Hearing on Regulatory Reform

Although regulation frequently has severe adverse market consequences, there are specific cases where safety-and-soundness regulation is necessary–especially when perceived government backing of financial institutions impairs market discipline. However, there is no historical precedence or evidence in today’s market that additional regulation is needed for any institutions that are neither government-backed nor designated as systemically significant. In cases where regulation is deemed necessary, successful rules must focus on increasing transparency, enhancing market discipline, and instituting counter-cyclical balances to keep market euphoria and panic in check.

I am pleased to have this opportunity to offer my views on the important questions on regulatory reform that Congress has asked you to consider. We are in the midst of a serious financial crisis, and there is a temptation to act precipitously, without thinking of the long-term consequences. It is cliché at this point that Congress does not act except in a crisis, and then it acts without thinking. Years later, lawmakers scratch their heads at the unintended consequences that they themselves have caused. It is encouraging that Congress has asked for your views on what new regulatory responses, if any, are necessary. I hope this panel will take adequate time to consider the issues discussed in this testimony, as well as others that are likely to come before Congress in the near future.

My testimony begins with a discussion of the deficiencies of regulation–its tangible and intangible costs–and then proceeds to a discussion of where safety-and-soundness regulation is necessary. Thereafter, for those areas where regulation is necessary–that is, in cases where the government is backing private companies–I outline what broad changes I believe should be made in regulatory policies. After that, I critique some of the ideas that have been advanced recently for a broadening of safety-and-soundness regulation beyond those financial institutions that are backed explicitly or implicitly by the government. Finally, I outline what I think should be the boundaries of business conduct or consumer protection regulation.

Regulation should be counter-cyclical and support market discipline

The overwhelming fact about traditional safety-and-soundness regulation–in which a government agency oversees the operations of individual companies–is that it has been a consistent failure. The S&L debacle, less than 20 years ago–when the S&L industry as well as almost 1600 commercial banks failed–was followed by the Federal Deposit Insurance Corporation Improvement Act of 1991 (FDICIA), which significantly tightened bank and S&L regulation. Now we have a mammoth financial crisis in which all depository institutions, both S&Ls and commercial banks, are the principal sources of financial weakness. Many have already failed, others will fail, and dozens have been saved from failure by the infusion of taxpayer funds. In other words, we have strong evidence that the current regulatory system has been not been effective in preventing financial breakdowns. What has been tried in the past does not work. It’s time to try a new approach.

In addition to its demonstrated failure in preventing financial collapse, there are other deficiencies of regulation that get very little attention because they are difficult to quantify or their failures are not as obvious. Here are a few of the issues that economists and others have always cited for their skepticism about regulation:

• The existence of regulation–especially safety-and-soundness regulation–creates moral hazard. Market participants believe that the government is looking over the shoulders of the regulated industry, and lenders are thus less wary that regulated entities are assuming unusual or excessive risks.
• Regulation reduces competition. The costs of regulation are more easily borne by large companies than by small ones. Large companies have the ability to influence regulators to adopt regulations that favor their operations over those of smaller competitors. This is particularly true when regulations add costs that smaller companies cannot bear. Regulation also may keep low-cost producers or foreign competitors out of regulated markets.
• Regulation impairs innovation. Regulatory approvals necessary for new products or services delay implementation, give competitors an opportunity to imitate, and add costs to the process of developing new ways of doing business or new services.
• Regulation adds costs to consumer products. Regulatory costs are passed along to consumers in the cost of services or products. These costs are frequently not worth the additional amount that consumers are required to pay.

