Wallison argues that it was we will never be able to eliminate moral hazard from banking simply because the banks are backed by the government.

How can we discourage banks from taking excessive risks? (Scott Sumner of Money Illusion)

Peter Wallison: Other than through regulation, it's almost impossible to prevent banks from taking excessive risks.  But there is one thing we could probably do that would help and that is to make sure that banks disclose more information to the market than they do today. 

Right now, banks disclose a lot of information to their regulators.  They disclose some information through their financial statements, but none of that gives creditors of banks enough information to understand whether the banks are really taking risks or not.  So, we are relying in almost all circumstances on the regulators to prevent this risking and as it's turned out, I think it's fairly clear that the regulators are not very good at understanding the kinds of risks that the banks are taking.  So, we have to rely on creditors and to rely on creditors, we should be giving them a lot more information.

But we will never be able to eliminate moral hazard from banking simply because the banks are backed by the government. 

Should there be a minimum downpayment requirement on mortgatges? (Scott Sumner, Money Illusion)

Peter Wallison: I think that's a good idea.  That used to be the case.  We required minimum downpayments of 20%, then went to 10%, and then it went down to as low as no percent and that's one of the reasons why we have so many problems today. 

A 20% downpayment would be a good way to start and that's the way it used to work.  It turns out that the downpayment is the key to determining how sound a mortgage will be.  The lowest downpayments result in the highest defaults, the highest downpayments are the ones that maintain solid prime mortgages better than any other system. 

What happens if regulators are still unable to unwind non-banks? (Dan Indiviglio, The Atlantic Business Channel.)

Peter Wallison: You know, the different between a bailout and unwinding is a little hard to understand in some cases.  If we're talking simply about the question whether an institution can be liquidated, and I think that's what most people mean by "unwinding."  It's very hard to imagine that that can't be done by some organization, like the FDIC, or any other kind of organization. 

Bailouts, however, are done not because an institution has failed, but because it's important to keep the institution going, and that has much more to do with the size and importance of the institution and what its effect will have on the rest of the market if it's allowed to fail. 

I don't know exactly how we can say an institution can't be unwound, it can always be closed down.  The question is whether we want it to be closed down, and in my view, very large commercial banks can create the kinds of, we'll call it a systemic breakdown, if they are not bailed out.  Other kinds of financial institutions do not raise those questions for a variety of reasons, but non-bank financial – I’m sorry, bank financial institutions, that is, those that are commercial banks backed by the government could under some circumstances create serious systemic breakdowns and have to be bailed out.  What we have to do, I think more than anything else, is through regulation and through the kinds of transparency that I talked about before, try to make sure that the market understands the risks that these institutions are taking so that they are deprived of the funds that they can take risks with because the market is well aware of how those funds are being used.  If we don't do that, if we are relying entirely on the regulators to prevent these firms from taking excessive risks, well, I'm afraid we will have to engage in bailouts because we must keep some of these very large institutions alive under some circumstances. 

How should capital ratios be determined? (Dan Indiviglio, The Atlantic Business Channel)

Peter Wallison: I'm afraid I have to go back again to talk about transparency here because it is exceedingly difficult to imagine any kind of formula that actually captures the risks that banks are taking.  Under Basel 1, we had a formulat with a number of buckets and we cast all commercial loans into one bucket and all mortgages into another bucket, but in fact, they're not all the same.  They do not all involved the same kind of risk.  So we have to have a more refined system. 

Under Basel 2, there was an effort to allow banks to use their own methods of risk measurement in order to determine what's risky and what isn't.  Again, I don't think that is going to work out very well.  We can see pretty clearly that banks have incentives not to disclose fully the kinds of risks that they're taking. 

So, what I think we must do, and it's probably the most sensible thing to do, is to try to get much more information out into the public so creditors understand the risks that are being taken by banks, and they will then apply the kinds of market discipline that will keep banks from taking too many risks.  

What is the minimum size at which a financial institution exhausts all economies of scope and scale? (Mark Thoma of The Economist's View)

Peter Wallison: I don't think anyone can possibly know the answer to that question.  There have been a number of studies by economists of this question; what is the largest the bank can be in order to be efficient in terms of economies of scale or scope, but many of those studies have been flawed.  And I think we ought to give up attempting to determine what is the right size for a bank.  It actually could vary quite a bit depending on the kinds of assets that the bank holds. 

So, this is not a good way to pursue the issue.  I think a much better way to pursue the issue is to think about what happens to a bank that is not efficient in terms of its economies of scale or its economies of scope.  That bank is going to be out competed by other banks.  It will eventually go away. 

Let's think, for example of General Motors.  When I was on college, I was taught in my economics class that General Motors was so big and so powerful that it could never be broken up, never be out competed and that we needed a big government in order to control General Motors.  Well, what we found out was that a company called Honda, or other companies, can take care of General Motors if they produce a better product.  General Motors was probably too big, not well-managed and as a result it had to be rescued by the government. 

So, I think we ought to let the market take care of this problem of banks getter very, very large.  It doesn't necessarily mean that they're larger than they have to be in order to operate efficiently; the market will determine that by the kinds of competition that they have to face.    

What should be done to eliminate too big to fail? (Mark Thoma, Economist’s View)

Peter Wallison: If we're talking about banks, I don't think there really is an answer here except what I believe is the only effective answer.  We must have regulation of banks because they are backed by the government.  We're talking here about commercial banks, the kind that take deposits.  And we should have more transparency. 

If we are talking about non-banks and financial institutions, large non-banks and financial institutions like insurance companies, or securities firms, or hedge funds and so forth, I don't think there's anything we have to do about those.  I think if they fail, they should go right to bankruptcy and they ought to be treated as the bankruptcy laws treat failed companies.  I don't see that they will cause any problems, any kind of systemic problems because they failed.  In fact, we have evidence of that in Lehman Brothers.  When Lehman Brothers failed, yes, there was a problem in the market afterward, but it was not a problem that came out of Lehman Brothers being unable to meet its obligations.  It was a problem that came from the fact that no one expected a very large financial institution like Lehman Brothers to be allowed to fail, and when it was, they became very worried about who else they were dealing with who might be in bad financial condition. 

But the studies since Lehman Brothers have shown that all of Lehman Brothers' major counterparties, 30 of them, suffered no adverse results from Lehman Brothers' failure; and that's what I would expect.  So, we should not be trying to rescue non-bank and financial institutions.  We should just let them fail and focus entirely on the banks. 

Should Fannie Mae and Freddie Mac been allowed to fail? (Mark Thoma, Economist’s View)

Peter Wallison: Yeah.  I think the answer to that question is that I would not favor government intervention under any circumstances except in the case of a very large commercial bank.  But a large non-bank institution should be allowed to fail, and if we start interfering in the financial condition of non-bank financial institutions, we will introduce huge amounts of moral hazard into our system.  People will no longer be sure which company will fail, which companies will be rescued, and that will be far worse than anything that can occur if a major financial institution, non-bank financial institution should faile.