U.S. Senator Richard Shelby, ranking Republican on the Banking, Housing and Urban Affairs Committee, today addressed the Oxford Union regarding his position that taxpayers should not incur losses arising from private risks, especially those undertaken by entities deemed “too big to fail”:
“In my view, too big to fail is an indefensible, flawed concept,” Shelby said. “It is clearly undemocratic—treating financial institutions and their creditors unequally. It picks winners and losers and privileges the large over the small. Furthermore, it is inefficient and, therefore, bad economics. Allocating resources through government subsidies rather than through free market prices supports activities and risks that would not be undertaken without government support. By leading to bailouts, too big to fail punishes innocent taxpayers who incur losses from risks they did not undertake.”
Shelby also outlined measures that should be considered in the ongoing financial regulatory reform debate to end the “too big to fail” mentality and ensure that risk takers rather than taxpayers bear responsibility for their actions:
“Preventative measures, along with a nimble resolution regime and accountability, will help protect financial systems and economies from a repeat of the recent crisis,” Shelby concluded. “Carefully constructed rules governing financial markets can address the costly problems arising from a too big to fail mentality. It is time to reverse the trend of increasing government bailouts and return to market discipline and due diligence by investors. Most importantly, innocent taxpayers must be protected from government bailouts of private losses.”
The full text of Shelby’s prepared remarks are as follows:
PROTECTING TAXPAYERS FROM PRIVATE LOSSES
Mr. President, thank you for inviting me to speak this afternoon at the Oxford Union; the world’s most famous debating society. It is humbling to stand in this historic place where many distinguished ladies and gentlemen have also stood.
In this forum for debating and discussing controversial issues, I will address the clearly controversial issue of whether taxpayers should incur losses arising from private risks, especially those undertaken by entities deemed “too big to fail.” My position is that they should not assume the risks or the burdens of the losses.
To support that premise, I will:
• Describe the evolution of governmental support for creditors of private institutions;
• Examine government support for Government Sponsored Entities (GSEs);
• Identify government support for private activities as well as institutions;
• Explore preemptive and reactive strategies to stabilize our financial institutions, thereby protecting taxpayers from assuming losses from risks they did not undertake.
First, let us discuss governmental support in historical context.
I. Evolution of Governmental Support for Private Institutions. As the U.S. banking system was initially conceived, depositors and other creditors of failing banks were not protected by any sort of official public safety net. While a central bank may have provided liquidity in times of crisis, that support was limited to solvent institutions with sound collateral.
The Great Depression changed that calculus as the central bank was unable to provide sufficient liquidity to stabilize the banking system and maintain depositor confidence. As a result of customer runs on banks and widespread bank failures, in 1933, the U.S. Congress created the Federal Deposit Insurance Corporation (or FDIC) to insure small depositor safety.[i] Presumably, with depositors assured of receiving funds on demand, in adverse times they would have no reason to create a run on a bank. In short, deposit insurance provides assurance that deposits are safe. Depositories, in turn, fund the FDIC through premiums.
Governmental Support for Uninsured Creditors. In the 1970s and 1980s, a new model began to emerge as banks moved into broader areas and failures of large institutions began to test government willingness to support uninsured creditors. A turning point came in 1984 when Continental Illinois National Bank faced bank run pressures. Fearing that the bank’s failure would lead to instability, the government acted to prevent losses for virtually all depositors and even uninsured bondholders. FDIC resources were used to recapitalize the bank, moving governmental support well beyond the explicit and narrow legal requirement to insure only small depositors for losses.
In the years subsequent to the Continental Illinois precedent, an implicit government safety net for creditors of banks and other institutions has grown.[ii] Most strikingly, in the recent financial crisis, massive governmental support expanded to a host of unsecured and uninsured financial and commercial transactions. The U.S. Treasury, Congress, FDIC and other regulators, and the Federal Reserve all responded to what was feared to be an imminent collapse of the financial system and the economy. Those governmental responses were largely ad hoc and piecemeal.
