Mechanism design theory has given economists a set of tools for designing institutions to achieve socially desirable outcomes. Unfortunately, the behavioral assumptions that these theories often rest are somewhat unrealistic. Testing these institutions in a laboratory setting gives us insight into what assumptions or properties of institutions make them behaviorally successful. Moreover these insights allow us to create new theories that offer, in principle, better actual performance. Thus, the interplay between experimental economics and economic theory seems vital in mechanism design to insure successful institutions. It is in this spirit that this dissertation precedes focusing entirely with mechanisms that were designed to achieve the Lindahl allocation in a public goods environment. The first chapter experimentally examines three such mechanisms in a laboratory setting. It finds that the mechanism that gets the closest to the Lindahl allocation is the one that induces a game with very strong stability of equilibrium properties. Unfortunately this mechanism also has some clear disadvantages: first, it is very complicated; second, payoffs to consumers while learning to play equilibrium are very low; and last, the mechanism gets more complicated when more people participate. The second chapter uses the insights from the first experiment to create a new institution which avoids some of the concerns outlined above while maintaining the strong stability of equilibrium property. The third chapter contributes a missing stability result into the literature. The final chapter of the dissertation experimentally compares the new mechanism introduced in chapter 2 with the most successful mechanism from the first experiment. The treatments in this experiment are designed to stress the above observed trouble areas.

This thesis serves as an introduction and overview of the broad and closely related fields of mechanism design, contract theory, and information economics. Each chapter is intended to provide a self-contained guide to the particular area of application -- examples include adverse selection, moral hazard, and auctions. The reader should benefit from the thesis in two ways: by understanding the general notions of the revelation principle, incentive compatibility, and individual rationality from the mechanism design theory as well as by examining the particular information asymmetry models in the individual areas. Powered by TCPDF (www.tcpdf.org)

Thesis advisor: Utku Ünver / This dissertation consists of three chapters. The first chapter consists of a survey of the literature on affirmative action and diversity objective in school choice mechanisms. It presents and analyzes some of the main papers on the subject, showing the evolution of our understanding of the effects that different affirmative action policies have on the welfare and fairness of student assignments, the satisfaction of the diversity objectives as well as the domain of policies that allows for stable outcomes. The second chapter analyzes the problem of school choice mechanisms when policy-makers have objectives over the distribution of students by type across the schools. I show that mechanisms currently available in the literature may fail to a great extent in satisfying those objectives, and introduce a new one, which satisfies two properties. First, it produces assignments that satisfy a fairness criterion which incorporates the diversity objectives as an element of fairness. Second, it approximates optimally the diversity objectives while still satisfying the fairness criterion. We do so by embedding "preference" for those objectives into the schools' choice functions in a way that satisfies the substitutability condition and then using the school-proposing deferred acceptance procedure. This leads to the equivalence of stability with the desired definition of fairness and the maximization of those diversity objectives among the set of fair assignments. A comparative analysis also shows analytically that the mechanism that we provide has a general ability to satisfy those objectives, while in many familiar classes of scenarios the alternative ones yield segregated assignments. Finally, we analyze the incentives induced by the proposed mechanism in different market sizes and informational structures. The third chapter (co-authored with Orhan Aygün) presents an analysis of the Brazilian affirmative action initiative for access to public federal universities. In August 2012 the Brazilian federal government enacted a law mandating the prioritization of students who claim belonging to the groups of those coming from public high schools, low income families and being racial minorities to defined proportions of the seats available in federal public universities. In this problem, individuals may be part of one or more of those groups, and it is possible for students not to claim some of the privileges associated with them. This turns out to be a problem not previously studied in the literature. We show that under the choice function induced by the current guidelines, students may be better off by not claiming privileges that they are eligible to. Moreover, the resulting assignments may not be fair or satisfy the affirmative action objectives, even when there are enough students claiming low--income and minority privileges. Also, any stable mechanism that uses the current choice functions is neither incentive compatible nor fair. We propose a new choice function to be used by the universities that guarantees that a student will not be worse off by claiming an additional privilege, is fair and satisfies the affirmative action objectives whenever it is possible and there are enough applications claiming low--income and minority privileges. Next, we suggest a stable, incentive compatible and fair mechanism to create assignments for the entire system. / Thesis (PhD) — Boston College, 2014. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Economics.

