In the first chapter, I present a theory of measurement of preference intensity and use this measure as a foundation for utilitarianism. To do this, I suppose each alternative is experienced over time. An individual has preferences over such experiences. I present axioms under which preferences are represented by an experienced utility function equal to the integral of instantaneous preference intensity over time and unique up to a positive scalar. I propose an ethical postulate under which social preferences are utilitarian in experienced utilities. In the second chapter, I consider a classic public good provision problem when the government has the power to tax its citizens. In this environment, participation constraints need not be satisfied. I replace such participation constraints with a weaker condition, which I call no-extortion, that limits the ability of the government to extract funds from its citizens. It is well known that there does not exist any strategy-proof, efficient, and budget-balanced mechanism. In fact, any strategy-proof and efficient mechanism that additionally satisfies individual-rationality or universal-participation fails to raise any revenue in large populations. However, replacing these conditions with no-extortion yields a positive result. There exists a simple, detail-free mechanism that is strategy-proof, efficient, extortion-free, and (asymptotically) budget-balanced in large populations. Furthermore, among all strategy-proof, efficient, and extortion-free mechanisms, this mechanism is undominated and uniquely maximizes ex-post revenue (minimizing any potential, though unlikely, budget deficit). In the third chapter, I propose a general framework with which to analyze the optimal response to crime. Each criminal act, detected with some probability, generates a random piece of evidence and a consequent probability of guilt for each citizen. I consider a utilitarian government with no further moral constraints. In particular, I assume no upper bound on punishment—such a bound can only rise endogenously from the utilitarian objective. No matter how individuals respond to punishment, optimal punishment is non-decreasing in the posterior probability of guilt. Moreover, if individuals respond to punishment in traditionally assumed ways, optimal punishment only punishes upon the realization of the most-incriminating evidence. I consider this a repugnant conclusion, leading me to question the standard assumptions on individual behavior in the context of crime.

Creator

• Fryxell, Loren

DOI

• https://doi.org/10.21985/n2-tjsj-5y25

Subject

• Economic theory

Language

• en

Alternate Identifier

• http://dissertations.umi.com/northwestern:15144
• etdadmin_upload_744314

Date created

• 2020-01-01

Resource type

• Dissertation

Rights statement

• In Copyright