Direct Indexing: A Paradigm Shift in Investment Management

On episode 34 of The Wealth Cast, Chas is joined by Dave Twardowski, Vice President and Head of Investment Strategy at Vise, an intelligent automation platform that delivers personalized investment solutions. Dave discusses how he works at the intersection of tech and finance and shares how direct indexing is transforming the industry.

Listen here:

The summary below has been created by a professional transcription vendor upon review of the recorded presentation. Please excuse any typos as well as portions noted to be inaudible.

Hello, and welcome to The Wealth Cast. I’m your host Charles Boinske. On this podcast, we bring you the information that you need to know in order to be a good steward of your wealth, reach your goals and improve society. Today I’m joined by Dave Twardowski, Vice President and Head of Investment Strategy at Vise. Dave is a mechanical engineer by background and received his Masters of Science in Mechanical Engineering from the University of Michigan, and then his PhD in Engineering Sciences from Dartmouth College. After initially working on projects such as designing flight control algorithms for the Blackhawk helicopter at GE Aviation, Dave transitioned to the investment management industry, with time spent prior to Vise at Dimensional Fund Advisors, and Avantis Investors. Dave and I are going to discuss the evolution of investment management to include one of the most recent developments, which is called direct indexing. We’re going to discuss what that exactly means, how it’s implemented, and what the future may hold for investors. I hope you enjoy the conversation.

Dave, welcome to The Wealth Cast. I’m really pleased to have you here today.

Thanks. Pleasure to be here. Thanks for having me.

You’re very welcome. We’re going to eventually start talking about direct indexing and what that means in this conversation, but I’d like to start sort of at the beginning for you in terms of your own personal experience. I noted in your bio that you started your career out of school programming software for Blackhawk helicopters for GE—GE Aerospace, I suppose. And today you’re overseeing the implementation of direct indexing strategies. And I wondered if you could talk a little bit about what that migration was like for you and how that happened?

Yeah, and it’s a great question. Probably not so obvious why jet engines lead to direct indexing. But I think it’s really about the math and the coding and the science behind it more than it is the domain. 

Yes, starting at GE, it was involved in some fascinating problems there with controls. So basically, what you’re doing, you know, you have fuel in and rotor speed out, and you’re trying to write algorithms to control the jet engines. And largely, the physical properties in the model of the engine is known. So what you have to do is to be able to control that. What was really interesting to me was the inverse problem, where you don’t actually know the model. It’s not as physical or as concrete as a jet engine. It’s something like—maybe it’s statistical, maybe it’s learning behaviors, or taking data and inferring models using statistical methods. 

That’s kind of why I went back to school, and just really enjoyed taking data and trying to build up models and infer or build statistical models of what’s happening from the data and then in a robust and reliable way. And we did that through a lot of domains. We did that in the security domain. So we worked and collaborated in some partnerships over the summer with the NSA. And we also did that in the finance domain. So I think, you know, it was really just looking for toy datasets and things to play around with, to build models and to understand more. 

And one day, you know, we were talking there and we called up NASDAQ. They said, Hey, we’d like to do some research on your datasets. And this, of course, would never happen in the finance industry. They’re like sure, we love academics, we’ll give you the entire terabyte dataset for free. And so that was—And that was, you know, a data mine, a big goldmine of things that we could explore. And that was right around the time of the flash crash, if you remember May 2010, I think. 

And so it’s just a fascinating time to do research on behaviors of high frequency market participants. What are they doing? What is the causal effect on the markets? And so we quickly started doing a lot of work on that and then realized, like, hey, we need somebody really involved in finance. There’s a lot of domain knowledge that we’re missing here. And so my advisor, we were talking about it and you know, got connected. Really, he’s like, Yeah, I know somebody in finance. He knows a thing or two about finance and, you know, connected me to Ken French. And that’s an understatement, right? So, yeah, worked with Ken and collaborated on some of those projects. 

And gradually, that was kind of my transition into finance in many ways. Started in grad school and obviously, learned about Dimensional and then started working there and was there for about eight years. And really took what we were doing in grad school and started first applying it to trading. So we did a lot of trading research and developed algorithmic trading, measuring costs globally, and then transitioned to portfolio design and retirement planning design as well. And then, gradually, as I explored other things that interested me, I came across SMAs, and really was fascinated and worked at Avantis for a couple years and really started to work in SMAs there. 

