The economic effect of “sticking” in nine experiments

People often consciously choose the worst possible outcome for themselves. They are guided not by rational economic principles or other non-monetary factors, but by what is known in economics as the “sunk cost effect.”

Author:Nikolai Borisovich Pydyk,Department of Corporate Governance and Finance of the Higher School of International Business at the Academy of National Economy under the Government of the Russian Federation.

A guy wins a free ticket to a football match for his favorite team in a lottery. He doesn’t want to go to the match alone, so he convinces a friend to buy a similar ticket. Just as they are about to head to the match, a terrible storm breaks out. The lottery winner looks out the window and says, “In this weather, I could just get washed away from the stands. Forget about the match. It’s better if we stay home.”

Or maybe (especially if he has an economics background), he says, “The costs I will incur as a result of the storm far outweigh the enjoyment I expect from the match. To hell with this match. It’s better to stay home.” Regardless, the lottery winner refuses the prize he has received. And what about his friend? He’s just furious! He demands that they go to the match immediately: “I don’t want to lose the $12 I paid for the ticket.”

From the perspective of rational economic behavior, the friend is behaving somewhat strangely. After all, if the costs of attending the match exceed the benefits, it would be in his best interest to refrain from going. Only the marginal costs should influence the final decision about attending the match, not the sunk costs. However, from the friend’s point of view, he is in deep trouble having paid $12. This $12 is referred to asthe costs of sticking (sunk cost), and my friend’s behavior —the effect of sunk costs (sunk cost effect)Текст для перевода: ..

Examples of sunk costs are found in abundance across various fields. For instance, at the end of 1981, a special committee of the U.S. Congress was trying to determine whether to continue subsidizing the construction of the Tennessee-Tombigbee Waterway project.(Tennessee—Tombigbee Waterway Project), which required significant funds for its completion. Although the project turned out to be so costly that it should have been immediately terminated, Congress decided to allocate additional subsidies. Here are some of the arguments that the senators relied on.

“The liquidation of a project that has already seen an investment of $1.1 billion would be an unprecedented case of taxpayer money being squandered,” said Senator Denton during the hearings.

“The completion of the Tennessee-Tombigbee project will not be a waste of taxpayers’ money. However, abandoning the project at its final stage would be a senseless waste of already invested funds,” said Senator Sasser.

A similar story happened with the Soviet BAM (Baikal-Amur Mainline). Just a few years after construction began, BAM turned into an investment project that was more profitable to liquidate immediately rather than continue. Nevertheless, construction continued, and it remains unfinished to this day.

The controllability of sticking costs can be understood from the following example. When discussing why nuclear energy will prevail in the future, a senior official from the U.S. Department of Energy explained it in simple terms:

“In the end, no sane person would dare to shut down a project worth $2.5 billion. As a result, no administration would risk closing a national nuclear energy program that has already seen an investment of $200 billion. The trick in our industry is to lay down as many nuclear power plants as possible and do it in a way that environmental activists remain unaware. By the time they catch wind of it and try to revoke our license, we will already have millions of tons of steel in the ground, and no one in their right mind would dare to stop us.”

The official’s last statement makes sense only if we understand “common sense” as the usual way people think. In this case, such thinking is irrational regardless of how convincing the arguments in its favor may seem. Let’s reiterate once more: current decisions should not depend on the costs of sticking to them.

Alongside the monetary costs of sticking, one can encounter numerous non-monetary costs of sticking in practice. Let’s clarify this with examples:

• Should I continue to invest in these unfulfilled relationships? I’ve already put so much into them.
• Should I stay at this terrible job? I spent a whole year to get this position.
• Many bad movies end up being watched to the end simply because, when a person realizes how bad the film is and should leave the theater, they feel influenced by the time and money they’ve already spent. These minutes and money (the sunk costs) compel them to wait for the end of the movie, hoping it will be better than the beginning.
• Many military conflicts persist simply because one side refuses to engage in negotiations and demands the continuation of hostilities until complete victory is achieved. The need for total victory is justified by the fact that too many sacrifices have already been made for it. It’s a curious logic.

Let’s demonstrate some of the costs of involvement. The research was conducted among college students. None of them answered more than one question. The number of respondents is provided below immediately after the question wording.

The sticking effect: Experiment 1

You spent $100 on a ticket that allows you to spend your weekend at a ski resort in Michigan. A few weeks later, you bought a similar ticket for $50, which is only different from the one you already have in that it specifies a ski resort in Wisconsin.

You believe you would have much more fun on a weekend in Wisconsin than in Michigan. But when you go to put your newly purchased ticket in your wallet, you discover that both tickets are for the same weekend. It’s too late to sell or return the tickets you bought. You have to choose which ticket to use. Which ticket would you choose?

• A $100 bill in Michigan — 33 people.
• A $50 ticket to Wisconsin — 28 people.

According to the principles of classical economics, when making a decision, an individual should consider only the actual costs and benefits expected from each alternative.

Guided by this principle, one would expect that every person would choose the trip that brings them the most pleasure—the trip to Wisconsin. However, only 46% of all respondents chose this trip. This means one thing: the postulates of classical economics, which suggest that 100% of individuals would prefer a trip to Wisconsin over a trip to Michigan, are proven false. Clearly, the costs of getting involved played a significant role here.

