The Sunk Cost Fallacy

March 25, 2011

tags: Arkes, Blumer, Facebook, Farmville, Kahneman, loss aversion, social games, sunk cost, Tversky

The Misconception: You make rational decisions based on the future value of objects, investments and experiences.

The Truth: Your decisions are tainted by the emotional investments you accumulate, and the more you invest in something the harder it becomes to abandon it.

You can learn a lot about dealing with loss from a video game called Farmville.

You have probably heard of this game. In 2010, one in five Facebook users had a Farmville account. The barrage of updates generated by the game annoyed other users so much it forced the social network to change how users sent messages. At its peak, 84 million people played it, a number greater than the population of Italy.

Farmville has shrunk since then. About 50 million people were still playing in early 2011 – still impressive considering the fantasy megagame World of Warcraft boasts about a quarter as many players.

So, it must be really, really fun. A game with this many players must promise potent, unadulterated joy, right? Actually, the lasting appeal of Farmville has little to do with fun. To understand why people commit to this game and what it can teach you about the addictive nature of investment, you must first understand how your fear of loss leads to the sunk cost fallacy.

Studies by psychologists Daniel Kahneman and Amos Tversky going back to the 1970s show you don’t equate loss and gain. Loss is more powerful. When they had subjects gamble in the lab, they noticed people tended to demand the promise of a payoff of at least double what they risked before they agreed to the terms of the game. Loss, they reasoned, was gain times two.

Source: www.kimclips.com

Outside of the lab, the pain you feel when you lose cash is twice as strong as the joy you feel when you gain an item of equal value. This is why marketing and good salesmanship is often all about convincing you what you want to buy is worth more than what you must pay for it. You see something as a good value when you predict the pain of loss will be offset by your joy of gain. If they did their job well, somewhere in your Byzantine perception you feel as though you won’t lose at all. Emotionally, you will come out ahead. Unless you are buying something just to show others how much money you can burn, you avoid cringing when you fork over your earnings.

Imagine the apocalypse is upon you. Some terrible disease was unleashed in an attempt to cure male pattern baldness. The human population has been reduced to 600 people. Everyone is likely to die without help. As one of the last survivors you meet a scientist who believes he has found a cure, but he isn’t sure. He has two versions and can’t bear to choose between them. His scientific estimates are exact, but he leaves the choice up to you.

Cure A is guaranteed to save exactly 200 people. Cure B has a 1/3 probability of saving 600, but a 2/3 probability of saving no one. The fate of hairlines and future generations is in your hands. Which do you pick?

Ok, mark your answer and let’s reimagine the scenario. Same setup, everyone is going to die without a cure, but this time if you use Cure C it is certain exactly 400 people will die. Cure D has a 1/3 probability of killing no one, but a 2/3 probability killing 600. Which one?

Source: http://hvzmovie.wordpress.com

When Tversky and Kahneman presented these two scenarios to doctors, minus the bit about baldness, the majority chose Cure A in the first scenario and Cure D in the second. The catch is both situations are the same, just framed differently.

All their logic and rationality was chewed up when their fear of loss was activated. The wording in the first scenario makes it seems as though you get to save 200 people before the first one dies. The overall loss isn’t emphasized. In the second, it seems as though you lose 400 people before you save anyone. The fear conjured by the emphasis on losing lives in the reframed scenario makes it seem acceptable to take a serious risk. That’s how much you hate loss.

When you lose something permanently, it hurts. The drive to mitigate this negative emotion leads to strange behaviors. Have you ever gone to see a movie only to realize within 15 minutes or so you are watching one of the worst films ever made, but you sat through it anyway? You didn’t want to waste the money, so you slid back in your chair and suffered. Maybe you once bought non-refundable tickets to a concert, and when the night arrived you felt sick, or tired, or hung over. Perhaps something more appealing was happening at the same time. You still went, even though you didn’t want to, in order to justify spending money you knew you could never get back. What about that time you made it back home with a bag of tacos, and after the first bite you suspected they might have been filled with salsa-infused dog food, but you ate them anyway not wanting to waste both money and food? If you’ve experienced a version of any of these, congratulations, you fell victim to the sunk cost fallacy.

Sunk costs are a favorite subject of economists. Simply put, they are payments, investments or costs which can never be recovered. An android with fully functioning logic circuits would never make a decision which took sunk costs into account, but you would. As an emotional human, your aversion to loss often leads you right into the sunk cost fallacy.

Loss aversion is one of your strongest drives. You know a confirmed loss lingers and grows in your mind, becoming larger in your history than it was when you first felt it. Whenever this clinging to the past becomes a factor in making decisions about your future, you run the risk of being derailed by the sunk cost fallacy.

Hal Arkes and Catehrine Blumer created an experiment in 1985 which demonstrated your tendency to go fuzzy when sunk costs come along. They asked subjects to assume they had spent $100 on a ticket for a ski trip in Michigan, but soon after found a better ski trip in Wisconsin for $50 and bought a ticket for this trip too. They then asked the people in the study to imagine they learned the two trips overlapped and the tickets couldn’t be refunded or resold. Which one do you think they chose, the $100 good vacation, or the $50 great one?

Over half of the people in the study went with the more expensive trip. It may not have promised to be as fun, but the loss seemed greater. That’s the fallacy at work, because the money is gone no matter what. You can’t get it back. The fallacy prevents you from realizing the best choice is to do whatever promises the better experience in the future, not which negates the feeling of loss in the past.

Kahneman and Tversky also conducted an experiment to demonstrate the sunk cost fallacy. See how you do with this one.

Imagine you go see a movie which costs $10 for a ticket. When you open your wallet or purse you realize you’ve lost a $10 bill. Would you still buy a ticket? You probably would. Only 12 percent of subjects said they wouldn’t. Now, imagine you go to see the movie and pay $10 for a ticket, but right before you hand it over to get inside you realize you’ve lost it. Would you go back and buy another ticket? Maybe, but it would hurt a lot more. In the experiment, 54 percent of people said they would not. The situation is the exact same. You lose $10 and then must pay $10 to see the movie, but the second scenario feels different. It seems as if the money was assigned to a specific purpose and then lost, and loss sucks. This is why Farmville is so addictive people have lost their jobs over it.

Farmville is a valuable tool for understanding your weakness in the face of loss. The sunk cost fallacy is the engine which keeps Farmville running, and the developers behind Farmville know this.

Source: Farmertips.com

Farmville is free, and the first time you log on you are transported to a netherworld patch of grass where you float above an abeyant young farmhand eager to get to work. His or her will is your will, and his or her world is empty save a patch of land ready to be plowed and a crop of vegetables ready to be picked.

Wading into the experience, you feel the game designers have made every attempt to turn your head toward the screen in a way which brings no attention to the grip on your scalp. It is all your choice, they seem to be saying, no one is forcing you to proceed. Here, harvest these beans. Hey, why not plant some seed? Oh, look, you could plow a patch of land, you know, if you want. A loading bar appears and then quickly fills as you watch your grinning Aryan-ish avatar with his messy-on-purpose haircut virtually dirty his digital overalls. The cheery music, which sounds like the cyborg interpretation of clumsily extracted memories from the brain of a reanimated Old West piano player, drones on and on. The moment the loop restarts is difficult to pinpoint.

Within a few minutes, you’ve done everything which can be done on your first garden, but there are hints all over the screen portending a fully functioning Texas-ranch-sized megafarm, should you plant your seeds well. Once you learn you must wait at least an hour or so to continue, you start clicking around and find you have coins and cash which can be spent on trees, plants, seeds, an impressive bestiary of jaunty fantastical creatures and a bevy of clothes, devices, buildings and props. You have just enough currency when the game starts to buy a caramel apple tree or some honeybees, but the nice stuff like pink tractors and magic waterfalls, will have to wait until you’ve played the game a while. If you stay vigilant, checking back throughout the day to see how close your strawberries are to being ripe or if a wandering animal has visited your feed trough, you can earn more virtual currency and advance in levels and unlock more stuff. You’ll need to plant and plow and harvest to advance, most of which is also an investment in something which must be harvested…later.

This is the powerful force behind Farmville. Playing Farmville is a commitment to a virtual life form. Your neglect has consequences. If you don’t return, your investments die and you will feel like you wasted your time, money and effort. You must return, sometimes days later, to reap the reward of the time and virtual money you are spending now. If you don’t, not only do you not get rewarded, you lose your investments.

To stave off these feelings you can pay Farmville real-world money or participate in offers from their advertisers to negate the need to tend to certain things, reverse the death of crops and expand your farm ahead of schedule. You can also ask your friends to help, since the game has tendrils reaching deep into Facebook.

Although all these strategies will keep the fallacies at bay for a few days, they also feed them. The urge to stay the course and keep your farm flourishing gets more powerful the more you invest in it, the more you ask others for help, the more time you spend thinking about it. People set alarms to wake up in the middle of the night to keep their farm alive. You continue to play Farmville not to have fun, but to avoid negative emotions. It isn’t the crop you are harvesting, but your fallacies. You return and click to patch cracks in a dam holding back something icky in your mind – the sense you wasted something you can never get back.

