Stephen T’s Blog Spot

A program is underperforming and the evidence keeps accumulating, but the case for pressing on is always the same. We have already put so much into this. We cannot stop now. It sounds like prudence and stewardship. It is precisely backwards. Whatever you have already spent is gone whether you continue or not, which makes it the one thing that should carry no weight in deciding what to do next.

This is the sunk cost fallacy, one of the most robust findings in behavioral science. A sunk cost is any money, time, effort, or reputation you have already spent and cannot recover. The rule for a sound decision is narrow and a little cold: weigh the future costs of each option against the future benefits, and nothing else. Your past spending explains how you reached this decision point, but it has no place in the comparison, because nothing you choose now can bring it back. Hal Arkes and Catherine Blumer showed the pull of this in 1985. People who had paid full price for a theater subscription attended more shows than people who got the identical subscription at a discount. The future value of each play was the same for everyone; the only difference was that skipping one felt, to those who had paid more, like wasting what they had spent.

That phrase, not wanting to waste what we spent, is the heart of it. Abandoning an investment forces us to book the loss, and losses hurt roughly twice as much as equivalent gains feel good, so we continue partly to postpone the pain of admitting the money is gone. We also do not want to look wasteful or inconsistent. And when the people deciding whether to continue are the same people who launched the effort, a further force takes over. Stopping is not just a financial write-off; it is a public admission that the original decision was wrong. Barry Staw named the result escalation of commitment: throwing good resources after bad, precisely because so much has already been thrown.

The dynamic feeds itself. Each new investment enlarges the sunk cost, which raises the pressure to keep going in order to justify it, which invites the next investment. The Concorde, the supersonic jet that two governments kept funding long after it was clear it would never pay for itself, gave the fallacy one of its names. The more they spent, the more unthinkable it became to stop, which is exactly the wrong response to spending that could never be recovered.

There is one honest complication worth stating, because it is often used as cover. The past is not always irrelevant. What you have already built, and how far the work has come, can be real information about whether the future looks promising. The discipline is to separate two very different things: the informational value of past progress, which is legitimate, from the emotional weight of past spending, which is the fallacy. Ask whether the work so far genuinely improves the odds from here. Do not ask whether stopping would waste what came before.

For those of us in this field, the trap runs through both sides of the work. In evaluation, one of the most valuable things a study can do is give decision-makers cover to stop, to move resources from what is not working to where they will do more good, which is the opportunity-cost point from an earlier post. Yet evaluations are often commissioned, half-consciously, to justify continuation, with the sunk cost supplying the emotional case. And in business development, the same fallacy shapes capture. A pursuit you have chased for a year, with real bid-and-proposal money spent, becomes hardest to walk away from at the very moment the win probability has collapsed, because leaving means writing off everything you put in. A disciplined decision, in either setting, ignores what is already spent and asks only what the future holds.

So here is my question. When you decide whether to continue a program or a pursuit, do you weigh what you have already invested, or only what it will cost and return from this point forward?