Minimum wage impacts on wages and hours worked of low-income workers in Ecuador

Abstract

Abstract Minimum-wage policy aims to raise the real income of low-wage workers. Low-wage individuals may be adversely affected by minimum wages, however, although the empirical evidence on this point is not without controversy. We analyzed the effects of the January 2012 increase in monthly minimum wages on the wages and hours worked of low-wage workers in Ecuador. Individuals could have chosen to enter occupations covered by minimum-wage legislation or those that were not. We applied a difference-in-differences estimation to account for potential self-selection bias. We also relied on exogenous variations in minimum wages by sector, industry, and occupation. We constructed individual panel data from a household panel and performed estimates that also accounted for potential sample-selection bias. The results suggest a significant and positive effect on the wages of treated workers, increasing them by 0.41–0.48% for each 1% increase in minimum wages, relative to the earnings of control workers. Our results also suggest that effects varied by type of worker: (i) women workers received lower wage increases, and their hours worked were significantly and negatively affected, both of which may suggest a failure of the minimum wage to reduce the gender wage gap at the bottom of the distribution, and (ii) the hours worked by young workers were significantly and positively affected, a result that is in agreement with results found elsewhere in the literature. These results persisted after applying robustness checks to account for different control groups, full- vs. part-time jobs, separate regressions for heterogeneous groups, and tests for potential attrition and sample-selection bias. The range of effects observed across disparate groups of workers suggests areas in which policy change could be useful. The income-compression effect we found suggests that further studies should address the effects of minimum wage on the drop in income inequality observed in the data.