Prof. William Goetzmann: Learning from Financial Market History (EP.248)

30:49And which of the two? Are you better off holding, let's say, just car company stocks, right? But holding car company stocks from China, from Japan, from South Africa, from Brazil, and compare that to what if you put all your money in one country, but you diversified across industry, okay? So it turns out you're much better off. When I say much better off, you have lower risk without giving up too much return. You're much better off having diversified your portfolio across countries, even if you have to hold only one industry, like the cars. That's a pretty interesting result. Why is that? You know, is it due to the fact that we have these political borders?

31:40We have different macroeconomic policies in each one of the different countries? How come a car company in Japan is not going to be highly correlated to a car company in North America, right? Why? So, I mean, there are forces that should drive the two of them together. Since after all, when we decide to buy cars, we go, well, maybe I'll take a Japanese car, but I heard the key is good and blah, blah, blah. The product market's integrated. How come the equity markets aren't integrated? So I don't know. That's an interesting problem. But that's the finding that we have that Geert has found before with other people.

32:18But we've been studying it particularly in that early time period from 1870 to 1929, because we're going to see if that holds true in the past, and whether it's due to trading in different markets as opposed to just having cross-border differences. Let's shift to your work on bubbles. How have you defined asset price bubbles in your work on the topic? I've defined asset price bubbles in the most naive manner possible. And I'll tell you what it is, but I'll tell you why first. Some of the most sophisticated and interesting financial theory has gone into trying to define a bubble in an asset market.

33:04Is it based upon overlapping generations of beliefs about the stock market? Is it based on some behavioral biases. My colleagues through the years have gotten very clever at designing mathematical expressions of what a bubble is and then studying those. So I said, that's not my comparative advantage. I mean, I'm not a sophisticated theorist. So I'm going to start with a definition of a bubble that most people can relate to, which is, I think it's a bubble if the stock market doubles in one year and then gives back that return or more in the next year. I mean, come on, that's got to have been a bubble and we control for inflation, right?

33:54So once I do that, the world is simple. Then I can study bubbles and you can tell me whether they're due to one bias or one problem or the Federal Reserve, whatever you want, but at least got a simple definition of a bubble that I can study it. And how do you differentiate a bubble from a boom? Well, it's real simple. A bubble starts with a boom and then it's followed by a crash. That's it. Followed by a bust. So we take the bubble and we say one side is the way up and the other side is the way down. Okay. So boom is the run-up without the way down afterwards. Yeah. The good news without the bad news.

34:38Now you've looked at this at a ton of data. Can you just briefly describe the data that you've used to look at booms and bubbles? Sure. I guess over the last, I don't know, maybe 15 years, a lot of people have been able to collect stock prices, stock returns in many different countries. And I've been guilty of that myself because I've been interested in this historical data collection. And so we're getting to the point now where you can get access to stock indexes for more than 100 countries, 120 countries probably. Well, some of them start really early, like the UK starts in the 1600s. And so if you have the world's stock returns at your fingertips, and so one of those databases is called global financial data, then you can really study the many different paths that markets have taken.

35:35And you can deal with the survivorship issue and things like that. And then another database that I was able to look at is a database collected by Dimson, Marsh, and Staunton. And that is a database that starts in 1900 with about 20 odd countries. And then lately, I think they've even expanded it annually. And so now we got over 120 years of very clean data that those scholars have constructed. Anyway, that's the data that I said we can use that to study booms, crashes, and the combination of the two, which are bubbles. Therefore, what was relative frequency of these booms and bubbles? Well, the good news is that bubbles don't happen very often.

36:23Thank goodness. And in terms of what you mean by a bubble, I mean, even in the United States in 1929, the run up to 29, and then the crash thereafter, it didn't happen in one year. So even to capture that idea of that bubble and also the crash of 1987, for example, I kind of had, or 1990 in the tech bubble, I kind of had to expand the definition of what a bubble was to include a run-up over three years and a crash over the next three years. So in any case, in my data, that one year up and one year give back, that bubble happened not much more than 1 % of the time, if I can remember correctly. Very rare.

