The Psychology of Organizations · Decision-Making
Escalation of Commitment
Why leaders pour more resources into a losing course precisely because they chose it.
Every organization has the project everyone privately knows should stop, and keeps funding. It is rarely a failure of intelligence. It is a predictable feature of how capable people respond to a decision that has become theirs.
Research Foundation
- Organizational behavior and decision research
- Behavioral economics
- Cognitive psychology
- Executive and enterprise field experience
- The Builders Build Framework
Key Takeaways
- People commit more to a failing course of action precisely because it is theirs. Escalation of commitment turns a recoverable mistake into an unrecoverable one, and it grows stronger the more public and personal the original decision was.
- The engine is loss aversion and the sunk cost effect: honoring resources already spent that cannot be recovered, and refusing to book a loss, so the organization keeps paying to avoid admitting it already paid.
- Because escalation is a failure to update in the face of new evidence, it is an Adaptability failure, and the fix is structural: stop criteria set in advance, and separating the people who decide to continue from those who chose to begin.
The plane that could not be stopped
The Concorde was a marvel and a money pit. Well before it entered service, it was clear to its British and French government backers that the supersonic airliner would never recover its costs. They continued anyway, year after year, in part because so much had already been spent and abandoning the project would mean admitting that the spending had been wasted. The pattern became so emblematic that behavioral economists named it the Concorde fallacy. The decision to continue was not driven by the future the plane would produce. It was driven by the past it had already consumed.
Every organization has its Concorde: the initiative everyone privately knows should stop, that keeps receiving budget and belief because stopping would make the earlier commitment look like a mistake. The psychology of why is precise and well documented.
Why the losing bet gets more, not less
Barry Staw's foundational work showed that decision-makers escalate commitment to a chosen course specifically when they feel responsible for the initial choice, throwing good resources after bad to justify the original decision. Kahneman and Tversky's prospect theory supplies the engine: losses are felt about twice as intensely as equal gains, so a leader will take a bad gamble to avoid locking in a certain loss. Arkes and Blumer demonstrated the sunk cost effect directly, showing people persist with a course because of unrecoverable past investment that should be irrelevant to the forward decision. Joel Brockner integrated these into a model of escalation as a self-justification process. The through-line is that the more the decision is yours, the harder it is to reverse.
No one funds a failing project because they are foolish. They fund it because stopping would require saying the earlier decision was wrong, and organizations make that the most expensive sentence anyone can speak.
Why organizations escalate hardest of all
Individuals escalate; organizations escalate with force. The person who launched the project is often the person judged on it, so admitting failure is a personal cost, not just an accounting one. Reputations, budgets, and careers become attached to continuation. Reporting flatters the project because no one wants to deliver the bad news that ends it. The organization thus builds a structure in which the truth that would stop the project is the most expensive thing anyone can say, and so it does not get said.
What this means through the Four A's
Escalation is an Adaptability failure, the refusal to reconfigure once the evidence has moved. Adaptability is the capacity to update structure, not just tactics, and stopping a committed course is the hardest update of all because it admits the earlier configuration was wrong. The research points straight at the remedy, and it is structural rather than motivational. Decide the stop conditions before you start, when no ego is attached. Separate the decision to continue from the people who chose to begin. And make killing a failing project a mark of good judgment, not a career wound.
From bias to constraint
The pattern beneath this is general: an individual bias, left unexamined, is adopted, reinforced, and encoded until it becomes a structural limit on the whole organization. The Four A's are where a leader interrupts the chain.
Individual perception
A person reads a situation through their own biases and priors.
Shared interpretation
Colleagues adopt the same read, and it begins to feel like fact.
Group reinforcement
Cohesion and deference harden it; dissent is quietly filtered out.
Process and policy
The interpretation is encoded into how decisions and work get made.
Organizational constraint
What began as one bias is now a structural limit on the whole system.
The Four A's intervention
Attention
What are we failing to notice?
Alignment
Where are our interpretations diverging?
Authority
Who can challenge or change the pattern?
Adaptability
What evidence would cause us to revise?
Evidence matrix
| Claim | Research | Field evidence | Four A's |
|---|---|---|---|
| Responsibility for a choice drives escalation | Staw (1976); Brockner (1992) | Owners defending their own initiatives | Adaptability |
| Loss aversion and sunk cost fuel persistence | Kahneman & Tversky (1979); Arkes & Blumer (1985) | Budget follows past spend, not future value | Adaptability, Authority |
| Bad news that would stop it goes unsaid | Royer (2003) | Projects nobody will kill | Adaptability, Attention |
Executive reflection questions
- Which project in your portfolio would you not start today, knowing what you now know, but keep funding anyway?
- For your biggest bets, did anyone set explicit conditions under which you would stop, and are you tracking them?
- Who would pay a personal price if a major initiative were killed, and are they the same people deciding whether to continue?
- In your culture, is stopping a failing effort treated as good judgment or as failure?
Builder actions
Design the stop before you need it. At the start of any major commitment, write down the conditions under which you will halt or reverse, while judgment is still cold and no reputation is attached. Separate the continue decision from the originators by giving a different group the authority to review and kill. Fund in staged tranches tied to evidence, not in a single act of faith. Run a scheduled kill review that asks the forward question, would we start this today, and treat a well-reasoned stop as a win worth rewarding. The goal is to make reversing a committed course a normal, respected move rather than an admission of guilt.