Patrick O'Shaughnessy on Custom Indexing and the Portfolio of You

This podcast is brought to you by Wells Fargo Advisors. Looking for a firm your clients can't outgrow? Wells Fargo Advisors offers financial advisors full balance sheet solutions with wealth planning, investment products, banking, and lending that meet the needs of affluent to ultra-high net worth clients. Build your practice your way and on your schedule with independent and employee model options designed to meet your needs today and tomorrow. Learn more at joinwfadvisors.com.

What do Morgan Stanley, BlackRock, Schwab, Vanguard, and Franklin Templeton have in common? They're all rushing into direct indexing, which is projected to grow at an annualized rate of more than 12% for several years, which is faster than ETFs, mutual funds, and SMAs according to Cerulli Associates. Here to explain what's driving this and what this means for advisors and their clients is Patrick O'Shaughnessy, the founder of custom indexing platform Canvas, and host of the popular and well-regarded Invest Like the Best podcast. Patrick, welcome.

Greg, thank you so much for having me.

My pleasure. Now, there's a lot to cover and I'm hoping and assuming most of our listeners know what direct indexing is, but let's just be safe and cover that. So can you provide a clear explanation of what direct indexing is and what custom indexing is?

Yeah, I was going to say not to throw a wrench in the conversation right from the jump, but I do think that that difference between direct and custom indexing is important, so I'll address both. Direct indexing has been around for a very long time. Probably the earliest example of it is a firm called Parametric, which has been doing this since the '90s. And the concept is really, really simple. So if you believe that the S&P 500, for example, is a good investment, a direct index version of the S&P 500 would be instead of buying the ETF or the mutual fund, You in a separate account for each client, let's say, or each investor, buy the underlying holdings of the S&P 500. So you're getting roughly the same exposure and you do tax loss harvesting on top of that investment. So you try to generate after-tax returns that are 1%, let's say, better than the S&P 500 itself. So all you're doing is taking advantage of the fact that some stocks go down, others go up. The ones that go down, you sell some of, generate a usable tax loss for the benefit of the investor, reinvest it in similar securities, have a similar diversification profile and so on. Really, direct indexing in my mind is synonymous with this idea of tax loss harvesting. For some reason, it's a different name, but it's basically that core concept. I think custom indexing, the thing that we're excited to have brought to the world with the Canvas platform takes that quite a bit further, and custom is the operative word here, which is really to have tax loss harvesting as a key part of what you do. So always be trying to generate the best after-tax returns, but instead build an index that is completely customized to the individual investor based on their circumstances and their preferences. So what does that mean? Well, everyone's got a different tax situation, everyone's got a different risk situation, everyone has different preferences for kinds of securities they may want to own more of or less of. ESG might be one way of thinking about these emerging preferences. Everyone's a little bit different. That's probably the number one thing that we've learned launching Canvas is that some people, most people in fact, that use the platform, 80% or so, do some tweak to their custom strategy that we haven't seen before, meaning the DNA of their strategy is completely distinct and unique, sort of like a unique fingerprint or something. We don't force them to do that. Customization is being adopted very organically to adjust for all sorts of things from tax profile to ESG preferences to asset allocation and strategy and factor preferences. Our vision of the future is one where everyone has their own strategy that's tailored to them, which makes it easier to stick with, easier to adjust and ride through changing times. That's what we're excited about. Tax loss harvesting and direct indexing is a critical component of that. I would just say it's sort of the last generation of this same concept.

What's behind the rise of custom indexing? Because it's really fascinating how many firms are really just flooding in full bore. Not that long ago, there was— I don't know if you want to say a lot of skepticism, but it wasn't crystal clear that this might happen, at least to many. I guess not you, certainly. Tell me what's behind the rise.

