Arnold Kling  

What's Different About Banks?

The Weiner Affair... Market Discipline vs. Regulati...

I've been at a seminar discussing "too big to fail," so bank regulation issues are top of mind.

I think that the most important thing that differentiates banks or bank-like firms from other firms is that it is easy for short-term results to be misleading. That is because banks borrow short and lend long.

When you invest your money with a financial institution, you do not know exactly what you are getting. You do not evaluate the assets of the institution, you do not know if it is well run, etc.

There are plenty of other products where we face this problem. Computers, cars, cell phones, and so on. But with those products, all consumers get quick feedback. If a computer manufacturer or retailer sells defective products, word will get around pretty soon. Reputation matters, and because word gets out quickly if the firm misbehaves, reputations are reliable.

In banking, you can get along for a while with a really defective strategy. There are many ways to write out-of-the-money options that pay off a little bit most of the time and every once in a while suffer catastrophic losses. You can maintain a good reputation for a long time, even though you do not deserve it.

What this means is that consumer-to-consumer feedback is not sufficient to police the banking industry. Instead, specialized monitors are needed to check up on what banks are doing. These could be private monitors, or they could be public monitors.

But who watches the watchers? If we had a system of private deposit insurance, or some such, how can we know that this system is being well managed? If we look at the securities rating agencies as models of how a private monitoring system works, then I think there is much to worry about.

Putting government in charge does not solve the problem. As we know, during the housing bubble, Congress and regulators were putting pressure on lenders to make more loans and to use lower credit standards. Again, this does not inspire confidence.

Comments and Sharing

COMMENTS (8 to date)
Mike writes:


Between banks and between regulatory regimes (public and private).

It'll never happen.

kebko writes:

I don't get the comparison of private deposit insurance and ratings agencies. One has money on the line; the other doesn't.
I don't see any reason to think the private deposit insurers wouldn't come up with decent monitoring protocols and methods for reducing risk (convertible bonds, etc.). Not that it's ever going to happen.

ChrisA writes:

I don't know the answer either. I bet though, that if banks were allowed to fail instead of being rescued, then a solution would be found by the market. Perhaps the solution would be that there would be no banks, or only very narrow banking with all funding (mortgages included) provided as equity. Or maybe the market would be able to find some way to properly evaluate a bank, there would be plenty of incentive for a bank in a system where bank failures were frequent, to find a way to advertise its prudence in an unfakable way. But we will never get there with our current system of government supported bailout.

Since a bail out system appears inevitable, why not make it a feature and require banks to pay bal out insurance to the Government. At least the healthy banks would then have some incentive to point out dangerous behaviors by competitors.

E. Barandiaran writes:

Just to make clear: the relevant theory to the problem you identify is the theory of incentives (also known as the theory of contracts between a principal and an agent). Although one could argue that it's a problem of credence goods, the critical issue is the incentives an agent has to perform as the principal expects.

Indeed the most challenging problem of incentives is democratic government, much more than banking. As lefties have been learning, Obama is the latest pin-up "political Madoff" (sorry David Henderson, it's not Anthony Weiner).

Les writes:

Arnold states the problem very well.

But private (not public) deposit insurance would solve the problem, because the private insurers would have skin in the game, since they are risking their own money - not taxpayer money.

Therefore rates charged for private deposit insurance would cover the risks involved. Those private deposit insurers charging rates that are too low would be driven out of business. Those private deposit insurers charging rates that are too high would attract competitors charging lower rates.

steve writes:

"Congress and regulators were putting pressure on lenders to make more loans and to use lower credit standards. Again, this does not inspire confidence."

How much pressure do you need to put on banks when they are making billions? Why wouldn't profits that were large enough to account for about 40% of the growth in our economy be sufficient motivation?


Lord writes:

It likely solves the problem more than you imagine. It solves it each time things do not go wrong and when they go wrong but only until caught and corrected. It solves it when people take responsibility and accept consequences. It does not and cannot when people do not want it solved and are unwilling to face failure. The solution is the ballot box but it rarely works in advance, and often not even in retrospect when the guilty try bamboozling the ignorant and shifting blame from themselves.

Jonathan Bechtel writes:

A couple thoughts:

A). You seem to be re-stating the gist of Tyler's message in the "Inequality That Matters" and from your post I'd imply you agree with him.

If I recall correctly, you disagreed with him when he first wrote the article. What changed?

B). I think you're right. I also think democratic processes behave sub-optimally for the same reason. The two are uniquely similar in that way.

C). To me the best policy solution to this problem would be to get rid of publicly subsidized deposit insurance.

I worked as a consumer banker for three years, and I came to appreciate the behavioral psychology on loss-aversion. It's really frickin' true. The guarantee means everything.

But in spite of that, people really could care less about what their bank is doing behind closed doors. They don't match the right hand with the left.

I think it's politically impossible, but it's the only thing I can think of to resolve the issue.

Of course, if you live in a Gary Gorton world, the decreased transaction friction from informationally-insensitive deposits outweighs the incentive problem and you should leave it in place anyway.

Comments for this entry have been closed
Return to top