Abstract
In this comparative study a modified industrial relations system concept, combining both environmental, perception and operational influences on the system of industrial relations, is used to test the hypothesis that the technical and market contexts of industrial relations ('constants') produce similarities, whilst the power context ('variables') produces differences in industrial relations rules across national systems. The focus of the study are the rules of conflict resolution at plant level. A combination of methodologies was used in the empirical study including literature surveys, expert interviews and a questionnaire survey of plant-level actors. The findings show considerable differences in both the type of conflicts arising and resolution processes in West German and British plants. In West German plants, conflicts took the form of collective procedural rights initiated by the works council in response to management decisions and were resolved by a process described as 'cooperative constitutionalism'. In Britain, conflicts take the form of collective procedural and substantive interests initiated by employees or their representatives in active pursuit of their own interests as well as in response to management decisions and are generally resolved by a collective bargaining process. The findings point to a complex interaction between constants and variables in their influence on industrial relations rules. Whilst constants have greater explanatory power in the British system of conflict resolution at plant level, variables in the form of the state and the collective bargaining parties have influenced the West German system to a significant extent. The hypothesis of a logic of industrialism' across industrial societies does not have sufficient explanatory power in view of the considerable differences outlined in the study. The modified industrial relations system concept used in the study is seen as a useful theoretical tool for generating hypotheses on industrial relations phenomena, which can be tested empirically.