Gerard O’Reilly serves as Co-Chief Executive Officer and Co-Chief Investment Officer of Dimensional Fund Advisors LP and is a Dimensional Director. He was named Co-CEO in 2017 and has been with Dimensional since 2004. Along with Co-CEO Dave Butler, Gerard guides the firm’s vision and strategy through daily oversight of the company’s people and processes. As Co-CIO, he drives innovation across Dimensional’s investment offering with fellow Co-CIO Savina Rizova. Since joining the firm, Gerard has been instrumental in the evolution of Dimensional’s systematic approach to creating and implementing investment solutions.
He was formerly Head of Research, managing the firm’s rigorous, scientific approach to interpreting, testing, and applying research in portfolios. He is a member of the firm’s Investment Research Committee, which reviews potential investment strategy enhancements, and a member of the Investment Committee, which oversees portfolio and account management. He also serves on the Boards of Dimensional’s US mutual funds and exchange-traded funds.
Prior to joining Dimensional, Gerard obtained his PhD in aeronautics from the California Institute of Technology. He holds a master of science degree in high performance computing from Trinity College Dublin in his native country of Ireland, where he also received a first class honors degree in theoretical physics and was a recipient of the Foundation Scholarship.
What if “just buying the market” isn’t the end of the story? In this episode, we are joined by Gerard O’Reilly, Co-CEO and Co-CIO of Dimensional Fund Advisors, for a deep dive into what really drives net investment returns. Gerard returns to the Rational Reminder podcast to explain the key principles that differentiate Dimensional’s approach from traditional indexing—and why implementation, flexibility, and detail matter so much more than investors might think. We explore the concept of hidden costs in index investing, how index reconstitution and trading frictions erode returns, and the nuanced decisions that shape a market portfolio: defining the market, excluding low-returning stocks, optimizing tax efficiency, and more. Gerard breaks down how Dimensional’s rules-based, evidence-backed process improves outcomes through smart exclusions (like IPOs and high asset-growth firms), precise trading, securities lending, and better handling of corporate actions. From the dangers of chasing low fees to the surprising benefits of thoughtful execution, this conversation is a masterclass in next-level investing.
Key Points From This Episode:
(0:01:07) Why Gerard was invited back: Dimensional’s approach to hidden costs and net returns.
(0:02:38) Looking beyond “index good, fees bad”—why investors should dig deeper.
(0:04:21) Gerard’s background: From Caltech rocket scientist to Dimensional co-CEO.
(0:06:22) How Dimensional differs from market-cap weighted index funds.
(0:08:42) Four components of net returns: Two increase returns, two decrease them.
(0:12:45) Defining the market: Free float, liquidity thresholds, and dynamic inclusion.
(0:17:52) How small-cap index definitions can create return differentials as high as 10%.
(0:22:03) What securities Dimensional excludes—and why: low-profitability growth, high asset growth, IPOs, and REITs.
(0:29:26) Why IPOs are excluded for 6–12 months and the mechanics behind inclusion.
(0:33:16) Why Dimensional’s exclusions aren’t like traditional active management.
(0:35:09) The “Great British Bake-Off” analogy: baking better portfolios with the same ingredients.
(0:38:13) How securities lending boosts returns—and how Dimensional does it better.
(0:42:09) Managing corporate actions (like M&A) to reduce cash drag.
(0:45:18) How Dimensional deals with buybacks and new share issuance.
(0:47:29) Momentum, short-term reversals, and securities lending fees as trading signals.
(0:50:36) Why Dimensional may lend out stocks that have negative momentum.
(0:52:42) How trading costs affect net returns and Dimensional’s execution edge.
(0:56:06) Hidden costs of indexing: Index fund rebalancing and price impact.
(1:03:19) Why focusing solely on fees is misleading—and what “value for service” really means.
(1:06:18) DFUS: A case study of Dimensional’s market series outperforming index funds.
(1:08:44) How Dimensional builds portfolios with intentional tilts toward higher expected returns.
(1:12:35) What excites Gerard: Expanding access, ETF innovations, and global growth.
Read The Transcript:
Ben Felix: This is the Rational Reminder Podcast, a weekly reality check on sensible investing and financial decision making from two Canadians. We are hosted by me, Benjamin Felix, Chief Investment Officer at PWL Capital and Cameron Passmore, Chief Executive Officer at PWL Capital.
Cameron Passmore: Welcome to episode 360. Today is another amazing conversation. Ben, we had a chance to talk with Gerard O’Reilly, who second time on the podcast.
He is the Co-Chief Executive Officer and Chief Investment Officer, I guess, Co-Chief Investment Officer at Dimensional Fund Advisors. We've been working with Dimensional for 23 years this fall since it came to Canada. Before then, our firm, we were working with them using their research for a couple of years prior to that.
Obviously, regular listeners know this. We're big fans of who Dimensional is and what they believe in and some of the ideas that they've brought forward over the years. Why don't you set up why you Gerard to come back on the podcast and I'll tell a bit of a reminiscent story afterwards.
Ben Felix: People know that we did a relatively recent episode with Marco Sammon from Harvard talking about the implicit costs of index reconstitution. When market composition changes and indexes reflect those changes and index funds follow suit, that there are some pretty meaningful costs to those transactions, to those changes relative to a more patient and flexible approach to investing, which is similar to what Dimensional does. It's similar to the approach that they take.
I found Marco's paper to be just fascinating. Gerard had given a talk at the Canadian Advanced Conference that I attended last year on the components of net returns. Instead of just looking at the fees, which are important and we talked about that in the episode, but instead of just looking at the fees and even just the market exposures, just thinking about all of the different things that can contribute to the net returns that an investor in a fund, an ETF or a mutual fund ends up receiving.
That talk from Gerard was great and I just thought it fit so cleanly with what we've been thinking about with respect to Marco Sammon's research. Hidden costs sounds like such a click-baity thing to say, but the harder to see in the implicit costs of index investing, but just investing in general. This topic fit really well with that.
We spoke to Dimensional about having somebody on to talk about it. They decided that this is such an important topic to them and such an important topic to investors that they wanted to supply us with Gerard as the person to speak about it, who's probably the best person in the world to speak on that topic.
Cameron Passmore: Exactly. What he said at the end really struck home with me. Investors looking beyond the wrapper and learning about the solution.
I think back to the mid-90s when we went fee-based and realized that we knew this, but being paid by the product was sub-optimal. Once you disentangle the compensation from the product, all of a sudden, wow, we can look at indexing. Once you look at indexing and follow the ideas, this is certainly at the time, all roads led to Dimensional because of people behind it and the ideas behind it.
So many people since then in the marketplace act about indexing good, fees bad, let's just go to broad market, simple index fund, which is much better than what was often offered or suggested or recommended to investors. I'm okay with that, but just to put this index good, fees bad, without digging deeper into the components net returns, I just think you're just hitting the easy button and not taking that extra step to learn more. This is the perfect conversation to learn more about those components and not just simply hit the easy button.
Ben Felix: Totally agree on all that. Gerard talks about this too, that the pendulum may be swinging back a little bit where I think investors are a little bit more open-minded now to the fact that indexing is great and it's taken us a long way from high fee active management, but it's not necessarily the best way to invest. It's not the only way to invest in a low cost systematic and evidence-based way.
When he mentions the fees, it blows your mind.
Cameron Passmore: I think back 25 years ago, the fees would be whatever, 200, 300 basis points. Now you get these unbelievable, whether it's indexing or what Dimensional does for three to nine basis points on an institutional level or a bit more in a retail level, it's incredible time to take advantage of what's been learned.
Ben Felix: Should we talk briefly about Gerard's background? Probably. People who listened to the last episode of him would know this, but he joined Dimensional in 2004, became the co-CEO in 2017 and works alongside Dave Butler in that role.
He leads the global investment strategies and the firm's overall direction. He completed a master's in high-performance computing and then a PhD in aeronautics from Caltech. He's quite literally a rocket scientist.
He went from academia to finance when he learned about Dimensional, which led to a role in their research department. Then he's progressed through the ranks at Dimensional from there, went from the head of research to the co-CIO and ultimately the co-CEO. He's a brilliant guy, as you can hear from the conversation.
His perspectives on portfolio management and how Dimensional compares to indexing, just the general thoughtfulness that you can hear through the conversation about how Dimensional implements portfolios, I think, is really useful, valuable information to hear.
Cameron Passmore: I agree. As I said, he's a pretty cool guy. Again, full disclosure, we've worked with Dimensional for many years.
We use them regularly with our clients in their portfolios. They've not paid for this podcast appearance and they do not pay us anything. This is a professional, fully arm's-length relationship.
