Introduction

The story of procrastination due to hyperbolic discounting is an intriguing one, but it seems to require the procrastinator to be rather dim—or in the words of economists, “naïve.” Owing to hyperbolic discounting, people might realize that they should go on a diet but put off the diet until tomorrow given the steep discount they give to utility of consumption the next day. But when tomorrow arrives, that steep discount is applied to the next day, resulting in putting off the diet one more day. Exactly how many days in a row can people put off their diet before they realize their behavior is preventing them from achieving their goal? Similarly, how often can one put off studying until the last day before the exam without noticing how the behavior affects one’s performance? And if the decision maker does start to recognize the problem of time inconsistency, what would his reaction be?

Consider first the would-be dieter. After several days of wanting to go on a diet, but putting it off “just this once,” the dieter might realize “just this once” has become an eternal excuse. What he needs is some way to enforce his current preferences on his future self. The dieter might then race through the kitchen and decide to throw away all foods that would tempt him to break his diet tomorrow. Ice cream and cookies are thrown out, with maybe a few eaten along the way, so that tomorrow the dieter will not be able to break his diet except through extreme exertion—enough exertion that it would not be attractive even to a hyperbolic discounter. If the would-be dieter can make the cost of breaking his plan high enough, he will implement the plan, obtaining the long-term goal at the expense of the short-term splurge.

Commitment mechanisms are ubiquitous in our economic lives. Some might exist as a way to allow actors to assure others that they are negotiating in good faith, such as a contract for labor, but others seem incompatible with the rational model of decision making. Commitment mechanisms reduce the set of possible choices available in the future. In all cases, a rational decision maker would consider a reduction in the choice set to be something that at best leaves the decision maker no better off. But then, no rational decision maker would ever consider paying to engage in a commitment mechanism. Consider the familiar story of Odysseus, who with his crew must sail past the Sirens. He is warned by Circe that the Sirens’ song is irresistible and that all who hear it are led unwittingly to an ignominious death as their boat is wrecked on the jagged rocks near the island. No sailor has ever escaped their temptation, even though he might have known of the legend. Odysseus commands his men to fill their ears with wax so that they cannot hear the song and thus cannot be tempted, though they would still have the choice of where to steer the ship. Odysseus further commands his men to tie him to the mast so that he cannot escape. While thus tied to the mast, he could be exposed to the temptation but would not have the choice to steer the ship toward the Sirens’ song. He could thus become the only man to have heard their song and lived. By restricting his choice set, he could indulge in the song without giving in to the temptation that accompanied it. In this case, he could consider himself better off for making the commitment because he expected to do something that would harm him if he was not thus committed. In other words, Odysseus must have believed he would make the wrong choice in order for being tied to the mast to make sense. In a similar sense, commitment mechanisms can be used by those who recognize their own time inconsistency to tie themselves to the mast.