How Investors Fall Into Bias Traps with Economists Richard Thaler & Alex Imas

33:44So this is kind of, Nudge is kind of fascinating because in the winner's curse, you talk about things very much related to what happens in investing. So there's loss aversion and the status quo bias and a variety of different things. Let's talk about what are the issues that most relate to, as Alex said, behavioral finance as opposed to behavioral economics, what do we think are the biggest factors that explain irrational human behavior in stock and bond markets? So I think there's a few things that kind of people documented in the late 90s, early 2000s that have just replicated and just became bigger, if anything.

34:34So the disposition effect is one of them. So the disposition effect, this is Sheffern and Statman. They came up with a paper in 1985 documenting it originally. Thierry O'Dean has this giant data set that he published in 1999 documenting it in a bit of a larger sample. And then now it's been replicated in Finland all over the world. And it's this tendency for people, you know, when I buy a stock, it goes up in price. What do I do? I sell it. I want to realize my gains. Same stock goes down in price. This is now a loss. What do I do? Hold on to it. So it's this tendency to realize your gains and hold on to your losses.

35:13Peter Lynch, by the way, 40 years ago, used to call that cutting your flowers and watering your weeds. That was his expression for it. So it was visible to a guy running a fund at Fidelity in the 1980s. Yes. And it's and this is just talking about like, are people learning? I mean, apparently not, because it's like it's again, you I bet you you download Robin Hood data from today, you're going to see it show up. So that's the and this is kind of the tendency, you know, what feels good when you're when when you own a stock selling it at a gain and, you know, telling your friends, hey, you know, I bought that thing for 90.

35:51It's 120. I just I just made a lot of money. You know what feels worse telling your friends? I bought it at 90 and I sold it at 60. So you just kind of hold on to it, hoping something happens. Maybe some people even double up, buy more shares just to break even. So the disposition effect, this kind of tendency for individual behavior to realize gains, avoid losses. The other thing is, in my view, this is kind of the bigger principle, is limited attention. So there's a lot of stocks out there. Which ones are people buying? And this is not just retail investors. This is bigger institutional investors, too.

36:29It's the ones that are covered in the news. We were talking about availability bias earlier. What are the things that are coming to mind? Things that have recently been covered. Maybe you heard an earnings announcement call or something like that. These attention-grabbing stocks that are much more likely to go into people's portfolios. It's because people aren't evaluating the entire universe of stocks whenever they're thinking about something to buy. So let's address that because the United States happens of all countries, not only has such a large stock market, but the home country bias is so acute here.

37:06And you don't hear a lot about foreign companies all that often. You mostly hear about local companies, local CEOs, local products. how significant is that sort of bias in people's portfolios being not only overloaded with their own country, but hey, if you're in New York, you can have more finance companies. If you're in San Francisco, you have more tech companies. If you're in the Midwest, you can have more manufacturing companies. It's more extreme than that. If I'm working for a specific company, I have more of that stock. When if anything, you should have less. Yeah, I think one campaign that has been moderately successful is I think fewer companies are foisting stock of their own company onto the workers.

37:55It used to be the match was often paid in company stock. Well, GE was notorious and they lost half a trillion dollars of employee investments because of their match. Well, and Enron. Enron. Zero. One of my friends, their father, he was working at Enron. He was a risk manager. FYI. Oops. Just not a very good one. Huge percentage of his portfolio was Enron. Although you could be the greatest risk manager there, the bosses were not listening to you. Right. But they compounded it by putting their employees' money in the 401k and run stock. So they get fired and their retirement money goes poof. Right.

38:46Unbelievable. But, you know, looking back at what people were owning, I mean, there is that, you know, people in the 401k element, but people who were working there were freely buying Enron stock. right according to economic models you should be diversifying you already have a bunch of enron stock in your 401k you shouldn't be taking your discretionary spending and buying more enron stock and that's exactly what was happening you know this home bias applies all around the world at least the u.s is a big country i wrote a paper once about the swedish uh social uh sort of 401k plan. This was a risky move because I was making fun of it.