37:08Booms were, if you just separated the first half and the second half, the booms were, of course, more frequent. But even there, not too many markets doubled in value in inflation-adjusted terms in one year. But when it does, that's when you want to know, oh my God, what's going to happen? So when it happens to you, you want to be able to figure out, is it going to crash the next year? Should I be selling out my stock position and patting myself on the back for doubling my money? And that's when you want to know what happens. I'm going to keep you in suspense. I know. I want to know the answer.

37:50Yeah. So, okay, for this one year up, one year down, what we found is that you doubled your money in one year, it's 50-50 whether or not you're going to double it again the next year or give it all back. It's like a coin flip. And so, you know, that's risky, but here's why it's that risky, because any market that has doubled in one year is a very volatile market. So by definition, we have sorted, I've sorted on the most volatile markets that can double. And so you can look at that and say, well, I don't know. I just made a bunch of money, but I don't want a coin flip of whether I can give it back or double it again.

38:32That's up to you. But there's no evidence that I found that the market has a tendency to reverse. In other words, it doesn't follow the old story of what goes up must come down. It's not a Daedalus and Icarus situation. And so that's really pretty comforting. Now, when you extend the horizon out, if you say to yourself, well, what if I hold on to my stocks for another three to five years, then the story gets even better. You're much more likely to have long-term returns that are positive, even after a boom like that. So that's kind of extraordinary. What would you say are the lessons for investors who may have become concerned about a run-up in a market that are maybe thinking about timing that market?

39:22Well, right now, wouldn't it be nice to have that luxury, huh? It's like, whoa, what we would love to see is another kind of super stock market that we had in early 2020s and so forth. But when that happens if you're a taxable investor, you have to think, do I sell out now and then pay my capital gains tax, but sleep comfortable at night? Okay. Well, the capital gains tax could be really expensive. That's expensive insurance for a good night of sleep. So that's the taxable side. I would say the statistical data suggests to us that it's a coin flip, whether it's going to keep on going up or go back down.

40:04Now, the penalty you pay, you have to trade off that risk aversion versus the cost of insurance. Now, what's interesting on the other side is, suppose that you saw something which we saw in, let's say, 2020. The COVID crash scared the heck out of everybody, right? And then you want to know, should I bail and go into something safe or even something like gold, which is not even future dollars? It's just a commodity. I should throw Bitcoin in there, too. So anyway, there, when we look at that pattern, we find that if it's a really big crash, the probability of a rebound is actually pretty high. That's exactly the opposite story, though, from what we found with the boom.

41:01So if it's a big crash, if it drops by 50 % over the course of one year in real terms, then we find that 50 % or 40%, we find that there's sort of a gain of a third to a half back in terms of value over the next year. And well, we've looked at different horizons as well. So there tends to be a rebound when there's a crash. Not all the time. If it's only a 10, 20 % decline in value, it could keep on going down. But there's this sense of recovery from the most extreme of the stock market crashes that we've been able to document. What explains the negative bubble behavior? Well, we didn't test it.

41:48We tested a lot of theories about it. those theories include, is it due to passing geopolitical shocks, internal turmoil, banking crises, and that kind of stuff? To some extent, the decline is associated with those problems. And with a banking crisis, you might see a flight to liquidity, a need to sell risk assets and for banks to be able to buy safe assets. And so that could be associated with a drop. But we don't really have any good fundamental explanation for the rebound, other than just the notion that whatever caused the crash was temporary and non-permanent. And so it's a little bit unsatisfying to say what condition, because I can't tell you, What conditions should I look for that will help me predict a rebound?

