Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
In many collective action problems individuals' contributions increase the probability of a collectively favorable event rather than affect the quantity of public goods provided. Such problems, that we refer to as collective prevention games, remain largely unexplored in the literature. We fill this gap by setting up an experiment where subjects' contributions increase the probability of a fixed collective benefit or reduce the probability of a fixed negative externality. Our main result is a substantial increase in cooperation in the probabilistic loss environment compared to the deterministic one. We explore some behavioral mechanisms that could drive this result.