Why Diversification Beats Stock Picking

It's almost a a mindset, right? Do you start with nothing and then figure out how you would add in a very small number of stocks? Or do you go literally the opposite way, which is, I'll start with everything and then I may want to refine based on research, based on expected return, based on, you know, different preferences. That's fine, but those are two sort of philosophical views on how you might want to approach it. Yeah. Give me a reason not to hold this. Welcome to the Informed Investor, where we break down the latest financial headlines, bringing in research and insights to help you separate the news from the noise. Welcome to the Informed Investor. Today we're talking how many stocks you need in your portfolio? The show is brought to you by Dimensional Fund Advisors who is bringing financial science to investing. I'm Mark Gochnour and I'll be joined today by the dynamic duo of Jake DeKinder and Wes Crill. Good to be back on the show. I'm excited to be here. I thought for a second you were gonna call me Jake and him Wes. I was kinda looking at you when I said Jake- I mean, you did get in late last night. So I was you know, a little tired. But I feel like we've talked about this topic though before. I'm excited. It does seem eerily familiar. Yeah, yeah, yeah. It is a great topic to talk about. I'm gonna start with some headlines. Are you over, under, or perfectly diversified? Over-diversification how much is too much? Jake I think you are probably perfectly diversified. You mean with my investments or in life or what? Investments. I am not going into your personal life. And diversification is not one. Alright, let's talk a little bit about how many stocks do I have to have in my portfolio. Now I went to the source to start this out and then we'll come to you guys. I went to ChatGPT and here's my question. How many stocks do I need to own to be diversified? The answer, 20 to 30 stocks, you capture most diversification benefits, 30 to 50 stocks, diminishing returns, risk reduction barely improves beyond this point. How do you feel about that, Wes? It's just funny because you would've thought this was a settled argument like 60 years ago, where we had this emerging concept of Markowitz and Modern Portfolio Theory, where it suggested you would want to have the market and these days when you can access the entirety of the market, I mean over 40 countries around the globe, more than 10,000 stocks out there just on the equity side, you can do it very inexpensively, very easily in high liquidity fashion. So it's not really clear why you would not want to have as broad a diversification as possible. I'd agree with you on that, I mean it's almost like, to Wes' point, it's like they, Chat looked up articles from the birth of modern finance, right? Which was almost like just a basic math problem, but there was so many other things that they were failing to take into consideration. Also, the evolution to his point, like well think about, think about where we were 60 years ago and where we are today and the opportunity set for investors. I mean it's just absolutely remarkable what you can do in a low cost, broadly diversified, 40 plus countries allocation. So it's weird that still when you look that up on Chat that it's directing investors to, you need 30 stocks. Yeah, but if you go back, I started Dimensional what, almost 28 years ago and taking the CFA, you've taken the CFA and you've probably had in a lot of your different studies, I mean that was a big part of the curriculum was how many stocks do you have to own? And it was 30 names and so it wasn't that long ago where that was kind of the prevailing view. But that's where they're defining risk as standard deviation, right Wes, where they're saying, hey, in 30 stocks, your standard deviation can be making up numbers 14.8 to 15.2. Like it's not meaningfully different if you add another 500 stocks to that. Yeah, I mean it's certainly not a number they're pulling outta the hat. It's like if you were to simulate different strategies and add stocks sequentially and just plot out where their standard deviation is, it becomes kinda asymptotic or starts to level off once you get to 20 or 30 stocks. I know I probably shouldn't have used that one. You looked at me. Can you define that for me? Not a chance. No idea what that means. Levels off like they say. But I think to your point, it's a very, what I'll call overly simplistic view of diversification benefits. Well that was my point of what I meant by it's a math problem, right? Which is if you take a very basic view, you work with this opportunity set, you look at it this way and you say if I go from 30 to 50 to 70, there's sort of this decreasing benefit around standard deviation. But again, that's the point of it's a very limited view of what you might wanna think about as an investor. Well you go back in time too, it was much more expensive to go out there and transact in stocks, right? The spreads were a lot higher. You had commissions that were really, really high and so maybe that was part of it, the expense of it. So you did want to narrow your number of holdings, but let's run with that for a minute just on the 30 stocks and just think