Testing the Economic Model of Crime:The National Hockey League's Two-Referee Experiment

Abstract During the 1998-99 season, the National Hockey League randomly varied the number of referees used across games, seemingly providing a rare opportunity to test directly the deterrence model. Combining experimental parameter estimates with an economic model, there is little evidence that the rate of offending changed substantially with the addition of a second referee. The reason, however, appears to be that the second referee had little impact on the probability of punishment. As a consequence, the experiment ultimately turns out to be of limited use for testing deterrence.