You Say You Want a Revolution - A Model on the Implications of Prospect Theory on Voting Behaviour

Abstract: In this thesis we create a voting model with a prospect theory foundation, introducing reference-dependent risk preferences, loss aversion, and probability weighting in the individual utility function. Classical voting models represent perfectly rational voters, we argue that voter behaviour can be better modelled by introducing prospect theory preferences, especially recent political trends. We consider a voting strategy for voters, and a campaigning strategy for parties; we find stable aggregate equilibria in a stochastic setting where the policy platform will yield either a positive or a negative outcome, with a 50-50 chance. We show that these equilibria depend on the distribution of the voters, on the degree of risk aversion, on the valuation function and on probability weighting. We explore the dynamics of our model, where we allow our variables to evolve over time. We find that when expectations outpace reality, disruption becomes inevitable. An extension is explored where we vary probabilities, finding that voters prefer disruption to the status quo even if disruption represents a less-than-fair bet. We show how introducing prospect theory characteristics into the decision making of voting can help explain the success of disruptive parties and shed light on some empirical anomalies in recent political trends.