The Joe Walker Podcast - On Radical Uncertainty, "Phantastic Objects", And Blowing Bubbles - David Tuckett

Episode Date: October 16, 2019

David Tuckett is a psychoanalyst, Professor, and Director of the Centre for the Study of Decision-Making Uncertainty at...See omnystudio.com/listener for privacy information.

You're listening to the Jolly Swagman Podcast. Here's your host, Joe Walker. Hello there, ladies and gentlemen, boys and girls, swagmen and swagettes. Welcome back to the show. One of the joys of doing this podcast is that sometimes I get to introduce to you ideas you haven't yet encountered but should have and people you haven't yet heard of but should have. I suspect for many of you, one of those people will be my guest, David Tuckett. David is a professor and the director of the Center for the Study of Decision-Making Uncertainty, which sits in the Faculty of Brain Sciences at University College London. David is rare not only in that he's an interdisciplinary scholar, but also in that two of his main disciplines happen to be economics

and psychoanalysis. David is known for spawning a new field of research known as emotional finance, and to give you a sense of the regard in which his research is held, consider that two of David's friends, George Soros, the billionaire investor, and Mervyn King, who was governor of the Bank of England from 2003 to 2013, have both taken and applied some of David's ideas. Indeed, I originally discovered David through the writings of Mervyn King, but several economists also independently recommended to me David's book, Minding the Markets. One of those economists and a former guest of the podcast was Timo Henkel, who recommended the book during my Housing Bubble Week series. In my quest to understand Australia's housing bubble

and to understand bubbles more generally, a lot of signs seem to be pointing towards David Tuckett as holding some important answers. Specifically, answers to questions like, can we quantify animal spirits in order to predict bubbles and crashes? And questions like, how should we think about narratives in markets? The last question is a question which I became intensely interested in after my conversation with Nobel laureate Bob Schiller. For example, during our conversation, Bob suggested that theories about Asian buyers driving up prices might underpin domestic sentiment in the Australian housing market. In this episode with David, I develop my theory of this narrative about Asian buyers a little further and provide some new

evidence supporting it. In essence, this conversation is about getting real about human nature and how we behave in markets. I was thrilled to speak with David. He at least met my expectations and as you'll hear is not only a genius but also a lovely guy. So without much further ado, please enjoy our conversation. David Tuckett, thank you so much for joining me. Pleasure. I'm very excited to speak with you. I've read your fantastic book, Minding the Markets, which was actually recommended to me by several people.

And here we are. And I thought perhaps we could begin the conversation by framing it with a simple formula. And the formula is this. The price of an asset, assuming it has an infinite maturity, can be described as the sum of two components. And those two components are the fundamental component and the bubble component. So in other words, the price or P at some time T equals the fundamental value, or FT, plus the bubble value, BT. So the price in this equation is the actual price we observe, and that means to work out whether a bubble exists,

we only need to look to see whether the fundamental value is below the asset price. And if it is, then voila, we've found a bubble, which is that difference between the asset price and the fundamental value. So it sounds pretty simple, but there's a big problem here and that's measuring fundamentals, of course, which we could define as the present value of the future stream of cash thrown out by the asset. But as soon as we mention that word future, things begin to get a little bit tricky. So, I want to come to this idea with you, but just before we do that, let's quickly talk about your background, Dave, just to introduce you to people. So, you've developed this field now known as emotional finance uh but before that you became very interested in psychoanalysis so just tell

you know when when most people hear psychoanalysis they think freud and jung and pseudoscience how does modern psychoanalysis compare to those founding fathers? Well, so I think you're quite right about that. Most people do not understand what psychoanalysis is about and probably pick up a received opinion from some of its critics, perhaps also didn't know much. modern psychoanalysis is fundamentally about the proposition that thinking and feeling are interconnected so any thought you might have about anything significant we now know will also if you look at someone's brain you know in experiments and so on, you can see that thoughts generate or at least light up bits of the brain that have feelings. And in particular, every single brain network that's activated will go through the primitive part of the brain, which we share with mammals, all mammals, which ignites either an approach or

avoidance response. So all the time, at a primitive level, we are evaluating the world and our thoughts and other people's thoughts in terms of whether we want to move towards them, get them, have them, relate to them, or move away from them now of course it's more complicated than that but that is a true statement of the basic level now Freud recognized this he did start out as a neurologist which is perhaps not well understood and in fact he turned or invented psychoanalysis when using the neurology of the day, he was unable to explain why patients of his suffered from hysterical paralysis. That's literally they couldn't move their arm because it didn't seem to work from a neurological point of view. And so he came to the idea that what they basically were suffering from was ideas, ideas they could not themselves know or express.

So I won't go back into Freud, but in fact, his view of this matter has largely been confirmed by what is now called the field of affective neuroscience. People like Damasio, Panksepp, Mark Soames, if you read all that stuff, it's extremely interesting. And it certainly would suggest that Freud was foresighted. Of course, it wasn't consistent with the neuroscience that came afterwards. And indeed, some of Freud's neuroscience is actually wrong, but most of it's wrong in the right direction. So I became interested in psychoanalysis when, as a young economics student at King's College, Cambridge, we were in those days, economics wasn't like it is now. We also were taught sociology, politics, political sociology. And we were in fact introduced at that time to a man called R.D. Lang, who was a psychoanalyst, who at the time was a very hot

guru, used to know the Beatles and the Rolling Stones and people like that. And I became interested in his work. And then through that psychoanalysis, after I left Cambridge, I pursued two tracks. One track was as a medical sociologist, which was pursuing sociology because I found economics too abstract. And the other track was learning to be a psychoanalyst. Then in 1990, quite a lot of time later, when I was running a small, what would now be called startup company, which was producing a CD-ROM of psychoanalytic publications, in this room I'm sitting now, a couple of people came through the door who were leading venture capitalists, actually, and investment bankers working for whatever it was called, Kleinwert Dresden, I think it was called then.

And they wanted to take over this company, give us a lot of money and create a dot com. And the more I listened to what they were saying, they took us out to some expensive dinners and so on. We came to the view that they didn't know anything about psychoanalysis, publishing or the Internet, and decided not to accept the check. And, of course, around 18 months later, as everybody knows now, the whole dotcom thing, which was a bubble burst. This got me interested in what were bubbles really about? And this allowed me at last to bring together psychoanalysis, sociology and economics and to form this new set of ideas. Fantastic. So let's come back to that question I left hanging at the very beginning. Why do fundamentals make it so hard to tell whether or not a bubble exists.

There's a very important concept known as Knightian uncertainty, named after Frank Knight, who introduced it in a book published in 1921 called Risk, Uncertainty and Profit. Tell us what Knightian uncertainty is. Well, that's a very good question, actually, because knights' writing is very rarely read, especially by economists. They refer to knightian uncertainty in a fairly simple way, making a distinction between risk and uncertainty for knight. Risk is where you can reasonably put probabilities on expected outcomes. Uncertainty is where you can't.

And then what happened is that Friedman came along and deploying Savage's ideas of Bayesian probability, subjective probability, said, well, you can always have a subjective probability, which is kind of true. Therefore, we can forget this distinction. Now, in fact, what Knight was talking about was much more fundamental. He was essentially talking about the fact that the dynamics of the world, the human world in this case, producing the outcomes we observe are actually unknown to us and constantly changing. So, for example, to give you an example, if you try to look at statistics on finance and try to derive conclusions, you know,

with a vast amount of data you can get about financial movements, in order to draw conclusions, you end up having to make assumptions that the finance world has been more or less like this since the Stone Age. And, of course, it's made fundamental changes in its dynamics, in its institutions and everything since then. So Knight was really talking about the fact that the future is yet to be made. And actually, we try to imagine it, but we usually fail. So you probably know the famous statement of Henry Ford.

If I'd asked my customers what they wanted, they would have said a faster horse. Right. Or I can remember because I went on to interview a whole group of fund managers in new york in 2007 i can remember going one of them told me about the apple iphone which at that time that was pretty much unknown it's not that long ago only 12 years ago and i went to the Apple store after the interview and saw all this. And at the time, people had been selling Apple shares, actually, because they hadn't been doing very well at that point. And nobody realized what was going to happen. And, of course, certainly nobody realized the kind of networked society

we now have pretty much as a result of those particulars so it the key idea here is that is that value depends on imagination yeah and actually what what economics and some other sciences completely fail to understand is that you don that information doesn't grow on a tree where its meaning is absolutely clear and you know exactly what it means. Most information has to be interpreted by the subject, has to be made sense of. And it's because it has to be made sense of. So, for example, I once rode in a car with a managing director of Siemens, which is a big electronics firm in Germany, worldwide firm. And we were chatting on the way to a conference and I asked him, you know, how do you value your company? And it's perfectly

clear from his answer that in a massive company like that with numerous divisions doing all kinds of things, it's a total guess. It's based on all kinds of conventions and accounting itself, as we discover when there are these accounting scandals, is extremely difficult to do because valuing the future is fundamentally problematic. Yeah. Dave, just to kind of pause and drill into this idea of why the future is radically uncertain, you mentioned the Henry Ford aphorism that if I'd asked my customers what they wanted, they would have said a faster horse. Steve Jobs used to love quoting that. There's a related idea which I get from Kyle Popperper's book the poverty of historicism but it probably

has roots that go back further he says that the the future is fundamentally indeterminate because the course of history is in large part based on new technology we can't know the technology the future since if we did we would already have invented it therefore the course of the future is fundamentally indeterminate so we have this idea of and you know even even if we could keep tabs on every project in every entrepreneurs garage happening around the world at the moment it still wouldn't really give us a good handle on the technology of the future because technology turns corners uh through innovation we find new applications there are many examples of this you know um viagra is one

example the internet is another penicillin is yet another um you know trains originally came from pumps used to you know steam trains from pumps used to pump water out of rivers um so there's many examples of this but is there i mean that, that's one reason for knightian uncertainty. Are there other reasons in this category of nonlinear chaotic dynamics in an economy and in human social life? Well, I think you can express it in quite a few different ways. So one aspect is that uncertainty is really the other side of free will. So, you know, going back to the utilitarians of the 17th century, you know, they tried to work out for Hobbes and Locke and people.

