Academic journal article Contemporary Economic Policy

Inflation History and the Sacrifice Ratio: Episode-Specific Evidence

Academic journal article Contemporary Economic Policy

Inflation History and the Sacrifice Ratio: Episode-Specific Evidence

Article excerpt


Monetary authorities often face the task of controlling the rate of inflation. A typical example is the disinflation policy of the early 1980s in the United States. Reducing inflation, however, usually has some cost. The magnitude of that cost is often summarized by the sacrifice ratio, the percentage of a year's real gross domestic product (GDP) that must be forgone to reduce inflation by 1 percentage point. Policy makers would like to know how high the cost of a disinflation will be when they commence reducing inflation.

Economists have suggested a wide range of determinants of the sacrifice ratio. In this study, we focus on the initial level of inflation and propose a new measure (inflation history measured using a geometric lag) that better captures the inflation environment at the start of a disinflation. This inflation history variable allows us to reconcile the contradictory evidence from episode-specific and cross-country studies.

Ball, Mankiw, and Romer (1988) examine the effect of the average rate of inflation on the short-run trade-off between output and inflation in a cross-country study. They show that higher average inflation increases the frequency of price adjustment and therefore makes the Phillips curve steeper. Low-inflation countries will have a relatively flat Phillips curve and a large sacrifice ratio, while high-inflation countries will have a steep Phillips curve and a small sacrifice ratio. (2) This relationship appears to breakdown when examining the episode-specific studies, which have found an insignificant relationship between average (initial) inflation and slope of the Phillips curve. (3)

The purpose of this article is to reconcile Ball's (1994) episode-specific results with Ball, Mankiw, and Romer (1988) cross-country results. Ball et al.'s finding implies that the Phillips curve faced by policy makers depends on the average rate of inflation and that the slope of the Phillips curve changes when the average rate of inflation changes. However, the fact that the effect of inflation on the Phillips curve slope cannot be found at the episode-specific level makes it difficult for policy makers to know the potential costs of a disinflation.

This article finds a negative relationship between the past inflation history and the sacrifice ratio from episode-specific analysis and helps to reconcile the previously inconsistent findings. We measure inflation history using a geometric lag model of inflation. Including the inflation history variable along with traditional determinants of the sacrifice ratio, we find that the higher historical inflation has been the smaller the sacrifice ratio is during a disinflation.

The article is organized as follows: in Section II, we construct a variable of inflation history with a geometric lag model. In Section III, we identify disinflation episodes in annual and quarterly data for 17 Organization of Economic Co-operation and Development (OECD) countries and calculate the sacrifice ratios. We show that the sacrifice ratio depends on inflation history in Section IV. Section V checks our findings by using alternative measures of the sacrifice ratio. Finally, Section VI concludes.


A. Motivation

Ball, Mankiw, and Romer (1988) show that trend inflation influences the output-inflation trade-off. Their finding has a practical implication for the conduct of monetary policy: by looking at the average rate of inflation, policy makers can improve their prediction of the cost of a disinflation. In countries with low inflation, the short-run Phillips curve will be flat and the sacrifice ratio will be large. In contrast, in countries with high inflation, the short-run Phillips curve will be steep and the sacrifice ratio will be small.

While the cross-country analysis strongly supports the prediction that higher inflation makes the short-run Phillips curve steeper, many researchers have asked if the same result holds in episode-specific analysis. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.