Nobel Prize winner Daniel Kahneman is one of the founding fathers of behavioral finance. Although he holds a doctorate in psychology, not economics, he has had a profound effect on the dismal science. These days economic actors — that’s you and me — are not seen as rational, but rather human and prone to cognitive biases. This simple observation holds significant implications for the theory and practice of finance, ranging from the reliability of the efficient market hypothesis and the capital asset pricing model to listening to a company presentation at a sell-side conference, speaking with investor relations professionals, building financial models, determining when to buy or sell securities, and even deciding how to optimally organize an investment firm.
So wouldn’t it be nice to know what the good doctor knows? At the recent 65th CFA Institute Annual Conference, Kahneman distilled much of his research findings into bite-sized portions. What follows is a summary of his talk.
System 1 vs. System 2 Thinking
For starters, Kahneman identified two main ways that our minds process the world: System 1 and System 2. He calls System 1 thinking “fast thinking,” and it is characterized by snap assessments of situations, subconscious thinking, and thoughts processed in the brain’s amygdala. Think about how your mind works when listening to a speech, for example. The mind does not process every single word for its importance. Instead it has a laissez-faire quality to it. Kahneman believes that what we call intuitive thinking is an example of System 1 thinking. In contrast, Kahneman calls System 2 thinking “slow thinking.” This type of thinking is associated with the brain’s prefrontal cortex. It is characterized by deep analysis and is thus taxing. System 2 also monitors System 1 for when additional analysis is needed.
Kahneman provided an easy to understand example that highlights the differences: If he wrote “2 + 2 =,” your mind, without much effort, most likely responds “4.” This is System 1 thinking. Whereas, if he were to write “24 x 17 =,” your mind most likely switches into a very slow mode of operation, which is System 2. Our minds are still working at a System 1 level as we try to frame the problem. That is, we quickly know that the result of the multiplication problem is likely greater than 100 but less than 1,000. We also know that it is a multiplication problem and that we are going to need to call upon that part of our brain that can calculate the result. All of this is taxing and demanding.
Another important observation made by Kahneman is that our minds like associative thinking. That is, the brain likes to respond to causes. A cause will lead to the activation of a whole nebula of other ideas stored in, and understood by, the mind.
An example of this is “priming,” and to illustrate his point, Kahneman shared a humorous anecdote: One evening, his wife commented that a man with whom they had dined was sexy. After a few moments, his wife followed up by saying, “He doesn’t undress the maid himself.” Clearly, Kahneman was upset by this statement, having been primed by his wife’s first admission that she found the man attractive. But Kahneman had in fact misheard. His wife had actually said, “He doesn’t underestimate himself.”
Yet another example of priming was provided by the results of one of Kahneman’s experiments, in which people listened to sentences spoken by a man with an upper class English accent who then said, “I have tattoos all up and down my back.” This set of contexts registers in the brain as an incongruity; this is System 2 interrupting System 1. The sequence of cause and effect is set in motion because most people bring up the stereotype of aristocracy when they hear the upper class English accent, which is then upended by the reference to tattoos.
This provides us with further information about the characteristics of System 1. Namely, System 1 searches for causality. That is, it interprets the world and it creates coherent interpretations of the current situation and context. For example, the sentence “She approached the bank . . . ” is ambiguous, but the brain does not register ambiguity; instead, it picks an interpretation of the incomplete sentence and runs with it.
Another classic example is the following sentence that helps illustrate the brain’s preference for association: “How many animals did Moses take into the Ark?” Do you understand the problem with that sentence? If you did not, it was Noah, not Moses, who took animals into an ark. This works, in part, because the System 1 brain hears Moses, which activates the “biblical information” nebula in our minds, priming us for other biblical information. The problem would not exist if, instead of Moses, the protagonist were changed to, say, Bernice.
Despite the many examples of our minds having two components, System 1 (subconscious) and System 2 (conscious), most of us subjectively believe that we are System 2 thinkers. And, in fact, System 2 spends a lot of time covering up much of System 1’s thinking, according to Kahneman.
One implication is that two people experiencing the same thing most often interpret it in two different ways. If you want a penetrating way of identifying how someone thinks about an issue you should ask them: What was your first association? That is, what was your first interpretation? Kahneman said that if you know this then you can explain many of the phenomena uncovered by behavioral finance researchers.
People are Overconfident
How does the associative system generate feelings of confidence? The feeling of confidence corresponds to the coherence of the story that is generated in the associative system (i.e., System 1). In other words, the judgment and subsequent feelings of confidence are a response to the quality (coherence) of the story and not to the quantity of data contained in the story. This combination of factors, confidence in the presence of data to the contrary, leads to overconfidence.
Kahneman said day traders churn because they tell themselves coherent stories. But even Fortune 500 CFOs tell themselves stories that are much more coherent than the facts justify. And, in fact, CFOs at Fortune 500 firms are more overconfident than CFOs at smaller firms. Kahneman pointed out that men are much more overconfident than women (which reminds me of the old adage: If you want something said, ask a man, but if you want something done, ask a woman).
Our Minds Seek Coherence
Another deleterious effect of System 1 is that it tends to generate answers in situations in which there is an absence of information. Put differently, we are not often stumped by a question or situation. When we are later asked to justify the answer we came up with, we provide all kinds of answers.
On average people know very little about the stock market and yet they feel that they know a lot about it. Within the investment profession itself, it is very clear to Kahneman that people feel that they deliver alpha even if they do not — this is overconfidence at work.
Further proof of System 1 looking for coherence is that people often answer questions that they were not asked. Often there is an intuitive answer, but it interferes with the story you have told yourself. This is evidence of the friction between Systems 1 and 2.
What Can Be Done With Behavioral Finance?
So what can be done by knowing the many tenets of behavioral finance? “Very little; I have 40 years of experience with this, and I still commit these errors. Knowing the errors is not the recipe to avoiding them,” Kahneman said. In fact, he feels that organizations can improve the quality of thinking, but that individuals cannot do as much.
There are, however, several things individuals can do. Kahneman said there are situations in which we are more likely to make a mistake. So, when you encounter these situations, you want to slow yourself down.
One example is when numbers are mentioned, such as in a negotiation. In this situation what happens is a form of priming, known as anchoring, where any number that is mentioned tends to become plausible just because it was mentioned. Kahneman recommended that if the other side in a negotiation comes up with a number that is out of your range, make a big loud dramatic scene because this disrupts the priming effects of System 1 for both sides in a deal. If you are not in a negotiation but want to do better around anchoring, ask a lot of questions about the numbers and ask people to justify their conclusions. A final tip to help individuals improve their decision making: Rather than ranking options, look at the individual issues involved one at a time, and then consider all options only at the end, right before making a decision. This helps to ensure that System 2 is involved in the decision making.
For organizations, Kahneman recommended better quality control over decision making within firms. For example, to conduct better meetings, ask all of the participants to write down their opinions on a slip of paper before the issues are discussed. This helps to disrupt the priming effect set in motion when the first person to speak in a meeting ends up determining the flow of the discussion.
Organizations can also overcome the overconfidence bias by using a technique known as “conducting a premortem.” In a situation in which an important decision has been made but not yet implemented, get a group of people together that are knowledgeable and conduct a one hour conversation with them by saying: “Well, we implemented the plan, and it’s now one year later, and it was a disaster. Now you have a piece of paper, and I want you to tell the story of the disaster.” In other words, Kahneman advised that you have to legitimize dissent to improve decision making in an organization.
In conclusion, you cannot take the human element out of human decisions or institutions. But an understanding of how the human mind’s two systems work can help all of us make better investment decisions and build better investment organizations.