Episode #28: “There is Literally No Logical Reason for Anyone to Have a Preference for Dividends”
Guest: Larry Swedroe. Larry is a principal and director of research for Buckingham, an independent member of the BAM Alliance. Previously, he was vice chairman of Prudential Home Mortgage. Larry was among the first authors to publish a book that explained the science of investing in layman’s terms, “The Only Guide to a Winning Investment Strategy You’ll Ever Need.” He has since authored seven more books. In his role as director of research and as a member of the firm’s Investment Policy Committee and Board of Managers, Larry regularly reviews the findings published in dozens of peer-reviewed financial journals, evaluates the outcomes and uses the result to inform the firm’s formal investment strategy recommendations. He has had his own articles published in the Journal of Accountancy, Journal of Investing, AAII Journal, Personal Financial Planning Monthly and Journal of Indexing.
Topics: As we recorded Episode 28 on Halloween, it starts with Meb referencing his costume from the prior weekend’s festivities. Can you guess what it was? He stayed true to his financial roots, dressing as Sesame Street’s “Count von Count.” (Sorry, no photographs.)
But the guys jump in quickly, beginning with the subject of Larry’s 15th and latest book – “factors.” Larry tells us that the term “factor” is confusing. He defines it as a unique source of risk and expected return. So which factors should an investor use to help him populate his portfolio? Larry believes there are 5 rules to help you evaluate factors: 1) Is the factor “persistent” across long periods of times and regimes? 2) Is it “pervasive”? For instance, does it works across industries, regions, capital structures and so on. 3) Is it “robust”? Does it hold up on its own, and not as a result of data mining? 4) Is it “intuitive”? For instance, is there an explanation? 5) Lastly, it has to be “implementable,” and able to survive trading costs.
The guys then switch to beta. Larry mentions how valuations have been rising over the last century. He references how CAPE has risen over a long period, and points out how some people believe this signifies a bubble. But Larry thinks this rising valuation is reasonable, and tells us why. Meb adds that investors are willing to pay a higher multiple on stocks in low-interest rate environments such as the one we’re in.
Next, Meb directs the conversation toward a sacred cow of investing – dividends. He asks about one particular quote from Larry’s book: “Dividends are not a factor.” Larry pulls no punches, saying, “there is literally no logical reason for anyone to have a preference for dividends…” He believes investors over overpaying for dividend stocks today. He thinks it’s unfortunate the Fed has pushed investors to search for yield, inadvertently taking on far more risk. Dividend stocks are not alternatives to safe income. There’s plenty more on this topic you’ll want to hear.
Eventually the conversation drifts back toward market values. Larry tell us that when the PE ratio of the S&P has been around its average of 16, it has about a 7% expected return. So now that the CAPE is roughly 25, and the expected real return is around 4%, some people are shouting “Sell! Huge crash coming!” Larry disagrees and tells us why.
But the guys just can’t leave dividends alone. They swing back toward the topic, with Larry telling us the whole concept of investors focusing on dividends literally makes no sense. If you want a dividend, create your own by selling the commensurate number of shares.
This leads Meb to discuss a research study he did in which he asked if he could replicate a dividend index with companies that don’t pay any dividends. His research revealed that not only could you, but you can do much better. The takeaway was that for a taxable investor, this investing strategy would be far more efficient.
As the conversation progresses, Meb asks Larry about portfolio construction. Specifically, when he’s building a portfolio, does he pick out individual factors and hang with them for a decade, or does he want to review annually and tilt away, or go multi-factor?
Larry tells us to invest only in something you have a strong belief in. Why? Because every factor can have long stretches of underperformance, so you need to be committed. People think 3 years is long… 5 is very long… and 10 years is an eternity. Larry doesn’t agree. And if you chase returns across shorter time periods, you’ll likely get awful returns.
Next, Meb steers toward the momentum and trend factors, asking what Larry thinks. Larry says “put them to the test.” So he walks them through his aforementioned 5 criteria.
There’s far more in this episode that you don’t want to miss: the correlation of value and momentum… trading costs… the use of CDs in your fixed income allocation… corporate bonds and an eroding risk premium… the state of the ETF industry 10 years from now… There’s even a warning – if a former Miss America is pitching you a mutual fund, beware… In what weird context does that advice apply? Find out in Episode 28.
To listen to Episode #28 on iTunes, click here
To listen to Episode #28 on Stitcher, click here
To listen to Episode #28 on Pocket Casts, click here
To listen to Episode #28 on Google Play, click here
To stream Episode #28, click here
Comments or suggestions? Email us [email protected]
Links from the Episode:
- Your Complete Guide to Factor-Based Investing: The Way Smart Money Invests Today
- All of Larry’s ETF.com columns
- The BAM Alliance
- Twitter @LarrySwedroe
- “Detecting True Factors in the Factor Zoo: Assessing the Economic Rationale behind Equity “Factors
- Triumph of the Optimists: 101 Years of Global Investment Returns
- Modigliani–Miller theorem
- “The John Bogle Stock Valuation Model“
- “How Much Are Those Dividends Costing You?“
Running Segment: “Things I find beautiful, useful or downright magical”:
Transcript of Episode 28:
Welcome Message: Welcome to The Meb Faber Show where the focus is on helping you grow and preserve your wealth. Join us as we discuss the craft of investing and uncover new and profitable ideas all to help you grow wealthier and wiser. Better investing starts here.
Disclaimer: Meb Faber is the co-founder and Chief Investment Officer at Cambria Investment Management. Due to industry regulations, he will not discuss any of Cambria’s funds on this podcast. All opinions expressed by podcast participants are solely their own opinions and do not reflect the opinion of Cambria Investment Management or its affiliates. For more information, visit cambriainvestments.com.
Sponsor: Today’s podcast is sponsored by FoundersCard. I usually don’t spring for paid membership programs, but this one is a little different. The offering is targeted entrepreneurs and business owners, and the card enables premier benefits from leading airlines, hotels, lifestyle brands and business services. A few of my favorite benefits include free access to MailChimp Pro, Dashlane Premium and TripIt Pro. You can even get big discounts of services I love like silver car, 99Designs, Apple and AT&T.
My favorite though are the travel benefits, where you get an automatic status such as Hilton Honors Gold, American Airlines Platinum and Virgin America Gold. While I often use the great app Hotel Tonight for travel, the FoundersCard discounts can be massive too. If you go to founderscard.com/meb, podcast listeners can sign up for the discounted $395 a year with no initiation fee, and that’s a saving from the normal cost of around $600 per year. Again, that’s founderscard.com/meb.
Meb: Happy Halloween listeners. We have a very special guest today. Mr. Larry Swedroe, welcome to the show.
Larry: Hi there. Good to be with you.
Meb: Larry is calling in from St. Louis. I’m here in Los Angeles and I already saw the answer to this because we had Larry on Skype, but Larry, no costume yet? You’re not dressed up for trick or treaters yet?
Larry: I think I’ve passed that stage. I actually have my 65th birthday coming up in a few weeks.
Meb: Well, congratulations. We actually, we never get any at my house in LA which is a little strange because we have a very kind of populated neighborhood. But I was dressed up as Count von Count if you remember from the Muppets this past weekend which was…I’m still getting a little purple face paint out of my ears this morning. Most of the listeners will be very familiar with Larry. His writing ends up kind of many different places online, certainly on etf.com as well as all the financial journals. He works as Director of Research at Buckingham which he’s been at for almost looks like 20 years now, and is a rapidly growing investment manager, manages many billions in AUM.
We’re gonna have a bit of a problem here with the audio because Larry is originally from the Bronx. Speaks a little faster than I do, so most of my listeners, because I speak with such a kind of drawn out monotone, listen to the podcast at two times investment speed. It’s gonna be frustrating because Larry’s gonna be a little quicker to listen to. All right, we’re gonna get started. Larry just put out his 15th, I think, co-authored or authored book. Is that right? Is it 15?
Larry: Yeah, that’s correct.
Meb: My gosh! That’s a whole book shelf. It’s called “Your Complete Guide to Factor-Based Investing,” and I read this over the past week. It’s an excellent book. It’s co-written with Andy Berkin who is at Bridgeway, and this is the second book y’all done together?
Larry: Yeah. Our prior book was “The Incredible Shrinking alpha” which told the story behind the fact that it’s becoming persistently more difficult for active manages to generate alpha. Dropping from about 20% of them doing so on a statistically significant basis 20 years ago pre-tax to about 2% today pre-tax.
Meb: That’s not too high of a number. This book, by the way, is great, and the way that it goes and we’ll talk about it today is it’s kind of both a history and a survey of factor-based investing. It’s a very popular topic these days. Great foreword in the beginning by Cliff Asness, but Larry, why don’t you give us a start as a little bit of an intro? Maybe you guys go through a ton of different factors but you start out with saying look, maybe explain to our listeners a little bit about what we call this factor zoo and the five things you guys were kind of looking out for factors, and maybe even the beginnings and we can even get into a little cap in as well.
Larry: Sure. I think a good place to start is the word factor can be confusing for people not familiar with the literature. I think a good way to think about a factor or an asset class is that it is a unique source of risk and expected returns. It’s got certain common characteristics. We know, for example, value stocks, they can be defined as companies that have low price to earnings or low price to cash flow or low prices to book.
The question is in the literature there at the 600 factors or what at least purport to be unique sources of risk which doesn’t seem even possible that there are that many. In the literature, we ask the question, “How do you determine which factors you ought to investigate, consider investing in and include in your portfolio?” We came up with five key decision rules.
First of all, a factor should have a premium and it should add some explanatory power to the cross-section of returns of equities or bonds as the case may be. Once it meets those two criteria, it also has to be number one, persistent across long periods of time and economic regimes. That means we don’t wanna look at a 20 or a 30-year period and maybe you just got lucky because the risk didn’t show up to that period. We wanna see very long periods of time.
We also want it to be pervasive as the second big criteria. By that, we mean we wanna be sure it’s not a data mining result. We didn’t just tell a computer to find a trait that happens to work. The way you can reduce that risk, of course, is one that exists over a very long period of time, that’s the persistence, but it’s even better if it’s pervasive, meaning, for example, value works across industry, sectors, countries, regions and even asset classes where whatever is cheap tends to outperform whatever is expensive.
We want it to be robust, the various definitions, again, to reduce the risk of data mining. As I just mentioned, value works whether we look at price to book which is the most common metric in the literature, but it works on things like EBITDA, cash flow, sales, dividends. That really helps you to be confident it will persist, but it also happens to be robust across asset classes as well. Value works across stocks, bonds, commodities, etc. Persistent, pervasive, robust, we want it to be intuitive so that we have a reason to expect it will persist, it’s not just, again, data mining. We prefer it to be a risk-based explanation because obviously you cannot arbitrage away risk intuitively.
For example, stocks are riskier than bonds, nobody debates that. Small companies, most people would accept are riskier, nobody debates that. But there are other factors in the literature, momentum being the lead one where there is no risk explanation, and therefore we have to rely on a behavioral want and you wanna make sure that that behavioral want is subject to what in the literature is called “Limits to Arbitrage” that prevents sophisticated investors from correcting the mispricings that result from poor human behavior, and we identify what those limits to arbitrage are.
Finally, it has to be implementable. It has to survive trading costs. By that I mean let’s say the smallest 10% of stocks had a 5% premium, but the bid offer spread was 6%, it wouldn’t do you any good to try to exploit that. Those are our five criteria along with having a premium and adding explanatory power, and then we have identified eight factors that meet those criteria.
