financial crisis

Download PDF

Chester S. Spatt^†

It is a great pleasure to provide keynote remarks at this weekend’s conference on “Complexity and Change in Financial Regulation” at the Harvard Law School. In my comments tonight I will try to build from my experiences as a former Chief Economist of the U.S. Securities and Exchange Commission (SEC) from 2004 through 2007 and as a close observer of the evolution of financial regulation in the aftermath of the financial crisis to focus upon the “Complexity of Regulation.” While our financial system is itself very complex, our financial regulators would benefit in many cases by designing simple and robust approaches that build off of basic principles and that emphasize the role and importance of economic incentives and markets.

While I recognize that to some degree complexity in financial structure breeds complexity in regulation, often the causality is reversed. Complexity in regulation leads to complexity in financial structures and systems, particularly in light of the efforts of market participants to mitigate the costs and complications induced by regulation, including attempts to engage in regulatory arbitrage. Consequently, much of the costs of regulation in my view are associated with its intricacies. It also is useful to recognize that complexity in regulation leads to huge entry barriers associated with the cost of regulatory compliance. Instead of addressing “too big to fail,” this can lead to maintaining “too big to fail” institutions. This is a connection that appears to be underappreciated by our financial regulators.

Given these broad perspectives, you might not be surprised to hear that I feel that the extent of complexity in financial regulation has been, to a degree, excessive. For example, regulators often would do far better in accomplishing their regulatory goals by adapting relatively simple standards and principles that force market participants to internalize the consequences of their actions.

Economic principles emphasize the importance and power of relatively robust regulation, but in many contexts there has been little effort to try to utilize robust alternatives. An example that I will emphasize is the case of bank capital. While bank capital standards have risen in the aftermath of the financial crisis, they have not risen to levels that would internalize the full costs of risk-taking by heavily-leveraged institutions.^[1] Note that the terminology “bank capital” refers to the bank’s equity, as distinct from its debt financing, which just reflects how the bank funds or finances itself. But the essence of the “too big to fail” problem is the government backstopping the risk-taking by banks by preventing failure and default on debt in the face of systemic risk. Consequently, the pricing of debt on an ex ante basis does not sufficiently penalize the firm for the costs of its risk-taking. In contrast, governments have been willing to allow the value of a bank’s equity to collapse. High capital standards would essentially provide a solution to the “too big to fail” problem as it would internalize the costs of failure within the bank’s funding cost, rather than seeking support from taxpayers. A high capital or equity requirement forces the firm to deleverage. But as financial theory teaches a) equity and debt are substitutes in funding the firm and b) the overall cost of funding the enterprise is not substantially influenced by how the firm chooses to fund itself—at least absent governmental subsidies.^[2] This is a basic precept in financial theory, the Nobel Prize-winning Modigliani-Miller theory, understood by most recent MBA graduates. While some observers suggest that equity capital is dramatically more expensive than debt,^[3] this is inconsistent with the basic principles of efficient pricing on which our markets are built and does not reflect the true underlying riskiness and marginal costs of equity and debt.

Requiring a bank to have sufficiently high capital, such that the capital would bear any risks that the bank creates, would be a much more robust approach to the “too big to fail” problem in my judgment than the routes designed by our legislators and regulators.^[4] For example, as a byproduct of the financial crisis regulators have not attempted to scale back mega institutions and indeed, have permitted many to grow through acquisition of weak institutions (to address the immediate problems of the day). While the Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank Act) requires that systemically important institutions have “living wills”^[5] and the FDIC be empowered with “resolution authority” to resolve mega institutions,^[6] does the marketplace and the pricing of debt contracts reflect a belief that the largest financial institutions would be resolved cleanly in the event of a perceived systemic crisis? The answer to that is no. The basic point that I would emphasize is that the complexity of these institutions, as illustrated by the difficulties with “global resolution,” undercuts the credibility of Washington’s solutions to “too big to fail.” I should note that a fundamental source of complexity is the global nature of our institutions and the national structures to regulation, which leads to considerable regulatory competition and only limited coordination.

