What is a sunk cost, and how do I stop it driving my decisions?
Money and time already spent cannot be recovered by spending more. Why that is so hard to act on, and the one question that reliably strips it out of a decision.
- Difficulty
- beginner
- Time
- 10 min
- Read
- 3 min
Short answer
A sunk cost is anything you have already spent that you cannot get back whatever you do next. It should have no weight in the decision, because it is identical on every path. The test: if a stranger handed you this situation today, with no history, what would you do? That answer is the right one, and the difference between it and your instinct is the sunk cost talking.
Sunk cost is the reason people keep repairing a car that has already cost more than it is worth, keep going to a course they hate, and keep pouring money into a project that stopped making sense two stages ago. It is not stupidity. Abandoning something feels like admitting the earlier spending was wasted, so continuing preserves the story that it was not — at the price of wasting more.
Step by step
- Name the money and time already gone.Write it down explicitly: the repairs so far, the deposit, the hours. Naming it turns it from an ambient pressure into a specific number you can then set aside.
- Ask what is genuinely recoverable.Some of it may not be sunk at all — a deposit that is refundable, parts that can be sold, a course that can be transferred. Recoverable money belongs in the decision; unrecoverable money does not.
- Apply the stranger test.Imagine you have just been given this situation with no history: this car, this project, this half-finished room, as it stands today. What would you do from here? That is the decision, because the past is identical whichever way you go.
- Compare only forward costs against forward benefits.From today: what will it cost to continue, what will it cost to stop, and what do you get in each case? The eight hundred pounds you already spent appears in neither column.
- Separate the money from the lesson.Stopping does not mean the earlier decision was foolish — it usually means circumstances changed or information arrived. Write the lesson down if there is one, then stop paying tuition on it.
- Beware the near-completion trap."It is almost finished" is a genuine argument only if finishing is genuinely cheap. Half-done projects routinely have most of the cost remaining, so price the remaining work rather than trusting the feeling of being nearly there.
Tips
- Sunk cost applies to time and effort as strongly as to money, and time is the one you cannot rebuild.
- Watch for it in reverse too: refusing to spend on a repair because you already spent on the thing, when the repair is genuinely the cheapest way forward.
- If you find it hard to be objective, ask someone with no stake what they would do. They cannot see the sunk cost because they did not pay it.
Common mistakes
- Continuing to avoid admitting a mistake — The mistake is already made and already paid for. Continuing does not undo it; it adds to it, and it makes the eventual stop more expensive.
- Counting past spending as an investment that must be protected — Money spent on a depreciating repair is not a stake in an asset. Nothing you do next recovers it, so it cannot be protected.
Questions people ask
So past spending is always irrelevant?
The money is. The information is not — a car that has needed three expensive repairs is telling you something about its future reliability, and that absolutely belongs in the decision. Use the history as evidence, not as an obligation.
What about a deposit I lose if I pull out?
If it is genuinely forfeited either way, it is sunk. If pulling out costs a penalty you would not otherwise pay, that penalty is a real forward cost and belongs in the comparison.