42:47Which is sort of what a real investor would like to know. Does it suggest that big crashes like that are more discount rate related than cash flow related? It certainly does to me because, you know, a sudden shock is more likely to be related to these pretty frequent variations in the interest rates or even just the discount rate that you would apply to future cash flows. earnings and dividends seem to be a little bit slower to evolve. Although, again, let's look at COVID. Suddenly, businesses that had healthy expectation that people would be doing their holiday travel, suddenly, it was just gone overnight.

43:30So you can never really have a complete story. But I agree with you that a lot of variation in asset prices is likely due to variations in expectations and tied to the discount rate. I think that that's probably something that is driving the temporal nature of a portfolio of stocks. To keep going on crashes, how well does the frequency of actual crashes line up with investors subject to crash expectations? Well, that's not easy to test because subjective crash expectations means that you got to be asking people about what they think is the probability of a stock market crash. The good news is we've been doing that for more than 20 years.

44:23And it all started with my colleague, Robert Schiller, who's a Nobel laureate and one of the creators of behavioral finance. right after the crash of 1987, Bob had the idea to mail out questionnaires to individual investors and institutional investors and ask them, like, what do you think is the probability of a stock market crash of the magnitude of 1987 or 1929? Or where do you think the Dow index is going to be over the next six months or a year? Or is this a good time to be in the market? I mean, he asked them lots of questions and they answered them. A little bit more than 20 years ago, our center at Yale, when I say ours, it's the International Center for Finance, we started doing this on a monthly basis with exactly the same questions.

45:12And this was for Bob's research. And now we have this rich set of data about people's views on the market. And the big takeaway for me, once we started looking carefully at it, is that people thought that the probability of a catastrophic stock market crash in the next six months tended to be somewhere between, I don't know, between 10 % and 20%. So imagine that. Imagine the anxiety of walking around thinking a 1929 crash could happen in the next six months, one in five. One in five chance it could happen. I mean, it's amazing. So either people didn't understand that their probability assessment of 20 % chance of this happening in the next six months, maybe they didn't understand probabilities, but maybe people that were answering these questions had a dark view of what the future will bring on a systematic basis over decades.

46:16But by definition, there are only two crashes on the magnitude of 1987 and 1929 in the US market over the period since 1929. So almost, right, almost 100 years. Two times it happened in 100 years, that means it's 2 % chance. Now, we could fiddle with that to be more realistic about volatility, this and that. But if it happened twice in 100 years, now I'm asking you if it's going to happen in the next six months, your answer should be, well, the probability is about 1 % and it shouldn't be 20%. So the thing that worries me when I look at these, oh, and by the way, institutional investors as well as retail investors had very high probabilities.

47:06Institutional investors, a little bit more practical, but they were still pretty elevated. And it kind of makes you worry that people are so frightened that a crash is going to eat up their savings that they're leery of putting their money in the stock market. And then that means they're going to give up that equity premium just for the safety of avoiding a crash that probably isn't going to happen in the next six months. What role do media narratives play in investors' beliefs that there'll be a crash, do you think? Well, media narratives, first of all, Bob Schiller, he had this idea about narratives being the essential means by which people pass along economic ideas.

47:51They could be very short narratives. Like I mentioned, what goes up must come down. We all recognize that. And then if I tell you, if I said, the market operates the same way, what goes up must come down, you're suddenly thinking, well, that just makes so much sense. A lot of things are like that. That could be misleading because I just told you what goes up doesn't necessarily come down. But that narrative could be very powerful for a lot of people once they start spreading it around the internet or once journalists start to say, how am I going to tell a story that's going to really get people, we'll help them understand things or we'll sell newspapers or telling a story is the way that we humans interact with each other.

48:36So I've done with Bob and also with the co-author, Docile Kim, is, well, we started by looking at all the newspaper articles in the Wall Street Journal and the New York Times that mention something about the stock market. And 20 years ago, you couldn't do that. But now, with natural language processing techniques and big data methods, you can do it. And that allows us to see whether there's some relationship between negative news about the stock market on a daily basis. So we can figure out what the, we call it sentiment, but is it a negative article about the market or a positive article? And that allowed us to test the theory that you're describing, which is, well, how does the newspaper affect people's beliefs?

