Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We explore the possibility that demand for costly commitment may prove unnecessary and thus excessive. In an online experiment, subjects face a tedious productivity task where tempting YouTube videos invite procrastination. Subjects can pay for a commitment device that removes the videos with probability less than one, allowing us to compare their willingness to pay with realized material and psychological costs of temptation. A significant share of subjects do overestimate their commitment demand, being overly pessimistic about their performance when tempted. Even more subjects underestimate demand, and total realized losses from revealed undercommitment is greater than from overcommitment. Thus, undercommitment dominates and is inconsistent with pure noise in stated demand.