erroneo.us
Search all guides⌘K

Error of the Day · Thinking

The sunk cost fallacy: the money is gone either way

Finishing the bad movie, the dull book, the doomed project: it feels like honoring your investment. It's actually spending more to avoid admitting a loss.

Published By erroneo.us Editors3 min read4 sources

WrongWe've already put two years into this project, so we can't quit now.

RightIgnoring the two years we can't get back, is the next six months worth it?

Memory trick Ask: if I were starting fresh today, would I choose this? The past cost is the same either way.

What people get wrong#

You order a full-course dinner. By dessert you're stuffed and no longer enjoying it, but you keep eating because you paid for it. That scene opens a 2019 PLOS ONE paper on the sunk cost fallacy, and it captures the error exactly: the money is spent whether you finish the dessert or not. Eating it only adds discomfort to the bill.

A sunk cost is any money, time or effort already spent that you cannot recover. The fallacy is letting it drive what you do next. In the definition the PLOS ONE authors quote from the classic 1985 study by Hal Arkes and Catherine Blumer, the sunk cost effect is the tendency "to continue an endeavor once an investment in money, effort, or time has been made."

Economic theory is blunt about the rational version: decisions should be guided only by future gains and losses, because past costs don't change the outcomes of the choice in front of you.

Where it shows up#

  • Entertainment: sitting through a bad movie, or a whole season of a show you stopped liking, because you've "come this far."
  • Money: holding a losing investment until it "gets back to even," or pouring repairs into a car because of the last repair bill.
  • Work: keeping a failing project alive because of the budget already burned. Researchers call this escalation of commitment.
  • Life: staying in a degree, job or relationship mostly because of the years already invested.

It also has a nickname, the "Concorde effect," after the supersonic airliner. A 1999 review in Psychological Bulletin used both names in a pointed title: "Are humans less rational than lower animals?"

Why it happens#

Quitting turns a cost into a visible loss, and losses sting. Research points to emotion as a driver. The PLOS ONE study found that people's negative feelings about a specific situation predicted how strongly they fell for the fallacy, and the link was stronger under cognitive load, when people leaned on gut reactions instead of deliberating. Participants who were asked to justify their decision resisted the fallacy better.

Imagination plays a part too. In a 2016 study in Psychology and Aging, people who kept investing in a failing plan tended to entertain thoughts of eventual success, or to embellish the scenario in ways that made continuing seem reasonable. Older participants were more willing to cancel the failing plan, and less likely to dwell on hoped-for success.

The fix#

"We've already spent $200,000, so we have to finish."

"Will the next $50,000 buy something worth more than $50,000?"

Only future costs and benefits differ between your options; the past spending is identical in both.

Practical moves:

  • The fresh-start test. If someone handed you this project, ticket or stock today, at no cost, would you choose to continue it?
  • Write the reason down. Justifying the decision explicitly helped people resist the fallacy in the PLOS ONE study. If your best argument mentions what you've already spent, that's a warning sign.
  • Focus on thoughts and feelings, not rescue plans. The Psychology and Aging study found that prompting people to reflect on their thoughts and feelings reduced sunk-cost bias more than either prompting them to think about how to improve the situation or giving no prompt at all.

Is sticking with it ever rational?#

Yes, and the distinction matters. Past investment can change the future math: a half-built house may now be cheap to finish, and two years of training may make you genuinely good at the job. Continuing for those reasons is forward-looking, not fallacious.

Researchers also warn against seeing the fallacy everywhere. A 2022 paper in Science Advances showed that behavior which looks like a sunk cost effect in some experiments can be produced by a perfectly rational, reward-maximizing decision maker, thanks to a statistical artifact. The test isn't whether you persisted. It's whether you persisted because of what you couldn't get back.

Sources

Every factual claim on this page links to one of these. Checked October 28, 2026. See how we source.

  1. Dijkstra & Hong (2019), PLOS ONE: The feeling of throwing good money after bad journals.plos.org
  2. Strough et al. (2016), Psychology and Aging: Reducing sunk-cost bias in a life-span sample (PMC) pmc.ncbi.nlm.nih.gov
  3. Arkes & Ayton (1999), Psychological Bulletin: The sunk cost and Concorde effects doi.org
  4. Ott et al. (2022), Science Advances: Apparent sunk cost effect in rational agents (PMC) pmc.ncbi.nlm.nih.gov
ThinkingCognitive biases: the predictable ways good minds go wrongA sourced guide to 20 common cognitive biases: anchoring, confirmation bias, sunk cost, Dunning-Kruger and more, with examples, fixes and replication caveats.ThinkingLogical fallacies: how arguments break, with examplesStraw man, ad hominem, false dilemma, slippery slope, post hoc and more: 20 logical fallacies explained with plain examples, Latin names and how to respond.MythsMoney mistakes and personal-finance mythsMinimum payments, credit-check myths, skipped 401(k) matches, fee drag, market timing, forgotten subscriptions and scams: 13 money mistakes, with sources.