The Joe Walker Podcast - Rational Minds Part 3: Rethinking Bubbles - Vernon Smith

Episode Date: December 19, 2020

Vernon Smith won the Nobel Prize in Economic Sciences in 2002. This is his second appearance on the show.See omnystudio.com/listener for privacy information.

Men, wrote the Scottish journalist Charles Mackay, think in herds. It will be seen that they go mad in herds, while they only recover their senses slowly and one by one. My name is Joe Walker and three years ago, trying to understand Australia's obsession with residential real estate, I began researching housing bubbles. I read almost everything I could find. Increasingly, I came to view the question of bubbles as not just a topic of interest to the fortunes of my country, but a vehicle to explore deep questions of human nature. There's a long literature that gleefully lays bare the madness of crowds, from Mackay to Galbraith, Minsky and Kindleberger to Schiller and Chancellor. But it left me with a nagging question. No one in a bubble ever thought she was crazy. So what is going on here?

In this series, I'm using the prism of financial bubbles to tackle an eternal question. What does it mean to be a rational person? I'll be guided by five world experts who will show me that we're not quite so befuddled as popular narratives would have us believe. I'm inviting you to come with me on this journey, to reconsider what you might have been told, and to give rational minds a second chance. I used to be kind of sane. I used to act kind of normal. I couldn't complain. Now it's never the same. This series is proudly sponsored by Blinkist, an app that condenses the key insights from non-fiction books into 15-minute text and audio explainers.

The ad read for this episode says that I should focus on some books that I've been enjoying on Blinkist recently. But instead, I'm going to start with a book that I did not enjoy, and that was The Behavioral Investor by Daniel Crosby. on Blinkist recently. But instead, I'm going to start with a book that I did not enjoy, and that was The Behavioral Investor by Daniel Crosby. To quote one of its blinks, you're not as rational as you think. Of course, the whole point of this podcast series is to question what it really means to be rational and to show you that you're more rational than behavioral economists would have you believe. But that's the beauty of Blinkist. It helps me screen books that I agree with, that I disagree with, that I'm indifferent to, that I don't understand, that I want to learn more from, so that I can read or

hear their key ideas before I've even bought the book. Okay, so on to some books that I've actually enjoyed. I've been listening to Blinks for Apollo's Arrow by Nicholas Christakis, Other People's Money by John Kay, Fishing for Fools by Schiller and Akerlof, but the one that I want to focus on is perhaps the most challenging and therefore interesting, and that is Immanuel Kant's Critique of Pure Reason. Now, this is notoriously heavy going. Reading the Critique of Pure Reason can feel like running headfirst into a brick wall. So I was curious to see how it translated into Blinkist's Blinks, and they did an outstanding job. Now, a life hack that I learned from a very smart friend was to buy the second edition of the

Critique of Pure Reason. In the preface, Kant has his Copernicus moment. In a few short pages, he synthesizes rationalism and empiricism. And if you can read just that, if you can get that under your belt, you'll be doing very well. So what I advise is combining the preface to the second edition of the Critique of Pure Reason with Blinkist's blinks that condense the key insights from the entire opus. If you can do those two things, you'll be going a long way to understanding an infamously indecipherable text. So if you want to read all of these blinks and more, you can sign up at blinkus.com slash swagman.

You're listening to the Jolly Swagman Podcast. Here's your host, Jo Walker. Ladies and gentlemen, boys and girls, swagmen and swagettes, welcome to part three of this five-part series on rational minds. Time to take stock. This series does not deny the existence of bubbles in asset prices. Far from it, as regular listeners should expect. Instead, it mounts a counter-attack against the madness of crowds literature, against the notion that bubbles are

manias driven by lunatics. But wait a minute, I hear you say, how can you have a bubble without irrationality? I get the appeal of thinking of bubbles as extraordinary popular delusions. It makes us feel clever. We feel as if we're in possession of a great secret that only the smart can possess. That somehow we, a beacon of calm contrarianism, have levitated above the swirling mass of lemmings. I get it because I was once like that. But now I question whether it's really true. As I was studying housing bubbles and reading the stories investors told at the time, it became increasingly difficult to think of the participants as idiots or insane. It's easy to point the finger ex post and laugh at those who got it wrong. But is it fair? After all,

bubbles are booms that went bad, but not all booms go bust. This episode and the next episode in the series attempt to bridge the seeming chasm between the existence of bubbles on the one hand and the claim that the Madness of Crowds story is wrong on the other. The basic argument of this episode is that when people react to each other, whether cynically or naively, they can form bubbles together in a way that can be described as individually rational but collectively disastrous. To be sure, there's plenty of room for error at the individual level. Homo economicus remains a myth. But it's time to rethink bubbles. Because while stories of irrationality in financial markets are sexy and salacious, the truth is perhaps more disturbing. Bubbles are the inevitable outgrowths of collections of people trying to cope in an irreducibly complex world.

