The Sunk Cost Trap: Why You Keep Investing in Losing Positions

You’re eighteen months into a project that was supposed to take twelve. The budget has overrun by 40%. The team is fatigued. The market conditions that made the project compelling have shifted. Every objective indicator says this should be stopped — or at minimum, fundamentally restructured. But the room shifts from whether to continue to how to recover what has already been invested “We’ve come too far to stop now” isn’t an argument. It’s a reflex. And it’s one of the most expensive reflexes in professional life.

The resources you’ve already spent — the time, the money, the political capital, the emotional investment — are gone. They cannot be recovered by continuing. They cannot be recovered by stopping. They are, by definition, sunk. The only question that matters is whether the future investment is justified by the future return. But that question is almost impossible to hear when the past investment is screaming.

The research

Hal Arkes and Catherine Blumer published the definitive experimental demonstration of the sunk cost effect in Organizational Behavior and Human Decision Processes in 1985. In one memorable study, participants who had purchased a theatre season ticket at full price attended more performances — including ones they didn’t enjoy — than those who had received a discount. The plays were identical. The only difference was the size of the past investment. The more people had paid, the more they felt compelled to “get their money’s worth,” even when the experience itself was no longer rewarding.

Arkes and Blumer ran multiple variations and found the effect was reliable across domains: people continued eating food they’d paid more for even when full, persisted with failing investments longer when more capital had been committed, and resisted abandoning projects in proportion to the resources already consumed — regardless of the project’s future prospects. The past expenditure had no bearing on the future value. But it had an overwhelming bearing on the decision.

Barry Staw, at UC Berkeley, identified the organisational dimension of this trap in a landmark 1976 paper in Organizational Behavior and Human Performance. He called it “escalation of commitment” — the tendency for individuals and organisations to increase their investment in a failing course of action precisely because of prior investment. Staw found that decision-makers who were personally responsible for the initial investment escalated more aggressively than those who inherited the situation. The sunk cost wasn’t just financial — it was reputational. Admitting failure threatened the decision-maker’s self-image and public standing.

Joel Brockner expanded this analysis in a 1992 review in the Academy of Management Review, identifying multiple reinforcing mechanisms: self-justification (continuing validates the original decision), completion effects (the perception that the goal is “almost” achieved), and modelling (when leaders escalate, subordinates follow). The trap is not a single cognitive error but a system of interlocking biases that make continuation feel rational and stopping feel like failure.

The mechanism

The psychological engine behind sunk cost reasoning is loss aversion — the same mechanism Kahneman and Tversky identified in their 1979 prospect theory. Stopping a failing project doesn’t just mean absorbing the future losses. It means realising the past losses. As long as the project continues, the prior investment exists in a kind of psychological limbo — not yet a loss, because the final outcome is still undetermined. The moment you stop, the loss becomes real and final.

Richard Thaler, writing in the Journal of Economic Behavior & Organization in 1980, introduced the concept of “mental accounting” to explain how people psychologically frame costs and benefits. People maintain separate mental accounts for different expenditures, and they resist closing an account at a loss. The theatre ticket isn’t just a past purchase — it’s an open account that the brain wants to balance. Attending the show “uses” the ticket and closes the account. Not attending leaves the account permanently in the red.

This explains why the sunk cost effect intensifies as the investment grows. Small sunk costs produce mild discomfort. Large ones — years of effort, millions in funding, careers built around a particular direction — produce a pain response that is functionally identical to physical injury. The anterior cingulate cortex, which processes both physical pain and social or financial loss, activates when people contemplate writing off a major investment. You’re not just making a calculation. You’re experiencing pain. And the brain’s natural response to pain is avoidance — which, in this context, means continuing.

The escalation trap is further reinforced by the confirmation bias. Once committed, decision-makers selectively attend to information that supports continuation and discount information that supports stopping. Every small positive signal is amplified (“see, it’s turning around”), while every negative signal is contextualised (“that’s a temporary setback”). The two biases work in concert: sunk cost reasoning keeps you committed, and confirmation bias keeps you believing the commitment is justified.

The question “would I start this today, knowing what I know now?” isn’t a thought experiment. It’s the only question that separates the future value of a decision from the gravitational pull of its past.

The practical implications

Covering the past investment is a literal technique, not a metaphor. When evaluating whether to continue a project, physically remove the historical cost information from the analysis. Create a decision document that starts from today: here is the remaining investment required, here is the expected return, here is the probability of success. If the remaining investment wouldn’t be approved as a new project, continuation is sunk cost reasoning in action.

Personal responsibility amplifies the trap. Staw’s research shows that the person who made the original decision is the worst person to evaluate whether to continue. They have the most to lose from admitting the initial decision was wrong. Wherever possible, have the continuation decision reviewed by someone who wasn’t involved in the original commitment. Their assessment will be less contaminated by self-justification.

Watch for the language of sunk costs. “We’ve come too far.” “We can’t waste what we’ve already put in.” “We just need a little more to make it work.” These phrases are diagnostic — they frame the decision in terms of past investment rather than future value. When you hear them, from yourself or others, flag them explicitly. The language itself is a signal that the reasoning has been captured by the sunk cost trap.

The bigger picture

Organisations are particularly vulnerable to sunk cost reasoning because the mechanisms that reinforce it are embedded in organisational culture. Persistence is valued. Quitting is stigmatised. Leaders who kill projects are seen as having “failed,” while leaders who push through adversity are celebrated — regardless of whether the adversity was a solvable obstacle or a clear signal to stop.

This creates a systematic bias toward overcommitment. Projects that should be stopped after six months run for three years. Strategies that should be revised are defended with increasing desperation. Careers are invested in directions that stopped making sense years ago, held in place by the accumulated weight of prior choices.

The most valuable skill in decision-making may not be the ability to commit to the right path. It may be the ability to abandon the wrong one — to look at what you’ve already invested, acknowledge that it’s gone, and ask the only question that matters: knowing what I know now, would I start this today? If the answer is no, the sunk cost has already told you everything it helps to know. The only remaining decision is whether you’re willing to hear it.