It's a pleasure and an honor to speak this evening to the members of American Bankruptcy Institute. Given what is going on in Washington now, this opportunity couldn't be timelier. I want to talk to you tonight about a subject that is now right at the top of the Congressional agenda, and its outcome will depend on the correct interpretation of events that occurred last year.
What is the meaning of the Lehman bankruptcy?
We all know what happened after Lehman Brothers collapsed. There was a period during which the financial markets simply stopped functioning. No one had ever seen anything quite like this before. Banks and other financial institutions almost completely stopped lending--to one another or to anyone else. They began to hoard cash, and to the extent that they made loans at all it was for extremely short terms.
In response, the U.S. and other governments had to guarantee interbank lending and otherwise assure banks that they could safely lend to one another. Gradually, the interbank lending markets returned to normal, but no one who went through that period wants to do it again.
The Obama administration has turned the Lehman bankruptcy into an example of what will happen the next time a large, "interconnected" financial firm goes into bankruptcy. They argue that what is needed is a government-run system that will resolve failing or failed financial institutions in an "orderly" way, so the kind of crash that followed Lehman will not happen again. The often explicit message is that the Lehman bankruptcy itself caused the crash that followed it.
Is this plausible? I think not. There have been a number of explanations of what happened after Lehman, but the least plausible one--based on the evidence that we have--is that Lehman's bankruptcy caused the financial meltdown that followed.
John Taylor, an economist at Stanford University has pointed out that, beginning in the summer of 2007, the financial markets exhibited unusual concern about counterparty risks. Spreads on interbank lending rose suddenly on August 9 and stayed there with a few fluctuations until the Lehman collapse, after which they went even higher. What this means is that as early as mid-2007 the banks were concerned about the financial condition of their counterparties--particularly other banks. This was well before the rescue of Bear Stearns in March 2008 or the collapse of Lehman in September. Taylor then shows that immediately after the Lehman collapse spreads widened somewhat but did not reflect any special panic in the market. It was only when Secretary Paulson and Fed Chairman Bernanke went to Congress with a proposal for a $700 billion bailout plan that spreads in the interbank market rose dramatically and the financial crisis began. This explanation suggests that it was the markets' sudden realization that the U.S. government had no coherent policy that caused the panic of 2008 and the collapse of interbank lending.
Professors David Skeel and Ken Ayotte, looking at different data, compared the market reaction to Lehman's bankruptcy and the Fed's rescue of AIG. They found that the market seemed almost equally shaken by the AIG bailout--two days after Lehman--than by the Lehman collapse. They point out that there was a larger fall in the yields on Treasury bills following AIG than following Lehman, and that the TED spread, a measure of credit risk in the market, saw a larger percentage increase after AIG than after Lehman. However, a measure of volatility called the VIX--informally known as the fear index--rose slightly more percentage-wise on the Lehman bankruptcy than on the AIG bailout. These data points would suggest that the market meltdown that followed the Lehman bankruptcy was not caused by the Lehman bankruptcy alone but by a lot of unprecedented events occurring that week.
At the least--given the similar reactions to the Lehman bankruptcy and the AIG rescue--it is not clear that a government resolution system would prevent a similar market reaction in the future. After all, the AIG rescue was a kind of government resolution system, since the government took control of AIG and provided the funds that would keep AIG operating.
My own assessment is somewhat different. I see the market meltdown that followed the Lehman bankruptcy as a result of the moral hazard created by the rescue of Bear Stearns six months before. The Bear rescue led the markets to believe that all financial institutions larger than Bear would also be rescued, and this led to a great deal of complacency about the counterparty risks indicated by the sharp rise in market spreads in the summer of 2007. When Lehman was not rescued, the underlying concerns about counterparty risk came suddenly to the fore, and every large financial institution--but especially banks--had to look at their counterparties much more carefully than before. Now, the possibility that their counterparties might be insolvent or unstable suddenly became a real concern. The possibility of a government rescue was now doubtful. The natural result was a hoarding of cash, a shortening of loan terms, a dramatic rise in rates where lending did occur, and the beginning of the financial crisis. This interpretation suggests that the Lehman bankruptcy was not responsible in any meaningful sense for the market meltdown that followed.
