This paper examines the impact of selected labor market changes on the decline in the unemployment rate in the 1990s. The first section provides an overview of aggregate unemployment trends, inflation, and price and wage Phillips curves. The second section examines the effect of demographic changes on the unemployment rate. The third section examines the impact of the 150 percent increase in the number of men in jail or prison since 1985 on the unemployment rate. The fourth section examines the impact of evolving labor market intermediaries (namely worker profiling by the Unemployment Insurance system and the growth of the temporary help industry) on the unemployment rate. The fifth section explores whether worker bargaining power has become weaker, allowing for low unemployment and only modest wage pressure, because of worker job anxiety, the decline in union membership, or increased competitive pressures. The final section examines the impact of the tightest labor market in a generation on poverty. Our main findings are that changes in the age structure of the labor force, the growth of the male prison population, and, more speculatively, the rise of the temporary help sector, are the main labor market forces behind the low unemployment rate in the late 1990s.
inflation
One of the basic tenets of Keynesian economics is that labor market institutions cause
downward nominal wage rigidity. We attempt to evaluate the evidence that relative wage
adjustments occur more quickly in higher-inflation environments. Using matched individual wage
data from consecutive years, we find that about 6-10 percent of workers experience wage rigidity
in a 10-percent inflation environment, while this proportion rises to over 15 percent when inflation
is less than 5 percent. By invoking a simple symmetry assumption, we generate counterfactual
distributions of wage changes from the distributions of actual wage changes. Using these
counterfactual distributions, we estimate that, over the sample period, a 1 percent increase in the
inflation rate reduces the fraction of workers affected by downward nominal rigidities by about
0.5 percent, and slows the rate of real wage growth by about 0.06 percent. Using state-level data,
the analysis of the effects of nominal rigidities is less conclusive. We find only a weak statistical
relationship between the rate of inflation and the pace of relative wage adjustments across local
labor markets.