Note: Josh Rosner, managing director of Graham Fisher & Co., submitted this written testimony for a March 30 panel for the House Oversight Committee that was cancelled. His testimony has been entered into the Congressional Record and will be available on the House Oversight Committee website in the near future. The text appears below..
Has Dodd-Frank Ended Too Big to Fail?
Almost three years have passed since the United States financial system shook, began to seize up, and threatened to bring the global economy crashing down. The seismic event followed a long period of neglect in bank supervision led by lobbyist-influenced legislators, “a chicken in every pot” administrations, and neutered bank examiners.
While the current cultural mythology suggests the underlying causes of the crisis were unobservable and unforeseeable, the reality is quite different. Structural changes in the mortgage finance system and the risks they posed were visible as early as 2001. Even as late as 2007 warnings of the misapplications of ratings in securitized assets such as collateralized debt obligations and the risks these errors posed to investors, to markets, and to the greater economy were either unseen or ignored by regulators who believed financial innovation meant that risk was “less concentrated in the banking system” and “made the economy less vulnerable to shocks that start in the financial system.” Borrowers, these regulators argued, had “a greater variety of credit sources and (had become) less vulnerable to the disruption of any one credit channel.”
In the wake of the crisis, and before either the Congressional Oversight Panel or the Financial Crisis Inquiry Commission delivered their final reports on the causes of the crisis, Congress passed the Dodd-Frank Act. The act claimed to end the era of “too-big-to-fail” institutions and sought to address the fundamental structural weaknesses and conflicts within the financial system. To falsely declare an end to Too Big to Fail without actually accomplishing that end is more damaging to the credibility of U.S. markets than a failure to act at all. The historic understanding that our markets were the most free to fair competition, most well regulated and transparent, has been the underlying basis of our ability to attract foreign capital. It is this view that, in turn, had supported our markets as the deepest, broadest, and most liquid.
In fact, Dodd-Frank reinforces the market perception that a small and elite group of large firms are different from the rest. While the act sought to reduce the risks that too-big-to-fail (TBTF) institutions pose to the financial system and the broader global economy, it is unclear whether any such meaningful reduction has actually occurred. Moreover, although not fully implemented, Dodd-Frank has not reduced the number of systemically risky firms or placed meaningful new limits on their size, interconnectedness, or leverage. In fact, since the crisis began the largest financial firms have become even larger. In 1995 the assets controlled by JPMorgan Chase, Bank of America, Citigroup, Wells Fargo, Goldman Sachs, and Morgan Stanley represented 17 percent of GDP; as of January 2011 these firms controlled assets equal to 64 percent of our nation’s GDP. Today, the five largest banks, which controlled slightly more than 10 percent of deposits in the early 1990s, control over 45 percent.
During the panic of 2008, regulators approved mergers of massive firms that left the banking system far more concentrated. Rather than protecting society from the underlying problems inherent in a system comprised of a small number of highly correlated, leveraged and concentrated firms who, by size and undue economic advantage, have the ability to hold taxpayers hostage to their failings, today our legislators and the Obama White House appear complacent and uninterested in reviewing actions taken in crisis or considering whether those actions have enhanced or reduced longer-term stability.
It is of course understandable to want to “move on” from the crisis. But if we continue on our current path – and, allow the debts of certain companies to be viewed as implied sovereign obligations, the artificial economic advantages these firms receive will accelerate the demise of small and vibrant community banks, which successfully diversified the nation’s risks during the crisis. Moreover, if we again find ourselves in crisis, there is a strong likelihood that, like Ireland, creditors will demand the U.S. government explicitly accept these banks’ obligations as sovereign obligations, leaving taxpayers rather than creditors to once again shoulder the costs, as they have been forced to do with Fannie Mae and Freddie Mac.
In the years before the crisis, regulators and legislators confidently espoused now-disproven notions that the orderly resolution of Long Term Capital Management, the giant hedge fund, demonstrated that no firm should be considered “too big to fail.” As a result, regulators were believed to have armed themselves with new analytical tools and systems so that no financial firm would ever take risks that would imperil the institution. This absurdity of thought permeated even the most global bank policy initiative, the intended move from a Basel Capital Accord to a Basel II Capital Accord. Where Basel required banks to reserve for “expected losses,” Basel II supported the premise that rational actors reserve for expected losses and should only be required to reserve for “unexpected losses.”
