Comments
A Conversation with Daniel Kahneman
16 min readJul 22, 2022
On 7 July 2022, Prof Daniel Kahneman, the Nobel-Prize winning economist and psychologist was interviewed by Prof Jan-Emmanuel De Neve, Director of the Wellbeing Research Centre at the University of Oxford.
The conversation took place at the inaugural Wellbeing Research and Policy Conference and includes fascinating insights on:
You can watch the session here and access all the recordings from the conference here.
The following transcript has been lightly edited for clarity.
JD: Most of you will know Danny and the fact that he received the Nobel Prize in Economics for his incredible work on judgement and decision making as well as pretty much starting the field of behavioural economics.
In preparing for this interview, I looked up Danny’s Google Scholar page and I couldn’t believe my eyes. Danny has 486,000 academic citations! I know I'm not raising the collective wellbeing of the academics in the room by saying this. For the non-academics, as an academic throughout your career, if you do really well, you can probably aspire to 10-15,000 citations. So we're all sitting in front of a legend here.
Most of you will have read his book, Thinking Fast and Slow, which brings all his wonderful research together. What is slightly less known is that Danny has a very impactful line of research on happiness and wellbeing which he coined as experienced utility. In the context of economics, we tend to talk about decision utility or revealed preferences.
That line of research that Danny started in the late 1980s and early 1990s inspired people like Richard Layard and Paul Dolan to really turn their attention to this line of research. So thank you, Danny, for inspiring what are now becoming mentors to many other people.
JD: Let me start by asking about your origin story. How did you come around to taking an interest in happiness and experienced utility?
DK: Okay, so long story. When I was working with Amos Tversky on prospect theory in the late 1970s, I invented a puzzle. Imagine that you have an illness that requires you to receive one injection every day. It's painful and you don’t adapt because it’s once a day. How much would you pay to reduce the number of injections from 20 to 15 or from 10 to 5?
It's immediately obvious that you wouldn't pay the same amount and yet you should pay the same amount. So clearly, there was a distinction here between the utility that is experienced, where 20 is clearly twice as much as 10, and the decision utility. So that was in the back of my mind and I knew that sometime I would get back to that project.
In the late 1980s and early 1990s, I actually got to the question of whether people can predict their future enjoyment of things and whether people can correctly remember and evaluate experiences that they have had. There was a paradoxical result, which is quite easy to duplicate. We first found it in a questionnaire and then confirmed it in various ways.
If you take an unpleasant episode and you add more unpleasantness but diminishing unpleasantness, the memory (or the global evaluation) of the entire episode improves. So that led to the formulation of something that I called the peak-end rule. But mainly it led to the idea that experience and how you think about your experience are completely separate things. At the time, I thought that experience is reality and that what we think about experiences is just biased judgement. The biases are interesting, but I thought there is reality. And so that's how I began my work on wellbeing.
In the late 1990s, I was invited by Richard Layard to give a series of talks at LSE and I presented that work in one of the lectures. It was clear, or it appeared to be clear, that wellbeing research was being conducted with life satisfaction questions, but life satisfaction is a judgement. So I thought we can redo the study of wellbeing with experience and we'll find completely different things. So I engaged in that exercise with colleagues, we developed the day reconstruction method, and that's how I became interested.
JD: Throughout the conference, we've heard a lot about evaluative measures (such as life satisfaction) and experience measures (such as affect and happiness in the moment). We'll get back to that in a second but I want to dig a little bit into the story of how the economics world got to find out about your work.
I think you were the first one through in one of the very top economics journals. In 1997, the Quarterly Journal of Economics published a piece called Back to Bentham? Explorations of Experienced Utility. At the time, it was very much revealed preferences, decision utility, and the ordinal measure of utility that economists were using and to some extent are still using today.
What was the reception to that famous piece? What did your colleagues at Princeton in the economics department think? Did they dismiss this? Did they like this?
DK: I think it was ignored, like everything else the psychologists do. The Nobel actually made a lot of difference. After the Nobel it's been read much more but the initial reaction was very little interest I think. This was published after Amos Tversky’s death in a special issue in his memory, so it was barely refereed, if at all. If it had been refereed I'm not sure it would have been published.
JD: There is another famous paper by Andrew Oswald and others which was published a few years later in the American Economic Review which did go through proper review. And it almost didn’t make it because of very tough referees.
JD: You mentioned already that your interest in the field came through these colonoscopy studies and others with the peak-end rule and the fact that sometimes evaluations or memories of the experience don't quite match what actually happened.
