Sunk Cost Fallacy
Sunk Cost Fallacy: Irrationally continuing something because of what you've already invested β not because of its future value. "I've already spent $10,000 on this, so I can't quit now" is the fallacy. The $10,000 is gone whether you continue or quit; the only relevant question is whether the future expected value justifies further investment. Past costs are irrelevant to future decisions.
What Is the Sunk Cost Fallacy?β
A sunk cost is any cost already incurred that cannot be recovered regardless of future actions. The rational principle: sunk costs should be irrelevant to future decisions. What matters is the future expected value of each option, not the resources already invested.
The fallacy: humans systematically violate this principle, allowing sunk costs to influence decisions. The mechanism is loss aversion (Kahneman and Tversky): exiting a failing project forces you to realise a loss psychologically. Continuing keeps the loss as a "paper loss" that might be reversed. The choice to continue isn't rational β it's a psychological defence against accepting loss.
This error appears at every scale: individuals finishing bad movies, failing relationships, and unpleasant meals; companies continuing failed projects and bad acquisitions; governments continuing losing wars long after the evidence for futility is overwhelming.
How It Worksβ
The rational framework:
Decision: continue or exit the project?
Relevant: future expected value of continuing vs. exiting
Irrelevant: amount already invested (it's gone either way)
The fallacy mechanism:
1. Resources invested β psychological ownership and loss aversion
2. Exit = realising a loss (painful)
3. Continue = keeping the possibility of recovery (less painful today)
4. Result: continuation despite negative expected value
Warning signs:
β "I can't quit now β I've put X months/dollars into this"
β "If I stop now, all that work was wasted"
β "We've already committed publicly β we can't back out"
β Escalation of commitment: doubling down to "recover" past losses
Three Real-World Examplesβ
Concorde Programmeβ
The British and French governments spent billions on Concorde development in the 1960sβ70s. By 1974, the programme had serious financial problems β it would never be commercially viable. Both governments knew this. But by that point, enormous sums had been invested, and neither government could politically accept acknowledging that the investment was wasted. Development continued for years past when it was economically rational. "Concorde fallacy" is now a standard term in behavioural economics for this exact pattern.
Staying in Failing Relationshipsβ
Research on relationship breakups finds that the longer a relationship has lasted and the more resources invested (time, social capital, shared property), the harder it is to leave β even after the relationship is clearly not working. People explicitly cite "but we've been together for X years" as a reason to continue rather than evaluating the relationship's current and future prospects on their merits.
Corporate Project Escalationβ
A classic study by Barry Staw (1976) asked business students to evaluate a failing investment scenario. Students consistently allocated more resources to a failing project when they had previously decided to fund it (versus when a "prior manager" had decided). The personal prior commitment triggered sunk cost reasoning. In real companies, project escalation and the difficulty of cancelling committed initiatives β even those clearly failing β is one of the most frequently cited causes of resource misallocation.
When to Recognise Itβ
π¨ Sunk Cost Fallacy is likely operating when:
- Your primary reason for continuing is the amount already invested
- You're thinking about "not wasting" resources already spent
- The expected future value of the project is clearly negative but you're reluctant to exit
- You feel more committed after additional investment fails
β Countermeasures:
- Conduct a "zero-based" evaluation: "If I hadn't invested anything yet, would I start this today?"
- Separate the sunk cost acknowledgment ("that $X is gone") from the future decision
- Pre-commit to exit criteria before projects begin
- Ask someone without prior investment history to evaluate the decision
| Pairs well with | Why |
|---|---|
| Loss Aversion | Loss aversion is the primary driver of the sunk cost fallacy |
| Commitment Bias | Both cause irrational continuation of existing courses |
| Two-Way Door | Recognising decisions as reversible reduces sunk cost reasoning |
| Pre-mortem | Pre-mortems can reveal sunk cost reasoning before it compounds |
Common Misuses and Limitationsβ
Claiming all prior-investment reasoning is a fallacy. Not all consideration of past investment is the sunk cost fallacy. If past investment provides information about future returns (experienced teams, established infrastructure, proven technology), it is relevant. The fallacy is when past investment is considered regardless of its implications for future returns.
Ignoring switching costs. Sometimes continuing has value beyond the specific project's return β relationship capital, institutional knowledge, team continuity. These are legitimate future-value considerations, not sunk cost reasoning. Distinguish between "the past investment makes me feel like I should continue" (fallacy) and "the past investment created assets that make continuing more valuable" (not necessarily a fallacy).
Related Modelsβ
| Model | Relationship |
|---|---|
| Loss Aversion | Loss aversion is the emotional driver of the sunk cost fallacy |
| Status Quo Bias | Both cause irrational resistance to changing course |
| Opportunity Cost | Sunk cost fallacy ignores opportunity costs in favour of sunk costs |
Frequently Asked Questionsβ
Is the sunk cost fallacy always irrational?
Almost always in pure resource allocation decisions. There are edge cases where commitment signalling has strategic value β if others observe your willingness to follow through on sunk investments, they may make more credible commitments to you. But this is a strategic consideration about the future, not a sunk cost argument. In most personal and business decisions, the fallacy is pure irrationality driven by loss aversion.
Why is the sunk cost fallacy so hard to avoid?
Three reinforcing factors: (1) loss aversion β exiting means realising a loss; (2) consistency bias β we want to behave consistently with our prior commitments; (3) social accountability β we've often told others about the investment and feel we'd look bad for abandoning it. The fallacy is hardest to avoid when all three apply: you've publicly committed, invested significantly, and would face judgment for exiting. Pre-committing to exit criteria before investment begins is the most reliable protection.
Does the sunk cost fallacy apply to time as well as money?
Yes, and often more powerfully. People are as unwilling to "waste" time already spent as money already spent. Research finds that people will continue reading a bad book or watching a bad movie significantly longer if they've already spent more time on it. The mechanism is identical: the time is gone whether you continue or stop; only the future time cost and value are relevant. The principle applies to time, emotional investment, and social capital, not just financial resources.
Further Readingβ
- Thaler, R. (1980). "Toward a Positive Theory of Consumer Choice." Journal of Economic Behavior & Organization β introducing sunk cost reasoning in behavioural economics
- Staw, B. (1976). "Knee-Deep in the Big Muddy: A Study of Escalating Commitment." Organizational Behavior and Human Performance
- Arkes, H. & Blumer, C. (1985). "The Psychology of Sunk Cost." Organizational Behavior and Human Decision Processes
Apply with AIβ
π Identify sunk cost reasoning in your decisions with MindMax β
This page is part of the MindMax Mental Models Knowledge Base.