49:26And what we found is that, let's say the stock market went down yesterday, then comes out a negative news article. people really change their expectations about a crash. They think that the crash is a lot more likely if they've just got a negative news article. So then what's also interesting is suppose the stock market went up. Suppose then some journalist writes a positive article about the stock market. You'd think that that would have the same kind of effect on people. Well, it's not as big. The good news, good news situation only has a moderate impact on their optimism. But the bad news, bad news really makes the pessimism shoot through the roof.

50:11So is there a lesson here about long-term investors and that they should perhaps ignore the financial media? Well, I will say that what I've just described to you is about the individual investors, but the institutional investors are pretty savvy, actually, on this. So they're less influenced by the media. So I think that it's important to step back when you're thinking about investment decisions. It's important to step back from the flow of daily news and to go back to some basic principles that you base your investment decisions on. Not everybody's going to have the same principle because not everybody's going to have the same needs.

50:49But you say to yourself, a basic principle is the reason why stocks generate a higher return than bonds is because they're basically they're riskier. And that higher rate of return is because it's a compensation for the risk that you're bearing. So you say, OK, the stock market went down yesterday. What does that tell me with respect to the principle on which I'm investing? I'm in the market because I'm getting a premium for uncertainty and risk. Yesterday, there was a realization of the market that just tells me that it is risky as I expected, okay? And that I have to be in the market for the long term to take advantage of that risk premium.

51:41So instead of thinking to yourself, oh, it's suddenly risky, I'm frightened, I don't like risk. Actually, you do like risk as long as you can tolerate it over the long term, because that's where your benefits are coming from for investing in the stock market. So sometimes it's a bitter pill to swallow because you think, oh, my God, I've got to sell. But the investors that sold after the biggest stock market decline in U.S. capital market history. And by that, I mean, 1987, in October, the market in the US declined by 21, 22%, like in a day, okay? The people that sold after that were sorry forever, because the stock market over the next year or so climbed back out of that hole.

52:33The people that sold out on that day never were able to benefit from the recovery that that market had. Will, I want to come back to bubbles as we come near the end of our questions here. What did you learn about bubble dynamics by hand collecting data from the 1720s? I love that question because I love history. There was a great bubble in 1720 that was tied to the excitement about the Atlantic trade and also the excitement about how it was possible to use the corporate form to create corporations to do all sorts of different kinds of businesses. And that happened in London, Paris, in the Netherlands.

53:11So there was a cross-border boom in stock prices that was as high as eight times to 10 times the price. So you could have bought a stock in 1719 in Paris or London or Amsterdam, and then seen that thing just grow double and double and double. And so here's what we learned. First of all, in that case, we call it a bubble because as you might imagine, all the prices of these stocks dropped. And so in the next few months after they boomed, they crashed. And on the economic side, my colleagues and I who studied this figured out that the bubble had some rational basis, which is that the economy really had changed at this time and that it allowed investors in all those countries really for the first time to participate in this potential giant engine of growth, which was the Atlantic economy.

54:11So that was one driver. My lesson learned there is that when you see a boom in asset prices, or if you see a bubble, go and look for some kind of innovation, financial innovation, technological innovation, because that's probably what's describing this sudden shift in expectations about the stocks. And we can think of a lot of bubbles through time. I always think back to the tech bubble of a couple decades ago, as a time when suddenly people thought, well, with personal computers and other electronic devices, we're going to have huge gains in productivity and things like that. Now, we know that it did transform our economy, but not all stocks benefited from it.

55:03But nevertheless, that enthusiasm drove a lot of people to invest in these new stocks. And sometimes those stocks did bubble. The other thing I learned is that when the stock market has crashed, everybody's looking around for whose fault it is. And they're also making fun of those people that had the belief that the economy had changed. There's a little bit of schadenfreude or schadenfreude, the sense that somebody else, not you, was in trouble. And we saw that after the great financial crisis in 2008. You know, we saw people saying, okay, who are we going to throw in jail for this? Somebody has to be punished.