But then in 2018, Danny did an interview with Haaretz magazine and the title of the piece was, Why Nobel Prize Winner Daniel Kahneman Gave Up on Happiness. As a title, this was obviously very attention-grabbing. Luckily there is a byline, yet now he considers life satisfaction of greater importance. As you can imagine, in the field of wellbeing science that was circulating quite widely.
You may not have fully transitioned I gather but how do you feel today about which of the two measures could be an optimal criterion for policymakers or individuals to strive to optimise? Is it the experience or is it the memory of the experience?
DK: So I started out with the position that it's the experience and that life evaluation is just a biased judgement. And I held that position for quite a few years.
We had a group that I set up to get economists interested in wellbeing. Andrew Oswald and Andrew Clark were there but they were pretty isolated. I brought the late Alan Krueger, a very well-known economist at Princeton, and a whole group, to develop an instrument to measure experience. That was the day reconstruction method and it enabled us to look at, quite specifically, what predicts life satisfaction and what predicts emotional happiness. And it turns out they are different in obvious ways.
So life satisfaction is largely predictable as conventional success. Our idea of success in life is widely accepted and people measure themselves roughly by that criterion. Happiness seems to be determined, at least in our work, primarily by who you are with; by whether you're with people you like, in a pleasant interaction. Income was much more predictive of one than it was predictive of the other. There were those very large differences.
So we knew that. But then something hit me that I should have seen earlier about life satisfaction, which is that life satisfaction is what people actually want. When people set goals for themselves and make long-term decisions they're serving their remembering self. They’re evaluating the outcomes, they’re anticipating their memories, and this is what they want to maximise. In short, people want to have a good story for their life.
So I discovered that I had a theory of wellbeing that didn't correspond to what people wanted for themselves and that seemed extremely awkward. It's not that I gave up on happiness, I gave up on happiness as the solution. And it's not that I adopted life evaluation. It's just that I became totally puzzled and baffled and I didn't know how to solve it so I moved on to other things.
JD: Hopefully this conference brings you back in to some extent. But can I push a little bit on that point? How do we know that people may prefer the story of their life over the experience of it?
DK: When you look at the determinants of life satisfaction, those turn out to be goals that many people aspire to. They turn out to be how people evaluate each other, as well as how they evaluate themselves. It's also obvious that if we wanted to maximise our experience, we wouldn't know how to do it because all we get to keep from our experience is our memories. Maximising experience is a very difficult thing. It's an art that may be learnable or teachable, but it's very different from life satisfaction and what brings life satisfaction.
JD: I want to change tack a little bit. Back in 2012, I received an email from you. As a junior scholar this is the kind of stuff that you never forget and I’ve heard of a number of junior scholars that, after publishing a paper, received a note from you congratulating them on the paper and pushing their thinking a little bit further. So I want to thank you for that and I know the other junior scholars at the time also really appreciate that.
In that email, which I looked up again to get the actual text and it's something that has resonated with me ever since, you wrote:
“I'm increasingly troubled by the problem of labelling results by one pole of a dimension which reflects deep linguistic habits rather than the structure of the data. For example, I suppose that being very short is more likely to make one miserable than being very tall to make one happy but the relationship would still be described as connecting happiness to height.”
I've seen that also picked up in studies where a lot of the action is often driven by the bottom end of the wellbeing scale, yet it's described as matching X to wellbeing or happiness. Can you elaborate a bit on this because I know it still is something that you care much about?
DK: Oh, yes. That's one of the few points where Richard Layard and I disagree. I have always thought that we should be studying misery and that the objective of social policy should be to reduce misery rather than to increase happiness.
In part, this was because the dominant view of what ‘increasing happiness’ was at that time (15-20 years ago) was positive psychology which aims to change the way you think about your life. This is nice in a way, but it's also a very conservative message in some sense. It doesn't so much matter to improve your circumstances or to change your life as to make you happier with whatever it is that you have.
So I was questioning that. I thought that reducing misery should be the objective and that the focus on happiness is really linguistic. It's like we measure length, not shortness, and the focus on happiness was guiding us in a different direction if we were thinking of policy objectives.
That's what I thought. Let me say that this conference shows that I was wrong. I thought that if you focus on happiness, you wouldn't focus on reducing misery. It turns out that Richard Layard focused on happiness and the reduction of misery. It's very clear in this conference, and in the book that he’s written that I've just read, that many people are thinking about this. But in terms of talking to the public, happiness is a much better word than misery. That's clear.
