Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Question · Open

I kept my startup going two years too long because of what I had already put in. Is sunk-cost reasoning always a fallacy?

Rohan Desai Fellow
· 4 replies
Action & agency#burnout#commitment#decision-making#sunk-cost
Follow

Its coordinates: 6.24 Philosophy of Action

Everyone in my world knows the phrase "sunk cost fallacy". Money and time already spent are gone whatever you do next, so they shouldn't affect your decision. The classic study, by Arkes and Blumer in the 1980s, found that people who had paid full price for theatre season tickets went to more plays than people who had been given a discount, as if the money already paid obliged them to suffer through the rest.

I knew all this, and I still kept my company alive for two years after the numbers said stop. I told myself it was about the team, about the customers, about not quitting. Mostly, I think, it was that I couldn't bear for the five years before to have been wasted.

Now I'm not sure that last feeling was simply irrational. Some philosophers point out that past investments can matter for reasons beyond money: a promise to a cofounder, the meaning of a long effort, the kind of person who finishes things. A marriage in a hard year isn't a sunk cost problem. Or is it?

So: when is honouring what you've already put in a fallacy, and when is it a form of integrity? And how do you tell which one you're doing while you're inside it?

LikeAnswer
SaveFollow4 answers

Join the conversation: members reply here, and joining is free. Reading is always free.

Join free to reply Sign in

4 answers

  1. Dana Whitfield

    Fellow

    In organizing we have a version of this: campaigns that should have ended a year ago but carry on because "we've come too far". The test I use is whether the reason looks backwards or forwards. "We've spent so much" looks backwards, and is the fallacy. "People trusted us to see it through, and abandoning them now would damage that trust for the next fight" looks forwards, and is a real reason.

    Most of the integrity reasons survive the test. The pride reasons don't.

  2. Rowena Abellana

    Fellow

    Gabriel Marcel wrote about fidelity as "creative": keeping faith with a commitment is not just repeating the original decision, it is continually re-making it. On that view your marriage example isn't a sunk cost case at all. A good marriage in a hard year is being chosen again, not paid for.

    A company can be like that too. But if you were no longer choosing it, only refusing to lose it, that's the difference. It sounds as though you know which one it was.

  3. Rohan Desai

    Fellow

    Rowena, "refusing to lose it" is uncomfortably exact. Dana's backwards-forwards test I've written on a card and put on my desk. The hard part, which neither of you can fix for me, is that inside it, the backward reasons dress up very convincingly as forward ones.

  4. Bernard Tremblay

    Reader

    Rohan, I kept a dairy farm going for at least five years longer than the numbers said I should, so I recognise everything you describe. In Quebec you need quota to sell milk, and my quota was worth more than my land. Selling would have been rational for years. I didn't, partly because my father had paid for it over half his life.

    I think Dana's backwards-forwards test is right but too clean for a family business. Some of my backward reasons were forward ones in disguise: my son might want to farm, so the farm had to survive until he decided. That was a real reason. "My father paid for this" was not.

    What finally made the difference was saying the reasons out loud to my wife at the kitchen table. The sunk ones sounded foolish in daylight. The forward ones didn't. Maybe that is Rowena's creative fidelity: you have to choose it again, in words.