Center for Financial Innovation and Stability

Notes from the Vault

Larry D. Wall
April 2010

• "Too big to fail" policies are not about bank size per se but rather about the impact of financial firms' failure on financial stability and the real economy.
• Abolishing all government authority to engage in too big to fail policies may only delay bailouts in the event of major financial instability that affects the real economy.
• Critical stumbling blocks to the true elimination of too big to fail policies include the problems of resolving cross-border financial groups and dealing with the too-many-to-fail problem.

Congress tried to end "too big to fail" with its passage of the Federal Deposit Insurance Corporation Improvement Act of 1991 (FDICIA) (Wall 2010). Unfortunately, those efforts have not withstood the challenges posed by the recent crisis, and the persistence of too big to fail is having a variety of undesirable consequences.

The mere perception of the existence of too big to fail policies is sufficient to distort portfolio allocations toward more risky loans and investments. The perception of too big to fail also creates a competitive advantage for firms that are or grow to be too big to fail. The combination of more risky investments with a higher proportion concentrated in institutions that are too big to fail increases the likelihood of financial instability. Further, the fiscal costs of implementing too big to fail policies distort the allocation of governmental resources by diverting current and future tax receipts to recapitalize failed firms.

Simple solutions will not work as advertised
Given the problems with too big to fail, the seemingly obvious solution is to forbid the government from assisting failing financial firms. Unfortunately, this simple solution by itself is unlikely to change either the perception or reality of too big to fail. The key is that too big to fail is not really about protecting large financial firms per se but instead is about preventing a sudden loss of financial services that would have a significant adverse impact on the real economy. The loss in services could arise directly because of the firm's size or placement in the financial system or indirectly through the failure's impact on other financial firms or markets.

While a "just say no" policy to too big to fail could prevent the government from taking timely action to prevent the failure of the first firm, that is not necessarily the end of the story. The government can always reverse policy and provide blanket credit guarantees or capital injections.1 If a firm's failure does significantly damage the financial system, failure to follow too big to fail policies results in the worst of both worlds. As the firm's failure spills over into the real economy, the pressure on the government to protect the rest of the financial system will be intense, and the result will likely be the simple, easy-to-understand policy of guaranteeing all financial firms. Moreover, market participants can forecast this outcome. Thus, while they will be concerned that their firm would be unprotected if it were the first to fail, they could take comfort in the likelihood of blanket guarantees if their firm is not the first to fail.2

Once the too big to fail problem is understood as that of preserving the continued provision of financial services, it is immediately obvious that some other proposed alternatives to too big to fail are not sufficient to end too big to fail. For example, policies designed to make commercial banks safe from failure will not end too big to fail if critical financial services are provided by nonbank firms. In the most recent crisis, Bear Stearns and AIG were nonbank financial firms whose failure was deemed such a risk to the economy that extraordinary assistance was provided. When a third nonbank firm, Lehman Brothers, did enter bankruptcy, its failure caused shock waves that cause severe damage to the financial markets and the real economy.

Benefits of special resolution authority
What is needed to end too big to fail is to create the perception and reality that the failure of any firm or group of firms will not cause financial instability that would substantially reduce real economic growth. The bailouts and costly failure of nonbank financial firms during the recent crisis arose in large part because their failure could be resolved only through the bankruptcy courts using laws designed for nonfinancial firms. The bankruptcy courts are set up to provide a fair distribution of a firm's claimants. These laws are not designed to take into account the effects of a firm's failure on the rest of the economy. Commercial and savings banks are not resolved under the bankruptcy code, however, but instead by the FDIC (Dwyer 2010). The agency's procedures have proven reasonably efficient for creditors while providing a mechanism that can take into account systemic externalities.

The need to adopt a new mechanism to resolve failing nonbank financial firms is now widely accepted, though exactly how best to do it remains somewhat controversial. For present purposes, it is sufficient to note that creating a special resolution procedure for financial firms that takes account of the financial stability concerns could go a long way toward ending too big to fail.

Limits of enhanced resolution authority
Unfortunately, however, special resolution procedures cannot take us all of the way to ending too big to fail–type policies because we do not yet have good solutions to two problems: the resolution of cross-border banking groups and the "too many to fail" problems that arise when a substantial fraction of the financial system is in distress—especially if that distress arises from a common risk factor.

The issue with cross-border groups is that U.S. legal authority stops at its borders. The United States has no authority to resolve subsidiaries with foreign charters. This issue is an important problem because the boundaries between domestic and foreign operations are often fuzzy, with these groups tending to act more as integrated entities than as a collection of independent firms. This distinction could potentially pose problems in resolving U.S. financial groups that may be depending upon foreign subsidiaries for important services. The United States would also face problems should a foreign group with systemically important U.S. operations become distressed. Foreign governments would face similar concerns about the continued operations of distressed systemically important subsidiaries owned by U.S. groups in their country.