Among the paradoxes in Dodd-Frank is this one: Key elements of the Act that seek to reduce risks to the system – branding institutions as “systemically important”, increasing their exposure to risk assets, and implementation of a subjective, and untested resolution regime – actually increase risk to the system and even accelerate the moment of our next crisis. Furthermore, the broad discretion Dodd-Frank confers to regulators, combined with the fact that regulators so miserably failed us in the most recent crisis as well as the history of legislative and regulatory capture by the industry , only serves to increase uncertainty and promise a future relationship between the government and financial system that is not only corporatist, but promotes cronyism.
IDENTIFYING “SYSTEMIC”—DANGEROUS TO DO
Dodd-Frank requires that the Financial Stability Oversight Council identify and designate certain financial firms as “systemically important”. These firms automatically include all bank entities with over $50 billion in assets and such other firms that the FSOC determines to be systemically significant. Once branded as “systemically important, these firms will be subject to “enhanced supervision” by the Federal Reserve and, in case of failure, could be subject to a special resolution regime under Title II of Dodd-Frank.
“Systemically important,” firms will also be required to submit a resolution plan that provides regulators with a road map to their resolution under a Chapter 11 bankruptcy process. This is particularly ironic, given that they are being asked to craft their own resolution using a regime that has been determined to be ineffective for them and to which they will not be subject unless, on the “eve of bankruptcy,” the Treasury Secretary, Federal Reserve, and FDIC fail to agree to place them in a Title II receivership regime. It also demonstrates that the Title II Orderly Liquidation Authority is wholly unnecessary, since regulators will only allow those firms that have an adequate plan for resolution under Chapter 11 bankruptcy to maintain its existing risk profile.
Far more troubling, though, is that this part of Dodd-Frank implicitly expands the taxpayer safety net for large institutions and does so explicitly for systemically important financial market utilities, “the implicit support or guarantee provided by government to creditors of banks that (are) seen as ‘too important to fail.’” In fact, like the government-sponsored entities before them “such banks could raise funding more cheaply and expand faster than other institutions.” Just as there was no explicit support and a strong denial by government officials of the implicit government guarantee backing the GSEs, these banks and their creditors remain of the view “if they were sufficiently important to the economy or the rest of the financial system, and things went wrong, the government would always stand behind them.”
While Dodd-Frank’s identification process is intended to reduce the moral hazard, it is more likely that “when the government singles out particular institutions or markets as being especially critical to the stability of the system, moral hazard concerns may well follow. A perception that some institutions are ‘too big to fail’ may create incentives for excessive risk-taking on the part of those institutions or their creditors. For that reason, part of an effective risk-focused approach is the promotion of market discipline as the first line of defense whenever possible.”
Given the massive failures of the Federal Reserve, which had almost six months’ notice after the collapse of Bear Stearns to plan for the possibility of failure at Lehman Brothers, Merrill Lynch, AIG, Wachovia, GMAC, Citigroup, and Indymac, it is unclear what has led the authors of the Dodd-Frank Act to believe that enhanced supervision by the Federal Reserve would prevent similar outcomes in the future. Furthermore, given the lack of unanimity among the members of the FSOCs to designate firms, there is ample reason to question the Fed’s resolve to direct firms that fail to submit an adequate resolution road map to begin to sell or disgorge assets or businesses as required by Dodd-Frank. Given the conflicting roles of the Federal Reserve, as monetary authority and prudential regulator, and its monetary policy mandates of price stability and full employment, it seems unreasonable to expect that they could or would embark on such a path.
It is especially problematic that Dodd-Frank allows for ambiguity when defining institutional failure. The manner in which one is allowed to fail determines and defines its “going concern” value when alive. Every firm must be able to fail under the same regime—a different resolution regime for a select group of firms will create incentives for creditors of those firms to treat them differently in life than the “less important” firms. It was this ambiguity that created the incentives for Lehman to make itself less able to fail and thus less easily resolved.