And none of us is entirely immune. My guest for this episode is a man who has thought about bubbles more than most. Vernon Smith is the father of experimental economics and spent a career generating and studying bubbles in his lab. In 2002, Vernon shared the Nobel Prize in Economics with Daniel Kahneman. Vernon, by the way, was 93 years old when we recorded this. He remains an active scholar. Indeed, before and after recording, he was telling me about a manuscript for a new book that he's currently working on. I admire Vernon Smith. If you want to learn more about Vernon and his life, you can listen to the first conversation I recorded with him, which is episode number 99. Swagman and Swagettes,

without much further ado, here is Vernon Smith. Vernon Smith, welcome back to the podcast. Thank you. It's great to be here. It's been a while and it's so good to see you again. This time we are going to dig a little deeper into the topic of rationality. And I think you're someone who is uniquely placed to help me learn about whether we are rational, what rationality even means. And in that vein, I was hoping firstly to ask you, Vernon, what is rationality in economics? Well, rationality in economics, as it has come to us, down to us since the 1870s, is simply Max Hugh. You know, people are represented by utility functions defined over the actions they take. For example, quantities of goods they buy or quantities of goods they sell.

And they're out to maximize that utility function. But the problem, and that whole theory is kind of interesting, because if you look at what Jevons and Volros, the originators of those models, were talking about, they were already imposing on demand the idea that there's only one, the idea of a law of one price in a market because they're taking prices as given and people are responding choosing optimal quantities. So they're defining demand and supply in a way that's sort of anticipating what you ought to be able to predict, you see, without having to impose it on the theory.

You see, you go to market and you don't know whether there's going to be one price in the market. And in fact, the typical market is not at first. It takes a while to settle down. So I think there was an approach to rationality that had a lot of appeal because now we were going to ground rationality where it belonged in individuals, because you see everything comes from individuals, right? All groups. It seemed like the right, you know, the thing to do. Well, but if you look at the way the classical economists thought about how prices form in a market, you see a rather different picture. You see, Adam Smith, the way he modeled buyers is, well, they come to market, and you know

what? They have a maximum willingness to pay for sugar or olives or cereal, whatever it is they're buying. And what's the evidence? Well, the best evidence is in an auction where a bunch of people go to, buyers go to an auction, and it's an English auction, and prices are bid up. And you'll notice at the beginning, there's lots and lots of buyers in there raising prices. Well, as the price gets higher and higher, that thins out.

Finally, there's only one buyer, and that's the person who buys the item. Well you see people are dropping out because as Smith would say that the price is exceeding their maximum willingness to pay. So he's grounding you see demand in this notion that that buyers have a have a cutoff value above which they're not going to buy. And they're in there trying to buy cheaper than the value. They're always trying to buy cheap. Sellers are trying to sell dear. So that's the way he thinks about a market, you see, as going forward.

And where do prices come from? They come from the group. You know, it's buyers and sellers, higgling and bargaining by different rules under different institutions that make prices. And, you know, you might say, well, wait a minute, I don't see much variability in price in the supermarket. I go in there and I see a bunch of posted prices. Well, but suppose you're standing in line and there comes an announcement over the public address system that milk is being discounted by 25%. Now, why is the market doing it? Why is the manager doing that? He realizes he's got inventory buildup in his milk,

the inventory of milk, and he needs to clean it out. So he cuts the price. And so if Adam Smith were standing by you, he would observe that some people that didn't have a quart of milk in their basket would go get one. So some who have one would go get a second. So he's carefully observing, you see, what people do and inferring from that, you see, what's going on. Well, that's the process whereby prices, you see, get made in a market. And that was the classical view. And that somehow got mislaid, you see, in the marginal utility revolution when we had all of this calculus for deriving demand and supply curves. And that classical view whereby prices emerge out of human interactions,

but not by conscious human design, that is what you would call ecological rationality? Yes. And yes, that is. It's the, you know, it's the group. It's still bid auctions. You go into most retail stores and it's a posted price. It's kind of take it or leave it posted pricing. Well, that's actually in economic history, that's fairly new. You see, if you go back before about 1875, 70 to eight, 1870 to 80, you started to have the retail mass retailing start to come in.

Before that you, you did all your buying in the general store. There was a manager. He had all of the different products and a lot of the pricing was negotiation because you're buying in bulk perhaps or something like this. So you have some buying power and so there were negotiations. Somebody else may have less. So that was the way it worked. But then with mass retailing, you see you now had department stores,

and the people who are vending the goods are just minimum wage clerks. Well, you don't expect them to negotiate prices. So what do you do? You have to have a price policy. So centrally, management decides, here's all the prices of these goods, and the clerk simply vends the goods. They don't do any price making. Well, but that makes it, but then that you're, what are you going to do about all your mistakes? Well, what we have is the end of the month clearance sales. You

see, we have, we have a practice and an institution that comes out of that system. In addition to F.W. Woolworth and R.H. Macy in the department store business, you had Sears Roebuck, you had catalog sales. Well, you print a catalog once a year, and you've got to say what the prices are. And they're fixed. Well, you make mistakes. So what did Sears always do in the spring? They put out a sale catalog.

