Source: Smart Investors
Near the end of an in-depth interview in 2019, Daniel Kahneman's phone rang. He had forgotten to put it on silent.
The caller wanted him to give a speech or review a book. Kahneman answered, listened for a while, and before hanging up said, "My rule is never to say 'yes' over the phone. I'll get back to you tomorrow."
Shane Parrish, who was conducting the interview, quietly noted the moment. Afterward, he followed up with a question.
Kahneman responded that he knew his intuition would often lead him to hurriedly agree when faced with a sincere invitation. He had tried rejecting them outright, but every time he did, the other party's immediate response was always to renegotiate: What about a different date? Or a revised timeline?
So, he set this rule for himself: to leave a 24-hour buffer period before making any decisions.
This rule encapsulates a lifetime of Kahneman's research.
As the 2002 Nobel laureate in economics, the author of *Thinking, Fast and Slow*, and someone who spent half a century studying human decision-making biases, influencing countless top investors, Daniel Kahneman candidly admitted that despite a lifetime of research, he was no better than others at avoiding bias.
Relying solely on personal willpower to make better decisions is not very realistic.
What he truly believed in were processes, algorithms, structured scoring systems, and delaying intuition—any design that could postpone the rushed 'yes.'
Kahneman passed away on March 27, 2024, at the age of 90. This conversation was one of the few in-depth interviews he gave during his lifetime.
As usual, we delve into history and organize it. People in different stages and states of life will resonate differently.
For instance, Kahneman says that when we see someone do something 'illogical,' we tend to attribute the reason to their personality rather than the context they are in.
This insight is very useful for understanding the situation of fund managers. When we look at Buffett's sixty years of investment success, we often overlook his resolution of a key issue — placing himself in a completely self-consistent scenario through long-term insurance float.
The matching of liabilities essentially reduces those coercive factors that distort the decision-making of fund managers.
The section about agents is also intriguing. Agents are free from the endowment effect in buying and selling because they do not hold a sense of loss. Brian Johnson’s statement was adopted by Kahneman: 'I wanted to prove that algorithms take better care of me than I do myself.'
This could be considered the psychological basis for buying funds, especially quantitative funds.
Regarding the source of beliefs, Kahneman is candid, stating that people believe in things mostly not because of evidence but because those we trust share the same view. He admitted that his perspective on climate change was formed this way.
In 'Thinking, Fast and Slow,' there is a more complete version of this statement: Many important beliefs do not stem from evidence but from the shared opinions of those we love and trust.
Returning to intuition itself, Kahneman says that if you truly want to significantly improve your decision-making level, it comes down to one word — algorithms. If algorithms are unattainable, then slow yourself down.
When Sean asked him how to 'postpone intuition,' Kahneman provided a concrete method: when evaluating an investment project, instead of immediately asking 'Do you support or oppose it?' go over each dimension individually, reach a certain consensus on each dimension, and then make an overall judgment. Once an overall impression forms, much time will be spent validating that impression rather than seeking real evidence.
This brings to mind Buffett's 2015 conversation with Cunningham from Arizona State University.
Buffett said, "In a way, I consider myself a journalist. I assign myself topics, such as in 1973, whether The Washington Post Company was worth $22 per share. That was my working hypothesis, and then I went out to find the facts without making judgments or choices based on preconceived positions. He emphasized not jumping to conclusions and letting assumptions dictate the content of the report itself.
There were quite a few such insights in this dialogue.
The Intelligent Investor selected and meticulously translated the most insightful portions, adding some annotations.
1. On the distinction between 'happiness' and 'life satisfaction'
Sean: You made an interesting distinction between 'happiness' and 'life satisfaction.' Could you elaborate on how you see it?
Daniel: Of course. In fact, the term 'happiness' itself is quite vague, meaning different things to different people.
But one relatively reasonable interpretation is that it refers more to emotions, to how you feel at this moment, to the emotional tone of your life. Are you comfortable? Are you enjoying the present? Are you living as your true self?
On the other hand, 'life satisfaction' is an entirely different matter. It’s the evaluation you make about your overall life situation when reflecting on your life. Most of the time, we don’t think about these questions; we simply live day by day. But occasionally, you pause, look back over the past few years, and ask yourself, 'Am I satisfied with my life?'