WHAT HAVE WE DONE? —TITLE II LIQUIDATION RATHER THAN BANKRUPTCY!
It is often argued that the bankruptcy of Lehman precipitated the broader contagion in financial markets. While this is a convenient narrative, it is neither accurate nor is it justification for the resolution regime created by Title II of the Dodd-Frank Act.
In reality, it was neither the failure of Lehman Brothers nor any supposed mortal deficiency of the Bankruptcy Code that necessitated bailouts. Rather, I would argue, it was a panicked reaction of regulators who rushed to pay out the creditors of AIG, frightened markets, and exacerbated the crisis. After all, within days of its failure, much of Lehman was sold to Barclays and a relatively orderly bankruptcy process ensued.
Are there problems with the Bankruptcy Code that need to be addressed? Yes, but these problems were not unknown prior to the crisis. In April 1999, in the wake of the failure of Long Term Capital Management, the President’s Working Group on Financial Markets recognized that the derivatives termination rules in the Bankruptcy Code work well to facilitate the continued functioning of derivatives markets when a financial firm fails, but that in very rare instances, where the failing firm has concentrated a great deal of risk, the rules may not be adequate in mitigating market volatility. The Working Group stressed the need for new bankruptcy rules but it did not, however, recommend an entirely new and overreaching resolution regime.
The failure of regulators and legislators to address these identified problems for almost a decade make the actions of regulators in 2008 seem even more pathetic. After all, their predecessors had recognized the nuances of market volatility nearly ten years earlier. In this light, it is unsurprising that regulators responded to their ongoing uncertainty by panicking and indiscriminately handing out massive bailouts.
Analyzing and fixing these sources of volatility and potential market instability is very important. The Bankruptcy Code remains the answer, and the first choice even in Dodd-Frank, but when bankruptcy is deemed unlikely to work, the Fed, Treasury, and FDIC would be forced to consider their “Orderly Liquidation Authority” (OLA) under Title II. If, on the eve of bankruptcy, the Treasury, Fed, and FDIC fail to agree to place a firm in Title II resolution regime (“turning the three keys”), this would result in that firm being nonetheless subject to the bankruptcy process. As a result, it is critical to amend the Bankruptcy Code in a manner that would provide the necessary tools to handle such an event. Logic requires one to ask: “If the bankruptcy process remains the fall-back outcome, wouldn’t it be necessary to fix that process?” If it were understood that fixing that process is necessary, why would Congress not merely accept that such a fix would mitigate any necessity of a separate, Title II regime for these firms? The answer appears clear. If regulators fail to turn the “three keys” and the Bankruptcy Code is understood to be inadequate to manage an orderly resolution of the ailing firm, then the Fed and Treasury would likely go back to Congress, as they did on the eve of the crisis, and demand an emergency allocation of governmental funds to support the financial system.
Importantly, in addressing weaknesses in the Bankruptcy Code, Congress will have to analyze how to improve the derivatives termination rules in a way that is responsive to potential volatility and market instability. For instance, Congress should consider new mechanisms that facilitate a transfer/sale of a failing firm’s derivatives book before a counterparty termination kicks in. Before making such changes it is important to remember that the landscape of the derivatives world is changing significantly due to the new derivatives requirements in Dodd-Frank. Clearing requirements will mean that a vast majority of market exposure will be concentrated in clearing-houses and that the bi-lateral model is no longer the rule. As long as these clearinghouses can withstand the shocks, then these problems that regulators have been concerned about for a decade will not and cannot be systemic events.
Instead of making surgical fixes to the Bankruptcy Code, Congress and regulators created the poorly designed OLA. At its heart it is a bailout regime. Shortly after the seizure of a large firm under Orderly Liquidation, the government may disparately treat similarly situated creditors of the financial institution that are deemed “systemically important”. The Fed may also deploy a broad-based lending program to facilitate the cash needs of other market players. (Supporters of Dodd-Frank like to say that it is not a “taxpayer bailout” because the government will recoup these initial bailout expenditures by taxing unrelated private financial institutions in the years following the bailout.) It is, nonetheless, a government-run giveaway where regulators, mostly unchecked by judicial review, get to decide who receives liquidity that would otherwise not be available.
