The Role of Monetary Policy in the Current Macroeconomic Environment

William Poole*
President, Federal Reserve Bank of St. Louis

Prepared for Panel on Promoting Economic Growth: What Monetary Policy Can And Cannot Do
National Association For Business Economics 43rd Annual Meeting
New York City
Sept. 10, 2001

* I appreciate comments provided by my colleagues at the Federal Reserve Bank of St. Louis. Robert Rasche, Director of Research, and Kevin Kliesen, Economist in the Research Division, were especially helpful. I take full responsibility for errors. The views expressed are mine and do not necessarily reflect official positions of the Federal Reserve System.

In taking up the topic of this panel discussion Promoting Economic Growth: What Monetary Policy Can and Cannot Do I'm tempted to refer you to Milton Friedman's 1967 Presidential address to the American Economic Association, "The Role of Monetary Policy," and then sit down. But I won't. What I will do is organize my remarks around some of the ideas noted in Friedman's famous lecture.

Friedman's lecture is directly relevant to today's situation because the economy is suffering from a real disturbance in the form of sharp decline in demand for high-tech equipment. To what extent can monetary policy deal with this real disturbance? What are the opportunities and what are the dangers?

Before proceeding, I want to emphasize that the views I express are mine and do not necessarily reflect official positions of the Federal Reserve System. I thank my colleagues at the Federal Reserve Bank of St. Louis for their comments, especially Robert Rasche, Director of Research and Kevin Kliesen, Economist in the Research Division. I retain full responsibility for errors.

Real and Nominal Variables

In his presidential address, Friedman emphasized that monetary policy ultimately only affects nominal variables, such as nominal interest rates and the price level. The effects may be seen in both level form and rates of change. Consequently, the central bank cannot expect to be successful in targeting any real variable. The effects on real variables, such as real GDP growth and the unemployment rate, are transitory in nature and not very predictable.

The way I've stated this proposition is not quite right, because there is ample evidence to support the view that the monetary instability damages economic efficiency and thereby reduces economic growth. The converse of this observation is that monetary stability promotes higher real growth. Moreover, timely policy actions can help to stabilize real activity-that is, reduce the variance of real growth. Friedman was very skeptical that activist policy could be systematically successful; he argued, often and eloquently, that the best we are likely to be able to do is to adopt steady money growth as the policy rule. I believe that we have evidence from U.S. monetary policy since 1982 that it is possible to adjust policy in a stabilizing way. However, I'll simply assert and not argue that point here.

My framework is this. First and foremost, the central bank must maintain a commitment to low and stable inflation. If the central bank does not achieve that goal, no one else can. If inflation comes unstuck, all sorts of other problems will arise. Second, within the confines of the goal of low inflation, the central bank has some flexibility to lean against fluctuations in output and employment. However, the central bank ought not to pursue the goal of stabilizing economic activity so aggressively that it runs any substantial risk of compromising the goal of low inflation.

Finally, in leaning against fluctuations in growth and employment, the central bank ought not to have goals for levels of the economy's growth and unemployment rates per se. Within a wide range, we don't know what the economy's equilibrium rate of growth is, and what rate of unemployment will clear the labor market in the long run. We run the biggest risks of a major monetary policy mistake if we attempt to target the levels of real variables.

The Tech Tumble

Growth in the aggregate economy has slowed to a crawl this year, but the composition of demand has been uneven. Residential structures investment has been pretty strong; consumption growth, though lower than last year, has held up O.K. The tech sector has taken a real tumble.

The allocation of production across various goods is not something the Federal Reserve can control. A couple of years ago, when many farmers in the St. Louis Fed district were suffering from drought, I often heard pleas that the Fed should help agriculture by lowering interest rates. My answer was always, "the Fed cannot make the rain fall." Today, with all the excess telecom capacity, I offer the same sort of reply: the Fed cannot make college kids call home more often.

The economy is working through a period of excess production capacity in information technology and related equipment. Some of the adjustment will take care of itself as demand recovers from temporary weakness. I do not mean to minimize the problem by using the word "temporary." The decline in demand has been large and has been ongoing for about a year now. We have no guarantee that demand will rebound quickly next quarter or the one after. Still, in time, investment in high-tech equipment will recover in the normal course of events.

Besides this cyclical adjustment, however, there may be some longer-run adjustments that will eliminate certain firms. We don't know what business models will work in the Internet world, and with any new technology it takes a while to figure out what models yield reliable earnings over time. The Fed has no way to address the problems of this or any other specific sector of the economy.

What the Fed can do, at least to some extent, is prevent problems in specific sectors from becoming general problems. That is exactly how to view monetary policy this year. Tech investment is down, but housing investment and consumption have been maintained pretty well. Declining interest rates have certainly assisted in supporting aggregate demand.

Cushioning the effects of the tech tumble on other sectors is no mean accomplishment. So far, things have gone reasonably well considering the magnitude of the disturbance. This point is an important one. Given that monetary policy cannot determine the sectoral composition of output, success must be measured not by the speed and extent of revival in high-tech manufacturing but by the performance of the aggregate economy. No one knows for certain whether the eocnomy can escape an actual decline in real GDP, but the fact that we have done so to date and that many adjustments are now well along suggests that we have an excellent chance of doing so. And I say these things despite the dismal employment report last Friday.

Dealing with Uncertainty

If the economic boiler has a hole, with weak tech investment draining away steam pressure, it seems logical to turn up the fire to keep the pressure high enough that the economic locomotive wheels keep turning. But how much should we turn up the fire? Is the tech hole in the boiler growing, or healing itself?

How should policymakers cope with this uncertainty? One of Friedman's important points -that knowledge of how monetary policy affects the real economy is incomplete -requires the Fed to be cautious in responding to current developments. In a 1968 statement that rings as true today as it did then, Friedman said: "We simply do not know enough to be able to recognize minor disturbances when they occur or to be able to predict either what their effects will be with any precision or what monetary policy is required to offset their effects. . . Experience suggests that the path of wisdom is to use monetary policy explicitly to offset other disturbances only when they offer a 'clear and present danger.' "

A great source of strength in our current situation is that the market, as best I can tell, holds rock-solid expectations that the trend rate of inflation will remain low, in the neighborhood of where it has been in recent years. It seems unlikely that economic behavior today and in the near future will be driven by expectations of rising inflation. Still, that fact does not mean that inflation cannot rise. Moreover, inflation could-I am not forecasting that it will-creep up even though demand is not rising vigorously. For example, we've seen substantial wage increases in the airline industry; if airlines are to be profitable over time they will have to recover those costs through some combination of productivity gains and price increases, even though travel demand is not currently strong.

The Fed monitors the economy very carefully. The Beige Book process is a valuable supplement to the formal statistical information we follow. Watching current data, and gathering anecdotal information, helps us to identify changes in economic conditions in timely fashion. Nevertheless, my own conviction is that we should not rely too much on current observations on the state of the economy-both activity and inflation-but watch carefully direct measures of the thrust of monetary policy itself. We have ample evidence from history that the effects of monetary policy actions are stretched out over time, and that those effects are not easy to predict from current price and production data. I believe that there is information in the monetary aggregates on the thrust of monetary policy and that we ignore that information at our peril.

The Market as Ally

Economic policy of all sorts works better when it harnesses market forces, rather than fights with them. Monetary policy is no exception.

In fact, because monetary policy works through the financial markets and because these markets are forward-looking, market anticipations of policy actions play a significant role in the effectiveness of monetary policy. When discussing what monetary policy can and cannot do, the issue of improving the predictability of policy is of great importance. One way to see this point is to imagine that we have already identified a policy rule that is effective in achieving policy goals and highly predictable. Now imagine degrading that policy by adding a purely random component to it. Doing so not only would induce inappropriate policy settings but also would provoke inappropriate market responses because the market would have a more difficult time figuring out the direction of policy. Working to reduce the component of policy that appears to the market to be random and unpredictable will, therefore, pay significant dividends.

If the markets are confident that the central bank will take appropriate action, the timing of that action is not critically important. Bond yields began to fall last year long before the Fed first cut the intended federal funds rate in early January this year. The benefits of last year's declines in long rates in supporting this year's housing and consumer durables expenditures are clear.

For the bond market to contribute importantly to economic stabilization, as I believe it does, Federal Reserve policy must be predictable. Milton Friedman thought that predictability required steady growth in monetary aggregates, but clearly the current system has also proven to be highly predictable. I do not mean to imply that short-term interest rates can be predicted long in advance, for rates should and do respond to events that cannot themselves be predicted. But if the Fed's policy actions in response to events is highly predictable, then the market can respond efficiently to those same events. Further, to the extent that the market can predict certain events better than the Fed can, interest rates will change sooner than the Fed could change them.

Our understanding of this process is incomplete, but I think we've made a lot of progress in recent years. One of the important things that the Federal Reserve can do is to continue to improve the flow of information about what it is doing and why. The Fed took a highly productive step in February 1994 when it began releasing its policy decision promptly at the conclusion of each FOMC meeting. The St. Louis Fed is devoting its October research conference to the issue of Fed transparency, and I'm hopeful that continuing research on this topic will shed new light on how the Fed can improve its communication with the market.

Summary

To summarize my argument, monetary policy is above all responsible for the economy's trend rate of inflation. That is every central bank's central responsibility.

Success in keeping the rate of inflation low and stable yields many dividends. One is that market expectations of inflation remain low and change little over time. That environment provides maximum scope for the central bank to adjust policy to cushion real disturbances-to reduce the variance of real growth and employment. However, it is dangerous for policy to attempt to hit real targets; we know that in the long run, monetary policy determines nominal and not real magnitudes.

In today's economy, we do not know where the unemployment rate will settle after the economy's growth resumes, nor do we know what that growth rate will turn out to be. I am an optimist on long-run growth, but know that uncertainty over the rate is considerable. For that reason, setting monetary policy to achieve a particular rate of growth is hazardous.

Finally, we should not underestimate the importance of the markets in contributing to economic stabilization. The market is inherently monetary policy's friend, and we should do everything possible to improve market understanding of the course of policy.

Back to top