39:30And there was some award that might happen. But anyway, Sweden is 1 % of world GDP. Tiny, tiny GDP, tiny stock market. And they weren't putting most of their money in Swedish stocks. It's crazy. They ignored the other 99 % of the world. But that just goes to show you the bias. So the obvious question is, if you two were advising a portfolio manager, what sort of behavioral principles would you emphasize for them to build a robust portfolio? Well, as you know, one hat I have is I'm involved in a company that does this. That also has your name on the door. It also has my name on the door, Fuller and Thaler Asset Management.

40:21And now let me say, I cannot name a single stock we own. And no one at the firm would think it's a good idea for me to be making suggestions. We buy small cap stocks. So if we owned Apple or Tesla, I might know it. but we don't buy any big stocks. So they're mostly companies you've never heard of, and I've never heard of them. But this is because you've identified a behavioral issue that is now reflected in the model that they use to purchase. Right. So we're not a quant shop, which is a little unusual for a firm that's run by some academics. But each strategy is based on a bias. So there's one that's based on overreaction.

41:17There's one that's based on underreaction. So we try to find stocks that we think the rest of the market is making a mistake about. And then we forbid the portfolio managers from forecasting earnings. because they're going to be, you know, do we think with our 30 employees that we're going to make better forecasts than fidelity? It's crazy. But we think we have an advantage because we're trying to predict something else. We're trying to predict the mistakes. It's like you're a baseball fan. If there's a pitcher that is a sinker ball pitcher, so the Alex this means the ball goes down as it approaches the plate.

42:09He's foreigners but you know if there's a sinker ball pitcher you and I can predict batters are going to hit ground balls because they are fooled the ball drops and if you hit it slightly above center the ball goes down so you don't have to be able to hit a ball to know it's going to go down. And so we don't have to be able to forecast earnings to predict that other people are going to be predicting too high. I want to bring this back to the book because one of the concepts underlying the book was, hey, there's a reproducibility issue in social sciences. How well have these anomalies and the theories you built around them, how well is this held up?

42:57How robust and reproducible are these findings? And it turns out very. Talk to us about what what you guys discovered when you were revisiting all of these principles that were first written about 20, 30 years ago. Yeah. So as you mentioned, there's a there's a some might call a crisis of reproducibility in social science more broadly. So this is psychology, some sociology, etc. And the worry is that, you know, these anomalies that were published in the 80s and 90s, these are the bedrock of the entire field of behavioral economics. And you might be worried, like, look, maybe these things don't reproduce.

43:35And there's two ways that they can't they don't reproduce. One, you run the same experiment again, and it doesn't work. It was P-hacked, as I said, like small sample sizes, no incentives. The second way it might not reproduce is that it literally only reproduces in the exact conditions it was run originally with college students at low stakes. You go out in a different population with people who are a bit more sophisticated, know what's going on, and it doesn't work. So what we did in the book was to say, look, first, let's take the exact same experiments and run them again. Everybody knows about the original anomaly.

44:09So maybe they don't work because people are like, ah, this is a loss aversion experiment. I know what's going on. I'm not going to do this. This is the endowment effect. I'm not going to do this. So we just replicated them directly on a completely different platform. So we used an online crowdsourcing platform called Prolific. Basically, everything works. Everything works. And we, you know, you don't have to take our word for it. If you go on the website of the book, we posted all of the results of our replications, but also instructions on how you can do it yourself. So if somebody is like, I don't know about these guys, run them yourself.

44:46And people are still loss averse. They still have the endowment effect. Things like the conjunction fallacy, the Linda problem that Richard was talking about, all works still. The second part is this external validity part. Does anybody other than college students display these effects? And that's kind of the updates part of the book. And the answer is yes. You know, the loss aversion has been in the myopic loss aversion part that's been used to explain the equity premium puzzle. That's still reproducible. We also do a bunch of out of sample tests of the of the anomalies that didn't use experimental data.

45:25And, you know, that replicates out of sample, too. So people aren't learning. the psychology is the same. You know, one of the columns was about the equity premium puzzle. We didn't include this in the book because it's a little wonky, but the equity premium is just the difference in returns between stocks and bonds. The equity premium puzzle is how big it is. And theory says it should be like less than 1%. And historically, it was about 7%. percent. And the article about that was in the early 80s. So we've had 40 years of data since the puzzle was announced. The equity premium, exactly the same.