how crazy that is really for a strategy. Yeah, well if you have over 40 countries in the world, so I'm not a math major, but I think that implies that in a 30 stock portfolio you can't have one in every single country. So you gotta say, okay, which countries am I gonna pick? Which countries am I going to exclude? I mean even in developed markets, you know, the average return spread between the best performing and the worst performing country is over 50%. So that means there is a pretty big opportunity cost for getting it wrong. I mean last year, you know, if you had to choose between let's say Denmark and Spain, which one are you leaving on the table? Denmark. I'm taking Spain. See this is the expert at market time right here. You chose it based on the food, didn't you? I did. And what did we say before? The beaches too. The beaches, yeah yeah. Not a lot of good beaches presumably in Denmark, but you know, and you got the call correct by the way because it was a difference between, you know, having a negative return and over 80% in the case of Spain, the best performing development market last year. So the point here is that you know, the diversification, okay, it sounds good in theory that you're gonna level off your volatility for a simulated strategy, but if you have to start picking countries, let alone sectors, even within a sector, the individual companies which can have big variation, even this year with Mag 7 in Q1, I mean there was enormous return dispersion between the best performing and worse performing. So that's a lot of things that can go wrong when you start to narrow it down to 30 stocks and for what? It's like you gonna answer your own question. You guys probably don't. I was waiting for it. You probably don't remember Ricky Waters, a football player back in the day. I do remember Ricky Waters, yeah. Who say basically for who for what? When they were asking him why he didn't take a hit from a defender. Oh okay. He was in it for himself. I love the little history lessons you bring it. But actually, I mean it's a great point. Think about that. I'm just gonna say 10,000 plus stocks around the world. We know there's even more than that, right? And you're like, okay, so I'm gonna pick 30 of them. Like really think about that. And what you're actually setting up is you're setting up traditional active management right there and you go back to the very first studies that were run on traditional active management back in the late 1960s. And what you saw is that it didn't outperform, right? And all the way through as the studies continue to go on. So I mean not only do we see the evidence, but really just think about that logically. Like what countries are you gonna invest in? What sectors what I mean and when are you gonna get rid of those companies as well? That's an important point. So you gotta dump some and then buy some new ones. Which ones you gonna pick and then for how long? So let me give you some numbers here and I'm just going to run with kind of what you said up there is to say, alright, I gotta pick a country. Maybe not all of them, but I gotta pick one. Well okay, I have my country, now what? Do I just pick one stock within that country? So here's one recently for China. China, this is year to date returns through April, US dollar, China's down 6% yet you think, "Oh I'll just go get a massive company within that marketplace that'll represent generally the Chinese stock market." So I go buy Tencent Holdings. Well that happens to be down 22% in the first four months. So the risk of getting it wrong is massive. And then going back to your point then, okay, well what if I look at a sector, okay, which stock within a sector do I buy? So this is gonna be a little bit longer around time period here, but if you go 10 years, let's say you just go get a tech stock 10 years ago, Nvidia, I could have picked them, they're up 72% annualized over 10 years, which is a crazy annualized number or Intel up 15%, Netflix up 26%, Disney's up 0.9%. So I mean the risk of getting things wrong, whether it's a sector, whether it's a stock, whether it's country's massive in a small number of holdings like that. Well, and we haven't even really brought in the idea of implementation. You know, when you really think about that in terms of like super small number of names versus more broadly diversified. And when you start thinking about that of how would you actually manage a fund and would you want to have flexibility when you have to go to market and buy and sell stuff. So I think again, going back to some of those original studies, Markowitz and all that of the 30 stocks is is that, I mean that kind of leans towards that academic side, right? Like there's a reality of managing money in the real world and broad diversification can help you tremendously with that. Yeah, I mean we've just been talking about the returns of the securities and standard deviation, but there's other elements of value adding potentially bring to the table. If you don't have diversification constraints, we've talked about securities lending many times on the show and you know, you don't know always which companies are gonna deliver the highest securities lending revenues on into the future. Oftentimes