Well, I think if you study the history of investing, at large, it actually is a history of technology changes. If you like me were to view something like the limited liability corporation as almost like a legal technology that unlocked all sorts of incredible things in the world of business and shared risk, you could argue that Bogle and the index fund were one such innovation, that the mutual fund itself, commingled vehicles were another such innovation, and that really the march of investing technology or innovation has been towards better solutions for actual end investors. So lower costs, you know, better sharing of risk, more diversification, like all the benefits that we've seen come originally with the mutual fund, then with the index fund, then with direct indexing. And I would just say this is just the natural next stage in that same evolution, that it's enabled completely by software and technology. So the reason this is coming out of nowhere is even 3 or 4 years ago, this would have been basically impossible to do at scale. But with all the modern technology and software tools, we can have a situation where we've got thousands of accounts where they all have their own unique strategic settings that we have to maintain and rebalance and trade against, etc. It's entirely because of what's enabled by software. I think this is just the natural progression in an era of progress that's been happening dating all the way back to the first mutual fund. It's just providing a better solution to the end investor. What are the dimensions of better? Well, it's return, obviously. It's risk. It's cost. I think Bogle's great innovation or realization was that cost matters and costs compound. You want these things to be very low cost. Tax loss harvesting or tax alpha, whatever you want to call it, is like a way of getting more dollars after tax back into the investor's pocket. I don't think this is anything that interesting in the sense that It's just a continuation of trends that have been in place for a long time that are trying to deliver a better overall solution to the end investor.

Now let's talk about custom indexing vis-à-vis ETFs. I believe you once asserted that custom indexing will unbundle the ETF. Can you speak to that and also, you know, to what extent and what might be the timeline or trajectory? I mean, difficult question admittedly, but—

Sure, I just think it's very simple. So if you buy that there are benefits of customization, meaning that people want to make tweaks to their strategy, they want to own more of a value factor or more of a momentum factor, or they want less of Europe in their portfolio for some reason, or whatever the thing is that they want to tweak for whatever reason. If you buy that customization is good, then really the question is, let's compare ETFs or fund structures to separate account structures. The benefits of an ETF are that there's big scale, right? You can share the scale economies generated from a BlackRock or something and have much lower prices to access, say, the S&P 500. There's a scale advantage and there's a tax advantage in higher turnover strategies. If you put a momentum strategy, a pure momentum strategy behind an ETF wrapper, it's a much better tax consequence for the end retail taxable investor or taxable investor of any kind than if you did it in a separate account. The other end of the spectrum is that it's one size fits all. By definition, it's a commingled vehicle,. So you can't make adjustments to an ETF strategy to map onto an individual investor's needs or not even an individual pension fund's needs. It could be any investor. There's no changing them. It's one size fits all. And there's potentially tax disadvantages, which I'll get to in a second, at the very low turnover end of the spectrum, which is what tax loss harvesting was, you know, created to capture and deliver to investors again, but dating back to the 1990s. So my concept of unbundling the ETF is If it's an ETF that has relatively speaking low turnover, which most of them do, and there are benefits of customization and technology allows separate accounts to not be that expensive and they're not. Our marginal cost per separate account is extremely low. I would hope that we can get it below $100 on an annual basis. The scale economy advantage from the operation side is just, if not totally gone, on its way out to a large extent. The after-tax benefit of low turnover strategies with tax loss harvesting is a real advantage versus a commingled vehicle. I don't forecast that ETFs are going away. I mean, they're definitely not. They're great vehicles for tons of people. They'll be long-term holdings for tons of people. But we've already seen in many cases, our clients coming into us and saying, "We have an allocation to," the blank dividend ETF, can you replicate the strategy, which of course we can. We're quants and all the index methodologies are public. Usually, we can improve the strategy from a factor standpoint and then unlock all these benefits of loss harvesting and customization on top of it. So I'm not just guessing at this unbundling. We've literally had it be a major request and done it many times where we've gone after the principal goal of, let's say it's dividends, we can design a better dividend strategy and then you get these other benefits on top of it. That's what I mean when I say unbundling is for low turnover strategies where customization can be powerful, why wouldn't you go with a separate account and get— have your cake and eat it too?

Is there a certain client profile for which direct indexing might not make sense to have not enough assets to justify the added complexity or work involved?