How does that get you get any kind of order in society? Because the original theory was was divine, divine right, which was the king. And these people, of course, were arguing against kings. And so they came up with the idea, didn't they, which became known as utilitarianism. But if you analyze utilitarianism, it's a very strange kind of thing, because on the one hand, it says that human beings are free to choose. They can choose their utility, et cetera, et cetera. But on the other hand, it says that actually order arises in the system without any external

king or someone. And this is the theory of how the market works. But actually, as people like Marshall, the economist, but also Max Weber, Durkheim and Pareto demonstrated in various different ways, ultimately, if you adopt that utilitarian perspective, order, in fact, comes from the what's called the kind of environment. So something in the environment constrains us, right? Now, clearly, that is not the case with humans. If you allow humans to be free, then, as you were saying in your examples, we can invent stuff. So it arises from human interaction, which then produces inventions, I would say.

And of course, inventions are not necessarily in and good of themselves. This critically depends, you know, if I have a good invention, or you do, or you've got your lovely blog, at any moment, people may decide it's not interesting, or they may decide it's not interesting or they may decide it's terribly interesting so it's an interaction always between your idea and how you produce that idea and whether people like it so that is a fundamentally dynamic and complex system and of course uh if you extend it and you think well you know this is the way you're doing your blog now, but some way we haven't yet imagined someone else could be doing it, you could suddenly be knocked out overnight, so to speak. So when we come back to the question of the fundamental value

of sticking with that example of your enterprise, it is based on some estimate we make about how many people want to listen and how much they want to pay and stuff like that, and what your capability for expansion is. And at the same time, it's based on the assumptions we make about other people coming along who are your competitors, which could produce almost infinite variety of possibilities. And at the same time, it's based on the people who are going to buy shares in your Walker Inc. or whatever, how they assess those very difficult, how they make those judgments. So I, although I think it's kind of useful to think of fundamental value somewhere in this mix, the underlying problem is that you cannot know fundamental value except via human perception, human narrative, and making what amount to guesses. I would actually argue that all statistics is actually in the end

based on heuristics. So you make, for the reason I gave with the Stonehenge earlier, that we're never really in a situation where we know the full universe of data, or in other words, know the true probabilities. Wow. This is such a profound idea. There's a lovely quote of yours, which is, there's no such thing as fundamentals, only narratives about fundamentals. Yeah, exactly.

That idea is just so profound. When I first read that quote, I think it was in a Mervyn King speech. Mervyn King, the former governor of the Bank of England, who was quoting you. I thought this is so important, albeit pithy, and not enough people seem to understand this idea. Before we move on, I thought just to help drive home this concept of knightian uncertainty for everyone which is kind of going to be the backdrop to our whole conversation i thought i'd read the the john maynard canes quote uh because i think he

summarizes it very elegantly so he he also discussed this idea idea in a response to critics of his general theory in 1937. So the famous Keynes quote quoting here is, The game of roulette is not subject in this sense to uncertainty. The sense in which I am using the term is that in which the prospect of European war is uncertain, or the price of copper and the rate of interest 20 years hence. About these matters, there is no scientific basis on which to form any calculable probability whatsoever. We simply do not know. End of quotation. Exactly. Although he is still there in that quote,

he's still using the framework of probability, which, I mean, it's not that I disagree with the quote, but it's like then when people start talking about black swans and things like that, they kind of have an idea that the black swan is on a probability distribution, whereas in fact it isn't. If you see what I mean. Yes, yes, I do. So just riffing off that idea for a moment. I mean, does a world of nighty and uncertainty give us carte blanche to believe whatever we want about fundamentals? Are some narratives more defensible than others? So this, I think, is a very good question because it gets to two things. The underlying anxiety that academics at least,

and perhaps also professionals, encounter if they start taking this idea seriously, because they start to think, oh, my God, we know nothing. They become nihilistic. And that's pretty worrying. The second aspect of it is that economics is based on, modern economics anyway, is based on a set of propositions in what my colleague Lenny Smith calls model land. So in model land, you can write down easily an equation as you suggested earlier where you've

got a fundamental value you can write down all sorts of things you like and you can you can create an internally consistent system which will proceed and give you answers but without asking the question well how do you ever know what fundamental value is. So from this point of view, the idea of the model land definition of how the economy works has become so dominant that economists tend to divide. They tend to think of sort of rational economic behavior and what they call behavioral, where behavioral is something that deviates from the model. Really, it should be the other way around. The model should be seen as something belonging in an abstract world. And the person who's very interesting here is A.N. Whitehead, the physicist who talked about misplaced concreteness.

Models are, in fact, all imaginary. I don't know if you've come across the work of the German sociologist Jens Beckert. But he talks about what are called calculative devices. So this would be spreadsheets, all kinds of risk models, macroeconomic models, weather models, cost-benefit analyses model. All these things are seen as calculative devices, which of course can be quite useful, but they are ultimately imaginations of the future. And if we understand they are imaginations, which is Whitehead's point about misplaced concreteness, we then are much more likely to understand where they may be going wrong, i.e. through failures of imagination in one way

or another. So to go back to what you were asking,, or the panic that if you accept uncertainty, kind of anything goes, is based really on the proposition that there's an alternative. That is that in the model there is. But as soon as you come out of model land into real life, how do people make sense of the world? Well, they make sense of the world using narratives as one of the primary cognitive means at their disposal. And as a result, everything is a narrative, including everything is a set of ways of trying to put things together and imagine the future. So going back to the key aspect you asked me about, can we be more confident in one narrative than another, right? That's really what you've asked. That is what I now think is the central issue in all decision-making science. So I have something

which has come out of my book, which is called Conviction Narrative Theory, in which the center, the central focus is on how do people come to believe the, if you like, models of the future that they're creating. And I think we can there talk about good process and bad process. We can never talk about the decision being right, because by definition, if there is night in uncertainty, you can never know if you're right. But you until afterwards, and of course, economics cheats, as you know, by reversing this, this particular thing, said many years ago but that uh if you accept that um uh there has to be a narrative a conviction narrative in order to allow you to act because when you act you know if you set up a company you you have ideas that this is going to be very successful it's going to give you this that and the other a narrative but you also know that you could face loss possibly disastrous loss so it's going to give you this, that, and the other narrative, but you also know that you could face loss, possibly disastrous loss.

So it's how people manage that that is at the heart of all decision-making, and you can have a process that can make that better or worse. You may be able to use a model to tell you that actually certain things are impossible. I think 19 uncertainty doesn't stop you saying that certain things are impossible, that you can, I think, 19 uncertainty doesn't stop you saying that certain things are impossible, right? You know, for example, with climate science, certain, you know, the relationship CO2 to this, you know, has a particular kind of thing. But you always have to retain this idea of uncertainty so that it's in that it's in that uncertainty that bubbles form.

The reason why there are bubbles is because people don't know the future. And, you know, if you if you consider the value of Amazon in the dotcom bubble and its value now, they weren't wrong. They were wrong that lots and lots of different companies could all become equally valuable. So let me summarize what you've said to make sure that I have it correctly, Dave. Departures from Bayesian inference are not irrational, as the behavioral economists would have us believe, in a non-frequentist world. Yeah. And yet there are still ways of judging which narratives we should have more confidence in than others

using some sort of correct process. Exactly. Okay. I think a very good example of this is if you've read Schiller's book, particularly if you're interested in housing, Irrational Exuberance. I have read it, yes. So Schiller, unlike most other people, did talk about all this before it happened. So he gets a lot of credibility for this.

Although Eugene Farmer would say he's just lucky in where we're being fooled by survivorship bias. Well, he might. But the fact is that the book itself carries out a very forensic examination of, you know, for example, the idea of the Internet superhighway, in which he points out that goods will have to be delivered via motorways and transport systems and so on and so forth. So he does what I think you just mentioned, is that he subjects an idea of the future to what we know. And now, of course, you can say, oh, well, but maybe there'll be a completely different way of transporting goods around.

Right. Yeah. But then you'd be constrained by some aspect of the laws of physics. In other words, I think the whole point here is that there's not been an adequate recognition that scientific thinking is always based on hypotheses and evidence for them which give you confidence, but they are never actually true once and forever that is there'll be a context in which most scientific findings don't work okay so to gain confidence is not a zero one sort of choice there's always degrees of confidence which might think we're back to base and kind of the bayesian logic is fine. The difficulty is not the Bayesian logic. It is the question of whether or not you, for example, do you ever have

or could you ever have the data from which to derive this particular procedure? Okay. So let's explore this idea with another example. So, Dave, I attend roughly one Australian property investor seminar per month purely as an anthropologist. And I missed one of my favorites, which is I Love Real Estate because it has a very charismatic guru, Dymphna Beholt. They had a two-day summit about a month ago, one weekend. But I listened to, she recorded her main presentation and released it as a podcast. So, I was able to listen to that. And she asked the audience in that podcast, what would the median Sydney house price be

in 25 years' time if we experienced the same rate of growth that we experienced in the previous 25 years? And the answer she gave was $6.349 million. Now, I actually went subsequently to another seminar, this time in Parramatta, which is in Western Sydney. This was only a couple of weeks ago. And they actually said the same thing. And it turns out, you can actually trace this back to a report by Aussie Home Loans, which came out recently where they argued that it was an extrapolated median house values by 2043 for Australia and for all the capital cities. And for Australia, they extrapolated median house values by 2043 for Australia and for all the capital cities. And for Australia, they extrapolated the value to be 2.9 million.