it's very small companies and sometimes they are from sort of less liquid markets like emerging market countries. So again, you know, the more constraints you have in your pro, that's not to say that everyone should be holding every single stock in the market. I think we can probably think of some examples where, you know, you would want to exclude some of the securities from the overall market, but having unnecessary constraints in play can really limit what you're able to do from implementation return standpoint. It's almost a mindset, right? Do you start with nothing and then figure out how you would add in a very small number of stocks or do you go literally the opposite way, which is I'll start with everything and then I may want to refine based on research, based on expected return, based on you know, different preferences. That's fine, but those are two sort of philosophical views on how you might want to approach it. Yeah, give me a reason not to hold this. Yeah. Well expand on that. So let's talk about some of the stocks that you may want to remove from that opportunity set. Maybe there's some with lower expected returns, just a little bit about that. Yeah, I mean we have decades of evidence that especially in the small cap portion of the market companies with high valuations and lower profitability are ones with very high balance sheet asset growth year over year, these have lower average returns. We also expect that based on their characteristics through valuation theory. So that suggests that there is gonna be a subset of stocks that we are willing to, if you call it giving up some diversification, we're willing to exclude them in the pursuit of higher expected returns. And I think it's important to think about that is what is the goal of this investment solution? And you know, once you have the goal that can dictate how much or if you want to deviate from the overall market portfolio. I mean the idea of sort of do you give up diversification, do you think, I mean all of these things are sort of saying what are the trade-offs that you're weighing there? And I think, you know, the challenge again with a super concentrated portfolio is look, if you throw Nvidia in there and you get it right 10 years ago, yeah guess what? Your concentrated portfolio is gonna look really good barring maybe some of your other ones not doing so well, right? And so yeah, you're almost like swinging for the fences and maybe it pays off, but what we know in terms of the evidence of that's a really, really hard route to go. Whereas again, it doesn't mean you have to own every stock in the world, there are trade-offs there, but you may want to think about swinging a little bit more towards that broadly diversified versus, I'm gonna take a flyer on a couple of stocks. So let's run the evolution of that. So that was a pretty common theme back in the day to invest in just a small number of names. We talked about some of the limitations of that. So that led to then this idea of well let's just expand the opportunity set but not buy everything, right? So maybe we start sampling. You see that a lot in index funds where, because it still can't be costly to go buy everything, right? So they may perhaps say, oh let's buy a subset of that or you see other managers perhaps buy a subset of sectors and things. So let's talk about then some of the trade offs as you alluded to there of a sampling approach. Yeah, well that's where you get into the operational constraints of it. So again, you know with an index fund, I mean if you have sampling or basically selecting stocks almost at random from these countries and you could miss out on the ones that deliver the return premiums from those countries. You know, we had an experiment we looked at where over the last 15 years have you missed out on the top 10 return contributors each year in the global equity portfolio dropped your return from a little bit over 10% to about 7%. So there is a big potential cost to to pay there. And then if I'm a traditional active manager, well I might have constraints in my process just because of my investment philosophy where I'm not gonna hold everything out there in the marketplace. So again, you know, you think about what those, those two investment styles are constraining you with. Well, having flexibility in your investment process is really a key here because then I can access all of these different countries that thousands of stocks in a way where it's not imposing excessive costs on my portfolio. I love the point that Wes brings up there. You know when people are like, "Oh the market did well," it's like, yeah, guess what a subset of stocks did well that drove the market return. And that even goes down into, as you start thinking about the premiums, which we've talked about size and value and profitability. You know, when someone's like, oh value did well be like it's a subset of value stocks normally that did well that drove it, it's a subset of small. So it's really a question of if we know that's what we've seen from the evidence, do you think you can pick the ones ahead of time that are gonna deliver it? Think about then if you say, okay, start with that one idea you guys had. If you start with the entire