That’s what we call life satisfaction.
Here, 'satisfaction' does not simply mean 'not bad,' but rather a holistic contentment with your entire life state.
Sean: So how do we balance the two? Should we pursue happiness when we are young and seek satisfaction when we are older?
Daniel: I have never thought about it that way.
When I first started researching this topic, I leaned more towards 'happiness.' I believed that immediate emotional experiences were the most authentic, and life satisfaction was just a story people tell themselves. So initially, I thought happiness was what truly mattered.
However, as we conducted further research, we found that the conditions that make people happy are not the same as those that lead to satisfaction.
Happiness is often social in nature: whether you are with people you love and who love you back, and whether you have connections and relationships. Life satisfaction, on the other hand, is more traditional and worldly, being more closely related to factors such as career success, income, education level, and social status.
Sean: Did your research also reveal this finding: after annual income exceeds $70,000, people’s happiness does not significantly increase, but their life satisfaction continues to improve?
Daniel: Yes, that was research I conducted with Angus Deaton from Princeton.
We found that for 'happiness,' which refers to the overall tone of daily emotions (including both positive and negative feelings), higher income does not endlessly increase one’s happiness. However, poverty significantly increases suffering.
In the United States, there seems to be a threshold, which we estimated at the time to be around an annual income of $70,000. Beyond this level, additional income does not noticeably enhance your emotional well-being.
However, 'life satisfaction' is entirely different. There is no ceiling to life satisfaction; it does not stop rising after reaching a certain income level. Generally speaking, the higher the income, the higher people's subjective evaluation of their own lives.
In many cases, money actually serves as a 'medium of success,' symbolizing your position within the social system and representing a subjective sense of achievement.
Sean: Do happier or more satisfied people live longer?
Daniel: We have very clear evidence that being happier is indeed beneficial to physical and mental health, enabling you to live longer and enjoy a higher quality of life. Life satisfaction also plays a similar role.
However, it is difficult for me to make an exact comparison between the two at this point.
(Note: In the 2010 study by Kahneman and Deaton, they used Gallup data to separately measure 'emotional well-being' and 'life evaluation,' providing a very clear conclusion: high income continuously raises life evaluation, but emotional well-being in the U.S. sample tends to plateau at an annual income of approximately $75,000; meanwhile, low income significantly amplifies pain and negative experiences. This conclusion later became the basis for many discussions about how 'money cannot buy happiness.')
However, this '75,000-dollar threshold' has since sparked an interesting academic debate. In Matthew A. Killingsworth's 2021 study, large-scale experience sampling data revealed that both 'experienced well-being' and 'evaluative well-being' continue to improve with rising income and do not plateau after $75,000.)
2. Make good behaviors easier and bad behaviors harder
Sean: Can we change behavior? If so, how?
Daniel: I am not entirely sure about this premise. I believe changing behavior is extremely difficult.
Of course, there are some techniques and methods worth trying. However, anyone overly optimistic about 'behavioral change' is, to some extent, indulging in self-deception. The matter is far more complex than most people realize.
Sean: Isn't this the entire meaning of marriage?
Daniel: Perhaps. But if both partners in a marriage constantly try to change each other, I believe the relationship is unlikely to lead to happiness.
Sean: So lowering expectations might actually bring more happiness?
Daniel: Exactly. Even if you have expectations, don't think about 'changing' others because it's highly unlikely to succeed.
Sean: The strategy I can think of is to make good behaviors easier and bad behaviors harder to perform.
Daniel: That is precisely the core idea. I've always believed that this is one of the most important insights psychology has offered so far.
If you want someone to shift from behavior A to behavior B, there seem to be two approaches on the surface: you can push them, or you can ask, 'Why haven’t you done this yet?'
When you ask, 'Why haven’t you done this yet?' what you encounter is what Kurt Lewin referred to as 'resistance.' There are reasons why people have not yet reached the state you desire for them.
Lewin viewed behavior as a state of 'equilibrium.' Some forces drive you forward, while others hold you back. For instance, your speaking volume or driving speed—these are behavioral outcomes shaped by external forces in balance.
