Arnold Kling  

Bank Regulation is a Political Economy Problem

More on the Late Paul Samuelso... Book 1 on Reason TV...

Paul Volcker says financial innovation has not been a good thing.

Now, I have no doubts that it moves around the rents in the financial system, but not only this, as it seems to have vastly increased them.

You cannot get financial regulation right unless you think of it in terms of political economy. Instead, if you think in terms of "how can we optimally regulate so that banks have only incentives to do great things and no incentives to take excess risk," you are on the wrong track.

The political economy challenge is to constrain the regulators. The problem is, "how can you design a system in which regulators do not extend government guarantees in a way that leads to abuse?"

A short answer might be, "Get rid of government guarantees." But I think we're too far down the road with deposit insurance to take it away.

Given that we have deposit insurance, the key is to limit its scope. You cannot do that unless banks are small. Big banks have a credible threat that says, "If you let me fail, the whole system is going down." Policy makers do not have a credible counter-threat. The only hope is to break up the big banks before a crisis hits. As I've said before, it is absurd to suppose that you will exercise "resolution authority" to close a big bank during a crisis and at the same time say that it is unthinkable to break up big banks now.

Comments and Sharing

COMMENTS (9 to date)
Paludicola writes:

I imagine that it also helps to think of financial regulation as hopeless in the long run.

Mencius Moldbug writes:

Closing out deposit insurance is trivial. Deposit insurance is that common, yet nefarious, instrument called a "loan guarantee." These funky contingent liabilities can be eliminated with a trivial zero-sum balance-sheet transformation.

In a loan guarantee, C (the Fed) guarantees the loan of A (the depositor) to B (the bank). Ie, if B defaults, C pays A.

To eliminate C's contingent liability, replace this structure with a loan from A to C and a matching loan from C to B.

Thus, in the case of banking, the individual has a deposit "account" at the Fed, just as only banks have today. The Fed in turn loans money to the banks, as is its wont. If B fails, C takes the hit, just as it does today.

This restructuring of loan guarantees into simple loans preserves the risk structure perfectly, and makes the real financial arrangement much more clear. Of course, no one really wants the latter...

Tom West writes:

I think way too much emphasis is placed upon moral hazard. Is the consensus really that much of the crisis would have been avoided if there weren't implicit and explicit gov't guarantees?

Because from what I could see, the possibility of a real crisis was essentially discounted by just about everyone that mattered until it was too late. Gov't guarantees didn't come into it until it was already too late.

However, without the bank system rescue, there'd likely be a big hole where most of the Western economies now stand (okay, stagger).

The sums that are being gambled with are essentially so large, that it doesn't matter what happens to the direct losses, the consequences of those losses are enough to devastate the economy.

Honestly, if people knew the gov't wasn't going to intervene, they might well gamble more. After all, if the big boys fall, you'll go with them no matter how "safe" your investments, so you might as well try to make hay while the sun shines.

(Or to use another analogy, when your next door neighbour is gambling with high explosives, it doesn't particularly matter whether the government will rebuild his house if he makes a mistake, the whole block will be a crater.)

david writes:

Nationalizing the banks without compensating its shareholders would be a pretty good policymaker counter-threat. Policymakers could then auction the bank off immediately after recapitalizing it.

Note the similarity to bankruptcy theory.

The lack of existing counter-threats is only due to an unwillingness to consider direct applications of state power (the N-word, as they say); the end result is the use of elaborate excuses that aggravate liberals and libertarians alike, and rent-seekers cheerfully exploiting the consequent loopholes in regulation.

Prakhar Goel writes:

Building on MM's post:

Getting rid of deposit insurance is indeed simple. Here's another way which gets rid of the retarded private-profits-socialized-risk structure that deposit insurance currently has:

Halve the guaranteed amount and double the fees. Add a subordinate debt requirement. Do the same for the next ten years and publish this plan before hand. By the time we hit year five (or probably even earlier), the banks would have moved off.

This is (as MM implied) not a problem of practicality. This is a problem of political economy (as Arnold states with much insight). Right now deposit insurance benefits three very powerful groups. It benefits populist politicians because deposit insurance can be easily made to look like a good way of helping out the "little guy." It benefits large money holders because they can farm out their deposits and effectively have infinite deposit insurance -- private profit with socialized risks. It benefits large deposit holding financial institutions because it frees them from the discipline that naturally arises out of competing on the dimension on risk.

All this and more has been better stated by Calomiris on an earlier interview here on econlog.

wm13 writes:

"Too big to fail" is not the issue. Look at the money market funds: they aren't that big, they don't pay high salaries (by Wall Street standards), they don't promise mega-returns to either sponsors or investors, and all their capital is in the form of equity, yet the government was forced to bail them out.

Incidentally, I thought the person who responded to Paul Volcker was correct. The problems we have seen have come from home mortgage securitization, repo financing, and commercial paper, not from CDS or other more recent innovations. So you can argue about whether financial innovation is valuable--you could ask the same question about Windows 7--but it isn't the source of our current problems.

mark writes:

There is a sound alternative to breaking up large institutions. It is to require such institutions to replace short term institutional size debt with a subordinated long term debt that their regulator can convert to equity if the equity value drops below X% of the debt. The market will price that security and price the equity and police the problem itself. Cap the amount of short term debt at X% of the outstanding debt. And require every such institution to insert in every derivative contract a right to extend payment over multiple years.

Mike Sproul writes:

If you must have deposit insurance, why not limit it to something like $20,000 per depositor? People who have more than that can invest it in stocks, trust deeds, etc., and take their chances like adults.

CJ Smith writes:


How do you reconcile a libertarian's defense of the right to control their assets free from government intervention with the statement, "The only hope is to break up the big banks before a crisis hits(?)" Why wouldn't that be not only governmental interference, but a non-compensated taking? This comment is also directed at Mencius (who would make the Fed a super-bank), Prakhar (who merely tweaks the existing scheme and assumes that increased costs aren't passed along to depositors along with the reduced coverage), Mark (who would require government regulations to compel unsecured subordinate debt obligations that convert to equity in a failed or failing business), and Mark (who also merely tweaks the numbers).

Those who advocate termination of FDIC:

How would elimination of FDIC lead to less risky banking practices? By reducing the amount of the banks equity not allowed to be put at risk? By removing protection from banking customers who simply want to deposit demand funds without doing complex financial due diligence on their bank? By removing any form of security for depositors' funds, which are a either completely unsecured debt on the part of the banks or subordinated to secured debt well in excess of the banks' equity?

Banks that take risks and fail right now don't have any responsibility to replace customer's deposits - any amount not covered by FDIC is 100% at risk. Removing FDIC would just expose the depositors to greater risks of a total loss. Presumably depositors could self-insure, but isn't that what they are doing now in a round about way? Depositors would now have to be insurance gurus in addition to being financial analysts on both the bank and their insurer.

FDIC is not designed to protect large investors, it's designed to protect mom and pop shops that want convenient safe access to demand funds, not an investment vehicle. The fact that its abused by large investors doesn't make the program invalid.

We can critize FDIC as an abused governmental program - it is. But I have yet to hear a counter-proposal that would protect the mom and pop shop owner or the individual demand depositor without equivalent governmental intervention of some sort, or the imposition of financial risk and extraordinary transaction costs on these consumers. In my opinion, that's the "political economics" of the siutation.

Comments for this entry have been closed
Return to top