This paper makes an attempt at testing the external validity of corruption experiments by moving from the lab in a developed country, to where it matters the most, the field in a developing country. In our experiment a candidate proposes a bribe to a grader in order to obtain a better grade. We find the direction and the magnitude of most treatment effects to be statistically indistinguishable between the lab and the field. In particular, increasing the graders'.wage reduces in both environments the probability to accept the bribe. Finally, we identify several micro-determinants of corruption (age, religiousness, ability).