Sunk Cost Explainer
A sunk cost is a past expenditure of money, time, or effort that cannot be recovered regardless of the next choice. Because every available option shares that history, the sunk cost does not change the comparison of future alternatives. The rational question is whether the expected benefits from continuing exceed the avoidable future costs and opportunity costs from this point forward. This does not mean past spending was unimportant when it was approved; it means irreversibility changes its relevance to the next decision.
Imagine a company has spent $2 million developing a product and now expects another $800,000 of launch costs. Updated research indicates the launch would generate only $500,000 of incremental cash benefit. The $2 million is sunk. Continuing merely to “recover” it would add an expected $300,000 loss; stopping prevents that additional loss even though the original investment remains disappointing. If parts of the original project can be sold, refunded, or repurposed, those recoverable amounts are not fully sunk and should be included.
The sunk-cost fallacy occurs when prior commitment increases willingness to continue despite weak forward-looking economics. Reputation, pride, political pressure, and fear of admitting error can reinforce it. Businesses may keep an unprofitable product, facility, campaign, or software project because of money already spent. The opposite error is to abandon too quickly without considering switching costs, completion value, learning, contractual penalties, or strategic options that still affect future outcomes.
Good governance separates accountability from continuation. A company can investigate why the original forecast failed while independently choosing the best action now. Stage-gate reviews, predetermined stop criteria, outside review, and forecasts that exclude irrecoverable amounts can reduce escalation of commitment. Decision-makers should identify which costs are avoidable, which proceeds remain recoverable, what the best alternative would earn, and how uncertainty changes the range of outcomes. Past losses deserve learning, not automatic repetition.
The resource has already been committed and cannot be recovered under any current option. Its amount is therefore the same across the forward-looking choices.
Refunds, resale value, cancellation penalties, and reusable assets are not sunk to the extent they still change with the decision.
Compare expected future benefits with avoidable future costs and the opportunity cost of alternatives. Do not add irrecoverable spending to one side.
A project can be worth completing after a bad start, but only when its remaining economics support that choice.
Use stage gates, stop criteria, independent review, and updated forecasts to limit the influence of pride and prior commitment.
Review the original decision separately for accountability and learning; that review should not predetermine whether to continue.
Explore more "Explainers"
Discover additional explainers across politics, science, business, technology, and other fields. Each explainer breaks down a complex idea into clear, everyday language—helping you better understand how major concepts, systems, and debates shape the world around us.