And the thing that really was the ‘aha’ moment was, you know, we had conversations with advisors for a half hour, or it was scheduled for a half hour, and we’d sit down with them, and we’d start talking about SMAs. And it quickly turned into a two or three hour conversation, and they were starting from one person in the room to five people in the room. And we just realized, like, wow, this is something that really could be game changing in the investment space—just the ability to personalize. 

And so that really resonated with me. We also had other folks flying halfway across the country saying, we’re really interested in this, can you do XYZ? And so you just had a sense that there was a massive demand there. And I think in a lot of ways, it was fascinating to me, because in the market, if you look at some of the direct indexing, it’s like you have technology companies that have come up, and have provided those direct index solutions. You also have asset management companies that are incredibly good at building portfolios. What fascinated me was that intersection between the two—can you build a company that has tech roots? Because you have to really have good technology to scale personalization. Can you do that, while still building really solid, robust portfolios with a financial science lens? And that’s really where we are today at Vise.

That’s really interesting. Just to step back for a second and for the listeners, just to define SMA. SMA is just an abbreviation for separately managed account, for those listeners who haven’t heard it before. And in the early days of my career back in the mid 80s, or so, this was a hot topic, because there was this idea that individual investors could hire money management firms, and they would personally manage a portfolio for you. And the degree to which it was actually personally managed is up for debate. And I think not very personal is the truth behind that. And in addition, they were very expensive. It was not unusual in those days to pay 3% for a separately managed account, which today would just be—no one would do it. 

But I think it’d be interesting just to put things into context a little bit for the listeners, sort of the evolution towards what we now call direct indexing, or what we’re going to refer to in this conversation as direct indexing. I know we’re going to define that a little bit more. But, you know, when I started my career, a quote unquote, widows and orphans portfolio was 30 blue chip stocks that pay dividends. That was the appropriate thing to do—all US stocks, by the way—that was the appropriate thing to do for widows and orphans. And today that would be looked at as basically malpractice. Right? I mean, it’s not nearly diversified enough, there’s too much individual security risk, it’s too expensive to trade, all those sorts of things. But over time, we migrated from sort of individual securities to pooled asset vehicles, like mutual funds, and ETFs came along, exchange traded funds became cheaper and more tax efficient for investors. And now it seems like on the horizon, or just at the horizon, is this concept of direct indexing? So let’s define that a little bit about, you know, in terms of what that actually means, and then we can add some further color. 

Sure, what I would call direct indexing is the ability to design, implement and provide intelligence on portfolios in a personalized way that helps investors achieve their end goals.

So does that from your perspective, that you know, again, just comparing and contrasting the traditional separately managed account to a direct indexed approach. We’ll forget about all the things that happened in between but just for clarity, just use those two things as the book ends. In the old days, you would build a portfolio, and that portfolio would be a selection of X number of stocks that were actively managed by the firm sponsoring the SMA with a specific objective in mind. The big selling point was that you control the taxes in the portfolio—you the investor. 

But most of those strategies were actively managed, which sort of became old fashioned pretty quickly with the advent of all the research that came along from academia over the subsequent 20 years. And now we’re sort of moving towards separately managed accounts in name only. Under the hood, they’re totally different. Because you’ve got the ability to build that portfolio with the characteristics that you think jive with what’s come out of academia in terms of expected returns and diversification, etc. And you can do it much more cost effectively, because trading costs have dropped significantly over that intervening 30 year period as well, correct?

Yeah. And I think that ties into like, really what you’re getting at—what has changed that allows that transition from a concentrated basket of stocks, or bonds, or other securities in an SMA, you know, really started in the 70s, to what we know direct indexing to be now. And part of what you said is custodian fees. 

So at the end of 2019, likely influenced by some of the innovations that happened in the market, custodians drop their cost of trading stocks and ETFs to zero. That was a huge deal, because now you have a collection of stocks, where every time you went to trade them, you didn’t have to pay a fixed cost. And so the cost of that is just dropped dramatically. You know, what you’re also mentioning is some of the financial science innovations over the last 30 or 40 years. And first it was, you know, some of the efficient portfolio design and some of that influenced the index funds and started the index funds in the 70s. 