Governmental Responses to the Recent Crisis. The Treasury and Congress enacted legislation enabling a hastily constructed $700 billion Troubled Asset Relief Program, known as TARP. I opposed the TARP bailout. Even in its final form, it was a mere outline and not a fully specified plan. In my view, TARP vastly amplifies problems of moral hazard and makes it more difficult for the government to resist bailing out private risk taking in the future.[iii] Moral hazard also grew when the FDIC put American taxpayers at risk by guaranteeing trillions of dollars of private debt.
Similarly, the Federal Reserve contributed to creating moral hazard by vastly expanding use of its discount window to fund a variety of financial market participants, including some over which it had no oversight.
The Fed also created new lending facilities to direct liquidity and credit to markets that were deemed most stressed and systemically important. The Fed ballooned its balance sheet from a pre-crisis level of around $800 billion to over $2.1 trillion through credit extensions and purchases of risky private assets, GSE debt, and U.S. Treasury debt.
Some Fed actions were taken in concert with the Treasury. While cooperation within government during a crisis is comforting, it can blur the distinction between fiscal policy and the central bank’s monetary policy and lender of last resort functions.
True to the dictum set forth by British writer Walter Bagehot in Lombard Street, in a crisis, the lender of last resort should lend rapidly and freely to solvent firms, against good collateral, and at penalty rates.[iv] True to proper application of fiscal policy, however, any time taxpayers risk loss through government purchases of assets of failing private firms, Congress and the Treasury should decide whether to assume those assets and should be held accountable for those decisions.
The Rationale of Too Big to Fail. In the recent crisis, actions by the Federal Reserve and the Treasury raised questions about central bank independence and proper accountability for use of taxpayer funds. Underlying governmental responses to fears of systemic instability was the rationale that some entities are too big to fail. Under that notion, governments support uninsured creditors of large entities who are not explicitly entitled to such support, out of fear that large losses will cause widespread, undesirable effects.
In my view, too big to fail is an indefensible, flawed concept. It is clearly undemocratic—treating financial institutions and their creditors unequally. It picks winners and losers and privileges the large over the small. Furthermore, it is inefficient and, therefore, bad economics. Allocating resources through government subsidies rather than through free market prices supports activities and risks that would not be undertaken without government support. By leading to bailouts, too big to fail punishes innocent taxpayers who incur losses from risks they did not undertake.
Too big to fail is not, and never should be, a policy.[v] Being big is not inherently bad for a financial institution. Being big and being backed by government, however, is perilous. Therefore, I have opposed bailouts, whether for financial institutions or commercial activities like automobile production.[vi]
Estimates of the costs and inefficiencies from applying the too big to fail approach to institutions are elusive, as they are embedded in implicit subsidies to those institutions.[vii] Known costs are, however, quite large. According to some estimates, in the recent financial crisis losses to banks and other financial institutions alone have cumulated to well over three trillion dollars.[viii] Those losses undoubtedly contributed to the severe economic downturn felt around the globe, as credit dried up and credit-starved businesses of all sizes cut production and employment.
II. Government Support for GSEs. The flawed rationale of too big to fail used as justification for support of large institutions is exemplified in the U.S. experiences with its mortgage-market GSEs. Prior to being bailed out by the government, the GSEs were hybrid institutions with the public policy goal of expanding home ownership and the private goal of maximizing profits for shareholders. A too big to fail treatment of the GSEs, Fannie Mae and Freddie Mac, led to a systemically threatening accumulation of risks and an eventual bailout. [ix]
The U.S. housing market has long been highly dependent on the GSEs as a source of mortgage finance. Perceptions of government backing allowed Fannie and Freddie to borrow at implicitly subsidized low rates and use the funds to buy assets that earned high rates for their private stockholders. Government guarantees dulled market discipline, and the private profit motive led to vast expansion of the GSE portfolios and a housing bubble.
The government’s bailout response to failure of the GSEs validated market expectations of government backing. While we do not know the ultimate cost, the U.S. Treasury initially pledged as much as $200 billion to the two mortgage giants. Furthermore, the Federal Reserve bought close to $1.5 trillion of GSE debt and mortgage-backed securities, partly out of fear of widespread adverse effects on the housing market.
