This thesis focuses on finding Bayesian equilibria for sellers in virtual auctions using real currencies. Existing literature for real-world auctions is examined from the perspective of economic theory, game theory, and pricing strategies. Next, computer science theory is reviewed to identify applications of real-world auction models in video games. Finally, the video game Diablo 3, the first to have a real currency auction house for virtual goods, will be examined as a case study. This thesis contributes to the known literature by analyzing the Diablo 3 real currency auction house and identifying seller strategies to be applied in future virtual auction houses and economies using real currencies. JEL Classification C73, D44, D58 Keywords repeated games, auctions, general equilibrium Author's e-mail waheedbrown@hotmail.com Supervisor's e-mail gregor@fsv.cuni.cz Abstrakt Tato práce se zameruje na hledání Bayesovské rovnováhy pro prodejce ve virtuálních aukcích, které používají skutecné peníze. Na úvod se prozkoumá existující literatura o aukcích z hlediska ekonomické teorie, teorie her tvorby a cenové strategie. Dále budou analyzovány príspevky z informatiky, které povedou k zjištení zpusobu aplikace aukcních modelu ve videohrách. Hlavním príspevkem bude analýza video...

This thesis consists of three essays in economic theory, two on search models with information acquisition and one on repeated games when precise information about discount factors is unavailable. In the first essay, I develop a model in which optimal costly information acquisition by individual firms causes adverse selection in the market as a whole. Each firm’s information acquisition policy determines which customers it serves, which in turn affects the distribution of remaining customers and hence other firms’ incentives. I show that when information acquisition is ‘smooth’, the adverse selection externality due to each firm is dampened, and in equilibrium all firms make positive profits. By contrast, with lumpy information acquisition, only a limited number of firms are profitable. I establish that my results apply to a broad class of continuous-time information acquisition processes. The second essay explores information acquisition in labor markets. Noting that African-Americans face shorter employment durations than similar whites, we hypothesize that employers discriminate in acquiring ability-relevant information. We construct a model with a binary information generating process, ‘monitoring’, at the disposal of firms. Monitoring black but not white workers is self-sustaining. This ‘bad’ equilibrium is not merely a matter of coordination; rather, it is determined by history and not easily reversed. The model’s additional predictions, lower lifetime incomes and longer unemployment durations for blacks, are both strongly empirically supported. In the third essay, we investigate the possibility of repeated games equilibria that are robust to the discount factors. We prove a negative result which shows that a sizable part of the set of feasible individually rational payoffs can never be produced by such equilibria. We find the cutoff defining this region and interpret it as a limit on the ability to punish deviations when future rewards for randomization cannot be finely calibrated. Furthermore, we present a robust folk theorem to support payoffs in the complementary region with strategies that remain Subgame Perfect Nash Equilibria at all greater discount factors.

In strategically rich settings in which machines and people do not fully share the same preferences, machines must learn to cooperate and compromise with people to establish mutually successful relationships. However, designing machines that effectively cooperate with people in these settings is difficult due to a variety of technical and psychological challenges. To better understand these challenges, we conducted a series of user studies in which we investigated human-human, robot-robot, and human-robot cooperation in a simple, yet strategically rich, resource-sharing scenario called the Block Dilemma, a game in which players must balance fairness, efficiency, and risk. While both human-human and robot-robot pairs typically learn fair and cooperative solutions over time, our results show that these solutions tend to be different when communication is permitted versus when it is not. While people followed a less risky and less efficient solution, pairs of robots followed a more risky but more efficient solution. This difference in humans’ and machines’ behavior appears to influence human-robot cooperation negatively, as our studies show that human-robot pairs did not frequently produce either form of cooperation without communication. These results speak to the need for machine behavior to be better aligned with human behavior. While machines may behave more efficiently and produce better results than people when following their own calculations, machines may often better facilitate human-machine cooperation by aligning their behavior with human behavior rather than expecting human behavior to become more efficient.

This thesis presents two chapters on credence goods and one on ongoing partnerships in an infinitely repeated game. The chapters on credence goods focus on the welfare and efficiency of equilibria in overcharging models of credence goods, something which has not been explicitly addressed before. The chapter on partnerships presents a theory explaining ongoing partnerships as solving a commitment problem for clients. There is a small literature on partnerships, and this chapter represents a novel but complimentary approach to that literature. At core, chapters 2, 3 and 4 of this thesis ask the following questions respectively: Do competition and information increase welfare in credence goods markets? How do customers in credence goods markets discipline experts from committing fraud? Can these strategies be welfare ranked? Why do ongoing partnerships exist? What problem do they solve?