And for Sydney, it was the same as these seminars, 6.349 million. So, I think they may have got it from Aussie home loans. if you assume a 3% annual wage growth to 2043, a median Sydney value of 6.349 million implies a price-to-income ratio of 26x. Now, in a world of 90 and uncertainty, I can't say that's not going to happen, but it feels pretty implausible to me. So, you know, they're within their rights to say that's not going to happen, but it feels pretty implausible to me. So, you know, they're within their rights to say that because we don't know what's going to happen. Is there any way in

which I can say that that extrapolative belief is irrational? Well, let me just try and think my way into this i mean the the underlying point is that is that they are they've chosen a beginning and end point uh to create their average value to extrapolate i suppose they in other, they've looked at between then and then and what actually happened. And they've then said, well, it's going to go on happening like this till whatever it was. So, I mean, I think as you know, if you change the end, if you change, you know, as most fund managers know, with quant models, you know, if you mess around with the end and beginning dates, that may well change the dynamic.

But the underlying idea here is that the dynamics either are understood, the dynamics producing this outcome, either are understood or don't matter. Yeah. And that's a big assumption okay because it seems to me that um that the only way as you say knight in uncertainty but the only way we could envisage this happening would be with massive relative house price inflation to other things because otherwise who's going to live in these houses um and and of course it's possible there'll be massive inflation i suppose of that kind um but what kind of investment have you really got there yeah i mean i mean i think what what they're trying to do there is create a fantastic object and all fantastic objects um have some logic so it's you know you can't a fantastic object actually yeah i'll ask you about those uh a moment. Yeah. I mean, the other, this is related to your idea of it depends where you set the period.

But the other issue is, you know, real estate cycles are historically long. Yeah. And we have a national house prices index, which was composed by Professor Nigel Stapleton going back to 1890. So, Australia has a long-run prices index. A lot of people choose to start their analysis very recently, which coincides with a large run-up in household debt and a big boom. And to me, that's a little bit like landing in or visiting Seattle on a sunny day and thinking that it never rains there. Well, I mean, also also 89 is only what 50 years

ago exactly and 50 years may be a long time in our lives but it's not it that goes back to the point i'm making earlier it's a very short time yeah for deriving any city for doing any serious statistical work because i don't know how many downturns they've been in the last 50 years in the housing market but probably three or four serious ones yeah or less so you really can't tell uh from such it's my point a short data series is doesn't allow you to do the kind of statistics that you would need to do if you're to draw those conclusions so to come back to your original point it's poor process yeah exactly gotcha what what what has happened is you can so you can say well they're trying to imagine the future

and then they're trying to um uh convey this imaginative view to the people they're trying to imagine the future and then they're trying to um uh convey this imaginative view to the people they're trying to get to to buy their shares or whatever and they're doing a lot of excitement about it but a uh one way of looking at whether we think about you know and that's fine but one way of thinking about it further is exactly to look at the data, look where it's coming, consider alternative viewpoints and see what we think. Is this a useful way of imagining the future? Yeah. short so as you know in the big short there's one guy who did start shorting the u.s housing market very early on i think around 2003 or something right well and as kane said it all depends on the size of your wallet because he he was a very successful investor who built up a large list of

clients i think he's an ex-medical student yeah dr michael barry that's right and he you know he did in fact get it right yeah but the intermediate experience lost in most of his investors and much of his mind so so it's one thing you know to it that assumption about whatever whenever it was you said, is not leaving out that there's likely to be a lot of variation in between where you've got to hold your nerve. Exactly. There's a related idea, which I think Paul Samuelson articulated very well.

He wrote this in 1957. I'm quoting here. He said, I've long been struck by the fact and puzzled by it too that in all of the arsenal of economic theory we have absolutely no way of predicting how long a bubble will last to say that prices will fall back to earth after they reach ridiculous heights represents a safe but empty prediction why does some main why does some manias end when prices have been ridiculous by 10 while others persist until they are ridiculous to the tune of hundreds of percent well exactly and obviously you can you know there are technical um technical things you can talk

about there such as eventually, you know, various relationships become too strained. But ultimately, value depends on imagination. And my fundamental point is that given that values depend on narratives, the most, the big point is a narrative can change very quickly. The real world, that is a number of houses in the world in this example doesn't change very quickly at all So it's the narrative that shifts around Etc Which which would seem to explain the excess volatility we observe in financial markets Yeah to explain the excess volatility we observe in financial markets. Exactly, yeah.

Dave, I just want to stop at this point and talk about it because it seems like a good point to do that before kind of moving on with the core of the discussion. And there's a debate about rationality, which is fairly famous, at least in esoteric academic circles and on the one hand we have daniel kahneman and amos toversky all of their work the the program of biases and heuristics and then on the other hand we have good gigorenza the german psychologist and to me it feels like it's it's kind of much ado about nothing. And the debate is kind of an aesthetic debate about should we label these departures from Bayesian inference irrational or rational?

You know, Gerd Gigerenzer says that Kahneman and Tversky gave heuristics a bad name and that heuristics can actually make us smart. We've just spoken about this idea now that it's kind of unfair to label certain behavior irrational or what behavioral economists call irrational in a context of nighty and uncertainty. But I just want to give the behavioral economists some credit for a moment in that whatever we think of these labels of they've been fairly successful at describing how human decision making works uh and i'll give you an opportunity to to comment on that in a moment but i just thought i'd discuss one example so i really like andre schleifer's work about extrapolation and speculation in bubbles he's he's done this work with a number of co-authors principally nicolai genioli they

have a great book called a crisis of beliefs which came out at the end of last year but they talk about how um so they form this model of speculation by actually incorporating kahneman and Tversky's representativeness heuristic. Yeah, I know the book. Yeah, I thought it was really neat. So, the way it works is some good news about an asset comes in, you know, for example, news about changed fundamentals or a price rise, and speculators, instead of updating their beliefs about where fundamentals could go according to Bayes' rule, what they actually ask themselves is this feels like a boom. It's representative of a state in the future, which is really, really good for the asset um and so they kind of fling these these pieces of news into the future and and overestimate the state of the world and in so doing they neglect uh base

rates do you think that's a good a good description of how the psychology of speculation actually works and do you think beyond this rationality debate, behavioural economics at least has some things to offer to us in terms of describing how humans decide under uncertainty? Well, I certainly think that Schleifer and Gennaioli and that recent book is potentially a big step forward. Wow. But I would see them as limited by the framework we discussed earlier. That is, they do not actually fully get it that what they call rationality only works in model land.

Yeah. So there's no such thing as rationality of what they're talking outside the model they don't apply it you don't apply it nobody applies it apart from a few econs right who are probably a bit autistic so you you cannot actually function like that. And that's because, for example, the use of the availability heuristic, right? What that means is that you're limiting your sample before drawing your conclusions. And you're saying what matters is the things I can see around me, right? Whether around you is however you want to define it, right? Now, as I've tried to explain to you earlier, that sampling activity is inevitable.

There are no situations virtually involving human behavior, which is very different than, for example, the behavior of microorganisms or chemical constituents, where you can have such a large number of interactions you can get towards infinity. But in this, which is what you need for these statistical operations. So what I think about the behavioral economists is this. They're beginning to understand the importance of beliefs, but they set beliefs within a framework. In fact, they try to suggest, if you read the last chapter of the book,

they try to link this to biology. Yeah. They try to make out it's a biological thing, and they completely miss out psychology and sociology. Exactly, yeah. So he talks about the biology of memory, and he talks about things like interference. Yeah.

Yeah. And they don't really – so I don't know if you've come across the word of my colleague Nick Chater. I have not. So Nick, who's one of the leading psychologists in the world, he's a Bayesian, right? But they use the Bayesian approach in a very different way. So he has a theory. They have a theory called decision by sampling, in which the suggestion, along with my other colleague here at UCL Carl Friston is one of the most well-known

neuroscientists. They're talking about Bayesian processing but in a very different way. So they don't assume that the prior that you bring to the table, so to speak, is necessarily an accurate prior. So in other words, they would describe human beings, and they think they can explain many experimental results, as given what humans have got to work with, they behave pretty, I would prefer the word sensibly. That is, they do, you know, they do. So, for example, why would you pay attention to what Kahneman and co. would call an outlier? So, you probably know the Linda problem. Yes. You know the Linda problem? Should we explain it for people? Yeah. So, the Linda problem, as I recall it, is that people are asked or people are told that Linda is a librarian.

And then they're told, but the way it's set up is very much part of how this all works. They're then told that Linda could be a feminist. She's either a bank teller or a feminist bank teller. That's right, yeah. And they tell you various things about it, and then they show that people give different answers, and most people give what they call the wrong answer. The wrong answer is defined by knowledge of statistical frequency. So by

definition, somebody being two things is less likely than being one thing. Because feminist bank teller is a subset of bank tellers. Yeah. So, you know, that's clear enough clear enough right but that's a piece of logical nonsense which is of no use in most human situations where humans actually know have some sense of the distribution in their relevant community right and you can do the same thing with irish people with red hair and black hair and all this kind of stuff. There are endless examples of this. And the suggestion, and this is why I agree with Giga Renza as well,

Kahneman Co. set this up to pretend their answer is the correct answer when it is not the correct answer locally. So the big thing here is the difference between local and general and this is why gigarenza's argument is so powerful that locally uh something is different than it is generally so locally for in other words it may be particularly relevant that bank tellers are feminists, if they're whatever it might be. And Gigerenza showed, well, he thinks that he showed you could make the fallacy disappear by presenting it in frequentist terms.

And you can. There's been experiments to show this. So that particular, now most of the things that Kahneman and Tversky are talking about, and I just like – I mean, I do think they're great. You know, they did great work, and I admire it tremendously the way they, you know, sat together in some cafe in Jerusalem and thought of, you know, an argument. It's great stuff, and there is no doubt that those kind of biases are well worth knowing about. So you're not saying that the representativeness heuristic doesn't exist? No.

But you need to judge it according to its local, and that's Gert's point, ecological, right? Local ecological adaptiveness. So they talk about heuristics as general things a heuristic if you can assume that you've got all the data that exists an optimization equation will always be better than the heuristic of course it will but what a heuristic heuristic, the idea about heuristic is it's locally valid. And in fact, if you you might be interested in another book by another of my colleagues called Robert Smith, it's called Rage Inside the Machine. He's a machine learning expert.

