The majority of new jobs in the U.S. is filled by workers coming from employment or from out of the
labor force (inactivity). Yet, because the number of job seekers in these groups is unobserved, they
are often ignored in empirical labor market studies. This paper, instead, uses latent-variable
techniques to estimate the aggregate matching function - a relation between hires, vacant jobs and job
seekers - while considering searchers from unemployment, employment and inactivity. Importantly, the
estimation allows for the (match) efficiency with which these three groups of searchers find jobs to
vary on average and over time. This paper finds that almost half of the rise in U.S. unemployment
during the Great Recession is explained by a drop in match efficiency of the unemployed. This
contrasts sharply with previous studies which found match efficiency to be quantitatively unimportant.
(Copyright: Elsevier)
E24
,E32
,J64
,Matching function
,Mismatch
,Estimation