Inflation and Unemployment

Having developed the AS-AD interpretation of the Keynesian system, it is an easy matter to generate a relationship between unemployment and inflation. Use Okun's law to replace output with unemployment, and then play around with the price variables to turn prices into inflation. What you end up with, an equation that relates the current unemployment rate to the natural rate of unemployment and the excess of actual over expected inflation, is known as the expectations-augmented Phillips curve.
In 1958, A.W. Phillips published a paper showing that money wage inflation and unemployment exhibited a systematic negative relationship in Britain from 1900 to 1957. This relationship was soon found to hold in other countries as well, and also if wage inflation was replaced with consumer price inflation. The empirical regularity seemed logical, and was quickly taken up by policymakers.
In 1968, Milton Friedman and Edmund Phelps, in independent work, argued that the Phillips curve relationship was not a logical one. In fact, the logical relationship was between unemployment and unanticipated inflation. They pointed out that as long as expectations did not change, the empirical Phillips curve would look like a stable relationship, but whenever expectations shifted so would the Phillips curve. Their expectations-augmented relationship predicted in particular that any attempt by central banks to exploit the apparent Phillips curve tradeoff between inflation and unemployment would be self-defeating. Eventually, expectations would adjust until unemployment fell back to its natural rate.
In the first part of this section, we look at the evolution of the Phillips curve from its inception in 1958 until about 1970. The first reading is Milton Friedman's 1968 presidential address to the American Economic Association, in which his ideas were first laid out. The second is a very helpful and concise study of how the ideas behind the Phillips curve evolved.
The second part of this section looks more closely at the natural rate of unemployment. How is it measured? Why does it appear to change over time? And why does it appear to be so different across countries? The main reading here, by Weiner, attempts to understand why the natural rate has been so low in Japan for so many years.
Download transparencies for this section here.
Review questions for this section can be found here.

Detailed Contents
1.  The relationship between inflation and unemployment
      From aggregate supply to unemployment.
     The Phillips curve (1958-60).
     The expectations augmented Phillips Curve. Friedman's presidential address
The evolution of the Phillips curve
An animated gif on the Phillips curve, from Brad DeLong
     The sacrifice ratio.
2.  The natural rate of unemployment (aka the NAIRU)
     Frictional unemployment.
         Determinants of the rate of job finding.
         Determinants of the rate of job loss.
Why is Japan's unemployment rate so low and stable?
     Application: Has the US natural rate declined? In search of the NAIRU
     Efficiency wage theories
     Institutional rigidities (skipped)
     The time-varying NAIRU (skipped)
     Application: Europe's unemployment and hysteresis (skipped).