Rewarding someone for desirable behavior is not always a good idea. For example, small children are often inclined to help adults. After having been rewarded, however, they are less willing to help - unless a reward is forthcoming. This phenomenon also occurs with adults and has been observed in various psychological and economical experiments.

Wendelin Schnedler and Christoph Vanberg (University of Heidelberg) propose a simple economic explanation: whenever a child learns that his help has a 'price,' it is in his interest to drive this price up - even if he would actually be willing help without reward.

This 'playing hard-to-get' only makes sense if the activity cannot simply be taken over by someone else. If a father wants his son to tidy the room, playing hard-to-get only makes sense if the father is not willing to tidy the room himself (e.g. because he is mainly interested in a tidy room). The son can put no pressure on a father who is willing to tidy the room himself. He has no bargaining power and playing hard-to-get to drive up the price will not work.

The theory applies to various situations in which people are self-motivated. It has been published under the title 'Playing hard-to-get: An Economic Rationale for Motivational Crowding Out' in May 2014 in the European Economic Review (68) on pages 106-115. 

It is typically argued that paying large bonuses will induce people to exert more effort. This argument is, for example, evoked to justify the large bonuses of managers. Wendelin Schnedler puts forward a theory in which success bonuses actually reduce effort. This may happen even if more effort increases the likelihood of getting the bonus.

The theory is based on the assumption that people are less willing to exert effort when they get richer. In other words, they prefer playing golf, spending time with their family or on their sailing boat instead of sitting in their office. Given this assumption, success bonuses have two effects. On the one hand, there is an incentive effect. Tying a bonus to successful performance means that the manager now has stronger incentives to exert effort. On the other hand, there is an income effect. The bonus means that the manager can expect a larger income - even if she does not increase effort. A larger income, however, means that the manager's willingness to exert effort decreases.

Incentive and income effect need to be weighed against each other. If the manager has relatively little effect on success and the probability of success is high to begin with, the income effect dominates. Then, introducing or increasing bonuses will decrease effort.

The theory has been published in February 2011 under the title "You Don't Always Get What You Pay For: Bonuses, Perceived Income and Effort" in the German Economic Review 12 (1), pages 1-10.  Photo: Damien du Toit (creative commons).

When the U.S. Agency of International Development (USAID) paid villagers 5 dollar for each delivery of a bottle full of Colorado potato beetles trying to fight an infestation of these beetles in Afghanistan, locals were found breeding them. Incentives were scrapped. A plethora of similar examples in which agents "game" incentives but incentive theory does not offer a proper definition of "gaming," let alone a description how gaming affects incentives.

In a recent research paper, Wendelin Schnedler fills the gap: Incentives are gamed if effort is exerted but not used beneficially. Based on this definition, he shows that gaming can be avoided by aligning incentives. But aligning incentives is typically not optimal because incentives have two functions: preventing gaming and shirking. The gains from less gaming have to be traded against the losses from more shirking. The theory of gaming and shirking is relatively generally applicable and can explain why incentives are sometimes not used even if they increase effort-like in the beetle example.