Meb: You know, some of the most famous ones that have been published in the literature certainly by a French firm and many others are kind of the big three, was market beta, size and value. Then you talk about two more kind of in the main, I guess I’d call it the big five. Do you wanna talk a little bit about any of those factors and kind of how your thinking may have evolved over the past few decades, but about particularly beta size value and then moving on to kind of some of the new or few you include that we’ll get to as well?
Larry: Sure. Well, I don’t think that we need to discuss beta all that much because it’s so intuitive, it’s persistent, pervasive around the globe, etc. The one thing I would add about beta is that we know the valuations have continued to rise over the decades in the US. We look at metrics like the CAPE 10, for example, the cyclically adjusted price-to-earnings ratio, and its 140-year average is only about 16 or so and today it’s sitting at 27. People think maybe because it’s so far above average that we’re in a bubble and people have been saying that for almost decades now. We’ve spent almost no time in the last 20 years except for the brief period around ’09, I think, where it’s been below its average.
There are many good reasons to argue why that’s wrong, that we should have expected a long-term drift up because the US is a much wealthier nation and capital is less scarce. And therefore you have to pay less for it if you’re a taker of that capital. We have much stronger regulatory regimes making it safer. We didn’t have an FCC before, 1934, I think, or ’33, the FCC was formed. That made it more risky. We have much more strict accounting regulations. We didn’t even have a Federal Reserve for much of the first part of the period. We know they’ve been successful at dampening volatility. One of the things we want to be careful about is how you look at these premiums.
Even if they’ve come down doesn’t mean that they will disappear, and for example the same reason market beta has come down because economic volatility is now less. I would argue that’s a good explanation for why the value and size premiums might be lower than they have been historically because they too are riskier companies who tend to do worse in bad economic times. I think that’s an important point we should talk about.
Meb: It’s interesting when you’re talking about the market beta because you mentioned in your book a reference…by the way there’s probably 100 or 200 references in this book. It’s the best part. It’s kind of like a graduate level class on equity investing, but one of the references was my arguably favorite investment book which is “The Triumph of the Optimists” and it looks back at 115 years of equity returns, and over, I think, 20 markets around the world. We always talk to our listeners and say it’s so useful because you see the US had good returns but certainly not the best and not the worst, but it also demonstrates kind of the outlier events that can happen but the equity risk premium was certainly strong in almost all these markets around the world despite the fact that it can go many decades under-performing certainly bonds or competing investments particularly on a real basis.
I think one of the things you mentioned about CAPE as well, you know, the US has also been in a very kind of mild inflation environment and investors traditionally are much more willing to pay higher multiples on stocks when you’re in that sort of 1% to 4% safe zone. When you get over that 4% inflation hurdle, that’s when people start to get really panicky about not wanting to invest in equities as well. It’s okay. We cover market beta and you started to talk a little bit about size and value, which are two of the most kind of popular and we don’t need to spend much time there if you don’t want because they’re pretty well studied. But one of the interesting things you did mention, and this didn’t happen until the appendix, but while we’re talking about value, was that there’s a lot of traditional value factors such as price to book.
You mentioned is one of the most pervasive and written about ever since the 60s or the Chris database that we have. The kind of modern era of factor and investing, but you put dividend in the appendix and said…and Larry is well known for being an empirical guy and he’ll say what he means. He said, “Dividends are not a factor.” There’s probably a lot of podcast listeners that almost crashed their car right now on the road and so maybe talk about what you meant in the appendix when you said dividend is not a factor.
Larry: Well, I should put out a warning here. I have learned that there’s a religious belief among many investors about dividend so it’s hard to have a conversation with them. Despite the fact that 60 or so years ago, I think 50 or so years ago, a paper was written saying that dividend policy is irrelevant, very simply laid out the case. And one of the examples of it is anyone can self-create a dividend if you want. For 60 years, it’s been an anomaly.
In fact, that is my favorite field of finances, behavioral finances, a wonderful paper by Hersh Shefrin and Meir Statman and they raise this issue, “What explains this behavioral anomaly?” There is literally no logical reason for anyone to have a preference for dividend yet people do. They provide some behavioral explanation for that, makes sense, but it can lead to people over-paying for dividend paying stocks and that’s exactly what I think has happened today, and it’s very easy to see in looking at the valuations of dividend strategy.
I think, Meb, is that it’s really unfortunate the Federal Reserve has with their zero rate policy pushed a lot of people to try to get more yield by taking more risks and they forget yield and return are not the same. We’ve had a ration to things that anything with yield, so high yield bonds, reads, and dividend paying stocks, as in theory, substitutes for a safe fixed income. Well, the research shows that that’s exactly what happens because investors make that mistake. They flood in, rates go up eventually, and then the money comes flying back out because we all know that these are not alternatives for safe fixed income, and then you get a big crash and investors are really taking lots of risks they shouldn’t be.
There literally is nothing in the literature or a logic for people to focus on dividends. Now, I will say this, dividends indirectly can be indicators of other factors. If you have a high dividend yield, you probably have a low price to dividend ratio and a low price to anything will get you a value premium. It turns out that if you rank stocks by dividend yield, you do get a premium, but it is by far the smallest value premium. It is a tiny fraction of what the value premium is if you rank by cash flow or price-to-earnings, price-to-book, price-to-sales, and growing dividends is an indicator of a quality company as you might expect or one that’s more profitable. Indirectly, you’ll get some exposure to these factors by doing it.
The key thing is, if you buy quality stocks that pay dividends or ones that don’t, you have no difference in returns, and if you buy value stocks and rank them by price-to-cash flow or earnings and you separate the dividend payers from the non-dividend payers, you get the same returns. In other words, dividends themselves don’t tell you anything but people believe they do.
Meb: You know, it’s funny. It’s one of the most ingrained beliefs in all of investing. Investors just love getting their dividend checks and we’ve written a lot about it and I’ve seen you have too in one of the behaviors of this sort of push into search for yield over the past 15 years, is that flows change factors, is kind of what we described and dividends in particular over the past 15 years, historically, we’ve traded it that what you call that valuation kind of overlap with value factors.
