What Do We Know about Job Loss in the United States? Evidence from the Displaced Workers Survey, 1984-2004

摘要: Introduction A defining characteristic of the U.S. labor market is its fluid nature. Half all new jobs (worker/employer matches) end in first year and, at any point time, about 20 percent workers have been with their current employer for less than one (Farber, 1999a). (1) This fluidity allows rapid reallocation across sectors response to demand shifts, and relatively small direct costs employers laying off encourages hiring face uncertain future demand. Rates employment growth dwarfed those Western Europe no measure because firms shedding compared counterparts European Union. However, this flexibility can impose substantial on who lose jobs. My goal here characterize level job loss losers over 1981-2003 period look changes both cyclical secular, types borne by various losers. Perhaps most comprehensive source information incidence United States Displaced Workers Survey (DWS), administered every two years since 1984 as a supplement Current Population (CPS). In article, I incorporate data from latest (January 2004) DWS, covering through 2003. (2) The tight 1990s saw dramatic reduction civilian unemployment rate an average 7.3 1980s low 4.0 2000. annual rose steadily 2001 recession beyond, reaching 6.0 Job worker displacement are particular concern period, given perceived continuing general weakness market, fears due import competition outsourcing jobs, There three important issues measurement interpretation that arise when compares rates calculated using DWS time. First, asks only single involuntary loss. survey does not capture multiple losses same worker. Neither it terminations "for cause." meant result business decisions unrelated performance employee (for example, plant closing, layoff, abolition job). Thus, calculate fraction lost least cause" relevant rather destruction worker-employer matches. From 1992, asked separations previous five years, while later separation years. use adjusted account change recall so reported three-year basis. adjustment detailed Farber (1997). basic wording key questions has changed inception 1984. may affected whether respondents would report circumstance but another year. (1998) (2004), additional debriefing 1996, 1998, 2000 investigate how question likelihood change. results analysis re-weighted present study. Based computed, adjusting question, find followed roughly pattern between 1981 2004. …