This site has moved to
The posts below are backup copies from the new site.

April 4, 2009

Economist's View - 4 new articles

"Why Creditors should Suffer Too"

Tyler Cowen:

Why Creditors Should Suffer, Too, by Tyler Cowen, Economic View, NY Times: The Obama administration's proposals to reform financial regulation sound ambitious enough as they aim to bring companies like A.I.G. under a broader umbrella of government rule-making and scrutiny.

But there is a big hole in these proposals,... the new proposals immunize the creditors and counterparties of such firms by protecting them from their own lending and trading mistakes.

This pattern has been evident for months, with the government aiding creditors and counterparties every step of the way. Yet this has not been explained openly to the American public.

In truth, it's not the shareholders of the American International Group who benefited most from its bailout; they were mostly wiped out. The great beneficiaries have been the creditors and counterparties at the other end of A.I.G.'s derivatives deals — firms like Goldman Sachs, Merrill Lynch, Deutsche Bank, Société Générale, Barclays and UBS.

These firms engaged in deals that A.I.G. could not make good on. The bailout, and the regulatory regime outlined by Timothy F. Geithner..., would give firms like these every incentive to make similar deals down the road. ...

What the banking system needs is creditors who monitor risk and cut their exposure when that risk is too high. Unlike regulators, creditors and counterparties know the details of a deal and have their own money on the line. ...

A simple but unworkable alternative is to let major creditors make their claims in the bankruptcy courts, as was done with Lehman Brothers. But that is costly for the economy and, after the fallout from the Lehman failure, politically impossible now. Instead, the key to effective regulatory reform is to find a credible means of imposing some pain on creditors.

Here is one possibility. The government has restricted executive pay at A.I.G. and banks receiving government funds, but this move fails to recognize that the richest bailout benefits go to creditors. Restricting compensation at these creditor firms would have more force — if it is done transparently, in advance and in accordance with the rule of law. A simple rule would be that some percentage of bailout funds should be extracted from the bonuses of executives on the credit or counterparty side of transactions.

Such a rule would make lenders more conservative, which would generally be a good thing. ...

Here is another option..., credit agreements should provide for the possibility of a future, prepackaged bankruptcy. Those agreements should require that the creditors themselves would suffer some of the damage — even if the government stepped in to bail out the afflicted firm.

There is a risk that these sacrifices will not be extracted when the time comes, but the prospect might still check the worst excesses of leverage. ...

This poses a very difficult public relations problem for the government, because the Federal Reserve and the Treasury do not want to discuss the importance of the creditors too publicly right now. ... The challenge isn't easy, and we can't start on it today, but one way or another a new regulatory plan has to move some risk back to creditors.

These seem like good ideas, but I'm not so sure that these options pose a threat that is credible enough to "check the worst excesses of leverage" as much as is needed. Given the propensity for bank runs in both the traditional and non-traditional banking sectors when depositors/creditors are threatened - they will want to get their money out of institutions that look to be headed for trouble, i.e. to stop being creditors, before the firm is declared insolvent and these restrictions cause them to incur losses that they would be protected from otherwise - would the government step in with guarantees that keep them whole even if, a priori, they had declared they wouldn't in order to avoid the negative fallout from such runs?

The possibility that the government would keep its word and enforce the losses, along with the possibility of being unable to get money out in time and then coming under the regulatory restrictions, ought to check behavior to some degree. But to get substantial effects we need to have a substantial probability that the government will keep its word about imposing losses. However, I'm not so sure that the belief that the government will keep its word is as widely held as is needed, particularly if too large to fail, politically powerful institutions are allowed to persist.

"How to Improve the Geithner Plan"

Sandro Brusco explains how the Geithner plan can be improved by having bank managers put part of their own compensation at risk when purchasing toxic assets, and by using future profits and any executive compensation above some minimum amount as collateral for the loans:

How to improve the Geithner plan, by Sandro Brusco: Two major criticisms have been moved against the Geithner plan

First, the plan fails to activate a true process of price discovery in the market for mortgage backed assets. Instead, the plan tries to fill the gap between the willingness to pay of the potential buyers and the price at which the banks are prepared to sell with a public subsidy. The presence of the subsidy, in the form of no-recourse loans at a presumably low interest rate, implies that the prices that will be determined in the auctions will probably be of little help for the future.

