Episode 273: Professor Samuel Hartzmark: Asset Pricing, Behavioural Finance, and Sustainability Rankings

27:15But what our paper suggests is maybe the best case scenario is where all of this money has been flowing into these kind of green investments, but hasn't actually shifted the cost of capital yet. Because if you kind of take our results with impact elasticity seriously, it suggests that it could have a counterproductive effect relative to this green transition goal. So you were mostly looking at divestment. How do you think your findings would change if you looked at ESG integration instead? I think that ESG integration would have a very similar intuition to what we're finding. And truthfully, it's something that in the next draft, we're going to kind of write up a bit more because I think the intuition of exclusion is very straightforward.

28:04And so we've kind of written it using that language. But the integration idea of just overweighting or underweighting, using ESG as an input more holistically in your portfolio, I think everything roughly holds in terms of what you'd expect in terms of the economic impact, influencing the impact elasticity, economically, why you think it might shift around firms' behavior. So I think that almost anywhere in the paper we talk about avoiding brown firms, if you just say overlaying an ESG strategy and taking it into account in your portfolio weights would lead to a similar intuition. So I have to ask the question that I'm sure many listeners have right now.

28:47So if somebody wants to have a positive impact on global sustainability, where should they be allocating their capital? based on your research? So that's a very interesting question. And I don't have like specific funds, but I think there's some broad things that our paper points to that investors can focus on. And I'd say this is a general theme, both with green finance and also some of the later papers we're going to talk about is try to be clear about what it is that you're really trying to achieve. So like if you are an investor who say buying sustainable products for moral reasons, I think the answer might be different than, say, buying sustainable products because you want to have an impact on the environment.

29:29So if it's, say, moral reasons, maybe it makes sense to kind of focus on exclusionary rules that are aligned with your morals. In terms of impact, you probably want to be focusing to some degree on funds that are actually engaging with brown firms because at the end of the day, that's really where the improvement has to come from. And there are some exceptions where maybe funds that currently look green in terms of emissions, someone who is kind of really focusing on impact would invest in. But in general, the improvements need to come from those brown industries. So agriculture is a big emitter.

30:09If you want a green transition goal, you both want to keep feeding people and you want agriculture to emit less. And so you would need to really engage with the agriculture industry. and find firms that maybe were relatively more green within agriculture, try to grow those firms, try to incentivize them to become more green, maybe avoid the firms that aren't doing that. But you can't just say, ignore all of agriculture and expect the insurance firms to kind of make up for the lost food. That would be bad. Also, there are certain things that sustainable investors seem to be trying to do, but aren't doing very well.

30:49So something that could make a lot of sense is to kind of invest in firms that are actually improving. You're kind of trying to incentivize to them to improve. They do improve. They do better. One of the things our paper shows is that it looks like green funds try to do this and also ESG ratings, but they don't do it in the right units. So they actually look at percentage changes in terms of firms that have improved as opposed to shifts in levels. Now, this is a big mistake because the brown firms emit like well over 200 times a higher emissions intensity than the green firms. And so if you treat 1 % the same across each of them, well, the brown firm that decreases 1%, that's much bigger in terms of helping global warming than 100 % impact on the green firms.

31:42But it turns out that people are basically focusing on this wrong unit. So that's so interesting. Yeah. But those I'd say are the big things is, you know, try to engage. Definitely think much more within product, within industry. And if you're going to reward for improvement, use the right units. Wow. Yeah. Super interesting. That paper is just, I'm sure ESG investors will find the paper very difficult. I think it challenges a lot of what's out there, right? Yeah, no. And I, I think too, it's really important. Like these are really important issues. And I think that people rightly have big concerns and really would like to be doing things to improve the environment, to combat these global warming, these just massive issues.

32:29But because of that, I think it's also really important to think hard about, well, how do we actually do that? And empirically look at what's going on and hopefully improve the strategies and hopefully improve the impact it has in the process. All right. I want to move on to another topic that people also get very emotionally invested in, which is dividends. Can you talk about how investors treat price returns and dividend returns differently? Absolutely. Dividend investors are very emotional investors. I actually had another finance professor who I was presenting one of these papers and he went Christmas and his father-in-law was a big dividend investor.

33:07And he was convinced that our arguments would convince him. And he reported back afterwards that they agreed to disagree, that it was too passionate, the love for dividends. So emotions run high, but how do people actually treat prices and returns differently? So the first thing to note is that before we bring in important real world things, they shouldn't be treated differently at all. So this is like a classic finding from the 60s whose rough intuition is correct. So you've got some sort of a dividend payout. Let's say stocks at$10 is going to pay a dividend tomorrow. Well, tomorrow that stock will be worth$9.

33:45You've got a dollar in payment. You're not richer or poorer. You still have$10. If it wasn't paying a dividend, you could have sold a dollar of the stocks. So it should be irrelevant. And then you bring in real world things like taxes and trading costs. And for most U.S. investors, most of the time, dividends will have a little bit of a tax penalty. So if anything, you should probably not necessarily love them. But so how do people actually treat them? Well, they view them as different and not together. And that leads to a big mistake. Because think about what I just told you, right? The thing was$10, now it's$9, you've got a dollar in cash, you're better or worse off.