These deficiencies–together with its regular failure as a protection for the taxpayers or the economy–suggest that regulation should be a last resort, employed only when absolutely required. In this connection, there are several circumstances that may meet this standard:

• When a company or an industry has the backing–implicit or explicit–of the government. Explicit backing exists, for example, with commercial banks. Implicit backing occurred when Fannie Mae and Freddie Mac were allowed to continue operating with government charters and other benefits that told the market that these companies would never be allowed to fail. In these cases, the wariness of creditors is impaired, and market discipline is reduced, allowing more risk-taking than would normally occur. Because of its adverse effect on competition, and its tendency to create taxpayer liabilities, government backing–explicit or implicit–should be avoided.
• When the failure of a particular company or financial institution will have systemic effects. There are elements of the self-fulfilling prophesy here. If we designate companies as “systemically significant” we will certainly allow them to become so, because the designation itself reduces or eliminates market discipline and allows them to grow faster than their competitors. In normal markets, it is very difficult for companies without government backing to become systemically significant, and currently the only companies that can be so considered are already regulated as banks.
• When there is a significant asymmetry of knowledge between a supplier of services and its customers. Personal insurance lines such as homeowners, auto, or life, are examples of this. The complex contracts required for this service are beyond ability of most consumers to understand. States or federal regulation is necessary in this case for consumer protection. The same may be true of mortgage loans. It may well be that some homebuyers do not understand the commitments they are making when they sign up for mortgages with low teaser rates and high resets. In these cases, regulation may be required to assure that the risks are made known to them in clear and simple language.
• When there is a regulatory failure of another kind, such as a harmful pharmaceutical product that could be marketed before the dangers could be known by those who use it.

Assuming, then, that regulation of safety and soundness (in other words, risk-taking) is appropriate for one of these reasons, we should be looking for methods of implementation that are more effective than the failed systems of the past. In searching for these new ways, we should keep two things in mind–both of which are founded on what we know about human nature. First, incentives matter. The people who are likely to be most effective in combating risky behavior in regulated companies are those who are most likely to be injured by it. In the case of safety-and-soundness regulation, those most likely to be injured by risk-taking are creditors. In general, creditors–unlike equity investors–get no benefit from risk-taking behavior, so they have an incentive to control and limit it. When they have the necessary information and are able to use it in a timely manner, creditors limit risk-taking by withholding funds or seeking higher interest rates on loans in order to compensate them for additional risk. This is known as market discipline. Good policy, then, would focus regulation on helping creditors exercise market discipline.

Another element of human nature with which we are familiar is the tendency to believe that when prices are going up they will continue to go up, and when they are going down that trend will also continue. In other words, as many have noted, we are subject to both irrational exuberance (manias) and panics. The current financial crisis is the result of both–a huge real estate bubble that drove housing prices well beyond the ability of people to reasonably afford, and then a collapse in prices–driven by panic–which is still continuing. The Fed is now desperately trying to avoid a deflationary spiral because of the tendency of people and businesses to sell all at once when they believe asset values are going down. So the second thing we ought to be looking for in regulation is counter-cyclicality–the incentive and the ability to act against, or encourage others to act against, the development of either bubbles or panics. This is very difficult in a political environment, because Congress does not want to stop the party when everyone is having fun (the constituents are happy), and adds to the panic by blaming everyone in sight (except itself and its policies) when things look bleak.

With these thoughts in mind, there are some things that can be done to improve regulation, where regulation is actually needed.

• Fostering market discipline through published metrics. Market discipline can be enhanced through greater transparency and more information for market participants. To some extent, this is already being accomplished through credit default swaps, which create a market-based assessment of credit quality that is accessible to all lenders. But regulation and regulators can help by working with analysts and regulated industries to develop metrics or indicators of risk-taking and then making sure that these are published regularly and accurately.
• Reducing government’s distortion of markets. The government’s participation in markets distorts outcomes and should to the extent possible be eliminated. The government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac were largely responsible for the vast inflation of the housing bubble, and particularly for the subprime and other weak mortgages that are now causing most of the mortgage defaults that are driving down housing prices and causing economic recession. The ability of Fannie and Freddie to use their government backing to raise large amounts of low-cost funds was highly pro-cyclical. As housing prices increased, Fannie and Freddie looked healthier and more prosperous, making it easier for them to raise more funds to push housing prices up further. If they had not had access to the government’s credit card, they would have had to compete for funds with other economic priorities, but the perceived safety of lending to Fannie and Freddie put them at the head of the line for new funds.