46:08It's only 1 % lower. And that's what we see basically everywhere. Everything's the same. So one of the concepts that people have challenged as not being very reproducible has been the concept of priming to some sometimes uh anchoring is is similar but that seems to be more reproducible but when i hear linda the bank teller story that feels like the framing of that is very much a priming when you hear about her as politically active and uh being involved in what how do you distinguish when you have these theoretical overlapping biases that all kind of interact with each other. So priming is actually a huge literature in cognitive psychology.

46:58Basic priming is very robust. So it's the idea of, you know, I say a bunch of words that start with a K, what comes to mind a word that starts with a K that's going to reproduce any day of the week. There's a special subset of priming research that was done kind of in the nineties, early two thousands that kind of took this to an extreme, which is, so here's an example. Let's say you're doing word search and there's a bunch of words that have to do with like oranges, palm trees, hot weather, like vaguely related with Florida. Right. And then that's supposed to prime in your brain old people. And the result, the dependent variable was that those subjects who had those words, they walked a little slower out of the lab.

47:41Right. I mean, that's, it's a little crazy, Right. Kind of tough to measure also. Yeah. So, I mean, it relied. There's a lot of degrees of freedom. The researcher can be looking in a certain direction, you know, and those tend to tend to not reproduce the sort of priming that. something like the Linda problem, for example, has. That's more in the kind of cognitive psychology wheelhouse of like, what do you think about when I describe a person who takes part in radical rallies? What comes to mind? This is a basic concept in memory, right? And the second part that I wanted to say is that priming, as far as like looking at the behavioral economics research priming is a really small part it was actually not really featured much in the book um but but the type that is uh that was used by uh you know uh twersky and kahneman it's much more in the wheelhouse of just basic cognitive psychology more like anchoring does anchoring still hold up oh yeah very well oh yeah yeah yeah and look one of the things when when i was writing those columns, I could pick anything.

48:51I picked big effect sizes. And some of the problems, you know, we talked earlier about the norm in economics to make models smarter and smarter. I think there was a norm in psychology for results to get cleverer and cleverer. Well, I thought Alex's paper where you randomly sell versus what was actually sold, that was a very clever setup for a paper. It was clever, but it wasn't – what I was deriving is a norm that the models assume people are being clever as opposed to designing a clever paper. Gotcha. We're all for clever papers. Okay. We like clever papers. So when I was choosing what columns to write about, I picked big stuff.

49:43And think about there's a well-known company that makes cinnamon buns and has the strategy of pumping the smell of that out into the airport. Now, let's say you're on a low-carb diet, just hypothetically. Hypothetically. You know, if you walk by that thing, that's priming and that works. And it's not – It's not clever. It just works. Right? It's a big effect size. So there are – so everything I wrote about was big. And it's because I wanted to pick things that I thought were well established. and so you know it i think if i had looked for cute little things then some of them would have failed to replicate you also pick things that people were actively attacking and adversarially trying to replicate at the time that you were writing it yeah i mean look take take the ultimatum game.

50:57That's one of the original columns and one that we include in the book. The game is very simple. I give Barry$100. I say, share it with Alex. You can give whatever proportion of the 100 you want to Alex. He says yes or no. If he says yes, he gets whatever you offered and you get the rest. If he says no, you both get nothing. Now, the standard economic model at the time predicts that Alex will accept anything because something is better than nothing. Barry knows that Alex will accept anything. And so he offers him a dollar and Alex accepts. Now, real people, only an economist would think that that's a really good prediction.

51:57Anybody who's not an economist is going to say, what are you kidding? I'm not going to take a dollar and give you 99. You didn't do anything to deserve that 99. So if you run that experiment, if you offer less than 20%, you're going to get rejected. And the profit maximizing offer is about 40%. And most people offer half. Okay? Now, there were big fights. There was a professor from Brit who was saying this was challenging game theory. And no, it wasn't challenging game theory. It was challenging the idea that the agents only care about money and don't care about being treated fairly. So let's address that because I love the evolutionary biology of this.