You’ve demonstrated on etf.com, I think, and elsewhere where you say, “Look, don’t believe us. Pull up a ticker of some of the largest dividend ETFs and look at the underlying valuation of that ETF.” And in many cases, instead of trading at a value discount to the overall market which dividends historically have done, they’re now trading at a valuation premium which is really the first time you’ve seen in the database all the way back to the 1960s.
Larry: Yeah. Even worse and we do focus on this in the book because we get to the issue of what happens to factors when they get published and everyone becomes aware of them. The 2008 market crash and the publication of the literature showing that low beta or low volatility stocks have produced similar returns to all the rest of the stocks but with much less volatility, and they have far outperformed higher returns with less volatility than high beta stocks. Now, that’s an anomaly for an efficient market that shouldn’t be, but it has.
Here’s what people are missing on, again, we lay it out and show how valuations matter. USMV is the most popular low beta ETF, and traditionally over 80% of the time, low beta stocks have been in the value regime, meaning they were low P/E, low price-to-cash flow types of stocks, and when they did so, they provided market like returns with low volatility. The other 20% of the time or so, they were growth stocks and then they don’t provide market like returns. They still provide low volatility but no longer market like returns.
In our book, we looked at the value metrics comparing DFA, Dimensional Fund Advisors, their large value fund at the time the P/E ratio was about 15, it was a full 30% or so higher for these low beta stocks. Now, they were even more growthy than a total market fund, let alone a value fund. And if there’s a value premium, now obviously you’re gonna way underperform. But even worse, they were trading at more than twice the book value and so people just rush in, they buy what’s popular, and they ignore the evidence which is why we laid out the case in that appendix showing that when low beta stocks look like growth stocks, they don’t do so well.
Meb: It’s interesting because investing behavior, and this has probably been going on for as long as investors have been involved in markets, is that they chase what’s hot and what’s working, and this plays out over such a long period. Where you mentioned earlier various factors and I’m thinking of price-to-book especially can go years if not decades of underperforming other factors as well as investment styles. It causes this kind of change in valuation of what becomes dear and then what becomes expensive and vice versa, and so thinking back to the late 90s, no one wanted dividend stocks, but of course that’s everything everyone wants now, you know, kind of rinse and repeat.
There’s been a lot of debates in the financial investing community about valuation of these factors and if it’s possible perhaps to take that into account. Whereas looking at it right now, you say, “Hey, dividends and maybe low vol are more expensive historically.” Kind of what’s your perspective on the Asness or not spectrum as far as you…you know, is it something you look at for say dividends and say, “Okay, that’s interesting, but I’m not gonna really change my investment approach?” Or is it something where you would take a step back and say, “No, I think actually we could start to do a little bit of tilting away from these factors based on them being expensive or cheap?” Do you have any opinions on that topic?
Larry: Yeah. I have a few and let’s come back because I do wanna touch on one important thing on the size factor, so remind me, we’ll come back to that. I’m much more in the Asness camp because the research makes it clear and the logic makes it clear. Here’s what I mean by that. When the P/E ratio or the S&P 500 has been around its historical average of about 16, it’s gotten a bad as 7% real expected return. It’s one of the reasons why the E over P or earnings yield ratio has been as good as any predictor we have of future returns, and the CAPE 10 just smooths that out and it’s been a good a predictor as we have.
You get an earnings yield just like a bond yield, predicts future bond returns. What this means is when the equity prices are now fogging to make the math easy, say the CAPE 10 is 25, so now you’ve got an expected real return of 4%, you get people like John Hussman and Jeremy Grantham who now for the last three, four years have been shouting at the top of the lungs, “Sell stocks. It’s way over-valued. We’re gonna see a 70% crash,” they said, and my point is I don’t think so. It certainly could happen, but not for the reasons they predict.
The fact that valuations are higher, one, I can lay out a case to justify that. We touched on higher wealth, better regulatory regimes, dividends are much lower now. That means if you retain more earnings, growth of earnings should be faster. That alone should push prices up, valuations up.
Meb: Let me interject there because that’s a point that I think is very often misunderstood because a lot of people like you mentioned, when it comes to dividends and buybacks, their brain starts to just shut down. Historically, as you mentioned and I agree with almost everything you’ve said, but historically, dividend yield has been in the mid fours, like it’s like 4.5%, 4.7% historically. Dividend growth about the same amount, so 4.5%, 4.7% historically, it changes over time. Dividend yields are now lower.
Let’s round up and say 2% on the S&P, and a lot of people use that as saying, “Hey look, this is one of the inputs into,” for example Boggles equation, which is stock returns or basically dividend yield, dividend or earnings growth, should historically be the same thing, and then of course change in valuation.
What they don’t understand, and a friend posted some research on this actually today on Twitter, is that because companies are now using buybacks as an alternate means for distributing cash flows, in any given year, it’s half to more than half of actual cash flow distributions, it changes the equation. Dividend yield can still be lower but it changes the dividend growth rate to be higher. It’s two sides of the same coin, but I think a lot of people underestimate the dividend growth rate side of the equation as a part of that whole.
Larry: Well, that not only is true, but think about how foolish it is. Do you really believe that stocks would have only returned 4.7% if no company ever paid out a dividend? I mean, that’s insane, right? Clearly, if you retain the capital, you should be growing faster. That is why the original papers by Modigliani and Miller said, “Dividend policy is irrelevant.” It should be. People think if you don’t get a dividend, then you don’t get a return. Well, crazy as you point out, you can buy back your stock and if I want a dividend, all I have to do is sell the same number of shares as the cash would have been paid out and a dividend, I can create my own dividends. This whole concept of focusing on dividends literally makes no sense and yet it’s focused on all the time. Now, it doesn’t mean dividends can’t be signals that people can use in some way.