III. Government Support for Private Activities as Well as Institutions. Too big to fail has expanded beyond financial institutions, to include financial activities as well. Some financial activities involve market participants who are neither large nor systemic individually, but are engaged in similar and highly interconnected activities that can collectively become systemically important. Fear that interruption of such activities would destabilize the financial system led governments to provide support for the activities to continue; even in the face of significant loses.
Recent examples of too big to fail financial activities include money market mutual fund investments, commercial paper issuance, asset-backed securities investments, and overnight “repo” transactions. As the largely unregulated shadow banking system increasingly created financial transactions with qualities of demandable deposit banking—but without insurance to protect against panics—the financial system again became vulnerable to widespread, destabilizing runs.
IV. Preemptive and Reactive Strategies to Stabilize Financial Institutions. In looking for ways to prevent economic meltdown and to respond when crises do occur, we must create new rules and shore up existing measures to prevent systemic threats. We also need a crisis management plan to resolve economic crises as they arise. Thus, we must preempt as well as respond.
Preemption. As a first step, I would offer the following checklist of preventative measures to consider:
• Oversight to identify potential spillovers of adverse effects of a firm’s failure on innocent parties;
• Pre-planning by regulators and policymakers to ensure readiness to respond to severely stressed conditions;
• “Stress tests” to measure systemic stability in the presence of large concentrated activities;
• Incentives and new rules to ensure adequate capitalization and sound management of large firms.
Incentives and new rules should limit reliance of large firms on short-term financing and create capital, liquidity, and supervisory requirements that grow smoothly with institution size, to reflect growing risks. Rules should require that large firms pre-plan, in “living wills,” how they can be unwound in times of stress. We should also consider promotion of hybrid “contingent capital” securities for large firms—securities that are debt in normal times but convert to equity in times of stress.[x]
Response. It would be naïve to assume that preventative measures alone will rule out expectations that some entities, being too big to fail, will be bailed out by government. Therefore, we need clear procedures for resolving failure of large financial firms and the unique short-term exposures present in many of their transactions. Such clarity, ex ante, about how large troubled firms and those with highly interconnected activities will be treated should promote due diligence and market discipline.
A resolution regime for large failing financial institutions will provide the clarity to creditors and to the market. Such a regime will operate similarly to bankruptcy proceedings, with clearly delineated procedures for settling claims. Whereas ordinary bankruptcy proceedings would likely be too slow to respond to short-term counterparty exposures, a resolution regime must be nimbler.
A well crafted resolution regime may require rapid access to liquidity. That liquidity should provide confidence to counterparties—reducing the need to redeem short-term claims in the face of dwindling assets. Short-term resources necessary to avert panics and runs must be available promptly, but must not lead to taxpayer losses. As with deposit insurance, assurances must come at a cost to those needing the resources, not at a cost to taxpayers. Those responsible for risk taking must bear these losses.
Accountability is essential to maintaining financial integrity in periods of severe stress. Regulators must be accountable, and sometimes even fired, for their mistakes and failures. The Congress and the President must be accountable for use of taxpayer funds, and private risk takers must be accountable for their losses. We cannot expect much from regulatory reform without accountability of regulators, the government, and market participants.
Conclusion. Preventative measures, along with a nimble resolution regime and accountability, will help protect financial systems and economies from a repeat of the recent crisis. Carefully constructed rules governing financial markets can address the costly problems arising from a too big to fail mentality. It is time to reverse the trend of increasing government bailouts and return to market discipline and due diligence by investors. Most importantly, innocent taxpayers must be protected from government bailouts of private losses.
I will be happy to respond to your questions.
[i] Initially, President Franklin Roosevelt was opposed, with concern that insurance “would put a premium on unsound banking in the future.” Recognizing the concern, the insurance system was constructed with strict supervision and regulation on depositories, to essentially monitor on behalf of depositors and prevent excessive risk taking, and with charges for the insurance provided.
