Commitment device
Also known as: Ulysses contract, Pre-commitment
Binding your future self to the plan your present self wants.
What it means
A commitment device is a voluntary arrangement made in advance that restricts one's own future options or raises the cost of deviating, so that a far-sighted present self can constrain a future self likely to be tempted. The mechanism addresses time inconsistency and present bias: because preferences predictably shift as temptation draws near, the present self pre-commits to make the tempting choice harder, more expensive, or impossible. Devices range from hard commitments that physically remove the option — automatic transfers into locked savings, software that blocks distracting sites — to soft commitments that merely add stakes, such as public pledges, deposit contracts, or financial penalties for failure. A subtle design point is that effective devices balance bite against flexibility: too weak and they are ignored, too rigid and they backfire when circumstances legitimately change, so the best ones impose cost on backsliding without making adaptation impossible. They matter for saving, dieting, exercise, addiction recovery, and deadlines, and field experiments confirm that even modest self-imposed stakes can meaningfully improve follow-through.
What the evidence shows
Field experiments give the concept its empirical spine. In the Philippines, Ashraf, Karlan and Yin offered bank clients a SEED account that locked deposits until a self-chosen goal; take-up was modest at 28 percent, but a year later those offered it had saved 81 percent more than controls. Gine, Karlan and Zinman's CARES contract asked smokers to deposit their own money, forfeited to charity if a later urine test found nicotine; those offered CARES were several points more likely to have quit, and the gain survived a surprise test at twelve months. Milkman's temptation bundling, allowing gripping audiobooks only at the gym, lifted attendance. The stakes need not be large, but these effects sit alongside low take-up: most people offered a device decline it.
Who actually wants one
Demand for commitment is itself a clue about self-knowledge. Standard models predict that only a sophisticate, someone who foresees that their future self will be tempted, will pay to tie their own hands; a naif who wrongly expects to resist sees no reason to. So the people who most need a device are often the least likely to buy it, and the take-up rates seen in field studies, frequently in the teens or twenties, reflect that gap. Bryan, Karlan and Nelson distinguish hard commitments, which remove the option outright, from soft ones, which only attach a psychological or reputational cost. Soft commitments lean on loss aversion and the wish to appear consistent, which is why a public pledge or a small deposit can move behavior with no external enforcement at all.
When it backfires
The honest risk is that a commitment binds you at the wrong moment. Because a hard device cannot tell temptation from a genuine emergency, a locked savings account or a rigid penalty can punish someone hit by a real shock, a medical bill or a lost job, rather than a lapse of will. Anett John's field experiment sharpened the worry: offered a savings contract with a self-chosen penalty, a majority of clients defaulted and lost money, apparently having picked stakes too weak to change the behavior yet strong enough to hurt. Mis-set incentives can leave people worse off than with no device at all. Sophistication therefore cuts both ways: you must predict not only that you will be tempted, but how much bite you actually need.
Designing one that holds
Good design starts by matching hardness to the problem. Reserve hard, option-removing devices for temptations you know you cannot out-argue in the moment, and prefer soft stakes where circumstances might legitimately change. Make the committed path the default, so that following through takes no decision and only deviating does; automatic transfers and pre-scheduled site blockers work this way. Set penalties large enough to sting but capped so a bad month stays survivable, and give the plan a date rather than an open-ended intention. Temptation bundling offers a gentler route, pairing the unloved task with an indulgence instead of a punishment. Above all, build an escape valve for true emergencies: a device that cannot bend to real need is one people quietly abandon, or come to regret having signed.
Examples
Putting savings in an account with a withdrawal penalty, or staking money you lose if you skip the gym.
Odysseus had himself lashed to the mast so he could hear the sirens and still not steer toward them; the rope did what his future self could not.
Charging your phone in the hallway needs no willpower at midnight. It needs one decision at nine, while scrolling in bed still sounds like something you could skip.
A worker signs up so every future raise is routed into retirement savings before it lands in her paycheck; she never feels the money as spendable, so there is nothing to resist.
A doctoral student hands a friend a signed check made out to a cause she despises, to be mailed only if her dissertation chapter misses Friday's deadline.
First described in Schelling; Thaler & Shefrin (1981).
Key references
- John, A. (2020). When Commitment Fails: Evidence from a Field Experiment. Management Science, 66(2), 503-529. doi.org/10.1287/mnsc.2018.3236
- Milkman, K. L., Minson, J. A., & Volpp, K. G. M. (2014). Holding the Hunger Games Hostage at the Gym: An Evaluation of Temptation Bundling. Management Science, 60(2), 283-299. doi.org/10.1287/mnsc.2013.1784
- Gine, X., Karlan, D., & Zinman, J. (2010). Put Your Money Where Your Butt Is: A Commitment Contract for Smoking Cessation. American Economic Journal: Applied Economics, 2(4), 213-235. doi.org/10.1257/app.2.4.213
- Bryan, G., Karlan, D., & Nelson, S. (2010). Commitment Devices. Annual Review of Economics, 2, 671-698. doi.org/10.1146/annurev.economics.102308.124324
- Ashraf, N., Karlan, D., & Yin, W. (2006). Tying Odysseus to the Mast: Evidence from a Commitment Savings Product in the Philippines. Quarterly Journal of Economics, 121(2), 635-672. academic.oup.com/qje/article-abstract/121/2/635/1884028
- Schelling, T. C. (1984). Self-Command in Practice, in Policy, and in a Theory of Rational Choice. American Economic Review, 74(2), 1-11. ideas.repec.org/a/aea/aecrev/v74y1984i2p1-11.html