Organizations are increasingly relying on team-based structures as work becomes more complex, nonroutine, and knowledge intensive. Teams comprised of individuals with diverse skills and expertise may be well suited to perform such work. However, as teams become more prevalent, organizations may struggle with ways to exercise control. Formal controls, with their focus on prespecified rules, performance targets, and hierarchical relationships, may be less effective in a teamwork environment than clan control, in which work-related behavior is motivated by shared norms and values, as well as a common vision, and individuals attempt to be accepted or “regular” members of a team. However, little is known about the antecedents of clan control. Much of the existing empirical research on the antecedents of control has been influenced by Ouchi's (Ouchi, W. G. 1977. The relationship between organizational structure and organizational control. Admin. Sci. Quart. 22 95–113) framework, which posits clan control is used when managers lack knowledge of means–ends relationships and are unable to measure outcomes. This paper adopts the view that clan control is a “people” or social process and argues that social capital, a construct that reflects connections and relationships among individuals, is a missing, key antecedent of clan control. In particular, we posit that the existence of social capital enables team members, as well as project managers, to facilitate clan control within a team, i.e., team-based clan control. A model is developed and hypotheses are tested using survey data collected from 95 information systems project teams. The results suggest that social capital assets are associated with team-based clan control. Recognizing that the project manager typically has a different level of organizational authority than team members, additional hypotheses are developed relating characteristics of the project manager to team-based clan control. The results of these tests suggest that team-based clan control is also dependent on the manager's knowledge of business processes and the application area, coupled with the extent to which he observes the behaviors of the project team. Thus, our results suggest that there are conditions under which team-based clan control is facilitated by the project team as a whole (including team members and project managers), as well as conditions under which team-based clan control is facilitated by project managers in their role as team leaders. Implications and suggestions for future work are discussed.
