The New Output-Inflation Trade-off


Carl E. Walsh

FRBSF Economic Letter 1998-04 | February 6, 1998

One of the hallmarks of economic analysis is the recognition that choice involves trade-offs. Whether it’s a consumer deciding if the roominess of a sports utility vehicle is worth the lower gas mileage, or a firm deciding whether lower wages of an overseas production facility compensate for the lower worker productivity, or Congress deciding whether a new expenditure program justifies the higher taxes needed to finance it, trade-offs must be faced.

One of the hallmarks of economic analysis is the recognition that choice involves trade-offs. Whether it’s a consumer deciding if the roominess of a sports utility vehicle is worth the lower gas mileage, or a firm deciding whether lower wages of an overseas production facility compensate for the lower worker productivity, or Congress deciding whether a new expenditure program justifies the higher taxes needed to finance it, trade-offs must be faced. The same is true in the conduct of monetary policy. Acting too slowly to head off inflation may risk an increase in expected inflation that will make subsequent moves to reduce inflation more costly, while acting too quickly may run the risk of slowing economic growth prematurely.

During the 1960s and early 1970s, many economists and policymakers believed a central bank could achieve permanently lower unemployment by accepting permanently higher inflation. Attempts to exploit such a trade-off to gain the benefits of lower unemployment were, unfortunately, self-defeating. As unemployment fell and inflation rose, individuals began to expect that inflation would be higher. Workers demanded more rapidly rising money wages to compensate for expected price increases, and firms were willing to agree to these wage demands as they expected to be able to pass through their increased costs by raising prices. Rather than remaining stable at a new higher level, the inflation rate continued to increase as long as unemployment remained below the economy’s natural rate. That experience has convinced most policymakers that no such trade-off exists. Instead, most agree that the average level of the unemployment rate and the long-run rate of real economic growth are determined by such fundamentals as technological change, population growth, labor market institutions, and the skills of the work force. These factors are unrelated to the economy’s average rate of inflation, so allowing average inflation to rise brings no long-run benefit in the form of faster growth or lower average unemployment.

This does not mean, however, that central banks do not face unemployment-inflation trade-offs as they implement monetary policy. In fact, recent research in macroeconomics has focused increasingly on an important trade-off involving output and inflation. Unlike the short-run trade-off between the level of output or unemployment and the level of the inflation rate that was a focus of earlier policy debates, the new emphasis is on the choice between the variability of output and the variability of inflation. The research on this variability trade-off suggests that attempting to keep inflation within a very narrow band may increase fluctuations in real output and employment. Conversely, attempts to smooth business cycle fluctuations more actively will lead to wider fluctuations in inflation. The nature of this trade-off, and even whether it really exists, is a subject of debate among economists. This Letter discusses the new output-inflation trade-off and its implications for the design of monetary policy.

Two propositions

Today, most economists and central bankers accept the proposition that there is no long-run trade-off between the rate of inflation and the level of unemployment. At the same time, many believe that policies designed to help stabilize inflation do have real consequences. John Taylor (1996) has summarized current thinking about these issues in the form of two propositions:

The first proposition, about which there is now little disagreement, is that there is no long-run trade-off between the rate of inflation and the rate of unemployment (p. 186).

The second proposition, and there is more disagreement here, is that there is a short-run trade-off between inflation and unemployment. I think that the short-run trade-off is best described in terms of a trade-off between the variability of inflation and the variability of unemployment; that is, in terms of the short-run fluctuations in the variables rather than their levels over time (p. 186).

It is this trade-off between the variability of output and inflation that represents the new policy trade-off. It is easiest to understand why such a trade-off might arise by considering the economic impact of an adverse aggregate shock such as a rise in the price of oil or private sector expectations about future inflation. The direct result of either would be an increase in inflation. If policy acts to bring inflation back on target quickly, inflation will be less variable, but output will fluctuate more around trend. If policy acts more slowly to bring inflation back on target, then output will fluctuate less while inflation becomes more variable. Acting to offset the inflationary impact of supply shocks leads output and unemployment to fluctuate more in the short run, while stabilizing output leads actual inflation to fluctuate more.

Simulation evidence

Much of our knowledge of variability trade-offs comes from simulations of models designed to mimic the behavior of the major industrialized economies. These models incorporate realistic inflation and output adjustment so that they can be used to study the variability trade-off implied by different rules for conducting monetary policy. Fuhrer (1997) provides an example of this type of research, employing a model of the U.S. economy. The evidence from simulations can be used to determine the nature of the volatility trade-off that arises under a particular policy rule and to evaluate alternative policy rules. For example, Taylor (1993) has suggested that recent Fed behavior is characterized by a rule that describes how the federal funds rate is adjusted in response to movements in inflation and the output gap. Using such a rule for determining the funds rate, together with particular values for how much the funds rate is adjusted as inflation and the output gap change, the implied variability of inflation and output can be determined. By then changing how much the funds rate is adjusted in response to inflation and the output gap, a different combination of inflation variability and output variability will be implied. Linking together the different combinations of inflation and output variability, a trade-off emerges. In a similar manner, the frontier associated with a different rule for adjusting the funds rate, such as one that responds to nominal income movements, can be derived. In addition, a change in the nature of the underlying economic disturbances would shift the trade-off frontier; an increase in the volatility of energy prices, for example, would lead to more inflation and output variability.

