The Role of Ethnic Heterogeneity on Corruption: Experimental Evidence from Kenya

Abstract

Using a one-shot sequential-move bribery game, this paper investigates the role of ethnic heterogeneity on corruption. The interest is premised on the high levels of corruption in Kenya coupled with the rising ethnicization of politics. When ethnic identity is made salient, when a potential bribe-giver and a third party are co-ethnic, the bribegiver is significantly less likely to offer a bribe to a non-coethnic bribee. Rather, a bribe-giver is more likely to offer a bribe to a co-ethnic bribee when the third party is also co-ethnic. A possible explanation for this behaviour is the bribe-giver's expectation of ingroup reciprocity from both the bribee and the third party. A bribe-giver anticipates that a bribe offered to a non-coethnic bribee might be more likely to be punished by a co-ethnic third party than a bribe offered to another co-ethnic. Thiswould be consistent with the notion of ingroup reciprocity in the sense that when a bribe is offered to a co- ethnic bribee, even though this hurts the third party, the disutility experienced is somehow less than when the bribe is offered to a non-coethnic bribee. In other words, even though the third party is adversely affected in both instances, there is some solace to be found in the fact that a fellow co-ethnic is benefitting from the bribe as opposed to a non-coethnic. The anticipation by a bribe-giver that the bribee makes a decision to accept or reject a bribe on the basis of ethnic consideration is mistaken since the bribee's decision is purely opportunistic. The bribee's decision is mainly based on the initial endowment and bribe amounts.