Minimum Wage Shocks, Employment Flows and Labor Market Frictions

Source: Arindrajit Dube, T. William Lester and Michael Reich, UC Berkeley, Institute for Research on Labor and Employment, Working Paper No. 149-13 July 20, 2013

We provide the first estimates of the effects of minimum wages on employment flows in the U.S. labor market, identifying the impact by using policy discontinuities at state borders. We find that minimum wages have a sizeable negative effect on employment flows but not stocks. Separations and accessions fall among affected workers, especially those with low tenure. We do not find changes in the duration of non-employment for separations or hires. This evidence is consistent with search models with endogenous separations, but explanations focused only on quits or only on layoffs are unlikely to explain the full complement of findings….

…We begin by showing that minimum wages have sizeable earnings impact for these two groups: a 10 percent increase in the minimum wage raises average weekly earnings by 2.2 percent for teens and 2.1 percent for restaurant workers. We find striking evidence that separations, hires, and turnover rates for teens and restaurant workers fall substantially following a minimum wage increase—with most of the reductions coming within the first three quarters of the increase. For a 10 percent minimum wage increase, the turnover rates decline by around 2.0 percent for teens and 2.1 percent for the restaurant workforce. In contrast to our results on employment flows, the minimum wage increases do not affect the
employment stock: our estimated employment elasticities are small in magnitude and not statistically distinguishable from zero for both teens and restaurant workers. In addition, while workers remain at their jobs longer, the same does not appear to be true for time spent between jobs. We do not detect changes in the average duration of non-employment spells for those transitioning in and out of jobs (with the caveat that this variable is measured somewhat coarsely). Finally, for the restaurant workforce, we also do not find any evidence of labor-labor substitution with respect to age or gender…