53:08Humans were cooperative social primates. We have neither fangs nor claws, so we had to come up with some way to stay alive. And it turns out cooperating in a tribe is very useful survival tactic. It seems that an inherent sense of fairness is somewhat built into all of us as well as social status seeking. So how much of this issue in economics derives from not understanding a little bit of evolutionary history? You know, it's a tricky thing. obviously we have evolved to be who we are there are some people who then say well that means whatever we do is optimal well maybe maybe not no that's stupid i mean we evolved on the savannah right right nothing about picking muni bonds from a large assortment you know uh amos tversky was famous for one-liners.

54:14And he had a one-liner about loss aversion, which was there may have been species that did not exhibit loss aversion and they're now extinct. Right? So if you're at subsistence, it's really smart to be worried about losing. It's an existential threat. Right. But, you know, the three of us, we could go several days without eating, some more days than others. Right. Right? So we're not at subsistence. And yeah, managing our own portfolios is something people have been doing for 30 years. Right? Yeah, the rich people, but they had their broker do it. Right? So there's no evolutionary history of how to manage a portfolio.

55:12And even saving for retirement, people didn't live long enough to worry about that. And if you were unlucky enough to reach my age, then you hoped your kids would take care of you, and they lived nearby. You know, then people started scattering and penicillin and, you know, so now we live long and our kids are scattered and they have no interest in having us move in with them. So people had to learn a very new thing and they needed some help. Really, really fascinating. So we didn't talk about from whence the title of The Winner's Curse comes from. Before we get to the future of behavioral finance, let's talk about The Winner's Curse.

56:09Tell us where the name comes from. The concept, I should say, the title of the book comes from the title of one of the chapters. And I picked it as the title back then because it's sort of a fun phrase and a bit intriguing. And the concept itself is interesting and important. The idea is this. Suppose you have a lot of people bidding for some object that's worth the same to everybody. And it could be when you do this as a demonstration in the class, you fill up a jar of jelly beans or coins and say it's 10 cents for each jelly bean. and it's$100 in the jar. Now we're going to auction it off. High bidder wins the$100.

57:02But they don't know what. They don't know it's worth$100. All they see is a lot of coins or jelly beans. What happens? Well, the average bid is less than$100 because people are risk averse. But the winning bid is always above$100 if you have enough people, because the most optimistic forecast is likely the highest bid, and it's too high. Now, this was not discovered by psychologists in the lab. It was discovered by engineers at Atlantic Richfield, ARCO. The energy company. The energy company, who discovered they were bidding for oil leases in what I continue to insist on calling the Gulf of Mexico.

57:58And they realized that the leases they won had less oil than they expected. And they said, gee, we thought we had world-class geologists. What's going on? Are they dummies? And then they realized that it's quite subtle that what you're trying to do is make a bid that will make you money if you win. And the if you win part, if there's 100 people bidding, gee, do I really want to win? Because maybe I misunderstood something, right? So that's the winner's curse. And it was found and replicated on bidding for oil leases. It's relevant in book publishing. When they're bidding contests for books. Yes.

59:00Yes. So let me see if I can clarify the way you're describing the winner's curse. So we're bidding for oil leases. We don't know exactly how much oil is going to come out of this hole for or area for the next 10, 20 years. And when there's many people bidding, all of which are advised by geologists, if you make a conservative bet, the odds are you're going to lose. But if you make a bet that's high enough that you're going to win, the odds are it's not going to be a moneymaker. Right. Coming up, we continue our conversation with Richard Thaler and Alex Emis discussing the book they have recently updated, The Winner's Curse, Behavioral Economics Anomalies, Then and Now.

59:46I'm Barry Ritholtz. You're listening to Masters in Business on Bloomberg Radio.

1:00:10I'm Barry Ritholtz. You're listening to Masters in Business on Bloomberg Radio. My extra special guests this week are Richard Thaler and Alex Emis, both of the Chicago Booth School of Business at the University of Chicago. I shared with you an article I saw recently. some real estate group did a study of tens of thousands of home transactions where there was a bidding war and they found something very similar. The winners of the bidding war ended up paying much more than the subsequent home value was determined by looking at comparable homes in the neighborhood. So is the purpose of an auction to identify something at a fair value where it's profitable for you, or is the purpose of an auction to win at any cost?