To say Farmville has been successful is a silly sort of understatement. It has led to the creation of a whole new genre of entertainment. Hundreds of millions of dollars are being generated by social gaming, and like so many profitable businesses, someone is hedging their bets against a predicable weakness in your behavior in order to turn a profit.

Farmville players are mired in a pit of sunk costs. They can never get back the time or the money they’ve spent, but they keep playing to avoid feeling the pain of loss and the ugly sensation waste creates.

You may not play Farmville, but there is probably something similar in your life. It could be a degree you want to change, or a career you want to escape, or a relationship you know is rotten. You don’t return to it over and over again to create good experiences and pleasant memories but to hold back the negative emotions you expect to feel if you accept the loss of time, effort, money or whatever else you have invested.

If you dropped your cell phone over the edge of a cruise ship, you would need James Cameron’s unmanned submarine fleet to find it again. Sure, you could spend a small fortune to retrieve it, but you wouldn’t throw good money after bad. Laid out like this, logical and rational and easy to pick apart, you can pat yourself on the back for being such a reasonable human. Unfortunately, the sunk costs in life aren’t always so easy to see. When something is gone forever it can be difficult to realize it. The past isn’t as tangible a concept as the sea floor, yet it is just as untouchable. What is left behind is just as irretrievable.

Sunk costs drive wars, push up prices in auctions and keep failed political policies alive. The fallacy makes you finish the meal when you are already full. It fills your home with things you no longer want or use. Every garage sale is a funeral for someone’s sunk costs.

The sunk cost fallacy is sometimes called the Concorde fallacy when describing it as an escalation of commitment. It is a reference to the construction of the first commercial supersonic airliner. The project was predicted to be a failure early on; but everyone involved kept going. Their shared investment built a hefty psychological burden which outweighed their better judgments. After losing an incredible amount of money, effort and time, they didn’t want to just give up.

It is a noble and exclusively human proclivity, the desire to persevere, the will to stay the course – studies show lower animals and small children do not commit this fallacy. Wasps and worms, rats and raccoons, toddlers and tikes, they do not care how much they’ve invested or how much goes to waste. They can only see immediate losses and gains. As an adult human being, you have the gift of reflection and regret. You can predict a future place where you must admit your efforts were in vain, your losses permanent, and when you accept the truth it is going to hurt.

You Are Not So Smart – The Book

If you buy one book this year…well, I suppose you should get something you’ve had your eye on for a while. But, if you buy two or more books this year, might I recommend one of them be a celebration of self delusion? Give the gift of humility (to yourself or someone else you love). Watch the trailer.

Order now: Amazon - Barnes and Noble - iTunes - Books A Million

Links:

Mother kills her baby over Farmville interruptions

City council member booted for playing Farmville

Using game design to create a better world

Farmville co-creator admits to using sunk costs

Games: Why Zynga’s Success Makes Game Designers Gloomy

Avoiding sunk costs in travel

Sunk costs and virtual goods

Sunk costs and Concorde fallacy psychological studies

Seth Godin on sunk costs

Welcome to Farmville: Population 80 million

Farmville gamemaker Zynga sees dollar signs

Jonah Leher on Loss Aversion

Farmville discussion at Hacker News

The evolutionary basis for the sunk cost fallacy

The push to keep going in war as a sunk cost fallacy

261 Comments leave one →

Will Conley permalink

March 25, 2011 4:05 pm

This is so good. I was mesmerized, reading this article. Frankly, my heart is pounding. It is just so true. I may have to commit this one to memory.

Reply
- David McRaney permalink*
  
  March 25, 2011 4:33 pm
  
  Thanks for reading.
  
  Reply
- Alex Shearer permalink
  
  March 28, 2011 3:40 pm
  
  I feel the same way. First time I’ve read anything on this site, but I’ll be coming back. Thanks for the article.
  
  Reply
- Sean Hudson permalink
  
  April 13, 2011 12:41 am
  
  Agreed. Also the first thing I’ve read on the site, definitely coming back! Such a simple example for such a complex human emotion; gives many thinking points. When you’re aware of such things, it’s harder to be blindsided by it in the future.
  
  Reply
- Setek permalink
  
  April 21, 2011 4:44 am
  
  I agree, though I do not play Farmville, the mention of a relationship being a sunk cost brought tears to my eyes.
  
  Have read all your posts, and love the insight into the psychological mess. Thanks.
  
  Reply
Barrett permalink

March 25, 2011 4:09 pm

“The catch is both situations are the same.” No they aren’t. Being guaranteed to save 200 people means that at most 400 will die. In the other situation you were guaranteed 400 would die, with no guarantee that anyone at all will live, so all 600 could die.

Reply
- mike permalink
  
  March 25, 2011 4:22 pm
  
  The cure A/B/Y/Z scenario has nothing to do with loss aversion. it has to do with diminishing sensitivity relative to a reference point. It is already a bit of a stretch to call the sunk cost effect an example of loss aversion (there are other factors). But this is just wrong.
  
  Risk preferences are determined by the curvature of the value function, not the slope. Loss aversion only describes the slope of losses relative to gains. If a gamble doesn’t involve weighing both losses and gains, then loss aversion can’t possibly affect the decision.
  
  Normally I like this website so I’m happy to correct you in detail, if you’re still not convinced, but either way, this is wrong.
  
  Reply
  - mike permalink
    
    March 25, 2011 4:23 pm
    
    btw I don’t sign onto these other criticisms that the wording is vague though. The question is perfectly clear (at least in the original), rather that your interpretation is wrong.
    
    Reply
  - David McRaney permalink*
    
    March 25, 2011 4:33 pm
    
    Hey, thanks. I’m always happy to get criticized so I can make these posts better. The ABYZ scenario has been used by psychologists to explain loss aversion for years, and I’m using it here to show the emotional quality of loss keeps people from being rational about sunk costs.
    
    Reply
- David McRaney permalink*
  
  March 25, 2011 4:27 pm
  
  It is implied the cure will save the remainder out of the 400 who die.
  
  Reply
  - Kaitain permalink
    
    March 28, 2011 12:43 pm
    
    I think the problem you’ve made for yourself here is that in the scenario described, you can’t really make this unambiguous without drawing attention to the equivalence of the gain-centric wording and the loss-centric wording, which would disrupt the intended sleight of hand.
    
    Reply
- Mitchell permalink
  
  March 27, 2011 4:59 am
  
  That is exactly what i was thinking.
  
  Reply
- Kaitain permalink
  
  March 28, 2011 12:42 pm
  
  Yes, the wording is not ideal. It would have been better to have scenario two as one in which a mad scientist was going to unleash an attempted plague on 600 healthy people, had two beakers with the guaranteed 400 deaths (beaker A) vs 1/3 none, 2/3 all kill probability (beaker B), and you had to choose which beaker was unleashed. The original Kahneman/Tversky paper makes it clear that each person either survives untainted or dies. The background of the plague here complicates the decision, because surviving but still having a fatal disease is not the same as “surviving” in a conventional sense.
  
  There is another complication which messes things up here: if the 600 plague-afflicted people are the last ones remaining on earth, 200 assured survivors guarantees the survival of the species. This is not an issue in the Kahneman/Tversky work. In such a situation, it would be a hard call choosing between:
  
  a) One guaranteed 21 year old male and one guaranteed 21 year old female survivor; everyone else dies.
  b) A 50/50 chance of everyone on earth dying or everyone on earth surviving.
  
  Reply
  - Kaitain permalink
    
    March 28, 2011 12:51 pm
    
    btw I realize that it isn’t stated here that the 200 survivors still have the plague…the more rational interpretation is that they would have been cured, but it IS ambiguous, and the intended effect of being startled by the direct equivalence of two situations where your intuition drove you to two different choices is reduced by that ambiguity.
    
    Reply
- Hel permalink
  
  March 28, 2011 6:07 pm
  
  Actually, it says exactly 400 will die. As in no more, no less. Perhaps the article was edited to make this more clear?
  
  Reply
  - Hel permalink
    
    March 28, 2011 6:09 pm
    
    d’oh, yes, I see several comments down that it was edited!
    
    Reply
Ishmael permalink

March 25, 2011 4:14 pm

I agree with much of this, but there are some flaws here if the wording used in these studies is the same as used here. (If some of these items were paraphrased, then they may not be issues.)

“Cure A is guaranteed to save 200 people. [...] Cure Y is guaranteed to kill 400 people. [...] When Tversky and Kahneman presented these two scenarios to doctors, minus the bit about baldness, the majority chose Cure A in the first scenario and Cure Z in the second. The catch is both situations are the same.”

No they aren’t. Cure A would save 200-600 people. Cure Y would save 0-200 people.
If the original wording was “guaranteed to save/kill *exactly* 200/400 people”, then this changes things, but by providing what appear to be minimums you change the possible ranges and thus the predicted effects.

“Imagine you go see a movie which costs $10 for a ticket. When you open your wallet or purse you realize you’ve lost a $10 bill. Would you still buy a ticket? You probably would. Only 12 percent of subjects said they wouldn’t. Now, imagine you go to see the movie and pay $10 for a ticket, but right before you hand it over to get inside you realize you’ve lost it. Would you go back and buy another ticket? Maybe, but it would hurt a lot more. In the experiment, 54 percent of people said they would not.”