Look, I'm a firm believer that investors should do what they can stick with and what they understand. So the extent to which this creates some real or perceived complexity that the investor just doesn't get— and look, there's sometimes there's tracking error in these strategies, so your performance won't be the exact same as the S&P 500 every year, every quarter, every month. So if that's a problem, like, just buy the SPY and keep it very, very simple. I also think that the current limitations are real in the sense that too small of an account, you know, doing that. We couldn't do this with a $10,000 account. You know, we're getting there and being able to do it with $100,000 account. Right now, our minimum is $250,000. So these are big accounts on average that we're doing it with today. That will change with technology though, and it's already— that march is happening. Like, there's no turning back on that. Like, fractionalized share trading, you know, the low cost of trading at most brokerages, all of these things are nice complements to custom indexing. So I would envision a world in which, you know, 5 years from now, to pick a random number, this is accessible to somebody in a lot of ways, maybe not in every way, but in a lot of the same ways that has a $10,000 account and we're working our way towards that.

Right. I know Fidelity with its FID Folios is coming out with a very low minimum. It seems like, right, the trend will be lower over time. Yes, agreed. Also, in terms of the growth, according to a report by Morgan Stanley and Oliver Wyman, direct indexing is expected to control $1.5 trillion in assets by 2025. And that's roughly triple levels from 2020. So yeah, the trajectory is pretty clear directionally at least.

Maybe if I could offer a comment there, I think that's drastically understated. And the reason I say that is, first of all, these kind of industry-wide surveys I think are almost never accurate. Not because of the fault of the forecaster, it's just an unbelievably hard thing to forecast. There's so much going on that leads into the actual outcome. But from our own experience, we're seeing firms that adopt this first adopt it for like the most complicated client, let's say. Someone with the most special needs, some unique problem to solve, some concentrated stock position that we need to help them manage out of, which is something that custom index can do uniquely well. Then as soon as they adopt it, it starts to then spread across the whole equity book of the RIA that we are working with. The adoption is not linear. This is not a linear trend. This is a major practice decision that many RIAs are making. In some cases, we're seeing basically every new dollar of equity investments that comes through an RIA goes on to Canvas. When you have a platform like that, that becomes just like a policy decision, I think that the assets can ramp much faster than you would draw out in a linear extrapolation, which I think is how a lot of these surveys are built. So I would expect the number to be extremely high across the industry. Okay.

Can you briefly just talk about the creation of Canvas and the O'Shaughnessy Asset Management that did sale to Franklin Templeton?

Sure. So much of my interest in business could be summed up by studying the Amazon Web Services story, which I'm sure a lot of listeners will be at least somewhat familiar with or they'll have heard of it. The story is really simple, which is that Amazon always has been an incredible operations-focused business. With an unbelievable underlying complexity to get us all, you know, the 2-day shipping and the million other things that they do for us. And part of that was managing digital, you know, the digital side of the business. And they had built up incredible expertise in managing that for themselves. They even actually had a foray into potentially managing like websites for companies like Target in their very early days that was ultimately abandoned. But what they realized with Amazon Web Services was Okay, we've got a massive cost center that we can turn into a revenue opportunity because if we need all this servicing, you know, cloud, basically what we now call the cloud to run our operations, well, why don't we offer that same thing as a product to other companies? And our— we studied that in depth because we had built up like Amazon a crazy amount of what I would call cost center technology at OSAM to run our quantitative equity investment and research process. We had been around a long time managing many billions of dollars in active quantitative strategies, factor-based. I think the insight was, wow, my joke was always we'd built a Death Star to shoot a mouse. We had built this incredible technology platform for research and trading and rebalancing and backtesting and custom performance reporting and all this other stuff. We had built all this tech and we were using it for one very narrow purpose. The question became, what if we turned this over to our outside partners? Like, what if we gave them the keys to the desk, so to speak, and see what they did with it? And that's kind of the conceptual way that Canvas was born, was to say, we're using this for one reason, let's see what other people might use our same toolkit for. And, and what we found, as I said, is that they loved it. They started using it for all sorts of ways that we couldn't have imagined at the beginning, and it's proven to be very powerful as a result. So we built an initial web-based software platform that feels very much like building a Tesla online or something. So if you've ever— if you want to know what Canvas feels like to use, just go build it. Go build a BMW or a Tesla online. It sort of walks you through some very beautiful screens of software that are kind of easy to intuit and lead you through a whole bunch of different choices that you can customize as you build your unique strategy. And, uh, and that's how it was born. So it was born in you know, 2019 or so is at the end of 2019 is when we really started taking opening client accounts. It started to scale very quickly from, you know, no assets to, you know, over $2 billion in pretty short order. And we started partnering with firms we had never even been able to get in the door with before, to be totally honest with you, because we were providing a platform holistic solution, not just a point solution as part of their portfolio. So I think Franklin to give them enormous credit, saw the potential in our platform. Obviously, they bring ridiculous scale resources that we just didn't have. We were and are a relatively small team, 40 to 50 people. They're a trillion and a half of assets and relationship with every investor in the world and crazy data and technology resources and a senior team that wants to win in this category. I mean, they bring a bazooka to a knife fight, so to speak. And they approached us with this idea for partnership that we explored and consummated last year. And here we are today. So it moved pretty fast, but it was built, Canvas was built on a decade's worth of software and technology that we had built for our own purposes, sort of in the spirit of that Amazon Web Services story.

