Definition
The sunk cost fallacy is the tendency to continue an action or investment because of the resources (time, money, effort) already spent on it — even when the rational decision is to stop. Sunk costs are, by definition, irrecoverable. They should be irrelevant to any forward-looking decision. And yet they aren’t.
Why It Happens
Loss Aversion
Stopping feels like “wasting” what’s already been invested — and loss-aversion makes that waste feel like an active loss, not a neutral acknowledgment of past facts. System 1 experiences abandonment as losing the invested resources, even though those resources are already gone regardless of what you do next.
Commitment and Consistency
Having publicly or privately committed to something, abandoning it signals inconsistency — a social cost that feels real even in private decisions. (Munger identifies this as one of his 25 Standard Causes of Human Misjudgment.)
Narrative Identity
We define ourselves partly through our projects and investments. Quitting a project the self is attached to feels like losing part of the self.
The Rational Framing
The correct question is always: “Given only future costs and future benefits, is continuing worth it?”
Past costs are off the table. The question is purely prospective. The sunk cost should appear nowhere in this calculation.
Examples
Personal Scale
- Finishing a bad book or film because you’ve already started it
- Staying in a bad relationship because of years already invested
- Finishing a meal you don’t enjoy because you paid for it
- Continuing in a career path because of the education already completed
Organizational Scale
- Continuing to fund a failing marketing campaign (“we’ve already spent so much, let’s give it one more push”)
- Continuing a failing product launch to justify prior development spend
- The Concorde aircraft: Britain and France continued funding it long after it was clear it would never be commercially viable, because of prior commitments
Investment
- Holding a losing stock position to “get back to even” before selling
- Adding to a losing trade to “average down” without re-evaluating the thesis
Opportunity Cost Is the Real Measure
Beyond the irrationality of weighting sunk costs, continuing a failing endeavor also incurs opportunity cost — the value of what else you could be doing with the same time and resources. The source example: staying at a bad conference not only fails to generate value, but displaces time that could have been spent networking outside, working, or resting.
Connections
loss-aversion
Loss aversion is the core driver: abandonment feels like incurring a fresh loss on top of the already-spent resources. The emotional accounting is wrong but powerful.
mental-accounting
mental-accounting explains why sunk costs feel live: people maintain “accounts” for projects and resist closing accounts in the red.
high-agency
The high-agency framework’s “Attachment Trap” is sunk cost thinking applied to identity and beliefs — staying attached to a position, project, or self-image because of what you’ve already invested in it. High-agency actors update ruthlessly; low-agency actors protect past investments.
prospect-theory
prospect-theory explains sunk costs mathematically: the loss function is steeper for losses than gains, and framing abandonment as “losing the sunk cost” activates loss-aversion circuitry.
inversion
charlie-munger’s inversion tool is the antidote: “How would I guarantee failure here?” One reliable answer: “Keep doing things that aren’t working because you’ve already invested in them.” Inversion makes the fallacy visible.
Antidotes
- Zero-base the decision: “If I were starting fresh today with no prior investment, would I start this?”
- Opportunity cost accounting: explicitly calculate what else the remaining resources could accomplish
- Separate past from future: write down sunk costs, then physically set them aside and evaluate only forward-looking considerations
- “What would I advise a friend?”: distance reduces loss aversion