This thesis examines on a conceptual level whether, in Canada, government induced centralization of the bargaining structure is apt to contribute to a stabilization of industrial relations measured in terms of a reduction in strike/lockout frequency. A micro-model of strike causation, viewing the outbreak of industrial action as the result of a deliberate gamble of the bargaining parties, forms the core of the conceptual framework developed in this paper. According to this model centralization lowers strike/lockout frequency if it (1) raises the relative costs of industrial action (2) induces the bargaining parties to integrate macro-goals into their goal portfolio (3) modifies collectivity internal interest aggregation in favor of its moderate segments and (4) successfully internalizes intra-group conflict. On the basis of theoretical considerations and existing evidence, the paper then seeks to establish whether centralization can be expected to fulfill these conditions. The study suggests that the impact of centralization upon vertical interest aggregation takes the desired direction. Because of the design of the analysis a general prediction of the effect of centralization upon horizontal interest aggregation can not be made. The relative costs of industrial action in the 'average bargaining unit' are diagnosed to remain unaffected by centralization. While centralization of the bargaining structure appears to be a necessary condition for unions and employers to integrate macro-goals into their goal portfolio, the analysis can not show that it is a sufficient condition. Finally, contingent on the readiness of the government to actively back the rule of the majority, centralization is expected to internalize intra-group conflict successfully. / Arts, Faculty of / Political Science, Department of / Graduate