Have you been in a hopeless relationship? Struggling to go out to an event in bad weather, just because you've already bought a ticket with your hard-earned money? These are all examples of sunk cost effects, where a resource has already been invested (irrevocable) and is chosen to hold on. Decision-makers must remember that strategic decisions that are affected by sunk costs can have serious consequences.
Many notably high-risk decisions are due to sunk cost effects. For example, the decline of General Motors at the end of the last century is said to be due in part to management's reluctance to abandon a once-successful strategy. In the aerospace industry, Britain and France** invested heavily in the Concorde program, often seen as a continued waste of money after failure (the sunk cost effect is sometimes referred to as the Concorde fallacy). And in the political sphere, such as the protracted U.S. military operations in Vietnam and Iraq, illustrate that this effect is not only costly, but also costs tens of thousands of lives.
Many business economics and decision making courses emphasize that when making decisions about what's next, don't think about all the irrevocable sunk costs of the past. Decision-makers must remember:If strategic decisions are influenced by sunk costs, the consequences can be severe.
Evaluate the team
Whether it is susceptible to sunk cost effects
Researchers measure this effect by providing a variety of hypothetical scenarios and asking participants what they would do. But the scenario assumptions in experiments often do not cover all possible sunk costs (e.g., money, time, energy, emotion, etc.), and it is impossible to verify whether the answers to virtual scenarios can ** people's susceptibility to sunk cost effects in situations involving real-world benefits.
My recent study, "Sunk Cost Effect Assessment," with Daniel Sgroi and Anthony Tuckwell, addresses these issues and provides a new set of 8-question scales assessing susceptibility to sunk cost effects. Each scenario is a real-life everyday situation that allows everyone to put themselves in their shoes. All scenarios assume a variety of possible sunk costs. In most cases, there is more than one sunk cost, because our resources are generally interconnected. For example, many important decisions require more personal energy and emotions in addition to obvious and measurable costs such as time and money, which may be felt differently by different people. As a result, each scenario corresponds to only one class of resources, or only provides a scenario for a certain type of resource, which does not seem to meet the demand. Highlight the combination of different resources in a what-if scenario.
How do we do it?
The first set of questionnaires we asked participants to answer consisted of 18 hypothetical questions, covering a total of five different resources (energy, time, money, emotion, and confidence) from a variety of sources. For example, in the following scenario, the emphasis is on energy and time resources (keywords are underlined here for illustration):
You've been looking forward to this year's Halloween party. You've got the right cape, the right wig, and the right hat. You've spent a whole week cutting lots and lots of tiny stars out of paper to stick to your cape and hat, trying to make your look perfect. However, on Halloween, you think it would look better without these painstaking stars.
What would you do? Describe your practice with the scale below.
Based on the analysis of the questionnaire results, we selected 8 of the 18 situational questions to form the "SCE-8" scale. The resources involved in each story are listed in the Assess Sunk Costs chart. We analysed and identified which scenarios best described the changes in the data. It is important to note that the final selection of eight scenarios covers all the resources we have in mind, validating the idea that multiple resources should be considered when trying to assess sunk cost effects.
Answering each question is scored on a scale of 0 to 5, with 0 being the least vulnerable and 5 being the most vulnerable. The scores for all 8 questions are summed to give a total score of susceptibility (between 0 and 40). Looking at the data, susceptibility scores range widely, from 0 to 40, but the average score is below 10.
As shown in the chart, each of the eight scenarios assesses the extent to which participants are affected by the sunk cost effect, and each scenario targets at least one type of cost – energy, time, money, emotion, confidence. In scenarios where multiple costs are evaluated, there is often a category of costs that are particularly important to the decision, represented by black dots in the diagram.
The higher the score, the worse the decision?
To test whether the scores on the above scale can perform in an environment with real consequences, we asked participants to continue to participate in an experiment in which a monetary reward may be awarded. A group of participants can earn "assets" (a lottery ticket with a 10% chance of winning $10) by completing a heavy task that requires an investment of time and effort, among other costs. The other group of participants does not have to complete the task and only needs to choose between the two assets.
We asked the first group to choose whether to do the assets they earned by doing the task themselves, or to exchange them for better assets (with a 20% chance of winning $10). 2 out of 3 people chose the former. The second group all chose better assets. In other words, the majority of the first group insisted on choosing worse assets because they had invested time and effort into them, indicating that they were vulnerable to sunk cost effects. Not only did we develop a susceptibility scale, but for the first time, we found strong evidence of sunk cost effects in an experimental setting with material incentives.
Now we can see if there is a link between the susceptibility score and the actual susceptibility. We found that people with susceptibility scores above average (10 and above) were nearly three times more likely to be affected when it came to actual monetary benefits than those who scored below average (9 and below).
Experience and wisdom?
Our study also reveals elements of susceptibility to sunk cost effects. We had participants complete several psychometric tests for various cognitive abilities and found that experience or knowledge ("fixed intelligence") was more effective than raw computing ("fluid intelligence") in avoiding the sunk cost effect: in other words, intelligence was more important than intelligence.
This conclusion is in line with the general theory:To avoid decision bias, it's important to understand that you should restrain your instincts in the current situation. In contrast, once you realize that you are facing a sunk cost problem, you don't need to think much to avoid its effects.
However, if large enterprises and experienced leaders at the top are affected by sunk cost effects, even if experience works, it will not be enough to overcome this significant bias. Experience can help you avoid some pitfalls wisely, but no matter how experienced, consider the SCE-8 scenario assumptions and think about what you would do, assess your vulnerability, and use them as a reference to make better decisions in the future.
David Ronayne |Wen.
Shuoma |Translated by Zhou Qiang |Redaction.
True leadership is equal to "affinity + ability".
At the end of the year, it is said that those who are ready to resign will have these behaviors.