Some Observations on Monetary Policy

IN RECENT YEARS I have become more perplexed and more skeptical about the use of monetary policy as a tool for very short-run stabilization. At present the profession has one body of analysis that relies upon an elaborated Keynesian approach to outline the link between changes in monetary policy and changes in real economic activity. This linkage runs from open market operations to bank reserves and then through the money supply and other factors to short-term interest rates. Movements in shortterm rates are translated into changes in long-term rates through a variety of channels. Long-term rates in turn provide the link to business investment, the stock market, consumption, and homebuilding. This process has been described in great detail in models such as the FRB-MIT-Penn model. But one must qualify the results of these models by noting that the lags are very long and that many of the features of the outcome are very uncertain. The estimates of the response elasticities are subject to a wide range of disagreement. Some of the links, such as those involving the stock market, involve a great deal of uncertainty and possible interaction with other factors. These models do not lend much appeal to the notion of using monetary policy for precise short-run guidance. An alternative approach attempts to go directly from some monetary aggregate to economic activity, without specifying the details of the linkages. This approach has been illustrated in many papers in recent years. While the results are consistent with those of the structural models in a general sense, I am skeptical about the more detailed implications. In par-