Another fundamental flaw of OLA is that there are now two very different regimes under which a large financial firm can fail – the Bankruptcy process and the Dodd-Frank process. This is very unsettling to creditors and other stakeholders of the large firms because without adequate foresight of which resolution process the firm may enter, it is impossible for a creditor to adequately calculate one’s downside. (Even more troubling is the fact that stakeholders will have no idea which process will be employed until the firm is already seized by the regulators or has entered Chapter 11.) Regulators will argue that there is some level of certainty due to the fact that in either a chapter 11 or a Dodd-Frank Orderly Liquidation, a creditor must receive at least as much as the creditor would receive in a chapter 7 bankruptcy. However, this only demonstrates the regulators ignorance of how markets for distressed claims function, since there are multiple layers of analysis used by claims holders to determine the likely return on a claim not the “floor” return on a claim. In reality, Dodd-Frank and the Bankruptcy Code are two very different processes with very different outcomes for creditors. The value of a firm in its “going concern” state is dependent on the resolution process employed when it fails. All non-financial firms and most financial institutions use the Bankruptcy Code; commercial banks use the FDIA; broker-dealers use SIPA. There may be different systems for different types of firms, but there are not, and there should not be, multiple processes for the same firm.
In sum, the absolute worst thing that regulators can do is exactly what they’re doing now: signaling to the public and the markets, ex ante, which firms will cause systemic instability and then providing a U.S. Treasury–funded bailout scheme through the Orderly Liquidation Authority. Where investors have great certainty and clarity about the workings of the U.S. bankruptcy process, the Orderly Liquidation Authority’s dangerous subjectivity, increased opacity, preference for short-term creditors, and ambiguity in how it will treat similarly situated creditors will only increase the uncertainty among creditors of a failing institution and cause necessary risk capital to pause at precisely the time their capital is most needed.
WHAT SHOULD BE DONE? —BRING IN THE BOMB SQUAD
Beyond highlighted consideration of whether Title II or the Bankruptcy Code is a more efficient tool to manage the resolution of too-big-to-fail firms, there is ample reason to consider that neither would be effective in addressing the failures of our largest and most interconnected firms. Many of these firms have international businesses, foreign depositors, and foreign creditors and are subject to various legal regimes in different international jurisdictions. Without harmonization of international approaches to resolution there is little reason to believe that resolution by U.S. regulators or bankruptcy courts would be feasible. It is as difficult to imagine a scenario in which U.S. regulators would haircut the depositors in a foreign bank entity of a U.S. firm as it is to believe that the United States would have haircut significant foreign creditors of Fannie Mae or Freddie Mac.
Dodd-Frank misses a key reality. Rather than accepting a world in which instability is identified and left unchecked, prudential regulators should be encouraged to act on the notion that you don’t employ a bomb squad to sit around and wait for a bomb to explode; you engage them to dismantle it as soon as they find one.
While bankers and politicians are fast to warn that breaking up the big banks would reduce the competiveness of the U.S. banking system, this argument is fallacious on several counts. “Those who argue against a more proactive reduction in risk and size of TBTF institutions will, as always, revert to an argument that strikes a natural chord in every American’s heart: ‘Doing so would create an unleveled international playing field for our institutions relative to their international competitors.’ Level playing fields are a worthy goal, but this is not a relevant argument. Instead, this tired bromide must be resoundingly dismissed on several counts.”
Historically, large commercial enterprise was funded not by large banks but rather by syndicates of smaller banks and capital market participants. Resurrecting such a reliance on smaller firms would diminish the risk that if they failed, like Ireland’s or Iceland’s banks, our largest banks would imperil the solvency of the U.S. Treasury. Even Alan Greenspan, a former believer in the deregulation and consolidation of banking, had acknowledged, “If they’re too big to fail they are too big,” and highlighted the need to address the problem: “If you don’t neutralize that, you’re going to get a moribund group of obsolescent institutions which will be a big drain on the savings of the society.”