34:24But if you view, say, dividends as separate, well, you've got a dollar. This is awesome. This is free money. This is income. This is something you can live on. And firms kind of play this up. So they pay dividends in nominal amounts. They try to keep them at levels where they can kind of always pay that amount, maybe slowly increase it. So it seems like a safe, stable, nominal income stream. and there's the price changes. And that's what we think of as in terms of stocks. It can go up, that's awesome. It can go down, that's sad and risky. But you've got these kind of one thing you're holding, it's a stock, and you view it almost as two distinct assets.

35:04This risky price change component in the dividend component is kind of a safe, stable income stream. Between us, Sam, what causes people to believe dividends are so special? I think that some of it is that if you aren't, paying particular attention, it's not very easy to kind of directly identify this theoretical idea. So first off, empirically, it's totally the case that when the dividend gets paid, the price drops by roughly the amount of the dividend. And there's a big literature in the 70s looking like it's almost the whole amount related to taxable stuff, things like that. But it's like intuitively the price drops by the amount of the dividends.

35:45This is just true. But that's true when I've got a data set of like 10 ,000 dividend payments and I run a regression and you get this coefficient close to one with the T stat of like 30. But I'm some investor who's just looking at the five stocks in my portfolio. Well, stocks are really volatile. And so in practice, what happens? Well, you get a couple of basis point dividend payment on a day the market moves 2%, you have no idea how should you think of it as a rational person. Well, you should think in the counterfactual world, market's down 2%, but it would be down 2 % minus those basis point dividends if I hadn't received the dividends.

36:26But you don't really see it. And there's not really an effort to make you see it. So if you think about your brokerage statement, what happens with every brokerage I've ever seen is you get a dividend payment, and it just shows up in cash. And it's not very salient which stock it came from. It's not salient that the price of that stock should be higher. So if you aren't thinking about it, it looks like free money, like this volatile price has moved around. You don't notice where it came from. And that's why we call this the free dividends fallacy. You kind of view dividends as this free bit of money that's distinct from the price level, as opposed to the price dropping by the amount of the dividend, which in practice is exactly what happens.

37:09Yeah, that's crazy. So it sounds like a lot of it's related to salience, where the dividends are salient, but price changes are kind of obscure. Yeah, no, totally. I think a lot of finance and behavioral finance are things that are obvious when you have the big data set and are obvious when you think about the theory. But when you're sitting there with your app and your brokerage statement aren't obvious at all. It's just not very salient. It's hard to find kind of like, you know, what should people be doing? Well, they should be looking at how much more money you have. And we've got a measure of that.

37:44That is the total return. where you kind of consider that dividend payment because it's all just a pot of money. We want more money at the end. We know how to measure that. But then go and open any finance textbook and like page one will be like, this is a return. This is how we think about stuff. And then go and look at your brokerage statement and be confused that there's no return on there. And that just kind of makes the real world a lot more difficult. Super interesting. We rarely have conversations with clients about their income and their capital return because we report on total returns.

38:16I'd say that is improving. Like it is now easier to find that information. In one of our papers, I think it was Interactive Brokers, now reports total return in addition to the other things. But it was in their marketing that they did this. Like we do this. It's a competitive advantage that other people don't. Even sophisticated stuff. So maybe things are different. This was a few years ago. But Bloomberg terminals used to ship with the default being kind of price changes. And you could flip things into a total return thing. So it's like, it's not that it was hidden or anything, but the standard is just often price changes and dividends separately.

38:53And that's how they're shown. And people respond to price changes and dividends separately. Are dividends a safe hedge against uncertain fluctuations in price? No, absolutely not. But again, people kind of think about it that way, right? Because if you view a dividend as separate, then all of a sudden you can view this as kind of a stable asset that's uncorrelated with the price level of the stock. But because the dividend payment comes out of the price level, it's not a hedge at all. It would be like saying, OK, I've got$10 in my right pocket. I'm going to put a dollar bill in my left pocket.

39:32Now I've like hedged something. And it's like, no, you've still got the$10. So it appears that way, but it really isn't. And then, of course, at the extreme level, when prices actually go down a lot, that is when dividends do get cut. So it's really not a meaningful hedge in any way. And at the end of the day, all of this is kind of coming back to the same concept again of like, well, we know how to look at things. Like we look at things at returns because that's the economically meaningful way to do it. if you don't put these things together, all of a sudden you can start thinking about somewhat nonsensical things like this thing that's coming at the expensive price, being a hedge against the price, which doesn't make sense.

40:14I get that. And you just said that it's not a hedge because it comes out of the price, which makes sense to me. I've tried to explain this to people, like we talked to earlier with very emotional dividend investors, and it's really hard to get across. Do you have a way to explain this to someone in a way that's easy to understand? The hedge one is hard. So I think the base intuition of prices and dividends, like I think something like the analogy I give with like, you know, you've got say money in one pocket, you shipped it to the other. I think that is helpful. Yeah. I don't know about the hedges in particular, that one I've had less folks focus on in general.

