Open Collections

Abstract

Chapter 2 of this thesis studies the testable content of models of expectations-based reference-dependence. The main results of this chapter characterize a model based on Kőszegi and Rabin's (2006) preferred personal equilibrium. This model is shown to be behaviourally equivalent to a version of the shortlisting model of Manzini and Mariotti (2007) in environments without risk. Environments with risk motivate novel axioms that are conceptually consistent with expectations-based reference-dependence. These axioms are shown to behaviourally characterize a restricted version of the preferred personal equilibrium model of decision-making. Additional results characterize the choice behaviour of Kőszegi and Rabin's (2006, 2007) personal equilibrium and choice-acclimating personal equilibrium concepts. In the presence of background risk and under the Reduction of Compound Lotteries axiom, non-expected utility preferences cannot capture descriptively reasonable risk aversion over small stakes without producing implausible risk aversion over large stakes (Safra and Segal 2008). Motivated by experimental evidence, Chapter 3 assumes that compound lotteries are evaluated recursively. The main results of this chapter show that non-expected utility theories generate 'as-if' narrow bracketing over small-stakes gambles despite defining utility over final wealth, and can be consistent with empirically reasonable risk aversion over both small and large stakes. Chapter 4 uses list elicitation with varying probabilities to experimentally study choices between lotteries for a population of online workers. We document that list elicitation significantly diminishes risk aversion compared to binary choice elicitation. We show that this observation is consistent with a decision maker who has non-expected utility preferences, but when list elicitation is employed, reduces the compound lottery induced by her choices and the external randomization device used to determine payment. As a result, list elicitation distorts the inferences that can be drawn about non-expected utility preferences.