A Unified Approach to Measuring u*

摘要

This paper bridges the gap between two popular approaches to estimating the natural rate of unemployment, u_t*. The first approach uses detailed labor market indicators, such as labor market flows, cross-sectional data on unemployment and vacancies, and various measures of demographic changes. The second approach, which comprises reduced-form models and dynamic stochastic general equilibrium models, relies on aggregate price and wage Phillips curve relationships. We combine the key features of these two approaches to estimate the natural rate of unemployment in the United States, using both data on labor market flows and a forward-looking Phillips curve linking inflation to current and expected deviations of unemployment from its unobserved natural rate. We estimate that the natural rate of unemployment was about 4.0 percent toward the end of 2018 and that the unemployment gap was roughly closed. Identification of a secular downward trend in the unemployment rate, driven solely by the inflow rate, facilitates the estimation of u_t*. We identify the increase in labor force attachment of females, the decline in job destruction and reallocation intensity, and the dual aging of workers and firms as the main drivers of the secular downward trend in the inflow rate.