He he talks about what he calls the statistics heuristic. And in fact, most use of statistics is in fact heuristical because you make assumptions about the distribution which are actually guesses i was talking about with another colleague of mine is trying to predict the results of the nfl if you into that sort of thing and he's actually in the 95th percentile for doing it. And in order to be in the 95th percentile, you're either going to be there by luck, but if you're consistently there, it's by ignoring the true probabilities.

He could explain that. But it's basically realizing that in the NFL, in a season, there's a limited number of matches. And what you're trying to do, games and what you're trying to do is not to predict every the result of every match but to be at the top of the group of predictors yeah and i won't go through the logic, but this is another example of where local phenomena outrank a general position. So I think the problem here is the equation of normative statistical solutions with rationality. There are lots of different ways of being what I would prefer to call sensible. And the fact is that Kahneman and Tversky have got us in, and all the people who follow them,

have got us into the muddle where human beings are very bad decision makers, and yet look at the world, right? We've created all this stuff, etc, etc, which doesn't make sense so it's better to think of human beings as people who make decisions by sampling where what influences their sampling is local considerations emotions narratives etc so what you call sensible or good would call ecological rationality yeah exactly exactly yeah fantastic so you know if you're trying to solve your housing question earlier what what i would think is you you know you you would try to think about sydney and you try to think about you know what what is going to happen there and so on and so forth. And it would seem to me hard to suppose, unless there's massive inflation in the whole world, that there could be people living in houses in Sydney at those prices.

Don't ruin the party, Dave. Come on. Now, you mentioned that one of the problems you had with Schleifer and Giannioli's book was in the last chapter, they connect their model to biology, but not to sociology and psychology. What did you mean by that statement? Well, so i think that um have we already discussed the answer or did you mean something we haven't discussed well i mean i think it's slightly beyond the scope of this talk to go into in real detail but the under that the basic way to understand how people make sense of the world, of course, it rests on their biology, and the biology is important. And I take from the biology things like the role of emotion as a shortcut.

I mean, you know, something feels right, you know, it often is right. It's another kind of heuristic you know it's not always right but you have to stop at some point so they they they choose to go with because they've read and been interested in the the the bayesian work the bayesian brain, if you like, and they're using those ideas going back to biology, but without the social level. And that's actually the narratives that give the sense. Okay. So the reason I actually pulled up and asked you that was I thought you

might have been hinting at an idea which I articulated in my podcast conversation with Robert Schiller, where I said, so, you know, that whole program of Kahneman and Tversky biases were, those experiments were conducted on individuals in isolation. And I said to Schiller, we might kind of describe that as perceptual cognition. And then I said, I think we should hijack kahneman's metaphor of system one and system two and talk about system three or social cognition because it's not just the individual biases it's things like conformity contagion memetic desire social learning there's this this whole other world as well which is relevant to financial markets and bubbles so when you said they just focus on biology i thought dave's probably referring to system one um but there's also

psychology and sociology or what i would for shorthand call system three yeah although of course there's not much evidence that those systems actually exist yes that they exist or that they're distinct discrete categories yeah i mean the problem with those systems is that they appeal to people who think particularly for example in fund management people know that the market's very emotional they know it and many of the ones who talk to me said you know i've the aim here is to not be emotional. I think I would never give my money to anyone to invest if they didn't realize that emotions are very much part of it. That's a fascinating...

Which is not to say that we want them to be irrational, right? That's right. And that's a fascinating point you make in your book. And I want to come to this now, the role of emotion in financial decision making. Keynes was very famous in chapter 12 of his general theory for introducing this idea of animal spirits, which was that because there's no way of knowing what's going to happen in the future, it wouldn't be rational to make any decisions given that the future is unknowable. Therefore, the reason people act must be due to animal spirits. There must be something else motivating them.

There's another, you mentioned Damasio earlier in our conversation and you cite him on one occasion in your book. He obviously, he has this fantastic book called Descartes' Arrow, but he looks at people who have had brain injuries, who are essentially the economist's ideal, the econ. And he just, he, you know, people who have no emotion and you would think that they would be fantastic decision makers. And he describes these almost humorous anecdotes

of trying to make an appointment with one of these subjects. And they sit at the table for about half an hour while he's going through all the costs and benefits of each appointment slot on each day and which would be the optimal appointment. So it was total analysis paralysis. So why is it important that emotion has a role to play in finance? Well, so humans are fundamentally social, as you were mentioning,

and they are also fundamentally emotional. So a lot of the discussion in economics and other disciplines in a way is upside down. We evolved from being mammals and before that, all the rest of evolution before that. And the brain, so to speak, evolved and then the cortex evolved to improve survival and reproducing etc so the emotions are as i are at the heart of the or at the bottom of the basal brain and everything that happens goes through this because this is as if you look at an animal in a mammal this business of fight not fight but approach or avoidance is is fundamental now the emotional system is built on top of that so for example you can have guilt or

envy or rivalry these these are complicated emotions, but they would link to, you know, if you feel guilty, you want to run away. If you feel shame, you want to avoid. If you feel excited, you want to go towards. So I think of the emotions, it's the simplest way of doing it, because otherwise, you just end up with an endless classification that people argue about in at the end of the day from a point of view of finance you want to buy hold or stick right and and so when you're buying that's approach when you're selling that's um uh avoidance and when you're sticking you've got some sort of balance balance on so if if you then think of what people are trying to do that they're trying to anticipate as you said at the beginning the future price of assets and the thing about financial

assets which i think is different than than you know buying a television is that is that their value is continuously debatable so with a television you know once you've bought it and taken it home you know that's pretty much it although if you don't like it you could sell it but it would tend to be quite a large loss and you probably just get on with it houses may be intermediate in this because a house you buy, but you live in it. So, you get all kinds of benefits from being in the house. So, if the house is going up or down in value, unless you're very much at the margin with your financing, you don't, and of course, if you've got derivatives and everything, they're all ultimately based on narratives about what the underlying, as you say, fundamental is. And at the same time, narratives about what other people think the fundamental is.

So that's the Keynesian beauty contest the keynesian beauty contest is is always there and the guesses about the fundamentals are all are always there and they're kind of they're two slightly different problems um uh and in um and you can get information relevant relevant to either so while you're doing this processing and thinking of what's going to happen to this or the other share, the other point is always relative to some other or to others you haven't really thought of. So all the time, what I noticed in the first interview I ever did with a fund manager is it's just masses and masses and masses and masses of data and opportunities, which you can't possibly consider it all. Now, of course, nowadays, many people use computer screening to try and limit what they look at, you know, which flags up things are supposed to be undervalued.

But of course, that's just introducing a computer narrative. That's exactly what it's doing. It's filtering things that you've asked it to pick out. So what people do is they create a narrative such as I'm this sort of investor. I'm looking for this sort of opportunity. As I describe in the book, you know, this worked for me or this worked for something else. Can I, by analogy, find something else which is similar? So then they try and become more expert in particular areas,

whether using quantitative techniques or fundamentalist techniques. And they ultimately create stories. Now, as you're creating that story, your brain is evaluating that story because you're the great thing about narrative is you're projecting the future. And you are what you're literally doing is putting your body out into the future and experiencing whether your body likes the feeling of being there or not. Right. So, you know, if you're successful or if you're lost or if you're feeling guilty or whatever, this ultimately is felt, so to speak, in the body. So metaphorically, that's what you're doing and your brain is operating like that. So, of course, there's a cognitive aspect, but all the time the cognitive aspect is mixed with an affective feeling.

So when you get the conviction to actually do it, what's got to happen is more approach of emotion has got to be generated than avoidance. So I could give you give you an example I've written about recently. There was one fund manager with his team who was interested in buying a restaurant company in China. And it all looked good on the figures and so forth. They felt the fundamentals looked good. They felt that other people were undervaluing this and so on. But they weren't quite sure. So all that was generating approach emotions. But they weren't quite the members of the team to go and stand outside

some of these restaurants and count the number of people going in and out, right? And what they found was that the number of people going in and out was inconsistent with the data they'd got. So this generated anxiety, i.e. avoidance of motion, so they didn't do it. So this also introduces a very important idea that I don't think is in the book because it was developed after. There are attractors, which is the things, you know, this is a great, fantastic object, blah, blah, blah. And there are what we call counter repellers so in in the case of this restaurant story um if they'd managed to to go and be detectives and the detective work had supported their idea it would have repelled the doubt its function is to repel the doubt so what i'm trying

to say is that behind all this apparently statistical stuff, which is very important, what it's ultimately generating is emotion. That is either avoidance of emotion or approach emotion. Dave, I want to drill into what a narrative actually is in the way that you use it. And let me begin by distinguishing what I view to be two different types of narratives, and then I'll get your thoughts. So, we've been speaking about Robert Shiller, and as you're aware, he's got a new book just out called Narrative Economics. And in that book, he gives a kind of dictionary definition of narrative so to quote uh he says it's a story or representation used to give an explanatory or justificatory account of a society

or period uh etc and he expands on this in in his uh uh book but also i'll quote from his 2017 presidential address to the American Economic Association. So, a narrative is a simple story or easily expressed explanation of events that many people want to bring up in conversation or on news or social media because it can be used to stimulate the concerns or emotions of others. So, it has this viral and human interest element as well um so so i'll give you i'll give you an example of what i consider to be a shiller narrative uh in relation to the sydney housing market and bob actually mentioned this example more broadly in my discussion with him we were talking about theories of asian buyers which has been a feature in housing markets throughout the Anglosphere.

But it's certainly been a feature in Australia. And I have some really interesting data here. So, people, there's huge talk in Sydney, even among my friends and, you know, very intelligent people who believe that Chinese buyers were acting as the marginal buyers driving up prices in Sydney and Melbourne for a very long time. But that's actually contrary to the research put out by economists and Australia's key economic institutions. So, for example, in 2014, the Reserve Bank of Australia conducted a study where they concluded, and I'll quote here, Dave, overall, the available data suggests that while foreign residential purchases change a bit from year to year, they have been relatively steady and fairly low as a share of turnover in the housing market in Australia, and hence are unlikely to have been the main driving factor behind the recent increases in prices in Australian capital cities.

