47 – Thinking like an economist 14: The sunk-cost fallacy

People are very susceptible to feeling that past expenditures matter in current and future decisions. If you’ve spent a lot of money on something, wouldn’t it be a waste of that money not to use that thing? This sort of thinking can lead us into the “sunk-cost fallacy” and it can cause us to do things that are really not in our best interest.

Whenever Pauline and I go fishing together, there is a fair chance we’ll end up having a conversation about sunk costs. It goes like this.

David: ‘There are no fish here. Let’s go home.’

Pauline: ‘But give it a chance. You’re so impatient.’

David: ‘We’ve been giving it a chance for an hour and a half, and so far I’ve caught a piece of seaweed, and you’ve hooked my shirt.’

Pauline: ‘But it took us over an hour to get organised and down here. All that effort to get the fishing gear together, make a flask of coffee, get changed, drive to the shop, buy the bait, drive down here to the beach, rig up the fishing lines, and you want to give up already!’

David: ‘Those are all sunk costs. They shouldn’t have any effect on our decision to stay or go.’

Pauline: ‘Oh, sunk schmunk! Bloody economists!’

Pauline feels that the effort and expense that we’ve already invested in catching fish is quite relevant to the decision to stay or go. And she is far from being alone in feeling that sunk costs ought to matter. Would they matter to you? Consider these two scenarios.

  1. David’s enthusiastic praise convinces you to buy the latest Elvis Costello CD. On the way home from David’s house you stop at the music store, shell out good money for the CD, take it home and put it straight on, only to discover that the CD is damaged and won’t play at all. You have kept the docket. Do you go to the store and exchange the CD?
  2. David’s enthusiastic praise convinces you to buy the latest Elvis Costello CD. However, when he learns this, in his delight, David gives you a spare copy that he has just bought. You take it home and put it straight on, only to discover that the CD is damaged and won’t play at all. You notice that the docket is in the bag David gave you. Do you go to the store and exchange the CD?

The evidence is that more people would take it back to the store in scenario 1 than in scenario 2. In scenario 1, you’ve invested more effort and money in acquiring the CD. In a sense, you’ve got more of a stake in it. In scenario 2, you may feel something like ‘easy come, easy go’ (although, admittedly, this might be overcome by concerns about what David would say if he found out).

The tendency to be influenced by what you have already paid is called the ‘sunk-cost fallacy’. It is a fallacy because the additional benefit you gain by deciding to exchange the CD (that is, the benefit of acquiring a working CD) is the same in either scenario. The benefits and costs of that decision are not affected by what costs you have or have not borne in the past. That cost will remain on the negative side of your ledger regardless of whether or not you go back and exchange the CD.

The tendency for most people to fall for the sunk-cost fallacy is well established. In a classic paper called “The psychology of sunk cost”, Arkes and Blumer (1985) randomly distributed different levels of discount to people who were all buying tickets to the same theatre production. They found that people who paid more for their tickets were more likely to actually attend the performance. On average, both groups would have had approximately the same benefits from attending – the pleasure of seeing the performance – which all participants had judged was sufficient to warrant bearing the full purchase price plus the time and expense of getting to and from the performance. Once the ticket had been purchased, regardless of the price paid, the costs of attending consisted solely of the time and expense of getting to and from the performance. And yet people allowed themselves to be influenced by the sunk cost of ticket purchase.

It doesn’t just affect small decision either. Arkes and Blumer noted that that governments can fall for the fallacy in major spending decisions. Once a project has commenced and some money has been spent, even if cost estimates for the remainder of the project skyrocket, it is hard to shut the project down. They give the example of Tennessee senator James Sasser arguing for the continuation of a contentious and expensive infrastructure development, the Tennessee-Tombigbee Waterway Project.

‘Completing Tennessee-Tombigbee is not a waste of taxpayer dollars. Terminating the project at this late stage of development would, however, represent a serious waste of funds already invested.’

A classic expression of the sunk-cost fallacy! Whoever coined the old saying, “Don’t send good money after bad” understood the sunk-cost fallacy.

But it’s hard to fight. Just remember, sunk costs are sunk. By definition, you can’t get them back. Mentally, put them aside when considering where you go from here. Where you go from here is all that really matters.

David Pannell, The University of Western Australia

Further Reading

Arkes, H.R. and Blumer, C. (1985). The psychology of sunk cost, Organizational Behavior and Human Decision Processes 35: 124-140.

Noonan, K. (2004). Interview about The Delivery Man and Il Sogno, “Cool, that’s Elvis Costello”. Courier-Mail (Brisbane), 16 Oct 2004, http://www.elviscostello.info/articles/a-c/courier_mail.041016.php

Power, C. (2004). Review of The Delivery Man and Il Sogno, “Elvis grows up”. Newsweek 1 Nov 2004, http://www.elviscostello.info/articles/n/newsweek.041101.php


  • MEOW
    11 September, 2017 - 4:42 am | link

    It is not a fallacy per se. Au contraire. It is presented in the context of marginal analysis. As one sage economist said a student learns all the econ required in 101A and 101B. The next six years are spent making her/him analyze on the margin. How can one change the past?

    • 11 September, 2017 - 3:05 pm | link

      I don’t follow your comment. The sunk costs are not marginal costs. The sunk costs might be millions or billions or dollars (e.g. an investment in major infrastructure), which is clearly not marginal, but they are still sunk when considering whether it is worth spending the money to finish off the project. And the decision about whether to finish off the project is not marginal either. It’s about whether the overall benefits of finishing the project outweigh the overall costs of finishing the project.

  • MEOW
    5 January, 2018 - 1:27 am | link

    I did not say sunk costs are marginal costs. The projected intramarginal cost is the difference between the sunk costs and estimated expenditures required to finish the proposed task. Completing the dam is a marginal economic decision, as you accurately describe. Past expenses of taxpayer monies may merit an investigation as to legality and prudence?

    The assumptions of the equal enjoyment of a performance and the like opportunity cost for patrons attending a concert is questionable. Did these investigators ignore that interpersonal value judgments vary among individuals? (no two individual preference maps are identical)

  • MEOW
    4 February, 2019 - 2:23 pm | link

    A more universal example to give students a better understanding of sunk costs might be: Obsolescence is an unexpected reduction in the value of a productive resource from an unanticipated development of a competing resource (rather than by use, as with depreciation) when the competing resource is new and superior. This is a sunk cost, which no longer is relevant to decisions about the use of a resource.

    You could better express this using the wealth paradigm.

    • 19 March, 2019 - 6:31 am | link

      I don’t fully understand your example, but if you mean that investment in a technology that has become obsolete is a sunk cost and is irrelevant to future decisions, I agree.

  • Leave a Reply

    Your email address will not be published. Required fields are marked *

    Please solve this to show you\'re a real person *