Pii: S0361-3682(98)00060-9

Organisations increasingly face greater competition and uncertainty. One important organisational development is the creation of interand intra-departmental teams that improve both the speed and quality of an organisation's response. Successful teams require the empowerment of team members, an adequate information base, rewards for team performance, and the requisite abilities in team members. Management accounting systems can provide an integral part of the information base necessary for decision-making and rewarding performance. We investigate the incidence and importance of performance measurement for team performance. Team performance is positively associated with the variety and comprehensiveness of performance measures used. This relationship is enhanced if members participate in setting performance targets. Further, team performance is enhanced when team performance is given a greater weight in compensation. Finally, these e€ects are mutually reinforcing, such that team performance is substantially better when comprehensive performance measurement is combined with the participation of team members and a larger weight for team performance in their compensation. # 1999 Elsevier Science Ltd. All rights reserved. Today's organisations face complex and dynamic environments that have been attributed to increases in the globalization and competitiveness of the global economy (Mohrman, 1993). Several authors have argued that success in today's complex environments will be predicated on organisations becoming fast, ̄exible, and knowledge-based, and that one method of developing these characteristics is by employing teams (Cohen, 1993; Lawler, 1993;Mohrman&Mohrman, 1993). Teams can help organisations cope with these environments in two ways. First, teams improve an organisation's speed and ̄exibility of response. Second, team processes can improve the quality of the response. These bene®ts are achieved by focusing the skills and knowledge of appropriate organisational personnel on critical problems in a timely manner (Cohen, 1993). Both Lawler (1986) and Ledford (1993) identify four elements necessary to team success. These are: the provision of an adequate information base, rewards for performance, the extension of power to team members to make decisions, and the development or presence of the requisite abilities in team members. Management accounting systems can play an important role in team success because it can contribute toward ful®lling the ®rst three of these elements. Performance measurement can form an integral part of the information base necessary to team success and performance measurement is necessary if members are to be rewarded for team performance. Further, providing team members with the power to make decisions, Accounting, Organizations and Society 24 (1999) 263±285 0361-3682/99/$Ðsee front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S0361-3682(98)00060-9 * Corresponding author. suggests that team members participate in de®ning what constitutes performance and in setting performance targets. To adequately capture all the relevant information that team members require to succeed in managing complex environments, multiple types of performance measures are likely to be needed (Kaplan & Norton, 1992). This suggests that in a team context management accounting systems should be both expanded and modi®ed. Expansion is necessary to provide measures and systems that report directly on teams as an organisational unit. Modi®cation is necessary to incorporate the diversity of types of measurement necessary to capture team performance. If properly done, performance measurement should facilitate the e€ectiveness of the increasingly important organisational phenomenon of teams. The following questions are addressed: Should team performance be measured, and if so, how? Is it important to integrate team performance into individual compensation? What role does participation in setting performance targets play? Does employing team performance measures improve team performance? The results reported here are consistent with the organisational theory summarized above. We ®nd that as complexity increases organisations increasingly rely on teams. Teams that feature comprehensive performance measurement, with both ®nancial and non-®nancial performance measures, and encourage team members to participate in developing performance targets, perform better than those that do not. The performance improvement to be gained from performance measurement can be divided into direct and indirect components. Increasing the extent of performance measurement and participation directly increases team performance, presumably by providing useful information to the team members to help manage their tasks. In addition, there is an indirect e€ect of performance measurement on team performance because comprehensive measurement allows ®rms to increase the weight of team performance in team members' compensation. In turn, increasing the sensitivity of team members compensation to performance provides an incentive that leads to an incremental signi®cant increase in team performance. While we ®nd that measuring performance with only ®nancial or only non-®nancial measures or with less participation contributes to team performance, clearly the best team performance comes from combining comprehensive measurement with high participation. The latter combination provides an improvement in team performance that is markedly higher than would be suggested by simply summing the separate e€ects of performance measurement and participation. Thus, it appears to be critical to have all elements present to create the best opportunity for successful team performance. The paper ®rst develops the validity of the link between teams and complexity. This link forms the basis for the model and hypotheses concerning teams, performance measurement, compensation, and team performance. Participation in setting performance targets is incorporated as a moderating variable that in ̄uences the ecacy of performance measures both for setting compensation and for facilitating team performance. Next, we describe the sample and method. Then, the results are presented using both a typical path analysis that assumes linear relationships and interval scale variables, and a more general treatment that allows for non-linear relationships among the variables. The paper concludes with a discussion of the results and their implications. Teams can also be formed for motivational purposes, under the view that a more cooperative and socially interdependent work environment will motivate employees. Motivation within a team may also come from allowing the team a greater degree of self-management; however, allowing greater responsibility to the employee can be undertaken outside a team structure through job enrichment. The motivation rationale is complementary to the complexity rationale since complexity is likely to lead to the need for cooperative interactions to solve problems and greater responsibility for the employees, which in turn provides a legitimate basis for motivation. We stress complexity here because the organisational literature emphasizes that employee involvement is a solution to increased complexity and we can outline conceptual links between complexity and the need for diverse performance measurement. To the degree that teams are formed in non-complex environments purely for motivational purposes our model is incomplete; however, in this case there should be a bias against ®nding results for the hypotheses pertaining to performance measurement that follow. 264 T.W. Scott, P. Tiessen / Accounting, Organizations and Society 24 (1999) 263±285