Academic journal article Economic Quarterly - Federal Reserve Bank of Richmond


Academic journal article Economic Quarterly - Federal Reserve Bank of Richmond


Article excerpt

(ProQuest: ... denotes formulae omitted.)

In the United States during 2008-2009, as in previous episodes here and other countries, supplying funding to financial intermediaries and other firms was a component of the government's response to a financial crisis. Some of these funding initiatives have been characterized-and, in some quarters, heavily criticized-as being bailouts: transfers from the government, made to firms (and sometimes other entities such as city governments) or to their creditors in order to avert insolvency or mitigate its effects, that the recipients are not anticipated to repay. Note that this definition distinguishes bailouts from bona fide government loans.1 Henry Thornton (1802) andWalter Bagehot (1877) explained why it is good public policy for government to lend to firms (particularly to banks) in a financial crisis, and today that justification is widely accepted. Bailouts remain highly controversial, however.

Many economists perceive bailouts to be a costly manifestation of time inconsistency on the part of policymakers. That is, the government threatens that an entity that becomes insolvent must fail rather than being rescued, but subsequently, perhaps out of fear that insolvency would harm many people who bear no responsibility for it, the entity will be rescued when push comes to shove. Anticipating this denouement, the owners and managers of the entity make inadvisable investments, taking risks that they would have avoided if the threat not to assist had been taken seriously. This view, formalized by Kareken (1983) and more recently elaborated by Stern and Feldman (2004), is a cogent, prima facie reason to judge that bailouts are a socially inefficient form of government intervention in the economy.

Nevertheless, despite this logic, numerous academic economists, policymakers, and market participants argued publicly that the 2008-2009 bailout was an indispensable policy action. In their view, there was considerable risk that the economy would have suffered serious, long-term harm if the finance, automobile, and housing industries had not been subsidized. Presumably they were concerned that millions of people would face the sort of immediate harm that expositions of the time-inconsistency argument typically cite, but they spoke of a greater, more persistent harm. In their view, if government did not provide a bailout in circumstances where to do so was vital, then incentives for socially beneficial investment would be impaired in a way that might take decades to repair. This vision is the polar opposite of the time-inconsistency vision, which sees investment incentives being harmed by the occurrence of bailouts rather than by their nonoccurrence.

The goal of this article is to formulate an economic model, in terms of which the concern just described can be understood. This is a very limited goal. It is not even to provide a prima facie argument that conducting a bailout is likely to be good policy. To meet the goal, the model need only establish that a bailout would be economically efficient under some conceivable conditions in some economy that shares salient features of the actual one.

In an economy in which a bailout of firms might be efficient, there must be some reason for production to be undertaken by firms that issue financial claims against which they might default. This feature is necessary because, if there were no good reason for firms ever to become insolvent, then an optimal policy would be to prevent them from ever taking that risk, rather than to allow them to take it and to help them when insolvency occurs.

In particular, besides firms being able to do something for their investors that the investors cannot do for themselves, there must be some constraint on a firm's ability to issue financial claims that would only have to be paid in those states of nature where the firm had the capacity to pay them. The Modigliani- Miller theorem (cf. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.