• It is more difficult to make regulation counter-cyclical, but a few ideas should be considered:

• Mark-to-market accounting. Fair value accounting should be revised and reformed. As things stand today, we have an accounting system that is pro-cyclical rather than counter-cyclical. When assets increase in market value, they can be written up–in some cases adding to the bottom lines of financial institutions–but when the market turns sour or panics, assets are excessively written down, causing financial institutions to appear weaker than they are. A sensible accounting system–one that takes account of the realities of human nature and behavior–would deny companies the opportunity to write up assets when the market is in one of its euphoric stages (say, when prices have increased by a certain percentage over the last 12 months), and similarly would not require writedowns when prices have fallen precipitously during a rapid downturn, as in the collapse of a bubble. In these cases, alternative asset valuation procedures–such as discounted cash flow–should be used, and it should be made easier for financial institutions to declare assets as held-to-maturity during these periods. In normal markets, prices will fluctuate within a limited range and will rise slowly if at all.

• Regulators use of market indicators. Prompt corrective action (PCA), which was first introduced into banking law with FDICIA, has been modestly effective in preventing the natural regulatory tendency to forbear on closing weak or failing institutions. The trouble with it is that it depends on regulators making judgments about capital positions that they themselves control. It is too easy for regulators to accept the arguments of regulated entities that their capital positions are stronger than they appear. Prompt corrective action should be strengthened by requiring regulators to take account of spreads in the credit default swap markets. Another mechanism could require larger institutions to issue a special kind of subordinated debt that could not by law be bailed out. The interest rate on that debt would also be an indicator of the market’s perception of the risk-taking by the financial institutions that have such debt issued and outstanding.

• Counter-cyclical capital requirements. Regulators are subject to political interference that prevents them from acting counter-cyclically. This is a particular problem when markets are rising, regulated entities look healthy and consumers and investors are happy. The regulators who spoil the party will be very unpopular, and–as chairman Barney Frank has noted (in a different context)–you get no credit for harms avoided. There is no way to keep Congress or the administration from interfering in the regulatory process by punishing regulators for acting counter-cyclically, but regulators can be required to demand increases in capital during flush times, so that regulated companies are putting away reserves for leaner periods. This would be an effective counter-cyclical measure that would be responsive to the problem of pro-cyclicality in regulation.

• Counter-cyclical selling and hedging. There are groups in the economy that profit from betting against trends. These include short-sellers and hedge funds–to name two business models that are less likely than others to be carried away by the euphoria that occasionally afflicts market participants. Short-sellers are particularly despised when markets are falling, but no one pays much attention to them when markets are generally rising. In those conditions, they tend to moderate the rise of bubbles. It is particularly ill-advised to require an artificial restraint on short-selling such as the uptick rule. That has the effect of enhancing the likelihood of bubbles by limiting the moderating effect of short-selling. The regulation of hedge funds could have serious adverse effects if limits are imposed on their hedging against market downturns. This hedging activity can look like speculating in favor of losses for other investors and can induce regulators under political pressure to take steps to prevent it. The recent SEC activity in banning short-selling in the shares of financial institutions is an example of a regulator reacting to these pressures. To keep this from happening in the future, short-selling should be defined in the law and specifically protected. Of course, spreading false rumors in order to drive down a stock’s price should be punished.

These measures form a kind of conceptual hierarchy. First, we should recognize the deficiencies and failures of regulation. Then we should consider that, despite these deficiencies and failures, regulation is necessary when companies or financial institutions enjoy government backing. When some regulation is required, its focus should be on two objectives: (i) enhancing market discipline by requiring regulated entities to publish metrics and indicators of risk-taking that will alert creditors to excessive risk-taking, and (ii) structuring regulation so that it is counter-cyclical in its operation.