The costs of sticking: experiment 2

Experiment 1 was conducted using a simple survey, and of course, no real money was involved. Although many studies have found that the results of survey-based experiments closely mirror those of experiments involving real money, it is advisable to verify these results in an experiment that includes real monetary stakes.

For the experiment, 60 people were selected who approached the box office at the Ohio University theater to purchase an annual subscription for all the theater’s performances. After expressing their desire to buy a subscription, the cashier sold them one of three types of tickets, which were randomly chosen from the drawer. The first type of ticket cost $15 (its regular price), the second was $13 (a $2 discount), and the third was $8 (a $7 discount). The cashier explained the low price as part of the theater’s marketing policy.

It turned out that people who bought tickets at their regular price attended many more theater performances over the year compared to those who were lucky enough to purchase discounted tickets.

The sticking effect: experiment 3

Situation 3A.As the president of a major aerospace corporation, you have decided to invest $10 million in an R&D program aimed at creating a stealth aircraft. When your project was already 90% complete, a competitor unexpectedly entered the market and began selling similar aircraft equipped with radar-evading systems. To make matters worse, it turns out that the competitor’s aircraft are significantly faster and more economical than yours. Should you invest the remaining 10% of the funds to complete the stealth aircraft project?

• Yes — 41.
• No — 7.

Situation 3B.As the president of an aviation company, you receive the following proposal from one of your employees: to use the last million of your company’s research budget on a stealth aircraft development program. At the same time, you are aware that a competitor has already started their own program to develop a similar aircraft, which appears to be faster and more economical than the one you could build with the available resources. Should you invest the last million of your research budget in creating a stealth aircraft?

• Yes — 10.
• No — 50.

The only significant difference between these two situations is that in situation 3A, millions had already been invested, while in situation 3B, no investments had been made. If in situation 3B the majority of respondents considered the project a bad idea, in situation 3A the same respondents insisted on its completion.

It is clear that there are no reasonable reasons for ending the project. The only unreasonable reason that could explain such behavior is the costs of getting stuck.

The costs of sticking: experiment 4

Situations 4A and 4B are identical to situations 3A and 3B, except that at the end of each situation, a small note has been added: “Use a scale from 0 to 100, where 0 = no chance and 100 = definitely. Indicate a number on this scale that reflects your assessment of the aircraft’s actual financial success.”

The average score of 76 respondents in situation 4A was 41, while the average score of 82 respondents in situation 4B was 34.

These results lead us to conclude that when sunk costs are involved in a situation, individuals begin to reassess the likelihood that the airplane project will be successful.

It’s hard to say with certainty whether this revaluation is a catalyst for the decision to continue investing in the project or, on the contrary, a consequence of the decision to invest.

A possible flaw in experiments 3 and 4 may be that in situations 3A and 4A, the cost of creating the airplane is $10 million, while in situations 3B and 4B, it costs only $1 million. Respondents in situation 3B may have been reluctant to invest in the project due to their intuition that an airplane priced at only $1 million cannot be of good quality. ALSO WATCH:Competitive analysis of the industry and key success factors

In situation 3A, the respondents had no reasons for such concern, and therefore they were willing to invest an additional $1 million to complete the project. We will test this hypothesis in experiment 5.

The sticking effect: experiment 5

The situation used in this experiment is identical to situation 3B, except that the amount of 1 million dollars was replaced with 10 million dollars. The respondents’ decision regarding the construction of the airplane was as follows:

• Yes — 10.
• No — 50.

The data used in experiment 5 is identical to the data used in situation 3B. Therefore, it can be concluded that in situation 3B, the decision not to build the airplane was not influenced by the relatively low price of the airplane compared to its price in situation 3A.

The different responses to the questions in situations 3A and 3B are a result of the fact that when answering the question in situation 3A, individuals were influenced by the costs of getting stuck.

The Costs of Sticking: Experiment 6

Let’s consider another situation. On your way home, you stop by a store and buy a microwave dinner for just $3, while its regular price is $5. A few hours later, you decide it’s time for dinner and are ready to put the purchased meal in the microwave. Suddenly, a great idea pops into your head: invite a friend over to have dinner together and watch a good movie.

Your friend agrees, and you rush to that very store to buy a second dinner. But when you get there, you find out that the sale on $3 dinners has ended, and the price is now the regular $5. So, you have to buy the dinner for $5. When you get home, you toss both dinners into the microwave. Suddenly, just as both dinners are almost ready, the phone rings. It’s your friend: unfortunately, he won’t be able to come. You’re not hungry enough to eat both dinners, but you can’t freeze one of them anymore. You’ll have to eat one dinner and throw the other away. Which dinner will you eat?

• For 3 dollars — 2.
• For $5 — 21.
• Indifferent — 66.

