How Inflation Hawks Escape Expectations Traps

Article excerpt

Why did inflation increase so dramatically from the 1960s to the 1970s? That's a question economists are still debating. One possible theory, however, is that once people started believing inflation would rise, the Fed was forced to validate those expectations by increasing the money supply. Sylvain Leduc discusses this "expectations-trap" hypothesis and uses a direct measure of expectations to see if the theory is consistent with the data.

In the early 1960s, inflation in the U.S. was below 2 percent, but by the late 1970s, it was in double digits. Why the inflation rate increased so much over such a relatively short period is still highly debated. Among the different views, one is particularly controversial. The expectations-trap hypothesis suggests that inflation rose dramatically over that period because the Fed, by projecting a dovish image, painted itself into a corner: For whatever reasons, once the public started believing inflation would rise, the Fed was forced to validate those expectations by increasing the money supply in the economy.

According to this view, doing otherwise would have been too costly. This article discusses the expectations-trap hypothesis, then uses survey data on inflation expectations to see if a sudden rise in that variable could have led to a burst of inflation.

The expectations-trap hypothesis is controversial because it implies that the same set of economic fundamentals, such as industrial production and the unemployment rate, can lead to a drastically different inflation rate, depending on how the public interprets the data and their effects on future inflation. One practical implication of the expectations-trap hypothesis is that it becomes very difficult for theorists and forecasters to predict inflation rates because any inflation rate can be rationalized from a given set of economic fundamentals. The theory could be right or wrong, but in general, it's hard to tell from the data, since we don't know how people will interpret any given piece of economic news.

In this article, I will present an analysis that tries to get around this problem using a data set, maintained by the Federal Reserve Bank of Philadelphia, specifically designed to gather information on expected inflation. By using this direct measure of expectations, we can verify whether the theory is consistent with the data. The empirical analysis will show that the predictions of the expectations-trap hypothesis match the U.S. experience surprisingly well.

Obviously, the economy has changed substantially since the 1970s. The inflation rate has come down dramatically since the end of that decade; it averaged only 2.5 percent a year in the 1990s (Figure 1). Therefore, it may seem that understanding the causes of the inflation run-up of the 1970s would be mainly of academic interest. Yet, this is hardly the case. What triggered inflation to take off has important consequences for policymakers and the conduct of monetary policy today. Has a change in Fed policymaking kept inflation under control since the 1970s? Or has the structure of the economy changed to one favoring low inflation? The 1970s, after all, were much more turbulent than the 1990s. In the 1970s, there were two oil embargoes, and the Vietnam war was still going on. In the 1990s, there was an amazing increase in labor productivity growth. It is certainly easier to control inflation when you are in an environment of fast productivity growth that keeps production costs under control. But if policymaking hasn't changed, and we've just been lucky since the end of the 1970s, the corollary is that inflation can take off again when our luck runs out and economic conditions change. Thus, knowing the causes of the inflation run-up of the 1970s is relevant today.


The empirical work in this article demonstrates that changes in the economic environment are not the only reason for inflation's performance over the past two decades: The conduct of monetary policy must share some of the praise. …