55:46The financial system is at great fault. So after the 1720 crisis, just like after the 2008 crisis, there was a backlash against finance, against financial innovation, against investing in equities. The equity markets really kind of dried up in France and in the Netherlands. I mean, they didn't completely go away, but suddenly the world turned, and in many countries, they kind of turned against stocks investing for about a century. So we had this lapse from 1720 almost to about the middle of the 19th century when there wasn't a lot of investment opportunity in equities. Interesting. What are the lessons for investors who may be compelled to invest in so-called transformative technologies?

56:35You should understand that along with innovation and novelty comes uncertainty. And you don't know which of those stocks is going to really be the one that pays off. You don't know whether you're going to buy the next Apple, the next Tesla, or whether you have put your money into something that was going to go bust, even though it was an interesting or good idea. And that's been true for a long time, particularly, let's say, with mining stocks is a good example. There is always going to be a mining industry. There are always going to be discoveries of new gold sources or new mineral sources. But which of the companies is going to find that, you're never going to know in advance.

57:16So you have to think about investing in the new economy stocks a little bit like that. Even a diversified portfolio of them is going to have a lot more uncertainty than buying stocks of companies that are tried and true and have been basically making the same thing for decades. That's a good analogy. All right. Well, our last set of questions for you are related to your book, Money Changes Everything, which is fantastic. And we've mentioned it to our listeners a few times. How do you describe what money is? Well, money is actually pretty recent in human history, actually pretty recent even in terms of the time since writing began.

57:54Money, the way we think about it, we think about coins as money. And coins were an extraordinary invention because they're a system of transferring value that has nothing to do with accounts. It's memoryless. Once I hand you a dollar and then you use the dollar to spend, nobody knows where the dollar came from. So it doesn't require a system of keeping accounts. Now, when I say keeping accounts, When you use your credit card, when I use my credit card, what happens? An account somewhere in the ether is debited and another one is credited and it all comes back together at the end of the month and somebody gets a bill and all of this stuff.

58:34That's money that's based on accounting. But coins are money that didn't require that. But money actually, in a more abstract sense, actually, I shouldn't say abstract, it's very solid. Money goes back at least to about, I don't know, 4 ,000 years ago. In the ancient Near East, in the Tigris and the Euphrates area, the Babylonians had money. They just didn't have coins. Babylonians would use silver as their money, and they'd weigh it out. So you'll see in these cuneiform tablets, there will be a description about somebody lending somebody some money, and they'll say, well, this is pure silver.

59:13This is silver that hasn't been measured. They were using silver. Now, silver is interesting because it really doesn't have any practical value at that time. Now we can use it. You know, it has some other things we can use it for, but like for a while, photography. But silver was sort of a pure good that could be stored, transferred, and it was rare. It was mined. And so the Babylonians started to go and trade for silver just so they could have an injection of money into their economy. So that notion of suddenly in an economy, you had to have something more than goods and services that were real things.

59:53You had to have this kind of abstract way of valuing and storing that value. That was really the beginning of what we think of as money. So it wasn't really until about 1500 years after they started to use silver. Finally, we start to see coins that appear in Mesopotamia and in Greece and actually in China as well. So it was a financial revolution when we started to see coins, but money in terms of silver started long before that. Interesting. When you go back through history, would you say that money precedes trusted authorities or do trusted authorities precede money? Well, the way I think about your question is, is there money without a government that says that money has value?

1:00:43I agree, yep. And with commodity money, with silver and gold, even today, we think about that as a store of value that if the government disappears, it will still be something that one could use to buy goods and services. So which came first, the chicken or the egg? those early Mesopotamian societies that were collecting silver so that it could be used to pay for things and so forth, there wasn't one big country that was doing it. They were a set of trading cities and agricultural cities that sometimes were at war with each other and so forth. So there was a common monetary unit that existed even though there was no common government.