This is not the Obama administration's interpretation, and it is useful to ask what evidence there is for their view. Unfortunately for the administration, there is none. The administration has never been asked to defend its position, and thus it skates along quite well by simply pointing to the Lehman bankruptcy as the precipitating factor in the financial crisis. For example, when asked in his testimony before the House Financial Services Committee last week why a government resolution authority was preferred to bankruptcy for resolving large financial institutions, Treasury Secretary Geithner said "Look what happened to Lehman--in the wake of Lehman. The bankruptcy code was the only option available in that context and it caused catastrophic damage."
If the administration were ever challenged to produce evidence for a statement like this, I don't think they would be able to do so. In fact, the administration's own rationale for a resolution authority cuts against their argument that such an authority is necessary. To understand why this is so, it is necessary to listen carefully to the administration's rhetoric. Here, for example, is Secretary Geithner again, this time in his testimony before the same committee in September, describing the reasons for setting up a resolution authority:
The sudden collapses of Bear Stearns, Lehman Brothers, and AIG demonstrated that our framework for supervision and regulation of large, highly leveraged and substantially interconnected financial firms--and the government's toolkit for managing their failure--is profoundly inadequate.
I want to emphasize one word in this statement--the word "interconnected." This word appears in almost every administration statement about the causes of the financial crisis and in my view is the key to the administration's effort to set up a resolution authority that will supplant the bankruptcy system for large financial institutions.
Nonbank financial institutions may be large; they may be highly leveraged; they may also be complex; but it is interconnectedness that provides the basis for taking them over. This is because it is their interconnectedness that makes them dangerous--in the administration's view--to other financial institutions. Through interconnectedness, the failure of one large financial institution can--in a kind of contagion--adversely affect the financial condition of others. Again and again in administration statements about their regulatory reforms they point to interconnectedness among large financial firms that makes them systemically significant and underlies the administration's efforts to protect the financial system from another meltdown.
But I think the Lehman case stands for exactly the opposite. Interconnectedness is important to the administration's rationale because they are assuming that it caused the disruption that occurred after Lehman's collapse. Interconnectedness, they believe, permitted Lehman's losses to be transmitted to other firms--counterparties of or lenders to Lehman--that in turn suffered major losses.
However, as far as I can tell, there is no evidence of this--none. In fact, one of the remarkable things about the Lehman bankruptcy was that--with the exception of the Reserve Fund, a money market mutual fund that held a large proportion of Lehman's commercial paper--there are no examples of firms that were driven to insolvency or instability by Lehman's failure. This suggests that contrary to the administration's argument, and the conventional wisdom as reported daily in the media, large nonbank financial institutions--even though they are interconnected in some sense--can fail without bringing down others.
In other words, the administration's rationale for setting up a resolution authority, to the extent that it is based on the interconnectedness of Lehman, is not well founded. In fact, the evidence cuts the other way. None of this means of course that the bankruptcy system is perfect and in its current form is the best possible way to respond to the special problems that occur in the wake of the failure of a large nonbank financial institution.
Later in this discussion, I will turn to the possible reforms that might improve the way the bankruptcy system might work when it has to deal with the collapse of a large nonbank financial institution. However, at this point, I'd like to describe the flaws in a government resolution system, and why the bankruptcy system is superior.
The idea behind a resolution authority is that it is necessary to avoid the possibility that the failure of a large nonbank financial institution like Lehman or AIG could cause a systemic breakdown of the kind that followed the Lehman collapse. The flaw in this idea--and it is a serious one--is that no one can know, in advance, whether a particular financial institution's failure will cause a systemic breakdown. As a result, government officials will be given what is essentially unfettered discretion to take over and resolve any large nonbank financial institution.
In practice, they are likely to follow a "better safe than sorry" policy, taking over any financial institution that might create economic disruption of some kind but not the serious crash that would be a systemic breakdown. This is the tendency of regulators and others in government, because they very seldom get blamed if they rescue a firm--after all only the unorganized taxpayers are hurt in that case--but would be severely criticized if they were to allow a financial institution to fail and it turns out that some substantial economic consequence--such as a lot of lost jobs--is the result.
