Intrinsic motivation, health outcomes and the crowding out effect of temporary extrinsic incentives: A lab-in-the-field experiment

There is substantial international policy interest in incentivising healthy be-haviours. When considering extrinsic incentives, policymakers should be mindful of the potential for crowding out of intrinsic motivation. Focusing on monetary incentives, there is conflicting evidence on whether it is better to go big (Gneezy and Rustichini, 2000b) or to go small (Pokorny, 2008). In this paper, we inves-tigate the effect on intrinsic motivation of a range of extrinsic incentives, which vary by size and type, both during their application and after their removal. Ad-ditionally, we investigate whether intrinsic motivation predicts health outcomes. The laboratory experiment is comprised of a rich within and between subject de-sign that allows us to estimate a difference in differences model of the treatment effects. Our subject pool was a heterogeneous adult population. Subjects were given four time limited rounds of a real effort task. Round 1 measured intrinsic motivation. Extrinsic incentives were applied in round 2, varying between sub-jects. Extrinsic incentives were removed in rounds 3 and 4 to measure crowding out and persistence of crowding out. On average, we find support for the “pay –but do not pay too much” rule (of Pokorny, 2008). However, we find that “pay enough or don’t pay at all” (of Gneezy and Rustichini, 2000b) better fits the results for individuals with low motivation. The high powered monetary incen-tive is most likely to crowd out intrinsic motivation after its removal. Intrinsic motivation is found to partially explain waist-to-height ratio, highlighting the relevance of our findings for health policy.