Date of Award

8-2012

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Economics

Major Professor

Christian A. Vossler

Committee Members

William S. Neilson, Michael K. Price, Christopher D. Clark

Abstract

This dissertation consists of three chapters that explore individual choice and behavior. Chapter 1 investigates the incentive properties of advisory referenda using a particular form of NEU theory which replaces the independence axiom assumed in EU theory with two less-restrictive assumptions: betweenness and fanning-out. Betweenness replaces the independence axiom and allows for context dependent risk attitudes. The fanning-out hypothesis then governs the precise way in which risk preferences change given the unique circumstances in which values are elicited. When the assumption of independence is relaxed, an individual's response to an advisory referendum depends on how consequential she believes her response to be, and her beliefs regarding the existence of alternative proposals.

Chapter 2 utilizes laboratory experiments to investigate the behavioral dynamics pertaining to information acquisition and tax evasion. In recent years, a "service" paradigm, whereby tax authorities provide information about correct tax reporting to taxpayers, has shown the potential to further "encourage" correct tax reporting. The results show that the overall effect of a helpful information service is to decrease tax evasion. Further, an audit has the behavioral effect of lowering information acquisition rates and increasing evasion immediately after experiencing a penalty. This effect persists (although diminishes) in subsequent tax reporting decisions.

Chapter 3 builds upon the existing experimental literature regarding ``decoy effects” through an innovative design, which preserved the fundamental features of a consumer choice setting in the laboratory. The design involves choices with financial consequences, real consumer goods, and the ability for participants to opt-out. Through the novel experimental approach and econometric analysis, we demonstrate that the decoy effect is not an artifact of hypothetical settings, the decoy effect is not driven by forced choice, and that decoys do little more than sway individuals at the point of indifference.

Comments

This is the final draft of my dissertation.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS