The first chapter introduces the topic of coalition structures and stresses its importance and impact on game theory. Additionally, it presents and motivates the significance of the three following chapters. / The second chapter investigates how rational individuals partition themselves into different coalitions. We propose a notion that determines simultaneously the coalition structures that are likely to prevail in a game, as well as the feasible payoff configurations associated with them. Our solution concept is built in the spirit of von Neumann and Morgenstern stability, but it overcomes the overoptimism associated with it when employed in our context. Moreover, in doing so, we achieve consistency and resolve the problem of myopia embedded in previous notions. We prove existence for a general class of games, and investigate the efficiency of our solution. / The third chapter ascertains which partitions of players will emerge and what actions will these players choose under each partition, when they can negotiate with each other and their actions have externalities. Naturally, the environment is depicted by a normal form game. The solution is a collection of pairs, each consisting of a coalition structure and a strategy profile. Although this chapter addresses the same question the Ray and Vohra (1997) paper did, it does so in a manner that overcomes the problems embedded in their approach. In particular, we assume that once a coalition deviates it fears the worst, given that the non-members do the best for themselves. In doing so, we improve upon previous solution concepts (e.g. strong and coalition-proof Nash equilibria) by NOT assuming that all other players will stay put. Yet, unlike the Ray and Vohra approach, we do not endow the deviating coalition with the power of imposing its will on all of the other players. A general solution concept is defined and its properties (efficiency, etc.) and applications are discussed. / The fourth chapter analyzes cartel stability when firms are farsighted. It studies a price leadership model a la D'Aspremont et al. (1983), where the dominant cartel acts as a leader by determining the market price, while the fringe behaves competitively. According to D'Aspremont et al.'s notion of cartel stability, a firm will not remain in a cartel as long as it prefers the outcome where it is the only member leaving the cartel for the fringe. Such an approach implies that the firm is myopic since it ignores whether the outcome of its deviation is stable itself. Our notion captures foresight by employing a solution concept built in the spirit of von Neumann and Morgenstern stable set, yet adopting an indirect dominance a la Harsanyi. We show that there always exists a unique, non-empty set of stable cartels.
| Identifer | oai:union.ndltd.org:LACETR/oai:collectionscanada.gc.ca:QMM.36762 |
| Date | January 2000 |
| Creators | Diamantoudi, Effrosyni. |
| Contributors | Greenberg, Joseph (advisor) |
| Publisher | McGill University |
| Source Sets | Library and Archives Canada ETDs Repository / Centre d'archives des thèses électroniques de Bibliothèque et Archives Canada |
| Language | English |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Format | application/pdf |
| Coverage | Doctor of Philosophy (Department of Economics.) |
| Rights | All items in eScholarship@McGill are protected by copyright with all rights reserved unless otherwise indicated. |
| Relation | alephsysno: 001780143, proquestno: NQ69869, Theses scanned by UMI/ProQuest. |
Page generated in 0.0018 seconds