[pt] Este trabalho apresenta um modelo de conluio em leilões repetidos de primeiro preço estático, ou seja, um esquema de conluio que usa a mesma regra de conluio em toda a fase de cooperação, independente da história do superjogo. Neste ambiente, introduzimos um mecanismo de comunicação sem custo, por meio do qual os jogadores transmitem sua ordenação de preferências sobre os bens a serem leiloados no jogo estágio. A partir deste modelo, obtemos um aumento da receita esperada agregada por parte dos participantes do leilão frente a um modelo de conluio estático tácito. Além disso, o resultado de dominância do conluio dinâmico com comunicação - esquema no qual a regra de conluio depende de toda a história pública do superjogo - apresentado por Aoyagi (2002), frente a qualquer conluio estático não se verifica. Outro ponto a se destacar é que o refinamento do sistema de comunicação, por meio de um aumento do número de bens ordenados em cada mensagem, embora leve a um aumento da receita esperada, pode gerar uma elevação da taxa de paciência necessária para a manutenção do conluio. / [en] This paper models a Static Collusion in repeated First Price Auctions, that is, a bid rotation scheme which uses the same instruction rule throughout the collusion phase independent of the history. In this environment, we allow that bidders send cheap talk messages to each other about their ranking of preferences over goods that will be auctioned during the stage game. We can show that the introduction of the cheap talk messages improves the bidder`s aggregated surplus compared to tacit static collusion models. Moreover, the dominance of dynamic bid rotation schemes, as showed by Aoyagi (2002), on static versions is not necessarily truth. Another remark is that refining the cheap talk mechanism, by the extension of ranking chain, although generates larger expected receipts, may increases the minimum necessary patience rate to sustain collusion.

This PhD dissertation studies how market structures and economic incentives transform heterogeneity at agent levels into unequal economic outcomes. The first chapter studies the economic incentives that lead a country to specialise its production in specific segments of a supply chain, and how these incentives transform heterogeneity at the productivity level into wage differences between countries. This chapter presents an innovative framework that incorporates production networks to the Ricardian trade model. It describes the price formation mechanism that occurs along supply chains and how it induces countries to focus on the production of specific goods. Moreover, the model highlights the role of the network structure in the determination of prices, and uses it to explain how changes in the productivity of a country have consequences in the production decisions and wages of the other countries that produce goods in the supply chain. The second chapter studies the effects that the heterogeneity of income flows has over the implementation of collective agreements. Collective agreements are the primary mechanism by which communities cope with market failures. However, the lack of enforcement mechanisms generates coordination challenges. This chapter presents a theoretical framework that studies how inequality among individuals affects the participation incentives of the individuals and explains why agreements that balance the rent-seeking behaviour of wealthy individuals with the redistribution interests of the poor reduce the adverse effects of heterogeneity, and can even use it to create more robust agreements. The third chapter studies heterogeneity at the level of academic journals. This chapter models the interaction between authors and journals as a platform market and uses this model to explain how general interest journals compete against field-specific journals. The model provides new insights into the way in which general interest journals link the different publication incentives of journals across fields. The theoretical results explain why general interest journals tend to attract higher quality publications and how changes in the publication capacity of a journal, or the volume of research in a field, can affect the quality of ideas published in both field-specific and general interest journals. Finally, this chapter applies the previous theoretical results to understand how the Top 5 journals in economics obtained their central role, and how their influence has changed between 1980 and the present.

This dissertation consists of three essays on reputation and repeated games. Reputation models typically assume players have full memory of past events, yet in many applications this assumption does not hold. In the first chapter, I explore two different relaxations of the assumption that history is perfectly observed in the context of Ely and Välimäki's (2003) mechanic game, where reputation (with full history observation) is clearly bad for all players. First I consider "limited history," where short-run players see only the most recent T periods. For large T, the full history equilibrium behavior always holds due to an "echo" effect (for high discount factors); for small T, the repeated static equilibrium exists. Second I consider "fading history," where short-run players randomly sample past periods with probabilities that "fade" toward zero for older periods. When fading is faster than a fairly lax threshold, the long-run player always acts myopically, a result that holds more generally for reputation games where the long-run player has a strictly dominant stage game action. This finding suggests that reputational incentives may be too weak to affect long-run player behavior in some realistic word-of-mouth environments. The second chapter develops general theoretical tools to study incomplete information games where players observe only finitely many recent periods. I derive a recursive characterization of the set of equilibrium payoffs, which allows analysis of both stationary and (previously unexplored) non-stationary equilibria. I also introduce "quasi-Markov perfection," an equilibrium refinement which is a necessary condition of any equilibrium that is "non-fragile" (purifiable), i.e., robust to small, additively separable and independent perturbations of payoffs. These tools are applied to two examples. The first is a product choice game with 1-period memory of the firm's actions, obtaining a complete characterization of the exact minimum and maximum purifiable equilibrium payoffs for almost all discount factors and prior beliefs on an "honest" Stackelberg commitment type, which shows that non-stationary equilibria expand the equilibrium set. The second is the same game with long memory: in all stationary and purifiable equilibria, the long-run player obtains exactly the Stackelberg payoff so long as the memory is longer than a threshold dependent on the prior. These results show that the presence of the honest type (even for arbitrarily small prior beliefs) qualitatively changes the equilibrium set for any fixed discount factor above a threshold independent of the prior, thereby not requiring extreme patience. The third chapter studies the question of why drug trafficking organizations inflict violence on each other, and why conflict breaks out under some government crackdowns and not others, in a repeated games context. Violence between Mexican drug cartels soared following the government's anti-cartel offensive starting in 2006, but not under previous crackdowns. I construct a theoretical explanation for these observations and previous empirical research. I develop a duopoly model where the firms have the capacity to make costly attacks on each other. The firms use the threat of violence to incentivize inter-cartel cooperation, and under imperfect monitoring, violence occurs on the equilibrium path of a high payoff equilibrium. When a "corrupt" government uses the threat of law enforcement as a punishment for uncooperative behavior, violence is not needed as frequently to achieve high payoffs. When government cracks down indiscriminately, the firms may return to frequent violence as a way of ensuring cooperation and high payoffs, even if the crackdown makes drug trafficking otherwise less profitable. / text

