Sunk Cost Fallacy
mental-model proven
Categories: economics-and-financedecision-makingpsychology
Transfers
The tendency to continue an endeavor because of previously invested resources (time, money, effort) rather than on the basis of future value. Named by behavioral economists, the sunk cost fallacy is one of the most robust and replicable cognitive biases, observed in humans, organizations, and even in some animal species.
Key structural parallels:
- Irrecoverable investment — the defining feature: costs already incurred cannot be recovered regardless of what you do next. The money spent on a bad movie ticket is gone whether you stay or leave. The rational move is to evaluate only the remaining experience. But the felt weight of the spent money keeps you in the seat.
- Escalation of commitment — sunk costs compound. Having spent $1 million on a failing project, the organization approves another $500,000 “to protect the original investment.” Each new expenditure becomes the next sunk cost that justifies the next expenditure. The Concorde development is the canonical case: both governments knew the project was economically nonviable but continued because of the billions already spent.
- Loss aversion amplifier — sunk cost reasoning is powered by loss aversion (Kahneman & Tversky). Abandoning a project means realizing a loss — making it concrete rather than theoretical. Continuing preserves the hope that the investment will eventually pay off, converting a certain loss into an uncertain one. The fallacy exploits the same asymmetry as gambling: the pain of a definite loss exceeds the rational expectation of continued play.
- Completion bias — humans have a deep preference for finishing what they start. The Zeigarnik effect (unfinished tasks persist in memory) creates psychological pressure to complete. Sunk costs provide a rational-sounding justification for what is actually an emotional need for closure.
Limits
- Not all continuation is fallacious — the model is a diagnostic, not a universal prescription to quit. A half-built bridge, a nearly complete degree, a relationship that has survived real difficulties — these involve sunk costs, but the prior investment may genuinely increase the expected value of completing. The challenge is that the fallacy and rational perseverance feel identical from the inside. The model names the bias but provides no reliable method for distinguishing it from sound judgment.
- Reputation and commitment have real value — a leader known for abandoning projects at the first sign of sunk cost reasoning may be economically rational but will find it difficult to attract collaborators. Commitment itself is a resource. The sunk cost model treats each decision as independent, but in repeated games, your willingness to see things through affects others’ willingness to invest with you.
- The model assumes clear alternatives — “just walk away” presumes that the alternative use of resources is obvious. In practice, the resources freed by abandoning a sunk cost often have no clear better use, making the “rational” comparison impossible. The model works best in textbook examples where the alternative is cash in hand.
- Cultural variation — the model is grounded in Western rational-actor economics. In cultures where honor, face, and collective commitment carry weight, abandoning an investment may impose social costs that exceed the economic savings. The model frames these as irrational, but they are rational within a different utility function.
- Organizational sunk costs involve people — shutting down a failing project means laying off the team, admitting a public mistake, and destroying the careers of those who championed it. These are not sunk costs; they are future costs of abandonment. The model’s clean distinction between past and future costs is muddier in organizations than in individual decisions.
Expressions
- “Throwing good money after bad” — the folk version of the sunk cost fallacy, predating the formal concept
- “We’ve come too far to stop now” — the canonical sunk cost justification
- “We need to protect our investment” — organizational sunk cost reasoning dressed in fiduciary language
- “The Concorde fallacy” — the alternative name, after the supersonic jet program both Britain and France knew was uneconomic but continued funding
- “In for a penny, in for a pound” — folk wisdom that normalizes escalation of commitment
- “Don’t let sunk costs drive your decisions” — the prescriptive version, common in business education
- “Cut your losses” — the rational counter-prescription
- “I’ve already watched an hour of this movie” — the everyday domestic version
Origin Story
The concept has roots in classical economics (the distinction between fixed and variable costs), but the modern formulation emerged from behavioral economics. Richard Thaler’s work in the 1980s on mental accounting showed that people track sunk costs psychologically even when they know the costs are irrecoverable. Kahneman and Tversky’s prospect theory (1979) provided the theoretical mechanism: loss aversion makes abandoning a sunk cost feel like a new loss rather than an acceptance of an existing one. Hal Arkes and Catherine Blumer’s 1985 paper “The Psychology of Sunk Cost” provided the first systematic experimental demonstrations. The term “Concorde fallacy” was coined by evolutionary biologist Richard Dawkins in The Selfish Gene (1976) to describe the same phenomenon in animal behavior, though this attribution is debated.
References
- Arkes, H. & Blumer, C. “The Psychology of Sunk Cost” (1985) — foundational experimental paper
- Kahneman, D. & Tversky, A. “Prospect Theory: An Analysis of Decision under Risk” (1979)
- Thaler, R. “Mental Accounting and Consumer Choice” (1985)
- Dawkins, R. The Selfish Gene (1976) — Concorde fallacy in evolutionary context
Related Entries
Structural Neighbors
Entries from different domains that share structural shape. Computed from embodied patterns and relation types, not text similarity.
- The Cure Is Worse Than the Disease (medicine/metaphor)
- Analysis Paralysis (medicine/metaphor)
- Dunning-Kruger Effect (psychology/mental-model)
- Good Luck Reinforces Bad Habits (fire-safety/mental-model)
- Tantalus (mythology/metaphor)
- Zeno's Paradox (mathematical-reasoning/mental-model)
- Young Doctors Kill, Old Doctors Let Die (medicine/metaphor)
- Finger Trap (puzzles-and-games/metaphor)
Structural Tags
Patterns: pathforcescale
Relations: causeprevent
Structure: cycle Level: generic
Contributors: agent:metaphorex-miner