Naivete, Projection Bias, and Habit Formation in Gym Attendance

We implement a gym-attendance incentive intervention and elicit subjects' predictions of their postintervention attendance. We find that subjects greatly overpredict future attendance, which we interpret as evidence of partial naivete with respect to present bias. We find a significant postintervention attendance increase, which we interpret as habit formation, and which subjects appear not to predict ex ante. These results are consistent with a model of projection bias with respect to habit formation. Neither the intervention incentives, nor the small posttreatment incentives involved in our elicitation mechanism, appear to crowd out existing intrinsic motivation. The combination of naivete and projection bias in gym attendance can help to explain limited take-up of commitment devices by dynamically inconsistent agents, and points to new forms of contracts. Alternative explanations of our results are discussed.