The good news on cross-border resolution is that policymakers in the G20 countries recognize this issue as one that needs serious attention (G20 2009). The bad news is that finding an answer to the cross-border resolution issue will be especially difficult for at least two reasons: (1) there are wide disparities in the way countries resolve failed firms (not just banks), and (2) if a group of firms needs some assistance (even temporary liquidity assistance) to continue providing services, deciding how to apportion the burden of that cost and the resulting control rights over the distressed group would be very difficult to arrange.3

The second issue impeding the elimination of too big to fail is an old but underappreciated problem (see Acharya and Yorulmazer 2007). If an individual firm enters resolution or bankruptcy, other financial firms may expand their production of financial services or buy parts of the distressed firm. But if a large fraction of the financial system is failing because of similar risk exposures, those possibilities are more limited. Moreover, putting a large financial firm into resolution creates the risk of increased market pressure on other important financial firms. The two successful large financial firm resolutions in the United States since the early 1980s, Continental Illinois and Drexel Burnham Lambert, worked because the reasons for these firms' failures were idiosyncratic.4 However, policymakers worked to prevent failures in situations where a systemically important part of the financial system was exposed to similar losses, including the cases of the less-developed countries debt crises in the 1980s, Long-Term Capital Management's collapse in the 1990s, and Bear Stearns's failure in 2008. Unfortunately, there does not appear to be an easy solution to the too many to fail problem.

Next steps
A major goal of financial regulatory reform is to end too big to fail. While the creation of a special resolution regime for systemically important nonbank financial firms would be a huge step in that direction, two big obstacles remain: cross-border banking and the too-many-to-fail problem. Efforts to resolve both of these obstacles should be high on the agenda of policy analysts in and out of government. In the meantime, strengthening supervision is an essential, but incomplete and inefficient, substitute for ending too big to fail.

Larry D. Wall is a financial economist and policy adviser in the Atlanta Fed's research department. He thanks Jerry Dwyer for helpful comments. The views expressed here are the author's and not necessarily those of the Federal Reserve Bank of Atlanta or the Federal Reserve System.

_______________________________________

1 For example, Finland and Sweden provided blanket guarantees, and they, along with Norway, injected capital during the Nordic financial crisis in the early 1990s (see Sandal 2004). Similarly, in the United States two very large government-sponsored enterprises (GSEs), Fannie Mae and Freddie Mac, received massive capital injections in September 2008 to keep them from failing. This support of Fannie Mae and Freddie Mac occurred despite the lack of a preexisting bailout fund and despite federal legal requirements that their securities include language indicating that they are not guaranteed by, or otherwise an obligation of, the federal government.

2 Admittedly, policymakers will sometimes unnecessarily follow too big to fail policies when the financial system would have functioned reasonably well without the help. One reason they may do so is that the decision to follow too big to fail policies necessarily depends on forecasts of what might happen, not the certain knowledge of what will happen. Another reason is that costs of being wrong are asymmetric with much higher costs associated with not implementing too big to fail policies when they are needed.

3 See Lastra (2007) on disparities in resolution methods. See Mayes, Nieto, and Wall (2009) on some of the issues in apportioning costs of cross-border resolution and the resulting control rights.

4 Even in Continental's case the resolution was handled in a manner that protected all creditors from loss, perhaps in part because of concerns about potential runs on other large banks that had been weakened by problems with their loans to developing countries. See FDIC (1997, chap. 7) for a discussion of the Continental Illinois failure. See Saul (1993) for a discussion of the lessons from Drexel's failure.

References
Acharya, Viral V., and Tanju Yorulmazer. 2007.Too many to fail—An analysis of time-inconsistency in bank closure policies. Journal of Financial Intermediation 16, no. 1:1–31.

Dwyer, Gerald P. 2010. Too big to fail. Notes from the Vault, February. www.frbatlanta.org/cenfis/publications/vn_too_big_to_fail.cfm (April 2, 2010).

Federal Deposit Insurance Corporation. 1997. History of the 80s: Volume 1: An examination of the banking crises of the 1980s and early 1990s. www.fdic .gov/bank/historical/history/vol1.html (March 9, 2010).

G-20. 2009. Leader's Statement: The Pittsburgh Summit, September 24–25. www.g20.org/Documents/pittsburgh_summit_leaders_statement_250909.pdf (December 24, 2009).

Lastra, Rosa Maria. 2007. Cross-border bank insolvency. Law in Transition online, October. www.ebrd.com/pubs/legal/lit072h.pdf (March 3, 2010).

Mayes, David G., Maria J. Nieto, and Larry D. Wall. 2008. Multiple safety net regulators and agency problems in the EU: Is prompt corrective action partly the solution? Journal of Financial Stability 4, no. 3:232–57.

Sandal, Knut. 2004. The Nordic banking crises in the early 1990s—resolution methods and fiscal costs. In The Norwegian banking crisis, edited by Thorvald G. Moe, Jon A. Solheim, and Bent Vale. www.norges-bank.no/upload/import/publikasjoner/skriftserie/33/hele_heftet.pdf (March 18, 2010).

Saul, Ralph S. 1993. Drexel. Brookings Review 11, no. 2:41–45.

Wall, Larry D. 2010. Too big to fail after FDICIA. Federal Reserve Bank of Atlanta Economic Review 95, no 1:1–16. Reprinted from Economic Review, January/February 1993. www.frbatlanta.org/pubs/economicreview/er10n1_wall.cfm