40:53Why are dividends so hard? So first off, people love them. So it's emotional. It's not very salient. So they kind of get it wrong. Then it also gets conflated with other things that could be reasonable related to the characteristics of dividend paying firms. So there is a fundamental difference between, say, holding a stock because it's a firm that pays dividends and you like that for some reason. You could view it as a signal of ongoing profitability or is a value signal or any number of things that I'm not saying that's good or bad, but that is a coherent thing you could say. Like, I believe the dividend paying firms are a good, safe investment.

41:42You could argue they have alpha and I want to hold them that way. And we can debate, do they have alpha? But that's coherent. As opposed to this firm announced they're going to pay dividend next Tuesday, I really want to buy the stock today so I can get the dividend. And what I found is that I can start to de-bias that, like they're going to pay Tuesday, you shouldn't want it. But people then fall back on, no, but dividend paying firms are good investments or good firms. And then you've got to explain, okay, maybe they are, maybe they aren't. But if they are, and you're, say, a taxable investor, you probably don't want the dividend payment.

42:21So maybe you want to hold the dividend paying stock. Maybe that's your portfolio. And maybe you are an active trader, so maybe you still will receive it. But you shouldn't view the dividend as this extra benefit you get. It's something that for most investors is probably a tax hit that you want to avoid. But it all gets muddled, is what I've found. A hundred percent. You're so right because people exactly fall back on the argument that, no, look, dividend stocks have outperformed the market. And then when you try and say, but no, they don't have a five-factor alpha. And then that's just debt. The conversation can't go anywhere from there.

42:56Oh yeah. But then you'll get like, they're good firms, they're safe, they outperform the market. And therefore the dividend payouts ahead. You're like, no, no, no. But now we've just like broken that dichotomy of like dividend end payment versus dividend characteristic. And yeah, I feel like I've lost years of my life trying to explain this, but it's really tricky. And I think a lot of these behavioral finance type mistakes, one of the common themes is that these are things that if you write down the finance model, they're obvious, right? Like Miller and Medigliani is right. We need to add in real world frictions.

43:34And sorry, that's the dividend irrelevance paper. I don't know if I said that earlier. But then if you go to like a class of MBA students and start talking through this, you start getting like pushback to those ideas and you really see, no, like the core intuition is just wrong. And even common usage. So our paper, the Americans frequent flyer program is the dividends miles program. Like, or they may have rebranded, but it was like, why is that? Well, because dividends are awesome. This is the free money you get from flying. They didn't mean this is like the tax disadvantaged irrelevant program.

44:11They meant this is the free stuff program. And that's how people think about dividends in frequent flyer miles and in the market. Wow. When is the demand for dividends highest? So there's a few different reasons why people demand dividends. I think it comes back to that first question. So if you have the kind of wrong belief that dividends and price levels are separate, you suffer from this free dividends fallacy. So price changes are this kind of risky bet. And dividends are this kind of safe, stable, nominal income stream. So if you view the world that way, you're basically going to like receiving dividend payouts more when that safe, stable income stream looks better.

44:56So when does it look better? When the market's doing poorly, it's going to look better. In our paper, maybe the only time people didn't really like dividends was the tech bubble. That's like the only time we found in the last, I don't know, 40, 50 years or so, where it looked like there weren't systematically dividend investors, which was not active in the market then didn't really pay attention. But everybody I've kind of talked to seems to be like, no, it makes sense. No one cared about dividends in the tech bubble. and our data kind of bears that out. But so the flip side is when the market's doing poorly, clearly you want the safe, stable income stream.

45:31Also, when interest rates are really low. So you're kind of viewing this, again, wrongly as this independent payout, almost like a bond coupon. So when interest rates went towards zero, demand for dividends just went through the roof. And then the last thing is you kind of want this safe, stable, nominal income stream. So the safer you view it, like as long as you think it will keep being paid, or maybe if you think it will even grow, you'll like that even more. How does a preference for dividend yield affect the expected return of dividend-paying stocks? One of the things that's kind of interesting, when you think about the reasons that dividends appear better or worse, they're going to be systematic.

46:15And in general, any sort of behavioral bias or most behavioral biases, I should say, that result in, say, price impact are going to be systematic. So the interest rate's low for everybody. Everybody who kind of suffers from this free dividend fallacy and wants to get this dividend income stream is going to want to hold dividend-paying stocks. So what does that mean? Well, the price of those stocks is going to be bid up at the same time. And so kind of what we found when we look at the data is low interest rate periods, things like book to market ratios, price to earnings ratios, things like that.

46:55Dividend paying stocks relative to non-dividend paying stocks look much more expensive. And then we find that basically by doing so, you're shifting the valuations and you're shifting expected returns. So right after these periods where everyone's kind of piling into dividend paying stocks, returns to dividend paying firms are significantly lower. We find about, say, 2 % to 4 % per year kind of in the period moving from periods of high dividend demand to low dividend demand is kind of that calculation. So there seems to be a pretty sizable shift in terms of demand that does shift around valuations and does tend to lead to kind of underperformance after very high dividend demand periods.