[ii] The expanding implicit government safety net in the U.S. for private, uninsured creditors can be seen in developments over the past three decades, including: protection of nearly all depositors in a string of failures of thrift and savings and loan institutions during the 1980s and 1990s; regulatory forbearance and international debt workouts surrounding the Latin American debt crisis of the 1980s; central bank liquidity provision following the stock market crash of 1987; Federal Reserve assistance in creditor negotiations in the wind down of hedge fund Long Term Capital Management in 1998; government conservatorship of mortgage GSEs Fannie Mae and Freddie Mac; and massive open government support for a host of uninsured private financial trades over the past two years.
[iii] The moral hazard problem is well known in insurance. If you insure against a risk, your incentives to prevent adverse consequences of the risk from occurring are muted, and there is a hazard that you will relax your moral fortitude. The moral hazard problem also arises in the extension of credit. In banking and finance, for example, the problem arises when there is insurance, or expectations of insurance, against risks of losses. If someone insures you against financial losses that could arise from the counterparty on a loan that you make, you have diminished incentives to monitor the creditworthiness of your counterparty. Worse yet, if your counterparty expects insurance payments to its creditors in the event of losses, then your counterparty has little reason to forego risks and may take on riskier bets. The moral hazard problem can lead to excessive risk taking and losses to parties who did not undertake the risks.
[iv] Walter Bagehot, Lombard Street: A Description of the Money Market, New York: Scribner, Armstrong & Co., 1877.
[v] A discussion of too big to fail as a policy, as well as early warnings about the growing severity of the too big to fail problem, can be found in Too Big to Fail: The Hazards of Bank Bailouts, by Gary H. Stern and Ron J. Feldman, The Brookings Institution, 2004, and references therein.
[vi] Bailouts of U.S. automakers exemplify that too big to fail applies to commercial, as well as financial, firms. Automakers were bailed out in the recent crisis out of fear that failure would cause widespread economic instability. I opposed the bailouts, and not for the first time—I voted against auto bailouts dating back to the 1979 Chrysler case. While endowed with a solid workforce, many U.S. automakers have long suffered from poor management. Poorly managed companies do and should fail. Well managed and well capitalized companies do not fail.
[vii] A useful review of empirical evidence of subsidies from too big to fail treatment of firms can be found in “Does Bank’s Size Distort Market Prices? Evidence for Too-Big-To-Fail in the CDS Market,” by Manja V
[viii] See, for example, “Navigating the Financial Challenges Ahead,” Global Financial Stability Report, International Monetary Fund, October, 2009 (available at http://www.imf.org/External/Pubs/FT/GFSR/2009/02/pdf/text.pdf).
[ix] Fannie Mae and Freddie Mac are shorthand names for, respectively, the Federal National Mortgage Association and the Federal Home Loan Mortgage Company. Problems with the two had been known well before the recent crisis. For example, estimates of large implicit government guarantees to Fannie and Freddie, along with values at risk, can be found in “An Options-Based Approach to Evaluating the Risk of Fannie Mae and Freddie Mac,” by Deborah Lucas and Robert L. McDonald, Journal of Monetary Economics, 53, 2006. As of last year, Fannie and Freddie had grown to encompass roughly half of the $12 trillion U.S. mortgage market. Data on their portfolios are available at www.fhfa.gov. For a long time, I have advocated reform of Fannie and Freddie and proposed legislation for reform.
[x] Analyses of contingent capital can be found in: “Stabilizing Large Financial Institutions with Contingent Capital Certificates,” by Mark J. Flannery, 2009, University of Florida (available at http://ssrn.com/abstract=1485689); “Subordinated Debt and Bank Capital Reform,” by Douglas D. Evanoff and Larry D. Wall, in Bank Fragility and Regulation: Evidence from Different Countries, George Kaufman ed., pp. 53-119, 2000; and papers by the Squam Lake Working Group on Financial Regulation (available at http://www.squamlakeworkinggroup.org).