Using this hierarchy, it is difficult to see that extending regulation beyond GSEs, commercial banks, and S&Ls would be productive or useful. It would impair competition and innovation, raise consumer costs, and interfere with the market discipline that is instrumental in holding risk-taking in check. An exception might be made for the companies at the entry point to the mortgage finance system–mortgage brokers and other local originators. It may well be that their customers don’t fully understand the risks they are assuming when they accept complex mortgage products. The choice here would be to restrict or eliminate the use of these products, or to assure that those who sell them are licensed and that the risks are adequately disclosed. If we are realistic about human nature, it seems likely that no amount of disclosure will prevent a person from signing up for a mortgage with a teaser rate and a steep reset if he believes that housing prices will rise sufficiently so that he will be able to sell the home if he can’t afford the reset rate. So if we want to have effective regulation in this case, teaser rates should simply be prohibited. The same approach might be taken with high loan-to-value lending, negative amortization, or interest only loans. All these innovations will be abused if homebuyers believe that prices are going up.

If we want to go this far in regulating the mortgage business, there are a few other policies that would help to prevent the recurrence of a housing meltdown in the future like the one we are experiencing today. First, the prevalence of non-recourse mortgages throughout the United States has encouraged the phenomenon of people walking away from their mortgage obligations when the mortgage on their home is higher than the home’s value. If people understood when they signed up for a mortgage that they are personally liable on the note, they might be less inclined to take on mortgages that they may not be able to afford. In no other area of this or any other economy are people offered a free option to abandon their obligations when they turn out to be on the losing side of an investment.

Second, we allow mortgages to be refinanced without penalty at any time, so that when the value of the home has risen, homeowners are able to refinance and take some of that price appreciation out in cash. Prepayment of an obligation without penalty is very rare in commercial transactions. Because it requires the lender to hedge the risk of prepayment, it raises the costs of borrowing for everyone. Cash-out refinancing tends to reduce the equity that remains in homes, and when prices decline–as they inevitably do–minimal home equity coupled with non-recourse loans exacerbate the tendency of people to walk away from their mortgage obligations.

Third, tax laws permit the interest on home equity loans to be tax deductible, which in turn encourages homeowners to borrow against the equity in their homes to pay off credit card or other debt, the interest on which is not tax-deductible. This too reduces the equity in the home when the downturn comes.

If we want to be sure that there is equity in homes when the next bubble bursts, we should preempt state laws that permit nonrecourse mortgages and refinancing of mortgages without penalty, as well as those provision of federal tax laws that encourage homeowners to borrow against the value of their homes.

Critique of calls for broader safety and soundness regulation

Since the advent of the financial crisis, many observers have argued that the current crisis is the result of excessive trust in the ability of markets to regulate themselves. Occasionally, these critiques go as far as to claim that this has been the prevailing theory of the last 30 years and has been proved wrong by the financial crisis. The fact that some unregulated or largely unregulated institutions have failed during the financial crisis is cited as evidence that there is a need for greater government oversight of the financial system.

This formulation misstates the history of financial regulation, ignores the fact that the most regulated parts of the economy are the cause of the most financial trouble, and fails to explain why it would be necessary for the federal government and the taxpayers to prevent the failure of any company or industry that is currently not regulated.

Taking these points in turn, there has been no “theory” during the last three decades that private markets and private financial institutions could largely be trusted to regulate themselves. The theory that has prevailed over the last three decades is the same theory that has governed U.S. government policy on financial regulation for the last 200 years–that there is no sound policy reason for the federal government to regulate or protect the safety and soundness of any financial institution that is not backed in some way by the government. The idea that the federal government has in some sense withdrawn its regulation of financial institutions over the last 30 years–or that a different theory about financial regulation prevailed in the past–is entirely fallacious.

With the limited exception of the five largest investment banks (discussed below), federally-backed commercial banks, savings and loans (S&Ls), and the government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac are the only financial institutions that have ever been regulated for safety and soundness at the federal level. In 2004, in response to a demand by the European Union that securities firms operating in the EU have a consolidated home country safety-and-soundness regulator, the SEC assumed the role of safety-and-soundness regulator for the five largest investment banks then doing business in the EU. This proved disastrous, as all five took advantage of the appearance of government regulation to overleverage themselves. But with this one exception to the general rule, it is necessary to recognize that there has never been a theory that the federal government has–or should have–any responsibility for regulating the safety and soundness of financial institutions it does not back. The policy reasons for this consistent federal position–as outlined below–are overwhelming.