Clearly, we know companies hate to cut dividends because they get punished for it, so they keep dividends below where they think they’re sustainable so they never have to cut them or cut them rarely. But that doesn’t tell you much about anything else. That doesn’t tell you what the company would have earned with its capital if they had retained it or, as you said, bought the stock back. To me, it’s almost a hopeless cause because people have this religious belief. I’ve even had people make the argument that stock prices don’t fall when dividends are paid, and I once came up with an example of a company that had paid out over its life like $40 a share in dividends and the stock was at 40, so you mean to say the stock would not be 80 back on the balance sheet, right?
Larry: I mean, it’s ridiculous, but people fight over that.
Meb: We did a new investment study in the last few months and it was incredibly unpopular, it was like crickets. I’ll send it to you where we said, “Look, theoretically, could you replicate with dividends…” You mentioned, historically, they have outperformed the S&P by a percent or two yearly, but that’s largely because they have somewhat of a value tilt. I said, “Theoretically, could you replicate a dividend index with companies, but just companies that don’t pay any dividend?” We went back into the studies with Wes Gray and some others and showed that, yes, you could not only replicate dividends, but do much better than dividend stocks but with companies that pay no dividend. And the takeaway was that it could, theoretically, for a taxable investor, be much more efficient because you’re not paying that quarterly dividend tax. We used to joke and say, “We’re gonna launch an ETF called “The No Yield ETF” that would probably raise about $2 million total because no one would want an ETF with no dividends.”
Larry: Yeah. That’s exactly right. Here’s the full aude pod [SP]. As you note, if you’re a taxable investor, it’s much better to create a self-dividend because you only pay a tax on the portion that’s capital gain not the full dividend. It’s crazy but people ignore it. Now, what we do know is that there are a whole group of stocks that have lousy returns, they’re called lottery tickets in the literature. They tend to be very small cap stocks, they don’t pay dividends. They have very high betas and have very high, what we would call idiosyncratic or unique risks. They dramatically underperform the market.
They tend to be these penny stocks, stocks in bankruptcy, small growth companies that have high investments and low profitability, so of course they don’t pay dividends. All you have to do is screen those out. You don’t have to buy stocks that pay dividends, just screen out the junk and you get much better returns. That’s the key that investors should focus on and stop the focus on dividends. Because if you focus on dividends today, only about 30% of all stocks pay dividends and that means you’re eliminating a great amount of diversification benefit.
Meb: I think you made two important points and then we’ll kind of move on a little bit. I mean, one of the points is that if you’re looking at the lottery ticket stocks is that that is where a lot of these big winners come from. But if you don’t know that ahead of time, if you were to buy a basket of stocks that have kind of the lottery ticket characteristics, you’ll end up with a terrible basket that will lose money, but that’s kind of why people are attracted to them. It’s because they see, “Hey look, there’s stock that had these characteristics. I can make these thousands percents, yada, yada,” and that’s why they probably get bit up a lot. You know, it’s interesting. When you think of…did you have one more you wanted to say before we…?
Larry: Yeah, I do wanna just come back to the size premium. People point out that the size premium has been much smaller than the value premium. What they totally ignore, and that’s why we wrote an appendix on this, is the size premium is smaller because the definition is very different. For a value and every other, one of the factors in the literature, we break them down into the top 30%, the bottom 30% and kind of a middle or a core 40%. For value, you’re buying the cheapest 30% of stocks, and shorting the 30% most expensive. For whatever the reason, Sam and Friends decided to use the top half and bottom half for the size premium. Now, if you use the top 30% and the bottom 30% instead, since there is a monotonic increase in returns as you get smaller, clearly deciles 8 through 10 minus 1 through 3, those returns are a larger than 6 through 10 minus 1 through 5.
The fact there, premium as the academics do it today, taking the top half minus the bottom half, is an annual premium of about 3.3%. If, however, you took it as the way we define every other factor, the premium would be 5.22% which is basically identical to the value premium. It’s one of the strange things that people look at and of course as you noted, if you screen out for junk and get rid of those stocks, so you get rid of those lottery tickets and you buy the bottom 30% of the small stocks, you get very good returns.
Meb: Yeah. We often say in the podcast, we say, “Look, it’s not just me. Everyone is attracted to finding the perfect stock or finding the perfect portfolio,” and we often say, “It’s just as important to avoid the junk.” So avoid the you know top two core tile of expensive, disgusting stuff and skim that off does just as much to help your portfolio as picking the really cheaper one or trying to pick the really cheaper wonderful stock. If you’re thinking about these factors and so we’ve talked about a handful of them, as you’re building a portfolio kind of two questions, one is, you know, we mentioned a little bit about Asness or not, what’s your approach here? Do you say, “Look, we wanna pick out individual factors and we’re gonna allocate to them and take a step back,” and say, “We’re good to go with these for a decade because we believe in them.” Or do you say, “Hey look, maybe we’ll review this annually and if the things are going crazy, we’ll start to tilt away. Or do you go full multi-factor? What’s your approach in building a portfolio?
Larry: Well, first of all, I think it’s important that you should only invest in something you have a very strong belief in, because as you said earlier and we demonstrate in the book, every factor goes through decades, long periods of underperformance, even market beta has if you look at a 20-year rise, and 4% of the 20-year periods we have, have negative premiums. Unfortunately, investors, at least most, think that 3 years is a long time when it comes to investing, 5 years is very long and 10 years is an eternity. When anyone who is familiar with their research knows that when it comes to risky assets, 10 years has to be treated as nothing more than noise, so it’s really critical.
You will end up with much worse results than the very funds you invest in if you chase recency, chase performance. You need discipline. That’s number one, and number two, since I believe there’s no evidence, so I fall into Asness’s camp pretty much here, that there are people who can tell you when which factor will do well when, that you’re better off hyper diversifying. Diversifying across all these factors that we list because they all have premiums and they all have loaded negative correlation to each other, and that the evidence shows as we show in the book, even what are called naive portfolios that are won over in. If you have 5 equity factors, you put 20% of your portfolio in each of them, that is highly likely to give you a really good performance relative to almost any of the strategy. That’s rule number two, discipline, hyper-diversify.