And then, you know, you had factor investing and Dimensional was a huge influence on factor investing, said, you don’t have to have a market cap portfolio, you can tilt to higher returns. And then naturally, I think that started to integrate its way into SMAs and transitioning from concentrated to something where you had broad market like strategies with tilts to higher returns. Part of what needed to happen there, if you have SMAs, a number of different SMAs, is technology. So the technology is improved to go from concentrated portfolios to systematically managing hundreds or even 1000s of securities in an SMA format. So I think that the technology for scalability was another big one that led to those changes. 

There’s also an aspect of market demand. So the market is changed and the demands of those investors have changed, you know, that there’s a large transition now, in terms of who holds the assets. There’s a lot of millennials there that are looking for, really to express their values, their beliefs, some of their needs, through deep personalization. So if I wanted to invest in more environmentally friendly companies, how do I do that? How is it reflected in my portfolio in my own way? So I think that’s one of the things you’re seen, there is just a large market demand. 

There’s also I think, you know, the advisory space is competitive as well. And so thinking about it as a tool for advisors to better tailor portfolios to needs, risks and preferences. So there’s an advisor value added proposition there as well. And then you start to think about, so all these things kind of came in place, and I think, the custodian fees, and that confluence of market demand and technology started to, you know, bring this, this wave of what we now know is direct indexing. 

What may change in the future to maybe facilitate this more, I think, what you start to see is fractional shares. One of the things that is still kind of challenging today is the ability to invest lower AUM levels. But Fidelity, for example, already announced that they’re providing institutional access to fractional shares. And we know Schwab and some of the other custodians do this on the retail side of things already. But it’s really interesting to think about fractional shares being available for institutions. And now we can invest down to much, much lower, potentially, levels of AUM, and still have that broad diversified basket that you might have in a mutual fund or an ETF.

And you can do it cost effectively, because the big hurdle was small amounts of transaction with some minimum fee associated with that transaction made it impossible to execute those types of portfolios in the past, correct?

That’s right. Yeah. So it may not be an issue, as much of an issue if you have a few securities, but when you put, you know, 1000 securities and there are many securities and you’re evaluating every day, you may trade a number of stocks each day, that those costs add up. So the ability to trade essentially at zero cost in the stocks and ETFs is a critical component to make these cost effective.

Yes, yes. And the benefits of the customization in sort of today’s market really are not adding significantly to costs, right? Because you can trade in smaller lots—you can more easily with technology customize that portfolio cost effectively for both the provider and for the end user, I imagine, where you couldn’t do that in the past.

Yeah, the customization, I think it kind of comes back to just having a good frame, a technological framework, and also an investment framework to address customization. So like, being able to give the advisors and their end investors a variety of different options and levers, but still within a framework where you can apply technology and scale that and still have a robust offering. So you do have a one off customized portfolio, but it’s done through a financial framework that, you know, you can easily scale up and modify.

So just for the listeners’ perspective, or for their information, one of the applications for this might be, and correct me if I’m wrong, a concentrated portfolio of low cost basis securities that you want to diversify around, you can more easily do that today. Because you can customize this index to get the aggregate characteristics you’re looking for in a very systematic way that is very compatible with what the current research in academia and economics is. Does that make sense? 

Yeah, that’s right. I mean, part of it is, is the ability, like let’s say you had, you know, a client that just had a liquidity event, and you know, they have a deep exposure and maybe one stock. Most of their net worth is in a stock, I think it’s incredibly important to maintain diversification and be able to surround that with a broad diversified basket. And that’s really where customization comes in. 

Another important component of this is the ability to automate transitions. So you won’t, with this concentrated, large omitted gains, you won’t be able to remove that completely—it may take a number of years to do that. And so one of the things about direct indexing, and this is kind of why comes back to technologies, that automation to transition, that large concentrated position, and diversify it over time. That’s another component where you start to think about okay, like max cap gains, can I set this lever to say, I want my client, I don’t want to exceed this gain threshold in a year, and then have some automated rebalancing software basically transition you to that desired portfolio or target gradually over time in a tax efficient way. And then that rebalancer will also basically manage the gains, and then offset those with harvested losses. So the transition time will reduce, if you can take the harvested losses and offset the gains to do that.

Right. And these days, we do that most often by hand, so to speak. You know, our systems will tell us whether there’s a loss, and we go in and make the transaction and take advantage of it, or take a gain when it’s appropriate. But this newer, sort of next evolution of that will allow us to do that, or allow investors to do that in a more automated way.

