Electoral Risk and Vote Buying, Introducing Prospect Theory in the Experimental Study of Clientelism

Most traditional theories of clientelism assert that parties in need of securing electoral support invest in vote buying. We consider this framework limited because it assumes that losses and gains affect a party's decision-making process in a comparable way, and because it assumes that the decision-making process of clientelist political parties focuses only on absolute levels of utility while overlooking changes in outcomes with respect to a reference point. We hypothesize that parties are risk-averse in the domain of gains and risk-seeking in the domain of losses—i.e., losing an election hurts more than winning an election pleases. Unlike traditional theories of clientelism, we argue that vote buying is most likely when parties are probable winners or have experienced important losses in the past. After formalizing a theory of vote buying, we tested it by designing an economic experiment. Exploiting these novel experimental data, we show that prospect theory bridges important unexplained gaps in the literature.