What is our habitual approach? When we see someone hasn't changed, we try to exert pressure, which means increasing the driving force. However, Kurt Lewin's insight was that what you should really weaken is the resistance, rather than intensifying the driving force.
I always have a picture in my mind: a wooden board pulled by springs on both sides. If you want it to move in a certain direction, there are two methods. One is to add another spring to push it in that direction; the other is to remove one of the springs that is pulling it back.
If the board moves due to increased driving force, then in the new equilibrium, the springs will be compressed more tightly — resulting in greater system tension and stronger counterforce.
However, if you remove the resistance, after removal, the entire system becomes more relaxed with less tension.
I first read this theory when I was twenty, and I thought it was absolutely brilliant.
(Note: This idea is also very similar to Aesop's fable 'The North Wind and the Sun': The harder the north wind blew, the tighter the traveler wrapped his coat; the sun shone warmly and gently, and the traveler took off his coat willingly. The moral is 'Force often provokes resistance, while gentleness is more likely to bring about change.')
3. Behavior is situational: To understand a person, first look at the state they are in.
Sean: What psychological common sense do you hope everyone understands?
Daniel: Behavior does not necessarily reflect a person’s character; behavior is largely related to the situation.
If people act in a way that seems strange, you need to observe the situation they are in and identify the pressures within that situation that lead them to behave that way.
A well-known social psychologist referred to this bias as the 'fundamental attribution error.' It means that when you see someone act in a certain way, you tend to assume it is due to their personality. However, this may not necessarily be the case; it is more likely that the state they are in caused them to behave that way.
I also hope people understand that motivation is complex. When people do 'good things,' their motives are often a mix of good and bad intentions; similarly, when they do 'bad things,' their motives may also involve a combination of both positive and negative elements.
One of the significances of psychological education is to reduce judgment and increase empathy and patience. This is because passing judgment does not lead you anywhere.
The term 'fundamental attribution error' is commonly associated with the classic review article by social psychologist Lee Ross, in which he discussed people's tendency to overemphasize 'personality-based explanations' while underestimating 'situation-based explanations' when interpreting others' behavior.
(Note: From this perspective, we can understand why the liability side is crucial for investment managers. Matching funds allow fund managers to operate according to their investment philosophy and methods. However, the entry of short-term or return-expectation-mismatched clients could significantly affect or even constrain the fund manager's investment decisions.)
(We see Buffett’s 60 years of investment success but overlook his resolution of key issues, such as using long-term insurance capital to place himself in a scenario fully consistent with his investment framework.)
4. Allowing an 'agent' to make decisions on your behalf may result in more rational outcomes.
Sean: Why can we easily offer advice to friends, yet when we find ourselves in similar situations, we often fail to see clearly?
Daniel: Emotions interfere with clear thinking. For example, consider the endowment effect: when I try to sell you my sandwich, I typically demand a price higher than what I paid for it. The core reason behind this is that our perception of 'loss' is usually much stronger than our perception of 'gain.'
However, a very interesting finding is that if an 'agent' makes decisions on behalf of someone else, they do not exhibit this loss aversion. In other words, they tend to use the same price for buying and selling, which is closer to rationality in the economic sense.
This phenomenon is particularly crucial in the field of public policy.
The government, in essence, acts as an agent making decisions on behalf of society as a whole. As agents, they tend to focus on overall efficiency and the net outcome of policies. However, the issue lies in their frequent neglect of a reality: any reform may cause some individuals to 'lose their existing benefits.'
Experience has shown that 'potential losers' are often more actively opposed than 'potential winners.' This is also why many reforms either fail prematurely or, once implemented, incur costs far exceeding expectations. Because you must compensate the affected groups, and these costs are often underestimated in the initial stages.
Bryan Johnson once put it very bluntly, saying something along the lines of, 'I want to prove that algorithms can take better care of me than I can myself.'
This essentially takes the idea of 'outsourcing certain decisions to systems with less emotional interference' to the extreme.
(Note: Could this be considered the psychological basis for buying funds, especially quantitative funds, haha)
5. Beliefs are shaped more by 'people' than by 'facts.'
Sean: Is there anything that particularly hinders clear thinking?