Okay. And at the time of sale, is it correct that roughly 28, 30% of your assets were in direct indexing? Is that ballpark right?

Yeah, that sounds about right.

Yeah. Let's pivot a little to investing. What are some of the trends? What are you seeing? What are investors looking for right now?

Well, I mean, look, this is going to be an unoriginal answer because I'm sure everyone would say the same. With what's going on in the world, I think everyone is just incredibly mindful of risk right now. That comes after a period, a long period of time, even with the terrible vicious you know, instant bear market we had at the start of COVID where we bounced back very quickly from it, even taking that into account, markets have effectively been up and to the right. Like, the right decision has been to put more risk on for a very long time, and the right decision in the face of any instability in markets has been to double down on risk yet again. And I think what we're seeing now is, with inflation handicapping the Federal Reserve and what they can do to inject liquidity into the system, with the inflation that's resulted maybe from the fiscal response from COVID Everyone is just wondering, are the things that made that true still going to be true in the future? And there seems to be a lot of instability in the world, geopolitical, markets, valuations are still arguably pretty high. So that's the number one thing I hear on people's minds, on investors' minds is, how can we adjust for or create a lower-risk strategy. So like, for example, one of the most popular new things on Canvas has been to allocate to something that we call defense and stability. And the concept here is like better downside protection, lower volatility strategies as a factor that you can bake into your Canvas strategy if you're an investor. And, you know, it's been popular, I think, because of the core idea, but also because it's done really well through a period like this. When there's been vicious drawdowns in expensive growth stocks and fairly muted small drawdowns in, say, like value large-cap equities that are in the single digits still. This is not even really a correction in those kind of more defensive type stocks. That's probably been the single most popular trend or theme. Again, I realize I'm not saying anything unique or interesting there, but that is what shows up in our data.

Right. I mean, year-to-date, the NASDAQ's down close to 18%, S&P 500 about 12%, Vanguard just using VTV ETF is a proxy, it's down about 4 or 5%. So yeah, there's a real split there on the risk curve.

And if you go one click deeper into the non-earning or no-profit growth technology stuff, I mean, it's down 40, 50, 60% in many cases.

Yeah, I mean, I think the indexes actually are really concealing the amount of damage and pain that's out there. There is so much more hurt than people realize, or well, those who are feeling the pain well know. But if you look at how many— let's say growthy names, which would have rich valuations on a conventional measure, are just— so many are down 50% to 80%. I mean, dozens and dozens of stocks. It's pretty amazing. And a lot of them are very solid companies too.

I think— I can't remember who said it, but I love the idea that there's no company so good that a bad valuation can't ruin its investment prospects. And I think we're seeing that play out. You price something at 100 times earnings or 100 times sales in many cases, it's hard to earn a good return from that starting point. And that's what we've seen so far.

So one thing I was curious about now, a lot of tech companies often score well on counts of ESG. And what's really interesting with what's been going on in the market is that, well, as we've mentioned, tech stocks have— a lot of them have gotten battered. And then a lot of companies like oil stocks are doing exceptionally well. They're the only sector, I think, in the S&P 500 that's up this year. And even though coal companies are a tiny part of the economy, I think spot coal prices are at a record high and whatnot. So the interesting situation is that a lot of these ESG-friendly companies are suffering mightily. And a lot of the companies that, you know, frankly are not helpful for the environment are doing well. In terms of that tension, is there anything that you perceive that with people like sticking to their guns with ESG, are there people who are maybe picking, abandoning, let's say, stated principles and saying, "Screw it. I'm going to go with oil stocks. I think they're going to go up." I don't know if you have any insight into that. It's kind of an open-ended question.