Ben Felix: That's a good point. We're not paid by Dimensional in any way for this podcast or professionally. There's no conflict of interest other than us believing that this is a really sensible way to invest and us using their products.
Dimensional funds are available as ETFs in the US at least, which Canadians can buy if they wanted to as well. Just to say, we don't have a conflict to say this is good information or that these products are good. It just happens to be what we believe and we think it's good information for our podcast listeners to hear.
Cameron Passmore: All right.
Ben Felix: Good to go. Let's go to the episode with Gerard O'Reilly. Gerard O'Reilly, welcome back to the Rational Reminder podcast.
Gerard O’Reilly: Ben, Cameron, great to be back. Good to see you again.
Cameron Passmore: Likewise. Great to see you.
Ben Felix: Definitely. All right, Gerard. Most investors are familiar with market cap weighted index funds.
How do you articulate Dimensional's similarities and differences with that approach?
Gerard O’Reilly: Yeah, I always start with investment philosophy, Ben. There's two things about an investment philosophy. One, that you have one because you're going to go through time periods where returns are strong and when they're disappointing.
You need to have something to ground you so you can become a long-term investor. Two, it gives you a framework to make well thought out investment decisions. When you look at Dimensional's investment philosophy, there's probably three core components.
One is that a rules-based approach is a good approach. That means that you can articulate what the rules are to financial professionals. You can set expectations appropriately.
You can give tools for financial professionals to monitor that you've done what you said you would do. So we think a rules-based approach is a good approach. Second item, that prices are predictions of the future.
They're forward-looking. There's a lot of information in prices. Use it to improve returns or manage risk.
Prices change daily. That means that you need to have a daily process if you want to use the information and market prices effectively. And the third is that flexibility has value.
So markets change every day. You may want to adjust your portfolio a little bit every day. That flexibility can really reduce frictions, reduce implementation costs.
It has a lot of value if you use it in the right way. So when people think of Dimensional, you compare us to an index-based approach. We're both similar in that we start with a set of rules.
Dimensional's rules are probably, in my view, more detailed, more involved, more dynamic, more able to cope with different market conditions. And then we probably depart from there in that we have a daily approach because we're using prices to improve returns, manage risks every single day. So a little bit of turnover every day, unlike an index-based approach.
And we have flexibility in how we make those trade-offs day in, day out, whether it's what stocks we want to buy or sell, what bonds we want to buy or sell, how we trade those, how we minimize price impact. So all of those things make a difference in the long pull to net returns, and I think are a hallmark of what we do. We consider ourselves master implementers in terms of how we can take an investment thesis and bring it to life in the real world.
Ben Felix: That's what we want to focus on for the rest of this conversation is net returns, which is something we've been thinking a lot about. Can you talk about what the components of net returns for an investment product are?
Gerard O’Reilly: I can, and I'm going to start with a little analogy, a story from my misspent youth, let's say. Back in the year 2000, I was 24, so a lot younger than I am now. And my wife would probably argue with this, a lot more dumb than I am now.
She might say, no, not really. But I wanted to start learning how to ski. So I moved to the US in '98.
I was in grad school. Most of my friends were Europeans, French, Spanish, Swiss folks. They all could ski.
And I said, well, this season I'm going to learn how to ski. I never skied before because it wasn't common in Ireland. And so I didn't have a lot of money.
And so what did I do? I went on Craigslist and I looked to see what could I get for the least amount of dollars. And I ended up buying skis, bindings, boots, and poles for a hundred bucks.
And I often say that's about the most expensive $2,000 that I've ever saved because I took off not knowing anything about skiing, went off with my friends and within about five or 10 minutes, got the ski trapped, fell sideways and tore my MCL and ACL, so two ligaments in my knee, and got shipped down to the hospital. And they looked at the skis and they said, where'd you get those skis? They're from the seventies.
And the bindings had no release mechanism. Once you get your ski trapped and you fall, you were done for. It's an interesting analogy to investing because here's me, I knew nothing about skiing.
And rather than asking somebody for advice, what ski lens should I look for? What should I look for in bindings? What should I look for in boots?
How should the boots fit? What kind of lessons should I get? I just went straight ahead and tried to do it myself.
Now there's an example of where the price that I paid was not the cost of the entire experience. I had two surgeries, lots of PT. I still do extra work on my right leg to keep it up to strength with my left leg.
This is 25 years later. As I say, the most expensive $2,000 that I ever saved. Now, when it comes to the components, net returns, we often think of four groupings.
There's two that increase returns and two that decrease returns. The two that increase returns. One is asset allocation.
How do you weight the securities in the portfolio? What securities do you choose to exclude? And how do you make sure you get the maximum amount of market exposure possible given the mandate?
That's asset allocation. Then you have a second piece, which is implementation. How do you decide when to buy a security, when to hold it, when to sell it?
What do you do with securities lending to maximize the revenue for the end investor? How do you decide on different types of corporate actions and your stewardship approach to improve returns for the end investor? Those two groupings, they increase returns if you do them right.
Then you have two groupings that decrease returns. One grouping is around implementation costs. What are the explicit and implicit costs to trade?
What do taxes look like? Are there ways that you can make yourself more tax efficient? What opportunity costs or style drift do you have in the strategy?
They're implementation costs. And then you have the expense ratio. The expense ratio reduce returns.
Now, a lot of people focus only on the expense ratio. And they say, lowest expense ratio is what I want for my clients. And I think that you miss large portions of what drives net returns when the focus is only on expense ratio.
And we're going to get into a lot of that here today. But I often hear that, well, I'll go for the lowest commodity type pricing in areas where I don't think there's much value to be added, like the US market. And again, we're going to talk a lot about that today.
But even in the most liquid market in the world, there is still plenty to be gained by implementation. The analogy that I would use is, why would you put your clients in $100 skis? Why?
There's no good reason. It may work out just fine. And if you pay $2,000 on the skis, it may work out poorly.
There's no guarantee. But why would you not put them in the best position to succeed? And I think that's the same analogy here.
When you have a myopic focus on expense ratio, and you forget about the other components of net returns, you may be leaving returns on the table that your clients could quite reasonably enjoy.
Cameron Passmore: So let's talk about the market. What decisions go into defining the market when creating an investment product?
Gerard O’Reilly: So let's start at a very high level. At a very high level, really, it's how low do you want to go? And what do I mean by that?
It's what market cap do you want to include? What lowest level of price do you want to include? What lowest level of liquidity do you want to include?
Those types of decisions can go into defining what's the eligible market for this investment strategy. Differences in those decisions can lead to large differences in the number of securities that you might want to include in an eligible market. But that's kind of at a high level, there are some things that define the market.
Now let's dig down a level. Digging down a level, let's think about free float. What's free float?
Well, free float is this concept whereby a company may have a $1 million market cap, tiny company, less true, $1 billion market cap, still a small company, a billion dollar market cap, but not all of that market cap is available to outside investors. Let's say half of it is held by inside investors. And so the free float might be half.
And so how you choose to weight that in the market then may be at 500 million rather than a billion. Well, the free float and how it's done can make a big difference in portfolio weights. So I'll give you an example.
If you take the global market and you don't free float any security, the U.S. weight is about 50%. If you free float securities, typically the free float for non-U.S. companies is lower than the free float for U.S. companies. So the U.S. weight goes to about 65%. It can make a big difference in terms of the composition of the overall portfolio, something that a lot of people I don't think appreciate fully. The last comment that I make then is that the market is dynamic. It changes every day.
And therefore, whatever definition that you have of the market, it needs to take into account the latest information so that it's current. I'll give you some examples. Globally, there's probably about a thousand mergers announced a year in public markets around the world.
And that means that companies are being bought by other companies and that changes the nature of the market. You have to pay attention. Again, globally, if you think about IPOs, a little bit fewer in the U.S., maybe more outside, maybe on the order of about a thousand IPOs in any given year, globally, on average, sometimes higher, sometimes lower. When you look at a plain market portfolio and you think about the dividends and you think about the corporate actions and those types of events, you're talking about five to 10% turnover in a plain market portfolio. The market changes every day. That means that whatever your definition of should recognize that it's evolving, it's dynamic, it's something that needs to be looked at every single day.
How low you want to go, I think is the biggest input into what defines the market.
Ben Felix: It's so common to hear people say, just buy the market. But as you just noted, there's a lot of different ways to think about that. You mentioned the free float example, maybe other examples.
How much of a difference can there be between different definitions of the market?
Gerard O’Reilly: You might look to different index providers there for different definitions of free float, because every index provider, if that's what you're using as your step in or stand in for the market, will have a slightly different definition of free float and they'll apply different criteria. Some of them are a little bit subjective. Some of them are very reasonable.
