Two fans have tickets to a basketball game. A blizzard hits. One paid $200 for his ticket; the other got his free from a friend. Which fan is more likely to drive through the blizzard? Everyone says the paying fan — and everyone can feel why. But the logic is identical: both face the same drive, the same weather, the same game. The $200 is gone regardless.
The Framework
The sunk-cost fallacy is the tendency to continue investing in a losing proposition because of what has already been invested, rather than evaluating the decision based solely on future costs and benefits. The $200 ticket buyer drives through the blizzard not because the game became more valuable — it didn't — but because staying home "closes the mental account as a loss" (money spent AND game missed). The free-ticket holder who stays home only loses the game. The paying fan who stays home loses the game AND "wastes" $200.
The mechanism is mental accounting — Thaler's concept from Chapter 32. Each investment opens a psychological account. The account wants to close as a gain (money well spent) rather than a loss (money wasted). Continuing the failing investment keeps the account open — maintaining hope. Quitting closes it as a loss — which is psychologically excruciating because of loss aversion's ~2× weight. The rational response — "the money is spent regardless; what should I do from here?" — requires System 2 to override System 1's emotional accounting, which it rarely does.
Where It Comes From
Kahneman presents the sunk-cost fallacy in Chapter 32 of Thinking, Fast and Slow as a consequence of mental accounting and loss aversion. The corporate version is escalation of commitment: a company that has invested $50M in a failing project continues investing rather than accepting the loss. Kahneman notes that boards of directors often replace CEOs specifically because the new leader carries no sunk-cost mental accounts for the predecessor's decisions and can therefore cut losses that the previous leader couldn't. Research in economics and business programs has shown that graduate students who are taught about sunk costs are measurably more willing to walk away from failing projects — evidence that education works.
> "The sunk-cost fallacy keeps people for too long in poor jobs, unhappy marriages, and unpromising research projects." — Thinking, Fast and Slow, Ch 32
Cross-Library Connections
Voss's willingness to walk away in Never Split the Difference is the negotiation application of sunk-cost awareness. The time and emotional energy you've invested in a negotiation is sunk. If the deal isn't good on its current terms, walking away is the correct move — but the sunk cost of all those hours of discussion makes walking away feel like "wasting" the investment.
Fisher's BATNA analysis in Getting to Yes provides the sunk-cost corrective: evaluate the current deal against your Best Alternative, not against how much effort you've invested in reaching this point. The time spent negotiating is irrelevant to whether the current offer beats your BATNA.
Wickman's quarterly Rock review in The EOS Life creates a structural checkpoint for identifying sunk-cost traps: every 90 days, the team reassesses priorities without regard for prior investment. Rocks that aren't working get replaced, not continued "because we've already started."
The Implementation Playbook
The Sunk-Cost Test: Before continuing any commitment (project, job, relationship, investment), ask: "Would I start this today, knowing what I know now?" If the answer is no, the sunk cost is trapping you. The money, time, and energy already spent are gone regardless — the only question is whether the future returns justify the future costs.
Project Management: Build mandatory "kill or continue" reviews into every project at predetermined milestones. The review should evaluate the project based on current evidence and future prospects — never based on past investment. The reviewer should ideally be someone who wasn't involved in the original investment decision and therefore carries no sunk-cost attachment.
Investment Portfolio: Sell the losers, hold the winners. The disposition effect (selling winners to close accounts as gains while holding losers to avoid closing accounts as losses) costs investors approximately 3.4% per year in after-tax returns. The winning stock's gains are irrelevant to whether it should be sold; the losing stock's losses are irrelevant to whether it should be held. Only the future prospects matter.
Career Decisions: "I've been in this career for 15 years — I can't start over now" is sunk-cost reasoning in its purest form. The 15 years are gone. The question is: "Given my current skills, knowledge, and circumstances, what's the best use of the next 15 years?" The answer might be the same career. But if it's not, the sunk cost shouldn't be the reason you stay.
Organizational Leadership: When a project is failing, the leader who authorized it is the worst person to evaluate whether it should continue — they carry the heaviest sunk-cost burden. This is why boards replace failing CEOs: the new leader has no emotional investment in the old strategy and can evaluate the situation with fresh eyes.
Key Takeaway
The sunk-cost fallacy is loss aversion applied to past investments, and it keeps people trapped in losing positions across every domain of life. The rational prescription is simple: ignore sunk costs. The emotional reality is brutal: ignoring sunk costs means accepting that you wasted money, time, or years of your life, which activates the steepest part of the loss aversion curve. The bridge between prescription and reality is structural: build mandatory review points, involve people without emotional investment, and ask the "Would I start this today?" question before every continuation decision.
Continue Exploring
[[Mental Accounting]] — The narrow-framing system that creates the psychological accounts sunk costs trap you in
[[Endowment Effect]] — The related bias: ownership inflates value, making it harder to let go
[[Loss Aversion Ratio]] — The ~2× asymmetry that makes closing an account as a loss feel twice as bad
📚 From Thinking, Fast and Slow by Daniel Kahneman — Get the book