sunk cost
You buy a 50-dollar movie ticket, and twenty minutes in the film is awful. Should you stay just because you paid? The 50 dollars is gone either way — staying will not get it back, it will only cost you ninety more boring minutes. The rational move is to leave and do something you enjoy. That 50 dollars is a sunk cost: money (or time, or effort) already spent and impossible to recover, no matter what you do next.
A sunk cost is a cost that has already been incurred and cannot be undone or refunded. The key economic principle is blunt: sunk costs should be ignored in decisions about the future, because a good decision compares the future benefits and future costs of your options, and the sunk amount is identical across all of them. If a company has spent 10 million on a project and finishing it will cost 5 million more but is now worth only 4 million, it should stop — the 10 million is gone, and spending 5 to get 4 only loses more. Only the forward-looking 5-versus-4 matters.
Ignoring sunk costs is harder than it sounds, because humans hate to feel they wasted something. The sunk cost fallacy — 'throwing good money after bad', or staying in a failing venture, relationship, or war because of all you have already poured in — is one of the most common and expensive thinking errors there is. The discipline is to ask only one question: from here, what is the best use of my remaining resources? What is already spent is, economically, irrelevant to that answer.
An airline has spent 1 billion developing a new plane that, it now realises, won't sell. Continuing would cost 500 million more for a product worth only 300 million. The 1 billion is sunk; the smart, painful choice is to stop.
Money already gone should not drag you into spending more.
Sunk costs should be ignored when deciding what to do next. Letting them sway you is the sunk cost fallacy — 'throwing good money after bad'.