Daniel: Yes. We almost always have an immediate intuitive reaction to virtually every question. Just ask me a question, and a ready-made answer will pop into my mind instantly.
And these ready-made answers are precisely the biggest obstacle to clear thinking. We can't help but generate them.
Emotion is also a disruptive factor. I would even say that truly independent and clear thinking is almost impossible in many situations.
The reason we believe in a certain viewpoint is mostly not because it is backed by strong reasoning. If you ask me 'why I believe it,' I can certainly provide an explanation, but that explanation is not the real reason for my belief.
We hold these beliefs more because we trust certain people and, as a result, accept their views. Unless you are a scientist or undergoing similar training, your beliefs are rarely something you 'come up with on your own.'
Sean: But even for scientists, isn't it still difficult to achieve complete clarity?
Daniel: Yes, and the scope is very limited. Moreover, scientists are also influenced by emotions. You may cling to your own views, feel upset when others question you, and care about whether others think you are smarter. All of these can interfere with judgment. Therefore, I would say that true clear thinking is much rarer than most people realize.
Sean: Then, is there any way to do better when beliefs are just forming? For example, when we read a well-structured and logically clear commentary that aligns with our worldview, we easily adopt it as our own view. But in fact, we forget that this idea was not the result of our own thinking but rather 'borrowed.' Is there any method to remind ourselves of this?
Daniel: My view on climate change came about this way. I believe in climate change because I trust the people who say 'climate change exists.' Those who don’t believe it trust a different group of people.
Sean: But doesn’t this also mean that we could equally believe in 'fake news,' right?
Daniel: Of course. And I am more likely to believe 'fake news' from 'my side' rather than 'the other side.' Over the past ten to fifteen years, the quality of public discourse in the United States has indeed declined significantly. At least before, we had a consensus: facts matter.
Sean: Why do you think it has come to this?
Daniel: Political polarization has indeed had a significant impact, compounded by the fact that everyone now has the option to only consume information from sources they prefer.
(Note: In 'Thinking, Fast and Slow,' Kahneman makes a statement very relevant to his point here: Many important beliefs do not stem from evidence but rather from the fact that people we love and trust share the same views. It is almost the written version of his assertion that 'reasons are not causes; we believe in people first and then in ideas.')
If we place his remarks within the research context of social psychology and political psychology, they naturally align with two main lines of thought. The first is 'social identity theory,' which emphasizes that group identity shapes how we perceive ourselves and others, thereby influencing attitudes and judgments.
The second is 'identity-protective cognition' or 'motivated reasoning,' where individuals subconsciously interpret information in ways that better protect their group identity and values.
6. Don't overly rely on 'I just know at a glance'; follow procedures whenever possible.
Sean: You mentioned before that there are situations where we should trust our intuition and others where we shouldn’t. Generally, intuition is more reliable in stable environments where you can repeatedly practice, and feedback is fast and clear.
However, I feel that most decisions made within organizations don’t fit this kind of environment, yet we still heavily rely on experience and judgment. So, in such cases, what can we do to improve decision-making?
Daniel: First, I think you shouldn’t expect too much. Keep low expectations regarding 'improving decision quality.'
One fundamental principle is: slow down. Especially when you have a strong gut feeling of 'I am very certain,' it is crucial to consciously slow down.
Indeed, there are processes and methods that can enhance judgment quality, and we can discuss them further. But if you genuinely want to significantly improve decision-making, the answer is one word: algorithms. As long as subjective judgment can be converted into clear rules or algorithms, algorithms will almost always outperform human judgment across all scenarios.
Of course, delegating decision-making to algorithms comes with societal costs. However, in terms of 'decision quality' alone, the outcomes are better.
If you cannot achieve algorithmic precision, slow yourself down.
Different types of problems require different approaches. For example, for prediction-related issues, Phil Tetlock, the author of 'Superforecasting,' has studied who is better at forecasting and how to train such abilities. That represents one category of problems.
However, I am more concerned about another type of problem: situations where you need to evaluate, score, and compare multiple options. This is a type of judgment problem I encountered early on.
At the age of 22, I was doing psychological work in the Israeli army and was temporarily assigned to create an interview system. The previous approach was for interviewers to provide an immediate 'intuitive impression': Is this recruit a promising candidate?