Second, the plan places a lot of risk on the FDIC and the Fed. Given the nature of these institutions, if they end up substaining heavy losses, the general public will pay directly or indirectly a high price.

I believe that both criticisms are very valid. I have discussed elsewhere a possible alternative approach, and other sensible plans have been submitted. But the Geithner plan is what we have now. Thus, although maybe not optimal, we should try to see whether the plan can be amended in such a way that the two above-mentioned problems can be at least partially fixed. I believe this is possible, and here is what I propose.

First, to fix the price discovery problem we should ask the bank managers to lend their own money, alongside the FDIC and the Fed, for the acquisition of the toxic assets. Right now, it is claimed, the market for mortgage backed assets is not working because the banks put a value on their assets which is higher than what the potential buyers are willing to pay. The only way in which this makes sense, from a theoretical point of view, is that there is asymmetric information. The banks, or more precisely the managers who run the banks, know things about their assets that the investors do not know. If we want to reactivate the market the asymmetric information has to disappear. How do we get there? Well, if the bankers really believe that their assets are worth more than what the investors are willing to pay they should credibly signal their information. One simple way to do that is to participate in the debt financing of the assets' acquisition. The managers should be asked to put their money on the line, lending their money to the private investors who buy the toxic assets at the same conditions as the FDIC and the Fed. In fact, given that the managers have greater control and better information on the assets which are put on sale, their debt should be junior to the one of the FDIC and the Fed. How much should they put? For the operation to be credible it should be a sizable part of their wealth, this is the only way in which the market is going to believe that the bankers ''really mean it'' when they claim that their toxic assets are quite valuable. I don't know enough to provide a firm number, but a simple idea would be to look at their total compensation over the last five years or so, and ask them to invest a significant fraction of it.

Let me emphasize that the main goal of this measure is not to provide more funds for the acquisition of toxic assets. Rather, it is meant to provide incentives for a more realistic assessment of those assets. How exactly prices will be determined is still not clear, and it is the crucial point of the process. Even if the assets are sold through a competitive auction process, the final outcome will depend on parameters such as the total amount of assets supplied and the reserve prices which are set. The bank managers will presumably participate in the process of determining those parameters. Right now, managers have no incentives to make sure that the assets are correctly priced; on the contrary, they have all the incentives to make sure that the assets are overpaid. Since there is a public subsidy, the buyers may well accept to overpay. It does not need to be that way. If the managers have their own wealth at stake they will be careful in setting the parameters of the auction and in selecting the assets for sale. This will lead to more realistic values for the assets. Getting realistic values for the assets is the only hope to get the market to start working again.

Second, to reduce the risk for the public, the banks should put their earnings as collateral for the loans. Also, compensations for employees in excess of a certain threshold should be put as collateral. Under the current version of the plan, once the banks have sold their assets they are not responsible any longer for their performance. This is not a good idea. When there is asymmetric information the payoff to the seller should be linked to the future performance of the good which is sold. This is what happens, for example, when a car dealer offers a warranty on a used car. One simple thing that can be done is to use banks' future earnings to guarantee the debt of the FDIC and the Fed. The financing plan should specify that the banks are not allowed to distribute dividends until the debt is completely paid. Earnings should be put in a special fund. If the toxic assets underperform and the debt is not fully served, the money put in the special fund will go towards the service of the debt. In addition, all compensation to bank employees in excess of a certain amount should go into the fund. If and when the debt is fully served the employees will get their money and the shareholders will get their dividends. But if things go bad and there is default on the debt then the shareholders and the employees will have to participate to the losses. Asking the highest paid employee of the bank to provide their compensation as collateral would also strengthen their incentives to avoid overpayment of the toxic assets, thus additionally reducing the risks for the government.

I think that the two proposals can be defended purely in terms of their economic content. They lead to more credible prices and a better repartition of risk. However, since it is obvious that important political factors are at play here, let me add a couple of ''political'' considerations.