A closely related example is the Volcker Rule, which essentially bans proprietary trading by the major financial institutions.^[7] Some aspects of what is envisioned seem rather simple and clear-cut, but other features illustrate the inherent complexity of this regulation. On the one hand, it is relatively straightforward for the major financial services firms to give up their proprietary trading and hedge fund units, and indeed many of the Wall Street firms divested these activities in the immediate aftermath of the passage of the Dodd-Frank Act.^[8] On the other hand, the dealer and market-making function to facilitate the trading activities of firm customers is a central activity of financial intermediaries. Distinguishing trades that are undertaken to facilitate customer business from those that are motivated by the firm’s interest in creating trading profits can be challenging. At the heart of the regulatory proposals underlying the Volcker Rule is an attempt to develop “metrics” to try to distinguish such trades. But this is a tough distinction indeed—and at the heart of the complexity of Volcker. Indeed, recent headlines are highlighting the difficulty in implementing Volcker by the statutory deadline this summer.^[9] Recently, Congressman Barney Frank has even called for the implementation of a simplified version.^[10]

Yet what are we really trying to accomplish with Volcker? Most everyone agrees that we are not trying to block legitimate market-making, but instead we are trying to prevent the presence of “too big to fail” guarantees applying to proprietary trading and ensure that large banks and financial institutions bear the consequences of their risk-taking. Of course, the potential focus on complex “metrics” leads to considerable complexity in the proposed implementation and potential serious distortions in the market-making process. Ironically, government officials have exempted U.S. Treasuries from Volcker,^[11] appearing to acknowledge that Volcker would raise the liquidity costs of trading Treasuries—certainly suggesting that the costs for market making are real. Now European sovereigns are seeking similar relief for their bonds trading in the U.S.^[12] and not adopting analogues for firms supervised by their jurisdictions.^[13] Yet there is a potential broad solution for the types of risk-taking that Volcker is designed to confront and other types of excessive risk-taking, namely high capital standards.

Much of the focus on hard-to-implement and very complex standards, such as the Volcker Rule, would be far less significant if major financial institutions simply were forced to bear the consequences of their actions by high equity (and capital) standards. To the extent that it is apparent that government would not bail out equity, but will bail out creditors, high equity is crucial to ensuring the internalization of funding consequences and costs.

This highlights a central aspect to financial regulation in the aftermath of the financial crisis, namely that in many cases alternative approaches could be used to address key features of the same underlying problem. In the language of economics, alternative regulations would provide imperfect substitutes for one another. In the case of such substitutes there may be miscounting of the benefits relative to the costs in cost-benefit analysis. In these types of situations we might expect that the benefits, such as those associated with the reduction or elimination of “too big to fail,” may be substantially duplicated and thus overstated, while the costs of simultaneously adopting different approaches may be closer to additive. In effect, there would be diminishing returns to multiple approaches to a regulatory problem.

An interesting example of the theme that regulations can be substitutes for one another arises in the treatment of credit rating agencies under the Dodd-Frank Act. On the one hand, the Dodd-Frank Act directs regulators to remove references to credit ratings in various financial regulations.^[14] This limits the ability of rating agencies to sell regulatory treatment and limits the extent of systemic risk when a rating agency misjudges the underlying risk. Removing reliance on ratings for regulatory purposes suggests that regulators should be comfortable with rating agencies establishing their own norms, leading to competition in the definition of ratings. However, the Dodd-Frank Act also pursues a second theme with respect to rating agencies, stressing the importance of close supervision of the rating agencies by the regulator and relatively uniform standards and definitions.^[15] This is just the opposite of encouraging competition. Indeed, the case for uniform rating standards and definitions under tight supervision of the regulator would be much more compelling if the regulator were to use the ratings for determining regulatory treatment. As you can see, these two broad perspectives—reducing reliance on ratings for regulatory purposes and tighter supervision and more uniformity in standards—serve as substitutes for one another.