Placing this as a real-life situation, the $10 that you lost in the first scenario leaves a chance that you might recover it. In the second case, the ticket would presumably be on your person or in the theater in most scenarios, and you would have to choose to give up on finding it. Tickets are generally nonrefundable, so buying the second ticket guarantees a loss of $20. Buying a single ticket after having lost $10 again provides a range where you have lost $10-20. depending on the likelihood of your finding the “lost” $10 This variable could be removed by making the item stolen or damaged beyond use rather than simply lost; that way there is no chance of recovering them for later use.

I suspect that the numbers would still lean the same way with these items adjusted, but I wonder if the numbers would be as severe.

Reply
- David McRaney permalink*
  
  March 25, 2011 4:30 pm
  
  As a reporter, I am interested in getting these things explained correctly and simply. So I appreciate your comment a great deal. I have changed the wording to make the zombie scenario make more sense with the word “exactly.” If it still seems out of whack, I’m happy to take suggestions on how to better clarify it. Feel free to email me at davidmcraney@gmail.com
  
  Reply
Qon permalink

March 25, 2011 4:33 pm

I agree with @Barret and @Ishmael, but would also like to point out the that I was a bit Qonfused by 33/66 %. 33+66= 99… Since it would neither kill them or save them I assume that they aren’t affected at all and would of course die because of the disease.

btw I consider the logic and outcome only when I read these test question ;) (programmer)

Still, thanks for a good read!

Reply
- David McRaney permalink*
  
  March 25, 2011 4:38 pm
  
  Thanks. I’ve reworded this section as it was obviously a bit sloppy.
  
  Reply
Patrick permalink

March 25, 2011 4:50 pm

Very interesting article explaining this form of “universal weakness” pretty well. Love your blog. Always feel more smart after reading them, which is pretty rare over a web seemingly filled with pointless celebrity news and time wasting games.

Thanks a lot for these articles.

-patrick

Reply
- David McRaney permalink*
  
  March 25, 2011 4:57 pm
  
  Thank you for helping me keep this going. The book is due out soon. I’ll make a big deal about it once I know when.
  
  Reply
Brakhage permalink

March 25, 2011 5:10 pm

If you’re interested in further examination of the evils of social gaming I’d recommend game designer Jonathan Blow’s lectures on the subject – before I started listening to Blow I knew there was something underhanded about Farmville et al. but I couldn’t put my finger on why.

http://edtech.rice.edu/www/?option=com_iwebcast&action=details&event=2349

Reply
Jeff J permalink

March 25, 2011 5:19 pm

Check out swoopo.com – the most heinous deliberate exploitation of this I’ve ever seen.

Reply
- LiquidBeef permalink
  
  March 29, 2011 6:38 am
  
  Sites down. Is that one of those Penny Auctions? If so, you’re quite right.
  
  Reply
Pete permalink

March 25, 2011 6:22 pm

Great article, as always, very informative and interesting.
Thank you for all your hard work! Keep it up!

Reply
- David McRaney permalink*
  
  March 26, 2011 1:25 pm
  
  Thanks for sticking with me.
  
  Reply
David permalink

March 25, 2011 7:30 pm

Regarding the vacations, it should be noted that diversity of experience may be a factor. If you prefer to have eventually gone on both the ski trip in Michigan and the one in Wisconsin, then going on the Michigan trip should be preferred since it will be most costly to re-acquire. This may be a rational pursuit of one’s goals. The movie ticket scenario is a much clearer demonstration of the sunk cost fallacy.

Reply
Jane permalink

March 25, 2011 8:25 pm

Thanks — I enjoyed reading this. It made me think … but I haven’t slept for a couple of days, and my thinking is pretty murky. I read the whole article, and the comments, and still have just one question: People ski in Wisconsin?

Jane

From Utah. Seriously, Wisconsin?

Reply
Adrien permalink

March 25, 2011 8:28 pm

Do you have any pointers to references using sunk costs in a positive way ? From the top of my head, I guess this kind of investment could be used to entice someone to stop smoking, lose weight, etc.

Reply
- Jeremy permalink
  
  March 27, 2011 5:31 am
  
  One example at my university is a boot camp they run. It costs something like $60, BUT, if you attend and complete 80%+ of the sessions, you get your money back.
  I’m new to the whole concept, but I think that would be an example of sunk costs used positively.
  
  Reply
- Susie permalink
  
  February 22, 2012 12:06 am
  
  Some gyms in Europe use this positively. A membership costs nothing if you attend regularly; miss a week, and you have to pay.
  
  Reply
hk permalink

March 25, 2011 8:49 pm

Just a suggestion…Do psychologists always do the A/B/Y/Z questions in that order? It seems you should present the Y/Z question first, because having already read the A/B scenario, I was in tune to “the trick” and went with Cure Y.

Reply
- David McRaney permalink*
  
  March 26, 2011 1:27 pm
  
  No, it is usually presented as A B C and D. Hmmm. I think I’ll change it to match the original study.
  
  Reply
Akshay N R permalink

March 25, 2011 10:07 pm

Your blog is one of the very few things that makes my internet surfing worthwhile. And each post never ceases to get me thinking about how it has been affecting my own life. This post has done me more good than you can imagine. Really appreciate your effort into writing about this. Please do not stop. Also looking forward to the book…

Reply
- David McRaney permalink*
  
  March 26, 2011 1:49 pm
  
  I appreciate the comment. Thanks.
  
  Reply
Chris Abbey permalink

March 25, 2011 10:07 pm

The most troubling application is when you hear, “We can’t get out of , because then all our troops who died will have died in vain.”

The least troubling is when you keep watching a TV show that has gone all stankfoot.

Reply
- Akshay N R permalink
  
  March 25, 2011 10:12 pm
  
  No s**t!
  I remember the way I watched LOST all the way even after I knew it was trash after the 3rd season…
  
  Reply
- kirk permalink
  
  March 30, 2011 2:52 pm
  
  I used to work as a designer of IC chips in a company whose name rhymes with ‘Motorola’ and sat through about 10,000 hours of meetings where the gory details of ‘what we give up if we quit now when the end is in sight’ from some design team completely out in the weeds. I was in first line management, on the engineers side entirely, prayed for every design to be finished and knew from the first phoneme of the first word on the first powerpoint slide that after 59 1/2 minutes I would get up from my chair, turn on the lights and say: “Don’t ever give me a sunk cost argument, the project was cancelled before we came in the room.”
  
  Nobody.
  Ever.
  Got it.
  
  Reply
Chris Abbey permalink

March 25, 2011 10:11 pm

The words, ‘pick a war’ disappeared out of the previous post.

Reply
- David permalink
  
  March 26, 2011 1:23 am
  
  And yet, it was clear enough anyway…
  
  Reply
- David McRaney permalink*
  
  March 26, 2011 1:52 pm
  
  No idea how that happened.
  
  Reply
- Delphi_ote@yahoo.com permalink
  
  March 30, 2011 12:29 am
  
  Did you put it between greater than and less than signs? The software probably thought it was a bad HTML tag and removed it.
  
  Reply
WenatcheeTheHatchet permalink

March 25, 2011 11:45 pm

The best example in pop culture of the sunk-cost fallacy is how many people watched the Star Wars prequels.

Reply
- alex permalink
  
  April 16, 2011 10:38 pm
  
  Good example! Add to that the people who vehemently defend the films, because admitting to disappointment is so very difficult.
  
  Reply
Loku permalink

March 26, 2011 1:18 am

The misconception: Reading this misconceptions will make you be prepared for when you encounter this situation in real life.

The truth: You will harvest your sunflowers in two hours.

Reply
- WenatcheeTheHatchet permalink
  
  March 26, 2011 2:48 am
  
  Or in my case I’ll still listen to my roommate talk about Star Wars because, well, I watched the movies a lot when I was a kid. :)
  
  Reply
Arek permalink

March 26, 2011 3:18 am

Gosh, I love this blog.

Every misconception you present is easily explained and applicable in real life. I wish everyone on the interwebs would serve me valuable information like this.

Reply
Wok permalink

March 26, 2011 5:08 am

The reference to Concorde is not quite relevant. And, what is worse, the two last sentences of the conclusion are unclear! Otherwise, good job!

Reply
Don Gwinn permalink

March 26, 2011 8:51 am

This is exactly why I read all the books in the Twilight series (in my defense, I’m a middle school teacher, and everybody was doing it at the time. Hey, it was the Oughts, it was a crazy time. You have no right to judge me, man.)

I have you linked on my own blog, but in an interesting coincidence, I was just reading a forum where we were discussing sex-starved, unhappy marriages and how rare it is that the “high-desire partner” asks the “low-desire partner” the direct question, “Why are you still here?”

I submit that the answer often has a lot to do with sunk cost, and the HD partner may not know that term, but he probably realizes he might find an honest answer painful.

Reply
Gitona permalink

March 26, 2011 8:54 am

I enjoyed reading this article…probably my favorite in recent time. I understood what you were trying to get out…though the Cure part did take a reading over.

While I don’t play Farmville(and never will) there are other things in my life that represent the same fallacy.

Reply
Shae permalink

March 26, 2011 10:38 am

One time I got a giant diet soda that tasted awful, and I kept drinking it. Then I suddenly realized the very point that this article is making. The two dollars was already gone, the soda wasn’t giving me pleasure, I wasn’t dying of thirst, it wasn’t even providing calories. There was literally no value in continuing. So I tossed it out.

