Politicians and their promises in an uncertain world: Evidence from a lab-in-the-field experiment in India

In emerging economies, pro-social policy outcomes may be prevented by bureaucratic inefficiency, capture by elected or non-elected office holders, or by other hurdles. For local citizens, uncertainty about the true cause of such failures often prevails. We study the pro-sociality of politicians' decision-making in a modified dictator game with real politician participants in rural India. In our game, a recipient citizen does not know whether dictator politician capture or bad luck is to blame when receiving zero. Using a 2 x 2 design, we investigate how the combination of two non-monetary instruments affect politician behaviour in this hard to govern environment. The first instrument, a (non-binding) promise, is a commitment device; the second introduces a minimal relational lever between the politician and the recipient. We find that politician-dictator giving becomes dramatically more pro-social, from zero to 50:50-giving, when these two instruments are combined. Our results provide new insights about the scope for norm-based, lowcost mechanisms to tackle governance-related asymmetric information challenges in developing country settings.