The sale catalog is correcting all their mistakes. And you can kind of just see how with innovation, and certainly mass retailing was a great innovation, you see. With innovation, it starts to change how people interact and produce prices. So the institutions change. But that's all coming out of collectives and their interaction. That's just not – individuals, of course, are always in there trying to do the best for themselves,

but how do you do that when else knowing something about the boundaries of others? And you see, and it's the boundaries of each other that determine partly what we do. So it's kind of inevitably a collective process, even though it's completely decentralized. You see, there's no external intervention in this, in the typical kind of free market I'm imagining. The only restriction is property. You know, a seller can't represent something

other than what it is. Otherwise, the seller's in trouble. Either buyers won't come back to him or we now have a truth in advertising legislation and this sort of thing. And so we have so and also buyers can't steal.

If they take something, they've got to pay for it. And so it works on both sides. And so except for those kinds of thou shalt nots that are very, very ancient in origin, we are pretty free to interact. And that's different today than it was 100 years ago or 200 years ago. The great German psychologist Gerd Gigerenzer says that a heuristic is ecologically rational to the degree that it is adapted to the structure of an environment. How should we think about your use of the term ecological rationality in relation to Gerd's use of the term? The rules of the institution have to adapt to their environment.

So in that general store, you had a different price-making process than you had after Sears Roebuck got into the retail business and after R.H. Macy got into the retail business. And so that evolved. And you see, people today would find it strange that you would negotiate over over almost everything, that that might be something you would do but but it's still around you see you see in big ticket items you see if you're buying an automobile why most everybody knows you just don't go in and write a check for the list price you know you go in and the first thing you ask the guy what i do is i said well this is a cash deal. What's your cash price? Say, well, right away, he's giving you a different number. You see?

And so you say, well, that's too much. Well, what do you offer me? And also expensive appliances. There's still some of that goes on in the in the stores so you and gad are essentially talking about the same concept but just at different scales yes uh guard is talking about individual decision making and a lot of those are motivated by games of uncertainty, you know, a lot of them come from gambling games, okay, and sport games and this sort of thing where there may be some skill involved is not strictly a matter of, of probabilities. And,

and so that's what Garrett is talking about. And I, I quite agree with him. You see people, people find rules to operate because they can't, they, they find kind of good enough rules to get by and modify them if they find it's not serving them well. See, if I've got a rule and I find it's not working for me, I try another one. And so people, you know, they use trial and error to find something that they're better satisfied with. How should we think about speculative bubbles in asset prices? Well, of course, the first two biggest shocks in my life,

my experimental life, were first the discovery that markets for ordinary supply and clear to everyone and and then what we were expecting is people would trade at that fundamental value and then we would see if we could create create bubbles well our baseline bubbled. And we thought, well, what's going on here? It turns out, and people, it was an environment where

each person had an endowment of shares, shares and cash. The shares gave you a right to a dividend draw from a distribution at the end of each period. It was a simple four point distribution and we computed the expected value for people so that they could see what that was. And so here's the expected value for people so that they could see what that was.

And so here's the expected value per period. Well, you go into a market and they're going to trade for 15 periods. So what's the fundamental value of the first period? Well, it's 15 times the expected one period dividend value. And at the end of that period, you get a draw and you get a dividend realization. Then there's 14 periods. The next period, it should be 14 times. So we just explained this to everyone. We want to be sure everyone understood it. Well, the market didn't come off the stops at that level at all. In fact, these markets tend to start low, grow over time, go through the declining fundamental value and, and, and peak out maybe a few periods before the end and then crash.

And that, that was it. And so we just, this baffled us because, but what, what we learned from that is that you see, you can have common information on what something is supposed to be worth but that's not necessarily part of your experience you see and and you had people in there that saw the price rising and so they bought more and and the faster the price rise rose they were buying even more so these were we identified what we called trend-based or momentum uh traders in that market

and they were just they were just looking at price changes they weren't looking at dividends and then and then others they tended to buy below fundamental value and sell above. Well, they were kind of look more like the kind of rationalists we had expected everyone to be. But were the momentum traders irrational? You see a price trend. Well, are you going to get in on the trend or are you going to sit there like a dumb stick and do nothing so a lot of people get in there you see and we know it's not sustainable it can't last

you see it's got to turn around sometime and they all do well people are all playing that game that guessing game that is some of them you'll say so so that's kind of the anatomy of that bubble you got with inexperienced first-time subjects we brought them we'd bring them back a second time, put them in exactly the same experiment. Okay, what happened? We don't, the bubble doesn't go away, we just get a smaller bubble. And also the volume is less, we get a reduced volume, a lower amplitude price bubble. Okay. Bring them back a third time.

Now they're tending to follow fundamental value much more closely. So they get there. See, they get there from what they actually experience and what others do. That's what finally brings them in line, you see. And so in a sense, what's happening in the bubble experiments is kind of what happened in the first experiments, except it takes a lot longer for people to find equilibrium, you see. And well, if you're interested in trying to squelch housing bubbles, that's not very helpful. You say, well, people are learning, but they're learning too late, you see. Presumably, they're learning something

about the next bubble, but that doesn't help in this one. And of course people seem to have memories that fade over time so we get recurrent house bubbles, we get recurrent stock market bubbles and the problem with housing bubbles is that they're, you know, they're largely as, as Adam Smith would put it, he wasn't talking about housing, he was talking about the South Sea stock bubble. is too much of other people's money. Well, that applies to houses because we tend to buy them with a lot of other people's money, not a lot of our own money. And so that, see the difference between a market for lettuce and a market for houses is that you go to the lettuce market

and you buy it to eat it. You don't buy it to resell it. And no one is buying lettuce to resell it. Same thing with haircuts and hamburgers and hotel rooms. Most of the things we buy in consumer markets, we're buying them, using them, we're not buying them to resell them. So you can't get a difference between resale market value and use value. They're bound together, they're locked together, you see. But in something like housing, you see, or shares of stock, the resale value can

get untied from the dividend or use or basic value, you see. Homes have a shelter value. They have a rental value, you see. But they also have a resale value. So some people can buy them for resale, not necessarily to live in them. And if mortgage funds are plentiful and prices have started to move, then that can be a self-fulfilling kind of expectation,

very much like what happened in our simple laboratory asset trading experiment. I agree. I think the most important feature of asset price bubbles is that the items can be retraded. And especially if they have long lives like homes, which last, you know, 75 to 100 years on average, that means that the market value can be influenced by future price expectations of the market participants. So, you open the door to these expectations about the future. And I just want to hold that thought for a second. You mentioned the experiments that you ran where market participants were inexperienced,

but then after a few experiments, they learned. I understand there are more recent experiments which show that you can reignite bubbles even with experienced traders. Yes. Yes. In fact, we, I and Dave Porter and my co-authors did some experiments where what we did, we took people that had, let's see, they've been, they were all twice experienced in previous experiments. So we brought them back for an experiment in which we really, we gave them more cash. We gave them a sweeter dividend distribution, we kind of really we shocked you know that market quite a bit

and it definitely reignited a bubble. We got much more of a bubble in these markets than we would have gotten if they were just third time back in the same environment. But that took a lot of effort. It's interesting. It wasn't easy, actually, to get people that have been through two of these, and everybody's kind of had the experience probably of being burned you see he sold below fundamental value or bought above he did something that he realized didn't pay and so that he's trying to correct that that behavior so you've got people now that are in kind of correction mode and you put them in a new environment oh and by and by the way, they don't know that they all have been through two previous experiments. So you walk in, and here's people you haven't seen before.

Well, that's because we're drawing them in an environment where all they have to go on is kind of that learning they got from the previous two times. But they're not seeing the same people. They're seeing a different environment. It's a cashier-rich environment. I think we did two treatments, one with double and one with four times as much money. And, you know, they came through us. They gave us bubbles. So it was possible to do it.

And now, of course, we're talking about people that we're bringing back fairly soon, within a week or two, or even a few days. We're not talking about housing markets where it's a second generation, you see, of people that are now buying homes. And they're doing it for the first time. And maybe their parents are not even around anymore to advise them or caution them. And also, you know, young people, they tend to avoid, you know, discount the advice of parents. What do they know? You know, as they get older, they realize,

well, maybe the old guy did know something. You know, they start to learn. But again, you see, they're learning from experience. It's very hard to, it's very hard to people to learn from just being told things. Yeah.

You know, it's the difference, yeah, it's the difference between knowledge that and knowledge how. There's a great paper on that by Ryle, a philosopher. Oh, it's right after the Second World War, right at the end. Oh, I think it's in the Royal Philosophical Transactions. It's a great paper on, and it has to do with the distinction between knowledge that certain things are so and knowledge how. And he says the knowledge how is the knowledge of the craftsman. It's not articulatable knowledge, typical. A guy knows how to do stuff.