1:01:01Well, so there are two interesting aspects of that. One is, suppose you're these engineers and you've discovered this, what should you do? And if you're losing money every time you win an auction, you could not bid, but then you don't have any places to drill. And what did they decide to do? It's really clever. They decided to write a paper. In order to get people to stop overbidding. Yeah. Now, that's different than what the owners of Major League Baseball did. I was going to say, why didn't they just moneyball it? Why didn't they just start looking at ugly but productive? Well, what the Major League Baseball owners did is they colluded.

1:01:51Right. And they said, look, let's not bid anymore. Well, the salary caps that are in all these. No, no. There was just outright collusion when baseball players first became free agents. The owner said, hey, we're losing money on these crazy auctions. Let's just not bid. And then they got slapped down. Right. So what can you do? you can try to a good strategy is to bid very low on every site and in the data there were lots of sites that you could have gotten for a dollar really and because because the consensus was there's no there and were any of them productive than any of these some of them will be really So now let's bring this back to sports because you've written papers on NFL drafts.

1:02:52So the NFL draft, every year they have a draft for new players. The first pick is given to the team with the worst record the previous year with the idea that that's going to be a big advantage to them and will help them improve. Cade Massey, one of my former students, he and I wrote a paper showing that the first pick is actually not the most valuable. Because the league has a salary. The first player gets paid the most. And you can trade the first pick for the seventh and eighth picks or for five second round picks. And what we showed is if you trade the first pick for lower picks, you get more value.

1:03:52So now this is known like the Arco engineers publishing the paper. And yet there still seems to be this frenetic war for top one, top three, top five picks. Has the NFL learned any lessons from the research? almost nothing they so they have learned that you should they only trade up to get the first pick to pick a quarterback so that's smart because the quarterbacks are more valuable than any other player well of course you want that first pick so you could get a tom brady yeah except except Septon Brady was taken with 199th pick. And all the listeners who are football fans can have their list of people who were taken with the first pick and turned out to be busts.

1:04:50The Chicago Bears seem to specialize in that. Although, let's hope, this current guy is a new one. So how much of this, like I'm seeing this through the lens of my book, which I don't want to talk about, but how much of this is just how difficult it is to predict the future, to to have truly expert judgment about these very complex, very variable. So selecting quarterbacks, identifying oil leases, like it seems that supposed expert advice ain't all that expert. How much of this is aren't we better off just being a little more humble about our let's give up the top pick and have five second round?

1:05:39Somebody in that five is likely to be half decent, right? So let me stick to the sports for one second because there's one statistic from our paper, and we've just – coincidentally, we've been in a process of replicating that study. So I have the new data. But here's the statistic. Take all the players at a given position, say quarterback or cornerback or running back. Rank them in the order in which they're picked. and now ask the fourth guy, what's the chance he's better than the fifth guy? So for the whole thing. So what's the chance the earlier player is better than the next one? One over two, 10 over 11, five over six.

1:06:26Now, if they're perfect, it'll be 100%. If they're coin flipping, it'll be 50%. What do you think it is? I think it's less than 50%. I think it's negative. They know less than nothing. Right. That's right. I think it's in the 30s or 40s. Well, they're not that bad. All right. I mean, because if they were, then you could just, you know. No, no. You'd want to, what the George Costanza, you want to do the opposite. Right? That's right. So, no. They don't have negative knowledge. They have a tiny little, it's 53%. Okay. Okay. So, but that's your point, really, which is they think they know this guy is the next Tom Brady.

1:07:15And there's only a 53 % chance that he's better than the next one. And, you know, Patrick Mahomes, Josh Allen, think none of these were first picks. Right. That's right. Right? So. And this Mahomes kid is going to be a pretty good one. I think he might make it. Yeah. He's got some potential. Right. So and I think that is so getting off of sports. I think that your general point is exactly right, that people are look overconfidence. Danny Kahneman used to say that's the mother of all biases. And we fall into these traps because we think we know more than we do. and uh if we had some humility maybe if we listen to our spouses more often because at least in my house um my wife doesn't think that i know anything so she's always bringing me back to 50 percent.