This dissertation addresses the effect of information on firm and individual behavior. The first chapter examines the design of an optimal feedback mechanism by an informed principal and uses the results to explain why firms tend to assign coarse subjective ratings to their employees. When a firm has private information about an employee's ability, it can communicate this information through a subjective evaluation mechanism. I characterize the firm's optimal disclosure policy as a function of the worker's ability distribution and provide an algorithm to compute it. Further, I show that with some reasonable restrictions on the ability distribution, the firm's optimal strategy is always to reward the best workers, fire the worst ones, and assign one central rating to the rest. The second chapter investigates an informed principal's optimal feedback strategy in a dynamic setting. I first consider the case where both parties have non-binding outside options. In this case, if the principal ever wants to reveal any information, she will do so at the earliest possible stage. Moreover, the optimal disclosure policy can be characterized in the same way as in the static case. The same conclusion holds for the case where both parties have binding and constant outside options. I also discuss the case where both parties have binding and time-variant outside options. After incorporating firms' need to promote and/or to retain workers, the model is used to explain wage dynamics. The third chapter models a decision maker who "rationally" distorts his own belief to avoid the feeling of regret. People often suffer from regret when they realize that their previous choices were suboptimal. As a result, in a dynamic setting where information is revealed gradually, people are tempted to deny new negative information in order to avoid regret. At the same time, they are also aware of the economic cost of such belief distortions. A "rational" decision maker will optimally trade off these two concerns and choose his own belief accordingly. This tradeoff makes the past affect current decisions and hence can explain the sunk cost fallacy.

O principal objetivo desta dissertação é obter um Teorema do Envelope que permita mecanismos não diferenciáveis, preferências arbitrárias e que possa ser aplicado em modelos com múltiplos agentes. Nós alcançamos isto ao expandir a análise de Milgrom e Segal (2002), generalizando seus resultados para espaços de tipos multidimensionais. Dessa forma, continuamos permitindo que a regra de escolha (mecanismo) seja descontínua. Para obter nosso resultado, é necessário o uso do Teorema do Máximo de Berge e, consequentemente, devemos impor compacidade no conjunto de escolha. Inicialmente esta hipótese pode parecer forte, porém argumentamos que em aplicações _e muito improvável termos um conjunto de escolha aberto ou, principalmente, não limitado. Nós também identificamos condições para que a função valor seja absolutamente contínua e mostramos que sua representação integral também é válida para espaços de tipos multidimensionais. Inicialmente propomos uma generalização direta do resultado de Milgrom e Segal (2002), utilizando a hipótese de continuidade absoluta da função de utilidade do agente. Entretanto, esta exigência não possui muito significado econômico e é considerada pouco elegante por parte da literatura. Neste sentido, incorporamos uma hipótese adicional de diferenciabilidade da utilidade em todo o domínio que gera a mesma representação integral e possui uma maior interpretação econômica. Nossos resultados são, em geral, aplicados a modelos com múltiplos agentes, em especial Economia do Setor Público (provisão de bens públicos e taxação ótima) e teoria dos leilões. / The main objective of this dissertation is to obtain an Envelope Theorem that allows non-di erentiable mechanisms, arbitrary preferences, and that can be applied to models with multiple agents. We achieve that by expanding the analysis of Milgrom and Segal (2002) and generalizing their results to multidimensional type spaces. Thus, we continue allowing that the choice rule (mechanism) is discontinuous. For our result, it is necessary to use the Berge's Maximum Theorem and therefore we must impose compactness in the choice set. Initially this assumption may seem strong, but we argue that in applications there is an open or unbounded choice set is very unlikely. We also identify conditions for the value function is absolutely continuous and show that its integral representation is also valid for multidimensional type spaces. Firstly we propose a direct generalization of the Milgrom and Segal (2002)'s result, using the assumption of absolute continuity of the agent's utility function. However, this requirement does not have much economic interpretation and it is considered not very elegant in the literature. In this sense, we incorporate an additional assumption of di erentiability of the utility in all range that generates the same integral representation and it possesses a greater economic interpretation. Our results are generally applied to models with multiple agents, in particular Public Economics (public goods supply and optimal taxation) and auction theory.

<p>This dissertation collects work concerning the way individuals deal with imperfect information, both related to their knowledge of themselves and of others. The second chapter shows that bounded rationality, in the form of limited knowledge of utility, is an explanation for common stylized facts of prospect theory like loss aversion, status quo bias and non-linear probability weighting. Locally limited utility knowledge is considered within a classical demand model framework, suggesting that costs of inefficient search for optimal consumption will produce a value function that obeys the loss aversion axiom of Tversky and Kahneman (1991). Moreover, since this adjustment happens over time, new predictions are made that explain why the status quo bias is reinforced over time. This search can also describe the behavior of a consumer facing an uncertain future wealth level. The search cost justifies non-linear forms of probability weighting. The effects that have been observed in experiments will follow as a consequence.</p><p>The third chapter looks to understand how firms create and maintain long term relationships with consumers, or how procurement relations evolve over time, by studying a dynamic variant of the classical two-type-buyer contract in mechanism design. It is less trivial and more interesting if the utility determinant (or utility type) is not fixed or completely random, and fair assumptions are that it is either stochastic, or given by a distribution whose parameters are common knowledge. The first approach is that of Battaglini (2005), while the second is pursued in this paper. With two possible types of buyers, the buyer more likely to have a high utility type will receive the first-best allocations, while the other will receive the first best only if he has the high utility type. </p><p>The last chapter analyzes a dynamic procurement setting with promise keeping, where two firms (agents) with private information on their costs contract competitively with a principal. To this end, two models are proposed and the optimal allocations are determined. The agents face liquidity constraints, which induce distortions when high marginal costs are reported. We deduce that the principal uses promised utilities to incentivize the agents, which act as state variables in the recursive maximization problem. High cost types are allocated less than efficient quantities and the inefficiency of the allocation is relieved as the promised utilities increase.</p> / Dissertation

