Essays in Cooperative Game Theory and Public Finance

Book excerpt: This dissertation is divided into three essays. The first essay focuses on the question of enfranchisement of groups of individuals. Existing research work in the area has not adequately addressed this question. Indeed, previous analyses have focused on the value of a vote for a single individual. We considered a totally different approach whereby the central players are groups of individuals rather than single individuals. Within this framework, we have provided an answer to the question as to why groups of individuals struggle to gain the right to vote. We also provided an explanation of why the group in power may voluntarily extend the franchise to groups which have conflicting preferences. The second essay introduces the concept of coalition structure value to differential information economies. This extends the work of Krasa and Yannelis (1994) who introduced the concept of private value allocation to measure the information superiority of agents in an economy with differential information. The coalition structure value allows us to analyze the implications of coalition structures on the value of information. Contrary to our expectations, we found that the bargaining strength of an agent with superior information does not decrease if the rest of the agents collude and bargain as a unit with him. This is in sharp contrast to the one seller and two buyers example in full information economies where the buyers are better off if they bargain as a unit with the seller. The third paper generalizes the crowding types model introduced by Conley and Wooders (1994a) by allowing variable usage of local public goods within jurisdictions. We investigate the possibility of anonymous decentralizations of core allocations. It turns out that it is possible to decentralize core allocations with anonymous admission prices. However these prices are infinite dimensional. This leads us to investigate Lindahl decentralization. While it is possible to get nonanonymous Lindahl decentralization, the core is generally larger than the set of anonymous Lindahl equilibria.