1:08:22And she's usually right. My wife is from the same cut from the same cloth. So, so I want to bring Alex back into this. So when we're thinking about the future of behavioral economics and, and what this means for investors or regular people making financial decisions or and significant decisions, what direction are we moving in? Are we learning from all of this knowledge that's been accumulated, or are we just destined to make the same mistakes over and over again? I don't think we're destined to do anything. I think it's a choice to take, you know, read papers and look at papers on kind of published and financial journals where people are making mistakes and then to choose to say, like, look, I actually can correct this by having a particular decision aid or asking my spouse what to do or something like that.

1:09:22So, you know, a paper you mentioned earlier, we published this actually just last year called Selling Fast and Buying Slow. And in that paper, basically, we look at institutional investors. So thinking about who in the economy are least likely to be exhibiting behavioral biases, you know, Maybe retail traders, they're drinking beer in their basement while trading stocks on Robinhood. Maybe this is not the sophisticated people we want to be looking at. But institutional investors, the average portfolio in the data set was like$600 million,$700 million or something like that. We had a data set where we actually saw every single thing they did over something like a 12 - or 13-year period as far as what they're buying and what they're selling.

1:10:06And what we found is because the data is so rich, we can actually construct these counterfactual portfolios. We can say, look, I see what you're buying. What if you bought something else? So it could be something else from your portfolio. You can top something up or you can buy something new from the universe. On the other hand, we could say, look, same thing for selling. I saw you sold Apple. I saw you sold Samsung. Let me sell something else instead. How would that perform relative to what you actually did? And what we found is that on the buying side, people actually did really well. I mean, these guys are – Not really well, but better than random.

1:10:44Fund managers create some value in their stock selection when they're making purchases. But the flip side of that, not so much. No, not so much. We really wanted to be conservative. We didn't want to, say, compare them to the benchmark or something like that. We said, let's throw a dart at your portfolio and sell that instead of what you actually sold. So instead of selling what the manager wants to sell, you would sell something else randomly from the rest of the portfolio. Yeah, a random selling strategy. And the performance difference was how significant? Basically, same difference but in the opposite direction.

1:11:18Meaning? So they were losing a ton of money. So 100, 200 basis points on a random sell. Better performance. Exactly. And the way when I read that paper and wrote about it, the way I rationalized it or tried to conceptualize it was they're bringing a very objective quantitative approach to the stock selection issue. But it seems that their cells are filled with biases and squishy decision making. Is that a fair description? Yeah, exactly. So we found no evidence for heuristics on their buying decision. Like we couldn't find anything. Like they just seem to be very disciplined and principled about what they're buying.

1:12:00But on the selling side, we found literally the same biases that we found in the lab. Has there been any evolution or improvement in this recently? That's the question that I keep coming back to. It seems that, Richard, you figured a lot of these things out 25, 35 years ago. So are we any better at making unbiased decisions or are we still subject to the same foibles? I think you need something extra, right? You can't just say, I am not going to do this and I have decided not to listen to my psychology. That's what it would look like to be better. Choice architecture or building in some guide rails and defaults.

1:12:41This is where overconfidence comes in. When you read a paper about somebody doing something silly, your first reaction is not me, right? Yeah, that's them, that's not me. The blind spot. This is called the bias blind spot. Exactly. So this is a well-replicated finding when you ask people, to what extent do you exhibit a bias? I don't, obviously. I'm a smart person. But to what extent do other people? Of course. Other people are bad at selling. I'm really good. But in order to adopt choice architecture to help you out when making decisions, you actually have to have a lot of humility to say, look, these institutional investors, to say, look, looks like I'm not really doing so well on selling, I'm going to adopt some choice architecture so I don't suffer from these biases.

1:13:26Maybe I'll hire somebody else to help me out. Maybe I'll think longer or use the same sort of research technology for my selling as I'm doing for my buying. And because that requires humility, which most people don't have a lot of, that's really hard to do. So I think that's why we're seeing a lot of these biases just be perpetuated forward to to the point where we're running the same analyses now as we did 30 years ago and finding the exact same thing. So the question that, since we were talking about sports and lack of knowledge, and then you mentioned Robinhood, one of the things that's a little concerning is how some companies are putting our knowledge of biases and bad behavior to work for their own profit.

