This study is a registered replication of a field experiment on dishonesty by Azar et al. (2013). Their main finding was that most customers of an Israeli restaurant did not return excessive change; however, customers who received a higher amount of excessive change returned it more often than people who received a lower amount. Our study, which was conducted on a sample of customers of restaurants in the Czech Republic (N=219), replicated the results of the original study. The high excessive change condition increased the chance of returning the excess change by 21.7 percentage points (17.4 percentage points in the original study). The findings show that the psychological costs of dishonesty can outweigh its financial benefits. We similarly found that repeat customers and women were more likely to return the excessive change than one-time customers and men. The majority (70%) of customers in our sample returned the excessive change. We discuss the importance of field studies and replications of them in the further development of research into dishonest behavior.