Replication studies are considered a hallmark of good scientific practice. Yet they are treated among researchers as an ideal to be professed but not practiced. To provide incentives and favorable boundary conditions for replication practice, the main stakeholders need to be aware of what drives replication. Here we investigate how often replication studies are published in empirical economics and what types of journal articles are replicated. We find that from 1974 to 2014 less than 0.1% of publications in the top-50 economics journals were replications. We do not find empirical support that mandatory data disclosure policies or the availability of data or code have a significant effect on the incidence of replication. The mere provision of data repositories may be ineffective, unless accompanied by appropriate incentives. However, we find that higher-impact articles and articles by authors from leading institutions are more likely to be subject of published replication studies whereas the replication probability is lower for articles published in higher-ranked journals.