Thaler has described the self-control problem as a tension between what are essentially humans’ two selves: the myopic short-run self that wants everything RIGHT NOW! and the disciplined long-run self that can bear the costs of waiting for a payoff. In a recent working paper, Supreet Kaur, Michael Kremer, and Sendhil Mullainathan ask how the workplace can assist or sabotage the best efforts of our long-run selves.
They develop a series of experiments at an Indian data entry company in which workers can choose to be paid in two ways: 1) A simple straightforward rate with a payment for each data entry field they complete; 2) A commitment contract that allows the worker to set some data entry target. If they hit the target, workers receive a full payment for each data entry field completed. If they don’t make it, they receive half a payment for each data entry field completed.
Econs should choose the second contract, pick a target of 0 and earn, well, whatever they can filling out data entry fields. Humans who recognize their self-control problems might choose the second contract and set more ambitious targets in order to boost their performance.
When offered the choice of (the second) commitment contract, workers choose positive targets 35 percent of the time, with targets being set at a non-trivial level. In addition, simply being offered the choice of a commitment contract increases production and wages by 2.3 percent on average relative to the control contract.
Because the choice of contracts is randomized, the authors are also able to look at the social effects of fellow workers.
(H)aving a peer with above average productivity increases own productivity by 5 percent. We further find that the effect on productivity appears to operate through increases in work hours rather than productivity per unit of time.
Full pdf is here.