The Wall Street Journal reports on a new study published in the American Economic Review. Sebastian Kube, Michel André Maréchal and Clemens Puppe conducted an experiment, in which they compared workers' productivity when given a bonus vs. those in a control group. Actually, they set up two different types of bonuses: a 7 Euro cash bonus and a gift of a thermos worth 7 Euros. They found that the workers promised the gift as a bonus were significantly more productive than those given cash, whether told the value of the thermos or not.
Now, one could conclude that the researchers have shown that gifts might have a better incentive effect in the workplace than cash. However, I think we need to proceed with caution - a great deal of caution! Here we have a simple experiment with the bonus only worth 7 Euros. Would the same effect hold in the workplace if the amount of the bonus were much more substantial? That's not clear to me at all. So, while the experiment may be thought-provoking, I'm not sure it provides us practical guidance as to how to design incentive schemes for the workplace.