This dissertation consists of three essays in labor economics and contract theory. The first essay examines whether one’s wage is based on information about the performance of one’s personal contacts. I study wage determination under two assumptions about belief formation: individual learning, under which employers observe only one’s own characteristics, and social learning, under which employers also observe those of one’s personal contacts. Using data on siblings in the NLSY79, I test whether a sibling’s characteristics are priced into one’s wage. If learning is social, then an older sibling’s test score should typically have a larger adjusted impact on a younger sibling’s log wage than vice versa. The empirical findings support this prediction. Furthermore, I perform several exercises to rule out other potential factors, such as asymmetric skill formation, human capital transfers, and role model effects. The second essay analyzes the influence of macroeconomic conditions during childhood on the labor market performance of adults. Based on Census data, I document the relationship of unemployment rates in childhood to schooling, employment, and income as an adult. In addition, a sample from the PSID is used to study how the background attributes of parents raising children vary over the business cycle. Finally, information from the NLSY79-CH is examined in order to characterize the impact of economic fluctuations on parental caregiving. Overall, the evidence is consistent with a negative effect of the average unemployment rate in childhood on parental investments in children and the stock of human capital in adulthood. The third essay studies the bilateral trade of divisible goods in the presence of stochastic transaction costs. The first-best solution requires each agent to transfer all of her good to the other agent when the transaction cost reaches a certain threshold value. However, in the absence of court-enforceable contracts, such a policy is not incentive compatible. We solve for the unique maximal symmetric subgame-perfect equilibrium, in which agents can realize some gains from trade by transferring their goods sequentially. Several comparative statics are derived. In some cases, the first-best outcome can be approximated as the agents become infinitely patient. / Economics

This thesis deals with the contraception market in the context of decision making about children, trying to outline the consequences of this market for the economy as a whole. The first section introduces the market and defines the key role of the market for the women making the decision. The second part introduces the agents in the market with their incentives and motivations. The third part moves into closer detail paying attention mainly to choice problems and uncertainty problems. The forth part focuses on game theoretical framework and various strategies from which the players are choosing. The fifth part presents a range of models for situations in the market derived from "games" on the prisoner's dilemma principle. These games result in the hypotheses involving consequences of particular behaviour in the market. The sixth part tests the hypotheses from the previous chapter, using evidence from selected countries including Czech Republic, United Kingdom, United States, and China. The conclusion specifies future development that is uncertain but based on the current facts.

Artificial intelligence (AI) agents will need to interact with both other AI agents and humans. One way to enable effective interaction is to create models of associates to help to predict the modeled agents' actions, plans, and intentions. If AI agents are able to predict what other agents in their environment will be doing in the future and can understand the intentions of these other agents, the AI agents can use these predictions in their planning, decision-making and assessing their own potential. Prior work [13, 14] introduced the S# algorithm, which is designed as a robust algorithm for many two-player repeated games (RGs) to enable cooperation among players. Because S# generates actions, has (internal) experts that seek to accomplish an internal intent, and associates plans with each expert, it is a useful algorithm for exploring intent, plan, and action in RGs. This thesis presents a graphical Bayesian model for predicting actions, plans, and intents of an S# agent. The same model is also used to predict human action. The actions, plans and intentions associated with each S# expert are (a) identified from the literature and (b) grouped by expert type. The Bayesian model then uses its transition probabilities to predict the action and expert type from observing human or S# play. Two techniques were explored for translating probability distributions into specific predictions: Maximum A Posteriori (MAP) and Aggregation approach. The Bayesian model was evaluated for three RGs (Prisoners Dilemma, Chicken and Alternator) as follows. Prediction accuracy of the model was compared to predictions from machine learning models (J48, Multi layer perceptron and Random Forest) as well as from the fixed strategies presented in [20]. Prediction accuracy was obtained by comparing the model's predictions against the actual player's actions. Accuracy for plan and intent prediction was measured by comparing predictions to the actual plans and intents followed by the S# agent. Since the plans and the intents of human players were not recorded in the dataset, this thesis does not measure the accuracy of the Bayesian model against actual human plans and intents. Results show that the Bayesian model effectively models the actions, plans, and intents of the S# algorithm across the various games. Additionally, the Bayesian model outperforms other methods for predicting human actions. When the games do not allow players to communicate using so-called cheaptalk, the MAP-based predictions are significantly better than Aggregation-based predictions. There is no significant difference in the performance of MAP-based and Aggregation-based predictions for modeling human behavior when cheaptalk is allowed, except in the game of Chicken.