Situational Support for Corruption: A Two-Part Field Experiment on Collusive vs. Extortive Bribery

This paper reports a two-part field experiment study that demonstrates that negative externality primes commonly used in anticorruption campaigns and academic research on collusive corruption are more effective in reducing participation in corruption among the minority of people who already reject the private-public trade-off inherent in collusive corruption than they are in changing the behavior of those who are tolerant of it. Using a random sample of 1,403 residents of Nairobi County, Kenya, we first administered an in-person vignette experiment that measures respondents’ attitudes regarding collusive bribes compared with more coercive types. Collusive bribes are judged more acceptable across multiple outcome measures because they provide the greatest private benefits, but collusive bribes are also rated as the most unethical and at least as harmful to public goods as coercive payments. Several days later, we administered a behavioral experiment to the same respondents, which provided an opportunity to collude with a corrupt representative of the research team to receive illicit additional compensation for participating in the study. A negative externality prime was effective in reducing acceptance of the corrupt opportunity among respondents who rejected the trade-off between self-interested benefits and harming public goods in the vignette experiment. A concluding discussion addresses the high overall acceptance rate of the collusive offer and suggests directions for future interventions. Lastly, we generalize our results to a middle-range theory of organizational misconduct involving transactions between corrupt agents and the public.

Funding: This work was supported by the University of South Carolina Sonoco International Business Department’s Center for International Business Education and Research (CIBER) [Grant P220A220013-23], received from the U.S. Department of Education.

Supplemental Material: The online appendix is available at https://doi.org/10.1287/orsc.2024.18659.