Wages, Workers and Vacancy Durations: Evidence from Linked Data

This paper provides new evidence on the relationship between the duration of a vacancy, the characteristics of workers matched to the vacancy and their re-employment wages, using a novel data set that links vacancy data and matched employer-employee data. Our analysis generates two main findings: <i>First</i>, we decompose re-employment wages into fixed worker and firm effects, and find that there is a strong positive relationship between the duration of a vacancy and the worker wage-effect and robust negative association between the duration of the vacancy and the firm wage-effect. The latter finding is consistent with theories of directed search, where more job seekers apply to more productive firms. <i>Second</i>, we provide evidence that the longer a worker remains unemployed the more likely she is to be matched to a newly available vacancy rather than an existing vacancy. This finding directly supports the main prediction of theories of stock-flow matching and implies negative duration dependence in the set of job opportunities available to unemployed workers.