JD: Where do you see the research frontier? Genetics? Experience sampling? Twitter data? Where do you think the exciting new avenues for our understanding of wellbeing will come from?
DK: Whenever I’m asked the forecasting question, I’m reminded of a Hebrew proverb which said that prophecy is for fools. I really don't have any forecasts. It’s obvious if you want to know what’s going to be interesting in the next 20-30 years you should be looking at what graduate students are doing now. Personally, I am fascinated by the brain, by artificial intelligence, and by genetics. Those are big departures. But I wouldn't forecast that this is really what's going to happen.
JD: It's been a great privilege for me to be able to sit here and ask a few questions, but I want to share that privilege with everyone in the room.
Question from Prof Andrew Oswald (Professor of Economics and Behavioural Science at the University of Warwick)
AO: If you think about all the things that you discovered, not just pure ideas, what do you think was the thing that most surprised you? The one discovery that sticks in your mind as you were really not expecting that? Is there one?
DK: It's very hard to say because when you really understand something, then it seems that you always understood it. I think the discovery that excited me the most was that people make non-regressive predictions and ignore base rates when they're making predictions. That was quite new to me, and surprising and exciting. Other things that I’ve worked on are typically main effects that are pretty obvious to introspection so I haven’t had deep theoretical insights. I deal with obvious things.
Question from Dr Mark Williamson (co-founder and Director of Action for Happiness)
MW: What would be your top recommendation for policymakers to try and alleviate misery? So many people are struggling right now with low wellbeing. What would be your top insight for individuals to try and help them get further up that wellbeing scale?
DK: In terms of policy, I would say mental health and loneliness would be the first obvious concern and I think that they're widely accepted as goals for policy.
For individuals, I would say something very similar to what we just heard. Time is a very limited resource and spending it well is the most important decision that you’re going to make. Making decisions and constraining yourself by setting up habits and structures that make you use your time wisely would be my best advice (although I don't personally follow it much myself).
DK: You haven't given me the opportunity to say something that I wanted to say. That my main contribution to the study of wellbeing was really to get Richard Layard interested in it. That is really something that I feel very proud of actually.
I actually made another contribution that's not well known. I participated in the setting up of the Gallup World Poll. I wrote the emotion questions and it turns out, which I didn't know, that I convinced Jim Clifton (who directed Gallup at the time) to use the Cantril ladder, which I know John [Helliwell] doesn't like.
JD: I know that you've been reading the work of Jon Clifton, Jim Clifton’s son, and his recent piece in The Economist about the disturbing trends that we're seeing in negative emotions. Do you want to elaborate a little bit on that?
DK: It's clear from the Gallup data that it's a huge change. It's a 10% increase over the last decade in negative emotions (from 25% to 35%). When you see a finding of that magnitude, your first thought ought to be is there an artefact but what I gather is that actually there are many data from other sources that broadly confirm it. This is huge, because things are really very stable in the domain. So to see a 10% change in absolute terms is enormous.
JD: I find that personally very disturbing too, in no small part because it's against the backdrop of growth and low unemployment. The pandemic obviously exacerbated this trend and now, at the end of the pandemic, it has come back to the negative trend that it was on. It's something that I think we, as wellbeing scholars, should really be sounding the alarm bells on just like Andrew Oswald did this morning or Jon Clifton is doing in his pieces. These negative emotions are such a rising trend that it should be acted upon.
DK: Yes, I certainly resonated with Andrew’s talk this morning that this looks like something that's getting ready to explode, certainly in the United States.
Question from Dr Caspar Kaiser (Research Fellow at the Wellbeing Research Centre)
CK: I know that you're not a prophet but you were involved in both the experience sampling method and the day reconstruction method. So I'm wondering, how do you think we will measure wellbeing globally in 30 years?
DK: Well, I have no idea really. One thing that we didn't anticipate when we did the day reconstruction method, we thought of it as a substitute for experience sampling which was impractical. Then shortly after, the iPhone came on the scene and the problem was solved.
We're going to continue to measure experience that's clear. It's also clear already, so I'm not making long-term prophecies about 30 years from now, that wearables are going to come. We're going to know a lot more about what happens to our bodies and this is going to happen soon. It's happening now. Certainly, those things are going to interact and be used together.
Comments from Prof Paul Dolan (Professor of Behavioural Science at LSE)
PD: You would expect me to pick up on your comments about life satisfaction versus happiness.
First of all, on what it is that people want. On a sample of a few hundred LSE students (which is not representative in any meaningful way) we found that 60% of them would choose life satisfaction over happiness, but 40% would choose to be happy day-to-day over having high life satisfaction so it's not obvious that people would choose life satisfaction.