Also from the Treasury in 2016, quoting again here, the majority of price growth experienced in recent times does not appear to be attributable to increased foreign demand. And they actually went on in this treasury report and quantified it, which is really interesting. They said, the increase in prices attributable to foreign investors is small when compared to the average quarterly increase in property prices of around $12,800 in Sydney and Melbourne during the study period. Now, across Sydney and Melbourne, the models which we consider to be the best specified indicate that for a typical postcode, foreign demand increases prices by between $80 and $122 on average in each quarter. So, the empirical data

show that the impact of foreign investors in the Sydney and Melbourne housing markets was limited. Now, compare that with a study conducted by Dallas Rogers and a few other academics in 2016, where they surveyed almost 900 Sydney residents. And they found that people almost believe the exact opposite when it came to the role of foreign investors. So, in answer to the question, based on your understanding of Sydney's housing market, what do you think determines house prices? And you could select up to three from a range of options. 64.4% of respondents picked foreign investors purchasing houses in Sydney, which was higher than any other factor. And 77.9% either disagreed or strongly disagreed with the statement, foreign investment has no impact or very small impact on Greater Sydney's housing market. So, what I'm saying is that the facts were suggesting that foreign investment didn't play a large role in price run-ups, but everyone thought that it did.

To me, that's an example of what Bob Schiller calls a narrative. It's a story that's used to justify economic events, which can kind of feed back into prices if people believe that that story, you know, is going to support or elevate prices going forward. And then it feels like there's a different type of narrative as well, which is more inward looking. And another academic, Chris Martin, who's at the University of New South Wales, wrote this interesting paper a few years ago on Australian property investor seminars. He started going to them before I did. And he looked at the narratives of self-development and personal growth that both gurus and attendants use to kind of conceptualize their journey as property investors. And he was

looking at kind of mapping Joseph Campbell's monomyth over these seminars with the guru as the mentor and the person would go on this journey of self-development through property investing. So, there's the Shiller narrative, which kind of explains or justifies fundamentals. And then there's this other narrative, which seems to orient someone's emotions towards the asset class. My question for you is, is that a fair fair distinction and how do you define narrative in the way that you use it so i think we have to unpack quite a few things to to consider this this this question okay so the first point is is that i think that that at one level, Schiller's thinking could be applied to this housing example you've given very well, as indeed to the dot-com thing.

And I think we're there talking about what I would call a shared story. I think it's critical to make it a shared story i think it's critical to to make it it's a shared story i did i think i talk about this at the end of my book in in when i formulate a theory of how markets work but it's it's a shared story so it's not it's not everyone making the same mistake at the same time by coincidence no i mean i think i think the way i would think of it is that for all these as he puts it, go viral. That is to say, for whatever reason which we can come into, people all start to believe that story. And when I say all, a lot of, you know, the group grows. And how that happens, I don't think we understand a lot about, but it has to do with how far a story to sell your housing company or whatever.

If that doesn't resonate with some of the prevailing explanatory models of people who you want to sell to, it's never going to fly. I mean, I've done work in health education which use this particular model. So this story, the kind of story that Schiller's talking about, I think this is a real phenomenon. And I think it's a major step forward that he is putting this forward. However, what I think is problematic about is what I've already said, which is, like all economists, he frames this inside an irrational, rational basis. And that is very misleading for the following reasons. And we do then have a problem with the word narrative being used in lots of different contexts.

But in the term that I use, narrative, narratives are a fundamental aspect of human cognition, which start very early and what humans use to make sense of the world and to talk about the sense they make of the world to other people. So, you know, the psychologist Jerome Bruner discusses this and how, you know, very young children, I don't know if you have children, but small children, you know, they go to school, they come back and tell you stories about it. And you ask them, you know, and so there's very little in life that is not put in story form. You know, where have you been today? How did you get on with David Tuckett? He'll tell some story. I mean, that's the way you, you know, and you obviously won't tell them everything we said so you will tell some story about what comes to you with being the key thing that that interested you or disgusted you or whatever in the encounter so so it's necessary to see uh narratives as the

fundamental way in which people organize their world and particularly their social world and communicate to each other right and for the reasons i've indicated earlier narratives always are have both a social uh element so particular narratives at any one time are normal within particular social groups and then they may spread out beyond that group and you know go to other groups so in the case of the you know foreigners are taking over the housing market kind of idea that idea appeals to set i would imagine this the sort of thing one we really need to understand much more about is that that appeals to certain kinds of narratives that tend to be around in a society, e.g. foreigners taking over. Yeah, there's a xenophobic undercurrent.

But it also links into some other deeper narratives about Australia being – so I call this the liberty of distance narrative, that Australia is this attractive safe haven for foreign capital. So there are – and that is what – That's numerous. Yeah, that is what Schiller would call a constellation of narratives. So new narratives make sense in the context of existing narratives. Yes. And I think the new – another point about the new narrative so the new narrative has to has to so to speak link up with enough old stuff or or prevailing beliefs or as i

call them explanatory models way people explain the world in the particular context where this narrative is being is being told it has to do enough of that to work. And of course, there's a very good study, which is in my book about the people who studied internet IPOs and, you know, people who added dot com and then later they took dot com away. And that made a huge difference to the valuation. Well, there the word dot com was a whole narrative, wasn't it? Which behind it, you know, meant all kinds of things. And it had a different meaning, though, in the run up of the boom to afterwards when, in fact, you mentioned dot com and people wouldn't buy it so that's why they took

away dot instead of it being walker.com it became walker and people were more prepared to buy you this is a very hard set of studies so there is all that but i would place that inside two two things so one is the the what i've said about narratives being the way we make sense of the world and also also the way we see the world unfolding in future so that's where i think schiller's right that there's always a causal explanation somewhere inside a narrative as there is inside a novel in fact the great novels play with different causal models and so on you know and there's english literature and stuff comes in so there's always a so it's a view of the future and it generates emotions that's fundamental to it and generating these either

exciting or anxiety producing feelings based on what I said earlier about emotion. Now, but all of this is necessary because of uncertainty. So it's precisely because there are no fundamental values and we don't know what the value of house prices is going to be or we don't know whether atomic fission is going to be a great success or whatever they all the other ideas are a fantastic object only operates in an area where we do not know got it now as we've discussed before there may be ways in which sensible analysis of something can cause doubt about some of these narratives. But the underlying fact is we don't know. this type of behavior, like generally behavioral economics does, as bits of behavior that are

dysfunctional compared to the way we think the world ought to work. Whereas I see these bits of behavior as fundamental coping strategies with the reality of life and uncertainty. And I see the models as simply telling us something about how the world could work and of interest in that sense. But as you probably also know, the famous models like the Arrow Debreu, which became treated as though they were facts, were actually worked out to demonstrate just how many bloody silly assumptions you'd have to make. But it's true, true right the same with

hicks's work the same with savage's work actually on bay you know savage actually has a phrase which i'm not going to get exactly right here but in which he says to apply this to large worlds actually call them grand worlds, not larger, is ridiculous. Yes, yes. So I raised two species of narratives and used examples from the Australian housing market. The first type of narrative was narratives about fundamentals and the example was people's theories about Asian buyers driving up prices. The second type of narrative was the way people view themselves

as property investors going through this journey, their kind of life story. And we've agreed that for the purpose of this conversation, it's really only the first type that is relevant. You and Sheila broadly agree on the importance and the effect of those type of narratives about fundamentals. You sometimes call them explanatory models, which is terminology that comes from anthropology. But where you and Schiller disagree is you dislike this kind of program of irrationality, which is his driving agenda behind his version of narrative economics.

Because it means that you don't see that these are not exogenous. They're still trying to treat them as essentially exogenous behavior as an exogenous variable. When in fact, behavior is endogenous to the entire situation. And this becomes particularly true when you think about the way innovation is treated in economics and in economic growth theory. Because actually, there would be no innovations. I mean, if you go back to your point about canes and animal spirits, there would be no innovation if people didn't go beyond what is known. And actually, the word intelligence, what it actually means is to go beyond the data. Yeah, yeah. Schiller actually draws an analogy between narratives and innovation

in his 2017 presidential address to the AEA. He says that we can kind of think of narratives like innovations and think of them as exogenous shocks in that sense. And I guess the other reason we should be concerned about whether we start labelling things as irrational is that once you do that, you open the door to the nudging state, which is a whole other conversation. Yeah. And I mean, there's no problem with nudging if there's a general agreement that whatever behavior you're trying to nudge is desirable.

So, I mean, for example, in the UK, nudge was used to, you know, they redesigned the tax form in a way that led to more people filling it in tristly but unless we think tax dodging is great that's a that's a non but you can't nudge people to do the right thing with investments you might i mean you can nudge them there is the nudging to make them put more money into their pension which which perhaps makes, if we're all agreed about that, that's fine. But you can't nudge people into making decisions under uncertainty because you don't know what the right decision is. Yeah.

As long as we're talking about investment decisions, nudging is not really relevant here. Yeah. Dave, finally, I want to turn more directly to the question of bubbles with you. And I thought first, before we speak about bubbles, we should go through three fascinating ideas in turn, ideas which I learned from you. And we'll get your definition, your explanation of each. And then finally, we'll tie them into a theory of bubbles. So the three ideas are fantastic objects, divided states, and group feel.

So let's start with fantastic objects. You've kind of mentioned this a couple of times already, but we haven't discussed it properly yet. And when we say fantastic objects, that's with a P-H. So P-H-A-N-T-A-S-T-I-C. So what is this notion of a fantastic object? Well, when I first looked at the dot-com bubble, and then when I look back at all the other bubbles from the South Sea bubble onwards, what you see is something very compelling.

Now, the word object in this is object as in philosophy. So it doesn't mean a physical object. It means an idea. It could be an idea, an object, whatever. So gold can be a fantastic object. A girlfriend can be a fantastic object. And advertisers basically try to create fantastic objects left, right and left. So it's a representation in your mind.