So that’s—it’s really just what you just said, because I lost track of how many advisors, you know, that I’ve had a couple conversations like, well, you know, 2020 rolled around. And you know how we had that big drawdown, partly driven by COVID. And towards the end of the year–I was talking to advisors, like, yeah, I missed my opportunity to harvest losses, and it went back up. And it’s like, That, to me is a very solvable problem with automation. And that’s something that the industry can think and get better on, is providing the automation, the rebalancing software, to know when things draw down, and to be able to realize that that is a situation where you can harvest losses and be opportunistic about that.

Yeah, I think the automation would lower the cost of actually performing that function as an advisor, or as a wealth management firm, that potentially is very valuable. But it’s, I think we would agree that, you know, focusing on your costs, and being disciplined and being broadly diversified, and washing your taxes are like the four things, the four most important pillars of portfolio management, and we’ll leave aside the factor part, the size, value, market exposure, but if you can really nail those four things, you’re likely to do better than the typical investor, and whatever do better means, but those are controllable issues. And, you know, the more we can do that, the better off people are.

Yeah, that’s right. That’s really where that science comes in, is how do you blend in those trade offs to achieve a robust portfolio on a personal level? There’s really no free lunch here and we want to achieve broad diversification, we want to—I will start with an index fund but you know, if clients, given their risk tolerances, maybe want to tilt to higher returns, those are trade offs. And then if you think about harvesting losses or managing taxes and then another component is managing turnover. 

This kind of gets into some of the hurdles. It’s not a silver bullet. I think direct indexing, we have to be careful. And if you have higher turnover portfolios in an SMA, you will incur more gains. And so that kind of counteracts some of the losses harvested. So it’s not just the ability to harvest losses in perpetuity, we have to be careful about how much we harvest. Because eventually, you know, the more you harvest, the more you build up embedded gains. So you have to manage that trade off. So it’s really about capturing higher returns in a diversified way, the trade off between gains and losses, and then mitigate costs.

Yeah, it’s clear, it really seems to be sort of the nexus of all these ideas that are percolating through, have percolated through finance for—well, during my career, 30-35 years or so. They’re sort of coming together in this solution. That’s the way it appears to me. So it’s really interesting to learn about it from you. And I know, you know, the industry is sort of on the very edge of introducing this concept. I don’t think—if I bet you if we did a Google search in 2019 about direct indexing, the amount of hits you would have gotten is a fraction of what it would have been today. 

It’s a fascinating time to be involved in this space. I think, you know, it seems like there’s not a week that goes by that, you know, maybe I don’t see an M&A or you don’t see another direct index company come up. And that’s all really exciting. And I think there’s also a ways to go; there’s a lot of improvements that I think we can make as well. This kind of gets into some of the other things that we have to improve—reporting and benchmarking is a big one. So when you talk about customizing portfolios, every portfolio looks different. So how do you report on that. Custom benchmarking and using allocation weights to drive benchmarking weights is a really important concept so that you can understand that—the performance differences there. 

I think another interesting thing to think about is asset class exposure. There’s definitely a benefit to having commingled vehicles, and single stocks and bonds in the same portfolio. Certainly, you know, if you’re looking for single exposures to single stocks through ADRs, and US domiciled securities that are actually foreign companies that, you know, you can achieve that exposure in some regions—international, but for example, small cap international or emerging markets. Maybe there you can capture a smaller percentage of emerging markets through US domiciled securities. So interesting to think about other listings that may become available in the US to provide further exposure to different regions.

Right now, it’s a constraint, right? It’s a bit of a constraint on the direct indexing.

Yeah. And that’s why it makes a lot of sense. ETFs can, you know—you have international custodians, you can get exposure to a variety of different asset classes. The role of SMA is, I think the optimal balance is a blend between single stocks and bonds and commingled funds. And that’s because single stocks, you have all these personalization benefits. You can apply losses to different areas, your portfolio, but the commingled vehicles, you can get exposure to different asset classes. You can’t currently get exposure with single stocks right now, primarily because you’re working with US custodians. So you’re accessing different regions around the world through US domiciled securities ADRs. So there are benefits there to holding ETFs as well.

And then for clarity, when you talk about individual stocks, you’re talking about individual stocks in the context of a diversified direct index portfolio, not concentrating a portfolio in one wonder 10 Different companies. You’re talking about the broad diversification of those individual stocks, not stock pick, per se.

