8.4 Tournaments and promotions in teams
While the literature on tournaments provides strong support for the proposition that incentives to increase the value of the firm result from providing differential reward for differential effort, the application of this result in teams is less clear-cut. In teams, individual effort may not be observable and the output of each team member may be complementary to that of other team members (each member's output is increasing the output of other team members). This has led to a presumption that rewarding all team members equally may be optimal. The arguments marshalled in support of equal rewards include lower transaction and monitoring costs, increased peer pressure for equal contributions within the team, and avoidance of moral hazard problems (for example, Milgrom and Roberts, 1992).
However, the presumption for equal treatment of team members has long been questioned. Lazear (1989) observed that it was far from obvious that equal pay had these effects, in part because equal pay does not account for heterogeneity in agents' ability and performance (that is, all team members will not be of equal ability even if the manager cannot identify the differences in ability), and the lack of a link between pay and the individual's marginal product can provide incentives for free-riding. Recent work has provided a coherent theoretical framework for Lazear's observations.
Winter (2004), for example, outlined a model of teams in which the maximum level of efficiency is achieved when differential (not symmetrical) rewards are applied to individual team members with complementarity between the outputs of individuals. In contrast, the maximum levels of efficiency are achieved when symmetrical rewards are applied to teams where there is substitutability between the outputs of individuals. Georg and Kube (2010) provide experimental evidence to support this observation.
The intuition behind the findings of these authors is that asymmetric rewards facilitate coordination in teams where outputs are complementary, because the fact that one team member stands to get a large reward means that at least one team member will exert substantial effort. Where effort is complementary, the other individuals anticipate this high level of effort from the highly paid member, and exert higher levels of effort, anticipating that other members of the team will react in the same way. Conversely, with symmetric payoffs team members anticipate that other team members will not exert a high level of effort, and they choose low effort.
In the case of teams where outputs are substitutes and rewards are unequal, all team members expect the team member receiving the lowest reward to shirk. As a result, all team members reduce effort. However, equal rewards result in each team member expecting that the other team members will exert effort.
The general application of this literature is to say that the design of teams and the reward structures may influence the level of output achieved. Unequal rewards for members of teams (for example, if they have different levels of bonus for the same level of output) may not reduce efficiency and will promote efficiency compared with equal rewards in cases where effort in the team is complementary. This appears to be relevant to the public sector in New Zealand, where, as Scott (2001:27) notes, there has been a tendency to argue that individual performance bonuses (as one type of differential reward) might be counter-productive in an environment where individual performance is not easily determined and teamwork is essential for organisational performance.
