You're 45 minutes into a movie you don't like. You could turn it off and do something enjoyable. But you've already invested 45 minutes, so you keep watching.
You've spent two years in a career that makes you miserable. You could change course. But you've already put in two years, so you stay.
This is the sunk cost fallacy — the tendency to continue a course of action because of resources already invested rather than because of future expected value. It is one of the most studied cognitive biases in behavioral economics, and it affects decision-making at every scale from personal choices to billion-dollar corporate strategies to military campaigns.
Why it's irrational
Sunk costs — money, time, effort already spent and irrecoverable — should be irrelevant to forward-looking decisions. The only question that matters economically is: given my current situation, what choice produces the best future outcomes?
The 45 minutes already watched can't be unspent whether you continue watching or not. The only relevant question is whether the next 90 minutes of this movie is worth more than whatever else you'd do with them. The past investment is gone either way.
Continuing because of past investment is allowing irrelevant information to influence a decision. It produces worse outcomes by definition — you're choosing based on something that doesn't change the forward-looking calculus.
Why we do it anyway
The psychological mechanisms behind sunk cost thinking are well-documented.
Loss aversion. Kahneman and Tversky's prospect theory established that losses feel roughly twice as painful as equivalent gains feel pleasurable. Walking away from a sunk cost feels like accepting a loss. Continuing feels like preserving the possibility of recovery. The pain of acknowledging the loss drives continued investment even when that's irrational.
Commitment and consistency. Once we've publicly committed to a project or position, abandoning it requires admitting the original decision was wrong. The desire for self-consistency — and to be seen as consistent by others — creates resistance to changing course.
Effort justification. The harder something was to achieve, the more we value it. This is adaptive in some contexts (things worth having often require effort) but produces sunk cost thinking when it leads us to overvalue poor investments simply because we worked hard on them.
Waste aversion. There's a strong cultural norm against waste. "Don't waste" is taught early and deeply. Stopping a project feels like wasting everything already invested, even though the invested resources are equally wasted whether you continue or stop — you just waste more by continuing.
Real-world consequences
The sunk cost fallacy isn't just a personal finance problem. Its effects scale.
Military campaigns. The escalation of US involvement in Vietnam fits the sunk cost pattern: each increase in commitment was partly justified by the need to honor previous investments. The more had been spent, the harder it became to withdraw. Political scientists have documented similar dynamics in numerous conflicts where early losses generated pressure to continue rather than cut losses.
Corporate projects. The Concorde supersonic passenger jet continued development well past the point where engineers and economists recognized it was unlikely to be commercially viable, partly because the British and French governments had already committed so much that stopping felt politically impossible. The Concorde fallacy is now used as a term for sunk cost thinking in project management.
Relationships. People stay in poor relationships partly because of time already invested. The calculation "but we've been together for five years" isn't a reason to stay — the five years are spent regardless. The relevant question is what the next five years look like.
The zero-based mindset
The corrective to sunk cost thinking is evaluating decisions as if starting fresh. "If I hadn't already invested in this project, would I choose to start it now based on what I know?" is often clarifying. If the answer is no, the past investment is a reason to continue only if you believe prospects have improved.
This is easier said than done. The psychological forces driving sunk cost thinking are strong and largely automatic. Awareness of the bias reduces but doesn't eliminate it.
What helps: separating the decision from the person. The sunk cost fallacy is easier to see in other people's decisions than your own. When evaluating your own decisions, try to articulate the forward-looking case for continuation without reference to past investment. If you can't make that case, the sunk cost may be doing more work than you realize.
The money is spent. The time is gone. The question is always: what now?