Read an Excerpt
Ending Government Bailouts as We Know Them
By Nicholas F. Brady, Darrell Duffie, Joseph Grundfest, Richard Herring, Thomas M. Hoenig, Thomas Jackson, William F. Kroener III, Charles S. Morris, Kenneth E. Scott, George P. Shultz, Kenneth Spong, Johannes Stroebel, Kimberly Anne Summe, John B. Taylor, Paul Volcker
Hoover Institution PressCopyright © 2009 Board of Trustees of the Leland Stanford Junior University
All rights reserved.
Make Failure Tolerable
George P. Shultz
These are tough times for the U.S. economy and for many others around the world. Tense moments in the last half of 2008 produced unprecedented actions that, according to recently published detailed accounts, were taken without the benefit of reflective strategy, on a case-by-case basis, and in an environment of panic. The result, especially in the United States, has been massive bailouts of faltering organizations with consequent commitment of huge amounts of taxpayer dollars and heavy involvement of the federal government through ownership in customarily private sector activities: selecting boards of directors and chief executives, regulating pay, and otherwise influencing corporate behavior. The American people are clearly upset about these bailouts. In the view of many, the people who created the problem should pay a penalty instead of being bailed out by the taxpayers. Who would disagree with that sentiment?
Difficult times are still with us and clearly lie ahead. Unemployment is high, the Fed has unleashed every trick in its bag (and even some that no one realized were in its bag) to stimulate the economy, government spending seems out of control, tax rates are rising with the clear prospect of more to come and with their well-documented disincentive effects, and protectionist actions are all too evident. Remember, the 1930s were characterized by the heavy tax of virulent protectionism and an increase in the top marginal income tax rate from 25 percent in 1932 to 80 percent by 1936.
WHAT TO DO?
The way to proceed is to set a strategy designed to produce growth based on the vigor of the private sector with inflation under control. One essential pillar of that strategy must deal with the current bailout mentality. The right question is, How do we make failure tolerable? If clear and credible measures can be put into place that convince everybody that failure will be allowed, then the expectations of bailouts will recede and perhaps even disappear. We would also get rid of the risk-inducing behavior that even implicit government guarantees bring about. "Heads, I win; tails, you lose" will always lead to excessive risk. And we would get rid of the unfair competitive advantage given to the "too big to fail" group by the implicit government guarantee behind their borrowing and other activities. At the same time, by being clear about what will happen and that failure can occur without risk to the system, we avoid the creation of a panic environment.
Here are a few ideas that can help make failure tolerable.
1. The first is to make a careful assessment of just what systemic risk means and how it comes about. In recent times, the words "systemic risk" have taken on the impact of a yell of "Fire!" in a crowded theater. Careful analysis is essential. My own experience with labor disputes thought to be national emergencies and a few other so-called failure situations tells me that the problem can be overestimated or can be reasonably contained. So, what are the size dimensions of the problem? Remember, markets can handle lots of size. What are the kinds of interconnections that cause trouble? Are certain kinds of activities so risky that they need to be reined in somehow? To what degree does excessive leverage create problems? Can capital requirements be structured in such a way that any risk is borne in important ways by the person deciding to take the risk? Are some activities too risky to permit financial organizations to use them for their own accounts?
2. How might intervention deal directly with the issues posed by a failure rather than by using a bailout to prevent the failure in the first place? Such action depends on the earlier analysis of what creates the risk. Then these questions arise: How can these risks be dealt with directly? What can be learned from other areas, such as the handling of major labor disputes, about how to handle systemic risk?
3. The phrase "too big to fail" implies some sort of restriction on size that would place an unnatural limit on reach and capacity. Actually, the difficulties of managing very large, disparate, and complex organizations tend to limit size. Competitors tend to cut them down. That is the history of conglomerates in the United States. Nevertheless, financial institutions present special problems because, by their nature, their activities can affect large sectors of the economy.
4. So, an escalating schedule could be required of necessary capital ratios geared to size and matched with escalating limits on leverage. The presumption here is that size happens because it brings advantages. Since size implies a certain risk to society, some additional costs would also be appropriate. Therefore, increased capital and leverage requirements are justified. Alternatively, or simultaneously, well-defined and compelling specific capital ratios and leverage limits could be related to the riskiness of the activity undertaken.