Like for example, for Thai securities, there's foreign withholding limits. So there's only a certain amount of a company that foreigners can hold. So if you're an index provider providing an index for investors all over the world, you're going to take into account that foreign withholding limit.
But then there's probably a little bit more subjectivity in how much might be held by insiders and things of that nature that you may have slight differences across index providers. To give you a sense, just in terms of how different it might be, not just associated with free float, but in general, if you were to look at different U.S. broad market indices, so different U.S. broad market indices, you probably have something like the Russell 3000 or the CRSP market index, some indices like that, or MSCI broad market index. And you look back over the past 20 years, you would have had on average a return difference between those three indices, between the highest returning one in each year and the lowest returning one of each year of about 30 basis points or 0.3 of a percent as the average. But it's been as high as about 1%. And that's in the U.S. with just a single market, where the free float probably lines up more similarly across the different providers. You go to non-U.S. developed, that average difference jumps up to about 1%, where you'll have differences in any given year of 3% or 4%. And there you would look at FTSE indices, you look at MSCI indices, S&P indices, they all have broad market indices. And if you go to emerging, it jumps up to close to 3% is the average return difference between the highest and lowest on any given year, with some years as high as 6%, 7%. That gives you an idea of how these small differences in definition can lead to differences in returns in any given time period.
Ben Felix: Yeah, it's crazy. And those were total market examples. How much more pronounced can those differences be if we go to like value or small cap value indexes?
Gerard O’Reilly: Much bigger. And when you go to small cap indices, let's take the U.S. as an example again in small cap indices. If you look back, you take the CRSP small cap indices, the Russell 2000 and S&P small cap index, and you look at the highest to lowest each year.
And then you average that over the past 20 years, it's been about 5%. And in some years, it's been higher than 10% that difference in terms of the highest returning index to the lowest returning index. And you can see that in characteristics, like if you look at the end of last quarter, which would be the end of quarter one 2025, and you were to look at something like the CRSP index, its weighted average market cap was above 8 billion, whereas the Russell 2000 was probably closer to 3.5 billion weighted average market cap. So there's two small cap indices, but because of the way they define the small cap market, they end up with very different return outcomes.
Cameron Passmore: So how does Dimensional define the market?
Gerard O’Reilly: We start with how low do we want to go? And typically, when you look at a Dimensional portfolio or Dimensional strategy, we typically go lower than many others. So in the U.S., you might go to 10 million free float market cap. Outside the U.S., maybe 50 to 75 million free float market cap. So that's typically lower than many indices out there. And I'll get to why that's important in a few minutes.
Then we're going to look at other things like exchange listing standards. Things that we're looking for there are what are the investor rights and investor protections and the amount of disclosure and so on in each exchange. Typically, we end up with a smaller number of exchanges than maybe in some indices because of that criteria.
And then we're going to look at where stocks may be close to being delisted. So for example, stocks trading at two bucks or less here in the U.S. and that changes dynamically, but we're cognizant of we don't want to include a lot of things that may be close to getting delisted in the universe. And then you may look at other things like free float and how much is available for outsiders.
Because if let's say a company is 90 percent held by insiders, there's worry that the expected return for the insider might be different from the outsider because they're tunneling assets to their own benefit. So there's those types of considerations that would go into effect. If you put all that together and you say, what does a global portfolio look like?
And then you can look at some of our strategies. But probably on a global portfolio, so that's U.S., Canada, developed markets outside of North America, emerging markets, you're looking at about 13,500 different securities, different stocks. If you look at the MSCI ACWI IMI, so that's all country.
IMI means large to small. You're looking at about eight to eight and a half thousand stocks. There you're talking about quite a large difference in the number of stocks that would be included in our universe.
And you say, does it make a difference? It's not a massive difference in market cap, but here's why it makes a difference. When you look at implementation, when you have more securities to trade on any given day, it gives you more flexibility of what to trade.
And so it's helpful, and we'll get to this later, driving down trading costs. When you look at securities lending, in some markets, securities lending is driven by what's called general collateral. And that's a volume game.
It's how much can you get out and loan at any point in time. But some markets are driven by specials and that's not a volume game. You never know which stock is going to go on loan at special.
So emerging markets is very much driven by specials. And so there, if you have a broader universe, you probably can do better on your securities lending revenue. And then there's also a tax consideration.
What's the tax consideration? Because in strategies, you may want to harvest some losses. You may want to realize gains in a particular way.
It gives you more flexibility to do that. I think that where the trade-off is, is the cost of holding such a broad universe. And that goes beyond custodial costs, because we have very, very robust systems to take in current information about all those securities, about their financials, and news checks about all of those stocks.
We can do in-depth analysis on all of those stocks. And that really enables us from an internal management perspective to actually be able to efficiently manage a portfolio with that number of securities. So the cost is not too high for our strategies, but the benefits in our view are there and positive.
Ben Felix: So you've got this big, deep universe of securities. How does Dimensional decide which, if any, securities you're going to exclude from that universe?
Gerard O’Reilly: There's a few things that go into that decision, Ben. One would be expected return considerations. Dimensional uses historical data not to understand the past, although it does help us do that, but to predict the future.
So is there something that we see in the data over the past 20, 30, 40, 50 years that we say we would have expected to see it there because of the way we believe the world works? We saw it there. That allows us to quantify the magnitude.
And we would expect to see it going forward. So there may be small groups of stocks where we expect the return of those stocks to be much lower than the rest of the market. And you're not giving up too much diversification by excluding those stocks.
So it's an expected return consideration. Another consideration is a tax consideration. Are there stocks that are less tax efficient than other stocks?
And if there are, can we put together a strategy that excludes those stocks and one that only includes those stocks so that advisors can choose to put the one that's less tax efficient in a tax qualified account, and then the other strategy in a taxable account so that they can have a better after-tax set of returns for their clients? That's the second consideration. Another consideration on that front might be asset category, where something is kind of a different asset category.
And we'll get into REITs in a moment, but REITs are kind of a different asset category where you're passing through over 90% of your rental income straight through. It's more of a pass-through vehicle. And you're not retaining as many earnings.
It's not like a typical C corporation. And so it's kind of giving you exposure to direct investment in real estate at that rental income. So kind of like a different asset category.
So should you treat it a little bit differently? So those are some of the considerations that would go into deciding what should we include in the universe once we've defined how low will we go and what shouldn't we include.
Cameron Passmore: Can you expand on that a bit Gerard and talk about what types of securities you actually exclude at Dimensional?
Gerard O’Reilly: Yeah, we can. So let's go down the list a little bit. And let's talk about the expected return consideration.
If you look among small cap stocks, and you look at those small cap stocks that historically have had very high relative prices, so they look very growthy. They're trading at a high price relative to a company fundamental like book value or earnings. And they also have very low profitability.
Well, those stocks have underperformed the rest of small cap stocks by about 5 to 10% a year historically. How much of the broad market that those stocks represent? About 1%.
They're not a massive market cap. But they do have significant underperformance. And it matches with what you would expect.
Let's suppose that you think the world is kind of a fundamental valuation framework of explaining things. And you have something with a very high price, very low profits. If profits predict future profits, and they do, then that's low expected cash flows to shareholders.
That means low discount rate. So you would expect those stocks to have low discount rate, low expected returns, and they have had low historical returns. So we exclude those from purchase.
Another grouping is small cap stocks with high asset growth. So how do you grow your assets by a lot? You retain a lot of earnings, you issue a lot of stock, you issue a lot of debt.
That's how you grow your assets. It turns out the companies that tend to do that over the past 3 to 12 months tend to continue to do it over the next 3 to 12 months. Well, guess what?
If you're retaining a lot of earnings or issuing a lot of new stock, there's fewer cash flows for shareholders. So high asset growth, all else equal, would imply lower expected returns. And historically, they have underperformed the rest of small cap by about 5 to 10% a year.
So again, that's about 1% of the total market cap. We say, okay, let's keep those out of the portfolios or out of a broad market portfolio. So that would be a couple for expected return reasons.
And there's more, but we can dig in there a little bit later. If you look at a tax reason or an asset category reason that actually goes together, you might think of something like REITs or real estate investment trusts. When you look at REITs, and I'm going to use a US analogy, but this varies by region.
