By THOMAS F. COOLEY AND GARY D. HANSEN *
Money is incorporated into a real business cycle model using a cash-in-advance constraint. The model economy is used to analyze whether the business cycle is different in high inflation and low inflation economies and to analyze the impact of variability in the growth rate of money. In addition, the welfare cost of the inflation tax is measured and the steady-state properties of high and low inflation economies are compared.
Current controversies in business cycle theory have much in common with the macroeconomic debates of the 1960s. Twenty years ago Milton Friedman and Walter Heller debated the issue of whether “money matters. ” In the ensuing years the methods of business cycle research have changed dramatically but the questions have remained much the same. In particular, the issue of how much money matters is as timely now as it was when Friedman and Heller discussed it. In this paper we take the question of whether money matters to mean three things: does money and the form of the money supply rule affect the nature and amplitude of the business cycle? how does anticipated inflation affect the long-run values of macroeconomic variables? and, what are the welfare costs associated with alternative money supply rules? These are quite different questions and each implies a distinct sense in which money can affect the economy. Herein we describe a model economy that can be used to address these sorts of questions. The setting is similar to one suggested by Robert Lucas (1987) where money is held due to a cash-in-advance constraint. We use it to provide estimates of the welfare cost of the inflation tax and to study the effect of anticipated inflation on the characteristics of aggregate time-series.
Early equilibrium business cycle models were influenced greatly by the monetarist tradition and the empirical findings of Milton Friedman and Anna Schwartz. They were models where unanticipated changes in the money supply played an important role in generating fluctuations in aggregate real variables and explaining the correlation between real and nominal variables (for example, Lucas, 1972). More recently, business cycle research has been focused on a class of models in which fluctuations associated with the business cycle are the equilibrium outcome of competitive economies that are subject to exogenous technology shocks. In these real business cycle models, as originally developed by Finn Kydland and Edward Prescott (1982) and John Long and Charles Plosser (1983), there is a complete set of contingent claims markets and money does not enter. Considering the importance attributed to money in earlier neoclassical and monetarist business cycle theories, it is perhaps surprising that these real models have been able to claim so much success in replicating the characteristics of aggregate data while abstracting from a role for money. This does not imply that money is unimportant for the evolution of real economic variables, but it is true that the exact role for
*W. E. Simon Graduate School of Management and Department of Economics, University of Rochester, Rochester, NY 14627 and Department of Economics, University of California, Los Angeles, CA 90024, respectively. We would like to acknowledge helpful comments from Steve LeRoy, David I. Levine, Bob Lucas, Bennett McCallum, Ellen McGrattan, Seonghwan Oh, Ed Prescott, Kevin Salyer, Tom Sargent, Bruce Smith, three anonymous referees, and participants in the Northwestern University Summer Research Conference, August 1988. Earlier versions of this paper were titled “The Inflation Tax and the Business Cycle. ” The first author acknowledges the support of the John M. Olin Foundation.