The “dollar auction” is a useful illustration here too.

Reply
Verlene Caliguire permalink

March 26, 2011 10:53 am

Thanks for post. I keep more interesting publications. Been following blog for five days now and I should say I am beginning to like your article this site. I need to know how can I subscribe to your blog?

Reply
kennyth permalink

March 26, 2011 1:45 pm

This is a nice article; however, I object to the implications of the concluding paragraph. You write that “studies show lower animals and children do not commit this fallacy,” which makes them sound more rational than the rest of us. As you say, “Wasps and worms, rats and raccoons, toddlers and tikes, [...] can only see potential future gains.” Most psychology and animal behavior research shows that the adult human is one of about three creatures that can imagine the future or the recollect the past. I think it’s a safe bet that most animals and small children are not making decisions about potential future gains.

Reply
- David McRaney permalink*
  
  March 26, 2011 1:57 pm
  
  Hmm. Yes, I see how the word “future” could muddle the point there. I will scrub it.
  
  To your point about animals and children being more rational, yes, they are more cold and logical in some ways – machine like. I’m suggesting with emotional maturity comes some drawbacks. An emotional human is able to regret and fear loss, but at the expense of sometimes over-fearing and over-regretting it.
  
  Reply
Daniel F Moisset permalink

March 26, 2011 3:35 pm

Nobody has mentioned yet Cow Clicker; besides being a parody of Farmville, it was specifically designed as a social experiment about these issues: http://www.bogost.com/blog/cow_clicker_1.shtml

Reply
ChaozUT permalink

March 26, 2011 5:48 pm

“The fallacy makes you finish the meal when you are already full.”

Whenever my mom gives me too much food and I don’t finish it, I am told that I am wasting the food. However, I tell her that regardless of whether I eat it or not, the food has already been made and that, whether I eat it or not, I’m still going to be eating roughly the same amount the next day, so if I stuff myself and finish eating the rest of the food given, all that I’d achieve is to fatten myself up. I guess my mom is a victim of the sunk cost fallacy? @_@

Reply
S. Bauer permalink

March 26, 2011 6:40 pm

This is one of the basic emotions which causes countless amateurs to lose a fortune at the stock market.

For example, when the dot-com bubble burst, most people knew that their hot internet stocks were more or less worthless. The only rational action would have been: Sell immediately.

But this was impossible to do for so many people because of the emotional investment. They payed $100 per share, and selling at $50 per share would mean to admit the loss. So they hold their stocks, in the feign hope that the price will come back, some day. “Sunken cost fallacy” at its best.

Interestingly, at the end of bear markets, a new emotion sets in: Panic. This is when stock prices have fallen for months to ridiculously low valuations. Now the pain of loss gets over a threshhold and becomes unbearable. In these market phases, everything is sold in panic, even solid companies with good earnings, and people swear to never ever touch stocks again. Usually, stock prices start to climb shortly afterwards…

Reply
Brandon permalink

March 26, 2011 10:15 pm

David, I really enjoyed this post. Keep it up.

Reply
C. Bergsten permalink

March 27, 2011 8:29 am

Nice article, but I take issue with the “choose the cure”-scenario. I do not seem to be alone in this, but I do belive that I have a somewhat different perspective.

Here there are, for the moment, 600 humans left on the planet. Using Cure A/C, 200 people will survive, while using Cure B/D there is 1/3 chance that all 600 will survive and 2/3 chance that none will.

For me, it’s not the way you formulate the premise that makes the difference. The numbers come out the same. The issue is that they’ve taken the premise too far to the extreme and here is why.

From genetics we know that you need roughly 160 extremely healthy, unrelated and promiscous humans, 90% women, in order to rebuild a self-sustaining population. If you do not meet those requirements everyone will die in a few generations because everyone would eventually become infertile due to inbreeding. Adding other illnesses, accidental and intentional deaths and the situation grow even grimmer.

Assuming that any apocalyptic disease will not leave a population of those exact proportions, but rather keep to standard population division of approximately 50% men, 50% women with a distribution from the young to the elderly (of course fewer survivors in the young and elderly than in a standard population, as with most diseases), you can immediately see that cure A/C is just a complete extinction, delayed a few generations. Even with the total 600 survivors, having enough people to meet the requirements for sustained survival is borderline.
So regardless of how you choose to formulate the scenario, I would summarise the result as follows:

No Cure: Complete extinction at the rate of the initial apocalyptic disease.
Cure A/C: 200 Cured survivors, extinction within a few generations.
Cure B/D: 2/3 probability of immediate extinction.
Cure B/D: 1/3 probability of 600 cured survivors with a small chance of getting a sustained surviving population.

When looked at from this perspective, there really is only a single acceptable option, isn’t there?
Though I might add that the scientist that developed the cure should know, or be able to extrapolate, this. The fact that he seemingly doesn’t call to question the quality of both proposed “cures” anyhow.

While it is useful to work with extreme situations in experiments, it may easily take things too far. Unexpected factors may invalidate the result to some degree. The above information immediately entered my mind upon reading the scenario. I expect that in this I am somewhat of an aberration, but this perspective is worth considering, wouldn’t you agree?

Reply
- Annie permalink
  
  March 28, 2011 11:20 am
  
  @C. Bergsten: You’re completely missing the point of the example. Also, I suspect you are a douche.
  
  Reply
  - John permalink
    
    March 28, 2011 1:14 pm
    
    @Annie: I actually thought it was the best comment yet. Yes, the commenter is making a separate point than that made in the article, but I thought it was interesting and worth reading and I learned something from it.
    
    Your comment, on the other hand, was the first bit of incivility that I noticed in this thread. I’m not sure exactly what makes someone a “douche”, but you might consider pondering that question for a few minutes.
    
    Reply
    - ladeeG permalink
      
      November 3, 2011 10:04 pm
      
      I think that the point that Bergsten was trying to make is that oftentimes, decisions that we make are influenced by a whole host of factors, including but not limited to emotional investment, and that simply framing things in an economic cost/benefit way can be over simplifying the issue. For instance, in the taco example, perhaps the person who bought the tacos has an ethical issue with wasting (so called—ick) food. Maybe they’re super Boho and they actually can’t afford to buy something else, because it’s just not in their budget, and they need the ten bucks that they spent for subway fare. In the vacation example, perhaps they took the more expensive trip on the assumption that they would be able to get a similar price on the Wisconsin vacation in the future, so in that case they actually WOULD be minimizing their losses. These models are too simple to account for all the rationale that goes into human decision making.
      
      Reply
  - C. Bergsten permalink
    
    March 29, 2011 7:12 am
    
    @Annie
    I know very well what the intended point was with the scenario in question. What I am saying is that because the numbers was taken to such an extreme the scenario can’t be used to make the point while still expecting accurate results. The available outcomes simply aren’t what was intended. You’d then either need to answer the questions “how large a proportion of the test subjects will have the knowledge presented? Would that scew the results beyond acceptable limits?”, or change the scenario to a slightly less extreme version.
    I feel this is a valid course of inquiry when someone is trying to make a point by example.
    
    As for me being a douche. I’m constantly surprised how people seem to make these extraordinary conclusions about another individual, only based on how the reader themselves interpret emotional content from a short piece of writing.
    I’m not an author with experience and knowledge on how to write to infer a specific emotional response but rather I write clinically. I’m not even from an english-speaking country. This problem is, as it happens, the reason why smiley-figures were invented, but I digress.
    I could make a similar assumption about you, based on your short response… That you’re a rude person with a strong fanatical core, possibly of some extreme religious upbringing where dogma, and not discourse, is the norm.
    Does this actually describe you? Probably not, but it’s very easy to interpret you like that, based on your response. What I do, is give you the benefit of doubt. You’re probably a nice person, much like any other, that just happened to stray somewhat, intentionally or unintentionally.
    One short piece of writing does not say everything about who, or what, a person is. So, please, next time, courteosly give the benefit of doubt. Your interpretation may not be the intended one. But then again, that was the point I was trying to make with my original comment on the scenario in question, and so we’ve circled back to the beginning.
    
    Have a nice day.
    
    Reply
    - StAB in the dark permalink
      
      March 29, 2011 11:53 am
      
      @C. Bergsten: Your point is interesting, but I disgree with you. Isn’t the original point that aversion to loss caused people to pick different outcomes when the problem is framed differently? The point is not whether A/C or B/D is the better choice, but that if you chose A in the first scenario you should choose C in the second since the outcome is the same. Or, using your extinction outcomes and assuming a preference for the continuation of the species, the obvious choice should be B in the first scenario and D in the second. The point of the experiment is not affected by B being a better choice than A. Whether you think A or B is the better choice for the first scenario you should be making the same choice in the second scenario if you are choosing rationally.
      
      If anything, if either A or B is an obviously superior choice, then it should be easier to choose it in both scenarios.
      