It's not a rock that you stub your foot on. No, exactly. It's a representation of your mind of something to which you attach implicitly a narrative of it being absolutely fantastic. And its origin is in the attachment system within human beings, where it works, for example, between a mother and a baby. The mother thinks the baby's fantastic, and the baby thinks the mother's fantastic. And as a result, you get this attachment. Later, it works, you know for for love affairs and partnerships and so on now initiate so you need to think in terms of a fantastic object relation where it's the idea in your mind relating you to the object to the idea a dot com a house project, a girlfriend or boyfriend.

In a fantastic object, the object, whatever it is, is the object is idealized. That is, it's made super wonderful, and we don't know how it works, but it must hit bits of the brain, et cetera, that really set off hormones and so on. It's part of the human reproductive system, ultimately. Now, a normal relationship to something under uncertainty, which is most of life, is that you sometimes feel this is fantastic or, shall we say, very good, and sometimes it isn't. And so there's bound to be up and down, you know, it's not the same all the time. So, for example, George Soros, I once gave this idea to George Soros, who you probably

know is a well-known investor. You interviewed him, right, for your research? I did. Well, no, not as part, no, I didn't. No, he wasn't one of the people I interviewed. He was one of the people I discussed it with. Mervyn King introduced me to him. But he's since become a friend of mine.

And I was presenting this idea of the fantastic object at the time of the Euro crisis, and he became interested in the idea of the euro as a fantastic object. So if you think of the history of the euro, the idea of the European Union was very much an idealized idea of people coming together, not fighting each other anymore in Europe and a wonderful thing. And it gradually developed when the European, they expanded. We got more people and more countries into it. It worked quite well. And then at a certain point, they came up with the idea of having the single currency, the euro currency, which again was seen as this wonderful, wonderful thing.

Now, that's fine to, you you know to get something as difficult as that between countries where some have to give things up and you have to share just as in any relationship you need probably some degree of idealization to get it going but what normally happens in a normal decent relationship is that idealization gives way to real experience so where you really let's say you're doing what you do you really it goes from just being an idea to being an experience which satisfies you in reality or sometimes frustrates you but you like increasingly from a real experience of you know mostly liking it or if it's not that you give it up now with the euro instead of going to becoming a real thing which would require the institutions and so forth that are needed to back it up and also

a recognition of what you have to give up if you want to have this. Like in a partnership, you have to give things up to achieve a mutual relationship. What happened is it becomes denigrated. So this is a typical thing. And by denigrated, it goes from being idealized to being the most disgusting thing you've ever thought of. And this is also what happens to many ideas. For example, dot com. So in that thing I mentioned already, first adding the name dot com, then later taking

it away, the dot com went from being an idealized fantastic object to being a denigrated object. Now, the denigrated object is no more realistic than the fantastic one. I can remember in my interviews interviewing people and they were discussing fund managers in 2007. Yeah, that's right. And they were discussing whether or not they would invest in Vodafone. And one of them said, well, we couldn't invest in Vodafone because the senior guy at the end of the table said, we've done Vodafone, it hurt us terribly, we're not going there again. So I'm sure this is true.

So the fantastic object captures this particular relationship people have to things in which they invest emotional attachment. Now, what's the test for a fantastic object? For example, how would I confirm or disconfirm whether property is a fantastic object for the 1.3 million negatively geared Australian property investors? Well, so in 2014, my colleagues and I published a paper, I'm not sure if you came across it, in a journal called Social Networks. And in this particular paper, we looked at two things, Enron and Fannie Mae. I won't talk about Enron for the moment.

But in regards to Fannie Mae, we were, as you know, Fannie Mae was at the heart of the whole mortgage business in the US. And we looked at all articles in Reuters News over a period, I think, between 2004 and 2012 or something, in which we scored the number of words that could be rated as evoking approach emotion and the number of words evoking avoidance emotion in any article in Reuters which mentioned Fannie Mae. And this is a large number of articles, so it was done with an algorithm to look at it. And what we find is a massive decrease relatively in the number of avoidance words compared to approach words as you go up to and you'd have to look for

absolutely decide but as you go to around the beginning of 2006 wow and then it continues on up at the same gradient until uh about i think it's march 2007 and then it falls precipitantly you know when the crisis began yeah now we also tracked onto there the case schiller housing index and also the the share price of fannie mae well the share price of fannie mae and this excitement index, as I recall, went pretty correlated. The interesting thing about the House index was that that started to level off and then fall in 2006. There were, in fact, I think it was 15 months in which the sentiment index carried on merrily, becoming more and more excited and no anxiety words, despite this, what was happening to house prices.

Now this, and what you were saying about the Sydney thing reminded me about this. So what was happening is, despite the availability into the market of real information that if anyone had thought about it would have caused you to be most concerned and you would have expected journalists to be writing lots of articles you know you you know logically once you got the fall you know there should have been lots of articles full of anxiety instead the excitement just continued right so this is what we would call and the paper's got that name a divided state in which the only information that's been taken in

is information which uh confirms your hypothesis yeah but unlike confirmation bias it's similar obviously but it's the emotion that matters, right? So information that generates the emotion of going forward, you know, yes, I still want to marry her sort of thing. Information to that, and it'll have a particularly accelerating effect because there's nothing you know going against it to stop then of course when you get to the top then there is a complete collapse because the only emotions around are negative and so a divided state could be divided in terms of approach emotion and no avoidance emotion or it could be the other way around so sometimes you may get for example situations which are evaluated solely in terms of anxiety rather than in a balanced point of view so all this comes back again to uncertainty and given that uncertainty

can evoke all the time the possibility of loss or the possibility of gain, the emotions around it are highly influential. So a divided state is one where you only, somehow you're only looking at part of the picture generated by whichever emotion is dominant. Now, in a divided state, can you waver between approach and anxiety? Well, no. Well, strictly, I mean, this is the conceptualization. In a divided state, of course, if information – so, for example, John Kay told me that. I mean, do you know John Kay? He's a very well-known economist so john used to often be on tv during the um the dot-com crisis and he

would be on with one of these people who was massively you know talking the stock market up and john would be being skeptical and what john noticed was that nothing he said made any difference the people simply didn't hear yeah right so in what happens in a divided state is that although the information is there it's not attended to or it doesn't compute now there's a psychoanalytic explanation for this because this is what we find when we when we're talking to patients right and it's also well anyway it makes sense psychoanalytically but this is how it works for individuals and what i've recently done which is relevant to what also asked me group feel is extend this idea from the individual

to the way an organization is set up so it would seem that leading into the financial crisis, for example, some organizations, particularly in the finance world, were set up so that the negative information, the information that should have been causing anxiety was going off in the opposite. It was there, but it was not being attended to, as the Fannie Mae example gives you. You remind me of a moment I had in Texas last year. I was in a very small country town and I went into a gun store just to look around. And I got talking to the attendant and he asked me where I was from i said australia and he started asking me about the gun

we we have quite stringent gun laws here and and in fact it was a right of center politician john howard who initiated a huge gun buyback campaign after the port arthur massacre uh and he said to me um yeah it was terrible the gun buyback program, because after that, your homicide rate went right up because obviously none of the criminals handed their guns back in. And while I was at uni, I worked for an economist who actually had studied this question, the effect of Australia's new gun laws. And I knew not only that the homicide rate fell, but also suicide rates fell because people had less weapons available to them. So I actually knew the data. And I just said to this guy,

no, no, that's not true. Actually, the homicide rate fell. And his eyes almost glazed over. It was like I was talking to a zombie. He did not want to absorb the information. In the context of the housing market, my pet topic, as you can probably tell, there's a big narrative in Australia, which is that we could never have a housing crash because Australia has recourse loans.

That is, you can't post your keys back to the bank. The bank can come after you for your other assets. And so everyone talks about this. Everyone talks about this. This is the big point of difference between Australia and the United States, for example. And there's a guy called Peter Switzer who is kind of like a finance media personality I think

he has a mortgage-broking business as well he has a runs a company called Switzer Media and he's a pretty bullish guy he talks a lot about optimism and how like belief in the housing market is you know we should be believing the authorities like the Reserve Bank who says nothing's wrong and we should be optimistic for our economy so So, there is this kind of group feel element to his public pronouncements, but he does speak about this narrative of recourse lending as well and how this is kind of like a protective force around the Australian housing market. I caught up with him earlier in 2019 and I just took him through the facts, which was that if anything, recourse lending should make you feel more concerned because it means that banks are relaxed about the risks. Every major property bust has featured recourse lending from Spain to Ireland.

I mean, Ireland didn't even have proper bankruptcy laws until 2013. Going back further, Japan, Finland, Sweden. But also, recourse lending is state-based legislation in America. And I think it was like, you know, 39 states were actually recourse in the US, something like that. You know, roughly the other half were non-recourse, but everyone had it in their heads that US was non-recourse. And recourse states in the US included, at the time of the subprime mortgage bust,

Florida and Nevada, which had among the highest foreclosure rates in America. And the reason is like pretty simple theoretically if you can't afford to repay your mortgage you can't afford to repay your mortgage and and some uh economists at it might have been it might have been the dallas fed uh did a study study where they showed the effect of recourse on foreclosures was like, it was very minimal in preventing them. It had some effect on strategic foreclosures. But again, if you can't repay your loan, you can't repay your loan. But this narrative has kind of taken hold. I took Switzer through the facts and just sort of explained this.

