The religious sector in the United States is distinctive for the degree to which institutional change occurs not through the birth and death of organizations, but through mergers and schisms occurring among preexisting denominations. In this article we analyze mergers among mainstream Protestant denominations, as a means both to deepen our understanding of the dynamics of denominationalism and to broaden the field of discussion about organizational merger. We argue that patterns of merger in any sector are institutionally embedded: in the religious sector, the interaction of ingrained tradition and historical contingency influences the range of possible merger partners, the perceived advantages to merger and the power of actors that may encourage or discourage merger We pursue this argument using event-history data from denominations in four Protestant families between 1890 and 1990 to test hypotheses about the determinants of merger The analysis shows that merger is influenced by denominational identity (family, racial makeup), organizational characteristics (size, centralization, and membership concentration), and the diffuse influence of the ecumenical movement; many of these effects are time dependent, a result of long-term shifts in the roles of denominational elites.

The remarkable degree of religious diversity in the United States is produced more through schisms and mergers occurring to preexisting denominations than through foundings of new religious groups. In this article we analyze schisms, focusing on mainstream Protestant denominations over the period 1890-1990. We treat schism as an organizational process rather than an expression of purely doctrinal disputes; more specifically, we argue that schisms arise from within denominations in response to attempts by denominational elites to achieve organizational consolidation. Event-history techniques are used to test hypotheses concerning resource environments, effects of denominational reorganization, resource mobilization capacity, network linkages, and denominational centralization. Results show that rates of schism are influenced by prior structural changes, size, and linkage to a liberal federation; we find no effects of resource or institutional environments.