Behavioral economists have long recognized that humans have an extremely difficult time walking away from anything in which they’ve invested lots of time, money, or effort they can’t get back. But most research on the sunk-cost fallacy has focused on people’s attachments to their own investments.
A compelling new study, however, suggests that our aversion to losses and the dread of abandoning an expensive undertaking is so intense that we’re just as likely to keep plowing ahead with dumb decisions even when someone else incurred the costs.
In a series of experiments, Christopher Y. Olivola, an assistant marketing professor at Carnegie Mellon University’s Tepper School of Business, found subjects unlikely to walk away (or to recommend that others walk away) from a number of objectively unprofitable scenarios that someone else paid for.
Subjects said they were far more likely to keep watching a boring movie on a hotel TV if their traveling partner had already paid for it, to tell an unhappy cello student to keep playing an instrument she loathed if her husband had already splashed out on lessons, or to keep playing tennis (and aggravating a painful elbow injury) if a family member had already put down a pricey, nonrefundable club membership.
The effect holds even when we don’t have a pre-existing relationship with the person who went to the trouble. Virtually everyone surveyed said they’d keep eating an overly decadent cake they didn’t really have room for, even if all they knew about the person serving the cake was that they’d obtained it at great cost from a bakery far away.
Olivola stumbled onto this line of inquiry during his doctoral dissertation research, which explored the quirky fact that people donate far more money for charitable causes when the person soliciting them is also taking on some kind of arduous challenge, like running a marathon or dunking a bucket of ice water on their head.
“It’s puzzling why people subject themselves to pain and effort to raise money for charity when in theory, you don’t have to,” Olivola said. “Running marathons by itself doesn’t cure cancer or solve world hunger.”
It does not. But consciously or not, we value the effort—apparently to a degree that can warp our judgment. Olivola sees implications for this in scenarios far more consequential than those involving cello lessons or overpriced cake. The sunk-cost fallacy can keep a company investing in a dead-end idea that a previous leader committed great resources to. It can fuel a conflict for generations: If your ancestors spilled blood for this cause, how can you abandon it now?
The findings bring to mind Nobel-winning economist Richard Thaler’s famous line: “In order to do good economics, you have to keep in mind that people are human.”