What is escalation of commitment?
What a classic experiment on funding a failing division found, why the first decision holds on, and what research says helps a team stop.
Answer to the question
What is escalation of commitment?
Escalation of commitment is the tendency to put more into a course of action after it has started to fail, strongest when the person deciding chose that course. In Barry Staw's 1976 experiment, 240 business students who had chosen a failing division put the most money into it. Targets set in advance and judging process help teams stop.
Key takeaways
- In Barry Staw's experiment, 240 business students allocated research funds; those who chose a division themselves and then saw it do badly committed the most resources to it in a second round.
- Daniel Kahneman's Thinking, Fast and Slow treats an unfinished project as an open mental account that stopping would close at a loss.
- A preregistered replication gave partial support to Staw's findings on personal responsibility and decision consequences (Dhanda and colleagues, 2025).
- Simonson and Staw found three procedures most effective at reducing escalation: making negative outcomes less threatening, setting minimum targets in advance, and judging process and not only outcome.
- In three experiments, affirming an unrelated strength or a cherished value reduced reinvestment in a failing choice, while affirming the very ability the decision had questioned raised it (Sivanathan and colleagues, 2008).
Responsibility for a first choice can make a bad bet grow, a pattern called escalation of commitment. In a classic study, business students who chose a division to back, then watched it do badly, put more money into it than any other group, even though the results gave them every reason to pull back. The news pointed one way and the money went the other.
Barry Staw ran the experiment decades ago, and the students played a financial vice-president handing out research and development funds. Once the results came in, everyone received a second round of money to allocate. The students who had made the first call themselves, and who then saw poor results, committed the most.
How escalation of commitment shows up in daily work
In Staw's design the sunk money was the same for everybody. What differed was authorship. Some students picked the division themselves, and others were told which division had been chosen for them. Both groups saw the same poor results and faced the same second allocation. Staw linked the extra investment in this escalation of commitment to self-justification: for the person who made the first decision, adding more is a way of showing that the first decision was sound.
That difference between the groups is the useful part, because it separates the pattern from a plain dislike of waste. It is why the idea sits beside What is the sunk cost fallacy, which turns on spending already made, whoever made it, and not inside it.
Everyday versions are easy to picture. A product team has been building for a long time, the early numbers are thin, and the request is for more time to finish the feature. A household has started a renovation and keeps widening the scope to justify the first stage.
Authorship is the thread in each, and the same pull works on a smaller scale for the viewer who keeps going with a season that stopped being good, the case taken up in Why You Finish the Series You Stopped Enjoying. In each case a person or team owns the original choice, and the choice is still open for review. What keeps it open is not new evidence for the plan but the cost of admitting that the first call needs changing, a cost that grows with every quarter, room or episode already given to it.
Why the first decision keeps drawing in more money
Daniel Kahneman gives one reason for escalation of commitment in Thinking, Fast and Slow. A project in progress is an open mental account, and stopping it closes the account at a loss. He places the choice among bad options in the cell of his fourfold pattern where a likely loss makes people risk seeking, so continuing becomes the gamble that postpones booking the loss. The mechanics of that bookkeeping are set out in What is mental accounting.
Staw's account, built on a class of 240 students, adds who made the call. Kahneman's reading says why a loss hurts, and Staw's says why it hurts more to the person who chose the course. The two fit together, and neither requires bad intent. A manager who keeps funding a plan is protecting a judgement, often a sound one made on the evidence available at the time.
Theories of escalation of commitment had multiplied over more than 35 years, and few studies had weighed them against each other, so Dustin Sleesman and colleagues brought the determinants into one meta-analysis and assessed how well the main theoretical perspectives account for the behaviour. The responsibility effect itself has also been tested again, in preregistered replications of three influential studies on escalation and regret by Nandita Dhanda and colleagues, whose first study gave partial support to Staw's findings on personal responsibility and decision consequences. The reading that fits both results is that authorship matters and that the size of its effect depends on how the situation is set up.