opportunity said let's call it the market, you know, maybe you overweight certain stocks that have higher expected returns, maybe exclude some of those that have very low expected returns, like you said with low profit stocks or perhaps high asset growth stocks. But then with that said, you wanna buy 'em all because a couple things that jumped out where you guys said one, it gives you a lot more flexibility in your trading to manage your trading. You'd be more flexible around that. It gives you flexibility on securities lending revenue, right? Where maybe some of these smaller stocks is really where you earn the revenue globally, particularly in a lot of other countries like emerging markets can be big. And I think about it too of depending how you wanna phrase this thing, but it either by owning all those stocks, it reduces the chance of underperformance because you may, if you're sampling, I might not get the stock that did really well. To your point that usually a subset of stocks are the ones that drive the market. I guess the other way to think about that is you're increasing your odds of outperformance by having as broadly diversified portfolios you can get. Yeah, I mean even if you preserve your expected return, and we've run this kind of research experiments before, but even if you preserve the expected return at the same level but you're forfeiting diversification through a sampling, what you end up with is higher tracking error or return deviation versus the benchmark. And so that means to your point, you have a higher probability of underperforming over any time horizon. And so again that's another, it's another thing you're forfeiting when you don't have to, generally, especially if you have flexibility in your investment approach, you don't have to forfeit all these companies. Well and it's interesting you bring up sort of tracking error tracking difference, right? And I don't think people think a lot about that of okay, I'm seeing the broad market do X and if my portfolio performance is higher or lower a lot of times right, what are we getting it there? Now we're talking about discipline as an investor, right? And this is a really important concept as well of like, yeah, guess what, maybe you hit a home run, right? But if you've got sort of more variation, you've got more tracking or you're thinking about all these things, right? You gotta think about, okay, can I actually stick that out as an investor? And that's a lot of conversations we have with financial advisors and financial professionals where they're weighing all these trade offs around expected returns but also deviation from "the market." Because at the end of the day you gotta keep people in their portfolios. I think about the risk side of things then, you know, if you go back to those original studies like the 30 stock portfolio that was all about standard deviation, which is one measure, right? That's just about the volatility of the portfolio. One measure of risk, but you go back to what are the other risks perhaps for investors? You talked about one tracking error. Do I have the the mental aptitude that discipline to stick with it in case if I'm underperforming the market it could be outliving my money, it could be certain goals you have in retirement, standards of living you wanna maintain. I mean there's all kinds of different risks out there than just standard deviation which comes into play. I'm glad you mentioned the risk management aspect of it because this is where I think you see a little more explicitly the trade offs between you know, your goals and the level of diversification. It's easier to see it there because let's say I'm saving for a near term cash flow. Let's say I'm saving for down payment for a home in the next year. Well that implies that adding maybe a whole bunch of longer dated fixed income strategies to my portfolio are not gonna increase my reliability of hitting that goal. In fact, they might actually detract from my ability to meet that goal. And so in that scenario, I am willing to forfeit the potential benefit, diversification benefit, of having these longer and dated bonds so that I can increase the certainty of hitting my goal. So you know, again you go back to what is the goal within the constraints of that goal or within the definition of that goal, I want to be as diversified as possible, but that means in not necessarily adding more and more stuff that's inconsistent with my goal is gonna be helpful. Well talking about adding more and more stuff, you know we've seen over time that people will throw a bunch of different funds together in a broad allocation and not really understand what's inside of those funds. And so a lot of times at the end of the day you're like all of these funds or a lot of them own a lot of the same stocks. So you're getting at what your goal is in a really inefficient manner and you have this crazy overlap. You gotta understand what's actually inside of your funds and you gotta understand how those funds fit together in a complete allocation. You hear, well I'm diversified, I own 10 different managers. Get those 10 managers own the same stocks- They're holding max seven. You're really not. They're all max seven stocks. Okay listen, go