1:01:25So I'd say that the money preceded the government. But as soon as governments figured out how a important and useful money was, they started to figure out ways of controlling, creating money themselves, controlling it to the extent that they set up systems of taxation where they demanded that you pay in hard currency and so forth. Well, I mean, I don't have to describe it because that's the way life is today. But you see with early Greek coins back to 300 or 400 BC, a lot of times they'll have a face on it that will tell you that it's from this city or that city. It became something that was connected with the state in a way that just unstamped silver was not.

1:02:10What role does money play in finance? Well, I think that money is really a financial instrument. A lot of people might not say that, but here's why I think money is a financial tool. And finance to me is something that requires that you're passing money into the future, you're passing value into the future. Your investments in stocks are really about you pay for it today when you buy it, and then you're going to get dividends in the future, or you can sell it in the future. If you don't have that dimension of time, you probably are not talking about a financial transaction. Contract is something you agree on today, then there's delivery or some consequence in the future for the different parties.

1:02:58Anyway, long story short, I think money is a financial instrument because it is a store of value that you get it today and you can hold it as long as you want and then spend it in the future. It doesn't generate interest. It is subject to inflation shocks, but it's a way to preserve wealth through time. Now, it has other values as well because it's kind of a, well, it's a unit that you can use to measure relative values. You can decide whether two sheep might be worth one cow because you can figure out how much it costs to buy the sheep and to sell the cow. That's another benefit of it. But that fundamental technology of storing value by putting it into something that doesn't deteriorate and something that will be acceptable by society in the future as valuable, I think that's the essence of what money is.

1:03:55Very interesting. This is a question that I understand if you give a short answer, because it's literally the topic of your book. So I don't expect the full answer. What role do you think finance has played in the development of modern civilization? You're asking a really big question, which is what's the role of finance in the development of civilization? I think about finance as a technology, and it's a tool that has evolved since the first cities, at least, as a way of making things work better. So the most important thing for me is that without finance, you're not able to have big cities. People couldn't live in one place without a technology for storing value, transferring value, agreeing upon value, paying for services.

1:04:39and even today, cities survive by borrowing money, issuing bonds. The first cities that really couldn't live without finance were the first civilizations where you had a lot of people living in one place. But as finance developed, new kinds of cities evolved, cities that couldn't grow their own food but could go and finance ways of getting grain and bringing it to Athens or getting grain and bringing it to Rome. Finance made those possible. And when we think about what the future is, we're trying to imagine what cities might be like going forward. And the only way those cities are going to be built is through some financial arrangements that either through stocks and bonds or other forms of financing are able to bring a lot of capital together into one spot and build a lot of buildings or construct a lot of things for people to live in, create the infrastructure and so forth.

1:05:42So without finance, we wouldn't have that. We wouldn't have infrastructure. We wouldn't have density of settlement. We just wouldn't have the kind of advanced methods of getting along with each other and growing without it. Great answer. That's a great answer. And that's why I love that book. Our final question for you, Will, and it's a bit of a different one. How do you define success in your life? I've had a great opportunity really to pursue my research and my teaching. So I am thrilled with having been able to do both of those two things. I'm just blessed to be in a situation where I can pursue these questions that you've been asking me.

1:06:28I can explore them as a scholar, and then I can try and produce things that are useful to people and meaningful to people who are investing. So that has been a wonderful success for me. I also spend a lot of my time mostly teaching about finance, sometimes about financial history, and that's a rewarding experience to be able to engage with students and to learn together with them. So I guess I should say that I would define success as making more money, but that's always been a bit secondary. Being able to make things and experience things has been the greatest rewards for me. Incredible. Well, thanks so much for joining us.

1:07:10This has been a very fascinating experience. Thanks, Will. Thank you. Yeah, thanks, Will. And thanks for all of your research. It really is incredible. Yeah, I appreciate your interest in it. And this has been an interesting chat.