And again, it was like the eyes glazed over. It was like I was talking to a zombie. And I know I did not change his mind in any way. All of this information was one Google search away. The same for the people in the study you did in 2014, where news was available. You could check the case shiller index if you wanted but sentiment was still going up for a year and so this is what you call a divided state it is and and um and it it's quite a complicated business to understand how it's maintained

but one of the reason way reasons it's maintained is because it's very emotionally painful for this guy's switzer did he say yeah peter switzer the truth is emotionally painful and also would would would mean he had to do a lot you know you'd have to it's a bit like facing that your marriage is failing or, you know, any having to give up your beliefs if you've got really committed to them. And so you have to do what what we would call mourning. You know, so it's like if somebody dies, you at first it's very, very difficult, but you can you can mourn it insofar as you accept the reality, which is gradual. So it's unsurprising that when the information becomes available, it's not immediately reacted to. Yeah. And it's painful not only in an obvious physical or logistical sense, for example,

oh, great. This recourse lending thing is a myth now i'm gonna have to consider selling all of my investment properties it's also painful in a like an emotional and tribal sense um because if you if you change your belief then you risk alienating yourself from the original group and and that probably so we've spoken about fantastic objects, we've spoken about divided states, that probably brings us to group feel. Just before we do that, I think that, so, you know, Akalov, Shilla used to work with Akalov on narratives. Yeah.

And they're now slightly different. So I do some work with Akalov. And Akalov is much more interested in tying narratives to identity and things like that right which is what you just talked about because when when you have to make accept this new situation it involves changes in your identity changes in your feeling of your reputation or your actual reputation guilt shame all these guys there's a lot going yeah to keep your view that's that's part of the problem and so the australian property debate just briefly on that point is so ridiculously tribal bulls versus bears uh yeah but go on exactly so so i mean group so group think was identified by irving janice, as you know, in relation to – it's typically related to military disasters, the Bay of Pigs invasion and so on.

So somehow groups of people, decision makers, get themselves in a situation where they only relate to each other and they don't – the bigger picture doesn't come in right it actually is based on the work of this psychoanalyst wr beyond bion who who was in fact a tank commander in the first world war and was an exceptionally gifted guy who first came up with who did a lot of work on groups in the military and in and in other situations and what he noticed is the tendency of groups to do this inward-looking thing while they preoccupy themselves not with the task, what he called a work group, but became, just blocking on the word, they became a basic assumption group. That is a group where they're making some basic assumption which is away from the group, e.g. they're all there to please the leader or something like that. Or they must all be comfortable together.

So this is a profound psychological state which probably has its origins in the need for human groups to stick together, you know, many, many years in the past. And you don't necessarily want, you know, external influences or things like that. Anyway, group feel, I think, in effect, is the divided state passed from individuals to groups right and it becomes as you said uh if everyone else thinks this you don't really want to be different but remember that there are always some people in every group who in fact feel the opposite right there's some people you know you must have been in groups where someone makes a point of being difficult all the time and being so it's it's not like group think is an automatic thing it's a tendency right and of course um so but it's based the why related to why i call it group feel is that it's based not on the thoughts being the same.

It's based on feeling together. So, you know, think of a rugby game or I don't know what. You know, that feeling of being together is every bit as important as whatever cognitive reasoning there may be. And in that, yeah, I think there's one other thing. Yeah, so in relation to things like the housing bubble, you've got two things going on. You've got the fact that if you invest, or let's say the dot-com is better if you invest the shares actually go up if everybody's doing it so there's a there is a sort of logical feedback

and there's also this emotional feedback and these two things are very very powerful but what what is put out of action in these situations is curiosity. So you talked about those people, you know, when you talked in Texas, his eyes closing over. And that's exactly right. But you see, when I told you there was much earlier in this conversation, I said something to you about how when there's uncertainty, you really don't know. You didn't glaze over you did have a surprised reaction but you immediately looked curious that's the antidote curiosity where if you talk to george soros you will be astonished that he is able to be curious about and he's always interested in what's going on, right?

And he doesn't – so that is, I think, a fundamental characteristic which sort of goes against groupthink or divided states. So you look for – if you want to spot an organization or an individual that's in one of these divided states or groupthink, look at what happens when they're given new information Do they really? Have a look into it or do they dismiss it like I have a little story that might be of interest to you like your gun story I was actually in Australia in

2005 in the Elizabeth Murdoch lecture theatre at I suppose it was Melbourne University. And I gave this, it was the first time I ever spoke publicly about fantastic objects and divided states. And at the back of the hall, there was a very well-known Australian journalist, I can get you his name, who was a journalist for the Melbourne Age. And he put his hand up and he said he'd just come from some kind of investors meeting for an Australian mining company. And this was a mining company that had claimed some years before to have made some, you know, great strike of some minerals or whatever it was. And when questions were asked of this CEO,

he'd been extremely dismissive and had not engaged with the questions at all. And so he said, is that an example of a divided state? So not wanting to be sued i answered saying well i can't i can't uh comment on that particular situation but actually that's exactly what i'm talking about well two years ago two years ago i got an email from that journalist telling me by the way that company's just blown up it never was there wow but that also talks to your point about time he managed to get away with it for a long time yeah yeah it's probably a company that was uh short sold by one of my friends and and guests of the podcast john hempton who's a legendary short seller well so you know that's that's the thing that will be a good yeah yeah so dave i have this quirky little theory which is not statistically significant at all but let me speculate with you

and it's a nice counterpoint to this idea of group feel and it is this i've observed that in housing bubbles and i feel like the group feel is particularly intense with housing bubbles because it's an asset that's widely owned. And, you know, for most people, it comprises the majority of their net worth. So there is quite intense conformity in housing bubbles. And what I've noticed is that the people who, because calling out a bubble or being bearish on a bubble is an inherently contrarian position, therefore. Yeah. Many of the people who do it or do it successfully are either highly disagreeable individuals or are somewhere on the spectrum or have mild Asperger's.

I said this wasn't statistically significant, but there's a couple of examples. Earlier, you referred to Dr. Michael Burry from the Big Short fame. He self-diagnosed with Asperger's. There's also my friend Jonathan Tepper, who is the founder of Variant Perception. He has this reputation as being like the housing bubble whisperer. He grew up in San Blas, a suburb of Madrid, and then went to university in America, was a Rhodes Scholar at Oxford. And so he'd seen Spain, America and England. And he argued that Spain was in a housing bubble and America was in a housing bubble before they burst. And this year he went to a psychologist who told him that he had Asperger's and he told his friends and they said, you paid someone to tell you that um jonathan is also very famous down under for calling out the the recent phase

of the australian housing bubble which ran from about 2012 to 2017 um and he described his experience of coming down here in 2016 which is when he published his famous report with the short seller john hempton the report was called i know a guy who can get things done he described that experience as like slaughtering a cow on the streets of india the short seller John Hempton. The report was called I Know a Guy Who Can Get Things Done. He described that experience as like slaughtering a cow on the streets of India because the reaction was so vociferous. There's actually another gentleman who has broken the cardinal sin of being a foreigner and visiting Australia, and that's Vernon Smith, the Nobel laureate.

He was out in Australia in 2015 and said that Sydney and Melbourne looked like a pretty good bubble. And Vernon, by the way, self-diagnosed with Asperger's in 2005. I just think it's kind of funny that a lot of the people who, you know, there is this group feel in bubbles and then some of the individuals calling it out often have um some form of asperger's because they're not they're not so prone to to you know the social cues and and the group feel because of their asperger's well i i would say that so there's two things here there's to call it there's to call it out which if you read Galbraith's book on whatever, essentially on bubbles.

A Short History of Financial Euphoria. Yeah, I mean, he talks about the various, you know, that one of the features is there are always people calling it out at every stage. Yes. So that's, again, why the issue is not that suddenly someone calls it out. It's always been calling it and it's the issue is not that suddenly someone calls it out it's always been calling it and it's always being ignored that's yeah now to call it out i think you have got to be an outsider in some sense so that could be mentally outside or socially outside or

nationally outside yeah um uh and then if we move from just calling it to actually being willing to take a large financial position against it, then you're only going to be able to do that if you can manage the anxiety. Now that you may be able to manage the anxiety because as someone with Asperger's, you know, there's some actual reason that you don't get, you know, some physical reason you don't get anxious and so you can carry on. Or it may be that you can be anxious in, you can be in an integrated state, which I think, for example, I would say Soros was, and you can know that you're anxious. I mean, he talked to me about getting backache and stuff like that, etc. So he takes that as a sign that he should think a little bit more

about his investment. Is that right? He took it as a sign that he didn't understand what was going on and therefore he should go back to not taking you know not taking risks and do you call that superstition or he was taking when he was taking big risk he he he didn't feel that necessarily feel uncertainty then he just and anxiety then he carried on i would call that yes he turned he turned that into a heuristic absolutely okay And but I think that so I think that your theory about the and the last thing I'd say, of course, is are lots of people with that condition who wouldn't necessarily be any good at investing at all, as the as a Damasio type of example shows. So so I think the important thing is to focus on how do people manage to maintain their view in the absence of confirmation.

And this would apply also to someone like Steve Jobs or someone who's, you know, running a company. I mean, animal spirits is part of being able to depart from the herd, isn't it? So I don't think it's just a question of, you know, just, so to speak, being non-responsive. Got it. So we've spoken about fantastic objects, divided states and group feel. Before we speak about integrating these ideas into a view of bubbles, I just want to jump back very quickly to fantastic objects because the average person might not be able to do you know some natural language

processing to to quantify whether something has become a a natural has become a fantastic object in in the zeitgeist or in the markets but but there is an interesting little line from page 106 of your book, which might even be a heuristic for whether something is a fantastic object or not. And you say, quote, fantastic object stories are about exceptional opportunities to gain reward at much lower risk than usual for such gains, end quote. So for the average person who's trying to deduce whether something is seen as a fantastic object that might be a good little test

for those of us trying to work out whether the market is viewing some particular asset whether it's a dot-com stock or a tulip as a fantastic object we we might be able to think about, we might be able to ascertain whether the speculators are viewing it as an exceptional opportunity to gain reward at much lower risk than usual for such gains, even if we don't have a lot of resources to determine that. No, I think you're right. I think the sort of under uncertainty, you would expect a certain amount of volatility because you can't quite decide today you think differently. If you get, to tie these concepts together, what is your story or your anatomy for a bubble? I think a bubble is a moment when a shared narrative

about some new exceptional opportunity begins to get beyond the immediate group concerned with it, e.g. geeks interested in the Internet, begins to go out so as it becomes part of general discussion and kind of lots of people are talking about it at that point the narrative is likely to be uh more exciting with the anxiety getting a bit left out left out so the the underlying sign of a bubble is lack of anxiety. Does that mean we can ex-ante-identify bubbles and crashes? Well, that's what I've been working on with the Bank of England. I mean, I think that you can't tell which thing is going to be the next fantastic object I think it's very hard to tell that and to distinguish a fantastic object from a truly great

innovation but I do think that if such a thing is generating persistently excitement without anxiety given what I've said about uncertainty that is grounds for very forensic analysis the post-war era has been the era of national housing bubbles which i think is has coincided with financial liberalization, but also the creation and publication of national house price indices, which only really began around the 1970s, which has enabled the media to report on house price rises. House price bubbles are the worst type of bubble. They're macroeconomically devastating.