Since the benefits and costs of consuming the $3 and $5 dinners are equal, we would expect (according to the axioms of traditional economics) that any individual would be indifferent between the two. However, the costs of sticking with the $3 dinner increase the benefits of choosing the $5 dinner. Since the reasons that led some individuals to choose the $3 dinner are based on factors unknown to us, it would be reasonable not to consider these respondents. Of the remaining 87 people, 76% chose “indifferent.” Such results suggest that the predictions of traditional economic theory, which state that 100% of individuals should be indifferent between both dinners, are incorrect.

The choice of a large number of individuals opting for the $5 dinner (compared to those who chose the $3 dinner) can be explained by the sunk cost effect. People feel that since they have spent more money on the $5 dinner, the most economical decision is to eat it. However, it is clear that this is an entirely nonsensical act driven by the psychological impact of sunk costs. It’s a kind of false economy.

The sticking effect: experiment 7

Situation 7A.As the owner of a publishing house, you need to make a decision about modernizing your production. You can either purchase a new printing press for $200,000 or use the same amount to upgrade your fleet with new trucks. You choose the new trucks, which will be able to deliver your products twice as fast as the old ones, while maintaining the same costs.

A week after you purchased the trucks, rumors start circulating that one of your competitors is facing serious financial difficulties. In an attempt to raise some cash, this competitor approaches you with a request to buy his computerized printing press for just $10,000.

Compared to your old machine, this one will operate 50% faster and with half the maintenance costs. You know you won’t be able to sell your old machine to raise money for the new one. The old machine was custom-built for you and has unique features that cannot be modified. Your own savings amount to exactly $10,000.

Should you buy a computerized printing press from your bankrupt competitor?

• Yes — 49.
• No — 15.

Situation 7B.As the owner of a publishing house, you need to make a decision about modernizing your production. You can either purchase a new printing press for $200,000 or use the same amount to expand your fleet with new trucks. You choose the new printing press, which will operate twice as fast as the old one while maintaining the same costs.

A week after you purchased a new machine, rumors start circulating that one of your competitors is facing serious financial difficulties. In an attempt to raise some cash, this competitor approaches you with a request to buy his computerized printing press for just $10,000.

Compared to your new machine, this one will operate 50% faster and with half the maintenance costs. You know you won’t be able to sell your old machine to raise money for the new one. The old machine was custom-built for you and has unique features that cannot be modified. Your own savings amount to exactly $10,000.

Should you buy a computerized printing press from your bankrupt competitor?

• Yes — 43.
• No — 38.

Although acquiring a new printing press will lead to the same proportional increase in productivity in both situation 7A and situation 7B, a significantly larger number of individuals wanted to purchase the press in situation 7A. When respondents were asked to explain their choice, those in situation 7B who decided not to buy the press cited the following reason: “I already have a new, good press that cost me a fortune.”

It turns out that the respondents had significantly fewer motivations to buy a new machine if they felt that this purchase was just a repeat of one they had recently made. By opting not to buy a new machine, the respondents believed they were saving their money in this way.

This behavior has an interesting application: if the situation involves the respondent’s own money or if they bear personal responsibility for the investments, the impact of sunk costs will be much more significant than in situations where their own funds are not involved or where there is no personal responsibility. We will confirm this in Experiment 8.

The Costs of Sticking: Experiment 8

Situation 8A.The aerospace company ASME Airline Company has decided to invest $10 million in its R&D program.

The goal of the program is to create a stealth aircraft. Just when the project was 90% complete, a competing company unexpectedly entered the market and began selling similar aircraft equipped with radar-evading systems. Moreover, it turns out that the competitor’s planes are much faster and more economical than the aircraft that ACME is developing.

Should the aerospace company invest the remaining 10% of funds to complete the stealth aircraft project?

• Yes — 37.
• No — 21.

Situation 8B.The ACME Airline Company received a proposal from one of its employees: to use the last million of the company’s research budget for a stealth aircraft development program. At the same time, it became known that a competing company has already started its own program to develop a similar aircraft, and it seems that their plane will be faster and more economical than the one ACME could build with its available resources.

Should ASME invest its last million of research budget in creating a stealth aircraft?

• Yes — 2.
• No — 35.

The answers to the questions in situations 8A and 3A differ significantly from each other. When individuals made the decision to build the airplane on behalf of the corporation’s president, they were much more influenced by the sunk cost fallacy than when they made the decision from the perspective of an outside observer. However, even when the decision is made from the viewpoint of an outside observer, the effect of sunk costs remains highly significant.

In Experiment 9, we will slightly manipulate the personal involvement of the individual. In Experiment 1, the respondent paid for tickets to Michigan and Wisconsin. Will the endowment effect disappear if the individual receives the tickets for free?

The sticking effect: experiment 9

On a sunny morning, your phone rings, and the host of a local radio station informs you that you’ve won a free ticket to spend your weekend at a ski resort in Michigan or Wisconsin. You believe you’ll have a much better time in Wisconsin than in Michigan. Still, you call the travel agency to find out the cost of each ticket. It turns out that the ticket to Michigan costs $100, while the ticket to Wisconsin is $50. You need to make a choice.

Which ticket would you choose?

• A $100 bill in Michigan — 44.
• A $50 ticket to Wisconsin costs 42.

It seems that the sunk cost effect persists even when the subject isn’t spending any money.