During the 1960s, it appeared that there was a stable trade-off between the rate of unemployment and the rate of inflation. The Phillips curve, which describes such a trade-off, suggests that lower rates of unemployment come with higher rates of inflation, and that lower rates of inflation come with higher rates of unemployment. But during subsequent decades, the actual values for unemployment and inflation have not always followed the Phillips curve script.
There has, however, been a relationship between unemployment and inflation over the four decades from 1961. Periods of rising inflation and falling unemployment have been followed by periods of rising unemployment and continued inflation; those periods have, in turn, been followed by periods in which both the inflation rate and the unemployment rate fall. These periods are defined as the Phillips phase, the stagflation phase, and the recovery phase of the inflation—unemployment cycle, respectively. Following the recession of 2001, the economy returned quickly to a Phillips phase.
The Phillips phase is a period in which aggregate demand increases, boosting output and the price level. Unemployment drops and inflation rises. An essential feature of the Phillips phase is that the price increases that occur are unexpected. Workers thus experience lower real wages than they anticipated. Firms with sticky prices find that their prices are low relative to other prices. As workers and firms adjust to the higher inflation of the Phillips phase, they demand higher wages and post higher prices, so the short-run aggregate supply curve shifts leftward. Inflation continues, but real GDP falls. This is the stagflation phase. Finally, aggregate demand begins to increase again, boosting both real GDP and the price level. The higher price level, however, is likely to represent a much smaller percentage increase than had occurred during the stagflation phase. This is the recovery phase: inflation and unemployment fall together.
There is nothing inherent in a market economy that would produce the inflation—unemployment cycle we have observed since 1961. The cycle can begin if expansionary policies are launched to correct a recessionary gap, producing the Phillips phase. If those policies push the economy into an inflationary gap, then the adjustment of short-run aggregate supply will produce the stagflation phase. And, in the economy’s first response to an expansionary policy launched to deal with the recession of the stagflation phase, the price level rises, but at a slower rate than before. The economy experiences falling inflation and falling unemployment at the same time: the recovery phase.
In the long run, the Phillips curve is vertical, and inflation is essentially a monetary phenomenon. Assuming stable velocity of money over the long run, the inflation rate roughly equals the money growth rate minus the rate of growth of real GDP. For a given money growth rate, inflation is thus reduced by faster economic growth.
Frictional unemployment is affected by information costs in the labor market. A reduction in those costs would reduce frictional unemployment. Hastening the retraining of workers would reduce structural unemployment. Reductions in frictional or structural unemployment would lower the natural rate of unemployment and thus raise potential output. Unemployment compensation is likely to increase frictional unemployment.
Some economists believe that cyclical unemployment may persist because firms have an incentive to maintain real wages above the equilibrium level. Whether this efficiency-wage argument holds is controversial.