Now, here’s where I would fall slightly, just ever so slightly into Arnott’s camp and I like Cliff Asness’s line that if you’re going to sin, sin a little. I believe based on the evidence, and there’s some wonderful papers on, for example, trying to time the value premium. A fellow named Jim Davis wrote a paper and said, “Whenever the spread between value and growth stocks is wide, invest in value. Whenever it’s narrow than the historical average, switch to growth,” and if you could do that, you ended up with worse results than if you just stayed with value all the time. That’s because as I said always, there isn’t value premium all the time by definition. There should also be an equity premium all the time. It may be lower than it was historically as we said today. We think the equity risk premium is maybe 4% or so, but it used to be seven. But I don’t think you can use that to sell. Now, how do I, with the exception, when you get obvious bubbles, you may wanna act, but boy, you better understand that you could be wrong and wrong for a long time and maybe forever.
I’ve only made a couple of trades in the last 20 plus years. One of them was in 1998 when value stocks were about their historical average in price-to-earnings ratios and the market had gone up there starting to get to catastrophic levels. I personally sold all of my equities except for value around the globe. Now, I was wrong for about two years. Now, many people would throw in the towel, but over the succeeding next three years, we had the largest value premium in history by a wide margin, and we’ve had over the full term, a good value premium. But I think, one, unless you had the same convictions that I have, I wouldn’t do it, and I’d stayed with that strategy.
Meb: What was the client reaction during that period? Was that a painful two years for you or…?
Larry: Yeah, it was a very painful period…now, this was me personally. I did not do that for our clients. What our clients did is we maintained our allocations which meant that we were selling growth parts of our portfolio because we were now out of our rebalancing bands or tolerance ranges and we would sell growth expensive and buy value cheap. When 2000 to 2002 happened, we had less growth so we took less losses there and we had more value and picked up more gains, and for the last eight years, we’ve been doing the reverse, buying more value and selling growth.
Now, we do tilt our portfolios typically depending on the client, but most of our clients have about two-thirds exposure to value in a portfolio, and one-third more core or a market like. Because we believe very strongly in a value, in size premium, so we also tilt that half of the portfolio is small where the market is only 10% small. Here’s my advice. If you can’t resist the temptation to time the market, do it rarely and in small amounts, and only do it at real extremes and a perfect extreme would be approved to me was this, in 2000, tips yields were 4% and the earnings yield on the S&P 500 was under 2.5%. That meant the expected real return to riskless tips was much more than it was on equities. That to me told me that was a bubble.
Meb: That’s the thing. When you talk about bubbles, is it…? By definition, they’re pretty rare, and so most of the time you spend in this kind of normality zone of kind of either normally valued or slightly expensive or slightly inexpensive and it’s kind of boring honestly, but that’s the way that it should be. It’s pretty rare when you have a bubble like the US in the 90s or Japan in the 80s. You know, what we talk about right now with global valuations is we say most of the world is pretty cheap to actually…really, really cheap on the equity side. We think the US equities are one of the more expensive, but again, it’s not a bubble. I mean, bubble to me, you get CAPE ratio of above 40, which has happened many times in many countries around the world. It only happened once in the US which was in the late 90s. You know, 27, we often just means future expected returns are gonna be lower, but by no means is it bubble territory.
Larry: You and I are in complete agreement, Meb. You want to stick with your plan, rebalance and if you can’t resist, you know, don’t do what I did. I’m willing to take that risk. I have a very strong belief system. If you were 50% value, maybe you go to the 60% value. If you’re going to sin, sin a little.
Meb: Well, I think the important thing and this has been the biggest challenge for us is that how do we make it systematic? We’ve done tons of research that shows markets when they’re down 60% to 90%. Great time to be invested. Markets, when they go down multiple years in a row like we’ve had for commodities in emerging markets, usually a great time to be investing. But the question is, how do you put that into play so that you’re not shooting from the hips, you’re not doing it subjectively? You know, Arnott talked about looking to tilt towards markets that were down the most or had the worst trailing performance of 1, 3, 5 and 10 years, which I think is a…
Larry: That’s a value strategy. That’s what that is.
Meb: I think it’s a great example, but in my mind, you have to make it systematic to avoid trying to think about every month or week when things are going great or poorly to kind of go around and muck with it.
Larry: By the way, Meb, here’s another point, that strategy is also a momentum strategy, it’s called the reversal effect. When you have very long periods of either outperformance or underperformance, you tend to get a reversal. Momentum is really kind of a short-term strategy with a long-term reversal.
Meb: Perfect segue. Let’s talk briefly about the momentum factor because you have it and you’ve been doing a little more writing about both momentum and trend following, which by the way Larry did a wonderful job correcting me in one of my papers on the correct definition of momentum and trend following, which I think we’ve since fixed. You talk a little bit about momentum and trend following. What’s your perspective on those two types of strategies for both equities, but also all asset classes?
Larry: Well, I think the way to answer that question, first of all, is to put it to the test that we establish. Does it meet all the criteria of persistent, pervasive, robust, intuitive and implementable and is there a premium? One, the momentum premium and what’s called the cross-section momentum, which is a relative momentum. For your audiences’ benefit, what that means is when you’re executing a cross-sectional momentum strategy, you’re typically buying in whatever the asset class is. If everything is going up, so all stocks are rising, or all commodities are rising, or all currencies outside of a dollar are rising, you buy the 30% that are going up the most, and you go short, even though they’re going up, you go short the ones that are going up, the 30% that are going up the least. If everything is going down, you go short the 30% that are going down the most and long the 30%, the securities that are going down the least.
The reason you wanna do that is you want to eliminate the exposure to the other factors of market beta value or anything else. You wanna isolate this cross-sectional momentum which the literature shows has actually had a bigger premium and a higher Sharpe Ratio than market data, and it’s been more persistent over whatever time frame you look at. Whether it’s one 1:3:5, 10 to 20 years, that’s the evidence we present. And we also lay out the case, why there are behavioral reasons to expect that to continue to occur.