Yeah. The insight I would have is that the percentage of people— Again, the nice thing about having this platform is I don't have to guess at it. I just know the exact percentage of people that choose a given setting or don't. It's just objective. It's only about 15% or 20% of people that we've seen, and we think we have a very good cross-section of styles of investors, geography, like we have a good sample size here, including lots of investors or advisors that we work with that are really focused on ESG. So it's only about 15% or 20% of our accounts make active adjustments for something in ESG. It doesn't have to be oil and gas. Often it is because of its relationship to climate, which is sort of the biggest ESG issue we see. But it could be around things around gender or social causes or whatever. And, uh, the answer is no. We don't see or expect a ton of turnover of those settings as a result of what's going on in the world today, because typically for those investors who are obviously— they're a very vocal minority, meaning they care often very deeply about those settings, and in many cases it's the single thing they care the most about. Yeah, in their portfolio. And I don't think those decisions change lightly. Exactly. So I guess that would be my revealed answer is like it's a low percentage, but it's a very staunch low percentage that my guess would be stick to their guns. I think it's a much interesting question, much more interesting question and very much beyond my expertise or pay grade. Like, wait a minute, does this mean that the best thing for ESG would actually be to pump more oil right now? Like, I think that's a really interesting case to listen to and think about for the long-term, you know, health of the country and of the world and of the climate and everything else. Like I said, it's way too complicated for me to wrap my head around, but an interesting question for sure.

In terms of insights into what people are customizing their portfolios with, are there any emerging trends that you're seeing? Things— you mentioned you'll see things that surprise you.

Yeah, I would say there's one major one that's been— it's the major insight that I think we've had since we started this experiment, which is that basically 100% of people that are taxable have some tax setting set up in their account, in their strategy. And the big revealed insight, I would say, is that that extends far beyond just loss harvesting. So we started the conversation by saying direct indexing equals tax loss harvesting. That sort of is the value proposition. What we've realized is that there's tons of demand and value in other ways of managing taxes that aren't about generating losses. It might be a very popular one is limiting gains. We have lots of people that say, "We want factor exposure, but factors generate gains." Factors equal turnover and turnover equals taxes in the long run. We might see someone say, "I want an X% exposure to factor XYZ, but I want you to limit my incurred capital gains to $100,000," or whatever it is, some number, some raw dollar amount or percentage of the portfolio.? And that is an extremely popular feature. It's not loss harvesting, it's a tax setting and limitation. Another one is, help me transition from Portfolio A to the Canvas portfolio in the most tax-friendly way possible. Again, usually you're incurring gains, not losses as you do that, but there's an incredibly interesting way you can use technology to do that in a really effective way. Sometimes it's, transition me away from a single stock that I own way too much of. "And I don't want to go into an exchange fund. I don't want to do options hedging strategies. I want to solve the problem permanently. So pair losses elsewhere in the portfolio with strategic gains taken in my Accenture stock or my Facebook stock or whatever and make it a tax-neutral transition." Those are just 3 examples that we have seen pulled out of us over and over again since we launched Canvas. So I would say that the insight is, the trend is, tax research and tax strategies that aren't necessarily just about losses and alpha, but are about other considerations too.

I wanted to just bring up a couple of the— I don't know, concerns or criticisms of this. One is bringing up a behavioral aspect. I mean, one of the benefits of indexing is that you're not going to be tempted maybe to do a lot of meddling or playing with your portfolio too much. A little bit, set it and forget it, which often is a often the right thing, maybe not perfect, but generally good. Tell me about the potential risk for that or that temptation to maybe get involved and make too many decisions where maybe it's no longer a productive thing, right? There might be a point of diminishing returns.