47:41Wow. Those are big numbers. Very big. Crazy stuff. We talked about salience earlier. Can you talk about how mutual funds exploit the preferences of dividend investors? Yeah. So if you think of the Morningstar ratings, like Morningstar is not necessarily getting a benefit out of the salient fact, but you could think of there's something big that investors respond to that lacks some nuance. So you can give that to investors. And maybe if you are trying to maximize your profits, kind of do other things behind it that investors aren't paying attention to. So if you're, say, someone who really likes dividend yields when buying, say, a mutual fund, well, that's kind of a weird thing.

48:28So for a firm, there's all sorts of economic things that could be going on. So we talked about why, say, a dividend payout shouldn't be liked, but maybe there's characteristics of firms, things like that. Now, mutual funds are just pass-through vehicles. So they are just pots of money that hold assets. And so they receive dividends and they pay those dividends out to investors from their investments. Now, mutual funds tend to report the dividend yield on their fund from those dividend payments. Now, what makes mutual funds interesting is they aren't limited by any means to that amount of payment.

49:08Because in order to call something a dividend, they've got to have received it and pay it out. But they're a pot of money. They could always just send more money if what investors wanted is more of a payout. And that same Miller and Modigliani intuition holds of like that payout leaves the fund, the fund drops by the amount of the money. And so there's no reason why, I don't know, a fund with a 2 % dividend yield couldn't give 4 % money out. And as an investor, it's not clear why you should prefer like a 4 % dividend yield to a 2 % dividend yield or something like that. With that said, investors love dividends and they tend to get shown dividend yield.

49:50And investors could be making decisions to buy mutual funds based on that. Now, where this gets interesting in terms of providing a salient characteristic is if you're a fund, you can increase that number. So you can either just say whole dividend paying assets, which is, I think, what a lot of people tend to have in mind in terms of what that dividend yield is capturing. But you don't have to do that. So you could start trading with an eye to when those dividends are being paid out and start systematically trading to increase that dividend yield. And in doing so, you will increase the number that gets shown on that Morningstar page, the prospectus, things like that.

50:35Now, you haven't made more money. You haven't done anything like that. But if investors are going to respond to that, that is something that you could actually be doing. How much does dividend juicing cost investors? It costs a pretty sizable amount for a few reasons. So the most direct thing is just tax costs. So for most investors who are taxable, the specifics matter a little bit in terms of your level of capital gains, because qualified dividends, you pay a tax on the full dividend amount, but at a lower rate than capital gains. If you're at a massive capital gain, this intuition doesn't hold.

51:14But for most investors, you would rather sell a dollar worth of stock than get it through a dividend because you're going to pay tax on the full amount of the dividend versus only the capital gain on the stock you sold. Now, if you haven't held it very long and it's non-qualified, then that goes out the window. You're always better off selling some stock. So just in terms of taxes, we kind of have a back of the envelope envelope calculation from firms who are doing this juicing. We estimate a lower bound of about 60 basis points per year in terms of tax penalty. And that doesn't even begin to take into account things like the funds that are juicing are trading a lot more, which is going to cost you money.

51:56They also charge higher fees, which are going to cost you money. And we don't find any return outperformance. I think the paper finds slightly underperformance as well, too, which intuitively kind of makes sense. These are funds that are providing an economically nonsensical characteristic to a set of investors. If what you're really good at is saying providing alpha or something like that, that's not going to be your clientele. You're kind of catering to a slightly more unsophisticated crowd. What you're good at is selling these characteristics and your funds end up being more costly on ways that are more profitable to the funds.

52:33But also a lot of these folks are probably giving 1 % to the IRS, not even to the fund manager. So there's some big costs associated. It's so crazy. We heard about salience being sold through structured products from past guests and through thematic ETFs. These salient attributes are expensive. Oh, yeah. No, and I think the structured product is another great example of like, there are so many headline things that people like, right? Like a capital guarantee. It sounds great. And the same way, like a dividend yield, like higher is better. But for the structured products, no one ever goes, what am I paying for that capital guarantee?

53:10And, you know, except the hedge fund on the other side who's pricing it and doing quite well. On dividends, do you have any final words of wisdom for our listeners? First off, really think through the logic and don't dismiss the fact that a dividend payout does not make you richer. I'd say that is the first order thing. And it's really hard and it's really tricky. Many people will be on the internet trying to sell you something otherwise, but that base intuition holds and is true and is what's going on. In the second aspect is we really do understand what number you should be looking at, which is some number of total returns.

53:49It could be more complicated after tax total returns. There's further modifications, but as much as possible, try to make sure you are focusing on the right number. And then lastly, make sure you aren't conflating dividend-paying firms as a return predictor versus dividend payouts. So dividend payouts themselves are irrelevant to tax disadvantaged. And that fact is irrelevant or is orthogonal, not related to should I think of dividend-paying firms as a good value proxy or profitability proxy or something like that. That's just a totally separate question that shouldn't be conflated. That was a great segment on dividends.