Figure 1 illustrates the output-inflation variability trade-off for two hypothetical policy rules. The rule that produces the dashed trade-off frontier can be described as inefficient; for any given output volatility, the policy rule that produces the solid line results in lower inflation volatility. Once the efficient trade-off frontier has been found, policymakers then must weigh the relative costs of output variability versus inflation variability in choosing a point on the frontier. If inflation variability is viewed as more costly than output variability, a point such as A might be optimal, while point B would be optimal if the costs of output variability are assessed more highly. This two-step approach, finding the efficient frontier and then deciding which point to pick, is useful in separating two distinct aspects of policy choice. On the one hand, the structure of the economy and the nature of economic disturbances that affect it will define the efficient frontier. On the other hand, the factors that determine which point on the frontier to choose depend on an assessment of the relative costs of different forms of economic variability.

Other evidence?

The notion that focusing more on limiting fluctuations in real output will lead to more inflation variability is fairly intuitive. But does such a variability trade-off actually exist? Simulations of economic models reveal such a trade-off, but economists disagree about which model best captures the true behavior of the economy, and these disagreements mean that there is no consensus about the true trade-off faced by policymakers. It is also difficult to find evidence of the trade-off in the data from actual economies. There are several reasons why the empirical evidence is inconclusive. The chief problem is that each point on the trade-off frontier is associated with a specific way of conducting monetary policy. If policy has been conducted in a stable and efficient fashion over several years, then the observed volatility of output and inflation would provide an observation on a single point on the trade-off frontier. Evidence on just a single point does not provide information on the entire trade-off frontier.

One way around this problem is to look at the experiences of many different countries. If countries have similar economic structures, have faced similar disturbances, and have operated on the efficient frontier, but have differed in the choices policymakers have made between output and inflation stability, then historical patterns of different countries would provide evidence on the output and inflation variability trade-off. Unfortunately, actual economies have different economic structures, have experienced different disturbances, and have conducted policy in different ways. Thus, it is difficult to identify a variability trade-off using the historical experiences of a cross-section of countries.

Recent debates

Evaluating alternative policies in terms of their implications for the trade-off between output volatility and inflation volatility offers useful insights into some recent monetary policy debates. For example, the widely held consensus that monetary policy cannot have permanent effects on the level of the unemployment rate or the rate of real economic growth has led some to advocate that central banks focus only on maintaining low inflation. As a position about the long run, few economists would disagree. But, as Taylor’s second proposition suggests, many economists would argue that a single-minded focus on maintaining inflation within a very narrow band may lead to undesired real economic fluctuations. And conversely, attempts to smooth real fluctuations too actively will lead to excessively volatile inflation.

The variability trade-off is also important for those countries that have moved to an inflation targeting policy regime since it is critical for determining the appropriate width of the inflation target. New Zealand, for example, initially defined its inflation target as 0 – 2% inflation. In 1997, however, this was widened to 0 – 3%. The Bank of England has a target inflation band of plus or minus 1% around its target of 2.5% inflation. The output-inflation variability trade-off is one of the key factors in determining the effects of changing the width of the inflation band. If the trade-off frontier is steep, for example, then reducing the variability of inflation causes little increase in output variability. In this case, a narrow target inflation band would be appropriate. A recognition of the variability trade-off shifts the focus from the level of inflation (which should be low) to questions of how wide the target band should be.

Realizing that the long-run effects of monetary policy determine average inflation, not average unemployment or the economy’s real rate of growth, is critical to maintaining a successful policy. However, central banks still face trade-offs as they balance short-run inflation variability against short-run output variability, and this also needs to be recognized when evaluating the contribution of monetary policy to achieving macroeconomic goals.

Carl E. Walsh
Professor of Economics, UC Santa Cruz
and Visiting Scholar, FRBSF


Fuhrer, Jeffrey C. 1997. “Inflation/Output Variance Trade-offs and Optimal Monetary Policy.” Journal of Money, Credit and Banking 29 (May) pp. 214-234.

Taylor, John B. 1996. “How Should Monetary Policy Respond to Shocks while Maintaining Long-Run Price Stability? — Conceptual Issues.” In Achieving Price Stability, Federal Reserve Bank of Kansas City, pp. 181-195.

_______.1993. “Discretion versus Policy Rules in Practice.” Carnegie-Rochester Conference Series on Public Policy 39, pp. 195-214.

Opinions expressed in FRBSF Economic Letter do not necessarily reflect the views of the management of the Federal Reserve Bank of San Francisco or of the Board of Governors of the Federal Reserve System. This publication is edited by Anita Todd and Karen Barnes. Permission to reprint portions of articles or whole articles must be obtained in writing. Please send editorial comments and requests for reprint permission to