While political ideology supporting supposedly “free market” capitalism creates a natural antipathy to proactively breaking up private firms, we must recognize that, like the trusts of old, these firms have grown not by economic outperformance but by governmentally conferred benefits. This is the basis of our antitrust laws and the success of such proactive action has demonstrable benefits.
Even in the absence of a consensus to break up large firms, there are other solutions that would more properly align the incentives of the management and boards of directors of these firms with the interests of broader society. Rather than demanding the breakup of specific firms, Congress could pass legislation that would meaningfully reduce the risk of taxpayer support. The legislation should seek to cause those senior bank officials to act in a manner more consistent with the traditional partnership model of private investment banks. As has been said, “risk is the price you never thought you’d have to pay,” so the executives’ acknowledgment of the real and personal costs of excessive risk taking would be a significant step forward in self-regulation.
Where senior executives have real economic and reputational risk exposures, they are more focused on proper risk management. I would suggest that if an institution was found to need extraordinary government support, in the form of government asset purchases, debt guarantees, or borrowings extending beyond sixty days at the Fed window, the primary regulator should take “prompt supervisory action.” If the entity does not replace government support in a timely manner, the board and senior executives of the firm should be replaced at the earliest convenience of the regulator and then barred from employment as consultant, director, or employee of a regulated entity for a period of five years.
Just as limit on compensation for executives who received TARP funds created incentives for those institutions to repay TARP quickly, such a law or regulation would create incentives for officers and directors of a firm to proactively increase their risk management environment to a level at which they would never require any government support or cause them to consider selling off businesses and assets and shrinking the company to a size that could be efficiently and effectively managed. How could any regulator, policymaker, executive, or free marketeer argue against such a policy?
THE FALSE COMFORT OF RISK RETENTION
As I have written previously, recognizing failures in the process of securitization, legislators included provisions in Dodd-Frank intended to prevent loan originators and securitization issuers from selling loans to investors without regard to the quality of the loans. They reason that if issuers retain some ongoing responsibility and financial liability for the underlying loans they sell, then they will have a greater incentive to make better loans or at least make sure that the cost of those loans to borrowers will be priced relative to the risks market participants identify as inherent in the loans. To that end, Dodd-Frank generally requires financial regulators to ensure that loan originators and/or securitization issuers retain at least 5 percent of the structure of a securitization.
While the rule creates a standard and expectation of retention, it recognizes differences among the character of residential mortgages, auto loans, commercial loans, commercial mortgages, and other asset classes, as regulators deem appropriate. In recognition of these differences, Dodd-Frank permits, at origination, adequately capitalized third-party purchasers of the first-loss position to substitute for the retention requirements of the securitizer.
Does the risk retention ensure better underwriting, better lending, or safer markets? On the surface the logic is compelling. If a lender or securitizer knows he will have to drink the poison in the chalice he offers to others, then of course he would be unlikely to offer it. If, however, because of his belief in his own systemic importance or financial strength, the chalice bearer believes that he has an enhanced immunity to poison or that he will be first to receive an antidote, then perhaps he will ignore the disincentive to poison others.
More likely, as we saw in the past crisis, the same firms that poisoned their investing customers failed to recognize the power or dilution of the poison. After all, the banks that were in most dire need for direct government support were dying precisely because they had ingested large quantities of the toxic goods they had sold to others. Even with a relatively small 5 percent retention of each structure, we have seen that the different structures of similar underlying collateral were highly correlated. Thus, if securitization returns and grows, this part of the Dodd-Frank Act will have the effect of creating a future systemic risk and aid in the creation of another class of too-big-to-fail issuers.