In addition, to suggest, as many have, that the current financial problems of unregulated entities are in any sense or degree different from the current financial problems of regulated entities is obviously wrong. Assuming that we treat the investment banks as unregulated, there has been only one total failure among these institutions–Lehman Bros.–while four others have either been rescued (Bear Stearns) or sought shelter against the consequences of a profound loss of market confidence in their stability or solvency. It may well be that all these institutions would have ultimately failed, but this result must be compared with the failures of the many heavily regulated banks and S&Ls that have failed thus far, and particularly with the multi-billion dollar rescue of at least one bank–Citibank–that was overseen continuously for years by the Comptroller of the Currency. The argument that the failure of unregulated financial institutions was the result of their lack of regulation is clearly unsustainable; it completely ignores the fact that many more fully regulated entities have suffered the same fate. If regulation does not produce a better result than non-regulation, there is no reason to pursue it.

Finally, the reason that the federal government has not regulated currently unregulated financial institutions is that there has never been a sound policy reason for the government to prevent the failure of financial institutions that it is not backing, and there are many sound reasons for the government to stay its hand. At the outset of this testimony, I suggested four reasons for regulation in general. This includes what would be called business-conduct or consumer-protection regulation as well as safety-and-soundness regulation. In this testimony, I am discussing only safety-and-soundness regulation–i.e., regulation that is intended to keep financial institutions from failing, rather than regulation that attempts to assure that customers and clients are treated fairly.

What could be the reason that the government might want to regulate the safety and soundness, or–more specifically–the leverage, of financial institutions for which it has no financial responsibility? It is of course generally understood that the failure of companies is a good thing. Bad managements or bad business models are eliminated from the economy, making room for good managements and better business models. The losses of investors and creditors make them cautious about their investments and loans in the future, so that market discipline is enhanced. Why would the government want to prevent these salutary results from occurring? In addition, why would the government want to impair market discipline by creating moral hazard, reduce innovation and competition, and increase consumer costs–all well-known consequences of regulation? Among reasons for regulation cited at the beginning of this testimony, the only one that seems plausible as a basis for safety-and-soundness regulation is its interest in preventing defaults that might have system-wide consequences–i.e., the possibility that the failure of one institution causes other failures throughout the economy through a process of contagion. This is known generally as systemic risk.

It is important to emphasize at this point that this has never happened. There is no example in all of U.S. history where the failure of a non-regulated financial entity–securities firm, hedge fund, insurance company (if insurers are considered non-regulated because they are regulated only at the state level), finance company, or private equity fund–caused a systemic breakdown. In 1990, for example, when Drexel Burnham failed, there was no systemic result. Occasionally, the example of the hedge fund LTCM is cited. What we know of that event is that the Fed–fearful that a systemic event would ensue if LTCM failed–brought together a number of large lenders to LTCM and suggested that they rescue the fund, which they did. We don’t know what would have happened if they had not, and many scholars believe that the Fed overreacted. In any event, LTCM never failed and there was no systemic event. In order to maintain that there is now a policy basis for regulating firms and industries that have not previously been regulated, there must be a showing that there has been a change in the historical pattern–that now market conditions are different. I am not aware of any such showing.

In reality, there is no evidentiary basis whatever for arguing that the financial market is any different today than it has always been, or that the failure of an unregulated entity today would have a systemic effect on the economy as a whole. Reports in the media that the financial markets are now more “interconnected” are not evidence of any change. Financial markets have always been interconnected. That’s why financial institutions are called “intermediaries.” It is through financial intermediaries that money moves from where it is less useful to where it is most useful. There is nothing about the financial markets today that makes them more interconnected than they have ever been. It is true that money moves faster, and this makes it possible for investors and counterparties to move their funds more quickly from institutions which they regard as troubled, but this is not an indication that the markets are more interconnected, so that the failure of one institution will bring about the failure of others.