      Reply
      - C. Bergsten permalink
        
        March 29, 2011 5:31 pm
        
        @StAB in the Dark
        The real key in this experiment is the desire in the subject to pick the better choice. If there actually is an obvious better choice, then the phrasing is insignificant and the sunk cost fallacy isn’t applicable. You need a dilemma for it to apply. Cost vs. benefit. That dilemma is something you can influence through phrasing. I maintain that because, with the knowledge I put forward, there is no dilemma for the fallacy to exploit. No matter how you phrase it one of the two options is only cost and no benefit.
        Switching the scenario beyond this extreme, say by using 800/2400 instead of the 200/600 used, you immediately eliminate the issue I put forward (though there may be others here unmentioned).
        By using extremes you can and will add unexpected factors, but you’d best make sure they don’t influence what you’re trying to measure. Here it does.
        It is a point, disconnected to the actual fallacy the article concerns, but it stands.
        
        Reply
        
        StAB in the dark permalink
        
        March 30, 2011 8:51 am
        
        I concede that having that knowledge could have sabotaged the experiment. My contention is that since 72% of the respondents in the experiment chose secnario A and a majority chose scenario A and then Scenario D, the experiment was largely unaffected.
        
        Reply
        
        C. Bergsten permalink
        
        March 30, 2011 11:23 am
        
        You’re right, and I haven’t disputed that. They may have gotten away with it here, but the issue is still there. There is something to be said for critically examining things, especially since I contest that most people that read about, create or participate in these studies doesn’t question themselves or the content enough.
        
        At the moment I’m also curious at how thin columns we can get in our replies, due to the current website discussion format. ;)
        
        Reply
        
        Luisa permalink
        
        April 5, 2011 4:25 pm
        
        I know. That creeps my designer self out. Love the blog, jate the website.
        
        Reply
        
        Luisa permalink
        
        April 5, 2011 4:26 pm
        
        *hate. I can´t even see what I type.
        
        Reply
- Kyle permalink
  
  March 30, 2011 3:05 am
  
  Yes, this thought is a bit off-topic, but an interesting one. And that’s fine by me. I am glad to be at a place where people think and share their ideas.
  
  I think the difference between your conclusion and the test norms are your detachment from the existing individuals and attachment to the wellbeing and future success of humankind. I personally focused on keeping the largest number of survivors alive while not completely forsaking mankind on a no-loss gamble. But with your numbers, my answer wouldn’t have guaranteed humans make it, which I feel should have been the ultimate goal.
  
  Your “From genetics” section was pretty much all new information to me, much like things in this blog post. Do you have any links that explain more about that?
  
  Reply
  - C. Bergsten permalink
    
    March 30, 2011 12:23 pm
    
    @Kyle
    Yes, I believe you’re right. I would further say that with the prerequisite knowledge the kind of detatchment you mention will probably manifest itself in just about anyone.
    
    As for reading material and such. It’s been quite a few years since I studied these things myself so I don’t have it on hand, but a few searches for information about MVP estimates (Minimum viable population), Population bottlenecks, Population genetics etc. will give you a place to start. How deep you choose to go, is up to you.
    
    This though, is a decent summary of MVP:
    http://www.eoearth.org/article/Minimum_viable_population_size?topic=58074
    
    You’ll read that the estimates vary quite a bit (500 – 50 000 individuals), with a median around 5000. The variation is huge because of differences in environmental factors.
    Since humans most likely will congregate, and be able to actively control the breeding process, we can get a much lower MVP than the average species. It does however require our surviving women to breed like bunnies for a long time, and monogamy will simply not be possible.
    The MVP of 160, with prerequisites on the population makeup, that I presented is more or less a bare minimum, which requires everything to go right, for survival to be possible.
    
    Reply
    - Kyle permalink
      
      March 31, 2011 11:19 pm
      
      Good stuff. Thanks-
      
      Reply
      - Luisa permalink
        
        April 5, 2011 4:35 pm
        
        I really don’t care about the survival of the species, only the survival of my self. I assumed A/C cure wasn’t enough for everyone, but I would definitely get my dose, so I chose that.
        
        If we all die in a few generations, I don’t really care, I won’t be here anymore.
        
        Then again, if the world population is down to 600 people, what’s the point of living, anyway…
        
        Reply
Matt permalink

March 27, 2011 9:03 am

Interesting read. Well done, David!

Reply
Dylan permalink

March 27, 2011 6:05 pm

While I certainly wouldn’t doubt you have many valid examples of classical sunk cost fallacies, there are many cases that aren’t so clear cut – and I’d argue Concorde was one of them. It *is* possible that with the right combination of technological breakthroughs, good management and even an amount of good luck Concorde could have ultimately become a highly profitable and successful enterprise, and there was no point at which the costs that had been sunk into the project were completely irrecoverable, except of course when it went bankrupt.
One thing I think could help people deal better with sunk cost problems is accepting that in real life there are very few *truly* sunk costs – even losing a $10 bill can be a valuable experience, if it teaches you to look after your money more carefully (or invest in a new wallet!). And in the case of something like Farmville, just look at the time you spent on it as a way to keep yourself entertained or amused.

Reply
NomadicNeill permalink

March 28, 2011 8:11 am

Fun article, reminds me of the one in GamaSutra explaining all the various ways games are designed to get you addicted to them.

I wonder what the psychology of people who seem less susceptible to this is like?

I’ve stopped reading books that I don’t enjoy, walked out of the cinema etc.

Not sure if it’s because I’m more logical or just have a stronger focus on the future and better things I could be doing with my time, effort, money etc.

Reply
Rob Close permalink

March 28, 2011 8:18 am

Great article. Found it on Reddit, shared it on Facebook. Hope all my friends read it.

Reply
Bogdan permalink

March 28, 2011 8:43 am

Great post as always : )

A good way I trick myself into avoiding the “sunk cost fallacy” is to rephrase the situation thusly:

I pay $10 for a movie, and after 15 minutes I realize the movie sucks. I want to stay for the rest of it so I don’t “waste” that $10.

Then, I tell myself, “If I stay here, then I just paid $10 to suffer for another hour and 15 minutes… and it’s stupid to spend money to feel bad.”

Reply
Daniel permalink

March 28, 2011 10:23 am

You should start tweeting your articles again

Reply
Jon permalink

March 28, 2011 10:29 am

This post is so true! As I was reading it, the following real-life scenario came to mind:

1. I rent a movie from the movie rental store.
2. I receive a disc from Netflix.
3. I start watching a streaming movie on Netflix.

If the movie doesn’t seem all that interesting, I tend to do the following:

1. Finish watching the movie regardless. I can’t remember ever not finishing a movie I’ve personally rented. (Not counting a movie the wife wanted to rent.)
2. Probably finish the movie, but there have been a handful of times I’ve stopped a movie and just sent it back. Half felt guilty about it though. (Now I know why!)
3. Movie not drawing me in immediately? Push stop. Good movie but halfway through I’m tired of sitting around watching movies? Push stop.

In each scenario, if the movie is bad and I watch it anyway, I’m wasting 90+ minutes of my life that should be devoted to enjoyable entertainment. Yet when it’s a rental from the store (even from the $1 kiosk) for some reason I’ve always felt that I needed to finish the movie regardless.

Kind of funny if you think about it. :)

Hopefully I can remember the skunk cost fallacy next time…

Reply
Jack permalink

March 28, 2011 10:30 am

I’ve been basically trying to tell this to my mom for the past year. I know se wont listen, but I posted it on her wall nonetheless…

Reply
Dr. Pete permalink

March 28, 2011 10:38 am

The worst part about the sunk-cost fallacy is that, even knowing you’re doing it, you still can’t help yourself sometimes. I’m a cognitive psychologist by training, and I still got pulled into Zynga games. I never gave them actual cash, but I friended 100s of people to play Mafia Wars and spent what must have been in the realm of 100-300 hours on it. It sounds ridiculous to people who haven’t been lured into it, but that’s exactly the danger. We ALL think we’re “above” these mistakes. I knew the game was a colossal time-waster and, worse yet, that it wasn’t even fun to play. I was actually waking up stressed out about it.

It’s probably also the reason I’m a Cubs fan ;)

Reply
Klaris permalink

March 28, 2011 10:47 am

this is life(style)-changing!

Reply
Alli permalink

March 28, 2011 10:54 am

Man i sure hate the pain I is feel when I lose cash! Almost as much as the pain when you is no edit your article :)

Reply
Stubblychin permalink

March 28, 2011 11:12 am

I’ve had two examples of this recently, one a success.

Need for Speed: Hot Pursuit on Xbox360. Got great reviews and everyone seemed to worship it, so it bought it new (I usually rent). Although it seemed an ok racer, I made me incredibly angry, with cheating AI and fickle controls. I swore a lot. I kept playing it marginally longer than I should of because I’d bought it (if I’d rented it I would have sent it back much sooner). Thankfully I realised this was wrong and exchanged it for a violent foul-mouthed game called Bulletstorm, which doesn’t make me angry and doesn’t make me swear at the television (some irony there I think)

I think I might be in a stuck cost fallacy at the moment though with the Stephen King Dark Tower books. I’m on the last book and really struggling. I’m sticking with it so I can read the ending and because I want to complete the lot. I really didn’t like Wizard and Glass (3rd or 4th in the series) but read it because I read the previous books and hoped it was good as those.

