Democratization should reduce incentives to engage in corruption, by expanding the size of the winning coalition, heightening transparency, increasing accountability to the electorate, and multiplying the number of veto points required for a corrupt deal. Yet many young, consolidated democracies, such as South Korea, have recorded higher levels of perceived corruption following democratization.
I argue that the apparently higher level of corruption accompanying democratization results from overdependence on perception surveys to measure corruption. As democratization frees the press, more stories of graft lead the public to higher levels of perceived corruption without any corresponding rise in real corruption. A more effective measurement strategy is to objectively “follow the money,” by focusing on outflows of rents and any related personal receipt of favors by relevant officials. Applied to the Korean case, contrary to popular perceptions of increased corruption, objective measurement of elite corruption reveals that democratization produced a sharp reduction in the corrupt exchange of rents through industrial restructuring programs, bank lending, and government procurement contracts.