These remarks are abridged from a speech of the same title delivered April 9, 2013, to the Global Society of Fellows Conference in Richmond, Va.

In my opening remarks, I would like to discuss the problem people refer to as “too big to fail.” A wide range of reform proposals for dealing with this problem emerged following the crisis of 2008, many of which were considered in the debate leading up to the Dodd-Frank financial regulatory legislation of 2010. Some found their way into the act itself, but that has not put an end to the debate. In fact, an array of proposals have garnered attention in recent months, from breaking up large financial institutions to dramatically increasing their capital requirements. Despite the diversity of ideas, there appears to be widespread agreement on at least one of the goals of financial reform — ending “too big to fail.” This morning, I would like to discuss some work that I believe is essential to ending “too big to fail.” As always, my remarks today will reflect my own views and are not necessarily shared by others in the Federal Reserve System.¹

Broadly stated, “too big to fail” consists of two mutually reinforcing problems. First, creditors of some financial institutions feel protected by an implicit government commitment of support should the institution become financially troubled. Second, policymakers often feel compelled to provide support to certain financial institutions to insulate creditors from losses. Instances of such intervention reinforce creditors’ expectations of support and encourage reliance on highly liquid funding sources that make such support more likely. Creditor expectations of support, in turn, force policymakers to intervene in the event of distress, since disappointing those expectations by withholding support would provoke a sudden, turbulent readjustment of investor expectations regarding support for other, similarly situated firms. Perceived guarantees thus encourage fragility, which induces interventions, which encourages further fragility. The result is seriously distorted incentives to monitor and control risk and taxpayer-funded subsidies to financial firms that are widely viewed as deeply unfair.

To end “too big to fail,” therefore, we need to bring about two mutually reinforcing conditions. The first is that creditors do not expect government support in the event of financial distress. The second is that policymakers allow financial firms to fail without government support. If we can achieve that commitment and make it credible to market participants, we can improve private sector incentives to avoid fragile financing arrangements and limit risk-taking, thereby reducing the pressure for government intervention.

Allowing a financial firm to fail without government support for creditors does not mean that the operations and activities of the failing firm will simply come to an abrupt halt. Outside of the financial sector, most firms — even relatively large ones — fail by filing for bankruptcy, thereby initiating court-supervised procedures to either restructure obligations and continue operations or liquidate assets and distribute the proceeds to creditors. If it proves collectively more valuable to creditors to continue the firm’s operations as a going concern than to liquidate, bankruptcy provides a mechanism for doing so. Indeed, several large airline companies have gone through bankruptcy reorganization while continuing regular flight operations. Our goal should be for the bankruptcy of a large financial firm to be as much of a nonevent as the bankruptcy of a large airline.

A key provision of Title I of the Dodd-Frank Act requires that bank holding companies larger than $50 billion in assets, as well as certain nonbank financial companies, draw up detailed plans for their orderly resolution in bankruptcy, without government assistance, and submit them to regulators.² I believe that these resolution plans, also known as “living wills,” represent the most promising path toward ending “too big to fail.” Indeed, I believe that without robust and credible resolution plans, other financial reform strategies will be incomplete and likely to fall short.

Some observers see the Federal Deposit Insurance Corp.’s Orderly Liquidation Authority, or OLA, as a credible mechanism for resolving financial firms without bailouts. But under the OLA, the FDIC can make payments to creditors it deems “necessary,” and it can draw on funds from the Treasury to do so.³ Indeed, the FDIC has announced its willingness to do so in some circumstances.⁴ This ability, and willingness, opens the door for creditors to believe they might benefit from such treatment and therefore to pay less attention to risk than they should. If expectations of support for financially distressed institutions in orderly liquidation became widespread, regulators would likely feel forced to provide support simply to avoid the turbulence of disappointing expectations. We will have replicated the two mutually reinforcing problems that define “too big to fail.”

Many recent proposals to address the “too big to fail” problem would make structural changes to financial firms — imposing quantitative limits on their size, for instance, or prohibiting certain risky capital market activities.⁵ In my view, it makes perfect sense to constrain the scale and scope of financial firms in a way that ensures that they can be resolved in an orderly manner, without government protection for creditors. But how would you know you have chosen the right limits? Is size alone the issue, and if so, how small should you make them? If activities are the problem, which ones make them hard to resolve? And how do you know you haven’t gone too far and sacrificed valuable efficiencies that may derive from the current industrial organization of the financial system?

The only approach I can envision to answering such questions is resolution planning — that is, the hard work of mapping out in detail just what problems the unassisted bankruptcy of a large financial firm as it’s currently structured might encounter. Such maps would provide an objective basis for judgments about how the structure or activities of such firms need to be altered in order to give policymakers the confidence to choose unassisted bankruptcy in the event of distress, but without going too far and unnecessarily eliminating efficiencies associated with economies of scale and scope.

The Dodd-Frank Act requires that bank holding companies larger than $50 billion in assets, as well as nonbank financial companies supervised by the Fed, submit resolution plans annually to the Federal Reserve and the FDIC.⁶ A final rule implementing this provision was announced by the Fed and the FDIC in October 2011. A plan, or “living will,” is a description of a firm’s strategy for rapid and orderly resolution under the U.S. Bankruptcy Code in the event of material financial distress or failure. It must contain a detailed description of the firm’s organizational structure, key management information systems, critical operations and a mapping of the relationship between core business lines and legal entities. The heart of the plan is specification of the actions the firm would take to facilitate rapid and orderly resolution and prevent adverse effects of failure, including the firm’s strategy to maintain operations of and funding for their material entities. Firms are required to provide analysis under the three economic scenarios — “baseline,” “adverse” and “severely adverse.” Plans may not rely on the provision of extraordinary support by the U.S. government.

The Federal Reserve and the FDIC can jointly determine that a plan is not credible or would not facilitate an orderly resolution under the Bankruptcy Code, in which case the firm would be required to submit a revised plan to address identified deficiencies. A resubmission could include plans to change the business operations and corporate structure in order to eliminate deficiencies. If the Fed and the FDIC jointly determine that the revised plan does not remedy identified deficiencies, they can require higher capital, leverage liquidity requirements or restrict the growth, activities or operations of the firm. In essence, regulators can order changes in the structure and operations of a firm to make it resolvable in bankruptcy without government assistance.

Regulators last year required the first round of submissions from the largest U.S. firms and foreign-based companies with significant U.S. assets.⁷ These plans reflect considerable effort by covered companies to address requirements of the resolution plan rule. Through the process of developing the plans, firms have deepened their understanding of their own complex legal entity structures, their financial and operational interconnectedness, as well as the vulnerabilities that may arise in their resolution.

The living wills program will require more hard work and detailed analysis. But I see no other way to reliably identify exactly what changes are needed in the structure and operations of financial institutions to end “too big to fail.” I see no other way to achieve a situation in which policymakers consistently prefer unassisted bankruptcy to incentive-corroding intervention and investors are convinced that unassisted bankruptcy is the norm.

It is essential that we eliminate the government backstop implied by “too big to fail.” Ambiguous commitments induce fragilities that in turn induce intervention that expands implied commitments. Richmond Fed economists estimated that, at the end of 1999, about 45 percent of financial sector liabilities were explicitly or implicitly government guaranteed. At the end of 2011, as a result of the precedents set during the crisis, they estimated the figure to have grown to 57 percent.⁸ This fraction is likely to continue to grow unless we end “too big to fail.” And to end “too big to fail,” we need to do the hard work of learning what it takes to make all firms safe to fail without government rescues.