In this paper, we present both a theoretical and an empirical model in order to identify the effects of disability on wages. In the theoretical model we assume that the wage gap of a disabled worker depends on a permanent and a transitory productivity gap and the model predicts that the wage gap will be lower after gaining some work experience in the new job. We test this theoretical hypothesis using an exogenous health shock and matching methods associated with treatment effect techniques for policy evaluation. In all our specifications, we find that the reduction of the wage for the disabled is between 293 and 342 euros per month expressed in constant terms at 2010 prices (21-24% of the average wage of disabled workers) but this reduction is more than offset when we take into account both the disability benefits and the wage. As predicted in the theoretical model, we observe that the pay gap between the disabled and the non-disabled individuals falls over time once the transitory drop in productivity is disappearing. However, we observe a constant wage gap that remains over time and that corresponds to the permanent fall in productivity predicted by the theoretical model.