Ways a team can stop in time before losses grow
Itamar Simonson and Staw compared several strategies for reducing escalation of commitment to losing courses of action. Earlier work had mostly documented how people become locked in, and this comparison asked which ways out work. Three procedures worked best.
- Make negative outcomes less threatening to the decision maker.
- Set minimum target levels in advance, so that missing them triggers a change of policy.
- Judge decision makers on the quality of their process and not only on the outcome.
Each removes a reason to defend the first choice, which is the engine of escalation of commitment on Staw's account. A target written before the money is spent turns a later retreat into the plan working as designed. Judging process leaves room to say that a sound decision met bad luck, and a team that can say so out loud has less need to keep spending to prove otherwise.
Kahneman's book also endorses a method from Gary Klein, the premortem. Before a group has formally committed, the people who know the decision gather, and the leader asks them to imagine that a year has passed, the plan was carried out, and the result was a disaster. Each person writes a short history of that failure. Kahneman describes the value as making doubt legitimate, since even supporters are invited to look for threats they had not considered, and he calls it a partial remedy and not a cure. It is a close relative of the rehearsal methods weighed in Does Mental Rehearsal Help a Decision.
Niro Sivanathan and colleagues tested a different kind of help in three experiments on de-escalation. Participants who were reminded of their own worth through a cherished value, a high self-esteem score or an unrelated strength reinvested less in a failing choice. In one experiment, those who wrote about their most important value reinvested less than those who wrote about their least important one. When the affirmation praised the very ability the failed decision had questioned, commitment went up.
Common questions
What is escalation of commitment?
It is the tendency to put more into a course of action after it has begun to fail, strongest when the person deciding chose that course. Staw described it in 1976 with 240 business students allocating research funds.
Is escalation of commitment the same as the sunk cost fallacy?
No. The sunk cost fallacy concerns spending that cannot be recovered. Escalation of commitment concerns the personal link between the decision maker and the first choice, which Staw's experiment varied directly.
Has the original finding been replicated?
A preregistered replication reported partial support for Staw's findings on personal responsibility and decision consequences, so the pattern holds with qualifications.
How can a team stop backing a failing plan?
Studies of de-escalation point to targets set in advance, judging process as well as outcome and making losses less threatening. Kahneman also describes the premortem as a partial remedy.
Sources
Kahneman, D. Thinking, Fast and Slow. Seed book; vault hub Books/Thinking Fast and Slow.
Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Performance, 16(1), 27-44. doi:10.1016/0030-5073(76)90005-2 https://doi.org/10.1016/0030-5073(76)90005-2
Simonson, I., & Staw, B. M. (1992). Deescalation strategies: A comparison of techniques for reducing commitment to losing courses of action. Journal of Applied Psychology, 77(4), 419-426. doi:10.1037/0021-9010.77.4.419 https://doi.org/10.1037/0021-9010.77.4.419
Sivanathan, N., Molden, D. C., Galinsky, A. D., & Ku, G. (2008). The promise and peril of self-affirmation in de-escalation of commitment. Organizational Behavior and Human Decision Processes, 107(1), 1-14. doi:10.1016/j.obhdp.2007.12.004 https://doi.org/10.1016/j.obhdp.2007.12.004
Sleesman, D. J., Conlon, D. E., McNamara, G., & Miles, J. E. (2012). Cleaning up the big muddy: A meta-analytic review of the determinants of escalation of commitment. Academy of Management Journal, 55(3), 541-562. doi:10.5465/amj.2010.0696 https://doi.org/10.5465/amj.2010.0696
Dhanda, N., Ziano, I., Ching, K. H., Phuong, A. K. L., Lo, Y. T., Yip, M. K., Shi, B. J., & Cheng, B. (2025). A replication and extension of three studies investigating escalation of commitment and regret aversion. Personality and Social Psychology Bulletin. doi:10.1177/01461672251345021 https://doi.org/10.1177/01461672251345021