According to Gibrat’s Law of Proportionate Effect, the growth rate of a given firm is independent of its size at the beginning of the period examined. While earlier studies tended to confirm the Law, more recent research generally rejects it. This paper reconciles these two streams of literature, taking into account the role of market selection and learning in reshaping a given population of firms through time. Consistently with previous studies, we found that Gibrat’s Law has to be rejected ex ante, since smaller firms tend to grow faster than their larger counterparts. However, a significant convergence towards Gibrat-like behavior can be detected ex post. This finding is an indication that market selection “cleans” the original population of firms, so that the resulting industrial “core” does not depart from a Gibrat-like pattern of growth. From a theoretical point of view, this result is consistent with those models based on passive and active learning, and can be seen as a defense of the validity of the Law in the long-run.