Purpose The purpose of this paper is to examine whether managers punish more and work harder in teams with peer monitoring when it is less costly to punish in a two-period, one-shot horizon. Design/methodology/approach An experiment is conducted in a two-period horizon with two treatments. The structure of performance measures makes it costless or costly to punish in the second period. Findings The results find punishing, contingent on first-period strategies, was significantly greater when it was costless compared to costly, as expected. Working, which is analogous to cooperating in prisoner dilemma games, was also significantly greater in the first and second periods when punishing was costless. Practical implications This paper is informative about the potential benefits of performance measures in dynamic team environments, which can be challenging and costly to develop. It adds insight into the design of self-discipline and tasks in teams which might help increase productivity. Originality/value This paper is related to the research on indefinite horizons, which attributes increases in cooperation to the existence of subgame perfect strategies to cooperate and potential gains from future cooperation. In comparison, this study examines the effects of the existence of subgame perfect strategies to work in isolation from the potential gains from future interactions. In addition, it examines whether their potential benefits depend on the cost of punishing when punishing is subgame perfect in a one-shot horizon.