Demand for domestic help services: Evidence from a natural experiment

Abstract

We investigate how an increase in subsidies for purchasing domestic help services affects the consumption of individuals who need assistance to live at home. Drawing on administrative data, we analyse the impact of a reform implemented in Quebec (Canada) in 2016, which made the program more generous for a subgroup of beneficiaries. For this purpose, we estimate a difference-in-differences lognormal hurdle model. We derive the corresponding average treatment effect on the treated for this class of non-linear models. Our results suggest that the price elasticity of the demand for subsidized domestic help services for the treated is around 0.74. The elasticity of monthly purchase frequency (0.59) is much larger than the elasticity of monthly purchase intensity (0.14). Based on our results, we determine a floor to the marginal external benefit required for the reform to be socially worth adopting.