It's very hard to teach someone else to do it, you see. And so what people do is they apprentice with a craftsman and just kind of hang around and watch him and gradually learn it because it's not easy to convey that knowledge. And I think people have, they learn to have some of this kind of knowledge in markets, you see. They don't know exactly. They can't defend everything they do, but like in the

supply and demand markets, they don't really, the first time buyers and sellers in those markets do remarkably well, you see. They have more trouble in a market where things are retraded, but

now you see the retrade price, resale price may get disconnected from use value. And so I think we're starting to understand, you see, why these markets work differently. I love that distinction between knowledge that and knowledge how. And so much of knowledge for the greatest part of our history was transmitted through those mentor-apprentice relationships and that's the knowledge how and a lot of that I guess is learned by imitation and culture as opposed to dictation and rote learning. I want to come back to this question of future price expectations. And you

divided market participants in bubbles into two camps. There are the fundamental value traders on the one hand, whose objectives are supported by long-term rational expectations. And then there are the momentum traders on the other hand. And in book with steven gerstad rethinking housing bubbles you describe the momentum traders as myopically rational what do you mean by myopically rational are we able to quantify that characterization or is it just just words Well, our way of quantifying it is to simply hypothesize that their buying behavior, their willingness to pay for shares is related to the rate of change in price. So if the price change is positive, they're buying. When it turns negative and starts to fall, they're selling. So that's kind of a, you understand that's a simple theoretical way of modeling it.

Most people out there are going to be, it's not going to be quite that simple, but it's a way of, you see, it's a way of capturing some of, some of the truth, a piece of the truth that we think we're, we're, that's going on out there. And so we take this, this part of the truth is people buying according to fundamentalism. They're buying in proportion to the discount from fundamental value. They're selling in proportion to the premium. So that implements not a perfect rational expectations guy,

but a guy who's on balance moving in the right direction. And then we model other traders as these momentum traders. Now, mathematically, you put those together and you get bubbles. The bubbles, though, depend upon how much weight, how many of the first type versus how many of the second type. So if it's mostly the first type, the second type can't get very far. They're swamped. And if it's mostly the second type, then the first part don't have much effect. And so you just have huge bubbles, you see, in those circumstances. So that helps us to at least grasp apart pieces of the truth that we

think are going on in these experimental bubbles. But we're not explaining all the data. There's still a lot of noise in that data. There's unexplained changes left over after you introduce these two types of traders. And I think that gives you a clue as to why people are always out there looking for ways of analyzing price histories and coming up with a way of making money. See, they're exploiting the fact that price deviations are not just kind of random noise around a tram. They have a structure, a lag structure that's systematic, and people are looking for ways to exploit that you see and make money And but anyway, but if you find a rule that makes money and you start to implement it

well It's going to tend to go away and the more more people exploit it, the sooner it'll go away. And so when that happens, you've got to look for a new rule. And people are always doing that. You know, there's stock market advisors that are always, you know, giving you new rules about how to operate, especially right now. You know, we got a big increases in the stock market.

And so people are forecasting that this can't last and they're trying to come up ways for you to trade that you'll make money on this. So I guess in the real world, rather than one person being a fundamental value trader and another person being a momentum trader, the same individual might waver back and forth between different strategies across time. There might even be other types of traders, like maybe people invest for ideological reasons, for example. Well, there are some people that do deep systematic studies of the effect of news on the stock market.

And you see there you're looking at kind of trying to look for the fundamental elements that may produce momentum or not. So if you get a run of news events and they're finding maybe different kinds of announcements have different effects on prices. So people are using very sophisticated computer models you see to get a finely grained, finely structured set of of micro variables that will give them clues to changes, you see, in the prices of certain stocks, some going up and some going down in response to those news. And people trying to make money maybe on both ends by investing in the ones that are going up and shorting the ones that are going down.

You know, those are elaborate versions of the simple theme, simple modeling that we did in our experimental lab bubbles. I said I thought that the most important characteristic of asset price bubbles was that the assets have resale value, which opens the door to future price expectations. And if that's true, I would posit that the second most important feature of asset price values is that the beliefs of market participants are not independent. Provided you agree with that second statement, my question to you, Vernon, is which beliefs are the ones that people learn from each other? Is it the beliefs about what fundamental value should be? Is it the beliefs about what are going to happen to prices?

Is it just more crude beliefs like this is a great investment class or this is the right investment decision? If people do learn from each other and imitate and that kind of behavior generates bubbles, what are the things that people are imitating oh no i i think uh that that's that's a really good question and and it it does the answer does importantly involve the extent to which people's behavior is correlated with each other. You see, the famous, you know, Galton was a statistician, a very early statistician, lived in London.

He went to a livestock show. And people were betting on the weight of a prize bowl. You could look at the bowl, okay? Look all around and feel him, whatever you wanted to do. And then you would make a bet on what that bullet weighed. Classic bull market. So, so Galton got all that data and he found that the mean of all those bits was just very close to the actual weight of that bull. So now that you see here is a wonderful example of the wisdom of crowds.

