Reducing Corruption Through Crowdsourcing


Inefficiency and corruption typify the delivery of public services in developing countries. In addition, citizen access to regular and transparent information on the specific problems encountered at government offices—which they could use to plan their attempts to acquire service—is rarely available, due in part to substantial personal risks individuals face for reporting corrupt behavior.

Project Vision and Strategy

This project develops and pilots an application to crowdsource information on the performance of Indian government offices, with the goal of improving the overall quality of service delivery. In this project, the team evaluates the impact of an innovative crowd-­‐sourcing application generically called the Citizen Feedback Model (CFM), which aggregates and publicizes user-­‐reported data on corruption and the quality of service delivery at Indian government offices. For each government office in the study, the team creates a score—similar to a restaurant score in Yelp—based on citizens’ reports about the quality of service delivery at that office. They then randomize the exposure of bureaucrats to their office’s score and assess impact on the quality of service delivery and extent of bribe taking, using data from citizen surveys. They will also evaluate the impact of providing citizens with CFM scores via a mobile app, as well as SMS messaging. We hypothesize that improving flows of information about the honesty of bureaucrats will influence the usage by citizens of specific offices and reduce the propensity of bureaucrats to ask for bribes. Our technology provides substantial opportunities to scale the intervention throughout India and contribute a solution to a pressing development challenge.

Lead Researchers