Previous studies about the effects of regulatory institutions on the outcomes of regulation have resulted in a lack of consensus on the nature of these impacts. This paper seeks to resolve some of this ambiguity by analyzing two dimension of electric utility regulatory outcomes, prices and reliability, with a broader panel of explanatory variables and with a Hausman-Taylor regression technique. The results suggest that elected regulators and deregulated electricity markets result in worse reliability outcomes for consumers without strong evidence that either institution secures lower electricity prices. Incorporating these insights into a theoretical model of regulation could give more detailed insight into how to create regulatory institutions that can optimize the outcomes of governance.

I study split-pie bargaining problems between two agents. In chapter two, the types of both agents determine the value of outside options -- I refer to these as interdependent outside options. Since a direct mechanism stipulates outcomes as functions of agents' types, a player can update beliefs about another player’s type upon receiving a recommended outcome. I term this phenomenon as information leakage. I discuss binding arbitration, where players must stay with a recommended outcome, and non-binding arbitration, where players are not obliged to stay with an allocation. The total pie is reduced if the outcome is an outside option. With respect to efficiency, I derive a necessary and sufficient condition for first best mechanisms. These are mechanisms that assign zero probability to outside options for every report received. The condition describes balanced forces in conflict (outside options) and is the same in the cases of binding and non-binding arbitration. I also show a strong link between conflict and information: when conflict exists, information leakage occurs. Hence, non-binding arbitration may seem more restrictive than binding arbitration. To analyze why this is the case, I solve for second best mechanisms with binding arbitration and find a condition under which they can be implemented under non-binding arbitration. Thus, I show that non-binding arbitration can be as effective as binding arbitration in terms of efficiency. I also examine whether the equivalence between binding and non-binding arbitration can cease to hold, and provide analysis of why this happens. In chapter three, the bargaining problem entails no uncertainty but rather envy. Players can feel envy about the allocation of the other player. The Nash Bargaining solution is obtained in this context and some comparative statics are shown. The introduction of envy makes the more envious party a tougher negotiator.

Prior research on consumer behavior in health insurance markets has primarily focused on individual decision making while relying on strong parametric assumptions about preferences. The aim of this dissertation is to improve the traditional approach in both dimensions. First, I consider the importance of joint decision-making in individual insurance markets by studying how married couples coordinate their choices in these markets. Second, I investigate the robustness of prior studies by developing a non-parametric method to assess decision-making in health insurance markets. To study how married couples make choices in individual insurance markets I estimate a stochastic choice model of household demand that takes into account spouses' risk aversion, spouses' expenditure risk, risk sharing, and switching costs. I use the model estimates to study how coordination within couples and interaction between couples and singles affects the way that markets adjust to policies designed to nudge consumers toward choosing higher value plans, particularly with respect to adverse selection.

Finally, to assess consumer decision-making beyond standard parametric assumptions about preferences, I use second--order stochastic dominance rankings. Moreover, I show how to extend this method to construct bounds on the welfare implications of choosing dominated plans.

I began this thesis because I was confused about economics. I wondered why there were so many different models. I didn't understand how they fit together. I was also confused by the assumptions being made. For instance, the assumption that humans are rational utility-maximizers did not seem to agree with my own experiences. With my director Dr. Edward Schlee's help, my thesis has become an inquiry into the state of economic methodology, both in theory and in practice. The questions that drive this paper are: How do economists choose between theories? What is the purpose of economic theory? What is the role of empirical data in assessing models? What role do assumptions play in theory evaluation, and should assumptions make sense? Part I: Methodology is the theoretical portion of the paper. I summarize the essential arguments of the two main schools of thought in economic methodology, and argue for an updated methodology. In Part II: A case study: The expected utility hypothesis, I examine methodology in practice by assessing a handful of studies that seek to test the expected utility hypothesis. Interestingly, I find that there is a different between what economists say they are doing, and what they actually seem to be doing. Throughout this paper, I restrict my analysis to microeconomic theory, simply because this is the area with which I am more familiar. I intend this paper to be a guide for my fellow students and rising economists, as well as for already practicing economists. I hope it helps the reader better understand methodology and improve her own practice.