And so in some regions, we will include REITs in a strategy, in some regions, we won't, but let's use a US example. So in the US, if you think about REITs as, let's say, being 2% to 3% of aggregate market cap in the US, give or take. And the yield on REITs is between 4% and 5%, whereas the rest of the US market is like 1.5%, some number like that. And then if you look at the tax rate that is applied to REITs, because in the US, they distribute a lot of what's called non-qualified dividend income, taxed at 40%, whereas most stocks will produce qualified dividend income taxed at 20%. So now you have this increased yield of about 3%, let's say, a tax rate differential of about 20%. So if you take REITs out, and you have a separate REIT strategy, put it in your tax deferred account, you save yourself about five basis points, give or take, on after-tax returns.
So there's a tax reason that you say, they're not wrong with REITs, we think they have about market-like expected returns. But let's do something from a financial engineering perspective that gives tools to financial professionals that they can optimize the tax profile of their client. So that would be another example that fits into the asset category plus tax consideration.
Ben Felix: Can you talk about why IPOs are excluded immediately after the IPO?
Gerard O’Reilly: Yes, we also exclude IPOs immediately after the IPO. Probably it's good to use an example there. And the example I would use is Rivian.
That's from a few years ago, it IPOed at about 100 bucks a share. In the first day or two, it climbed up to about 170 bucks a share. And then afterwards, it declined and declined and declined.
And now it's trading at whatever. It didn't have the best stock price experience of all IPOs. And that's a bit of an extreme, but it illustrates the point.
When you look at IPOs, there's a few things that stick out. One is that they often have good first week returns after the IPO lists. And then if you look after that in about the 6-12 months after, they often have relatively poor returns or underperform the market in the 6-12 months after.
Now, it depends on how you measure. So whether they use calendar or event based studies, you get slightly different outcomes. But when you look at their characteristics, their characteristics look like small growth, low profitability.
So that kind of gives you an idea of why you might expect them to underperform. But then there's other things that happen. So for example, at an IPO, the insiders might be locked up for the first six months.
And so in six months time, you may have a lot of stock that's coming to market. So a significant change in the free float over that first six months. And so that's an event that will happen.
You may have some aftermarket trading support for the investment banks right after the IPO is launched. If you can't get in for that initial first week, which most people can't, and if you were trying to get in, you have the problem of selection bias in that you're working with investment banks, are you going to get allocations to all the good ones, all the bad ones, or a mix of both? So there's some selection bias there.
So our view is that when it comes to investing, you almost always have more time than you think. And I think that's a good philosophy in life in general. And so we wait about six to 12 months and there's analysis done before including an IPO in the portfolio.
But that's the reason that we would exclude IPOs from the investment strategies.
Ben Felix: You mentioned six to 12 months. How is that decision made about when it's going to get at it?
Gerard O’Reilly: That's a portfolio manager decision. Within the firm, we have a financial data working group. So that provides all sorts of data across this broad universe that I mentioned that serves it up very efficiently to portfolio managers so that they can understand the specific facts and circumstances for each corporate event or each listing.
With that, then they can make an informed decision. Okay, after six months has passed, all the insider lockups are done. Things seems to be trading more like normal.
Is this the right time to include it or should we wait a few more months? So it's a PM decision.
Cameron Passmore: So I'm curious, Gerard, what proportion of the market or the defined universe is actually excluded in a typical broad-based market Dimensional portfolio like the core fund?
Gerard O’Reilly: Let's take the US and then I'll go outside the US for an example. Probably when you say, how low do we go? So that first piece, how big is the market in terms of number of stocks, it's probably about three and a half thousand, give or take.
Then when you take into account the REITs exclusion, IPO exclusion, when you take into account some of the small, low profitability growth exclusion, the high asset growth exclusion, you're probably down to about two and a half thousand stocks, give or take. Now that's not a massive percentage of market cap. So the small high asset growth and the small low profitability, that's probably a 2% between the two of them of market cap.
The REITs in the US is two to 3%, depending on how REITs are. The IPOs will vary over time depending on the activity in the IPO market. So you're getting pretty broad market cap coverage where you're probably at something like, I want to say around 55 trillion is what you might think of the eligible market cap that will be available in the US.
That number of names is quite similar actually for developed markets outside the US and emerging markets. There's about a thousand from the starting to the end. But there you end up with far more security.
Developed market outside the US, you're probably like four and a half thousand and in emerging markets, you're probably like 6,000 somewhere around thereabouts. So you're ending up with a lot of stocks and the vast majority of market cap, but you are refining. You're making what I would say a better starting point for investors.
Ben Felix: It's a pretty low proportion of market cap that's getting excluded. How impactful do you expect those exclusions to be to expected returns?
Gerard O’Reilly: Let's say you're doing a plain market cap weighting strategy and you say that small high asset growth and small low prof growth underperform the rest of the market by 5-10% and it's 2% of the portfolio. Okay, that's 10 to 20 basis points of expected outperformance. When you look at the REIT exclusion for the US as an example, I mentioned a number of five basis points in after-tax returns.
Okay, there's five basis points there. When you look at IPOs, IPOs have typically underperformed the broad market by on the order of a percent a month in that first 12 month period. So that will vary depending on how much IPO activity there is.
So all of that in a broad market portfolio, you might be talking on the order of 20 to 30 basis points somewhere in that neighborhood. But then when you go to an asset category, like a small cap, then it can be even bigger and those decisions can be more meaningful in terms of the level of return deviation that you will get year on year, but also in the level of expected outperformance that you would get from those decisions year on year. But in a broad market, 20 to 30 is probably not a horrible estimate.
I'm using wide range to make sure people realize that it's an estimate and all estimates of expected returns are noisy and therefore wide ranges are appropriate.
Cameron PassmoreI would guess that most listeners know that a small portion of stocks are responsible for a large portion of the realized returns. So can you talk about how risky you think it is to exclude parts of the market?
Gerard O’Reilly:
The first thing that you have to keep in mind with that, Cameron, is opportunity cost. I always love the analogy from Professor Kevin Murphy at the University of Chicago, which is, is it a bad thing if your kids watch five hours of television a day or a good thing? The answer depends on what else they might be doing.
If what they might be doing is you're teaching them mathematics, reading them poetry, they're learning a musical instrument, it's a horrible thing. If the alternative is that they're out in the streets, getting into a gang and getting into all sorts of mischief, it's a great thing. So it's all about the opportunity costs.
And the first thing that I think you have to keep in mind, Cameron, is that there is no strategies, there are no strategies out there that don't exclude some portions of the market. Every strategy, whether it's an index market or traditional active or Dimensional, all of them are excluding portions of the market. Dimensional is very intentional in what portions of the market is excluding.
It's excluding portions of the market that we think will contribute to returns over time. We're very intentional in how we do it. So I mentioned in the U.S. there's a thousand names. Those thousand names represent a few percentage points of market cap. Therefore, it's not like we're excluding Apple. Let's take Apple out of the portfolio.
And therefore, the performance of Apple can make a big difference on what their returns are over any given time period. You're spreading that over many, many names. And so you put yourself in a way better position to get the average outcomes that you've seen historically, because you're doing it in a very diversified fashion.
So I think that the way that you do it matters. And being intentional about why you're doing it matters and recognizing that there is no solution out there that doesn't exclude some portion of the overall market.
Ben Felix: You spoke to this just now, but any traditional active manager would say that they're also intentional about what they exclude. Can you talk about how these exclusions that we're talking about are different from traditional active management?
Gerard O’Reilly: I'm going to use another analogy here. Hopefully, it will be helpful for part of the conversation. Something that I love is baking shows.
Now, people often ask, why do you like baking shows? You don't eat any sweets. You don't eat any desserts.
You don't eat any puddings. You don't eat gluten. You don't eat dairy.
Why do you watch baking shows? I just like them. I don't know why I like them, but I do.
Maybe it's the forbidden fruit. I don't really know. In any event, I love the Great British Baking Show.
And I don't know if you're familiar with that show. But in each episode, there's three rounds of competition. And one of the rounds is called the technical.
And in that round, every baker is given the exact same ingredients. Every baker is given the same recipe. And every baker is told to produce the same bake.
So let's say, for example, it's a ciabatta bread, then they're all told to produce a ciabatta. But the recipe is not complete. So it might not tell them how long to knead the dough for.
Or it might not tell them what order the ingredients should be added. Or it might give them an oven temperature, but not tell them how long it should be in the oven. So the recipe is incomplete.
And then what happens? They get judged based on criteria, which are what does the crumb structure look like? What does the color look like?
How good is the bake? And who wins? The best baker.
And investing is not actually all that different. Whether you're traditional active, you're Dimensional, or you're index, we're all starting with the same ingredients. And we're focused here on market portfolios.
So we're all trying to produce the same thing. We're trying to produce the same bake. But it comes down to who has the technical expertise to produce the best bake, because the recipe is very unclear.
When it comes to investing, there is no perfect rulebook that says this is exactly how you should invest. It is very unclear. And so when you think about traditional active, and you compare it to Dimensional, what are some of the differences?
When we do something, it is based on very, very strong empirical evidence, and very, very strong theoretical evidence. So that when we do it, there's high confidence, and we can show you the evidence and the reasons why you should expect this in the future. Traditional active is predicated on a proposition for which there is no evidence to support.
That's a proposition that you can make money by outguessing market prices. We have seen the evidence time and time again, that show in the aggregate, that is not a winning proposition. And that show you can't identify those traditional active managers in advance, that may have enough skill to cover their costs.
That evidence is overwhelming, it's very, very strong. And so when I think about those differences, we're doing it in more of an engineered fashion, where it's either evidence about expected returns, or it's the goal of the portfolio, when people tell us, taxes are important to us, how can you make us more tax efficient? Okay, I can go back into the lab and figure out how to make your strategies more tax efficient without taking on too much tracking error for you, and then show you ways to locate assets to make you more tax efficient.
So it's a very different mindset and a value proposition, that then ends up being a better investment outcome, in my view, for the end investor.
Ben Felix: I would say that's a pretty important difference. How does securities lending affect portfolio implementation?
Gerard O’Reilly: So we kind of touched on this earlier, so let me go back and just in case there's not a lot of familiarity with securities lending, what is it? So let's suppose Dimensional is managing a mutual fund, or a Canadian fund, or any of those funds on behalf of the end investor. We have well diversified strategies, we have low turnover strategies, we're not beholden to any index, that flexibility is important, and I'll get to it in a moment.
So somebody comes and says, they want to borrow a stock, and they may want to borrow a stock to settle a trade, they may want to borrow a stock to short it, there can be many different reasons to borrow a stock. And we say, okay, you want to borrow a stock, what are you willing to pay the investors in this portfolio in order to borrow the stock? That's a market like any other, maybe some negotiation, but we get to a price of effectively how much revenue is the portfolio going to get for loaning out this security.
Now, if you look across Dimensional's equity complex, last year was about four basis points in return was added by securities lending revenue. That varies by market. The US market in a broad market portfolio is around two.
If you're looking at developed markets, it's about five to 10. If you're looking at emerging markets, it's about 15. It varies by market and can be meaningful in certainly different markets.
And they're all broad market portfolios, by the way. So that's all just broad market portfolio lending revenue, and it can vary by market. Dimensional does a few things that I think are important.
Number one, Dimensional does not keep any of the lending revenue for Dimensional, the advisor. We have agents, the agents have to be paid. All the remainder goes to the end investor.
None to Dimensional, the advisor. So there's no conflicts there. Other things that are important though, for the way that we invest is because we have such a diversified group of securities and that we have low turnover and that we have flexibility not to have to sell a stock at any particular time.
That means that we're very attractive to many of the borrowers out there. They like to use our portfolios to borrow securities from. Because let's imagine in an index fund and you're loaning out security XYZ.
You may not want to loan out as much of your position as you otherwise would, because what happens if you have a redemption and you have to sell everything pro rata because you're trying to manage tracking error. If we have a redemption and one stock is out on loan at a very high fee, well, I don't have to sell that stock right now. And that type of characteristic is attractive to borrowers.
And so you can command higher fees, better utilization. And as I said, general collateral is more of a volume game. And right now the U.S. is more of a general collateral game. You remember from GME from a few years ago and all the shorting and what happened to hedge funds. So the U.S. SEC lending revenue has gone off over the past few years. It may come back, but right now it's around two-ish.
Outside the U.S. in developed markets, it is a general collateral game again. So it's a volume game. Emerging markets is much more specials.
you have to have the diversified securities and then things just go on loan at a very high fee. And if you don't have them, you can't get that revenue. But if you do have them, you can.
Ben Felix: You mentioned what makes Dimensional attractive for people who want to borrow securities. Do you know how your securities lending revenue compares to other product providers?
Gerard O’Reilly: You can look at Morningstar and you can look at Morningstar category averages for those types of numbers. And the category average masks that there's lots of variation, but in the U.S. you're probably looking at a half basis point in the market portfolio. So not a massive amount.
It's a small amount in the U.S. When you go outside the U.S. to develop markets, it's probably closer to five. When you go to emerging markets, it's probably closer to 10 in terms of how more relative to the Morningstar category average, a Dimensional strategy would be getting in lending revenue for the end investor relative to what that Morningstar average is. And again, as I said, you will find variation across manager.
It will vary. So you'll just have to do your due diligence there to see what the lending program looks like and how much is contributing to the returns of the portfolio.
Cameron Passmore: How do things like corporate actions, proxy voting, engagement with portfolio companies affect the implementation and also the net expected returns?
Gerard O’Reilly: Those are important too. It's probably best to illustrate with an example. Remember I said earlier on that there's about a thousand merger announcements in a given year.
Not all of those are for cash, but many of them are, especially outside the U.S. Probably more than mergers outside the U.S. are for cash than in the U.S. But let's suppose you have a merger that's for cash. And the market thinks this is going to go through for sure. So the announcement happens, the stock that's the target, let's say it's trading at 20 bucks, the announcement is for 25, it's for cash.
And the market thinks this is going to happen for sure. That price pops from 20 to 25. And then how do you think it trades between the announcement and the consummation of the deal?
Trades like cash. When you look at small caps in particular, and you see the amount of merger activity, it's about, you know, in the order of 2% of small caps are kind of in play for mergers, especially for cash globally at any point in time. An example that we had from October 23, it was called EnergySmart, I think was the name of the firm.
It was in the Russell 3000 and it closed the deal in kind of January of 24. And over that time period, the Russell 3000 was up about 16%. So a pretty good return.
And it traded for cash during that whole period. So what do we do here at Dimensional? Well, we're looking every day for those types of announcements.
Then depending on the announcement, whether it's for all cash, whether it's stock and cash, we're going to do an analysis. And we're going to be looking at the price reaction in the markets to this announcement, because that gives us a lot of information about the probability that this announcement and this deal will actually go through. And depending on the market reaction, we're going to make a decision.
And that decision might be, the market is thinking this is going through at high likelihood. We've already gotten the price pop. We don't want to have 2% of our small caps with cash drag because they're all in M&A mode.
And so let's sell this stock way before the deal consummates. Then we're going to make decisions on how to sell it, because we have a lot of different tools. Are we going to try to get this stock out in kind so that it's a stock that we want to sell, but can we sell it without realizing a gain?
Oh, that's a nice tool to have. Or if it's in a Canadian portfolio, well, what's the CGRM going to be at the end of the year and how are we going to work that in to make it more tax efficient? So then there's potentially a tax question, but that's an example of a corporate action that you can add value to the portfolio.
And this goes back to, remember what we said at the start, market exposure in that asset allocation. This is a way that you can have hidden cash inside your portfolio that you don't know about unless you're taking care of the details and making sure that you're trying to minimize that kind of cash drag in the portfolio.
Ben Felix: That's super interesting. So there's an implied cash drag in equity portfolios from cash merger activity. Exactly.
Very interesting. We had a recent guest, Marco Sammon from Harvard on to discuss his research on how index funds incur what he calls in his paper, adverse selection costs. That's coming from rebalancing, index rebalancing into new share issuance when firms issue stock, IPOs, which we've already talked about, and then selling out of buybacks when firms buy back stock to match the index.
How does Dimensional deal with things like buybacks and new share issuance?
Gerard O’Reilly: So the new share issuance is very much aligned with the asset growth exclusion, because if a company is issuing a lot of new stock, then it's probably going to fall into that category of greater than 75% asset growth over the past 12 months. Those would be examples of securities that we would just exclude from purchase because those stocks typically underperform in the months after that new issuance has occurred. And again, it's back to that less cash for investors, all else equal, right?
It's back to that concept. So that would be an example there. And I think that helps mitigate that issue.
The other item though, and I think this is important to keep in mind, is that we have a piece of research that was done recently, which looks at either free flow changes or shares outstanding changes for companies in indexes, and what volume spikes look like around the time that those happen because of index activity. And what you typically see is on the order of 10 to 20 times, if you were to look at the closing volume over the prior 20, 30 days before the corporate event, you typically see about 10 to 20 times that volume in the day of the event. So it's kind of like the index and then the index funds are doing it.
That causes a lot of abnormal volume that can push prices. And which is a push prices. Well, in about the 10 minutes into the close, it pushes prices by three to five basis points.
That's a lot of push in 10 minutes. So I think that the other aspect then that you have to keep in mind with these is not just its implication for expected returns, but it's implication for liquidity and trading costs, because these types of events cause indexes on a frequent basis to reshift, to do a little bit of rebalancing, to keep track of those changes in shares outstanding. And that can push prices up a little bit when they're buying and push prices down a little bit when they're selling.
And that can be a different form of reconstitution effect that's not as popularized and not as well known as stocks coming completely in or going completely out of the index. It tends to be a bigger effect for those stocks, but can be a bit of a drag on performance.
Cameron Passmore: How do empirical realities like momentum, short-term reversals and high securities lending fees actually affect the expected returns?
Gerard O’Reilly: Let's define some of those so we're all on the same page. Momentum is the tendency of stocks that have outperformed over the past 12 months or so to continue to outperform and stocks that have underperformed to underperform over the next few months. When you think about the securities lending fees, what our research shows is that when stocks go on loan at a very high fee, they typically underperform, especially in small caps over the next few days.
And you might say, why would that be? Well, if they're going on loan at a very high fee, somebody probably knows something. They want to short it.
They're willing to pay a lot. And so that information is going to work its way into the price one way or another. That could be an example there.
And so how these things work into the process is that you have your buy and sell. So what you want to buy and sell, but then there's a timing of when you want to buy and sell. And I think that's, again, if you go back to our baking analogy, there's no recipe that says this is the perfect time to buy and sell.
So we ask ourselves the question, well, if there's 100 or 200 stocks that we want to buy on a given day, and we know that things like momentum and securities lending information is gone very quickly, like a stock is an upper momentum for a couple of months, and then it's not, or a stock is in high securities lending for a while, and then it's not, or vice versa. Can we factor that into the timing of when we're going to buy the stock? We're going to buy them all, all the ones that we want.
But as we have flows come in and go out, can we use that information to improve the expected returns over the short horizon so that we're not getting adversely impacted by some of these things over the short horizon? And that leads to better returns over the long pull as well. So you can think about it as our buys are overweight to upward momentum.
They're not motivated by momentum, but they're overweight to upward momentum. And our sells are overweight to downward momentum. They're not motivated by downward momentum, but they're overweight to it in the way that we would trade.
And the key piece of being able to do that goes back to something that we started with, which is the investment proposition. That prices change every day, flexibility adds value. You can only do this in an efficient manner if you're looking to adjust the portfolio a little bit every day, and you're not beholden to trying to lower your tracking error relative to some index.
That's the only way that you can take these types of things into account. And incrementally, they add value to the portfolio over time.
Ben Felix: Have you guys modeled out the sort of expected economic benefits of doing the timing screening on these trades?
Gerard O’Reilly: Yeah, and it depends. It's related to the turnover for the strategy. So for strategies that have lower turnover, it's going to be less.
For strategies that have higher turnover, it's going to be more. That's just the way that it is. As you vary the turnover strategy, you know, it's in the five to 20 basis point range, depending on the turnover, when you put all of those together, the momentum screens, the short-term reversal screens, the securities lending screens, that's what it may add over time is kind of some estimates.
Again, not super precise. It depends on the strategy, but they're pretty reasonable estimates.
Cameron Passmore: I want to go back to something you just said, Gerard. You talked about how expensive to borrow stocks have lower expected returns, yet getting SEC lending revenue is a good thing. How do you balance that off in a portfolio?
Gerard O’Reilly: When something goes on loan at a high fee, our view is as follows. If somebody knows something and it's material to the company, it's going to be reflected in the price sooner or later. Should we loan somebody a stock to get it reflected in the price sooner?
If they're willing to pay us so that we get paid when the price goes down, let's do it. We're not going to buy any more of it. So it's excluded from purchase while it's on loan at a high fee, but let's get paid while that information goes into the price of the security.
And it's interesting because depending on the time period that you use, the dataset that you use, the amount of underperformance is quite similar to the amount of lending revenue that you can get for the security. And so it can be helpful for the overall portfolio. We're not going to sell it necessarily.
We're happy to keep on getting the lending revenue. Then we're not going to buy any more of it over time. One classic example there is GME, because when that went through the roof, I think the annualized rate on that loan that we were getting for GME in the portfolios was like over 10%.
It was giving a lot of securities lending revenue. And we had about $60 million worth of exposure at the beginning of that January to GME globally in our portfolios across all of our strategies. And by the end of the month, we had a billion dollars worth of exposure to GME.
We didn't buy any more, but we had a billion dollars worth of exposure because of how much it went up. And we sold it all at that point because it was a stock that was in small cap portfolios, value portfolios, micro cap portfolios, and it was a large cap stock. There's an example of, we were happy to have it in while it was giving the lending revenue.
When it started to change characteristics so quickly, we were able to react. So we turned all of those paper gains into real gains for our investors. Now when you look at the difference to indices, indices held it up and they held it down.
Markets change every day. Prices change every day. If you're not reacting every day, you're going to leave money on the table for your clients.
Cameron Passmore: But you're using information you're getting paid to have, which is so interesting.
Gerard O’Reilly: It's quite an opaque market too. If you're not in the market, you may not know the information because a lot of the short interest in real time information is just very restrictive. You can't really see the short interest in real time. So it's kind of a proxy for that.
Ben Felix: It's really interesting. That is an interesting one. Can you talk about the main ways that trading costs affect net returns?
Gerard O’Reilly: This is another good one on the kind of implementation cost side of things because trading costs, there's explicit costs like the commissions or whatever it is that you might be paying, but then there's bid offer spreads and there's price impact, which are implicit. And the interesting thing about trading costs is they don't go to the expense ratio. They come out of the NAV.
So unless you have a process to ask the manager about their trading costs, you don't know what they are. The other interesting thing about trading costs is they're built into index returns because of the index managers. So trading costs are tricky.
It's important to manage them well. So how do you manage them well? Flexibility is your friend.
How you manage them well is that you have flexibility on what you can buy or sell on any given day. And how we do that is that a portfolio manager may generate a list of order candidates trading on that day based on the portfolio on that day. I might say, here's $10 million worth of orders, but I only want to execute 5 million.
And then the trader has flexibility on that day to do what we call participate, don't initiate. There's a lot of market volume that happens. Last year, if you look over 800 billion U.S. dollars was traded every single day in markets around the world. There's a lot of volume that happens. So if you participate, don't initiate, you can be part of that flow, never move prices. And that can be very, very helpful to the end return for investors because you're leaving less money on the table as you try to implement your strategy.
Cameron Passmore: And how significant are the trading costs?
Gerard O’Reilly: It can vary. So we do a whole ton of experiments on trading costs, and we have a lot of different charts that illustrate it in different ways. But one of our charts, we call it the lollipop chart because it's all these little straight line with a little circle on top.
It shows U.S. large caps, U.S. small caps, non-U.S. developed markets, large and small, emerging markets, large and small. And what you see in that chart is we try to estimate our relative price advantage to others in the marketplace trading. So what do we do
We run a set of experiments where we look at what we traded on a given day. So the actual name that we traded, the amount that we traded. And then we say, if we were to trade those and use other people's trade prices, because we sign all the trades, we use an algorithm to sign the trades, whether it was buyer initiated or seller initiated, what would it have cost us relative to what we executed?
So we're looking at how others trade relative to how we trade. And we're saying, what's the price differential? And in U.S. large caps is like 5 to 10 basis points and small caps is 10 to 20. When you go outside the U.S., then you're in that 20 neighborhood. And when you get to emerging markets, it might be as high as 30, that differential. And that means buying lower, selling higher than others.
And then how that impacts your overall returns depends on the turnover and how much trading you're doing. If you're doing very little trading, then it's not as impactful. If you're doing a lot of trading, it's more impactful.
And so I think that kind of gives you a way of understanding the trading cost relative to what else is going on in the market at that time.
Ben Felix: Those basis point figures were Dimensionals per trade price advantage over competitors that you're measuring. Yeah. That makes sense.
You mentioned this being less of an issue for lower turnover funds. We also talked a little bit about the price pressure on index fund reconstitution or index changes. How much exposure would index funds have to trading costs?
Gerard O’Reilly: The index reconstitution effect is very well documented. It's when a stock is added to an index, then the market knows that if that index has a lot of assets attached to it, then the managers who are trying to minimize tracking your reverses, the index are going to have to trade a lot of that stock, buy a lot of that stock on a very specific day. And if a stock is deleted from an index and there's a lot of assets attached to it, then the reverse is going to be true.
They're going to have to sell a lot of that stock that particular day. The market is happy to provide anyone a service as long as the people in the market providing the service are getting paid. How they end up getting paid is what you typically see is you see a price ramp up in excess of the index return for stocks that are going into the index and a price decline below the index return for stocks that are coming out of the index.
And that typically tends to reverse after what's called reconstitution day when all the trade in the index. And so then if you look at that, we have a number of papers on this that use very recent data the past 10 years to try and measure. And for those pure additions and pure deletions, and that's a subset of index turnover, it's not all index turnover, a subset of index turnover, then you find that there's about a 4% price run up in excess of the index return for the stocks coming in and about a 4% underperformance for the stocks going out that reverses.
So what does that mean? That the index itself has the trading costs of the index managers built into its return. You can nail the index.
That does not mean that you've had zero trading costs, because your trading costs have been built into the index return. You'll never ever see it if you don't understand where to look. One of the best examples I think was Tesla when it went into the S&P 500.
And that was, was that December 2020 or was it 22? In any event, that was one of the biggest additions to the S&P 500. And we have a strategy here in the US that's a large company portfolio that looks a lot like the S&P 500, but we don't do exactly what it does.
And we delay sales or purchases, or we go early when there's a stock coming in. And we did that for Tesla on that particular month in December. And that strategy outperformed the index by eight basis points in that month.
Most of it because we chose to include Tesla at a different time than the index did. So that gives you a sense of how that impacted the index returns itself in that given month. Eight basis points, by the way, is like four times the expense ratio for most S&P 500 index funds was spent in a month.
But even then, if you look subsequent to that, Ben, Tesla has changed the number of stocks that have been included in the index multiple times. And each one of those, you see little price pressures around when that occurs. So those things are built into the index returns.
And so then it's what people often refer to as the hidden cost of indexing. It's not in the expense ratio, but it definitely impacts the returns.
Cameron Passmore: Just looked it up. December 21st, 2020 was the date Tesla joined.
Gerard O’Reilly: 2020. There you go.
Cameron Passmore: So you mentioned expense ratio, and I think I know your answer to this question, but I have to ask it anyways. Fees are important, of course, and paying attention to fees are important. But do you think there's currently way too much emphasis on fees relative to these other components of net returns?
Gerard O’Reilly: I would say yes. And I agree, fees are important. And if you look at Dimensional over time, whether it's their Canadian fund lineup, whether it's our US fund lineup or OIC lineup or user lineup or trust lineup, you've seen that we've reduced our expense ratios over time, both the management fees and the other expenses.
And we've commitment to doing that over time. And I'm sure in the future we'll continue to do so. We look at that ongoing and make sure that we're competitive with what's out there in the marketplace.
But I think that there's a whole concept of value for service. Value for service is something that you have to have expertise to make a good assessment around. Are you getting value for the fees that you're paying?
Are you getting excess returns for the fees that you're paying if they're in excess of the index? You look over the past 20 years, over 80% of the Dimensional funds here in the US that were around 20 years ago, they're all still around today, by the way, have outperformed their prospectus benchmarks. That's a hell of a number, much larger than the industry, 100% survival where the industry is about 50% survival and over 80% outperformance where the industry is like 14, 15, 20.
And it's about doing these things to control what you can control. And so I think that for folks, all we ask is give us a chance to explain why we think all of these things are important. And hopefully we can convince you that we're giving you value for service, because just like you don't want to put your investors or your clients in $100 skis, not necessarily do you want to put them in the cheapest investment product that you can find, because that can be accretive to value over time.
And I'll give you a couple of examples there just to illustrate one long pull example and one kind of recent. We've had a strategy in the US for over 40 years. It's a small cap strategy and it's outperformed as prospectus benchmark Russell 2000 by about one and a half percent annualized compound return difference over a 40 plus year period.
Well, what does that mean in terms of returns? We say the market about 10% return means your money doubles every seven years. Not bad.
If you can take that 10 and turn it into 11 or 12, your money doubles every six. So over a 42 year period, you got double the money from what you started with. And that one and a half percent is net of all fees and expenses where there's no fees and expenses extracted for the index.
That's an example. And when we break that down, that's a small cap strategy. So some of the things we talk about are probably bigger, but around 50 to a hundred basis points came from the universe definition.
All the things that we were talking about that are very important for small caps. Some came from momentum screens. Some came from high asset growth exclusions.
Some came from the small low profitability growth exclusions. They all came from different sources that added up to about that one and a half percent. And different things were important different times because the strategy hasn't been the same the whole time period.
It's always been a small cap strategy, but we've gotten better. As we learn more, we improve our strategies. So our commitment is to get better.
At the same time, the expense ratios are coming down. So you're getting a better product for less, which I think is pretty good. The other example I'd use, Cameron, is over the past year, if you look at 2024, we've won in 2024 about 18 billion US dollars in new mandates from institutions that were leaving indexing behind and going to our approach.
So they weren't asking for a lot of overweight to size value profitability. They were asking for, build us a better market portfolio. And they used some of the existing mutual funds that we had that we had around for a long time as examples.
And the components of net return conversation was what was starting to change their mind to get them to dig deeper. Because there's a big trend in the institutional marketplace for a long time, where it's get your public market exposure at as low expense ratio as you can and spend all your time in the private market stuff. And I think that hopefully we've gotten a bit better at explaining the benefits of our approach and the value for service at Dimensional.
But I also think that the market is getting to see that due to their own experiences of what they've experienced, these large institutional investors in implementing these indices in-house of what some of the frictions are and how they leave some money on the table. And they're willing to go and increase the management fee they're paying because they feel that they're getting value for service.
Ben Felix: I find DFUS to be a fascinating product because it's kind of what you just said, take the market and do it a little bit better, addresses a lot of things we've talked about. Can you say how much of a difference in net expected returns you would expect between it and a total market index fund? And DFUS has a nine basis point expense ratio, I think, and say an index fund with a three basis point expense ratio.
Gerard O’Reilly: About five or six basis point difference in expense ratio. If you look at that fund, and actually that fund, Ben, is part of an overall series. We call it the market series here in the US.
So we have two US funds in that market series and one non-US developed and one emerging markets. And they all have very, very, I would say market-like characteristics. So they're not doing a lot of small cap and value and profit, but that's not what they're doing.
They're doing more of the things that we talked about, some exclusion, some implementation and so on and so forth. And they've all been live since January of 2021 because we started our ETF business, standalone ETF business here in the US in November of 2020. And we launched two in November and one in December.
So they've all been live since then. The one that you mentioned, DFUS, is probably on the order of 50 basis points outperformance since then. And the ones outside the US are even higher.
They're, I don't know, 50 to 100, somewhere in that neighborhood. I would say that's a little higher than I would have expected, to be honest with you. I think that we've just had a good run.
Things like small, low profitability growth and high asset growth have had really poor performance over that time period. So not holding them has been very, very helpful. But I wouldn't be too put off by 20 to 30 basis points in terms of performance, difference net of differences in fees and expenses, and especially after taxes here in the US in particular.
Like I give you an example, that REIT exclusion makes a difference to the after-tax return. But then things that we do around the timing of when we buy and sell securities, when they're paying a dividend, to make sure that the percentage of qualified dividend income is higher for our strategies relative to some of our competitor index funds, I think is also an advantage. So after-tax returns, we need 30 basis points, some number like that.
And for a five basis point or six basis point expense ratio differential, we think that's a pretty good deal. Back to our baking analogy, we're all making ciabattas here in the sense that we're making portfolios that look like the market, smell like the market, taste like the market. They don't have a lot of differences.
When the market's up, you kind of know what your portfolio is doing. But how do you squeeze that little bit out? Even though it's a broad market portfolio, and all of the things that I mentioned that are very systematic, they're all rules-based, we improve over time, they're all very, very rational.
It's not like we're doing outlandish things. We're just using common sense and how to approach the market. They've all added value over time.
So I think that expected outperformance, I'm always looking to put a number on for a strategy because returns are very, very noisy. So we've had a good run. We may have disappointing outcomes in the future, but I think over the long pole, 20 to 30, some number in around there, which could make a difference, makes a difference to your long pole wealth.
Ben Felix: I just ran the numbers for DFUS since June 14th, I think is when it listed as an ETF and then compared that to a total market index fund. It's actually closer to 100 basis points annualized since the conversion.
Gerard O’Reilly: There you go. And you know, it's interesting, the conversion itself, this is something that I think, you know, it's a bit of a challenge for financial professionals, but it's not an insurmountable one. That was a mutual fund before converted to an ETF.
And it's been around since maybe 2001 or so. So it has a long track record, but it's done slightly different things over that time period, which is why a conversation with the manager is important to understand its relative performance, because it was a tax managed mutual fund back when it launched. And so we did things around delaying the realization of long-term capital gains, trying not to realize any short-term capital gains, some things around how we manage the dividends, our exposure to companies paying dividends.
We had different types of security caps. We didn't have the small or profitable exclusions at the beginning. We didn't have the high asset growth exclusions.
And so when you look at it now versus its long pool history, it's a little bit different than it was then. And that's why a conversation with a manager, when you're trying to understand any fund, ours or anybody else's, I think is appropriate for a financial professional. I like to tell a story.
So I have a 10-year-old daughter and she's very active. She's into all sorts of things. One of the things that she likes to do is Brazilian Jiu-Jitsu.
And I don't get to go to all of our events, but I do get to go to the Jiu-Jitsu events because they're on the weekends and her training's on the weekend. So we go on a Saturday morning. And I don't know if you know, in Brazilian Jiu-Jitsu, their black belts are called professors.
Every morning before they start class, there's a little life lesson. And some of them are actually really good. I listen, I take notes.
And so one of these life lessons was the professor said, okay, what does practice make? And one of the kids put up their hand and said, perfect. And he goes, no, practice makes progress.
You can never be perfect. It was a really nice life lesson that we embrace here at Dimensional, which is we get better all the time. And so that is reflected in the strategies all the time.
It's where strategies aren't identical through time. The main themes of strategies are very, very similar, but how we implement gets better because practice makes progress. And that's what we're trying to do day in day out is get better at how we implement.
Cameron Passmore: Amazing. So you've been sharing information over the past hour. That's very interesting.
And I'm curious how you pull this all together when you build portfolios that have the tilts towards the higher expected returns.
Gerard O’Reilly: We've talked about portfolios in the market series. So now let's say we want to have a tracking area budget. What comes into play?
All the things that we just talked about still work, but now we're going to do a lot more weight deviation relative to the weight in the market. So in those market series, so what you mentioned, DFUS, you look at the individual security weights, they're very close to their market cap weight. But in Canada, we have a lot of cores.
We have a lot of vectors. So the cores now will take much bigger differences in terms of their weight relative to the weight in the market to pursue another area of value add, which is overweight, the highest expected returning securities in the market. And I think that's another tool or lever that you have in your toolkit.
And you just have to decide how much tracking error can you take as an investor, because the more tracking error you take, the worse at certain points in time you will feel. In the long pull, we think you're going to be happier, but you have to be able to live with the unhappiness to get the long pull happiness, because the more tracking error, the bigger the chance for underperformance, and that underperformance will show up at certain periods of time. That's just how it works.
And so when you look at the core to the vector, it's kind of a trade off of how different you want to be for your expected outperformance. So in core, we often quote like one to 2% expected outperformance with two to 4% tracking error. That's what we'll quote.
And vector, we might go two to three with 6% tracking error, something like that. So we can go a little higher. The example that I use for vector, because vector now is kind of a market portfolio, but not a broad market portfolio, and that is excluding more stocks.
So the stocks that will be included in a core are pretty much the same group as would be included in the market series that we were talking about. But vector, now you're going to exclude about 40% of the market cap. The way that I often describe it is if you look at the US market as the end of last quarter, the profitability, so that's aggregate profits divided by aggregate book value, about 60%, 0.6 profitability. You look at valuations and they change, but as the end of last quarter, first quarter 2025, you're looking at price to book ratios is like four and a half. PE ratio is 20 to 25. So people will say, yeah, very high valuations for the US market.
But the US market has about twice the weighted average profitability as developed markets outside the world. Because developed markets outside the world are about 0.3 weighted average profitability and about half the valuation. So it's not clear that the US was so much higher value because of the weighted average profitability.
So you say, what does value do? Value would typically take that profitability and cut it in half. So a value strategy in the US at the end of last quarter would have taken the profitability and turned it down to about 0.3. I would have taken the price to book ratio or price earnings ratio and also cut that roughly in half. So the vector, what it tries to do is kind of have a better blend of value and profitability. So you look at the profitability of vector, it's about 0.6, about the same as the US market. But then you look at the valuation ratio, it's about 30 to 40% of the US market.
Not quite like value, but getting there. So what's it doing? It's focusing on those stocks that are either high profitability, deep value, or both high profitability and value.
And so between the core and the vector, you can put those together to get a very tailored exposure to these size value profitability premiums, which also then contribute to expected outperformance over time.
Ben Felix: Super interesting. That idea of making a value portfolio and then you can see it if you look at any product and it's profitability just drops, but then seeing how the vector funds, you get that value, but your profitability doesn't drop below market. It's cool to see the implementation in action when you look at product characteristics.
You guys are doing tons of cool stuff. The products we just talked about are cool. All the stuff that we've talked about on implementation is incredible.
You guys got into the ETF business relatively recently and become a big player there. What are you most excited about at Dimensional right now?
Gerard O’Reilly: There's probably a few big things getting me going. One is that it really does feel like investors are looking beyond either the wrapper or the expense ratio. And we're back to what is the investment proposition?
What's the investment strategy first? And let's talk about those other things second. And I think that there's more of a focus on that, that I think is good to see because if you don't have that focus, in my view, you can leave money on the table.
And by the way, that money comes out of your client's pockets. That's not a good thing for the end investor. So I think that's good.
And we see that with institutions. We see that with financial advisors. That's a good direction of travel.
And I think that Dimensional is trying to contribute to that conversation in the marketplace and in the community and trying to push in that direction of travel. That's been exciting. We've done a lot of work on that front to improve how we communicate and to improve the set of strategies that we have for investors.
The other one that's a broad theme that I find pretty exciting, and this is something that Dave and I, so Dave Butler, he's the co-CEO along with myself. We've been together now as co-CEOs for almost eight years. So time goes by real fast.
One of the things that we said when we first got into the seat together was we want to make Dimensional convenient to work with. And in part, what that means is that we have to have all the vehicles that clients like to consume an investment proposition through. So one investment proposition, then choose your own venture and how you want to consume it.
And we've done a lot of work on that, whether it's launching ETFs here in the US, we're looking at use of ETFs, ETF access points for Australian trusts, or whether it's what we're working on right now, ETF share classes, which could be huge for the industry here in the US. And Dimensional is leading the charge on that to try and get that exemptive relief over the line with the SEC. SMAs, we've done a lot of work on separately managed accounts, tax managed separately managed accounts, a lot of FinTech solutions.
I would love to be able to take that to other places in the world outside the US. And I think that can be helpful, but it depends on when the infrastructure in each one of those markets is ready. And so that's not within our control, but something that we can probably push along.
That is kind of exciting is figuring out all the ways and the vehicles that we can put in place so that the financial professionals that we work with can choose what's best for them. Because what we do is we listen to what's important to them, and then try to build vehicles that meet those needs with that underlying solid core investment proposition. So those are two big themes that I think are important that drive a lot of what we do, and hopefully will improve over time and bring more things to the market that people find useful over time.
Ben Felix: Very cool. All right, Gerard, this has been a fantastic conversation. We really appreciate you coming back on the podcast.
Gerard O’Reilly: It's been a pleasure. Thanks again, and I love that you're always as detailed as you are, but I'm detailed too, so it suits.
Cameron Passmore: It's good to see. Meeting of the minds here, Gerard. Great to have you on and great to see you as always. It's a great conversation, lots of value added.
Is there an error in the transcript? Let us know! Email us at info@rationalreminder.ca.
Be sure to add the episode number for reference
Participate in our Community Discussion about this Episode:
Links From Today’s Episode:
Rational Reminder on iTunes — https://itunes.apple.com/ca/podcast/the-rational-reminder-podcast/id1426530582.
Rational Reminder on Instagram — https://www.instagram.com/rationalreminder/
Rational Reminder on X — https://x.com/RationalRemind
Rational Reminder on TikTok — www.tiktok.com/@rationalreminder
Rational Reminder on YouTube — https://www.youtube.com/channel/
Benjamin Felix — https://pwlcapital.com/our-team/
Benjamin on X — https://x.com/benjaminwfelix
Benjamin on LinkedIn — https://www.linkedin.com/in/benjaminwfelix/
Cameron Passmore — https://pwlcapital.com/our-team/
Cameron on X — https://x.com/CameronPassmore
Cameron on LinkedIn — https://www.linkedin.com/in/cameronpassmore/
Gerard O’Reilly — https://www.dimensional.com/us-en/bios/gerard-k-oreilly
Dimensional Fund Advisors: https://www.dimensional.com/
Episode 322: Marco Sammon — https://rationalreminder.ca/podcast/322Marco Sammon
Episode 198: Gerard O’Reilly — https://rationalreminder.ca/podcast/198