This is demonstrated by the failure of Lehman Bros. At the time Lehman failed, there was a strong adverse reaction in the financial markets. Banks stopped lending to one another, and the credit markets in general froze. We are still living with the results of that event. However, Lehman’s inability to meet its obligations did not result in the “contagion” that is the hallmark of systemic risk. As discussed below, no banks or any other Lehman counterparty seems to have been injured in any major respect by Lehman’s failure, although of course losses occurred. The markets freeze-up was caused not by these relatively minor losses but by fears on the part of banks and other financial institutions that the Lehman failure signaled unexpected weakness in other institutions and forced a sudden recognition that governments would not prevent runs by their own depositors and counterparties. These banks knew they would have to close if they could not meet depositor or investor demands for cash. Although there were media reports that AIG had to be rescued shortly after Lehman because it had been exposed excessively to Lehman through credit default swaps (CDSs), this turned out to be another canard. When all of the CDSs on Lehman were settled about a month later, AIG’s exposure turned out to be $6.2 million–a rounding error for this huge company. Moreover, although Lehman was one of the largest players in the CDS market, all its CDS obligations were settled without incident, and all the CDSs written on Lehman itself were settled for a cash exchange among hundreds of counterparties of $5.2 billion. There is no indication that any financial institution became troubled or failed because of the failure of Lehman.[1]

The question then becomes whether it is an adequate reason to regulate currently unregulated financial institutions because, in the future, investors and financial intermediaries like banks might again panic as a result of the failure of a large institution like Lehman. The pros and cons are easy to outline. I have already mentioned that there are good policy reasons to allow companies to fail, or at least to stop the government from preventing their failure. The reasons most often given for regulating currently unregulated entities are usually two: (i) regulation will induce more “transparency,” and (ii) the government rescue efforts in the current crisis have created moral hazard by suggesting to investors that many institutions are too big to fail. Sometimes this last idea gives rise to suggestions that a government agency should be empowered to identify “systemically significant” financial institutions and regulate them accordingly.

Turning first to the issue of transparency, I noted at the outset of this testimony that transparency is important in fostering market discipline, and suggested that the goal of regulation–where regulation is required because of government backing–should be to develop metrics or indicators of risk-taking and make sure that regulated entities publish these indicators accurately and on a regular schedule. Is the same requirement necessary for unregulated entities? The answer is no. Unless there is a policy reason for the government to regulate any entity, it should not do so. Government regulation that requires transparency already exists in the only area where it is necessary–disclosure required by the securities laws for companies and others that sell seek investments from the general public. Government regulation is not necessary to protect sophisticated individuals and institutions, which should know the questions to ask before they advance credit or invest in equity. If these institutions do not know the questions to ask, they should be eliminated from the market by failure, just as badly managed companies should be allowed to fail. Again, unless there is some policy reason for the government to prevent the failure of investors or creditors, there is no reason for the taxpayers to assume this obligation.

The final question, then, is whether the government rescue efforts during the current crisis have created so much moral hazard that the regulation of otherwise unregulated institutions is now necessary. This is a bit of bootstrapping by those who favor regulation. No hedge funds, for example, have needed a rescue by the government. And although many have closed their doors, none of these closures has created a systemic event. So whatever the government has done to rescue AIG or GMAC and to create subsequent moral hazard is not an argument for regulating hedge funds. It is an argument, perhaps, for regulating finance companies like GMAC, but that doesn’t seem like a particularly urgent need. At its base, the argument relies on a twisted bit of logic: regulated and nonregulated companies have been rescued; therefore nonregulated companies should be regulated.