You see, but it's also an example where presumably most of these estimates are independent. People are, you know, unlike, you see, what may be going on in some of these stock markets where people are all looking at the same fundamental things and inferring the same. I mean, they're all saying the same good news. And so their behavior is highly, highly correlated. So I think that's, and there's a, you know, there's a book, I've forgotten the author of The wisdom of crowds there's a great sir a wiki exactly james ricky and he's the one that points out um you you see he he asked what

what are the properties of those cases where crowds are wise and those cases where they don't look they're not so wise and one of one of them has to do with this independence you see and if people are making independent estimates then the crowd can actually beat individuals by a substantial amount because it's it's basically aggregating all that information you see in a way that the individual is unable to do. But if they're correlated, then it can give you a much different result. And also it helps you understand that correlation can help you understand why a bubble goes on and you'd think, wait a minute, it's surely past. Everybody surely can see that this is almost over.

You see, that was the housing bubble in the United States. Let's see, it started in August of 1997. I remember this very well. Peaked out, let's see, it was either the first or second quarter of 2005. I've forgotten now. I think it was the first quarter of 2005. That's a long run. And then it just collapsed.

And it didn't turn around until about 2011, 2012. And it finally sort of bottomed out. And that's what, of course, made the Great Recession so painful because you had this, the equity in homes that was widely held. And, you know, most people, if they have any wealth, it's in their home. That's where it is. It's not in the stock market. And you have a huge collapse, you see, in that value against debt. Debt is fixed and value is falling.

So the equity just gets hammered. And lots of negative equity all over the place. Well, you see, that's not in the national income account. That's not a part of GDP. You see, that's only current flows and incomes. And so GDP is not a good measure of, the decline in GDP is not a good measure of the pain in what Steve Gerstad, my co-author in that book, and I called balance sheet crises.

When household balance sheets are getting crunched badly, you see, and a huge loss of equity value, and people don't feel like spending, even though they've still got a job, you've still got income, but you don't feel as ebullient in that situation. So it's the kind of thing that you get in the Great Recession that we got in the Depression that's not much a part of the small recessions. Where housing values, they may dip slightly, but they hold up. And equity holds up, you see, pretty well.

So I think, you know, the sources of stability and instability in the economy are that are fairly easy to identify. And that is if it's a consumer non-durable, it's stable. See my original experiments were really about those kind of markets. People find the equilibrium in those markets very quickly in the lab, out there in the real world, spending on those goods, you know, hamburgers, haircuts, hotel rooms, that holds up. Even though your GDP is down, spending on those goods doesn't drop near in proportion to GDP at all.

And so that's, and listen, that is 75% of private product. 75% of private product can't, final private product can't be retraded. Well, thank goodness, because that gives us that inertia. The economy has a certain inertia,

you see, so that if houses are falling in value and the stock market is falling in value uh that source of instability is not totally dominating in the economy and that that helps to save us. Just to come back to this idea that in a bubble market participants beliefs are not independent. You could probably say that the more that behavior is correlated the more probable that it is that bubbles will form, right? Yes, I think that's true.

You probably, you know, if you thought about it, you could design an experiment that would kind of help to show that where people were being, whatever information you're feeding them, that's forecasts of earnings. Well, earnings help to determine dividends, say. Something like, see, think of it that way. Okay, you can have the information public and all get the same information or all people might get the information privately. So do things like that matter? You see, and you would try to get at factors that are associated with this correlation. And you try to design experiments so that by reducing that correlation,

you can show that the bubbles are not as severe. That would be the idea. Yeah. Yeah. I still don't feel like I have a clear sense of which specific beliefs people learn from each other. Is it like a vague belief that prices will rise? Which specific beliefs people learn from each other? Is it like a vague belief that prices will rise?

Is that the kind of belief that people are copying from each other in a bubble? Well, somehow that the news is related to the price of these securities. and you know news on the coronavirus 19 the pandemic you see that has ebbed and flowed and and that has that's

actually boomed some stocks I mean the biopharma, the pharmacological biopharma stocks that are into vaccines and treatments, they go up. Maybe when you have that sort of news, whereas the general market may be declining. Although pieces of that market may do better, because now we have a lot more mail order sales. So mail order sales companies, FedEx, they are the ones that bring this stuff to your door.

