Unemployment, Unsatisfied Demand for Labor, and Compensation Growth in the United States, 1956-1980

This paper presents two key facts which call into question the value of unemployment rates as barometers of labor market tightness. First, while both unemployment rates and unsatisfied labor demand proxies perform reasonably well on their own in compensation growth equations, in models which include both, only the unsatisfied demand variable appears to matter. Second, the past decade's outward shifts in Phillips plots can to a substantial degree be tied to outward shifts in plots pairing the relevant unemployment rate and unsatisfied demand proxies. The paper also provides results which indicate that Phillips relationships which are defined in terms of unsatisfied demand variables appear to be somewhat more stable than those using unemployment rates. Taken together, our findings have a clear message for those concerned with macroeconomic theory and policy: labor market pressure on wages can be more reliably assessed by looking at measures of unsatisfied labor demand than by looking at the unemployment rates on which most earlier analyses have focused.