Cross-sectional momentum has not quite as bigger premium but it’s 6% a year, which is larger than all but market beta and the cross-sectional momentum, and very importantly, it’s about 6% a year both up and down markets.
This trend following or absolute momentum, which means you buy what’s going up and short what’s going down, that too has added value because both of these are uncorrelated to all the other factors as we look at. So they’re going to improve the Sharpe Ratio of portfolio which is why I recommend using multi-factor funds and adding these factors into your portfolio construction. Again, nobody knows which factors will do well when, and therefore you wanna diversify. One last thing, very important for especially value investors to understand, is that value and momentum naturally are negatively correlated because how do you get to be a value stock? The prices have done poorly, prices are falling, you now have low prices to some metric. Well, that’s how you get to the negative momentum. A value fund would be buying it and a momentum fund would go shorting it. Clearly, you’re gonna have negative correlation between the two.
If you’re a value investor and you know there’s a momentum premium, you should really want to include momentum in the portfolio because it will dampen the volatility of the portfolio, and similarly, if you’re a momentum investor, you ought to love value because of that. The key is to own them both and you wanna own them both in one fund. Because otherwise, think how foolish this is, you have a momentum fund that’s going short of stock, the value fund you owned is going long gate [SP], you’re paying two expense ratios to each of those fund managers. They’re both incurring trading costs and maybe taxes and you have done nothing. That’s why we like funds such as the AQR multi-style premium funds because they gain exposure specifically to each of the factors by wading them in a portfolio rather than targeting them independently which does not make sense.
Meb: If you wanna find a list, Larry and Andy have a great list in the back of the book called “Implementation, Mutual funds and ETFs” that have a lot of examples of funds that we’re talking about today that’s useful for the price of the book alone. Look, let’s do one or two more quick questions. I don’t wanna hold you up all day. One of the interesting things that you guys talked about in the book was a little bit…we haven’t talked that much about fixed income today, but I actually heard you mention this on Barry’s podcast awhile back. You were talking a little bit, in this case, the book about corporate bonds but also about using CDs in an allocation which is not something that you hear most advisors talk about. Could you comment a little bit on that?
Larry: Yeah. First of all, everyone’s familiar with the equity premiums we’ve discussed. Very few people or less people are familiar that there are two bond premiums which explain virtually all of return. They are the term premium meaning you’re getting paid to take a duration risk or inflation risk, you could think it of, and also there’s a credit premium. The credit premium is defined as the difference in return between long-term government bond and long-term corporate bonds. That premium which is purely an index base has only been 30 basis points. Now, clearly, any mutual fund between its expense ratio and trading costs is going to have expenses of at least that 30 basis points which meant from an implementation perspective, you got no reward for taking the credit risk, and unfortunately, it also, to make matters worse, tends to show up at exactly the wrong time.
In 2000 and ’08 while Treasury bonds went way up in value, corporate bonds went down in value especially if they were high yield or junk bonds. The correlations don’t work well and that’s a natural phenomenon that you should expect because both equities and corporate bonds are claims on corporate assets, and when the rich shows up for equities, they also show up for bonds. Now, here’s the other thing which is the point you were making. In most time periods, CDs which have the same credit risk as US Treasury if you stay within the FBI city limits, but yield much more. Typically, I would say in most periods, 30, 40 basis points, in some periods and when you go out to CDs, as far out as 10 years which are now available. I’ve picked up yields of 80 to 90 basis points more and you save the expense ratio of the mutual fund. Why, if you’re in your own IRA you can control it, would you invest in corporate bonds when there is no evidence that the returns have been worth the incremental risk? They don’t mix well with the equity portion in your portfolio. You have implementation costs and you can get higher yields in CDs.
Now, inside a 401 K, you can’t likely buy CDs, so there you may not have a choice. There, we’d recommend to stick either with treasuries, government agencies or really the very highest great corporate bonds, I’d prefer you avoid them altogether because there’s no evidence that you get rewarded. One last point is one reason you have not been rewarded well is that call risk has not been well rewarded and almost all corporate bonds have that call risk which of course shows up at the wrong time. When rates are falling and stock prices are getting hit, you need that continued high yield from that 20-year bond you bought but it gets called and now it’s an overnight bond and your stock, you have to reinvest at much lower rates. We just don’t buy corporate bonds for our clients and portfolios, we build individual bonds and for taxable dollars or taxable investments, we buy almost exclusively CDs.
Meb: There’s a lot of cool FinTech sites that are starting to pop up and some of them have been around for a while for even individuals wanting to manage their cash balance. I know there’s one called MaxMyInterest, but others that will do such things as they’ll your cash balance in between banks depending on which has the highest rate in which case, you know, now many near zero, but there’s a handful that it’s fully protected up to 1%. That’s another idea out there for people. We’re gonna ask one more question. This one’s from Twitter, we asked people to send in a Twitter question and then start to wind down. Sam from Chicago wants to know what you think the ETF industry is gonna look like 10 years from now. It’s a pretty open-ended question, but fire away.
Larry: I always say my crystal ball is always cloudy when it comes to the stocks, so I’m gonna almost pontier [SP] too because I don’t believe in crystal ball gazing, but I will say this, here’s what I think the ETF industry should look like. One, probably 80% of them if not more should disappear, all these specialty ETFs and leverage ETFs and all of them. There is no role for them in any prudent portfolio. The ETF industry, however, does provide significant benefits for people who can identify the funds that are, one, low costs and we’re in a race to zero here, and I know how to analyze the fund to see if it’s giving them the beta exposure they want in each of these factors in an efficient way, and avoid the dumb beta mistakes. Like I don’t think you want to invest in single factor funds. You want to invest the multi-factor ones that give you those exposures as I pointed out.