Those may do well, even though the general market is going down. So people are looking in the news for impacts on particular industries and therefore particular stocks uh and and well and and when people are looking at all looking at the same news and have the same belief about how that news affects market stock prices then you get you got a lot of correlation to say that a crowd has gone mad is not to say that the individual participants are acting irrationally, correct? Well, let's see. Yes, I would agree with that in the

sense that, well, take the momentum traders. As long as you believe that that prices are going to continue to rise then it makes sense to be in there as as a buyer you see and there's nothing irrational about that except that you know it's hazardous because you know it can't last you see and and and so whereas the fundamental trader he's dealing with actions that are going to last you see he's if he buys below what he thinks is fundamental value and he's correct in that belief, then he'll do better doing and by that strategy than if he hadn't. So people act on these beliefs and if there's a herd effect, you see, a madness, as you say, in the crowd,

it may come from just a summation over all of those kinds of behaviors that may imply an excess, you see and in this context it's an excess and in the sense that people are really buying lots because prices are going up and well that's why prices are going up see that's both the both the cause and an effect and how do you and how does that cycle come apart that's not always easy to understand prior to 1970 most researchers believed that people are pretty good decision makers and indeed the most cited paper of that era was man as an intuitiveuitive Statistician by Peterson and Beach, 1967. But then there was this sea change and a focus began to be put on finding anomalies. Now, in your Nobel lecture, you said that if people in certain contexts make choices that contradict our formal theory of rationality, rather than

conclude that they are irrational, some ask why, re-examine maintained hypotheses, including all aspects of the experiments, procedures, payoffs, context, instructions, etc., and inquire as to what new concepts and experimental designs can help us to better understand the behavior. Has behavioral economics strayed from the approach that you outlined? Well, I guess yes and no. Well, I think, you know, I think one of the things that, let's take a particular game that has attracted a lot of interest by behavioral economists. And that's the trust game. You see, this violates the notion that people just look at their own payoff and are concerned to maximize their own payoff.

And so, well, let me just give you an example of that game. It's a simple two-move game, and it's played in extensive form. That is, player one moves first, then player two moves second, and they each have complete information on the payoffs, and they each see each other's move you see. Well player one moves first. If player move, if player one moves right, the game stops and they each get $10. Okay, but suppose player one passes to player two. Well, that $20 becomes, let me see, I'm trying to remember the parameters here.

It becomes $40 and and player two now is going to divide stakes of $40. If player two moves right, it's $15 for person one and $25 for himself. So let's see, that adds up to $40. Yeah, so I got the numbers right. Okay, or he can defect.

That would be called cooperation okay instead of moving right person move person two can move down and let's say that if he does person one gets nothing and two gets $40. In other words, he can just take all the money. Now, we first ran a game like that. I think it was about players in that first game, and we were astonished because, see, we deliberately chose parameters where person one should be very reluctant to pass to two because if one moves right it's ten dollars for sure. If he passes to two he could get fifteen dollars or nothing

and it's obvious that it's in two's interest to give one nothing. So well what astonished us was that half of the player ones passed the player twos, half of them. Player twos, three quarters played right and one quarter defected. 75% did what we, in those days, we called it reciprocating. I offered to do something for you by passing to you and increasing the, sweetening this pot, and so now you're going to return the favor.

That was kind of our way of explaining it at the time. But there were those who explained that by putting these payoffs in the utility function. explanation for why the twos move right was that not only was their own payoff in their utility function, but the payoff of the other person was in that utility function. Well, that bothered me and Kevin McCabe, who was my co-author on that original paper, because for one thing, they did nothing to predict that. They're just finding out that's what people do.

And then they're saying, well, here's the way to rationalize it. People care about other people's payoff. And in a positive way, you see. And to us, that was not satisfactory. factory because that see the Ptolemaic view of the universe you know originally was that the planets all moved in circles but they weren't simple circles so

one planet as it moves around another planet it's itself going around in a circle around the arc the circle that goes around the first

so that gives you epicycles well it turns out that's a very rich model you can fit that model to anything because you not only got different size masses, you have different radii that you can work with. So I mean, you know, there's lots of flexibility. So it's one of those models that has as many parameters as it has variables in it. And so you can, almost anything you observe,

you can find a set of circular radii, you see, that would explain why the planet Venus, say, looks like it does from the perspective of the Earth. And so all, of course, that got all of that. See, that was a theory that explained everything and predicted nothing and science has got to watch out, if it's explaining everything and predicting nothing, what is it doing, what's it contributing to your understanding, you see, so we had deep

methodological problems with that. And so the experimental economists and the so-called behavioral economists have had, now they did a lot of interesting experiments exploring that, and all of those were good experiments. And in fact, the right model ultimately has to be able to explain you see all of these experiments but we think actually we think

the theory of moral sentiments has got a better model than that because he has propositions on how people react when someone else does something good for them versus how they react when somebody does something bad to them. He's got propositions governing that, and we think they apply very nicely to that trust game. So that trust game helped us to rediscover Adam Smith's first book on the theory of moral sentiments.

And because he was modeling human sociability. He assumed, by the the way that everyone was strictly self-interested and the reason why we do good neighborly things is because we learn to follow rules that turn out. If people are good neighbors, they live kind of in peace and stability, you see. And if they're bad neighbors, they don't. We learn from the rules. But it's not because we're not self-interested.

