It's not that I completely disagree. Regulation is difficult. Take for instance the I.Q.-and-experience mismatch across the public-private divide. Those smarter graduates with MBAs and law degrees get hired by Wall Street for big bucks; their more-mediocre classmates wander off to work for the U.S. government for half the pay. Then, once the latter spend a while at the SEC or CTFC they naturally seek to cash in on their resume at some point -- after all, their experience, plus their insight into how a regulatory agency works, are valuable to Wall Street. They jump the fence and serve those they once regulated. So, in the end, the government doesn't even get the benefit of experienced mediocre talent.

Does that mean we should give up on regulation? Or regulate with the cynical conviction that our efforts don't really matter much, because Wall Street is just so politically well connected, so financially powerful, so much better endowed in both quality and quantity of people, that we don't stand a chance? This, to me, seems absurd. Why can't we instead challenge ourselves to do better?

Certainly, U.S. regulators often do appear hopelessly outmanned and outgunned. But they have one powerful advantage, too often overlooked: In a nation of laws, they define the rules for the playing field. In fact, they can define the playing field itself.

So what do we do?

Here are some philosophical thoughts toward smarter regulation. They are broadly prescriptive.

1. Move toward a more principle-based system of regulation.

One problem right now is that we lay out rules, in staggering detail, and anything not prohibited is generally assumed to be legal. That invites the creation of a loophole-seeking culture in the financial system. But why do we have to spell out everything? We don't always do this elsewhere in our legal system. The standard for determining guilt in a criminal case happens to be very subjective: "beyond a reasonable doubt." And what of the legal definition of pornography? It cites what an "average person" would find obscene applying "contemporary community standards."

Why can't we have more broad principles in financial regulation that say for instance, "If, when doing x, you do not follow basic concepts of generally acknowledged sound accounting (or some such), you can be prosecuted even if what you do is not expressly prohibited?"

Okay, I know, possible objections:

Accounting is complex; there are various methods of accounting. But still, there are basic principles that should be adhered to. Why can't these be enforced broadly? For instance, the treatment of special purpose entities in this crisis made absolutely no sense to me. How can a bank not hold regular reserves against an SPE's assets yet immediately pull the same assets onto its books as soon as the special vehicle gets into trouble? This shouldn't be allowed; basic common sense should tell us that. You shouldn't need a specific rule prohibiting it.

Principle-based accounting will create a lot of uncertainty, which will stifle innovation. Well, it's not clear that financial innovation has done us much good in the first place. The truth is, we probably don't need as much of this vaunted "innovation" as we now have. Still, it's worth noting that a more principle-based approach doesn't preclude being able to spell out certain rules, especially as they apply to simple, necessary functions of the banking system. Once you start innovating outside of that sphere, then you would be increasingly on your own.

2. Add criminal penalties. Subtract certain regulations.

Basel II, and regulatory regimes like it, invite a lot of mind-bending game playing for financial institutions. Under Basel's international standards, banks must set aside a certain percentage of reserves for assets of class x, but a different percentage for class y, and so on. So if you're a bank, you obviously have an incentive to transform your assets into favorable classes that require less in reserves.

But, as Mike points out over at Rortybomb, just because the regulators say you must have, say 8 percent capital on hand, doesn't mean you have to go down to 8.0001 percent. You can set aside 12 percent. Or 16 percent. You should be making that calculation based on the market, your collection of assets, and how much you think you'll need. Unfortunately, when that 8 percent number becomes enshrined by regulators, it then becomes a floor to evade your way around. So the banks try to get to 8, then to gain a competitive edge, they look for a way to secretly go down to 6 percent, and make more profits off the higher leverage.

Ironically, here you can argue that certain types of regulation make things worse (and indeed, some people are looking at that now: see Mike's full post above). What would be a way to solve this problem? Perhaps we could loosen and simplify certain capital regulations, but then put into place a hammer on the backend to make bank executives more subject to criminal charges (and subject to personal bankruptcies) when they gamble unwisely and lose.

After all, remember one significant thing about the landscape before all the investment banks went public by selling stock. Under the old partnership model, before they got to play with other people's money, they were a lot more careful. The partners had to eat the losses. So they were careful to stay on top of the risk.

3. Seek to create "natural brakes."

A "natural brake" is something that tends to slow you down, in a simple and automatic manner, without messy intervention. I would compare it to shifting into first gear when rolling downhill in a car. The lower gear automatically helps to slow the vehicle.

What are other natural brakes?

Housing bubble brake: Some mechanism that kicks in when average house prices climb high enough relative to average rents/incomes. This sort of brake means you don't have to try to judge that tricky, right moment to intervene on a bubble. Not only is that judgment hard to make, but it's complicated by political considerations and the fact that you're basically removing the punch bowl while the party is rockin' and rollin'. Not a popular move.

How might this brake work: When house prices reach a certain multiple of rents, then homebuyers have to put down x percent more in a deposit to buy a home (this isn't an original idea; others have proposed it). This acts as a nice natural brake because, when homebuyers are forced to increase their downpayment, more will shift toward renting, and the price of rents will nudge up, and the price of homes nudge down. A more-normal equilibrium will naturally be reached.

Short-selling is a natural brake for the stock market. It's often demonized during a down market, but the shorts take the air out of bubbles. They exert downward pressure on share prices. They keep owners of stock more honest and help protect them from getting burned in momentum-driven rallies.

What are natural brakes that could be employed by bank regulators? I'm sure there are some good ones out there. For home prices, as I noted above, we could start by creating a linkage between the market costs of buying vs. renting, using each to keep the other in a sensible range.