The Sunk Cost Fallacy: Why You Stay in Bad Situations Far Too Long

The sunk cost fallacy is the irrational urge to keep investing in something (a relationship, a career, a project) because of what you've already put in, even when the future returns are clearly negative. Arkes and Blumer's 1985 research showed that prior investment predictably distorts future decisions in every domain they tested, and that the effect gets stronger in high-stakes situations where the investments are largest.

The only rational question is "What is the best use of my resources from here on?" How much you've already put in doesn't enter into it. The sunk cost fallacy is the psychological mechanism that makes asking that question feel like betrayal, and it's behind the Concorde, relationships kept going for years after they ended, and careers continued long after the evidence said stop.

The economics of money already spent

Human brains can't accept one basic economic principle emotionally: a spent resource is gone for good, whatever you do next.

The $1.1 billion invested in the Concorde didn't get un-spent when the next billion was committed. The first billion's fate was already sealed. The only decision ever on the table was: "Given where we are and what we have now, what is the best path forward from here?"

The earlier investment has nothing to do with that question. It can't be recovered, so it should have zero influence on a future decision. Economists call these sunk costs, and a rational actor ignores them entirely when weighing future choices.

Humans aren't rational actors. We treat sunk costs as evidence of commitment, as investments that "deserve a return," and we treat walking away as the act that creates the loss, when the loss actually happened the moment the money was spent.

That inversion, believing that quitting causes the loss instead of acknowledging one that already exists, causes a huge amount of unnecessary suffering. It runs on the same cognitive-consistency wiring behind several other psychological tricks that actually work: once you've committed to a course of action, your mind fights to keep your future behavior consistent with it, even when the honest math says stop.

Three places sunk cost does the most damage

The Concorde is dramatic, but the sunk cost fallacy does far more damage in quieter, more personal areas.

Relationships

Research by economist Justin Wolfers found that one of the strongest predictors of people staying in unhappy long-term relationships is the number of years invested, not love. The longer a relationship has lasted, the harder it is to leave, whatever the current level of satisfaction.

The time already spent together feels like it should "count for something," and leaving feels like wasting it. Those years aren't on the table, though. They've already happened. The only question is what you want the next five years to look like.

Careers

Someone who has spent eight years building expertise in a field they've come to hate faces a particular version of this trap. The years of study, the professional relationships, and the certifications make a pivot feel catastrophically wasteful.

So they stay, and another five years pass. The original eight years bought five more of being stuck. The real cost of the sunk cost fallacy is those extra years of inaction, on top of the unhappy ones already spent.

This ties directly to knowing when to quit. Strategic quitting can feel like failure, but it's the economically rational choice to redirect your limited future resources toward something with a better expected return.

Projects and businesses

Startup culture turns the mythology of persistence against founders: "Never quit." "The ones who fail are the ones who give up." The survivorship bias is extreme. You only hear from founders who were right to persist, never from the much larger group who should have pivoted three years earlier. The mythology spreads partly through social proof; hear "never quit" from enough people and it starts to feel true whatever the evidence says. Social proof is one of a dozen well-documented psychological tricks that actually work, and recognizing it takes away some of its power.

The signal to look for is the answer to one question, and how much you've already invested plays no part in it: "If I had no prior investment in this, would I choose to invest in it now, knowing what I know?"

The grit paradox

Angela Duckworth's research on grit, the capacity to keep going through difficulty toward long-term goals, matters. Persistence really does predict achievement, and most worthwhile things involve pushing through a stretch of discomfort before results show up.

Popular culture has badly misapplied grit as a justification for staying in losing positions. Grit is for pursuing the right goal through temporary difficulty. It doesn't justify continued investment in a project whose economics are fundamentally broken.

One question separates productive persistence from the sunk cost fallacy: "Has new information emerged that changes whether this is viable, or am I continuing because of what I've already spent?"

If new evidence suggests the path forward is viable, you have a reason to persist. If you're staying because stopping feels like the loss, you've found the trap.

Can you avoid the trap in a realistic scenario? The Sunk Cost Casino drops you into a high-pressure setting where the house is designed to exploit your attachment to previous bets. Every scenario builds to the moment you have to choose: double down or walk away. Most people are horrified by how easily they fall in.

The two-step exit protocol

Getting out of a sunk cost trap means doing two things at once, which is why it's so hard.

Step 1: Get the emotion outside your head. Write down the total you've invested and look at it. Then write: "This is gone. It doesn't factor into what I decide next." Writing it down makes the intellectual separation more concrete.

Step 2: Zero-base the decision. Ask a specific, structured question: "If I had no history with this at all (no money spent, no time invested, no emotional attachment), and I could choose to start this specific thing or anything else, what would I choose?"

The zero-base question takes the sunk cost out of the equation, and the answer is often startlingly clear.

Conclusion: the loss already happened

The most important reframe for escaping sunk cost traps is that the loss happened when the investment was made on wrong premises, long before you thought about quitting. Looking at the sunk cost fallacy through relationship and career examples makes the point: quitting stops you from compounding that loss.

The Concorde cost billions because the people responsible couldn't feel what economists knew intellectually. The money spent arguing for continuation was being spent on decisions made in the present, not the past, and every dollar used to justify the next phase was taxed by a psychological fiction.

The aircraft flew beautifully, and it never made a cent.

When the economics are broken, walk away, however beautiful the aircraft.

Frequently Asked Questions

What is the sunk cost fallacy in relationships? The sunk cost fallacy in relationships occurs when you stay with an incompatible or toxic partner simply because you have already invested years of time, emotion, and energy into them, making quitting feel like a devastating loss.

How does sunk cost apply to your career? In a career context, the sunk cost fallacy traps professionals in industries or jobs they hate because they have already spent thousands of hours and dollars on degrees, certifications, and networking.

How do you overcome the sunk cost fallacy? You overcome it using a zero-base decision-making framework. Ask yourself: "If I had zero prior investment (no time, money, or history) in this current situation, would I choose to enter it today?" If not, it is time to walk away.

Sources

1. Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140. https://doi.org/10.1016/0749-5978(85)90049-4. Classic experiments demonstrating the sunk cost fallacy across domains.
2. Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–292. https://doi.org/10.2307/1914185. Loss aversion as the psychological driver of sunk cost behaviour.
3. Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Performance, 16(1), 27–44. https://doi.org/10.1016/0030-5073(76)90005-2. Escalation of commitment; self-justification in organisations.
4. Thaler, R. H. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior & Organization, 1(1), 39–60. https://doi.org/10.1016/0167-2681(80)90051-7. Mental accounting framework explaining why "paid for" costs feel real.
5. Knox, R. E., & Inkster, J. A. (1968). Postdecision dissonance at post time. Journal of Personality and Social Psychology, 8(4), 319–323. https://doi.org/10.1037/h0025864. Commitment increases confidence in past decisions, creating sunk cost resistance.