I conduct a two-fold study on the relationship between adverse selection and nonlinear pricing in competitive insurance markets. First, I reassess empirical evidence of adverse selection in life insurance with the Health and Retirement Study (HRS) data used by Cawley and Philipson (1999). Specifically, I evaluate the shape of the premium schedule and present indications of quantity premia beyond a certain coverage level. The observed pricing schedule appears like the "backward-S-shaped" curve described by Chade and Schlee (2012); I discuss why this result cannot be entirely explained by fixed costs of underwriting. Second, I critique the arguments against adverse selection in existing literature by modifying the Rothschild and Stiglitz (1976) model of competitive insurance markets. I present several existing models and a new framework to explain how adverse selection and quantity discounts can coexist in equilibrium. These modifications deviate from the standard models of competitive insurance, but produce plausible hypotheses with conclusions contrary to conventional theoretical results.

Economists, political philosophers, and others have often characterized social preferences regarding inequality by imagining a hypothetical choice of distributions behind "a veil of ignorance". Recent behavioral economics work has shown that subjects care about equality of outcomes, and are willing to sacrifice, in experimental contexts, some amount of personal gain in order to achieve greater equality. We review some of this literature and then conduct an experiment of our own, comparing subjects' choices in two risky situations, one being a choice for a purely individualized lottery for themselves, and the other a choice among possible distributions to members of a randomly selected group. We find that choosing in the group situation makes subjects significantly more risk averse than when choosing an individual lottery. This supports the hypothesis that an additional preference for equality exists alongside ordinary risk aversion, and that in a hypothetical "veil of ignorance" scenario, such preferences may make subjects significantly more averse to unequal distributions of rewards than can be explained by risk aversion alone.

One decision procedure dominates a given one if it performs well on the entire class of problems the given decision procedure performs well on, and then goes on to perform well on other problems that the given decision procedure does badly on. Performing well will be defined as generating higher expected utility before entering a problem. In this paper it will be argued that the timeless decision procedure dominates the causal
and evidential decision procedures. It will also be argued in turn that the updateless decision procedure dominates the timeless decision procedure. The difficulties of formalizing a modern variant of the ”smoking gene” problem will then be briefly examined.

This dissertation presents three essays in economics. Firstly, I study the problem of allocating an indivisible good between two agents under incomplete information. I provide a characterization of mechanisms that maximize the sum of the expected utilities of the agents among all feasible strategy-proof mechanisms: Any optimal mechanism must be a convex combination of two fixed price mechanisms and two option mechanisms. Secondly, I study the problem of allocating a non-excludable public good between two agents under incomplete information. An equal-cost sharing mechanism which maximizes the sum of the expected utilities of the agents among all feasible strategy-proof mechanisms is proved to be optimal. Under the equal-cost sharing mechanism, when the built cost is low, the public good is provided whenever one of the agents is willing to fund it at half cost; when the cost is high, the public good is provided only if both agents are willing to fund it. Thirdly, I analyze the problem of matching two heterogeneous populations. If the payoff from a match exhibits complementarities, it is well known that absent any friction positive assortative matching is optimal. Coarse matching refers to a situation in which the populations into a finite number of classes, then randomly matched within these classes. The focus of this essay is the performance of coarse matching schemes with a finite number of classes. The main results of this essay are the following ones. First, assuming a multiplicative match payoff function, I derive a lower bound on the performance of n-class coarse matching under mild conditions on the distributions of agents' characteristics. Second, I prove that this result generalizes to a large class of match payoff functions. Third, I show that these results are applicable to a broad class of applications, including a monopoly pricing problem with incomplete information, as well as to a cost-sharing problem with incomplete information. In these problems, standard models predict that optimal contracts sort types completely. The third result implies that a monopolist can capture a large fraction of the second-best profits by offering pooling contracts with a small number of qualities.