5. Understood and transparently used methods of delinking parts of large organizations could be developed so that if one goes haywire, the others can remain in business. Are you old enough to remember Christmas tree lights from long ago? When one light failed, they all went out. And the longer the string, the harder it was to find the guilty bulb and therefore the more time-consuming was the remedial action. Derivatives and securitization, so to speak, made the vulnerable string of lights even longer, increasing vulnerability and making the system more difficult to fix. The Christmas-tree-lights problem caused manufacturers to come up with a delinking system so that, these days, when one light goes out, the others stay on. If the manufacturers of Christmas tree lights are smart enough to do this, why shouldn't we be smart enough to work out delinking arrangements in the financial and corporate spheres? Obviously, limited-recourse suborganizations would have to be clearly advertised as such, so that those who play with whatever fire exists will know they could get burned.
6. Then there are organizations that grow because they are heavily subsidized. This is a deliberate process designed by government to encourage some form of activity such as homeownership. The widespread American instinct that homeownership is a good thing, that owners take care of their properties better than renters, and that people prefer to live in a nest they have created according to their own style of life are great virtues and arguably deserve some subsidy. The question is how to structure the subsidy. Tax deductions for interest payments on mortgages are one model. They are widely used and present no problem of abuse. The gross misfortunes generated by Fannie and Freddie, with their guarantees that represent large and somewhat invisible exposure, suggest that this broad approach is the wrong one. Keep the subsidy focused on the individual who has some real skin in the game, and cause lenders to keep at least some reasonable amount of their skin in the game.
7. Bankruptcy proceedings need to be examined carefully. Are different processes needed for different kinds of organizations? Do we need a system especially adapted to the financial services industry? To what degree do problems arise from slowness of application? If quicker resolution would be helpful, can some greater degree of automaticity or presumption be built into these processes? Of course, the key part of a bankruptcy reorganization proceeding is that the organization continues to function while the proceedings take place. This fact deals automatically with some of the risk factors. And the proceedings take place within an understood rule of law.
8. There has been considerable discussion of the contribution to the problem by certain financial instruments. Warren Buffett says derivatives are weapons of financial mass destruction. Securitization has been identified by many as a cause of problems because this process separates the originator of a risk from the consequences. And while risk may be spread, risk is also obscured in this process. Should something be done about these instruments? And what about other risky activities such as taking positions in private equity, hedge funds, or other trading activities? Should organizations like banks, with their access to credit from the Fed, be prohibited from trading in these kinds of presumed assets? Or, if they or other financial organizations do trade in such assets, they do so in the form of a mutual fund and not on their own account. Such a requirement would remove the risk from the financial organization. The holders of the mutual funds would bear the risks and would be entitled to the gains.
9. Recent problems got their start from a Fed-induced long period of exceptionally easy credit and a government-produced push for homeownership on terms (no down payment, no questions asked) that together produced excessive risk taking and mortgage originations. Can we expect government to act in a way consistent with prudent practice in the private sector, most especially the financial services industry?
I am attracted to Andrew Crockett's standards in his "Reforming the Global Financial Architecture":
There are four key prerequisites of an acceptable ... regime ... that permits the orderly winding down of a failing institution: (i) imposing losses on stakeholders that are predictable and consistent with avoidance of moral hazard; (ii) avoiding unnecessary damage to "innocent bystanders," especially when that would provoke a loss of confidence in otherwise sound financial institutions; (iii) minimizing taxpayer costs; and (iv) sharing equitably across affected countries any residual fiscal burden.
This conference is designed to help answer the kinds of questions I've listed earlier. The mission must be set out with clarity and urgency. The financial system is the central problem. The goal must be to remove the word bailout from our vocabulary. With that accomplished, one needed pillar for the strategy of growth without inflation will be in place.CHAPTER 2
Financial Reforms to End Government Bailouts as We Know Them
The policy workshop and this resulting book have an ambitious intellectual and practical challenge: setting out ways and means for ending government bailouts as we know them.
I want to paraphrase George Shultz when I say the object is to make the world safe for failure — failure of even the largest or "systemically important" institutions. That is an ambitious goal, but I share George's feeling that the bailout mentality has been reinforced and become pervasive after the unprecedented rescues that have taken place in the past year or more, not just in the United States but in the world. The result has been at great expense to the different fiscal positions of various governments and in terms of monetarypolicy actions. A by-product has been an unfortunate loss of credibility for central banks.
I wonder in looking at this whole problem of the financial system whether there isn't some correlation between the number of financial engineers and the number of damaging failures in the market.
In any case, we have a large problem, and I don't think we've seen the last of it. While zero interest rates may be necessary at the moment, they lead to some dangerous possibilities in terms of breeding more speculative excesses.
Let me try to deal with the particular issues of financial reform related to "too big to fail" and moral hazard. I was reminded in rereading a bit of The Wealth of Nations that Adam Smith had an answer to some of these problems. He worried about failures of banks — he thought they were overly inclined to involve themselves in speculative activities. The only good remedy he could think of was to keep them small. I'm not sure that what was possible in Scotland in 1776 is possible in the United States or other advanced financial systems today. I think we probably have to look for other answers, but if in the process we can keep the biggest banks somewhat smaller, that would help.
Let me try to clear up some definitions as you think about this problem. We talk about failure; we've had one clear failure in this period: Lehman Brothers. The stockholders lost everything. And while it remains to be seen how much the creditors lost, they certainly were affected. That was a failure situation without bailout. I think it's fair to say it had some repercussions.
Then we had a number of institutions here and abroad: Bear Stearns, Countrywide, Wachovia, Washington Mutual, Merrill Lynch — I'd include Fannie and Freddie on this list — that also failed in that they could no longer stand alone. As they merged, creditors were saved (though stockholders were wounded). They were too big to fail in terms of current attitudes but were not necessarily bailed out. It depends on what you mean by "bailed out," since those institutions no longer exist or won't exist over time. They are examples where assistance was needed to smooth a merger or to smooth liquidation, but they weren't permitted to continue.
Next we have a whole list of other troubled institutions that received large official assistance going beyond usual liquidity support by the Federal Reserve and/or the Treasury. They're still independent and operational; while some may be limping, they're still there. In most cases management is still in place, and some are doing quite magnificently in terms of compensation. That, it seems to me, fills every definition of a true bailout.
What about systemic interdependence? It's a fuzzy concept. Part of it means that the failure of some institutions, because they are so interconnected in obligations — in particular, to creditors and in terms of liabilities or assets — that their failure will cascade through the system and give rise to a wide-scale breakdown of the financial system. That was thought to be the case of AIG and, earlier, Bear Stearns.
But beyond the kind of mechanical spreading of break-downs and failures there may be an even more important psychological point: if the creditors or the stockholders hurt in a failing institution, you may have incipient panic among other institutions that otherwise would be able to stand on their own. The psychological panic — a classic run on banks — gives rise to a problem for the whole system. This seems to me one of the most difficult areas to deal with. We can talk about being tough on institutions or letting them fail, but as was the case in the fall of 2008, the officials were obviously concerned about the psychological effects of inducing runs on, or a sense of panic in, other institutions.
After that background, let me give a little picture of how I would like to see this situation approached. There are a lot of areas in which there is substantial consensus, at least in concept (some of that broad agreement will break down in practice) as to how to protect the financial institutions, the system, and individual institutions within the system from the kind of systemic breakdown to which I refer. It begins, I think, with risk management in the major institutions themselves. Obviously they did not do as well as they should have. Whether their incentives were correct or not is a large question. And I think there will be pressure — there should be pressure — on those institutions to improve risk management.
In the regulatory area, capital standards get a lot of attention. I think that's appropriate. I believe there's a limit on how far capital can be increased while keeping institutions economically viable, but there's certainly room for a more effective approach in this area. There have to be some limits on leverage in important institutions. We have had accounting problems. We've had the problem of off–balance sheet assets and liabilities — with more liabilities than assets, I'm afraid. Those kinds of issues ought to be taken care of by a more effective regulatory system. And while they may not require legislation in and of themselves, there are important questions of who does the regulation that obviously are the provenance of legislation and congressional policy.
We have the whole difficult area of compensation practices to deal with. I must confess that I admired a Wall Street Journal article in which Henry Mintzberg, a professor at McGill business school up in Canada (maybe the location says something), was arguing that the whole idea of stock compensation and stock options is mistaken. He argues that people ought to be paid in cash salaries, depending on a judgment as to their effectiveness over time. It's a short summary of a very persuasive article, and though I don't know how many people will be persuaded, compensation practices are important.
Finally, clearance and settlement practices in the derivatives area are of some consequence. The matter has caught everybody's attention after the AIG fiasco with credit default swaps, and the administration and others have been working hard to get settlement practices that can better assure outstanding derivatives can be taken care of in a way that will not unsettle the whole system and lead to systemic problems.
Excerpted from Ending Government Bailouts as We Know Them by Nicholas F. Brady, Darrell Duffie, Joseph Grundfest, Richard Herring, Thomas M. Hoenig, Thomas Jackson, William F. Kroener III, Charles S. Morris, Kenneth E. Scott, George P. Shultz, Kenneth Spong, Johannes Stroebel, Kimberly Anne Summe, John B. Taylor, Paul Volcker. Copyright © 2009 Board of Trustees of the Leland Stanford Junior University. Excerpted by permission of Hoover Institution Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.