This paper analyzes data from the 1999-2000 sweep of the U.K. National Child Development Study to investigate convicted individuals prospects in the labor market. Decomposition analysis makes it clear that convicted workers are in fact discriminated against as compared to non-convicted ones, both in terms of employment and wage. Adopting a simple theoretical model accounting for the problem of partial adverse selection in the hiring process, I show that discrimination can be explained not only in terms of economic stigma, but may also derive from incomplete enforcement in detecting crime and punishing offenders. In fact, while firms may apply economic stigma to recover the expected extra-costs from hiring convicted workers, rationally behaving firms may also need to recover unforeseen extra-costs deriving from the hiring of non-convicted offenders as workers. Further stigma increases with the probability of offenders to commit crimes and with the expected level of extra-costs in hiring convicted offenders, while it decreases with the probability of convicting the offenders.