First, it is clear that the administration is reluctant to ask Congress for additional funds. Bailing out the banking industry, whatever its economics merits, is clearly unpopular in the US right now. For practical purposes it is therefore prudent to consider the amount of public money that can be pumped in the operation as fixed. This implies that maximizing its effectiveness in removing the toxic assets from the balance sheets is even more urgent. Having the managers participate in debt financing would be a good move towards that goal. To start with, it is likely that it will moderate the overpayment of the assets. Lower prices mean that each dollar of investment will remove more toxic assets from the banks, creating more bang for the buck. Moreover, the prices will be more credible and private investors will be more willing to believe that they accurately reflect the knowledge that the bankers possess on the true value of the assets. They would therefore require a lower leverage for their acquisitions. This way the funds provided by the FDIC and the Fed would attract a larger amount of private funds.

Second, if the value of the toxic assets turns out to be less than hoped (''expected'' seems too strong a word here) and there is default on the debt these arrangements will provide political cover for the administration. Imagine a scenario in which the FDIC loans are not repaid in full but at the same time the banks distribute fat dividends to their shareholders and large bonuses to their managers. That would probably cause a political firestorm compared to which the whole discussion on the AIG bonuses would pale. And, differently from the AIG controversy, the blame would be entirely and squarely on the current administration. The issue may explode at any moment before the midterm elections of 2010 or the presidential election of 2012, with lethal electoral consequences. If the two remedies suggested above are implemented this will not happen. If the FDIC ends up losing money no fat dividends and no large bonuses will be observed. On the contrary, the administration will be able to take credit for its cautious approach and will be able to claim that the money lost by the public did not go straightly into the undeserving pockets of powerful and connected people.

Free Passes

It looks like I'll be at the same conference as Paul Kedrosky:

The excellent Milken Global Conference coming up on April 27-29, 2009, has fifteen free passes available. But there is a catch: You need to be recently unemployed, having lost your job within the last eighteen months. (It also helps logistically if you're in California, as the conference is at the Beverly Hilton in Los Angeles.) The Milken people will on April 10th draw the fifteen "winners" from a pool of folks who qualify,

It's a great gesture from the folks at the Milken Institute, as the conference (which I'm attending again this year) is usually chock-a-block with smart people and highly thought-provoking.

Here is the offer.

I'll add, from the email asking us to note this:

Global Conference is a tremendous opportunity for exposure to the latest trends and the ultimate networking opportunity. There are about 3K top level attendees from finance, government, business, entrepreneurs, philanthropy, non-profit and academic - from all 50 states and about 60 countries from around the world.

This is a particularly good opportunity for anyone who could benefit from the networking opportunities, one that, given the costly registration fee ($4,000), wouldn't be available otherwise.

As for the conference, I was there last year, and I am anxious to see how (and if) attitudes have changed. Last year, the sense I had from the sessions I attended was that there was begrudging acceptance that some sort of financial regulation was coming, and perhaps even needed, but we had to be careful about regulatory overreach as we got back to business as usual. You could feel the resistance to change in the air. It was a matter of getting through the crisis, and then reestablishing financial markets much as they were before (with some limitations coming from the new regulations, but as few as possible).

Will attitudes be different this time? I suspect not, partly because of the conservative slant of the attendees (e.g. Karl Rove, Steve Forbes, Rupert Murdoch, Sam Zell, and Bill Bennett, and economists like Gary Becker and Myron Scholes), and also because of who they are - people who do not want their activities to be hindered by rules. But I'm still interested to see if there is any sense that things must and will change, for their own mutual protection if nothing else, because if they are resigned to the need for change, that's a sign regulatory changes will be easier to bring about. And I'll be interested to assess attitudes more generally, e.g. about recovery from the crisis, though once again I expect lots of bullishness, and lots of "get the government out of the way and let us fix this thing" sort of attitude, but we'll see. I'm going on a press pass and I plan to ask questions whenever I can, so I'll let you know about that, and whatever else I find out.

links for 2009-04-04

No comments: