An Economist's Perspective on the Fed

Simon Research Highlights-February 1997

Economic Policy Issue

This issue of Simon Research Highlights is devoted exclusively to a speech given by Thomas F. Cooley, the Fred H. Gowen Professor of Economics, as part of the Simon School's 18th annual Economic Outlook Seminar, held on December 9, 1996, in Rochester, New York. Cooley is director of the School's Bradley Policy Research Center, which sponsors the Shadow Open Market Committee and the Shadow Securities and Exchange Commission. He is the author of several papers on monetary economics and the business cycle. His recent research has examined the Social Security system.

In his speech, Cooley argues that the one economic variable which the Federal Reserve truly can control is inflation. He states that economists are nearly in unanimous agreement that the Fed has succeeded over the several years at keep-ing inflation in check. He urges the Fed to now cap-italize on that success and set its sights on a goal of zero inflation.

An Economist's Perspective on the Fed

We are now in the 68th month of an economic expansion that began in March of 1991. To date, it is the third longest post-war expansion. The euphoria is such that The Wall Street Journal recently featured several articles that speculate about the possible end of business cycles. That would be terrific, if true, but dare I say this might be a bit of "irrational exuberance"? I have to point out that the demise of business cycles has been prematurely pronounced a number of times in the past. In the spring of 1929, the U.S. economy was in the midst of a long expansion, the stock market was booming, and the financial press was congratulating the Federal Reserve for finally doing a good job of managing monetary policy. We all know what happened later that year.

These same articles that speculate about the end of business cycles also argue that a lot of the credit for the current expansion should go to Alan Greenspan and the governors of the Federal Reserve System. Similarly, a recent New York Times survey of business executives showed that 67 percent of them think the Fed is the institution most responsible for the state of the economy. And if you read the financial press in the weeks leading up to a meeting of the Federal Open Market Committee (an event which happens every six weeks), there is enormous speculation about what the Fed is likely to do or should do. Indeed, prior to the September meeting, the Financial Times of London and its German equivalent both said that financial markets in their respective countries were anxiously poised, trying to Figure* out what the Fed would do! On Monday of that week, The Wall Street Journal reported that the Fed was likely to leave rates alone because there were few signs of inflation. On Tuesd

Given that the Fed is thought to have such a profound influence on the economy, it is not surprising that its deliberations are subject to so much scrutiny.

What I'm going to talk about this afternoon is what the Fed can do and what the Fed should do. It is clear that, by virtue of the fact that so many people believe the Fed has a big influence over the economy, it does have a de facto influence, in that people spend so much time and energy speculating about what the Fed will do. But to an economist who studies monetary policy, this is a bit surprising since our view of the Fed is that it has much less power over real economic activity--the production of goods and services--than most people attribute to it. What I'm going to present is the way an academic economist views monetary policy. You will find that it is a little different from what you read in the press.

There are really three issues that are the focus of most current discussion of Fed policy:

ï Should the Fed respond to the pressures to manage the economy?

ï Should the Fed lower interest rates to try to stimulate growth?

ï Should the Fed target zero inflation?

Managing the Economy The meetings of the FOMC have long been the subject of intense scrutiny by the financial press and there is every reason to expect that they should be. In recent years the scrutiny has become more intense and the reaction of financial markets seemingly more extreme. Do we need any better example than what happened last week, when a fairly innocuous rhetorical question by Alan Greenspan caused the Tokyo market to fall by 3 percent, the London market to fall by 2 percent and the U.S. market to open down 140 points?

What is slightly disturbing about the current climate is the expectation that, if the economic news is too good, if unemployment is too low or the economy is growing too quickly, the Fed must act. This "good news is bad news" mindset is the result of the mistaken belief in a trade-off between unemployment and inflation.

The financial press and many observers still believe the Fed can manage the economy, that it should react when the economy is growing too quickly, and that actions of the Fed can slow the economy down. Where is the evidence in support of this view? The search for channels by which monetary policy affects the real economy, and for compelling evidence that it does, has been the Holy Grail of monetary economics for decades. Never-theless, the empirical evidence is at best summarized as very fragile. My reading of it is that monetary policy doesn't have much of an impact on real economic activity under normal circumstances. Robert Lucas of the University of Chicago, who was last year's Nobel Laureate in Economics, wrote in his Nobel lecture: "So much thought has been devoted to this question and so much evidence is available that one might reasonably assume that it had been solved long ago. But this is not the case..."

One type of evidence that I have found persuasive is data from studies which relate the independence of central banks to their ability to control the price level, employment, growth and real interest rates. Let's look at some evidence based on a cross-section of countries.

Figure* 1 shows the average inflation rate over the period from 1959 to 1990, plotted against a measure of the independence of the central bank. This relationship is important because, unlike the Fed, many central banks are not independent of the government in power. An independent central bank is free to do those things that can be accomplished with monetary policy and is not bound to do the bidding of the government in power.

The very clear downward sloping pattern that emerges from these data illustrates that the more independent the central bank is, the lower the average rate of inflation. Banks that are free to try to control inflation seem to do so, and they achieve a lower variance in inflation as well, as is illustrated in Figure* 2.

But these banks are free to pursue other goals as well. What is the evidence that they achieve them? We see from the next several Figure*s that there is no corresponding pattern in the data for unemployment or for growth. The data show no pattern that could be interpreted as a relationship between, say, growth and the independence of the bank.

This is an argument that the Fed cannot succeed at influencing the real economy in the long run as the financial markets seem to expect. So the question remains: Why do Alan Greenspan and the members of the FOMC continue to act as though they can? One answer might be that they believe they can influence the economy in the short run. Alternatively, they may simply be trying to act as markets expect them to act, so as to behave in a predictable way and thereby preserve their independence--something that is periodically threatened.

The Growth Debate Another element of recent criticism of Fed policy has been the claim articulated by Lester Thurow, Felix Rohaytn, and the writers at Business Week magazine, among others, that the Fed is holding back the growth of the economy. The argument goes roughly as follows: Productivity increases have been much greater than those measured in official Government statistics and inflation is overstated. Therefore, official statistics overstate the inflationary pressures associated with recent episodes of robust growth. From this, these authors conclude that the Fed could ease its policy (lower interest rates) and thus stimulate growth without adding to the inflationary pressures.

This argument is pretty silly. First, the same set of arguments implies that we probably are growing faster but not measuring this growth very well. Second, there is no evidence to support the idea that Fed policy influences the long-term growth rate of the economy. Again, I point to the fact that there is no observed relationship between central bank independence and economic growth. Further, the presumed relationship between inflation and growth on which this argument is based is not supported by the data.

The Federal Reserve and financial markets do respond to reports of renewed growth and lower unemployment rates by raising interest rates. Interest rates fall when data suggest that the economy is slowing down. One reason for this is that market participants believe the Fed will not reduce interest rates if the economy grows at 21/2 percent or higher. The rationale is that either economic growth causes inflation and higher interest rates or the Fed acts on the presumption that economic growth causes inflation and raises interest rates. It doesn't really matter which. But again, the evidence is inconsistent with this view. As Figure* 5 shows, there is no discernible relationship between inflation and growth among the OECD countries.

Zero Inflation So far I've argued that the Fed can control inflation but cannot affect unemployment or growth. Let me now talk about whether they can do a better job of controlling inflation. The last element of the current discussion about the direction of monetary policy is the debate about pursuing zero inflation.

Virtually all observers agree that the Fed has successfully held inflation in check for the past several years. The FOMC gets high marks for its vigilance and for the fact that it has built a credible reputation for the Fed as being committed to keeping inflation low. But a question arises that has arisen many times before: Why settle for low inflation? Why not zero? This line of thinking has been given new life in recent months by Senator Connie Mack's push for the Fed to adopt an explicit target of price level stability, or zero inflation. More fuel has been added to the fires of this debate by a recent Brookings Institution study which argues that being at zero inflation would be costly because it would make the economy less adaptable.

There are three separate issues here:

ï What are the benefits of zero inflation?

ï What are the costs of moving to (or being at) zero inflation?

ï What is zero inflation?

The benefits of having a stable price level, as contrasted with a stable rate of inflation, are somewhat elusive and controversial. At zero inflation, there is no inflation tax on transactions or holdings of money. All estimates, including my own, which are often cited, suggest that the benefit from eliminating this distortion is very small.

There are indirect benefits that operate through the tax on capital gains, and these may be large. Capital income taxes are the most distorting form of taxation and, to the extent that nominal rather than real capital gains are taxed, inflation raises the effective tax rates on capital income. Most estimates suggest these costs could be very large. Nevertheless, at low levels of inflation, they are not huge.

The argument is sometimes made that a better way to fix this problem would be to eliminate, or at a minimum index, the capital gains tax. In other words, monetary policy should not be used to fix what really amounts to problems with the tax system. But this is not an argument against pursuing zero inflation.

Other benefits of zero inflation are that, with a successful policy, the cost of Government borrowing may be lower since there is now a risk premium for the uncertainty about inflation. In addition, the costs of contracting would be lower if parties didn't have to worry about inflation risk premiums. The ultimate extent of these benefits could be large.

The benefits must be balanced against the transitional costs of going to zero inflation--something known as the sacrifice ratio. This is the amount of output that must be sacrificed to achieve lower inflation. These are costs that arise because most wages are set in advance in nominal terms anticipating some level of inflation. In addition, many goods prices are set in advance in nominal terms.

Because wages and prices are fixed in nominal terms for a period of time, one cost of going to zero inflation might be a loss of employment and output because the real cost of labor and goods would increase. Some estimate these costs to be large, but that seems unlikely in the face of a credible commitment to zero inflation. It is important to remember that the costs would be temporary. The size of these costs would depend on how credible the disinflation policy is.

There has been a recent fuss over a new study suggesting that the costs of being at zero inflation may be large. This argument is completely unconvincing; indeed, it borders on nonsense. It is based on two kinds of evidence. The first set of evidence is based on the knowledge that when we look at data on wages we find that individuals do not experience nominal wage decreases. From this observation, it is concluded that it is very difficult to lower nominal wages in response to shocks to the economy. If it were impossible to cut nominal wages, this would make the economy less adaptable in a world of zero inflation.

But this evidence is meaningless. In a world with positive inflation, i.e., the world from which these data are drawn, one would not expect to see much evidence of nominal wage decreases. I doubt that autoworkers and air-traffic controllers in the early 1980's, or people who have been "downsized" in the recent wave of corporate restructurings, would agree with the conclusion that wages can't fall.

Finally, we need to consider the issue of what zero inflation is. At current inflation rates, we may actually be closer than we think. The reason is that the CPI is known to have built-in biases that tend to overstate the inflation rate. Overstated by how much is another question. The most common estimates suggest that it may be overstated by as much as 1.2 percent. This comes from three sources.

First, the CPI does not adjust adequately for quality changes. This biases the CPI upward by about .6 percent per year. Second, it doesn't adequately account for substitution effects, which represent changes in consumers' market baskets that bias the CPI upward by about .3 percent per year. Finally, there is something called outlet substitution bias, which basically means there is more competition in retailing that is not accounted for. This leads the CPI to overstate the cost of living by about .4 percent per year.

The bottom line is that inflation is the one thing the Fed really can control. We are relatively close to having a stable price level; getting there would not be difficult. It is time for the Federal Reserve to begin a serious discussion, mindful of the biases in the CPI, about targeting zero inflation. In the meantime, we should hope that Alan Greenspan will avoid after-dinner speeches, or, if he does speak, that he will go easy on the wine. And we should hope that the Fed will continue to act concerned without trying to respond to expectations that it should try to manage the real economy.

* Figures are available by e-mailing: [email protected]

Several Weeks after delivering this speech, Cooley was awarded a three-year, $200,000 grant by the National Science Foundation to support his ongoing research on Social Security systems.

Do you have ideas, comments or suggestions regarding this publication?

Please contact us at:

Office of Public Affairs William E. Simon Graduate School of Business Administration University of Rochester 2-341 Carol G. Simon Hall Rochester, New York 14627 Phone: (716) 275-3736 Fax: (716) 275-9331 E-Mail: [email protected]