This thesis addresses three problems where a focal agent's operational policies (inventory and capacity allocation) interact with marketing decisions. The first chapter studies how wholesale all-unit discounts may lead to products being shifted from authorized retailers to discounted gray market channels. Such discounts lead to discontinuous ordercosts which may induce buyers to order up to a threshold where they receive a greater discount. The buyer in this chapter is a reseller who makes purchasing decisions while taking into account inventory holding costs, how their resale price affects consumer demand and whether or not they divert inventory to the gray market. I analyze factors which determine how the reseller balances between lowering resale prices and diverting to the gray market, both of which lower costs by shortening the time inventory is held. Modelling the decisions as a Stackelberg game, the welfare of the authorized channel participants is analyzed. Of import, consumer welfare may decrease if a gray market emerges when holding costs are low. In the latter two chapters, the supplier sells a congested service. For example, this supplier may be a courier facing stochastic buyer arrivals. Buyers vary in their value for the service and how patient they are, so the supplier may improve outcomes by providing a menu of delay levels and prices. The system is modelled as a priority queue where congestion constrains the arrival rates at each delay level. In the first study, the supplier has aggregate market data. I model the problem as an optimization subject to incentive and congestion constraints. The novel contributions include a precise description of the optimal menu as a function of the supplier's capacity (the rate at which buyers can be served). Findings include existence of distinct capacity regions where the supplier utilizes service pooling and strategic delay. In the final chapter the related welfare maximization problem is considered. Sufficient conditions for optimal pricing are derived which depend only on operational information: the current revenue must be equal to the best-case revenue subject to current prices and congestion constraints. An associated performance measure is shown to bound deviation from maximum welfare and is used as a heuristic within an adaptive pricing protocol. This protocol is shown to converges to near welfare maximizing outcomes.

This thesis addresses three problems where a focal agent's operational policies (inventory and capacity allocation) interact with marketing decisions. The first chapter studies how wholesale all-unit discounts may lead to products being shifted from authorized retailers to discounted gray market channels. Such discounts lead to discontinuous ordercosts which may induce buyers to order up to a threshold where they receive a greater discount. The buyer in this chapter is a reseller who makes purchasing decisions while taking into account inventory holding costs, how their resale price affects consumer demand and whether or not they divert inventory to the gray market. I analyze factors which determine how the reseller balances between lowering resale prices and diverting to the gray market, both of which lower costs by shortening the time inventory is held. Modelling the decisions as a Stackelberg game, the welfare of the authorized channel participants is analyzed. Of import, consumer welfare may decrease if a gray market emerges when holding costs are low. In the latter two chapters, the supplier sells a congested service. For example, this supplier may be a courier facing stochastic buyer arrivals. Buyers vary in their value for the service and how patient they are, so the supplier may improve outcomes by providing a menu of delay levels and prices. The system is modelled as a priority queue where congestion constrains the arrival rates at each delay level. In the first study, the supplier has aggregate market data. I model the problem as an optimization subject to incentive and congestion constraints. The novel contributions include a precise description of the optimal menu as a function of the supplier's capacity (the rate at which buyers can be served). Findings include existence of distinct capacity regions where the supplier utilizes service pooling and strategic delay. In the final chapter the related welfare maximization problem is considered. Sufficient conditions for optimal pricing are derived which depend only on operational information: the current revenue must be equal to the best-case revenue subject to current prices and congestion constraints. An associated performance measure is shown to bound deviation from maximum welfare and is used as a heuristic within an adaptive pricing protocol. This protocol is shown to converges to near welfare maximizing outcomes.

