Writing back in 1873, Walter Bagehot made the observation that since “our credit system [is] much more delicate at some times than at others .. panics come according to a fixed rule, [so] that every ten years or so we must have one of them”.
The dramatic events of the last two years have forced us to reconsider some of the fundamental questions that preoccupied Bagehot and his contemporary financial economists: is a periodic crisis an inherent property of a competitive financial market? Is financial crisis a market failure? If so, what kind of policies should be implemented in order to diminish the social cost? Lastly, is neoclassical economics capable of providing a framework for the analysis of these questions?
There are three major approaches to the analysis of financial crisis. From McKay’s (1841) “extraordinary popular delusions” to Shiller’s (2000) “irrational exuberance” there is a view that financial crisis results from a black-out of human reason. We shall not consider this view here, if only because it is hard to square with a proper welfare analysis, which is one of our main objectives. Then, there is a view, initially formulated by Diamond and Dybvig (1983), followed by Morris and Shin (2004) and many others that crises result from a coordination failure, usually associated with a (defective) financial structure that has a built-in first-mover advantage, like demand deposits, or loss limits. Lastly, there is Fisher’s (1933) debt-deflation theory that builds on “two dominant factors, namely over indebtedness to begin with and deflation following soon after” (emphasis in the original text). Fisher’s own focus on the role of “distressed selling” hints that the main issue is not the nominal rigidity of debt contracts per se, but rather the drop (during crisis) in the relative price of capital goods and industrial output relative to the value of corporate debt, which drains away companies’ equity.
In the autumn of 1998, the US government managed, by sponsoring the rescue of LTCM, to stop the Asian-Russian financial crisis from spreading to the developed economies, with hardly any cost to tax-payers. That episode strengthened the view that financial crises may be a pure coordination failure. Unfortunately, such an optimistic scenario failed to materialize recently. Governments have committed large amounts of resources with the intention of stabilizing the markets, but the effectiveness of the policy is still debated. This strongly suggests that coordination failures alone cannot provide a complete explanation of financial crises. Rather, Fisherian elements such as high levels of indebtedness, fire sales and depressed prices of capital goods seem to be important in understanding financial crisis. At the same time, it seems that multiple equilibria should play a role in a theory of financial crisis, perhaps for cases where the magnitude of the shock is relatively mild.
We believe that many of the theoretical building blocks that are necessary in order to construct a satisfactory theory of financial crisis have already been worked out. In this paper we attempt to move in the direction of an operational, macroeconomic model that integrates theories of liquidity provision Alan and Gale (1998), Gorton and Huang (2004), Holmstrom and Tirole (1998) with theories that explain how depressed capital-goods prices may adversely effect macroeconomic activity, as in Kiyotaki and Moore (1997) or Goodhart et. al . (2004, 2006). By “operational” we mean an integrated macroeconomic corporate finance model, where financial contracts have a “realistic” interpretation and where most of the parameters can be identified by the financial statistics so that numerical examples can provide some idea about the quantitative fit of the model. Equally important, the model’s measure of social welfare should be easily interpreted in terms of standard policy targets such as GNP.
Download
PDF Ebook Liquidity, Contagion and Financial Crisis