Reply
Avinash permalink

March 28, 2011 11:15 am

The line:
“Outside of the lab, the pain you feel when you is lose cash is twice as strong as the joy you feel when you gain an item of equal value.”
shouldn’t it be:
“Outside of the lab, the pain you feel when you lose cash is twice as strong as the joy you feel when you gain an item of equal value.”

Reply
Linkoroo permalink

March 28, 2011 11:35 am

Wow! This is a great article. Some of these facts are scary. It’s also sad that some people are really deep in these kinds of games. They’re afraid of lost but they don’t gain anything :P

Reply
Damned permalink

March 28, 2011 11:36 am

Have you heard of a game Minecraft? In it you are a character in a world made out of blocks (stone, dirt, wood, etc), with different animals and enemies. In it you have freedom to mine blocks, use them to build forts, dig holes, explore diferent caves, create different items and use them in million ways, etc. After a while I found myself asking a simple question: ‘What is the point?’ and the answer was simple enough, there is none. The more stuff you build the greater the sunk cost feeling is. Some people probably enjoy it, as an oportunity to express themselves in a way, but I believe the truth is that after a while you just feel sorry for the time you lost and don’t want to quit… Sure hope I won’t get trapped like this again: ) Btw. great article!!

Reply
- Kaitain permalink
  
  March 28, 2011 12:48 pm
  
  But you could make that argument of any game at all: what’s the point of chess, for instance?
  
  I think the key here is that Farmville is a curiously dull and unrewarding game by most conventional metrics, yet people keep ploughing on with it.
  
  Reply
  - Kaitain permalink
    
    March 28, 2011 12:48 pm
    
    (Plough pun unintentional, but I’ll take credit anyway.)
    
    Reply
  - Vinternet permalink
    
    March 28, 2011 5:08 pm
    
    Actually, the main difference is that Farmville continues making money off of microtransactions throughout a player’s lifespan, so its developers are more motivated to intentionally bake these compulsion loops into its gameplay. When Minecraft, Pokemon, or Animal Crossing do it, it’s okay because it’s just clever game design, but when Farmville, WoW, or Call of Duty do it, many people cry foul because these games make money off of the subscriptions and future micropayments that this psychology helps to fuel.
    
    Reply
    - Kaitain permalink
      
      March 28, 2011 7:20 pm
      
      Only a small percentage of Farmville’s players ever pay any money. Obviously Zynga design it in the HOPE that players will pay money, but few of them ever do. (Many remain addicted regardless.)
      
      Reply
- Gerri permalink
  
  May 2, 2011 10:01 am
  
  Such a deep anwser! GD&RVVF
  
  Reply
- tshqnmeyhkk permalink
  
  May 4, 2011 12:11 am
  
  pTrcRT qyfmhoaunyjh
  
  Reply
jimmy permalink

March 28, 2011 11:46 am

Reminds of me of Tomigotchi, which was a digital toy representing an animal that you had to feed and walk and play with. It required constant attention.

Reply
- Luisa permalink
  
  April 5, 2011 4:52 pm
  
  I had one as a child. I remember grown ups were concerned that we (children) would get too involved in it, and stop studying, eating, sleeping…
  Well, my baby dinossaur died, and so did every baby dinossaur in my tamagotchi. I just knew they were just pixels and weren’t acctually getting hurt, so it didn’t upset me, and I never once thought of the time I had wasted. Guess I was too young for sunk cost.
  
  Reply
Shanoboy permalink

March 28, 2011 12:02 pm

I wanted to stop reading after the first few paragraphs, but I figured I had already spent a minute or so on the article, that I should just keep going so that that first minute wasn’t wasted in vain…

Nice article. :-)

Reply
Kaitain permalink

March 28, 2011 1:20 pm

Suggested altered wording for the second scenario:

“Ok, mark your answer and let’s reimagine the scenario. Same setup, everyone is going to die without a cure. This time, the scientist has two options, both of which will cure the disease but have a risk of overloading the body’s immune system and killing the recipient. If you use Cure C it is certain exactly 400 people will die from the overload. Cure D has a 1/3 probability of killing no one, but a 2/3 probability of killing all 600. Which one?”

Reply
Kaitain permalink

March 28, 2011 1:26 pm

Re: the general problem of the sunk cost fallacy. It would be useful if people could be taught to think of everything they buy as being an OPTION. In finance, if you purchase an option to buy a certain stock at a certain price at a certain time, you don’t have to execute it when the time comes. If you bought an option to buy 100 shares in Microsoft at $20 a share on July 1st, but when July 1st comes Microsoft is actually trading at $15 a share, there’s no point in exercising the option. Sure, you paid to have the option, but using it is clearly counterproductive.

Similarly, you should view vacations, meals, cups of coffee etc. the same way: as options. If you paid $500 dollars for a vacation in Hawaii, but when the time comes you’re in bad health and think you’d actually prefer to stay at home, you let the option expire, because it isn’t worth exercising. If you buy a coffee but start to feel like you’ve already drunk enough caffeine that day, let the coffee option expire unexercised.

EVERYTHING you buy is an option, and doesn’t have to be used if it doesn’t make sense to do so.

Reply
cidus permalink

March 28, 2011 3:05 pm

This is one of the best articles on this site ever. Congrats. =)

Reply
T. AKA Ricky Raw permalink

March 28, 2011 4:53 pm

In addition to the sunk-cost fallacy, I’d say the other big psychological phenomenon that comes into play here is Intermittent Rewards or Random Reinforcement. Every time you come back to your farm, there is something different. You could lose crops or have new crops appear.

http://therawness.com/the-compliance-recipe-part-3-intermittent-rewards/

Reply
Dirac permalink

March 28, 2011 6:52 pm

This is an interesting article. I wouldn’t be surprised if this was an evolutionary thing; limited resources means those who hold on to what they had were more likely to survive, those most likely to hold on to what they have were those who were most crushed when they lost things.

One thing I noticed is that of the two presented cure scenarios, the second set did not resonate in a way differently from the first one, I did the math and the first option seemed ideal given the circumstances. Perhaps this was because I was expecting this trick within the context of the article.

With Farmville, I remember playing it briefly due to the buzz I saw from other Facebook friends. I planted crops, returned to the game a day later, and saw that the crops had grown, died, and withered. This instantly turned me off to the game and that was the last time I played it.

I’d never played another game on consoles or PC where I had to manage anything in real time. I turn off the game, and when I come back, everything is roughly where it was when I left it. I wonder how much of the Farmville demo is made up of people who don’t really play video games? I don’t want to sound arrogant in saying that they might not know any better, but that might be the case.

Reply
Amanda permalink

March 28, 2011 7:05 pm

I honestly think Farmville is fun. If my crops wither, I just plow over them and start again. I don’t feel the need to buy cash to unwither them. I’ve left other Facebook games because I didn’t enjoy them any longer. Are you saying it’s impossible to enjoy these games at all?

Reply
Zé estevão permalink

March 28, 2011 7:25 pm

Brilliant analysis I must say! A real eye opener!

Reply
- Tom permalink
  
  March 28, 2011 8:08 pm
  
  This is a great article and has really hit me quite hard, i play a game online that i started about 4 years ago, over time it has lost its appeal to me and i simply dont find it fun anymore but because i have spent such an enormous amount of time in the game i feel that if i leave it i will be loosing all that time in my life when really, its already gone and will never come back.
  
  Even while writing this im fighting with myself whether or not to delete my account in the game knowing that ive spent all that time and money on it even knowing very clearly that this sunk cost fallacy is happening to me. Brains are really annoying…
  
  Reply
Terry permalink

March 28, 2011 7:46 pm

Sunk cost is why I skimmed this article right to the end, even though the point was in the first 4 lines.

Reply
Justin McAleer permalink

March 29, 2011 8:15 am

This is probably the most valuable post yet to the average person. I don’t find the taco analogy to be appropriate, though. That strays beyond merely the sense of loss and into the realm of pure waste on a tangible sense. If the tacos had been tainted or something, ie incurring some additional loss, that would go along. But not eating them simply because they didn’t taste good is indeed wasting food in an ethical sense, plus incurring the additional loss of paying for some other food. It seems irrational to me to make that choice.

Reply
Kitty permalink

March 29, 2011 8:44 am

I disagree. If I don’t like a movie after 10 minutes, I would just leave. Why pay to do something I don’t like and has no future benefit? Its the reason I quit Cityville. Why waste so much time clicking on stuff to get more stuff when we only have limited time on Earth?

Reply
- T. AKA Ricky Raw permalink
  
  March 29, 2011 6:38 pm
  
  That’s fine and dandy, and congrats for being apparently immune to the sunk cost fallacy, but the post isn’t about what Kitty would do, it’s about what the average person would do.
  
  Reply
  - Reader permalink
    
    April 15, 2011 12:57 pm
    
    I would leave, too, and have done. I think an attempt should be made to understand why certain make different choices. In addition, a walking out on a movie is a very low-stakes choice, so this fallacy would have to be analyzed in different contexts.
    
    Reply
Metaronic permalink

March 29, 2011 5:23 pm

I’ve attempted to use sunk cost fallacy as a motivational tool with limited success. I end up paying for a year of time at the gym and then I feel that if I don’t go to the gym, the money I spent is “wasted”. While I know intellectually this is not true, I don’t actively try to fight this impulse since going to the gym is good for me (and I would exercise otherwise).

