DURABLE INEQUALITIES An Experimental Investigation of Indian Caste

We conducted an experiment in rural North India that shows persistently that publicly revealing individuals’ social identity has significant effects on individuals’ behavior even when that information is unrelated to rewards. These results are of interest because they suggest the possibility that if there is a belief system that stigmatizes members of a particular social group and if individuals are publicly identified as members of that group, their behavior responds by conforming to the belief system. 198 low caste and 198 high caste junior high school male students in rural India performed the task of solving mazes under economic incentives. There were no caste differences in performance when the participants’ names or caste identities were not publicly revealed, but publicly revealing social identities created a large and robust caste gap. When a random draw was introduced to determine who within the group would receive rewards for mazes solved, the caste gap disappeared. The experimental design enables us to exclude socioeconomic differences as explanations of the caste gap in performance.