54:35Dividends are fun to talk about, and they are emotionally charged. Yeah. You talked earlier about price-only indexes and how the Bloomberg terminal ships with the price-only default. Can you talk about how common the use of price-only index data is? Yeah. So first off, just to make sure everyone's on board, a price index just looks at price changes and a total return index puts a dividend back into them, which is super important because as we've talked about in the prior segment, the price drops by the amount of dividends. We care about the total pot of money. You need to add the dividend back in.

55:13So it's important. It's actually big. So give or take a third of market performance is coming from dividends, about two thirds from prices. You got to put them together. Now, in terms of prevalence, since almost every major market index that you see is a price index in terms of the default version that gets shown. So if you open the Wall Street Journal, the Financial Times, you want to know how the US market is doing. The S &P 500 you see, the Dow Jones you see, those are price indices. They ignore dividends. And that's true with most international markets. The big exceptions are the DAX is actually a total return index that gets shown by default in the Brazilian Bovespa.

55:58Both are pretty modern indices. Now, the other thing that's a bit weird is kind of the right version of all these indices now exist. So if you Google S &P 500 total return, that will come up. But if you Google S &P 500, what comes up is the price index. You've got to actively search it out and find it. So we've talked a lot about salience. What is salient for almost every major index that you see? It's a price only return index, not a total return index. And why are price indexes so commonly used as a reference? That's a very good question. I mean, particularly in a world where we've got the other version, right?

56:41Like we've got the version that's kind of calculated correctly. So this is from a paper with Dave Solomon, who's also at Boston College with us. we kind of dug into a little bit trying to understand what's going on. And I think it's largely historical accident. So think the Dow Jones. The Dow Jones, you know, started as a railroad stock index in the 1880s. This was calculated with paper and pencil. And things like keeping track of dividends and returns and things like that isn't particularly straightforward and easy to do. The S &P 500 dates back, I think, to the, I figured it was the late teens, early 20s, was like the initial index that became the S &P 500.

57:21And similar idea, like you've got some poor guy looking at tickers, calculating it by hand. And a price index is an approximation. But why are we still doing it? I think to a first approximation, because that's how we've done it. That's how we've continued to do it. Like, so why does Germany, the DAX, do it the right way? Well, so they used to have the Frankfurt index through, I think, the late 80s, and then they created a new index that became the DAX. And so when they created the new index in a world with Excel and other type tools, they could calculate it the right way. But most of these indices are old.

57:57We keep doing it the way we're doing it. And I think it's just largely historical accident. How does reference to price-only data affect the media's coverage of the market? it. What's interesting about a price index is that it is predictably and systematically wrong based on how high dividends are that day. So that Miller and Medigliani idea of the price dropping by the amount of the dividend that works for firms, it holds for the market. So let's say that the market is paying a dividend yield today of 1%. That means that if you were looking at the total return index, that number you were looking at would be 1 % higher than a day with no dividend because, you know, you should be looking at it.

58:40You're looking at the wrong number. So what we show in our paper is we look at New York Times media coverage of the market and we say, well, what is the media coverage reflecting? Is it reflecting the actual market performance, the total return? Or is it the fact that reporters are looking at the number that's like blaring right out in front of them. Because again, none of this is secret. The real number is there. But what we find is that kind of financial coverage is significantly more negative, the higher the dividend yield is, controlling for the actual level of performance, which intuitively makes sense.

59:16You're looking at the wrong number, you're responding to the wrong number. That's what coverage is actually responding to. How does price return data affect fund flows? So what we look at there is kind of how are mutual fund investors allocating money to funds? And a lot of mutual fund performance is benchmarked relative to things. But before going to kind of the wrong things that's out there, I should say there are now a lot of examples of the right things of, say, total returns to a mutual fund and total returns is an index that are out there. For example, Morningstar has switched over and is doing that, and a lot of other groups are following suit.

59:55But something that's still pretty common to observe is looking at the price of a mutual fund, kind of the change in NAV, relative to the, say, S &P 500 price index. Now, this is now a bizarre mistake for a few different reasons. So one is we shouldn't be using these price indices. We should be looking at total returns. But the other is that a mutual fund price doesn't include dividends, but also doesn't include a lot of other stuff. So, you know, if you distribute capital gains, things like that. And so you've got, say, the wrong number for mutual funds, looking at the prices and not including the distributions, the wrong number for the market, but they're wrong in different ways, but they're commonly displayed next to each other.

1:00:42And so what we find is using this nonsensical benchmark that's commonly displayed, if you're a mutual fund who beats the S &P 500 price index, based on your just shift in price, you end up with significantly higher inflows. So it doesn't really make economic sense. It's the wrong benchmark, but it's commonly shown and people respond to it. That's crazy. So that's like investors responding to not just performance, which is wrong in its own way because performance chasing doesn't make sense, but they're responding to the wrong performance metric. Yeah. So people, they like beating the benchmark, even if it's the wrong benchmark with the wrong performance metric, but the commonly displayed performance metric.