back to the founding of Dimensional. When David and Rex go out and talk to these big institutions, what they found is a lot of them had 10, 15, 20 managers, whatever it was. And they're like, yeah all you guys own the S&P 500, you're literally just trading with each other. Wouldn't you want to own some small stocks there? So, and and again that's not a fault of individual investors. You see it sometimes with some of the largest institutions that might be out there. So you really gotta understand what's under the hood, what does my diversification look like? How do these things play together? And again, role of the financial professional and I mean hugely valuable in what they do. Hey, we're talking about stocks, most of this conversation, let's go back to the bond side of it. Wes, you brought that up a little bit there. How do we think about a, you know, number of names within a bond portfolio, does it matter there how much you're diversified? Well that's where, you know, in the example I was using about near term cash flow needs and the volatility of the portfolio where you know, shorter dated, like having a focus on shorter dated bonds of high credit quality might be sufficient to satisfy that goal. I don't need to add lower tiers of credit quality. I don't need to add longer dated bonds to the portfolio. I already have something that's helping me reliably achieve my goal and expectations. And so the other stuff might, if anything, be counterproductive. Even if you think about like if I have a portfolio that is focused on US government bonds, well adding a whole bunch of different securities that were issued by the US government might not be additive there. It's different if it's a corporate bond portfolio where you might wanna have lots of different issuers in there for diversification. But then if I get 15 different bonds from GE, does that make me more diversified than I was before? Or if I add in bonds from additional companies as well. So you have to think about whether you're actually additive and it's exactly like analogous to what you were saying, which is if you have a whole bunch of different portfolios but they're all holding the same stuff, I'm not increasing my diversification, I'm just increasing the cost and complexity. So almost similar on the bond side, start with your opportunity set and then reduce depending on what your objectives are. Yeah, with bonds I think it's a little more crystal clear just because bonds tend to be the risk management tool in your portfolio and if you have risks that are well specified, which many of us do in our everyday lives, then the appropriate asset class is more well crystallized. Then once that's the case, then you kind of know what the level of appropriate diversification is. Well it's also on the implementation side, it's really similar, right? If you do have that broad diversification then you have more flexibility around the stuff that goes in and outta your portfolio that can add value as well. So there's definitely differences between the stock and the bond market, but a lot of those concepts carry over. Alright, so let me summarize here and I'll almost go back to the beginning of the conversation, which is sort of the evolution of this notion of diversification where it started out with or 30 stock portfolio based on 60-year-old data and research, it's evolved into sampling, which I would go back to say that is still a suboptimal solution because it doesn't give you as much flexibility as you you can in the way you implement. So probably depending on the objective of the portfolio, you wanna diversify as much as you can. That's how I think about the conversation here today. I think that's a really good starting point. Again, it's you know, do you start with nothing and add in, you start with everything and then potentially trim from there. I keep thinking about the story you were telling about the origins of the firm where, you know, 45 years ago it was sufficient to make the argument about diversification to the point where they were even willing to tolerate the underperformance of the asset class and now we have some people saying you only need 30 stocks. It just kind of feels like missing the point. All right, last thing here is I look at your list there, Jake, in your notes, like is that like your shopping list of "Hey, I gotta buy flowers on the way home, get pet food." It definitely could be. Actually every single episode I walk in, I just write a couple of things down, which is markets work, risk and return and related and diversification is your buddy. Just keep those things in mind. You'll probably be just fine. We should make a statue or not a statue. We should make a little poster of that in the back here. Those are awesome concepts to always keep in mind. Alright, the dynamic duo. Thanks guys, that was fun conversation. Thanks for having us. Appreciate it here today. Thanks to all of you for joining the Informed Investor. We appreciate your time here today. Be sure to check out a newsletter from David Booth, who is our chairman here at Dimensional. It's called Stay Calm Investing. You can access the link there in the show notes. And with that, have a fantastic rest of the day. Thanks for joining.