Credit ratings point to a number of interesting examples of how governments think about the information generated by market participants.^[16] On the one hand, there is the desire to outsource aspects of the regulatory process, but on the other hand huge suspicion by government officials of the information that would arise from such processes with respect to sovereign credits. By way of illustration, this has emerged vividly on many fronts with concerns about the quality of sovereign credits. For example, in the aftermath of the downgrade of the United States by Standard and Poor’s, the Treasury said that S&P had engaged in “terrible judgment.”^[17] In Italy, the police even raided a credit rating agency in the face of an adverse judgment.^[18] The European sovereign debt crisis led to a ban on naked credit default swaps and short selling of financials in some European countries as well as attempts to undercut the definition of a “credit event” in Greece.^[19] There can be tremendous information in credit default swap pricing—at least if there is not an attempt to manipulate the underlying credit event. On other fronts, “the European Union’s (EU) application of Basel bank capital rules placed ‘zero risk weight’ on EU sovereign debt, as if these instruments were not subject to credit risk, creating huge artificial incentives for banks to hold these sovereign credits.”^[20] Instead of being motivated by investor protection, such treatment appears to be an attempt to increase the demand for sovereign debt and the ease with which incumbent officials can continue to issue it.^[21] Even the purchases of sovereign debt by the European Central Bank would inflate its demand, thereby artificially raising the price and complicating the problem of sorting out the natural private sector interest in the bonds.^[22]

This broad class of government policies, “by weakening the integrity of market pricing”, can reduce the sovereign’s own access to funding over time.^[23] It certainly would have been difficult to see how these could be acts of investor protection, which is a core justification for much of financial regulation. It is important for government officials around the globe to have a healthy respect for market institutions rather than suspicion about these arrangements, especially when applied to their government’s own credits. This would not require the type of complexity that appears to be the hallmark of much regulatory policy.

Central clearing of relatively standardized derivatives is another important focus under the Dodd-Frank Act.^[24] This is leading to a fundamental redesign of the trading of derivative securities and swaps and the required use of central counter-parties (CCPs) for relatively standardized instruments. The re-engineering of these markets is proving to be tremendously complicated. While I am somewhat sympathetic to the use of the CCPs as a way to reduce contagion associated with counter-party risk and to make the risk structure more transparent, it is not clear that the use of clearinghouses will actually reduce systemic risk.^[25] Risks are not eliminated when matched through a clearinghouse. Indeed, the incentives to trade with weak counter-parties would be heightened (because of more favorable pricing), and more generally, the clearinghouse would tend to attract transactions that it was mis-marking.^[26] If market participants perceived the risk of trading through the clearinghouse were low (e.g., due to a potential federal guarantee) market participants would increase their risk exposures.^[27] For these reasons, as well as the concentration of risk in the CCP, it is plausible that central clearing would raise systemic risk greatly when another crisis occurred and perhaps even raise the likelihood of a crisis.^[28] While many observers have pointed to the lack of clearinghouse failures during the financial crisis a few years ago, in fact in recent decades there have been a number of clearinghouse failures, and the nature of the risks that would be assumed by a swap clearinghouse would be huge compared to that in traditional clearinghouses.^[29] “Indeed, Federal Reserve Chairman Ben Bernanke attributed the [absence of such failures] during the financial crisis to ‘good luck’” and summarized the relevant takeaway as follows: “As Mark Twain’s character Pudd’nhead Wilson once opined, if you put all your eggs in one basket, you better watch that basket.”^[30] To avoid the potential collapse of a major clearinghouse would require very strong risk management and backstops from Washington, D.C. The complexities associated with the operation and design of a swaps clearinghouse will be vast.