1:14:13So when we see the gamification of investing with Robin Hood or just the incredible rise, not just of sports books and gambling, but you could bet on every play. It's reached a point where it's ridiculous. And there is robust evidence that especially young men are having all sorts of how can we how can we deal with what seems to be not a good use of choice architecture, but a bad use of choice architecture, at least as far as the public is concerned. Yeah, it's a very good question, Barry, and one to which I don't have a pat answer. I mean, it's tempting to say, look, all these sports betting apps and the gamification of investing are bad for people.

1:15:03On the other hand, people like doing it. They're mostly adults. And, you know, prohibition basically didn't work. Right. So I think some disclosure would help. It's difficult to find out what the odds are in a lot of these things. But it's a tough question. I had a conversation on this book with Nate Silver a couple weeks ago, and we talked a lot about sports gambling. He's a professional gambler. And he spent a year betting on NBA games and basically broke even. So, you know, if Nate can't make money doing this, chances are you can't. And, you know, my advice would be, look, if you really think you like doing this, do it on a small scale.

1:16:09You know? Don't bet the house literally. Don't, right. And the same with weekly options or daily. That's one of the most popular. One of our colleagues at the University of Chicago, she did an analysis of what retail traders are actually doing on Robinhood. And one of the most popular products, because it's pushed by Robinhood, is weekly options. And there's now end of day options where it expires. You have till four o 'clock to either make money or not. So, you know, if you want to risk one month's pay on that, fine. Just not every month. Yes. Yeah. That's your lifetime budget. And when it goes to zero, switch to something else.

1:16:55I'm a big fan of the cowboy account where you take three or four percent of your portfolio. And if you want to fool around with options, whatever, knock yourself out. And if it makes money, great. But like we've seen, you mentioned Apple. If it was your whole portfolio, you would never have been able to ride it to be a 5X or a 10X. You would have taken, I'm up 20 bucks. I'm taking the money off the table. Yeah, yeah. So, you know, people long ago would adopt the strategy of bringing a certain amount of money to the casino. Right. And then, of course, the casinos put ATMs on the floor. So it's a battle.

1:17:34but mental accounting, you have a gambling account. But that's it. Yeah. I mean, it would be better if it were zero. Right. But otherwise, set it up. That's something you can afford to lose. And when you've lost it all, stop. Don't go to the ATM. That's right. So I only have you for a few more minutes. I want to ask two of my favorite questions that I want to ask each of you that I ask all of my guests, starting with what sort of advice would you give a recent college grad interested in a career in either behavioral finance or economics? Alex, you're the more recent grad. What would your advice be?

1:18:16Get teched up. Really? Get teched up. I think that's the biggest kind of difference between even when I was in graduate school and when I'm seeing hiring pre-docs and RAs, the sort of work that you're doing in modern behavioral economics, modern finance, It just requires a different level of analysis. So is this learning to code or is this becoming a prompt engineer for AI? I think you still got to learn to code. I think, you know, I work in the applied AI group at Booth. You know, you still got to learn to code. And a lot of the sort of modern analyses that people are doing, particularly as behavioral finance, behavioral econ has moved out of the lab into the field.

1:18:54These data sets are huge. Machine learning, AI tools. the type of people who are getting hired are doing sophisticated analysis so it's still basically stem groups science technology engineering and math for people who don't know the acronym but applied specifically to the field yeah yeah and like you know you take your economics courses you take your finance courses take some cs courses on the side those are uh this is what i wish i would have done right when i was getting my phd it wasn't on my radar to take you know a coding class in the CS department, people coming out now, the ones who are really successful, you have to have good ideas.

1:19:32That's a necessary condition. It's not a sufficient condition. You need to be teched up to a level that I don't think we were seeing back when I was graduating. Really interesting. I'll reinforce that with the following. I think you need some practical experience. Because the part that you don't learn in our textbook is you get this gigantic data set and it's noisy and there are errors. And so learning how to clean up a data set, you got to learn that through experience. Really interesting. And our final question, what do you know about the world of behavioral economics today would have been useful back in the 1970s and 80s when you were getting started and in the 2000s when you were getting started?