Instead of creating further concentrations of possibly mispriced risks at the largest issuers, as Dodd-Frank’s risk retention rule does, it would make more sense to require that private issuers of securitizations finalize collateral pools prior to deals being sold. In the private-label mortgage securitization market, prior to the crisis, deals typically came to market before all of the underlying collateral had been identified or transferred to the pool. As a result investors had no way to analyze the specific collateral they were buying. It was this feature that supported mispricing of assets and allowed issuers to arbitrage deals by manufacturing the weighted average characteristics of loan pools. Just as regulation requires that an initial prospectus, with disclosure of all material information, be offered to prospective investors for a period before an equity offering could be sold, the same should hold true in the private-label securitization market. Issuers should be required to provide investors a standardized data-warehouse of loan-level information. This would serve two purposes: It would require the issuer to retain the risk for a period before the deal came to market and it would allow investors to inspect the loan-level collateral information and therefore appropriately risk-price the offering.
OTHER ISSUES TO CONSIDER
The Volcker Rule
Legislators included a rule, the Volcker Rule, to reduce conflicts and enhance safety and soundness by placing limits on the ability of large and complex banking firms to act both in their utility role as aggregators and allocators of capital and in a self-interested role of proprietary trading. After all, nobody would rationally support the notion of proprietary risk taking being covered by the safety net of implicit or explicit government support. This rule, in intent, is one step toward returning banks to their historic and appropriate narrower banking activities. Unfortunately, given the market-making roles of banks and investment banks, it is difficult to easily define which activities are consistent with their utility mission as liquidity providers to customers. Legislators left the determination and implementation of activities that would be considered proprietary to regulators. Unfortunately, there is no bright line that can be drawn to separate these activities and, as witnessed by the positive market action in the stock prices of these firms upon the release of the final rule, it appears that significant loopholes in the implementation will continue to allow these firms to engage in some significant level of proprietary trading activity. Still, even if the rule were sufficiently strong in implementation and successfully eliminated the proprietary activities of these firms, the firms would remain too large and interconnected to be seen as manageable in failure.
Office of Financial Research—A New Moral Hazard
The Dodd-Frank Act also established, within the Department of Treasury, a new Office of Financial Research (OFR). A key goal of the office is to “improve the quality of financial data and provide analytic support to the FSOC and its members.” The goal is to create a series of timely databases, available to the Financial Stability Oversight Council members and market participants, that would allow for more accurate regulatory and market assessments of systemic and firm risks. As stated in the OFR’s “Frequently Asked Question” paper :
The problem of monitoring systemic risk is closely related to the risk management challenge that individual firms face. To monitor risk in the financial system, positions in thousands of diverse financial products, involving thousands of individual financial firms, have to be aggregated across the entire financial system in ways that are meaningful. Standardizing the way financial transactions are reported, and the consistent use of robust reference data on the key characteristics of individual financial instruments and counterparties, can greatly facilitate this process for regulators and individual firms alike.
The OFR will, in consultation with relevant stakeholders, develop standards for financial data and publish reference databases of financial entities and instruments that will be made available to the public. As mandated by the Act, data security and confidentiality will be a top priority for all of the OFR’s data activities, including the publication of the reference databases. These reference databaseswill also likely be used by market participants.
The industry‐wide standards for financial data and reference databases will help the FSOC to monitor systemic risk and improve the efficiency and efficacy of risk management, reporting and other business functions at individual financial institutions. These actions will enhance both supervision and market discipline by giving both supervisors and market participants better visibility into the risks that individual financial firms take.
One of the key lessons that should have been learned from the crisis appears to have been left unconsidered: There is a great difference between data and information, and the value of data is only as good as the user’s ability to apply the data in a valid and informative manner. As Federal Reserve Chairman Bernanke has previously stated, it is unlikely that these databases and the information gathered will result in practically useful information. In fact, the collection of such information may increase the perception of a government support for the underlying products or institutions:
Given the complexity of trading strategies and the rapidity with which positions change, creating a database that would be sufficiently timely and detailed to be of practical use to hedge funds’ creditors and investors or to regulators would be extremely difficult. Collecting such information also risks moral hazard, if some traders conclude that, in gathering the data, the regulators have somehow reduced financial risk. The principle of consistency on which I am focusing today raises an additional objection to this proposal, which is that it would make little sense to collect data on hedge funds’ positions without gathering the same information for other groups of market participants that use similar strategies and take similar risks.