The suggestion that a government agency should be empowered to designate systemically significant institutions and regulate them more fully than others is a particularly pernicious idea. It is especially troubling because it seems to ignore the obvious consequences of such a policy. Even assuming that it is possible to identify systemically significant institutions in advance of their failure, what would such a designation mean? Clearly, the institutions involved would not be allowed to fail–that is the reason they are being designated as systemically significant. If these institutions are not going to be allowed to fail, they will have substantial competitive advantages over institutions that are not so designated. They will have easier access to capital and loans and will grow faster. Other, presumably smaller, competitors seeking the same advantages will consolidate in order to be considered within the category of the elect few that will not be allowed to fail. In other words, designating institutions as systemically significant will have essentially the same result as creating a new crop of government-sponsored enterprises like Fannie Mae and Freddie Mac–only this group will be an open-ended category that virtually any company could join if it could prove to the designator that its failure could be a threat to the system because of its size. The effect of this idea on our competitive system would be dire.

What’s more, the designation of certain companies or financial institutions as systemically significant, opening the possibility that government will keep them from failing–far from making the financial system more stable–will have the effect of increasing the number of failures. In his famous book, Manias, Panics and Crashes, Charles Kindleberger recognizes this problem:

When asset prices tumble sharply, the surge in the demand for liquidity may drive many individuals and firm into bankruptcy, and the sale of assets in these distressed circumstances may induce further declines in asset prices. At such times a lender of last resort can provide financial stability or attenuate financial instability. The dilemma is that if investors knew in advance that government support would be forthcoming under generous dispensation when asset prices fall sharply, markets might break down somewhat more frequently because investors will be less cautious in their purchases of assets and of securities. (5th edition, p 14)

It is not possible to cure moral hazard by injecting more of it into the economy. The fact is that systemic risk is context-specific. As noted above, when Drexel Burnham failed in 1990, the markets were stable and functioning normally. There was very little reaction and no systemic problems that arose because of this failure. However, in 2008, when there was doubt about the solvency and stability of most of the world’s major financial institutions, the failure of Lehman Bros. produced a significant effect, even though there is no evidence that Lehman’s failure to meet its obligations resulted in any substantial adverse effects for any other financial institution. For this reason, it is impossible to know in advance whether the failure of a particular institution will have a systemic effect and when it will not. The effect of a failure will depend on the nature of its relationships with other institutions and on the financial condition of those institutions at some future time.

The Lehman example seems to demonstrate that even when a major institution fails at a time of profound market panic, the actual systemic risks are minimal. Other institutions come to fear runs by their depositors and counterparties, but they don’t suffer life-threatening losses as the result of the failure. Nor is it possible to argue that regulation is necessary in order to prevent the failure of systemically significant institutions; if there were ever a systemically significant institution it was Citibank, and of course the government had to step in to rescue it despite the fact that the bank was heavily and continuously regulated. So even if it were possible to identify systemically significant institutions, and even if we were willing to bear the competitive and moral hazard consequences of designating these institutions as too big to fail, we would still not be able to avert their failure through regulation. It would still be necessary for the government to step in and rescue them–at a cost to the taxpayers that would certainly be higher than if they had not been designated as systemically significant.

So even though the government has created moral hazard by rescuing some financial institutions, this is not a sufficient reason to regulate any financial institution as systemically significant. The argument rests on the erroneous idea–demonstrated everyday in the current financial crisis–that regulation can actually keep institutions from failing. And if regulation will not prevent failure, why introduce the anti-competitive consequences, impaired market discipline, and instability that will flow from–in effect–designating some entities as too big to fail?

Regulation of business conduct

Up to now, the discussion in this testimony has been confined to safety-and-soundness regulation–government oversight of financial institutions for the purpose of preventing their failure or the need for a government rescue. There is, however, another form of regulation–known as business-conduct or consumer-protection regulation. As its name implies, in this form of regulation the government sets standards for the behavior of businesses when dealing with the public. The SEC’s rules on disclosure, the OCC’s standards for how banks deal with the public, and the Food and Drug Administration’s advance approval of pharmaceuticals are all examples of this kind of regulation. It has nothing to do with the financial condition of the company involved, but instead attempts to protect the public against specific kinds of harm. In the case of the SEC, the purpose is to make sure that the public receives adequate disclosure before purchasing a security; the OCC’s rules assure that banks do not use their superior knowledge to overreach customers; and the FDA’s prior approval process is an example of the government testing the safety of consumer products before they reach the public.

