Money already spent has no place in the decision. And yet it decides almost everything

21 September 2026, 12:53

In the 1970s Britain and France already knew that Concorde would never pay back what it cost. They carried on regardless. Not out of stupidity, nor even out of national pride: for a reason every business owner recognises instantly, "we've already spent too much to stop now".

That line of reasoning has a name in the economic literature and, ironically, it is called the Concorde fallacy. It says something simple and unwelcome: money already spent should not count in today's decision, and yet it runs it.

The same sentence, spoken on a factory floor

In companies it sounds like this. "That line has been losing money for three years, but we've put two hundred thousand euros' worth of moulds into it." "That machine runs at thirty per cent, but we paid for it." "That customer earns us very little, but we've been serving them since 2009." "Nobody uses the management software, but it cost a fortune."

All these sentences have the same structure: the past is used as an argument for the future. And they are all wrong in the same way.

The moulds have already been paid for. That money does not come back whether you close the line or keep it open: it is sunk, and it no longer belongs to the decision. The only question that matters is a different one: from here on, does that line bring in more than it consumes?

Why it is so hard, even for people who can do the maths

It is not ignorance. It is how the mind works in someone who has taken risks personally.

Closing means booking a loss, and booking a loss means admitting that the decision was wrong. As long as the line stays open, the mistake remains theoretical. Psychologists call it loss aversion: losing a hundred hurts roughly twice as much as gaining a hundred pleases. And that is why a business owner who haggles with every supplier down to the last penny can keep alive, for years, a product that eats into the margin.

There is an organisational side too: that line has someone in charge of it, and that person is good. Closing it feels like a verdict on them. It almost never is, but nobody says so out loud.

The counterweight is having the numbers line by line

The bias can be fought in only one way: by taking away the darkness it lives in. As long as margin is looked at only in aggregate, the loss-making line stays invisible, funded by the good ones, and the decision becomes a matter of opinion, which is always won by whoever has the most seniority.

When, on the other hand, revenue and costs are attributed to each line, product family or customer, the conversation changes its nature. Nobody argues any more about whether that product is "doing well": you look at how much it contributes towards covering fixed costs, and you decide.

Beware of one trap, though. Sunk cost does not mean the whole past is irrelevant: the residual depreciation is not a good reason to carry on, but the stock you free up by closing that line is real cash coming back in, and that very much belongs in the decision.

The thirty-second test

There is a question that takes the bias apart better than any spreadsheet, and it is worth asking it out loud:

"If I walked into this company today, with no history and no attachments, would I open this line?"

If the answer is no, then you are not keeping it for economic reasons. You are keeping it so as not to admit something. That is a legitimate choice, of course, but it has to be called by its name and, above all, it has to be costed: keeping it open for another year has a price, and you pay that price out of the margins of the lines that work.

What to do tomorrow morning

Take the product line or the customer that irritates you most when you think about it: it is usually the first one that comes to mind, and that is no accident.

Calculate its contribution margin over the last twelve months, that is, revenue minus the costs that exist only because that work exists. Then look at that number and answer the question above.

You don't have to decide today. You just have to stop deciding with yesterday's money.