They ruin balance sheets. They're associated with construction busts. Do you think how – I wonder how much you've considered housing markets specifically. And do you think there's something in housing that lends itself to being a fantastic object, perhaps more than other asset classes? Because houses are very tangible and you have these narratives like bricks and mortar is the safest investment, almost invoking the corporate reality of housing to support the investment thesis. So do you think there's anything inherently in housing as an asset class that makes it more liable to

be viewed as a fantastic object well so it's it's a it's a bubble you're talking about an actual bubble in house prices or a bubble in a bubble in you know derivatives etc around it because i think there's a difference i'm talking about a bubble in house prices. Right. So I think in order to say that a house is a fantastic object, you need to think a bit carefully about the theory that I'm putting forward because houses aren't exactly new. So what is new about it?

And what must be new about it is some idea connected to the housing, you know, like the one that they're going to go on valuing. They're going to be worth billions in whatever year it is. Because otherwise, I think you've got to have this newness because the newness creates the uncertainty because if it was as simple as we might like to think banks wouldn't be lending money people you know etc etc so so there has to be some element of newness um and then i think perhaps one of the issues with with housebuys bubbles is the feeling that develops that you can't lose so the banks lend to people I mean basically banking has become about

house mortgage finance and they hardly ever lend to most of the banks don't lend to businesses at all now, you know, not to startups and the real economy doesn't get much. So I think there's a there's a kind of collusion around the idea that you can't lose and it's that can't lose things. So, in other words, it's not so much the excitement that goes with a, you a new invention which which is the which is the excitement of the fantasy it's more sort of can't lose and we can live in it and this you know then maybe the idea of a home is a bit fantastic you know home as something in that sort i think that will be um but i know that you know when i hear my wife for example talk you know when she was you know wanting to move house or buy flats and things there's a tendency for people to feel you can't lose they don't go down yeah and so it's the lack

of knowledge about them going down which is partly mitigated by the fact you know that they've got in i live in london and the houses have gone up and down uh but when it's going down doesn't matter to us at all because we're not thinking of selling now you know if you're if you're in a situation um and it's the same old house, but when it's going up, maybe it's more, you know, you get the news, you get excited, etc. But I mean, I know that during the American thing in that one, you know, what was happening was that people's mates were buying second houses and third houses. And so it went beyond just, you know, buying your own flat to live in. Yeah. Well, that certainly happened in Australia.

So, you know, we're a nation of 25 million people. And according to the latest tax data, there are 1.3 million Australians who own at least one negatively geared investment property. That is, they're actively submitting to lose money in a property investment. You get some back in tax, but it's generally not enough to fully compensate the losses. And that by definition means that they're speculating, I think, speculating on the capital appreciation. It's about 300,000 Australians with two or more negatively geared investment properties. So I think you probably need to think about, you know, so if you're buying a house in Sydney and you're planning

on living in Sydney for the next 20 years, then it probably doesn't matter if you pay buying a house in sydney and you're planning on living sydney for the next 20 years then it probably doesn't matter if you pay too much depending how much do you buy because okay it goes down but it doesn't really matter um if on the other hand you're buying a second home or you've got an investment idea and you want to be able to sell that in the next few years so i don't know if you can split the market like that but i would tend to think you know i would think the dynamics are different depending which and and look at this um i'll share my screen with you just as a brief aside uh so this this chart can you see my screen? So this chart.

Oh, yeah, I got it. This is from Philip Lowe, the governor of the Reserve Bank. This is from a speech he gave. I think it was in March. So the green line is generally interpreted to be owner-occupiers. Now is a good time to buy a house. The lilac line is generally interpreted to be investors. Real estate is the wisest place for savings.

So we had price falls beginning around mid-2017 and prices started rebounding around mid-2019. So there's that price belief feedback for investors during the bust and at the same time, owner-occupiers think this is a rare mid-cycle opportunity. So the owners obviously think this is a cheap time to buy a house, don't they? Yeah, exactly. So I don't think we're in an unstable situation at the moment. No. Now, so Amir Sufi from the University of Chicago and Atif Mayan from Princeton have some new research showing that house price run-ups in the United States during, so this must have been 2005, 2006, speculators were getting more and more optimistic about price rises.

Meanwhile, expectations about house prices growth among the broader population were becoming more pessimistic so there was this the broader population were kind of raising their collective eyebrow at this core clutch of groupies who are still speculating on the asset until the so that seems to kind of run against the popular vision of a bubble as this sort of mania that grips an entire society till the bitter end. Yeah, no, I think it does. Yeah, absolutely. Does that accord with your work? Well, I think I've only studied the big bubbles, right, which are the ones which do extend over sectors and groups. But I think the issue, so I mean, that sounds like an investor group behaving like a basic assumption group.

So if it's a group of speculators, they have somehow walled themselves off from what everybody else is thinking. And they just listen to each other. That sounds to me. I mean, it is a very interesting paper that I could send you, but it's a totally different subject. It's an interview with people who were in, I think it was called the red brigade right uh in in italy and it's it's an interview with one of them in prison about you know what they did and one of the interesting things about it is that if you're a terrorist you obviously have to have to restrict who else you talk to and you you eventually get into more and more secure settings

where the only people you talk to are other people like you. And these people apparently really believed, like, that the workers were on the verge of, you know, everyone striking against the state. So they were able to maintain totally mad beliefs as true because the only people they talked about. And there is a famous story about traders and sardines. You must have heard that one.

I don't think so. Okay, so I'm not sure where I got this. But there's a story about, so traders who are trading, you know, forward prices for sardines. And at some point, the price of sardines goes into some bubble. And at that point, one of the traders actually eats a tin of sardines and says they're disgusting. And the other one says, look, these are trader sardines, not edible sardines. Apparently the real story, and which gets a bit to the same point. And certainly that's the kind of thing they were saying about, you know, POS, piece of shit about dot com, things like that.

So I'm sure that's a phenomena that people get into this excited in-group thing where they only they don't see the rest of reality. And that would be, you know, you call that a group feel or a DS state. Feedback is not working, and that will be for emotional reasons in part. Yeah. Finally, Dave, if there was one thing you would like central banks and economic policymakers to take away from your research uh what what would it be well i think it it would be to recognize that walking about talking you know to real people whether it's real people in the markets or real house buyers and understanding how they think and make decisions is actually the only reliable way

you're going to be able to make forward-looking policy so i mean i'm doing some work at the moment with the bank of england's agents do you know that so that the bank of england has 12 regional agencies which in and in each one there's two or three people whose job is to go out and talk to the businesses in that part of the country they go talking to businesses they do do 9,000 of these contact businesses. They call them every year. And I and a group of time they have one of their meetings about interest rates as a survey of business conditions or something. It's a rather similar thing at the Fed. They have something called the Beige Book, which is a similar thing.

Now, the fact is that there's quite a lot of evidence that these people know what's going on and even that they know more what's going on than they themselves realize but because of the rather dysfunctional set of theories that are called macroeconomics they don't they what they've what how they use the information themselves and how they pass it up is quite um is not so good i mean i've got a i did a talk on this which is on twitter um i can send you the link if you want on the preliminary work on it but the whole point the whole point here is that what these um agents are picking up around the country are the narratives the business plan narratives that businesses actually have and one of the amazing things about businesses is they're very quick

you know things happen and they quickly adjust so for example the the agency picked up before anyone else did the use of foreign workers in the UK economy, before anyone had realized, and before Brexit, right? And, you know, you could then think about what the effect of this would be. And there are other things. They picked up that small businesses were being starved by banks in 2009. And because if you're not finding workers, or if you're having to pay people too much or whatever businesses are quickly adjusting so what you're getting is real-time information

whereas what the uh npc are getting is statistical series based on statistics which are usually about 18 months lagging because we don't even know what GDP, you know, what is GDP today is actually, you can have estimates in three months time of what it is today. But those estimates are often revised. And there have been quite a few occasions and probably happened in Australia where the bank has thought there was inflation and put rates up, when in fact the economy is beginning to go into decline. So this all goes back to what I said about uncertainty models, because these models are all extrapolations of a fairly stable system. And when the system's stable, they do quite well.

But as soon as anything new is happening, they break down. So if you want to find out about what's going on new, you need to go out and see what's going on. And you also need to be willing to be a bit more taking risks. So, for example, obviously, you or i wouldn't want to base economic policy on one conversation with one businessman but it's better than a survey i would argue because after you've had one conversation you can put the word out to the rest of the agents is this happening where you are right and then And then having got the, because the biggest problem with dealing with uncertainty

is getting the new idea. Once you've got the idea, you can test whether you find it elsewhere. So listen to the narratives. Exactly. Dave, I think you're doing some of the most fascinating work in finance at the moment. And I've certainly learned a lot from

reading you and speaking with you. So glad I could share your work and thank you so much for your time. A pleasure. Thank you very much. Thank you so much for listening. If you got this far, you will recall that we discussed a lot of books and sources during that conversation for links to everything discussed you'll find those on my website www.josephnoelwalker.com that's my full name j-o-s-e-p-h-n-o-e-l-w-a-l-k-e-r.com you can also find me on twitter my handle there is at joseph n walker And if you enjoyed this conversation, please leave a rating and a review on iTunes. It really helps. I really appreciate it. Thank you for your time. Thank you for everything. Until next time. Ciao.