You make a really good point about time use which requires us to measure experiences more directly (what we do, who we are with, what we think about).
Your reasoning for moving to life satisfaction is that people are motivated by it. That’s interesting because one of the reasons that I got into happiness research is that you and others showed over many years that people are not very good at predicting how they’ll feel in the future. The motivation and the goals that individuals have are not necessarily the best yardstick to judge which measure to use because we know that people make all sorts of mistakes and errors in assessing happiness. So experiential data gives us at least some way of assessing to what extent the things that people want actually improve their experience.
DK: I think that what guides people in their long-term decisions is not the view of experience. If I'm asked, do I prefer something today or something tomorrow, then I prefer something today. That's clear, and I prefer experiences when they are short-term. I was referring to the long-term, and in the long-term, I don't even know what it would mean to try to maximise experience. Maximising success in life or maximising life satisfaction (I don't know which is a proxy for which) it's clear that what defines life satisfaction in terms of its predictors is really quite conventional success and much less experience.
Question from Noof Aljneibi (Director of Emirates Center for Happiness Research)
NA: Sometimes those who are interested in the field of wellbeing struggle to decide between defining wellbeing, measuring wellbeing, educating wellbeing, and trying to influence policymakers. What advice would you give to a new researcher like myself?
DK: I wouldn't give you any advice. There's a lot of diversity. People pick different things to do and the mix more or less works out and more or less corrects itself. As individuals, you really have the luxury of choosing your problems. That would be the advice I would give which is really trite.
Question from Prof Carol Graham (Leo Pasvolsky Senior Fellow at the Brookings Institution)
CG: I remember years ago on the National Academies panel, you were discussing anger as a very different emotion. We had a discussion about the negative emotions index earlier and then Andrew’s presentation (the idea that this is all going to blow up) and I agree. I mean, I've just finished reading How Civil Wars Start.
Jan mentioned the biggest increases in negative emotion are in stress and worry. As far as I can tell, we aren't seeing huge increases in anger. We're seeing increases in public protests. I was just wondering what you think of that. Particularly as you talked about this being explosive, potentially.
DK: Those questions about anger are personal questions. Did you feel angry yesterday about something that happened yesterday? But these questions are not measuring anger with the world. It's the kind of despair that we heard about earlier but it's not the kind of emotion that you would pick up in experience sampling necessarily. It‘s part of the way that people view the world and their place in the world. And that is clearly getting worse.
In part, but only in part, it's probably because of the decline in the importance of religion, which I think provided some framework. In the United States, people pretend to be religious but that's where things are the worst.
I know why you refer to me saying that anger is a different kind of emotion. That's because, in general, the negative emotions are associated with avoidance but anger is a negative emotion that's associated with approach. Physiologically, it is halfway between the negative emotions and the positive emotions and that’s what I was referring to.
Question from Amanda Janoo (Economics and Policy Lead at the Wellbeing Economy Alliance)
AJ: You were talking about how, particularly around emotional happiness, it's so much about who we are with, the importance of social connection and not feeling lonely. I was wondering if you've done any research or have considered also looking into a lot of the data around how our relationship with nature and spending time in nature has huge impacts on our happiness.
DK: Well clearly, that's something that studies in experience sampling are going to bring to light. Nature has a huge effect on how people feel. It’s the weather, trees, greenery, water. There are children in the urban poor who have never seen nature. Making that part of education and making sure that everybody knows the joys of nature, that I think is a worthy policy objective that's not about reducing misery. It's really about something positive.
JD: It's the wonderful work done by George MacKerron and mappiness and these kinds of experience-based sampling using geotags as part of iPhone pulse surveys that have really been able to show, very concretely, the proximity to green or blue spaces (like the Thames for example, in the World Happiness Report chapter) is very positively related to how you feel. Especially if it's with other people that you enjoy spending time with, and even more so if you're doing it actively by moving such as jogging together.
Question from Mike Nolet (Director at LiveBetter)
MN: Mindfulness is about being in the present moment; not worrying about the future or ruminating about the past. How do you marry that with a focus on life satisfaction or the memory of things when a lot of research is showing that the more mindful we are and the more we are in the present moment, the happier we are.
DK: Mindfulness is actually highly active. I mean, you're doing nothing but you're actually completely focused. I was interested when we were mentioning the amount of television watching that there is. Television and mindfulness are both passive physically but they are radically different. When you watch television, you’re not in the moment; you are nowhere in some sense. So being in the moment, is a very different art.
