Questions: Sunk Cost Fallacy and Escalation of Commitment

Question 1

A company has spent $10 million developing a product that market research now shows will lose money. The project director argues, 'We have to keep going — we've already spent $10 million.' What is wrong with this reasoning?

• A. Nothing — past investments create moral obligations to follow through
• B. The $10 million is a sunk cost and cannot be recovered either way, so it is irrelevant to whether future investment will be profitable
• C. The project director should gather more market data before deciding
• D. Continuing is justified if there is any chance of future profit, however small

Answer: (B)

The $10 million is already gone regardless of what the company decides next. Rational decision-making looks only at future costs and future benefits. The sunk cost fallacy treats unrecoverable past expenses as if they can be 'redeemed' by continuing, but no future action changes what has already been spent. The tempting option A (moral obligation) is precisely the self-justification mechanism that drives escalation — not a valid rational argument.

Question 2

Research on escalation of commitment finds that decision-makers who personally chose the original investment are more likely to keep funding a failing project than decision-makers brought in later. What psychological mechanism best explains this pattern?

• A. Original decision-makers have more expertise and correctly identify long-term value others miss
• B. Original decision-makers have access to inside information that justifies continued investment
• C. Personal responsibility for the original commitment amplifies self-justification — abandoning the project means admitting one's own earlier mistake
• D. Newcomers are less loyal to organizational goals and too quick to abandon viable projects

Answer: (C)

Staw's research demonstrated that personal responsibility for the initial decision amplifies escalation through self-justification: stopping means admitting the original choice was wrong, which threatens the decision-maker's self-image as competent and rational. Outside evaluators, not having made the initial bet, can assess future prospects more objectively. This is also why organizations benefit from rotating who evaluates ongoing projects — the evaluator's identity shapes what information gets weighted.

Question 3

The sunk cost fallacy tends to weaken as more is invested in a failing project, because people eventually recognize the loss and cut their losses.

• T. True
• F. False

Answer: False

The bias typically intensifies with deeper investment, not weakens. Each additional dollar spent raises the psychological cost of acknowledging the entire prior investment as wasted. The logic 'we've put in $500 million — we can't stop now' is more powerful than 'we've put in $50,000 — we can't stop now.' This escalating dynamic is why sunk cost reasoning can drive organizations to catastrophic levels of continued investment in failing projects. Recognizing the bias early, before large commitments accumulate, is much easier than reversing it after they have.

Question 4

From the perspective of rational decision theory, only future costs and future benefits are relevant when deciding whether to continue a project.

• T. True
• F. False

Answer: True

This is the normative core of the sunk cost concept. A sunk cost is by definition unrecoverable — no decision changes it. Including it in future deliberations amounts to double-counting: the loss is already realized. The diagnostic question 'Would we start this project today if we were beginning fresh?' operationalizes this principle by forcing decision-makers to evaluate future prospects without the distorting weight of prior investment. If the honest answer is no, continuation is driven by sunk costs, not expected value.

Question 5

A friend says, 'I have to finish this terrible book — I've already read 200 pages.' Explain why this reasoning commits the sunk cost fallacy, and what question the friend should actually be asking.

Model answer: The 200 pages are a sunk cost — the time spent reading them cannot be recovered whether or not the friend finishes the book. The fallacy treats that irretrievable past investment as a reason to incur additional costs (more hours of miserable reading). The rational question is: 'Given how bad this book is and what else I could do with my time, is finishing it worth the remaining time it will take?' If the answer is no, the rational choice is to stop — regardless of how much has already been read.

The sunk cost fallacy works by reframing continuation as 'getting your money's worth,' but this is an illusion — the past investment is gone either way. The friend's reasoning implicitly adds the sunk cost to the future-decision ledger where it does not belong. Cognitive dissonance is the mechanism: stopping means admitting the original choice (reading this book) was a mistake, which is psychologically uncomfortable. The rationalist solution is to reframe: prior investment is evidence only about the past, not a guide to the future.