GM and Chrysler are examples of companies that were clearly not systemically significant, but were rescued because they were politically significant. We can expect a lot more of this political favoritism for large companies if the government is given the power to take them over. The government's frequent use of its power will create moral hazard. The markets will come to assume that any large company will be rescued--that they are, in effect, too big to fail--and that will give larger companies significant advantages over smaller one. Market participants will understand that a large company that is too big to fail will be a safer credit than a smaller one that will go into bankruptcy if it fails. And creditors, accordingly, will provide funds to the larger companies at more favorable rates than to smaller companies.
With lower funding costs, larger companies will be able to take more risks and will come to dominate their industries, while smaller companies will be compelled to consolidate with others in order to make themselves eligible for the same government-supplied benefits. In other words, we will be creating Fannie Maes and Freddie Macs, and completely changing competitive conditions, in all areas of the economy where these large companies operate.
The administration and its supporters deny that this will be the outcome. They argue that their intention is to wind down the failed companies they take over. This may well be true, but not relevant. First, the legislation they have endorsed would give them the power to operate companies they take over and ultimately to return them to financial health. But perhaps more important is the fact that any takeover that is different from a bankruptcy will be more favorable to creditors than a bankruptcy filing, and market participants and creditors will know this quite well.
In a government-run resolution process, there is always opportunity for political influence--creditors that represent important constituencies, such as unions or state pension funds, will probably be able to negotiate more favorable treatment for themselves in a government resolution process than in a standard bankruptcy. In any event, the whole idea of a resolution authority is to allow the company to operate for an extended period without causing losses to creditors--that's how systemic breakdowns are avoided--and that's bound to be a windfall for creditors when compared to a bankruptcy.
In a hearing last week, Barney Frank--the chair of the House Financial Services Committee--said that the intent of the resolution system was to make sure creditors suffer losses: "we want to be creditor unfriendly," he said, raising questions about whether he understands how his own draft legislation would actually work.
So there are many reasons why a government-run resolution authority would be bad policy. None of these deficiencies in a government-run resolution system is present in the bankruptcy system, which provides certainty of outcome for creditors, excludes political influences, and leaves the decision about the company's future where it belongs--with the creditors.
But could there be changes in the bankruptcy process that make the resolution of large nonbank financial institutions more effective? Some work, as you know, has already been done on this question in Congress. H.R. 3310, introduced by the Republican leadership, would make a number of changes in the way the bankruptcy system functions for large financial institutions. Among other things, these reforms would (i) provide for a pre-filing consultation between the debtor and its creditors, to determine whether a forbearance would be appropriate, (ii) allow various government agencies to have access to the bankruptcy court to inform the court about the interests of any government agency or regulator in any regulated affiliate of the debtor; and (iii) allows the debtor to impose a stay under certain circumstances on the various derivatives and repos that would ordinarily be exempt from the automatic stay upon filing of the bankruptcy petition.
A number of other ideas are being discussed informally in Washington, but not formally endorsed by any anyone. One of these would provide for a special venue for bankruptcy filings by nonbank financial institutions, and perhaps a special bankruptcy court for financial institutions, so that if a filing were to occur the court would already have significant experience with the kinds of issues that arise when a nonbank financial institution files for bankruptcy. Another idea would allow the government to provide debtor-in-possession financing if no credit was available elsewhere and the government deemed it critical that a particular nonbank financial institution continue in operation.
In general, financial institutions raise difficult questions in a bankruptcy context. Unlike operating companies, their assets tend to be their franchises or brand names (the value of which may have been substantially reduced by the bankruptcy itself) and the skills of the employees, who may be inclined to seek other employment immediately upon the bankruptcy filing. Unless they are immediately or quickly perceived as able to honor their obligations, financial companies will not be able to carry on any kind of trading business. For these reasons, it may be more difficult to work out financial institutions than other kinds of companies, and the filing in bankruptcy could be a death sentence.
These aren't arguments for a resolution authority--since there is still no indication that the failure of a nonbank financial institution--no matter what its size--can cause a systemic breakdown, but the financial crisis and the threat of a resolution authority provides the occasion for a review of whether the bankruptcy code is entirely satisfactory for preserving as much value as possible that is present in a large nonbank financial institution when it fails.
In closing, I should mention that the extensive knowledge of the members of the American Bankruptcy Institute could be of great help to those of us in Washington who are working on these issues. I hope you'll consider contributing your ideas. There is very little time to lose.
Thanks for your attention.
Peter J. Wallison is the Arthur F. Burns Fellow in Financial Policy Studies at AEI.