In trying to identify simple and robust perspectives to guide the financial regulatory process, it is helpful to focus briefly on cost-benefit analysis. This has been an especially visible issue in Washington, D.C., in the aftermath of last year’s decision by the District of Columbia Federal Circuit Court of Appeals, striking down the SEC’s proxy access rule due to an inadequate and inconsistent cost-benefit analysis.^[31] The court’s ruling, following earlier decisions by the D.C. Circuit overturning the SEC’s independent mutual fund chair and director rule in 2005 and 2006,^[32] highlights what should be central criteria for financial regulation under administrative law and economic principles. It is challenging to do cost-benefit analysis right, but until last year’s decision there was relatively little inclination by financial regulators to take the challenge seriously. In my view, the development of evidence in support of rule-making and to determine the direction of potential rule-makings is quite important. Among other tools, this emphasizes the importance of natural experiments—perhaps even randomized ones—as well as serious before-and-after analyses to assess the impact of regulations. Indeed, one way to give before-and-after analyses teeth would be to “sunset” (end) the rule adoption after a number of years, so that its merits would need to be reargued, in part using the data generated from the initial rule adoption.^[33] More generally, excess complexity in the formulation of a regulation can be a serious impediment to the generation of meaningful evidence.

Adherence to economic principles is an important way to strive for simplicity in the regulatory process. While much of the focus of the Circuit Court’s proxy access decision addressed inconsistencies in the SEC’s cost-benefit analysis, at least some aspects of that rule also suggested conflict with economic principles.^[34] I thought that a particularly striking aspect of the overturned proxy access rule was that the coalition of shareholders who would be permitted to nominate candidates to be placed on the corporation’s official proxy ballot needed to own at least 3% of the company’s shares for at least three prior years.^[35] The latter restriction seems at variance with the traditional concept of the full transferability of ownership rights through a sale. Ownership of an asset is a claim to its future rather than something vested by virtue of history.^[36]

A final theme that I would like to highlight in my discussion of simplicity vs. complexity in regulation is the importance of transparency and disclosure. This arises at many levels. First, I think that it is very important that the decision-making process of the regulators be exposed to sunlight with serious opportunity for diverse perspectives at the regulator to be articulated. Such conversations would reduce the effective complexity of regulation by contributing to the understanding of the public. Yet remarkably, a recent Wall Street Journal article highlights the almost total absence of public meetings by the Board of Governors of the Federal Reserve in the last two years and even errors by not properly recording dissents.^[37] Transparency is important not only for statements about monetary actions, but also with respect to the rule-making (and along other lines for the detail underlying liquidity facilities).^[38]

Disclosure also is an important substantive value in the regulatory process for regulators to take seriously. Yet in some situations regulators appear to have discouraged disclosures. An interesting example along these lines is the limited disclosures by Bank of America during its acquisition of Merrill Lynch, with the apparent strong encouragement of the Federal Reserve and Treasury.^[39] Indeed, more extensive and informative disclosures to shareholders would likely have led to the rejection of the acquisition, heightening concerns about the possible collapse of Merrill Lynch. The example of stress tests also is an interesting one in which the concerns of systemic risk regulators and banking supervisors would not have emphasized disclosure and indeed, the absence of any disclosure until the latter stages of the 2009 stress tests was striking.^[40] Yet disclosure is a fundamental principle in America’s capital markets, and indeed to the extent that banks plan to address stress test results with a capital raising, it is crucial that investors be apprised of the planned uses of the funds (including providing a capital cushion) rather than being misled. Full disclosure also is very simple (especially compared to incomplete disclosure)—the structure of information that is available to market participants would be transparent and recognized by all.

Financial regulation benefits from an emphasis on simple rather than complicated rules that avoid creating needless distortions, undertake serious cost-benefit analyses, use transparent rule-making processes, and emphasize disclosure and incentives.

I appreciate your listening and reflecting upon my remarks and would welcome any questions that you might have. Thanks very much.