In fact, if you're not self-interested, how do you know you're person A and you're observing B and C? B does something that gives more of some good to C. Well, how does A know that C is better off? It's because he assumes that he's self-interested. In other words, we all take that for granted when we analyze these games. And in fact, you don't even know what it means to hurt or benefit someone, unless there's common knowledge of what's good and what's better and what's worse. So, so you have to have the self-interest in order to have good rules to live by.

So, and, and, and so Adam Smith is so clear on these things. So there's no, see, and people in the theory of moral sentiments are all self-interested. People in the wealth of sentiments are all self-interested. People in the wealth of nation are all self-interested. Okay, there's no difference in the behavior. But in the first book, he's talking about social networks and neighbors. First family, you know, kids, brothers and sisters, uncles, nephews, nieces, and cousins, as it goes out, why, you know, that we're maybe not as good of rule followers as we are with the people that we live more closely together with. The next-door neighbor is more likely to be the kind of person that, well, you know, you come home from work and you

forgot to bring in your trash barrel from the street. Your neighbor noticed it and brought it in for you. Well, that was a good thing to do because the street, that evening, the street sweeper was coming by and if there's any trash barrels still out there, you get a ticket, you get a phone. So here's people kind of looking out after each other.

And they tend to, people reward those good things. So there's a way of kind of understanding social interaction in terms of the rules we follow that have a very nice structure. They're not arbitrary rules. They have patterns to them. And Adam Smith is interested in articulating those rules. And so he gets, you know, he gets reciprocity out of that. Who above all should we be kind to? Those who have been kind to us in the past, you see.

So kindness begets kindness. He doesn't say it that way. He says kindness is the parent of kindness. That's just the way he puts it. And that, you see, that's that is, that's a description of reciprocity, to turn favor, to return a favor, you see, is in that

proposition. So it's, to me, it's just, it's so refreshing. This guy is just, he's a careful observer of everything that goes on around him. He's looking for meaning in the observations. He understands that order comes from rules. And he kind of learned he kind of learned that he and david they first learned that from isaac newton see he was he was very much the source of

of that way of thinking and so now how do you but there's other things orderly in the universe besides the motion of the planets. You know, there's order in the economy. There's order among social systems that involve living with our neighbors. And so how do you understand that order? And he begins first with the social world, writes a book on that, then he goes to the economic world and writes a book on that.

Vernon, to summarize what I think is the main lesson that I've learned from you about rationality, I would say that we shouldn't talk about rationality without referring to the particular scale, whether that is rationality at the level of the individual or rationality at the level of the collective. My final question to you is, have I learned the correct lesson from you? And is there like a final message you would you'd leave the audience with on thinking about rationality at different levels well joe i don't know i think i think you've you've got the lesson down very well and i think there's there's there's an understanding that we've arrived at through experiments, but there's an understanding

that also comes through the study of markets, both of those where people can't retrade the item. And the collectives there tend to find the prices fairly quickly. And also people have on the ground good smarts in those markets. They may not be able to model it in terms of rational theory. But what they have in those markets is pretty good models of each other. You see, practical, serviceable models of each other. So as a group, they tend to do very well. And when each is trying to do best for themselves, the group does best. Okay. Well, we come to asset markets and there's slippage from that. And the fundamental source of that is the fact you've got an item can have a retail value that's disconnected from its use value.

And that gap can actually last a long time and even grow. And you see that in housing markets where that flow of mortgage funds is an important part of other people's money, as Adam Smith would say, is helping to fuel that. And so it takes longer in that situation for the groups to get their act together. And they may get way off base, you see. And in the process, in the meantime, cause a lot of pain. And it's important public policy try to find ways to lean it against the wind rather than blow with it. You see, we unfortunately tend to blow with the wind

of policy because, you know, people want easy money when the price bubble is going very well in houses. They want to be able to borrow at low rates and pull the equity out of their home and buy a car. They want to do all that kind of stuff, but it may be destabilizing eventually. So, yes, I think this helps us to understand. And to me, it's fairly common sense. I think most of your listeners and viewers are not going to have any difficulty relating to these. They can relate it to their experience, you see,

and maybe thinking a little bit more outside the box than they might normally, you see, but they can see how these things work. So anyway, yeah, you're a great student, Joe. Well, you're an unparalleled teacher and it's always great to chat. Thank you so much for joining me. Okay, it's my pleasure. Thank you so much for joining me. Okay. It's my pleasure. Thank you so much for listening. I hope you enjoyed that conversation as much as I did. For show notes, including links to everything we discussed, you will find those on my modestly

titled website, josephnoelwalker.com. That's my full name, J-O-S-E-P-H-N-O-E-L-W-A-L-K-E-R.com. Please do subscribe to or follow the podcast, depending on which app you use, to ensure that you never miss updates when we release new episodes. The audio engineer for the Jolly Swagman podcast is Lawrence Moorfield. Our very thirsty video editor is Alf Eddy. I'm Joe Walker. Until next time, thank you for listening. Ciao.