--- title: "Sunk Cost Fallacy" description: "The sunk cost fallacy is continuing an course of action because of what has already been invested, rather than because of what it will produce from here. Past spending cannot be recovered by any future choice, so a rational decision compares remaining costs against remaining benefits alone." last_updated: "2026-08-02" --- # Sunk Cost Fallacy The sunk cost fallacy is continuing an course of action because of what has already been invested, rather than because of what it will produce from here. Past spending cannot be recovered by any future choice, so a rational decision compares remaining costs against remaining benefits alone. Continuing a course of action because of resources already invested, rather than judging it by its future costs and benefits. *Example: You're an hour into a bad movie and stay to the end 'because you already paid for the ticket.'* ## What it looks like **In a project review** > We've spent eighteen months and £2M on this platform. We can't abandon it now. *Why it fails:* The £2M is gone whichever way the decision goes. The only live question is whether finishing costs less than the value finishing delivers, and the past spending does not affect either figure. **At the cinema** > This film is terrible, but I paid for the ticket so I'm staying. *Why it fails:* The ticket price is unrecoverable either way. Staying converts one loss into two: the money, plus ninety minutes that could have been spent better. **In a personal decision** > I've been studying this subject for three years — it's too late to switch. *Why it fails:* Three years are spent regardless. The comparison that matters is the value of remaining years in this field against the value of remaining years elsewhere, minus the switching cost. ## Practice ### Question 1 A company has spent £800,000 on a product. Finishing it requires £300,000 more and analysts expect £400,000 in lifetime revenue. What does the sunk cost principle say? A. Abandon it — total spend of £1.1M against £400,000 revenue is a clear loss B. Finish it — £300,000 to obtain £400,000 is worth doing, because the £800,000 is unrecoverable either way C. Finish it, because abandoning would waste the £800,000 already spent D. Abandon it, because the original estimate was clearly wrong **Answer: B.** This question deliberately makes the sunk cost logic point toward continuing, which is where people trip. Only the £300,000 remaining cost and £400,000 expected revenue are decision-relevant, so finishing nets £100,000 more than stopping. Option A commits the fallacy in reverse by including the sunk £800,000. Option C reaches the right answer through the fallacious reasoning. ### Question 2 Why is the sunk cost fallacy sometimes called the Concorde fallacy? A. Because it was first identified by a French economist B. Because the Anglo-French Concorde programme continued despite both governments recognizing it would not be commercially viable, with prior investment cited as a reason to persist C. Because it applies mainly to aviation projects D. Because Concorde was ultimately profitable **Answer: B.** Concorde became the standard illustration because the commitment was public, prolonged, and defended explicitly in terms of what had already been committed. It is worth noting that political and prestige considerations were also genuinely in play, so the case is not purely a demonstration of irrationality — but it named the pattern. ### Question 3 Which of these is NOT an instance of the sunk cost fallacy? A. "I have to finish this book — I am already 300 pages in." B. "We should keep the vendor because we spent so long on the contract." C. "I will finish this course because the remaining three modules cover exactly the skills I need." D. "I have to keep the gym membership; I already paid for the year." **Answer: C.** Option C looks forward: the reason given is what the remaining modules will deliver, not what has been spent. The others each justify continuing by reference to past investment. Note that continuing is sometimes the right call — the fallacy is in the reasoning, not the decision, and forward-looking reasoning can support the same action. ### Question 4 What psychological mechanism makes sunk cost so persistent? A. People are bad at arithmetic B. Abandoning an investment requires acknowledging a loss, and loss aversion makes that acknowledgement more painful than continuing to spend C. People genuinely believe money can be recovered D. It only affects people who did not study economics **Answer: B.** Loss aversion is doing the work: stopping crystallizes the loss and makes it real, while continuing preserves the possibility that it was not wasted. This is why the fallacy is emotionally sticky rather than merely a computational error, and why simply explaining the arithmetic often fails to change the decision. ### Question 5 What is the most useful practical test for a sunk cost situation? A. Ask how much has been spent so far B. Ask whether you would start this project today, at this cost, knowing what you now know C. Ask whether other people would continue D. Ask whether the original plan was reasonable **Answer: B.** The fresh-start test strips out the history by construction — it asks about the decision facing you now, which is the only decision available. If you would not begin today with the remaining cost and current expectations, continuing is being driven by what is already spent. ## Frequently asked ### Is it always irrational to consider what you have invested? For the decision itself, yes — past spending cannot be recovered. But related forward-looking factors are legitimate: reputational cost of abandonment, contractual penalties, and knowledge gained that lowers remaining cost. These are future consequences, which is what makes them relevant, unlike the spending itself. ### How is sunk cost different from loss aversion? Loss aversion is the underlying tendency to weigh losses more heavily than equivalent gains. The sunk cost fallacy is one behaviour it produces: continuing to invest so as to avoid crystallizing a loss. Loss aversion is the mechanism; sunk cost is a symptom. ### Does the sunk cost fallacy apply to relationships and time? Yes — any irrecoverable investment can trigger it, and time and emotional investment often bind more strongly than money because they feel less replaceable. The same forward-looking test applies: would you choose this now, given where things actually stand? ## Related - [Framing Effect](https://www.criticalthinkingexercise.org/pages/glossary/framing-effect.md) - [Motivated Reasoning](https://www.criticalthinkingexercise.org/pages/glossary/motivated-reasoning.md) - [Hindsight Bias](https://www.criticalthinkingexercise.org/pages/glossary/hindsight-bias.md) - [Biases in Decision Making](https://www.criticalthinkingexercise.org/pages/exercises/biases-in-decisions.md) - [Cost-Benefit Analysis](https://www.criticalthinkingexercise.org/pages/exercises/cost-benefit-analysis.md) - [Lesson: traps-and-pitfalls/cognitive-biases-decision-social](https://www.criticalthinkingexercise.org/pages/learn/traps-and-pitfalls/cognitive-biases-decision-social.md) ## Sitemap See the full [sitemap](https://www.criticalthinkingexercise.org/sitemap.md) for all pages.