1:20:27So I think if we go back, we talked about the changes that I helped make in the retirement plans. And what I wish I had been able to accomplish more of is making retirement saving at the workplace available to the possibly 40 % of American workers whose firms don't offer that option. And there were plans around and they didn't get passed. I think the system they have in the UK is a reasonable model, which is there's a requirement that any firm with more than, I'm not sure the number, say 20 employees, has to offer a plan and automatically enroll people into the plan. And the government has like a generic plan they can use, like the government thrift program.

1:21:36So, and this is useful because big firms like Fidelity and Vanguard don't really want tiny accounts. Right. So make it easy for an employer. They don't have to do anything. They just have to let their employees enroll in this. And then when they change jobs, they can keep it there. Because the real problem is they go to work for a while at this firm, and they work there for a year, and they've saved$600. And then they leave, and they take the cash out. So we need automatic rollover. so that that's the piece of the puzzle that i don't know whether i could have done anything about but it's what i wish we could work on now so related to that what do you think of these new baby accounts every newborn in america next year gets a thousand dollar has to be invested domestically which you know we can have an argument we're talking about home country bias But still, you're starting every infant off with a portfolio.

1:22:46What are your thoughts on that? I don't know the details of how that's going to work. You could – my friend and sometime colleague when I'm in Berkeley, Ulrike Malmendie, has made a similar proposal in Germany where she's on the German Council of Economic Advisors. I think the idea is to give kids some experience with the stock market. And I think that could be useful. I don't know how this is going to wash out. And all of these things end up being tilted toward the rich. I mean, there were big reforms made recently to the retirement plans. One of the reforms was to let you wait longer to start making withdrawals.

1:23:43Who do you think that helps? People who are wealthy, healthy, and going to have a longer lifespan. Right. So most people start taking the money out at 59 and a half, raising the date at which you have to start. from 70 to 72 doesn't help anybody that's in any trouble. Yeah, that makes a lot of sense. Alex, what do you know today about behavioral finance that you wish you knew when you were getting started? I think when I was getting started, I wasn't really thinking about being able to target these big institutional investors and thinking about getting data sets on smart money. in the economy.

1:24:30So I think when I was starting out, I was really focused on lab experiments, I was really focused on kind of the data that's that was available. And if I was starting out now, I would I would start my PhD trying to get trying to get data sets that I eventually was able to get because the types of as far as like looking at my research, what has had the largest impact? What has had, you know, people in finance in in the professional world calling me up and saying, hey, what do you think about this or that? It's been looking at the population that people are actually interested in, which are the smart money in the economy.

1:25:04So I think, and this is, I think, you have this analogy of looking under the streetlight, where is a lot of behavioral finance - For the missing car keys, the junk on the streetlight. Where are you looking? Oh, it's where the data sets already are. So Terry O'Dean had this genius idea in 98 when he published his paper, what did he do? He made that data set available. Now it's easy to just, oh, I got an idea. Why don't I look at Terry's data set? Terry's data set is great, but it's three years in the 90s with a couple hundred retail traders. And that tells you about that specific population, but you can't have a field evolve looking at three years of retail traders with$10 ,000 portfolios.

1:25:48So I think if I was going to going into the field now and thinking about, you know, what have I learned? It's the power of getting data sets and running analyses on populations that are important for the economy and for finance. Really, really fascinating. Gentlemen, thank you so much for doing this. We have been speaking with Richard Thaler and Alex Emis, both of the Booth School of Business, about their updated version of The Winner's Curse. Strong recommendation. If you enjoy this conversation, well, check out any of the 592 we've done over the past 12 years. You can find those at iTunes, Spotify, Bloomberg, YouTube, wherever you get your favorite podcasts.

1:26:33And be sure to check out my new book, How Not to Invest, The Ideas, Numbers, and Behavior That Destroys Wealth and How to Avoid Them. I would be remiss if I didn't thank the crack team that helps put these conversations together each week. Alexis Noriega is my video producer. Sean Russo is my researcher. Anna Luke is my producer. Sage Bauman is the head of podcasts here at Bloomberg. I'm Barry Ritholtz. You've been listening to Masters in Business on Bloomberg Radio.