This thesis work is in the area of modeling trust in multi-agent systems, systems of software agents designed to act on behalf of users (buyers and sellers), in applications such as e-commerce. The focus is on developing an approach for buyers to model the trustworthiness of sellers in order to make effective decisions about which sellers to select for business. One challenge is the problem of unfair ratings, which arises when modeling the trust of sellers relies on ratings provided by other buyers (called advisors). Existing approaches for coping with this problem fail in scenarios where the majority of advisors are dishonest, buyers do not have much personal experience with sellers, advisors try to flood the trust modeling system with unfair ratings, and sellers vary their behavior widely. We propose a novel personalized approach for effectively modeling trustworthiness of advisors, allowing a buyer to 1) model the private reputation of an advisor based on their ratings for commonly rated sellers 2) model the public reputation of the advisor based on all ratings for the sellers ever rated by that agent 3) flexibly weight the private and public reputation into one combined measure of the trustworthiness of the advisor. Our approach tracks ratings provided according to their time windows and limits the ratings accepted, in order to cope with advisors flooding the system and to deal with changes in agents' behavior. Experimental evidence demonstrates that our model outperforms other models in detecting dishonest advisors and is able to assist buyers to gain the largest profit when doing business with sellers. Equipped with this richer method for modeling trustworthiness of advisors, we then embed this reasoning into a novel trust-based incentive mechanism to encourage agents to be honest. In this mechanism, buyers select the most trustworthy advisors as their neighbors from which they can ask advice about sellers, forming a social network. In contrast with other researchers, we also have sellers model the reputation of buyers. Sellers will offer better rewards to satisfy buyers that are well respected in the social network, in order to build their own reputation. We provide precise formulae used by sellers when reasoning about immediate and future profit to determine their bidding behavior and the rewards to buyers, and emphasize the importance for buyers to adopt a strategy to limit the number of sellers that are considered for each good to be purchased. We theoretically prove that our mechanism promotes honesty from buyers in reporting seller ratings, and honesty from sellers in delivering products as promised. We also provide a series of experimental results in a simulated dynamic environment where agents may be arriving and departing. This provides a stronger defense of the mechanism as one that is robust to important conditions in the marketplace. Our experiments clearly show the gains in profit enjoyed by both honest sellers and honest buyers when our mechanism is introduced and our proposed strategies are followed. In general, our research will serve to promote honesty amongst buyers and sellers in e-marketplaces. Our particular proposal of allowing sellers to model buyers opens a new direction in trust modeling research. The novel direction of designing an incentive mechanism based on trust modeling and using this mechanism to further help trust modeling by diminishing the problem of unfair ratings will hope to bridge researchers in the areas of trust modeling and mechanism design.

This thesis work is in the area of modeling trust in multi-agent systems, systems of software agents designed to act on behalf of users (buyers and sellers), in applications such as e-commerce. The focus is on developing an approach for buyers to model the trustworthiness of sellers in order to make effective decisions about which sellers to select for business. One challenge is the problem of unfair ratings, which arises when modeling the trust of sellers relies on ratings provided by other buyers (called advisors). Existing approaches for coping with this problem fail in scenarios where the majority of advisors are dishonest, buyers do not have much personal experience with sellers, advisors try to flood the trust modeling system with unfair ratings, and sellers vary their behavior widely. We propose a novel personalized approach for effectively modeling trustworthiness of advisors, allowing a buyer to 1) model the private reputation of an advisor based on their ratings for commonly rated sellers 2) model the public reputation of the advisor based on all ratings for the sellers ever rated by that agent 3) flexibly weight the private and public reputation into one combined measure of the trustworthiness of the advisor. Our approach tracks ratings provided according to their time windows and limits the ratings accepted, in order to cope with advisors flooding the system and to deal with changes in agents' behavior. Experimental evidence demonstrates that our model outperforms other models in detecting dishonest advisors and is able to assist buyers to gain the largest profit when doing business with sellers. Equipped with this richer method for modeling trustworthiness of advisors, we then embed this reasoning into a novel trust-based incentive mechanism to encourage agents to be honest. In this mechanism, buyers select the most trustworthy advisors as their neighbors from which they can ask advice about sellers, forming a social network. In contrast with other researchers, we also have sellers model the reputation of buyers. Sellers will offer better rewards to satisfy buyers that are well respected in the social network, in order to build their own reputation. We provide precise formulae used by sellers when reasoning about immediate and future profit to determine their bidding behavior and the rewards to buyers, and emphasize the importance for buyers to adopt a strategy to limit the number of sellers that are considered for each good to be purchased. We theoretically prove that our mechanism promotes honesty from buyers in reporting seller ratings, and honesty from sellers in delivering products as promised. We also provide a series of experimental results in a simulated dynamic environment where agents may be arriving and departing. This provides a stronger defense of the mechanism as one that is robust to important conditions in the marketplace. Our experiments clearly show the gains in profit enjoyed by both honest sellers and honest buyers when our mechanism is introduced and our proposed strategies are followed. In general, our research will serve to promote honesty amongst buyers and sellers in e-marketplaces. Our particular proposal of allowing sellers to model buyers opens a new direction in trust modeling research. The novel direction of designing an incentive mechanism based on trust modeling and using this mechanism to further help trust modeling by diminishing the problem of unfair ratings will hope to bridge researchers in the areas of trust modeling and mechanism design.