1:01:26And again, kind of common theme is people will look at what's in front of them without necessarily thinking too deeply about whether it makes sense or not. Wow. I mean, I guess it sounds a lot like the sustainability ratings that we talked about earlier. Yeah, yeah, no, exactly. It's like you've got something that's easy and out there. And so So I haven't looked recently, but in the paper, we have Yahoo Finance. What happens if you Google a fund? Well, it shows you the price path. And then what happens if you click a benchmark and it graphs the price index? And it's really obvious for mutual funds because capital gains distributions can be big.

1:02:01And so you see these kind of big deviations, you know, and that's the wrong benchmark. But it's what investors see and they respond to it. How does the prevalence of price only index data affect the return expectations that investors hold? There's this big puzzle in finance related to the equity premium, kind of like historically markets have returned this very high amount. Why is it the case that more people don't own stock? How has it been so high? Things like that. And so one aspect of this, which is consistent with this, not a whole solution by any means, is that people don't really realize how high the return to the market has been.

1:02:40And, you know, we said at the beginning that performance on the market has been about a third of dividends historically. So I think U.S. historical averages give or take 8 % annual returns. But if you look at, say, the S &P 500, it'll be about 5 % because that's the price only aspect. It kind of misses that chunk. And so if you're an investor and what you pay attention to. It's just how is the S &P 500 been every year? Well, you're going to look at a number that's systematically too low. And it leads to some interesting things. It's partially joking, but around the time of Brexit, there was some news coverage along the lines that like, the Brits were just sick of the Germans having better performance than them every year on the market.

1:03:25And of course, the DAX was a total return index and the UK indices were price indices. So the German market basically has this, say, few percent head start every year intended to beat them because they were just different numbers. But people perceived the German market to have higher performance just because the wrong benchmark was being used. Incredible. So let's end this segment like the other two with some advice from you. What can people do to avoid being biased by price index data? The good news with this one relative to kind of the things we've talked about before is the takeaway here is pretty easy, which is it looks like people are kind of looking at the price index, not really thinking about it and responding to it.

1:04:11And that's leading to suboptimal behavior because they're looking at the wrong number. So how do you solve this? Well, you look at the right number. To the extent that you can, you should switch your defaults to total return indices. If you want to look at market performance, you should seek out these total return indices. Once you've got the right number in front of you, you can go ahead and just keep on ignoring the details because now you're actually looking at the appropriate number and you can respond to that appropriate number. And it would be good, I would say, to the extent that the default is changed.

1:04:46I think that would just be a societal benefit. You know, like the Wall Street Journal should be looking at total return numbers. Reporters should be talking about it. Brokerage statements should be showing it. This is just something that the default should be shifted to the total return metric. That would solve a lot of these problems. All right, I want to move on to our last topic, which is one of our favorites, multi-factor asset pricing. Oh, I mean, everyone loves multi-factor asset pricing. Another crowd favorite. How can you not? Can you talk about the theoretical explanation for why some assets have higher expected returns than others?

1:05:21Yeah. So the rough answer is that some assets have higher returns than others and kind of standard asset pricing theory because of an insurance model. So it's like, what do you care about in standard asset pricing is you want to consume and you want to consume in the future. And you always want to make sure that you have enough consumption. You get all this good stuff. you have as much wealth money as you can. So how do you go about doing that? Well, you think about the correlations with assets and kind of what's going to happen in the future. Is your house price going to go down? Are you going to be fired?

1:06:00Is the market going to like all of these sorts of things? And then you value assets more that have less of a correlation or an inverse correlation with those bad states of the world, because you like assets that will pay off when things are bad. So why is it that, say, stocks have a higher return than bonds? Well, because when bad things in the future happen, the bonds, you think, are going to be better insurance than the stocks. Because of that correlation, the price today of stocks will be lower, all else equal. That will lead to higher expected returns, which relates this belief about kind of future states of the world.

1:06:40How do we know whether investors actually allocate their assets in a manner similar to the theory? So I'd say there's kind of two different answers to this question. So the first is, look, if you assume that we've talked through all of that intuition, but we didn't really mean it. It was just kind of some equations that we wrote down. This is what's known as kind of an as-if model, like people kind of behave as if this is what was going on. It doesn't really capture the motive. Then the answer is, we'll just take it to the data, see if it fits the data. The short answer I would say to does it fit the data in terms of those theories is no.

1:07:17There's now a very big literature that's really dug into trying to build versions of these models to fit the data better, but it's very, very difficult to do. The second, and I think much more interesting version of this question is we're taking this model seriously. So this model is really trying to capture the economic problem that investors are trying to solve. They really care about this motive. They really think about payoffs in that way. And if that's the case, I think the answer is, well, just we need to go and talk to people. I mean, to say, well, do you really care about this stuff? Is this how you're thinking about the world?