Yeah, I mean, here again, maybe I would appeal to some actual evidence that we've seen so far. If you compare our, what I'll call churn or turnover, us getting fired in our old active, high active share quantitative strategies, to the same experience in the same period with Canvas, you see a drastic difference. The Canvas accounts look much more like I think you would see at like a Vanguard account or something or a BlackRock account or a State Street account, which is fairly low turnover, that these are buy and hold like investors and you see buy and hold behavior in lack of account liquidations or turnover or changing ETFs or whatever, which I think is what you're asking about, these behavioral changes or arguably mistakes. We just haven't seen that with Canvas like we do in a traditional active management business. I think that will remain true. I think there's even a case to be made that there is better behavior with a custom index than you would see with an S&P 500 ETF or something like this because you can go in and make small tweaks, which we definitely have seen people do. It seems to be that that is the stress response, not liquidate the account and go to cash, but rather like, "Oh, geez, I have too much Europe right now. Can we dial Europe down with a click of a button from 30% to 5% or 10%?" Whatever it might be. We've seen that a good amount, and that seems to be their stress response behavior, which I would argue is good. It's good to stay invested for the most part in equities long-term and not try to time things in and out. There's this IKEA effect that I always call it, which is when people have a hand in building something, which they do in this case. They're choosing something that's tailored to them, they tend to value it more. It's like a version of the endowment effect. What we have seen so far in Canvas is unbelievably good behavior. Frankly, in our old business, we didn't always see that. Flows would follow performance. When we had a bad run of performance, we would lose a lot of money. When we had a good run of performance, after the good performance, we would get a lot of money inflows from customers. I don't think there's any way around that in traditional active management. We've seen the opposite of that with Canvas.

Hopefully that addresses that concern. That point actually resonates. I mean, if you buy a fund, actively managed or an index, it's very much transactional. It's really devoid of any emotional dimension or affiliation. It's simply trying to make money. To your point, with this, you're presumably investing in things that are near and dear to your heart. Let's say you do underperform a little bit, it might not be that upsetting to the investor because they're not in at 100% for the returns. Of course, they want to make as much as they can at the same time. Yeah, 100%. Okay, so I did want to ask you, since we're getting a little bit near the end of the program, to share an actionable idea if you can with advisors to help them better understand custom indexing and how it might be able to help their clients.

Yeah, so hard not to give a self-serving one, but I do think this is one that you could do it a lot of different ways that we think we'd be good at doing, but certainly wouldn't have to, which is to reorient yourself with what the factor profile of your investing accounts are. I think because factors sort of went out of favor, specifically the value factor went so out of favor, my sense being in the industry was that people kind of stopped caring. They stopped thinking that factor profile, the valuation of your overall portfolio or the quality of your overall portfolio or whatever, whatever the underlying factors that you're interested are, didn't matter as much. That what you were supposed to do is ignore those things and that returns were driven by these outliers that could trade at whatever multiple for however long. I think we're going to see a return to caring about the underlying factor profile of a portfolio. And very often the very first step we take is to, is to ingest everything that somebody owns— other funds, ETFs, single stocks, SMAs, whatever— combine them all and do an overall assessment of the factors in the portfolio. And very often what you see is a look of surprise on the other side, like, geez, I, like, I wouldn't have guessed that the overall thing looked a certain way, and yet it does. Maybe that's not what I want, or maybe it is what I want. I'm doing a great job. But I do think it's like this very it's an actionable thing that you can do to say like, what do I really own here? Like, yeah, I own Amazon and I own Facebook and I own some railroad and whatever else, but like, what is that? What is the DNA of the portfolio? And is that DNA aligned with what I believe about, about markets or the world? And I think that's, for advisors especially, a great way to have a good productive conversation with a client or prospect is to frame it in terms of portfolio factors. That's the actionable insight that I would suggest. Go do that for your own portfolio or with a client and see what you really own.

Right. Like almost doing a physical from top to bottom, right?

I agree. Yes. Very good analogy. Yes.

I'll start using that. Okay. Well, my guest was Patrick O'Shaughnessy. For more advisor-specific podcasts, please check out barons.com/podcast. For The Way Forward, I'm Greg Bartalus.

This podcast was brought to you by Wells Fargo Advisors. Looking for a firm your clients can't outgrow? Wells Fargo Advisors offers financial advisors full balance sheet solutions with wealth planning, investment products, banking, and lending that meet the needs of affluent to ultra-high net worth clients. Build your practice your way and on your schedule with independent and employee model options designed to meet your needs today and tomorrow. Learn more at joinwfadvisors.com.