Description:
In this paper, we evaluate a series of minimum wage hikes implemented in the early XXIst century in Argentina using administrative records of registered employment. We identify the effect of raising the minimum wage on job separations via a regression discontinuity design. More specifically, we compare the match destruction rates for a treatment group directly bound by the minimum wage hikes and a control group slightly out of its legal scope. We show that this method represents an improvement over previous ones because it reduces the incidence of Type-I error. We find that, when aggregated, these hikes had a precisely estimated zero effect on separation rates. However, the increases enacted in 2008 arise as an exception. They decreased separations by 4.8 percentage points (19%). These results suggest that the employment effects of minimum wages may not flow through employment destruction.