1:07:58Is this how you're investing money? And that should kind of tell you, is the model making sense or not? How reliable do you think that type of survey data are for understanding people's asset allocation motivations? I think it depends. There are certain questions that the answer is not very well. There's subtle nuance, there's difficult things, there's people that do subconsciously, and clearly we can't ask people about those things. So there are very big limits to survey data. Now, with that being said, you think about what we talked about in terms of asset allocations, fear of future bad states of the world, insurance motives, things like that.

1:08:43These aren't subtleties. These aren't nuances. These are like the first order obvious things. And that should be pretty easy to do. And so this is related to a paper I've got with Alex Chinko and Abby Sussman. And for example, in that paper, we ask some other first order obvious things. Like if you ask investors, do you like things with higher returns? They say, yes, please, with, you know, T stats of like 15. Like it is not subtle. Like people want higher returns. This is obvious. This is first order. Doesn't matter how you ask it. This like falls out of the data. If you say, do you like more risk?

1:09:21They say no, not at all. They don't like volatility. Again, I mean, T stats very high. Doesn't matter how you ask. But if you ask things related to correlations, you don't find it. Now, this is a bit related to kind of this big debate with Milton Friedman, where he has this idea of kind of as-if models, like we're kind of building these as-if models. He's got this famous example of a pool player who goes for a shot and he's a professional billiards player. And Friedman says, look, he can't tell you the details. He just kind of rubs his rabbit foot for luck and hits the ball. And we can model him as if he knows physics and all these crazy equations.

1:10:01And I think a lot of economists would bring that up as an example for why surveys aren't necessarily very good. But actually, I think that example explains exactly why in this setting you should be able to talk to people. It's like, think of a professional billiards player and go and talk to that guy and be like, are you playing billiards or basketball? Which ball are you currently trying to hit and why? What are the rules of billiards? The first order obvious things are first order and obvious and not very difficult to get at. And even if he can't explain which muscle is twitching or the nuance that is going on, if what's going on is kind of these first order obvious characteristics, it should be pretty straightforward to find that people care about this stuff.

1:10:48What do your survey results tell us about asset pricing? So almost every asset pricing that's been written down post kind of the late 70s, at its core has this insurance motive. So it's the idea that people deeply care about correlations of asset payouts is in consumption. And actually, I don't even know if I've mentioned it, but like the big takeaway in the paper is, look, people care deeply about means, people care deeply about volatilities, and we show massive swings to this insurance motive, and nobody cares whatsoever. We show a bunch of experiments showing that. We show versions of kind of open-ended prompts where we can say, write down a paragraph about what you care about.

1:11:34Nobody mentions it. It does not seem to be a kind of aspect of this decision in any meaningful sense. And that says that our asset pricing models are basically missing something fundamental. They're kind of putting in something artificial that shouldn't be there is the first bit. And then at the same time, they're clearly missing something because just like kind of taking these things on their own, you can't explain, like mean and vol alone can't explain asset returns. That's why things shifted to this kind of correlation idea. So this is something that maybe people should be thinking about, but asset prices are set by what people are actually thinking about.

1:12:16And so it suggests that trying to explain asset prices with this kind of correlation hedging motive as a core concern isn't likely to be a successful endeavor. And what the field has done is to take this core idea and layer further complexities on top of it. So the initial one is you care about correlations with consumption. Well, that didn't work. So then you put an interaction of that with another state variable and another thing in a more complicated way. And each of these is just you care about this core concern in a more complicated, more nuanced way. And given that the first order thing isn't there, the kind of second and third order complication seems not particularly likely to explain anything.

1:13:01What do you think are the implications for people pursuing higher expected returns with their portfolios through tilting toward these priced risks? I think it definitely shifts the potential interpretation of these risk factors. So let's take value, for example. Let's ignore the debate about like, is value gone? So let's say we expect value to outperform. Should I invest in value? Well, the consumption-based kind of version of that that you would say from the standard models is, well, look, how do your risks and correlations compare to the market? Because value is there, but it's there because it represents a kind of systematic, unhedgable risk.

1:13:45And so maybe if you aren't as exposed to this as somebody else, it could be a good thing for you to hold some value. Now, about half the people you talk to, you should have the conversation, look, value outperforms. But given your correlation with risk, you actually have a little bit more of a loading on kind of this value risk paper, you should be buying growth. I think in practice and industry, those conversations don't tend to happen that way. And you guys probably know much more about that than I do. But it's like most value investors aren't talking about correlations with systematic risk.

1:14:21They're talking about outperformance. And they're saying, look, this is alpha. Isn't alpha kind of a great thing? And so I do think that it makes it much less likely that what's going on in these risk factors is a systematic correlation with an underlying systematic risk. In the paper, we do do a version where we kind of ask about that explicitly and we find that people don't think about it in that way. And so, yeah, I think that's kind of the main implication there. You're right that those conversations aren't happening, at least from what we see. But you made me feel a lot better because I felt very guilty about not having those conversations ever since we talked to John Cochran.

