The theory was quite simple - if unemployment fell below some level (called the natural rate of unemployment), inflation would start to rise. This theory, called the Phillips Curve, has filled economic textbooks for more than 25 years. Unemployment, however, has fallen since the 1990s with no sign as yet of an increase in inflation. When unemployment first started to fall, with no pickup in inflation some economists talked of a shift in the natural rate of unemployment. As this trend has continued, some are questioning whether we need a new theory.

Historically, the relationship between unemployment and inflation seemed to be a useful tool for economists. For example, when the unemployment rate fell to 5.3 percent in 1989, economists predicted that inflation would increase. Inflation, which had averaged 2.4 percent between 1984-87, did jump to 4.1 percent in 1990. Tighter monetary policy, oil price shocks and the Gulf War raised unemployment to 7.6 percent in June 1992 and, in accordance with the theory, inflation fell. The Phillips Curve theory was not sufficiently precise to enable economists to accurately forecast the response in either unemployment or inflation. Part of the imprecision involved uncertainty about the level of the natural rate of unemployment.

As the recovery from the 1990-92 recession proceeded, unemployment started to fall - to 6.1 percent in 1994 and to 4.2 percent by 1999 and so did inflation. While economists held different opinions as to the level of the natural rate of unemployment, virtually no one believed it was 4.2 percent. What was happening to the great tradeoff? Some economists felt special circumstances, i.e., a decline in health care costs, the widespread use of computers; etc. had temporarily suspended the unemployment / inflation tradeoff. Others felt that the natural rate had in fact declined due to productivity growth and a better matching of jobs and vacancies.

Many economists now believe that the relationship between unemployment and inflation is far more complex than first believed. Evidence suggests that the tradeoff has not worked well outside the United States and even in the United States the relationship was neither very strong nor predictable. It is clear that the economics profession needs to develop a more sophisticated theory to explain inflation.

(Updated April 1, 2000)