Although, I suppose since I know that I am trying to trick myself into going to the gym, it’s not as effective.

Reply
Ancel De Lambert permalink

March 29, 2011 10:04 pm

I came across this fallacy a year ago. After reading the article I reflected on my life and made a decision. I bought a new computer, the ground up, and gave my old one to my brother. A year later, I freaking love my computer. It’s faster, rarely screws up, and had a short learning curve. Without knowing about this fallacy I would have kept sinking good money into a defunct machine uselessly upgrading in a losing-man’s game. That’s pretty much the reason I love this blog, reading about my human tendencies and then being able to say “stop that” when I catch myself doing something stupid later on.

Reply
Antoine permalink

March 30, 2011 6:03 am

I think you forgot Dan Gilbert in your sources : his talk about “Our Mistaken Expectations” on TED.
Just saying…

Reply
theotherone permalink

March 30, 2011 5:23 pm

I’m not so sure that the sunk cost fallacy is really uniquely human. Monkeys display at least some forms of loss aversion, see
http://www.q-group.org/archives_folder/pdf/spring2008/ChenBehavioralBiases.pdf

Reply
shotgunner permalink

March 31, 2011 6:06 pm

This is why I’ve forced myself to accept such losses or not to regard them as losses as often as I can. Sometimes it’s leaving something early enough, other times just saying “good riddance”.
For example, when I stopped playing a F2P MMORPG I’ve had invested a lot of time and some money into… well, as soon as I was sure I wasn’t returning to it, I logged in for a last time and gave my friends (who were still playing) the considerable amount of in-game gold I had. I told myself: “Well, you don’t play other games for ever, either. You stop when you go through their content or get bored. You got bored with this one, cherish the good times and move on.” I’m happy I did it, such little experiences help with the bigger ones.

Reply
FNWK permalink

April 1, 2011 1:17 am

I hope Mormons are reading this… it will hopefully resonate.

Reply
Nina permalink

April 1, 2011 9:15 pm

This reminds of some of the pitches on “Dragon’s Den” and “Shark Tank.” People who have sunk thousands of dollars int0 terrible products refuse to give up on their dreams, even after being told again and again why their idea will never fly. They just keep borrowing more money and wasting more time, piling more and more debt onto their worthless inventions. It’s honestly pretty tragic.

Reply
John permalink

April 2, 2011 10:00 pm

Just stumbled on this site, awesome… But I though I would just add my two cents to the author’s conclusion. He says that only adult people make the sunk cost fallacy, and that this is a biological byproduct of being an adult. But isn’t it cultural too? I remember my grandfather telling me to drink the milk in the bottom of my cereal so as not to waste. Movies, television, schools and our parents all drum into us the idea that if we “keep at it” the breakthrough is right around the corner, and if we stop we won’t realize how close we were. It seems like this theory of loss only happens sometimes, I don’t remember the last time I feared the loss of a disposable razer or a plastic straw. I’m no expert so I might be totally wrong, but just though I’d point out that conclusion and see what everyone thought!

Reply
Jonas permalink

April 3, 2011 7:06 pm

I have read quite a bit about cognitive biases, and I was already happy to have stumbled upon your website some how.
But after reading this post I am excited and a little scared at once. If all of your posts are this elaborate and complete, then you are my new blogging hero.

This is absolutely fabulous, kudos to you good sir,

Jonas

Reply
Daniel Reeves permalink

April 4, 2011 2:00 am

Here’s a short post about how some people hypercorrect for the sunk cost fallacy: http://messymatters.com/sunk

It also has a quiz to see if you fall prey to it.

Reply
Jeffrey Lin permalink

April 4, 2011 1:35 pm

Found this through a RT on twitter. Subscribing to your RSS now! Yeah learned a lot about sunk cost fallacy in college econ but this was much better explained…and by using examples everyone knows and probably experienced, I’m sure it’ll make more intuitive sense than just an “Economic theory” where people’s eyes glaze over.

Reply
Grega Laura permalink

April 5, 2011 2:56 pm

Excellent reading. Hope your book comes out soon.

Reply
sassa permalink

April 11, 2011 1:11 pm

I chose vaccine B, but for a different (selfish?) reason. I have noticed that my individual survival rate is 1/3 whatever the vaccine. This means that with probability 2/3 I don’t care how many people survive. At the same time, the probability of my close friends and relatives surviving with me, reduces exponentially if I choose vaccine A.

This opens a chasm of moral questions.

Reply
Reader permalink

April 15, 2011 12:53 pm

“The fallacy prevents you from realizing the best choice is to do whatever PROMISES the better experience in the FUTURE, not which negates the feeling of loss in the past.”

Isn’t this a key problem?

The future is uncertain; “promises” is not the same as “WILL definitely be better.” By contrast, people understand enough about themselves to know they will definitely feel a sense of loss. Perhaps some would rather assuage their predictable bad feelings rather than take a risk.

Reply
Nana permalink

April 15, 2011 2:28 pm

That’s why the Stock Market is so stressful.
That’s why I hate gambling.

Well, my decisions uses to be based on the pleasure of the present moment, even when them will affect my future negatively. Bad, very bad…

Reply
Fibbo permalink

April 18, 2011 8:40 am

I see a problem when applying the sunk cost fallacy to the cinema example above, or in fact any work of art that transcends mere “entertainment”: How can I be sure I’ll still dislike the movie after it’s finished? How can I be sure that I’ll perceive the time I’ve spend reading a novel as wasted when I only read the first 20 pages and found them boring? Personally, I always sit through movies or books, even if I don’t like them (well, except there is obviously no merit to them whatsoever), and I pride myself on doing so consciously.

I don’t deny that this is the SCF at work here, but I assert that in the fields of art and entertainment SCF does yield some positive effects. It’s happened to me often enough that I realised the ingenuity of a piece of literature or art only in retrospect, while thinking of it as dull or unappealing at first. Therefore, I’d rather give a movie, book or whatever the benefit of the doubt and risk wasting a few hours of my life than ignoring a potentially great piece of art because I was too fixated on my getting instant gratification.

To illustrate my point, here’s what Matt Groening has to say about his experiences with Captain Beefheart’s “Trout Mask Replica” album: http://www.beefheart.com/datharp/groening.htm

Reply
Stubblychin permalink

April 18, 2011 9:05 am

If anyone needs a really good example, you will often experience this when you first start playing Poker. Beginners usually get ‘invested’ in a hand despite it being highly unlikely that they will win the hand. They continue to throw good money after bad. This makes a huge difference in Poker.

If you can overcome the Sunk-Cost Fallacy in Poker you will do better as you can use the money to actually bet on a good hand, and maybe win the money back that you had sunk earlier!

Reply
Walmart permalink

April 18, 2011 11:58 am

Come and get dozens of discount products, including baby care, gorceries, games, and more at Walmart!

Reply
Roy A. Ackerman, Ph.D., E.A. permalink

April 23, 2011 8:29 am

This is a fantastic piece. I never articulated this well as the reasons why I would not play Farmville!
It’s the primary reason why people hate change. Even if things are not going well- it’s the sunk costs of years of doing something that worked (or almost worked).
Thank you for writing this piece!

Reply
Stephen Briggs permalink

April 25, 2011 11:50 am

I just found this website and I have to say, my god what a fantastic, intelligent, accurate and informative article, this is pure brilliance I could show this to thousands. Which I believe I will do.

Reply
Ronando permalink

April 25, 2011 8:34 pm

The sunk cost fallacy is a more of a motivational issue for those who are not creative enough to think their way out of a conundrum; something for those with too much “on paper” thinking to allow money to fly out the window. The examples provided to prove the point and validity of sunk cost generally require a certain amount of suspension of disbelief.

In your ski trip scenario, I picked the more fun $50 ski trip, without reading what the answer was that we were supposed or expected to pick. I would have then done a charge back on the $100 airline ticket. Why? Because I can, just like in real life.

Sunk cost is an opportunity for people to show off how smart they are, especially if the money that was at risk isn’t theirs or if they had the fortune to study economics in college. I did study economics and was never happy with everyone’s examples, scenarios or logic. Every scenario that was presented, when applied in real life, could be salvaged to some degree. If you have purchased too much inventory there is, for the most part, plenty you can do with it, when provided with resources to facilitate your decisions, even with perishables.

Lack of motivation or incentivisation is the real problem. Don’t be so anxious to roll over and let a couple grand, or a couple lives, wash down the drain for nothing more than the lack of (I hate to say it) turning lemons into lemonade.

Reply
Jonny permalink

April 27, 2011 7:30 pm

I would imagine this is the reason the “war on drugs” is still continuing.

Reply
Ann permalink

May 5, 2011 8:26 am

What an eye opener! Incredible stuff, really amazing. You’ve described it very clearly, thanks so much, it makes a lot of sense to me. I can totally see I have been acting this way more than I really care to admit. Seeing it here, for the first time, has opened a large window to my mind. This will most certainly help me understand, and change, my way of thinking to the very better.
Thank you so much, David.

Reply
Kris permalink

May 8, 2011 10:12 pm

This is so true, and I always do it and KNOW what I’m doing, hate myself for it a little bit, but CAN’T STOP.

