In recent years banking crises have afflicted many countries throughout the world. Regulators’ responses to these banking crises have varied widely, ranging from multiple bank closures (Argentina, United States) to widespread bank rescues (Bulgaria, Czech Republic, Hungary, Japan, Norway, Sweden). Why do regulators in some instances apply tough policies to troubled banks and in other instances rescue many of them? This paper argues that the notion of "too-many-to-fail" explains multiple bank rescues. If too many banks in an economy are financially troubled, the social costs of closing all of them may exceed the costs of rescuing them.

Yet regulators may realize that they risk being trapped in a "too-many-to-fail" situation, and they may take measures to eliminate this risk. One such measure is to weaken bank regulation. By weakening regulation (for example, definitions of bank solvency), regulators diminish their own ability to detect troubled banks but simultaneously increase the credibility of the threat to be tough with financially troubled banks that are actually discovered. Becoming soft in regulation can permit regulators to be tough in banking crises. Indeed, early in the U.S. savings and loan crisis, regulators weakened solvency requirements so that fewer banks would qualify as insolvent. In Japan an early regulatory response to the banking crisis was to relax rules for banks to qualify for advantageous tax treatment for loan loss provisions.

Regulators’ responses to banking crises thus depend on a number of factors, including how the banks themselves have handled their bad loans; what type of banking regulation was in place prior to the crisis; and the total number of banks suffering financial distress. Troubled banks often attempt to hide their loan losses by passively rescheduling loans in default. This behavior can worsen a crisis; regulators may not detect the problem until it has become serious or widespread. As the notion of too-many-to-fail suggests, the number of banks suffering from financial distress is also important. Although the well-accepted concept of "too-big-to-fail" can explain rescues of large, individual banks observed in different countries, it cannot explain the simultaneous rescue of several banks. In contrast, "too-many-to-fail" captures the idea that multiple bank closures can generate high social costs—arising from reductions in the output of firms deprived of financing, and from the significant resource costs required to impose tough, case-by-case policies on many troubled banks.

The proposition that regulators may have to become soft in order to be tough implies that it can be less costly for regulators to weaken banking regulations ex ante and to apply tough policies ex post to a smaller number of troubled banks (allowing some troubled banks to go undiscovered) than to detect more troubled banks but be forced to rescue them. This suggests that in emerging market economies and in economies in transition, where the risk of systemic banking crises is high, it may be impossible for regulators to implement stringent banking regulations without running the risk of a bailout of the entire banking system. The outcome may be only a gradual strengthening of banking regulation over time, as the risk of a systemic banking crisis subsides.

Randall K. Morck, David A. Strangeland, and Bernard Yeung, Inherited Wealth, Corporate Control and Economic Growth, WP 209, September 1998.

Steven M. Burgess and Mari Harris, Values, Optimum Stimulation Levels and Brand Loyalty: New Scales in New Populations, WP 210, September 1998.

Janet Mitchell, Bankruptcy Experience in Hungary and the Czech Republic, WP 211, October 1998.

Yordan Staykov, The Marketing System in Bulgarian Livestock Production—The Present State and Evolutionary Processes during the Period of Economic Transition, WP 212, October 1998.

Martina Lubyova and Jan C. van Ours, Effects of Active Labor Market Programs on the Transition Rate from Unemployment into Regular Jobs in the Slovak Republic, WP 213, December 1998.

Milan Vodopivec, Does the Slovenian Public Work Program Increase Participants’ Chances to Find a Job? WP 214, December 1998.

Jochen Kluve, Hartmut Lehman, and Christoph Schmidt, Active Labor Market Policies in Poland: Human Capital Enhancement, Stigmatization or Benefit Churning? WP 215, December 1998.