Abstract
The centralization of government in metropolitan areas has been aw often proposed but seldom attained goal. A review of the organizational assumptions upon which these proposals are based, and the findings of recent research, both suggest that a re-evaluation of the capacities of centralized and decentralized systems of government is in order. Past writings have provided little basis for explaining the viability of or understanding the characteristics of fractionized government. With a theoretical model of decentralized government in which service units compete under quasi-market conditions, it is possible to explore the operation of such a system and its capacity to respond to local and metropolitan needs.