For instance, earlier today I was eating a burger and fries, but I got full before I was finished. I’m sure everyone has been in this situation. But did I stop? Course not!

Ironically enough, I have no problem throwing out part of a sack lunch (or at least less of a problem). Maybe cause it’s cheaper?

But the example about ski trips… that’s iffy. In that case it’s more a matter of “All the money is spent already, choose the best option” vs. “You spent your money, you could waste it or not.” And really, it means I will be less hungry later, and thus eat less, and thus save, right? Despite the possible damage to my health.

Weird how much emphasis humans place on money. Sad is a better word actually…

Reply
Oxymoronity permalink

May 20, 2011 4:51 am

Thank you for giving me the final push I needed to quit Farmville.

Reply
aj permalink

June 1, 2011 1:14 am

irrelevant to thread:
it would seem to me that one of the few things that really does hold true in the human psyche is wherever theres an err theres an overly portentous prick ready to exploit said err to others of alike prick-ish-ness.

if you caught the irony in this,visit this site :http://blog.fnaggle.com/

Reply
Adam Caverhill permalink

June 14, 2011 10:45 am

‘a bag of tacos, and after the first bite you suspected they might have been filled with salsa-infused dog food, but you ate them anyway not wanting to waste both money and food?’

Is completely different actually, there’s a suggestable atlernative to the other scenarios, wasted money on concert tickets you don’ want to go to? You can do something else, there is no value in that concert. These terrible tacos still have value as food, even if they’re terrible. The objections are legitimate and not really fitting with sunk cost fallacy.

Otherwise great article : )

Reply
Kraps permalink

July 8, 2011 9:35 am

I’m not 100% sure but I think Steam sales (http://store.steampowered.com/) fall under this fallacy. Steam is know for monstrously insane sales where prices can be slashed up to 90%. How can someone pass up a top-level, triple-A game that’s 60% off, or a $10-$15 indie game that’s now $5 or less, regardless of whether you enjoy that type of game or have not played half of the 160+ games you already bought in previous sales?

Reply
Alok. D permalink

July 28, 2011 6:58 am

I’ve to say, this will be one of the few blogs I will thoroughly follow. It’s hard to encounter a truly subjective and informative article over the internet (at least for me, it has been). Thank you for such a wonderful post. I would have probably never thought about this, had it not been for your article. Now, that I have – it will be a good direction for me to have a reasonable doubt on some good and not-so-good things in my life.

Reply
discordianKnot permalink

August 31, 2011 2:52 pm

Would you say this phenomenon is a big part of what is being signified by the word “hope”? Take, for example, a parent of a child with a severe learning disability. The ideal mind might rather early on recognize that the prospects of changing the outcome for the child’s future are negligible compared the work and money and stress that the attempt will require . . . but who is going to want to look back on their life and admit they didn’t try their hardest to make things other that what they became? And all the moreso, the more engaged the emotions are – where one is doing it to live up to their parents’ expectations, to make a lover happy, for for the welfare of their children.

Is this is a fallacy, or is it the essence of being “human”?

Reply
Dorz Delos Reyes permalink

September 2, 2011 7:19 pm

Nice post! Get Farmville news here http://www.farmvillescoop.com

Reply
felipe1982 permalink

October 8, 2011 2:46 am

I laughed at pink tractor

Reply
ingwe permalink

November 3, 2011 5:56 pm

That’s why death is so tragic.

Reply
Andreas permalink

November 4, 2011 8:24 pm

While I agree with most in this article, I found your conclusion in the last paragraph a bit strange (that the desire to persevere is a fallacy). Without this “fallacy”, mankind would not be where it is today. If people would always act “rationally” in this sense and quit their endavours in the face of problems to minimize their losses, because they think that it is pointless to continue, none of the great accomplishments which require perseverence would have been made. Just think of Marco Polo. If he had put his many unsuccessful efforts or support for his voyage as “sunk costs” and rationally “abandoned” his plans, he would not have discovered America.
I understand that this is not the point of the sunk costs argument, but I think your last paragraph makes an unjustified extension to all acts of perseverence, which in my opinion cannot be based on the sunk costs argument.

Reply
- DiscordianKnot permalink
  
  November 4, 2011 9:24 pm
  
  I like the sentiment, Andreas, but Marco Polo didn’t discover America. He “discovered” that European trade routes were connected to China.
  
  Reply
Timothy Miller permalink

November 16, 2011 4:36 pm

Is the sunk cost fallacy a special case of loss aversion? Or is it the combined effect of several cognitive biases? I’m not certain.

I’ve also seen the sunk cost fallacy discussed as a problem in game theory. Sometimes called the “used car effect.” The more money you spend on repairing a used car, the less willing you are to sell it. Yet if you keep it, it needs more repairs, more expense, and the problem gets worse.

At the moment, it makes more sense to me to think of it as a cognitive bias, but I’m not certain.

Normally I think of game theory as an issue separate from cognitive bias. Maybe there’s some overlap I haven’t previously considered.

Reply
Julia permalink

November 19, 2011 12:30 am

I played Farmville like crazy and it was never, ever fun. You are spot on, as usual.

Reply
Gav permalink

November 20, 2011 11:28 am

While the other posts often leave me saddened about my mortal brain, this particular post makes me feel good about myself. I rationalize a great deal of my purchases well before handing over any cash. Subsequently, I don’t buy much stuff I don’t “need”, and was never conned into Farmville. Hooray!

Reply
joejoe permalink

November 24, 2011 11:37 am

Conclusions from the vacation thing are completely wrong and do not relate to the point. In fact, choosing 100$ so-so trip over 50$ good one now is completely rational. You can always go for the good one later, and it will likely still be cheaper.

Now, if definitely unique, concrete experiences that cannot be re-sold were suggested instead of vacations, the results would likely be different.

Reply
m jones permalink

November 27, 2011 10:17 pm

it’s funny but the whole investment idea is also seen as one of the biggest reason for people to cheat. the more a person has invested in a relationship, the less likely they are to cheat. they think of all the investment (time, feelings, etc) that they would lose if they cheated and it threatened their relationship so they are less likely to cheat if they see that they have a lot invested in the relationship.

Reply
Schueller37 permalink

December 13, 2011 12:43 pm

Though I am sure these situations come up often, I think there might be a counter argument to the sunk cost fallacy. I could consider myself an avid reader, but I bought Stephen King’s Bag of Bones a few years ago and I found the first 100 pages, or so, to be very difficult to, for lack of better words, “choke down”. It actually took me quite some time to push through those hundred pages (a few weeks at least), but once I got to a certain point the book became fantastic. If I were not to push through those first few chapters, I wouldn’t have found out how great the book actually was. I think what I wanted to get to was, does the idea of an unknown positive outweigh the known negative? Because if everyone were to realize a sunk cost and simply move on, would that not remove the drive for excellence?

On another note, I have played several different social games on Facebook and never really got attached to anyone of them. So I guess my question is, what would that say about my personality (if there is any relevance)?

PS. I love the blog and found the book to be very insightful.

Reply
A permalink

December 31, 2011 12:44 pm

I, for one, am glad that there is some ‘sunk cost’ fallacy within the human psyche. Why? Well, imagine a society without one! Imagine raising two young toddlers, one throws a screaming fit, the other gets violently ill and pukes all over your house. Someone without a pull towards ‘sunk costs’ would just run out of the house and never come back. OR, let’s say your partner gets cancer. Without a ‘sunk costs’ fallacy, one would just leave their partner rather than care for them when they need it most. Imagine the world in which we allowed the rational mind to dominate all the time, and I imagine a pretty heartless, cruel, violent, and lonely place. I feel this post romanticizes the rational and demonizes attachment and compassion, when in fact some sort of middle road is really where we should strive to be. Moreover, relationships, careers, and other life dilemmas are rarely so cut and dry as to have a simple, rational resolution. There are layers of conflicting feelings, reasons, and desires mingled together. Cutting and running doesn’t necessarily help the situation in the long run, although it may seem to be the easiest solution in the short run.

Reply
Chris permalink

January 20, 2012 3:44 pm

In regards to the $10 for a movie ticket scenario, I would argue that something you should consider is the distance traveled to the theater and the overall desire to see the movie in the theater.

Assuming this is a movie that I HAVE to see in the theater:
If the theater is less than a few minutes away, I would probably not buy the ticket, and would wait until I had budgeted for it again. However, if I have invested 30+ minutes driving there, and the gas to get there, I would get the ticket. Because the time traveled, cost of fuel, etc…Since I was going to see the movie anyways, those travel costs would have to be incurred again.

If the movie was something that I would be OK with seeing on DVD/Blueray, I would probably not buy the ticket regardless of distance traveled, cut my losses, and wait to see it on DVD/BR.

Reply
fast acting diet pills permalink

February 6, 2012 3:40 am

[...]the time to read or check out the subject material or internet sites we have linked to beneath the[...]

Reply
why wont my car start permalink

February 9, 2012 5:45 am

[...]always a massive fan of linking to bloggers that I enjoy but do not get lots of link enjoy from[...]

Reply
ahmet permalink

February 14, 2012 4:05 pm

great read. i am in a situation like this.

Reply

Trackbacks

Leave a Reply Cancel reply

Top Posts

SocialVibe

Follow “You Are Not So Smart”