'Until not much more than 20 years ago, economists frequently lamented the fact that they were limited in their empirical analyses to statistical assessments of market behavior, because controlled economic experiments were (thought to be) infeasible, unethical, or both. Much has changed in the...

We analyse policy makers' incentives to fight corruption under different institutional qualities. We find that 'public officials', even when non-corrupt, significantly distort anti-corruption institutions by choosing a lower detection probability when this probability applies to their own...