1:15:01Did John guilt trip you? It wasn't a guilt trip, but I mean, John Campbell, too. They gave us very strong explanations of the hedging motive and exactly what you just said, that maybe some people should be tilting toward growth. So then we've been thinking about how do we have those conversations? Who should actually be tilting toward growth? But yeah, anyway, so now I don't feel as guilty. The value literature, I'd say, really grew up kind of in the late 80s and the 90s. And this is long before I was a professor. But talking to people who were around that, I think the discussion was, look, we found this predictor of returns, especially in that period.

1:15:37Markets reviewed is quite efficient. And so clearly there was a systematic risk that this was responding to. And people didn't understand what it was at the time, but they were quite confident that they would have a good answer moving forward. And now 30 years later, a number of things have kind of been mooted. Most of them, I think, have been discarded. And it's very unclear what systematic risk value could be providing compensation for. And I think that's another, like all of these stories about efficiently priced risk and return mispricing. So the base intuition of our survey is that where there's return predictability in terms of compensation for risk, these shouldn't be mysteries.

1:16:22Like these should be obvious, straightforward things. And in many ways, the academic who writes the paper showing the return mispricing is the last person in the world to find out about this. So, like, what would value be in a rational risk factor type intuition? Well, I mean, look, for the last century, people have viewed these assets as kind of having correlations of a certain type based on whatever that systematic risk is. prices today reflect that future motive in a knowing way right like if you are going to say hedge oil prices because you're southwest airlines and someone goes and says why are you buying these futures contracts you're going to say because i'm hedging oil prices like it's obvious it's straightforward i'm paying to do that today that's what should be going on with something like a risk factor with the value premium, but that doesn't seem to be what's actually going on.

1:17:23People have not been able to tell us why it is this is there. People have been surprised about these empirical facts and whatever kind of their systematic surprise about return predictability that seems much more consistent with a mispricing story than a risk factor type story. We talked to Adriana Robertson about this too, and she'd done a survey with, I think, fairly similar, at least on this part results. Crazy stuff. And we've had those guests or we have talked about the hedging motive. We talked about the ICAPM in an episode and a lot of our podcast audience, a lot of the response was like, well, I don't think people actually think that way.

1:17:59Like I think if you ask somebody why they invest in a growth stock, it's not because they're hedging. It's because they think it's going to have higher returns. Oh yeah. We can also just like kind of read what happened in the past, It's like, go and read Graham and Dodd, like the father of value investing. And they aren't saying, look, this is an appropriately priced risk factor. They're saying, look, through careful research, you can find underpriced stocks. And in modern language, they're going to have alpha because they're mispriced. And that seems to be what most people think they're doing in value investing.

1:18:33Super, super interesting stuff. That was awesome. That was a fun paper. We hit a lot of interesting things and a lot of deep thoughts about asset pricing models, which was fun to think about, fun to do, fun to present, a lot of different opinions. So it was good. Cool. Our final question, which is a fan favorite, much like dividends, how do you define success in your life? You shot me this question, and it's a very difficult one. And it's one that I kind of wish that I thought more about in the moment. So I've got a six-month-old at home. So my time for deep reflection has not been as copious as I would like.

1:19:10So I'd say the first thing that I've come to, I'd say, as life has gone on, is that it's less about concrete goals. So I think my 18-year-old self would have really focused on things. You want to achieve X or you want, I don't know, a certain amount of money or certain like that. But I don't know, life throws curveballs, life's changing, life's uncertain. And so as much as possible, I think I'm now kind of focusing on more like concepts and broader goals that even if life changes, even if things are different, are still things that are achievable and are important and kind of hold throughout those different things.

1:19:52And so with that said, it's like I was trying to think of a non-cheesy way of like talking about broad goals. And I don't know, I think I can. It's like I've got these young kids, so I really want to be a broadly defined good father and a husband. Broadly defined, I would like to that others consider me a good person and that I treat people fairly, things like that. I'm very much an educator. So I hope that I can actually help people in some small amount to educate them to make better decisions. You know, I think finance is important, not just for hedge funds and alphas and things like that, but like, hopefully I can help people in some small way, I don't know, make better retirement decisions or do better in their portfolios and have some fun doing it.

1:20:39So hopefully do some things I find intellectually satisfying and kind of enjoy the ride as it goes. So I was struggling for something more concrete, but I think that was about the best I could have come up with these days. It's a great answer to the question. Thanks. It's a really good question. Now I want to go back to your other podcasts and kind of see what other people, there was the other thing is I didn't want to cheat. So I felt like I totally could have gone and like looked at the last thing of like 20 of your podcasts and correlated the most awesome answer, but I didn't do that. Oh, your answer was great.

1:21:12All right, Sam, that's it. This has been awesome. We really, really appreciate you coming on the podcast. No, no, this has been an absolute blast. And thank you guys again for spending the time to go through all the papers and come up with these great questions. It's totally my pleasure to chat and do this. So anytime, this is great. Awesome. Thanks, Sam. Cool.