Moral hazard in teams with subjective evaluations

This article studies moral hazard in teams where workers' wages depend on two types of performance measures: objective team output and subjective evaluations. The evaluations include both self- and peer evaluations. I find that when evaluations become less subjective, workers' wages should be more sensitive not only to their evaluations but also to the team's output. I also show that subjective evaluations should be used in relative terms only when the degree of subjectivity is not too high. The results also associate the introduction of a budget breaker with the level of subjectivity within organizations.