Yeah, I mean, the science is pretty clear. And this goes back decades. There is a huge benefit to diversification. You know, there’s a lot of idiosyncratic risk with holding a smaller number of stocks, and you’re really not compensated for that risk. And now, you still want to think about diversification. Even if you’re looking to invest due to preferences or needs or make exclusions, you always want to think about a broad perspective portfolio.

You know, as you were mentioning that, it reminded me of something that I used to hear when I first started in the indie industry back in the early 80s. There was this idea that diversification was bad, because you missed the opportunity to knock the ball out of the park by picking one or two good stocks. It was actually, it was sometimes termed “deworsification.” And now, the evidence is so overwhelming that that statement or that point of view was, it couldn’t have been more wrong. Right? From a financial research perspective, from an academic perspective, but it’s funny how, in my own career, that concept has been disproven and buried not once, not twice, but 1000s of times in research activities, etc, research papers. It just shows you how much things change over time. And it’s important to be aware of what’s going on with what’s coming down the road, which is the whole reason to have this conversation.

Yeah, I mean, if you want to think about as missing the opportunity to invest in the winners, well diversification also mitigates, when you inevitably invest in the losers. There’s both sides there. 

Yeah, there are. It’s a two sided coin for sure. When you flip that coin, you don’t really know which side it’s going to land on? Well, this is really helpful. Lastly, where do you think—if you were to shine up your crystal ball, which I know you don’t have, but if you were to have one and look forward, can you see yet what the next evolution beyond direct indexing might look like? Or what direct indexing might look like when it’s fully adopted and integrated?

Yeah, it’s hard to say exactly where the market is going. But some of the things that are really interesting to think about, and this goes back to the direct indexing definition, is thinking about it more broadly from, you know, an asset class or a collection of asset classes to holistic portfolio design. So direct indexing, its name, like extending from the idea of largely beta exposure. So moving to customizable tilts to higher returns is, I think, a really interesting concept, but still in that low turnover framework, so that you can be careful about you know, and current gains, and appropriately manage harvested losses with gains, I think it rolls into financial planning. 

So you think about, you know, basically taking time series of future cash flows and liabilities in an account, and directly incorporate that into single stocks and bonds that a client might own. So it really starts in that planning and meeting the investor more at their interface. But who better to do that than the advisor? And so that’s why we want to work very closely with advisors—I think it makes a lot of sense to work very closely with advisors, because at the end of the day, they’re the the best ones to elicit their client’s risk preferences, needs and goals, but then really meeting of that, that interface through financial planning. 

Another one is, so take a large cap portfolio or large cap asset class. There’s dividend payers and non dividend payers. What’s really interesting to think about, and this is really the optimization part, is splitting the dividend payers into a non taxable, and then the non dividend payers into a taxable, because, you know, at the end of the day, we can’t control dividends, and we’re taxed on dividends. But what’s really interesting is you can get a lot more granular by thinking about asset location and where those different components, even of one fund, if you split them, might sit in different accounts across a client’s portfolio. That’s another, I think, a really interesting level of tax efficiency, that would be pretty fascinating to explore in the future.

Yeah, that’s certainly the next level. I mean, of course, today, many advisors take asset location seriously, and they want to put the assets that make sense in IRAs, and those that make sense holding personally in those accounts. But being able to split a managed portfolio of, let’s call it large cap US stocks into their two components is an interesting possibility.

Yeah, and there’s a lot of things that I would put under, you know, just a broader umbrella of holistic solutions is some asset classes that aren’t really as accessible right now are thinking a lot about private equity. You know, can you make a platform with private equity, annuities? That’s a really important component of, I would expect a lot of clients’ portfolios, think about that in a more cost efficient, fee-only format. And I know there’s a lot of progress there as well in the industry. But putting that on platform, and thinking about that as a component of your financial plan in a personalized way.

Yeah, all of that sounds really interesting. Well, we’re gonna enthusiastically support, as we always do, the advances in finance, and this seems to be one of those advances that’s here, that is just being adopted by investors. And I appreciate very much getting your perspective. You are on the front lines of this for sure, trying to figure this out, how it’s going to work for investors, etc. So I wish you good fortune, the best of luck in doing so, because I think everybody wins if this turns out to be what we think it could be for investors, and I hope to have you back on to get an update at some point in the future to see how things have progressed.

Thanks. It was really an honor to be on the podcast and I really enjoyed the conversation and look forward to reconnecting in the future.

That sounds great. