Andy Clarke and Nelson Wicas: Index Funds Aren’t Really Passive, and Other Lessons for Building a Better Portfolio
The authors of ‘The Architecture of Wealth’ share insights from working with Jack Bogle and decades of research on diversification, indexing, active management, and behavioral bias.
Subscribe to Morningstar’s The Long View from your mobile device: Apple Podcasts | Spotify
Our guests on the podcast today are Andy Clarke and Nelson Wicas, co-authors—along with Ganesh Suntharam—of the new book, The Architecture of Wealth: The Art and Science of Portfolio Construction. Andy has spent more than three decades as an investment researcher and writer. He has worked at Morningstar and Vanguard, where he served as assistant to Vanguard Founder and indexing pioneer, John C. Bogle. Andy has published research in practitioner journals such as the Journal of Index Investing. He is the author of Wealth of Experience and a CFA Charterholder. Nelson has developed and managed quantitative equity strategies for more than 30 years. Most recently, he worked at Redpoint Investment Management, where he was a founder and research analyst. After completing his PhD in economics at the University of Pennsylvania, Nelson joined Vanguard to help establish its quantitative equity management function. Nelson later helped found Vanguard’s investment counseling and research team to produce portfolio construction insights for clients and internal teams. Nelson has been a visiting professor of economics at Haverford College in Haverford, Pennsylvania.
Episode Highlights
- The Architecture of Wealth
- Why Portfolio Construction is an Art and Science
- Bogle’s Approach to Active Management
- Why Index Investing Isn’t Passive
- The Birth of Modern Portfolio Theory
- Behavioral Finance, Investor Biases, and Impact of Education
- Why Index Funds Finally Took Off
- Evaluating Active Managers
- The Small-Cap Effect
If you have a comment or a guest idea, please email us at TheLongView@Morningstar.com.
Transcript
Amy Arnott: Hi, and welcome to The Long View. I’m Amy Arnott, portfolio strategist for Morningstar.
Christine Benz: And I’m Christine Benz, director of personal finance and retirement planning for Morningstar.
Arnott: Our guests on the podcast today are Andy Clarke and Nelson Wicas, co-authors—along with Ganesh Suntharam—of the new book, The Architecture of Wealth: The Art and Science of Portfolio Construction. Andy has spent more than three decades as an investment researcher and writer. He has worked at Morningstar and Vanguard, where he served as assistant to Vanguard Founder and indexing pioneer, John C. Bogle. Andy has published research in practitioner journals such as the Journal of Index Investing. He is the author of Wealth of Experience and a CFA Charterholder.
Nelson has developed and managed quantitative equity strategies for more than 30 years. Most recently, he worked at Redpoint Investment Management, where he was a founder and research analyst. After completing his PhD in economics at the University of Pennsylvania, Nelson joined Vanguard to help establish its quantitative equity management function. Nelson later helped found Vanguard’s investment counseling and research team to produce portfolio construction insights for clients and internal teams. Nelson has been a visiting professor of economics at Haverford College in Haverford, Pennsylvania.
Andy and Nelson, welcome to The Long View.
Andy Clarke: Thank you. Great to be here.
Arnott: Well, as I mentioned in the intro, you have co-authored a new book, The Architecture of Wealth, along with your third co-author, Ganesh Suntharam. I’m curious if you could tell us a little bit about the idea for the book and who the target audience is.
Nelson Wicas: The book was motivated by a course that I taught at Haverford College, and it has antecedents in work. When I first was hired at Vanguard in ’92, I was asked to create an investment course for the business executives at Vanguard, which I taught for several years. Later, I traveled around the world presenting Vanguard’s portfolio construction ideas and strategy ideas. I had that content, and I was leveraging it for the course. The case study for the course comes from work I’d done with Ganesh Suntharam, because he and I worked here over 12 years. All the case studies are actually based on strategies we built and portfolios we ran or projects we did for particular clients.
My connection to Andy is that Andy and I worked together closely from 2002 to 2007 at Vanguard, creating Vanguard’s first iteration of their institutional thought leadership program. Andy helped write the white paper series. He taught me a lot about writing and the process. I know from working with Andy that he’s super talented at creating very accessible content, no matter how complex the investment ideas are. When it came to writing a book, I was interested in providing the intellectual rigor you’d have in a sort of serious academic textbook, but I wanted it to be very accessible. That’s why I reached out to Andy to see whether he’d have time.
Initially, he was busy on other projects, but he said, “Come back to me in a year.” After going through teaching the course a second time, I reached out to him, and I think in a moment of weakness, he said, “Oh, I do have time.” So, we all created time in our schedules and ran at it. And I have to say Andy kept us moving along at a good pace. Everyone who’s talked to me is like, “You got this done in six months?” And they’re like, “That’s super fast.” I’m like, “Andy kept us moving.” It was fun to do, and my writing skills have improved tremendously from working with Andy.
In terms of what the target audience is, the target audience is undergraduates, MBAs, financial planners, serious investors; in terms of where the book is targeted, there are a lot of great books out there from Bogle and Malkiel and Swenson. There are also all these really great academic books such as McKinnon, Tuttle, and Bodie, Kane, Marcus, and Campbell, Lo, and MacKinlay. And so what we’re trying to do is take the intellectual rigor from those Campbell, Lo, MacKinlay-type books and Bodie, Kane, and Marcus and make it as readable and accessible, but more comprehensive than, say, the Bogle books are, or Malkiel’s book, or Swenson’s book. That was the goal, and that was the target audience.
Benz: Andy, let’s bring you in here. I’d like to hear about how you work to make the book as understandable as possible, even for undergraduates who aren’t necessarily steeped in investing. What was your process for making it so readable?
Clarke: I think some of it was just kind of well-known techniques to enhance readability. Things like short paragraphs in general, a lot of use of analogies and metaphors to explain concepts. Wherever possible, I tried to use vivid quotes from some of the academic literature. A funny thing is that if you look at some of this literature from the ’50s and ’60s, it’s actually very accessible. And sometimes there’s a little sense of humor in that, too. I tried to draw on that.
I guess, more generally, what interested me about the book was this potential to combine, as Nelson was saying, the empirical and theoretical rigor of a classroom text with this history of ideas. Something I was thinking about as we wrote about it was when you first learn about astronomy in high school or middle school, you don’t start with your protractor in a trigonometry textbook. You learn about the debate: Is the Earth or the Sun at the center of the universe? That makes it engaging and kind of opens the door to what might be a more technical exploration of these portfolio construction ideas. That’s what appealed to me about it. I tried to keep that in mind as we work through the different developments in modern portfolio construction.
Arnott: The subtitle of the book is The Art and Science of Portfolio Construction. I’m wondering if you can talk a little bit about why you think portfolio construction is both an art and a science.
Wicas: Well, the science obviously is coming from the mathematical frameworks of modern portfolio theory. But as we talk about in the book, the actual implementation of modern portfolio theory initially was a big bust because you need good inputs for expected return, the variance estimate, and the covariance estimates. As you know, modern portfolio theory is driven off of estimating
But the original technologies that people were using, so typical technologies, basically produced betas that were not predictive. You ended up, when you put them into mean-variance optimizers, if there were no constraints, you’d end up having everything going into a handful of stocks, which is completely contrary to the original ideas. That’s where the whole “art” started to develop, which is, in its simplest form, creating constraints—linear constraints to the problem of making sure that the industry weights match the benchmark, plus or minus some error term.
Making sure the weights in the portfolio match the weights in the benchmark for all kinds of risk factors, making sure that no individual stock is overweighted too much or underweighted too much relative to its market-cap weight in the benchmark. That sort of starts the art of it. But then again, there’s been a multidecade process of trying to make risk models work the way they originally intended. There’s all kinds of science that has gone into that and all kinds of art that’s gone into that, which is really beyond this book. That’s a whole ‘nother course in itself.
To give you an example, though, of a very simple idea, when you form a quant portfolio, you’ve got all kinds of input. All your stocks have multiple measures of different information sources, and it’s typically put together in an average or one number, and that’s fed into your optimizer, and then you get a portfolio built from it. However, what would happen with those kinds of portfolios is, a lot of times if you just did that, the portfolio ends up being, over time, more and more tilted to just a few of your inputs. The reason for that is that the data, the underlying data, turns over more slowly for certain signals or information sources than others. These slow-turnover signals end up sort of sticking in the portfolio and hanging out there. You may have this diversified set of quant inputs, but you end up having a portfolio that’s very much tilted to just a handful.
What people do is you build your portfolio with your optimizer, you look at your buy list, and then you look: Do I have a broadly diversified set of signal inputs going into all the stocks that I’m buying in all the industries? You can try to refresh the sample in a manual way sometimes. That’s to get to the art of making sure the sample that’s going to the portfolio and the sample that’s consistent at staying in the portfolio is representative of your overall investment process.
Benz: Nelson, the book mentions that Jack Bogle hired you to develop active quantitative strategies. Is there a misconception that Bogle only advocated passive management?
Wicas: I’m not sure it’s a misconception, but certainly once he retired, he became a leading advocate for indexing strategies. However, in the book, we do reference academic studies that show that Vanguard has produced a very impressive active track record of outperformance. That track record actually goes back decades. When we were there in the ’90s and the 2000s, that time period, that successful track record was in existence.
When I started at Vanguard, the viewpoint that he had—I mean, I was interviewed by him and this was conveyed in the interview, but also was conveyed in many meetings that we had with him and the people that ran the firm with him—is that while they understood the idea of market deficiency and the cost argument that supported indexing, they actually viewed that, to run a successful mutual fund firm, your clients wanted active strategies, so you had to offer them.
They wanted to build active strategies that, in the worst case, would basically give you an index and return with minor tracking error. That’s the idea: Indexing provides the floor for acceptable performance. And then they thought that if you built risk-controlled portfolios and you were trying to outperform slightly and you could have low costs, then they knew that could be a top-quartile-performing active fund. That was the thinking that went into how they built active strategies starting in the ’80s.
Again, what is not widely known is that Vanguard under Bogle was an active client of almost all the initial quant managers out there. Bogle also encouraged his son to go into active quantitative management. When I was hired, the interview process was that they actually explained that was their strategy; that they were looking to create an internal active quant group to manage the core holdings in their multimanaged active funds to lock those strategies to the benchmark and to slightly outperform. And then they wanted to basically create a core satellite structure where they had traditional active strategies surrounding that.
Again, the people at Vanguard who built the multimanaged strategies, starting in the ’80s and throughout the ’90s, built those strategies by building in risk control to the process. Typically, simple ideas for risk control are: Make sure you hold a lot of stocks, make sure you’re having stocks in multiple industries, make sure that you are style-consistent with your strategy of staying in your
Again, if there are 30 industries and you have three or four managers, you’re going to have 90 to 100 or 90 to 120 stocks that are all different from one another and representing different bets. Those ideas, which are talked about in the book, Vanguard wasn’t necessarily a pioneer, but it was a very early adopter of those ideas that were coming out of the consulting world.
Clarke: I guess I’d just add one more thing that made a big impression on me. I worked for Jack Bogle early in my career, and he was always very respectful and supportive of active managers. And every once in a while he’d get a letter from a manager, usually at a small-fund firm, low profile. The manager would say, “Look, I’ve been reading all your stuff. And you say I can’t beat the market, but look, I’ve done it.” Or, “You advocate low-cost investing, but look, my fund has higher costs than the Vanguard competitor, and I’ve outperformed.” And he would ask me to get data from Morningstar, and he’d write a very thoughtful response to the letter writer and make comments like, “Yes, I see that you’ve outperformed our Vanguard fund, but I note that your portfolio turnover is a lot higher than ours. You seem to have less conviction in your ideas than we do.”
But he’d end these letters with something like, “I wish you the best. I truly do.” That made an impression on me because I just imagined being the portfolio manager receiving that letter back from Mr. Bogle. When I sent it, I probably expected it to wind up in the circular file. But instead I got this very thoughtful response. He takes a look at my performance and then wishes me luck. I think he did have a lot of respect and was very supportive of people who were managing portfolios and doing the best they could to deliver the best results to their clients.
Benz: Andy, related to that, I remember Jack Bogle calling me once or twice while he was on the hunt for some data to underpin something he was working on, and it was always a very exciting moment to pick up my phone and have him on the other line. He was always driven by data and research.
I wanted to follow up on the point about passive investing because you write that the term passive investing is unfortunate because it connotes “listless ignorance,” in your words. Is there a better way to think about or describe this approach?
Clarke: I think I prefer just index investing, but I understand why people use passive because it’s useful for the contrast with active management, where people are making active decisions to buy or sell certain securities. I worry that passive does connote this kind of do-nothing strategy and doesn’t do justice to all the theoretical and technological innovation required to build these portfolios. I think index investing at least opens the door to an exploration of that innovation, so I prefer that term.
Someone said passive could lead you to believe that, well, what do you need to do to manage a passive fund? And the answer might be nothing. What do you need to do to manage an index fund? Well, actually, a lot goes into it. There’s this historical technological foundation, but even the day-to-day is complex. Yeah, my preference would be index investing.
Wicas: Along the lines with what Andy’s talking about, back in the early days in index fund management, the tracking era, the performance of the fund relative to the benchmark actually could be quite wide, and it actually varied significantly from manager to manager. In the ’80s and ’90s, there was this big drive, at least at Vanguard, to tighten tracking error up. Where I’m going with this is that to do that, it was anything but passive. It was super active; attention to all the details that were involved in it.
At one point, Gus was training me how to run the index funds because I was subbing for somebody—Gus Sauter—and one of the first things he was showing me was the cash flow forecasting model. This was in the day when money would come into the complex, but he wouldn’t necessarily be told within the day that $5 million was coming to a particular fund, or whatever the amount was.
He had to figure it out because that client was going to get the closing price at the end of the day. Gus would have to try to make sure that anticipated $5 million was already invested so there was no tracking error. He actually had built a regression-based forecasting model based on prior cash flows. He actually used the returns of the fund because he knew that cash flowed into the complex based on, at that time, when returns were running up. He had this very complicated forecasting model, and that was just one component of it. He was an early adopter of transaction cost modeling. I think they were perhaps one of the first clients of this consulting firm called Plexus.
I went to meetings every month with his team as part of the team, and they went into really minute detail on all of their transactions in the previous month, looking at where they had fallen short of their implementation goals. They would literally sit there and discuss: What were we doing? What were the signals? What was coming from the brokers? Why did we do that? They were constantly improving it. During the period of time that I worked for Gus, he ended up closing transactions; he was tracking inside his expense ratio, and there were times that the index funds were actually outperforming the benchmark. This is all due to this incredibly careful attention to detail.
Another thing that goes on that I’m sure you and your listeners are aware of is that when the index, the benchmark, is being reconstituted, there’s all kinds of active trading going on based upon trying to minimize the market impact that can happen when a benchmark gets reconstituted. There was a man in the group, Mike Buek, and I used to go to meeting after meeting, and I’m coming from an academic perspective, and I would listen to Mike talk about what the strategies were going to be this year.
As an academic-type person, I could not see how he was doing it. I literally started thinking he was like the Radar O’Reilly of index benchmark reconstitution. He was listening to everybody in the marketplace, and he’d come in with this incredible set of information that would then drive what we did. It was a very interesting and active process to see in practice.
Now, in the last 20 years, those ideas have gotten to be more broadly understood. And so, tracking across index funds has tightened. It was fascinating to me that—literally, it went on for 20 years—Vanguard’s index funds were outperforming everybody in the industry. It really had to do with Gus Sauter and his team and their careful attention to detail and the active trading they were doing.
Arnott: Yeah. Even the process of constructing an index is not straightforward. You write about some of the improvements that Gus advocated for in terms of how indexes actually get put together.
Clarke: Yeah. He first introduced ideas about indexes that would make them better measures of how real-world portfolio managers actively invested. It was things like adjusting indexes for the float in a particular security. A company might have a certain market weight, but if only a small number of shares traded, it would be very hard for the managers to hold those stocks at an index weight. He introduced ideas about the transition from, say, a growth index to a value index, a small index to a large index. I guess it was in the early 2000s; he worked with Center for Research and Security Prices at the University of Chicago—I know Morningstar recently acquired CRSP—to introduce new indexes that reflected this kind of real-world portfolio management behavior.
Arnott: Yeah. In terms of investment theory, the book really focuses on the past 75 years or so of academic research on investing. I’m wondering if you could talk about Harry Markowitz and why he was such a pivotal figure in the birth of modern portfolio theory.
Clarke: I think because he took these kinds of common-sense ideas, “Don’t put all your eggs in one basket,” and translated them into a statistical framework that was much more useful. He created this framework that allowed you to test how the combination of different assets would affect a portfolio’s volatility and its return.
You turned this rule of thumb into a tool that could be used with a lot more precision. I’d say the second thing that his framework introduced was this idea of risk as uncertainty. Risk wasn’t just a bad outcome. Risk was a range of possible outcomes. That was kind of a break, I think, from the way people looked at risk in the past. In the book, we talk about John Burr Williams, who wrote this book, The Theory of Investment Value. In his book, he talks about how he recognized that General Motors was riskier than a government bond.
To account for that risk, you needed to assign a higher discount rate to its cash flows when you’re trying to figure out its intrinsic value. But that process still leads you to a single estimate, a point estimate of what the security is worth, whereas the Markowitz framework models risk as a possibility of outcomes.
Benz: What are some limitations of mean-variance optimization as an approach to portfolio construction?
Wicas: Well, we touched upon this in that art and science question earlier, but it basically boils down to how to estimate beta. Original estimates of beta were fairly poor or fell short of what realized beta turned out to be. I can give you an example of the conundrum that existed.
For example, Gus Sauter and Jim Troyer, when they were hired at Vanguard in the ’80s, their task was to improve the tracking error of the fund. They looked at the borrower’s product, they looked at different kinds of specialized optimization tools with custom risk models, and they looked at what the tracking error would likely be with those products. And then, they actually built what was called a stratified sampler. The stratified sampler basically takes the benchmark, breaks it up into all kinds of buckets based on risk factors of
This is a very inelegant solution for a portfolio that’s inspired by mean-variance optimization. However, it was fully informed by modern portfolio theory. They knew exactly what they were trying to do about matching risk factors to track a benchmark, but it was done in a completely different manner. Gus and Jim Troyer and others used that tool for a long, long time while the technology of more elegant mathematical solutions slowly caught up. It wasn’t until the mid-2000s that they shifted over to a more standard risk model approach.
That hopefully gives you a sense of the transition, how long it took for the more “elegant” mathematical solutions to be as good as a very simplistic way of doing it.
Benz: The book notes that the first total return bond index didn’t come out until 1973. Did that help practitioners quantify the value of bonds as part of a portfolio?
Clarke: Yeah, I think it was a very important development in the history of portfolio construction and indexes. Before this total return bond index, we had a series of government and corporate bond yields going back, I think, to 1926 or so, but we didn’t have total return indexes. We’re missing a big part of the picture. Without the total return series, you couldn’t really model how bonds and stocks would interact in a portfolio, particularly if you’re looking at periods of economic or financial shocks such as a big spike in interest rates. I think this was a critical development in modern portfolio construction.
Arnott: The book also has a section about behavioral finance. I’m wondering if you could talk a bit about what you think some of the most important insights are from that branch of academic research.
Clarke: Yeah. I guess we think of it in two dimensions. There’s kind of the personal finance angle and then a portfolio management angle. I guess I would say in personal finance, just what we’ve learned about inertia and status quo bias has been huge in corporate retirement plans. As we’ve looked at these innovations—automatic enrollment, default into a target-date retirement fund, and automatic escalation and savings rates. Those have put Americans in a much better position to finance their retirement, and they’re capitalizing on this behavioral finance insight into our tendency to stay tied to the status quo.
Wicas: Look, behavioral finance is a fascinating part of academic finance. It’s been transformative in terms of how people think about the theory, how they think about strategies, and the entire stock market anomalies literature basically was driven by an interest in behavioral bias.
Now, behavioral finance is driven by how people think and decide. It’s drawing heavily upon the psychology literature. When you think about, there are all these different ways in which people have biases in how they think and decide relative to standard statistical thinking. There’s this certainty principle where basically people take high-probability events, and they treat them as if they’re certain. And low-probability events, they treat as being not likely to happen.
Again, sometimes that decision-making, it could be an 85% probability of happening, and they act as if it’s going to happen. And there’s a 30% probability that something is not going to happen, and they choose to act as if it will never happen, so you have that bias. Estimates of probability are also influenced by recency, how recent something has happened, and the future is even just like the past, or this notion of availability. If an event is available to your mind that you know the history of, you then think, “Oh, this future event is just like something that happened in the past.”
Now, where you see this play a big role is, for example, during the Great Depression, people were profoundly influenced by that, and the availability of that experience stayed with investors for decades. After World War II, there were many institutional portfolios where the prudent thing was to stay in bonds and to not invest in stocks because stocks were seen as being terribly risky because, essentially, if you were invested in stocks in 1929, you did not recover the same value that you had in 1929 until 1945.
That sort of availability bias in your thinking led many institutional portfolios, colleges, and other endowments to be heavily weighted toward bonds. And then they missed the huge postwar boom in equities. That kind of bias plays a big role. When it comes to stock selection models, I teach behavioral finance because it’s used to motivate the literature on the stock market anomalies in history. Often, when you actually are deep in the data, there’s actually often a rational explanation for things.
So, momentum strategies, investing in stocks that have done well in the past, have often been said, “Oh, this is irrational. This is a trend-following event.” You’re basically acting upon information that everybody has, and it pays off for you. But when you look at the data carefully—analysts’ earnings revisions, which is another stock market anomaly—if you look at stocks ranked based on the analyst earnings revision and you look at stocks ranked in momentum, they’re highly correlated. A lot of momentum in practice is the reaction to earnings being revised by analysts. If you combine the two, actually you will see that the stocks that are highly ranked in momentum that continue to pay off are the ones that have subsequent analyst earnings revisions.
The flip side of this, though, is that there’s an anomaly, it’s referred to as Sloan’s accrual effect, where accounting principles allow you to take accruals or the accrual principle and count book earnings before you’ve actually been paid because it’s when the sale occurred. Accruals can boost earnings in the short term, and it often reverses. What you see is that adjusting range for accruals often creates a reversal effect in stock prices. And that seems to be, it is actually persistent.
Sloan’s paper was first published as a working paper in ’93. It was published in The Accounting Review in ’96, and it ebbs and flows through time, but it’s still pervasive, and it’s in other markets. Where I’m going with this is that behavioral finance or the psychological biases do play a role, but financial markets tend to correct them perhaps faster than the theory would suggest.
Benz: I’m curious, Nelson, you mentioned that these historical experiences tend to create these biases in how people are approaching investing. As you think about investors writ large today and their experiences over the past 15 or 20 years, can you reflect on the biases that we might all have today based on what we’ve gone through as investors?
Wicas: It’s funny. In thinking about this interview today, I was thinking about how my experience has been with all kinds of people, how rational thinking has been taught all around the world to institutions and to well-healed investors.
This is not an example where the past influences things, but during the global financial crisis, I can remember being at my mother-in-law’s house, and she was a lady around 90 at the time. There was a cocktail party with her other 90-year-old friends. The GFC had just happened, and these sort of well-heeled old ladies were basically saying, “Nope, you need to sit tight, and costs really matter. Just sit tight, don’t make a change, it’s going to come back.” They’re all talking about this thing. “Yes, that’s so true. That’s exactly what we need to do.” It’s remarkable how investing behavior has changed through education.
The other thing that we emphasize in the course, when we talk about this cost argument for indexing, I say to students, “Look, this is not just something that we’re teaching you because it’s kind of interesting and Sharpe wrote this paper about it.” I said, “Look, this isn’t a set of ideas that the big firms like Vanguard and BlackRock and State Street—they’ve literally gone all around the world, and they’ve spoken to every large institutional client and many small institutional clients, and they’ve taught this information.” One of the things you were raising was about why has indexing become so predominant, whereas in the beginning, it was so hard to sell. My thought is it reflects how education can change people’s biases.
Clarke: Yeah. Christine, just thinking about your initial question, I think I just echo what Nelson said. Our recent experience, a bias might be that stocks always recover quickly. There have been long historical periods where stocks have struggled for a decade or more. That could be a bias that affects our behavior.
Wicas: That’s actually a really good example, Andy; because of my age, people seek investment advice and financial planning advice. I actually had a conversation recently, again, with a lady who was 77. She’s fairly well educated, but we were talking about what happens if we have a period of high inflation and how it’s going to crush people’s ability to spend from their portfolios. There is this general sense that we’re not likely to have that kind of experience.
On the other hand, the flip side of it was, we were talking about, well, what would you do? People were saying, “Well, you could sell real estate. Real estate really takes a bath during inflationary periods of time. You’d really have to hang tight on that.” There’s kind of this back and forth where there has been a lot of education that counteracts these biases at the same time. I do think this idea that we’re not going to have an inflation shock is perhaps in people’s thinking.
Arnott: Nelson, you mentioned the role of education in encouraging more people to adopt index-based strategies. I’m curious about why that took so long to happen. The book notes that back in 1995, index funds made up only about 5% of total assets in equity funds. I was covering funds back then. I think the general thinking about indexing was it was almost like this quirky corner of the fund world, and a lot of people sort of thought it was un-American almost. Why do you think there was so much opposition to indexing initially and why did it take so long to gain the type of adoption it has now?
Clarke: Yeah, I think there are a couple of things. First, just in general, great innovations sometimes take longer to be adopted than we would expect. You look at something like the telephone; Alexander Graham Bell invented the telephone in 1876, and it wasn’t really until the 1950s that half of US households had a telephone. There’s that slow adoption of new innovations.
Specifically in the case of index funds, they came from academia, and the original case for indexing was the efficient-market hypothesis, which probably just wasn’t believable to a lot of practitioners. That probably slowed the adoption. Later, the case for indexing became a cost argument. There were early articles by Charley Ellis, and Bogle was making this case. And then the iconic article by William Sharpe, “The Arithmetic of Active Management,” which showed that all investors as a group earn the market return.
Before costs, the average actively managed dollar is going to earn the same return as the actively managed index dollar. After costs, because actively managed funds have higher costs, the average actively managed dollar is going to trail the index dollar. I think that was probably a boost to index adoption.
Finally, I’d say just the accumulation of performance data. I remember when I first bought an index fund in the mid-1990s, I wasn’t super excited. I kind of thought, “Well, I guess I’ll earn the market return. Maybe that’s good enough. At least I won’t get hurt. Maybe I’ll outperform 51%, 52% of the alternatives.” And then you look, and you fast forward 10, 20 years, and you look at the reports from Morningstar, from Standard & Poor’s, and these funds were outperforming 80%, 90% of the peers. I think the accumulation of this performance data year after year just cemented the case for index and with investors.
Benz: For investors who want to continue to look for active managers who might outperform, how would you say investors could evaluate an active manager’s track record to give themselves more confidence that outperformance wasn’t a fluke, that it could have some persistence into the future?
Wicas: Again, in the book, what we’re really talking about is how to form risk-controlled portfolios, whether it’s an index strategy or an active strategy; risk-controlled portfolios are essential to lead to success. When it comes to actively managed funds, there’s this law of active management from Grinold and Kahn, where it basically boils down to the idea that you need to place lots of little bets.
When it comes to evaluating active managers, what’s done in the institutional space is essentially attribution analysis. They look at the track record, and they look at what they are holding and what paid off. And those payoffs, how do they sync up with known exposures to risk factors? What came from true stock selection where it wasn’t a risk factor driving the return? If your performance during a time is largely due to a risk factor exposure, it may well be that there was just luck; you were just in the right place at the right time, the risk factor paid off, but you rode the wave.
But if you have an investment process that clearly is synced up to be betting on a risk factor or betting within stocks within the risk factor, then it may be plausible that they have a repeatable investment process. What you’re really looking for is that people who are managers that are picking stocks that are paying off, that are not coming from just a risk factor, because then the manager is truly generating information in their analytical process that’s leading to placing bets that are starting to pay off.
Typically, when you’re looking at active managers, they do this kind of analysis. They then interview the manager, and they listen carefully to what their investment process is, and they sync it up to the analysis they’ve done. Again, people I’ve known who do this for a living, they essentially are looking to see: Does the manager know why his process paid off? Does the manager know what risk factors he was exposed to? Does the manager know what was luck and what was skill?
And then you can, depending upon how much information you get, you can then try to simulate the strategy quantitatively to see whether or not, over longer periods of time, that strategy’s going to pay off the way you think. Essentially, when it comes to manager selection in the active space, it has become a very quantitatively driven approach that, again, is heavily influenced by academic finance, and the idea is it came out of academic finance.
Clarke: Yeah. I would just add that as an individual investor who might not have access to all these analytical tools, you want to start with a fund that’s low cost, below average cost, that at least gives you a head start. And then try to understand the economic rationale behind a manager’s investment thesis. Why does this manager think a particular kind of stock will outperform? Are you able to see that rationale reflected in the portfolio holdings?
This is something that, as Nelson was talking about, you could do in a much more rigorous way with all these tools, but an individual can just see how the fund has performed during different market environments. If the fund is a value-oriented fund, does it tend to do well in value-oriented markets and does it tend to fall behind in growth-oriented markets? That’s not a problem. That just indicates that the fund’s doing what it says it’s going to do. I would start there as an individual.
Arnott: You touched earlier on some of the academic research on anomalies like the momentum effect and earnings revisions. One of the most widely studied anomalies is the small-cap effect, where if you look at performance before 1981, small-cap stocks had a pretty significant performance advantage, about 3.5 percentage points per year, which adds up to a huge advantage over time. Since then, they’ve had a much smaller performance advantage, actually closer to zero. I’m curious if both of you could just briefly weigh in on whether you think there is still a small-cap effect or whether there may be measurement issues that contributed to it looking better than it really is?
Clarke: I have read that some of these long-term studies didn’t adequately account for the transaction costs and building small-cap portfolios. If you were able to account for those, the outperformance would look much less impressive. I guess when I look at the graph of the periods of small-cap outperformance, to me, it just looks very spiky. I know there are economic stories that can kind of explain why you would expect higher returns from a small-cap company. They might be more financially vulnerable, and you should receive more return for that risk. I guess the spikiness of it just doesn’t give me a lot of conviction that it’s a risk factor that I would expect to outperform. I’m happy to have small-cap exposure in a total stock market fund, but I don’t have any conviction that would overweight small caps.
Wicas: Yeah, I mean the idea that what came out of the studies in the late ’70s about the small-cap effect, it was as if this is always going to persist. Therefore, you can substantially overweight small caps relative to the total stock market, and you’re going to outperform based on that. If it’s a risk factor, it’s going to come and go. That’s why Andy and I would naturally want to have a total stock market exposure in our portfolio because we don’t want to just be large caps because small caps will outperform. We don’t want to just be small caps because large caps will outperform.
However, in terms of the results that were reported in the ’70s, I believe there are studies that also look at the ’70s as a period of time where we had inflation shocks; you had the oil shocks. It was followed by accommodative monetary policy that made the inflation shock basically kept inflation going. It took till the ’80s before they were able to squelch down inflation. It’s possible that those inflation shocks were more easily accommodated by smaller firms, that they were able to pass on price increases and maintain their profitability, whereas big firms just found it very hard.
A classic example would be the auto industry. Before the oil shock in the early ’70s, the US auto industry had gigantic cars that got 10 gallons of a mile. The oil shock forced the industry to radically change what it was building. But it took a full 10 years before we got to the 1980s, when the US automakers were starting to build, like Chrysler was building K cars. Ford came out with the Taurus. I think GM was making—there was a car called a Chevrolet Celebrity that was a midsized car that was actually fairly well-made—and these were profit engines for the firms. Those are different reasons why we think of the small caps as having more risk factor than investment edge.
The other thing that’s gone on is since the ’80s, monetary policy has developed and evolved, and it’s been much more smartly implemented because they’re trying to be credible that they’re going to beat back inflation. And we saw this very recently. We had a spike in inflation during the coronavirus pandemic, and the Biden administration, basically, the Federal Reserve, Jerome Powell, was very adamant about trying to squeeze out inflation again. There’s just a difference in policy that’s happened over the last 40 years and might come and account for the change.
At the same time, like most investment managers or investors such as Andy and I and Ganesh, we all think that the market is less efficient as you go from large caps to small caps. When it comes to being active bets, all managers—quant managers and traditional managers—their active bets tend to be in the smaller of the large names, definitely in the mid-cap region and also the small-cap region. But the ability to implement it as you give in a small-cap region is very much a function of trading skill because there isn’t liquidity that’s there in order to get your positions on.
All this careful trading expertise that I alluded to that Gus Sauter and his team were so great at doing—our small-cap quant funds were very successful because they could easily get the positions on and you could capture the alpha that you saw on your backtest. Whereas I know other people that have worked other places where the trading skill is not the same and the alphas are not present.
Benz: One question we’ve been putting to a lot of our guests recently is what they’re listening to, what they’re reading. For both of you, are there any other books, blogs, or podcasts that you’ve put on your must-read list for investors in addition to your own book?
Clarke: Yeah, I have a few. I love Jason Zweig’s weekly column in The Wall Street Journal. And not to pander, but I listen to The Long View every week. Also Animal Spirits from Ben Carlson and Michael Batnick at Ritholtz Wealth Management, a regular reader of Ben Carlson’s blog and his books.
Other books I’ve really enjoyed—anything by the Bernsteins, Peter Bernstein and William Bernstein. I say if there’s a theme among the things I enjoy, they give me great insight, but they also have a sense of humor. Investing is a human activity, and we all have these foibles, and that’s sometimes sad, but it’s often very funny. So, like Bill Bernstein, you read one of his books, and you might find yourself laughing out loud as you read about portfolio construction. I always appreciate that, always appreciate a sense of humor. I’d say just in general, I just like reading about the industry and investing in general, but those are some of the names I’d highlight.
Wicas: I think my focus is much more narrow than Andy’s. I do pay broad attention to what’s going on in the macroeconomy in the US and also in the developed world. I’m largely reading the popular press to pay attention to what’s going on. I just have a strong interest in macroeconomics and macroeconomic forecasting, so I tend to pay a lot of attention to that. When it comes to investment reading, I have to say, I feel like I happen to find quantitative stock selection really interesting. I tend to read academic literature; I read about studies, and I’m kind of in the weeds about that kind of stuff.
Arnott: Well, thank you so much for joining us today. It’s been great talking with both of you, and congratulations on the book.
Wicas: Thank you.
Clarke: Great. Yeah, thanks so much, Amy and Christine.
Benz: Thanks so much, Nelson and Andy. We’ve really enjoyed today.
Arnott: Thank you for joining us on The Long View. If you could, please take a moment to subscribe to and rate the podcast on Apple, Spotify, or wherever you get your podcasts. You can follow me on social media at Amy Arnott on LinkedIn.
Benz: And at Christine Benz on LinkedIn or at @christine_benz on X.
Arnott: George Castady is our engineer for the podcast. Jessica Bebel produces the show notes each week, and Jennifer Gierat copy edits our transcripts. Finally, we’d love to get your feedback. If you have a comment or a guest idea, please email us at thelongview@morningstar.com. Until next time, thanks for joining us.
The author or authors do not own shares in any securities mentioned in this article. Find out about Morningstar’s editorial policies.
United States