I changed the approach by first defining six dimensions, all of which I devised myself, such as discipline and responsiveness. Interviewers were instructed to ask questions around each dimension, assign independent scores, and only after recording the scores for all six dimensions would they write a final 'intuitive score.'
At the time, many people resented this change, saying we had turned into machines. I told them their responsibility was to improve reliability, while I would take responsibility for the predictive accuracy.
Later, we validated that the new method was significantly more accurate than the old one.
More interestingly, that final 'close your eyes' intuitive score also proved valuable. It did not completely overlap with the average of the first six scores and provided additional information.
Thus, our final approach was to calculate the total score as half from the structured scoring and half from the intuitive scoring. This method was used in the Israeli army for decades, and even fifteen years later when I returned, they were still using it.
This experience became a core idea for my subsequent book writing.
The key is to delay intuition and avoid forming an overall impression too early. We usually form intuitions too quickly, and rapid intuitions are often not very good. However, if you wait until more complete information is available before allowing intuition to emerge, it will be better.
Sean: Then how specifically do you 'delay intuition'?
Daniel: Through structuring. For instance, when a group of directors convenes to evaluate an investment project, my suggestion is not to immediately ask everyone whether they support or oppose it, but rather to go through each dimension one by one.
After reaching a certain consensus on each dimension, make the final overall judgment. This approach not only clarifies information but also avoids initial impressions from influencing judgments.
Because once you have formed an impression, you will spend a lot of time validating that impression instead of looking for real evidence. If your initial impression is wrong, you will end up being 'wrong all the way.'
Therefore, independence is very important. Otherwise, it is like a group of eyewitnesses discussing together; they will quickly influence each other, diminishing the value of the information. What you really want to do is first collect each person’s original judgment.
(Note: When Buffett spoke with Arizona State University professor Jeff Cunningham on March 5, 2015, he mentioned that, in a sense, he considers himself a journalist. He assigns himself topics, such as 'Was the Washington Post Company worth $22 per share in 1973?' and 'Is BNSF Railway worth our paying $34 billion for?' He sets these as his working hypotheses, then seeks out facts without making judgments or selections based on position.
Buffett emphasized that one principle should always be followed: do not jump to conclusions; let assumptions guide the content itself.)
Sean: Then, within an organization, how can dissenting opinions be made easier to express and be heard?
Daniel: There are many methods, such as 'Red Team vs Blue Team,' assigning 'devil's advocates,' and so on. But the most crucial aspect is that, as a leader, you need to protect those minority voices.
This is because opposition is painful and comes at a cost. If you don't reduce the cost of 'expressing dissent,' then opinions won’t surface.
Sean: You have long studied cognitive biases, but you've also said that you are no better than others at avoiding them. This sounds somewhat disheartening.
Daniel: Exactly. When it comes to individual thinking, I am not much better than others. Relying on personal willpower to avoid biases is unrealistic. On the other hand, I have more confidence in organizations because they can establish procedures, slow down the pace, and create processes.
Many people draw up a 'checklist of cognitive biases' and go through them one by one. However, I am not optimistic about this approach. There are too many biases, and they often intersect and reinforce each other.
A more feasible approach is to recognize the specific situations in which you are prone to making systematic errors, and then avoid relying on intuition, substituting processes instead.
Take the anchoring effect, for example. If the other party starts with an outrageously high price during negotiations, it skews your perception of what constitutes a reasonable range. Knowing about this effect allows you to immediately dismiss that anchor and remove it from the conversation.
However, some biases are particularly difficult to guard against, such as our tendency to overconfidently predict extreme outcomes. Our brains are naturally poor at mean reversion.
One of my favorite examples is this: I tell you about a college student who could read fluently at the age of four, and then I ask you what her GPA might be in university. Most people immediately give a very high score, like the 90th percentile. But statistically, this makes no sense because the correlation between early reading ability and GPA is far weaker.
A rational prediction should regress toward the mean, only slightly above average. However, our intuition often jumps too far.
Occasionally, you can remind yourself, but as long as you genuinely care about the issue, it is difficult to avoid bias.
Editor/Jayden