Working Paper: CEPR ID: DP4567
Authors: Jennifer Hunt
Abstract: I study data on bribes actually paid by individuals to public officials, viewing the results through a theoretical lens that considers the implications of trust networks. A bond of trust may permit an implicit quid pro quo to substitute for a bribe, which reduces corruption. Appropriate networks are more easily established in small towns, by long-term residents of areas with many other long-term residents, and by individuals in regions with many residents their own age. I confirm that the prevalence of bribery is lower under these circumstances, using the International Crime Victim Surveys. I also find that older people, who have had time to develop a network, bribe less. These results highlight the uphill nature of the battle against corruption faced by policy-makers in rapidly urbanizing countries with high fertility. I show that victims of (other) crimes bribe all types of public officials more than non-victims, and argue that both their victimization and bribery stem from a distrustful environment.
Keywords: corruption; crime; networks
JEL Codes: D60; K40; O10
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Trust networks established among long-term residents (Z13) | reduce bribery (H57) |
| Victims of crime (K42) | more likely to bribe officials (H57) |
| Presence of peers of similar age (C92) | lower bribery rates (H57) |
| Crime (K42) | bribery (H57) |