If you do that, I think you’re likely to do well especially in taxable accounts. One last thing on ETFs. There’s recent research showing that trading costs are much higher than people think because when you go to execute, you’re likely trading on the wrong side. If they are undervalued, the pros are gonna be there driving the prices back up and you’re likely to pay above price, so people focus on the bid offer spreads when that’s not really what matters. It’s not only the bid offer spread, but where it is trading to its true value, it’s often or more often than people think in many of these ETFs particularly in smaller caps or less liquid stocks are not trading where they think they ought to be. You really want to be careful. Make sure there’s a lot of liquidity in these things. We personally won’t invest in any one that doesn’t have at least $200 million of AUM. Where there’s not several million dollars of trading every day, and we believe the fund sponsor is there for the long-term.
Meb: There’s definitely a lot of silliness in the ETF space. I mean, of course, it’s not just limited to ETFs, there’s a lot of pretty silly mutual funds as well. I remember there used to be…do you remember…and it may still exist. There used to be a mutual fund that would simply rotate on its exposure to the US stock market depending on when Congress was in or out of session, do you remember that one?
Larry: No, I don’t remember that one, but I do know that there is like three times the number of mutual funds as there are stocks these days, which clearly makes no…and about 7% of mutual funds literally disappear every year.
Meb: That’s a good thing. I wish that number was higher. But this congressional fund is funny because the historical research shows that it works, but as far as anomalies go, like how could anyone ever really bet on that, I don’t know, but they used to. One, they charge 2% a year for this service, and then two, is that they would hire like former Miss America girls to man the booths of the various investment conferences, so let that be a lesson. If a former Miss America is pitching your mutual fund that charges 2%, that’s probably two red flags for you right there. You know, it’s interesting. There’s a lot of things in the ETF space that I have on my wish list.
Certainly, I wish you could almost divide some of the products into two different categories. Like you mentioned, a lot of these triple leveraged and inverse funds, I almost wish you could call them something else so they don’t get thrown into the bucket of what we call the good ETFs. And I would love, and I don’t think it’s practically possible, but I would love for the ETF industry to figure out a way to transact at end of day NAV plus whatever a couple of pennies like mutual funds do. I think there’s challenges for that, but I don’t see why you couldn’t do it.
Larry: I agree. I’d say that that would make much more sense for people. There’s literally no reason for an individual to trade an ETF during the day. It’s actually a temptation that’s there because you can do it, that causes people to trade too much which we know is detrimental to their financial wealth. There’s very clear evidence in the research that the more people trade, the worse their results are.
Meb: If you could have applied the tax efficiency of ETFs and the mutual fund end of day structure, to me, that’s like the killer product. I can’t really figure out how to structure it. Maybe one of the listeners who’s smarter than I am will be able to figure that out.
Larry: Well, when you do figure that out, let me know and I’ll be your first investor because I completely agree with it.
Meb: I’ll work on it. One more question we like to ask all of our guests is going back this old post of things I love is that people may or may not be aware of when we ask our guest. Do you have anything particularly useful or beautiful that you think our guests would be interested in?
Larry: Yeah. There are two things I’d recommend. If you’re not familiar with and you’re into analyzing mutual funds and investments, there’s a website called portfoliovisualizer.com. They have a regression tool that allows you to look at what loadings or exposure a fund to each of these factors and it’ll also tell you whether the fund has been generating true alpha or not once you account for the exposure to these factors. So that’s number one. For those who are interested in real finance, I’d recommend following John Cochran’s blog, that’s the only blog, besides Professor Damodaran’s that I follow, besides my own, of course.
Meb: We’re gonna add links to both of those in the show notes. Portfolio Visualizer is a great way, also it has the ability to back test various allocations which is a lot of fun for investors to do particularly because we believe the exact allocation doesn’t matter a whole heck of a lot over time. Larry, it’s been a blast. Thank you so much for coming on the podcast today. Any final words for investors before we let you go back to start giving out candy tonight?
Larry: Well, here’s my advice. If you wanna do well as an investor, one, you need to get educated because education is the key to discipline. Without understanding, there can be no discipline because we know risks show up. Any risk factor or any investment strategy will undergo very long periods of underperformance that literally mean nothing. They are likely noise. That’s number one, and because of that, you want to do what I call hyper-diversify. Diversify, of course, as many factors or unique sources of risk as you can. Recently, for example, I’ve made significant investments in a reinsurance fund.
It was never available before, but to me, it’s almost a perfect investment because it has equity like returns and clearly the risks of insurance are going to be uncorrelated with anything related to the risk of stocks and bonds. That’s like a perfect investment. We are getting better tools all the time. The academics keep improving our knowledge, and the finance professionals keep coming up with new and better products that in many cases used to be the exclusive domain of the hedge fund world where you had to pay 2 and 20 to get them. Now many of these products are an index fund like cheap, but they’re no longer 2 and 20 and therefore I think they’ve become, you know, worthwhile investments for people to consider.
Meb: That’s great. If people wanna find your writing, where is the best place for people to go?
Larry: You can go to our company’s website which is bamadvisor, B-A-M-A-D-V-I-S-O-R.com. I write for three websites, etf.com, mutualfunds.com and also Advisor Prospectus.
Meb: Great. Larry, thanks so much today for taking the time. Listeners, thanks for taking the time to listen, we always welcome feedback and questions for the mailbag at [email protected] themebfabershow.com. As a reminder, you can always find the show notes and other episodes at mebfaber.com/podcast. You can always subscribe to the show on iTunes, and if you’re enjoying the podcast, please leave a review. Thanks for listening friends and good investing.
Sponsor: Today’s podcast is sponsored by the ride sharing app, Lyft. While I only live about two miles from work, my favorite means of getting around traffic-clogged Los Angeles is to use the various ride sharing apps and Lyft is my favorite. Today if you go to lyft.com/invite/meb, you get a free $50 credit to your first rides. Again, that’s lyft.com/invite/meb.