In each of these cases, the government is doing something that the public could not do for themselves–because they don’t have the knowledge, the time, or the expertise. Under the securities laws, for example, there is no need for a company to make the same kind of disclosure to a sophisticated investor that it makes when it deals with the public. The reason for this rule is that sophisticated investors should know the questions to ask, and requiring a certain form of disclosure for them would be a waste of resources and would raise the cost of capital for everyone.

This gets to the central question of when business conduct or consumer protection is necessary, and when it is not. Generally, as noted at the outset of this testimony, regulation is appropriate when there a great asymmetry of necessary knowledge between the parties to a transaction. Thus, a company has far better knowledge about the value of its securities than a member of the public that is being offered a share of stock, so the government seeks to redress this balance by requiring the company to make certain written disclosures to the prospective purchaser of its shares.

The mortgage meltdown has brought to light the fact that many people signed up for mortgages without understanding the nature of the obligation they were assuming. This seems to be a proper focus for regulation, even though anyone who has bought a home knows that there are already enough regulations to require paperwork several inches thick at every closing. Nevertheless, it would be useful to have a simple one-page form that would make clear to every buyer the relationship between his income and his monthly payment and would explicitly describe any increase in the monthly payment that might be in prospect. Requiring licensing for mortgage brokers might also be appropriate, just to be sure that they understand their obligations to the public when they sell a mortgage.

But there is generally no need to require regulation where service providers do not deal with the public, but rather work at arm’s length with people who have the sophistication to understand what obligations they are assuming. For example, credit default swaps (CDSs) are transactions generally available only to sophisticated buyers. They would not be participating in the CDS market unless they were able to assess credit, and no one would want them as counterparties unless they were sufficiently strong financially to meet their obligations under a CDS. To regulate the transactions between sophisticated institutions in the CDS market would be a waste of resources. It would protect institutions that don’t need protection and that should be allowed to fail if they do not understand the risks they are assuming.

Conclusion

Regulation is known to have severe adverse consequences in the form of impairing competition and innovation and raising consumer costs. Still, there are some cases where safety-and-soundness regulation–although often ineffective–is necessary, primarily where government backing of a financial institution eliminates or severely impairs market discipline. In these cases, regulation can be improved by making the activities of regulated entities more transparent. This functions as an enhancement of market discipline. Regulatory policies can also be structured so they are counter-cyclical, and thus tend to compensate for the human tendency to become euphoric about asset growth and to panic in an asset price decline.

Despite the frequent calls in the wake of the financial crisis for wider regulation of financial institutions that are not regulated today, there is no sound policy for following this course. Banks must be regulated for safety and soundness because they are perceived as backed by the federal government. But beyond banks and GSEs, which are similarly backed, there is no policy reason–other than preventing systemic risk–to stop financial institutions from failing (although it is debatable that regulation is even capable of doing this). There is no evidence in the current financial crisis that systemic risk, in the sense that the losses of one institution spread to others and cause their failures in a kind of contagion, actually exists. But even if it did, attempting to designate certain entities as systemically significant would have severe adverse competitive results, impair market discipline, and increase taxpayer losses when rescues actually occur.

Peter J. Wallison is the Arthur F. Burns Fellow in Financial Policy Studies at AEI.

Note

1. As an aside, CDSs are no different from loans. If, as AIG did, a company writes protection on a loan, what it has done is not essentially different from making the loan itself. The widely bruited concerns about CDSs are unfounded. There is no evidence that they create any more risks than lending itself. See, Peter J. Wallison, “Everything You Wanted to Know about Credit Default Swaps–but were Never Told,” Financial Services Outlook, AEI, December 2008, available at www.aei.org/publication29158.