Research on the consensus-performance relationship has typically used correlational data to examine the simple, bivariate relationship between top management team consensus and firm performance. The results of this research are equivocal. Recent theoretical work, however, suggests a number of 'third variables' that may provide additional insight into the consensus-performance relationship through the 'process of elaboration'. This paper presents theoretical models, based on recent theory building, that may be appropriate for incorporating these additional variables in future research.
CONSENSUSConsensus, defined as general agreement among all or most, is viewed as an important outcome of group decision making. Whyte (1989, p. 41) notes, for example, that 'the task, after all, of a decision making group is to produce consensus from the initial preferences of its members'. Similarly, agreement among strategy makers on a firm's goals and the competitive methods appropriate for achieving them has long been considered an important outcome of the strategy formulation process. According to the rational-normative strategy literature, for example, each step of the sequential planning process must result in top management team (TMT) consensus (Andrews, 1971; Anso6 1965). In the incrementalpolitical view, goals and methods are determined through bargaining and negotiation processes that lead to consensus regarding a course of action (Allison, 1971; Braybrooke and Lindblom, 1970; Cyert and March, 1963;Quinn, 1980). Despite its appeal from both the normative strategy literature and incremental-political perspective, the attainment of consensus by a TMT can have both positive and negative consequences for an organization. On the positive side, consensus may result from the open sharing of information and the expressing of opinions and perceptions pertaining to a firm's competitive environment, goals and strategies. Such interactions may help resolve differences and lead to both a common understanding and a strong commitment to strategy. Such favourable consequences may be facilitated by Brodwin and Bourgeois's (1 984) 'collaborative' and 'cultural' approaches to strategy formulation and implementation, or by Shrivastava and Grant's (1 985) 'systematic bureaucracy model', wherein consensus building is enhanced through problem centred interaction, shared norms, and formulating multiple problem-solving sets within a highly structured system of decision making. Despite such desirable consequences, achieving consensus may also, under certain conditions, be detrimental to an organization. For example, in interpreting his research findings, Bourgeois argued that managers should heed Janis's (1972) exhortation to avoid the perils of groupthink -that is, a firm should emphasize diversity in both vision and direction among its top managers, for in consonance comes insulation, arrogance, tunnel vision, blindness, and Watergate-style feelings of moral omnipotence. Diversity in perception, however, removes blinders, and diversity in goal...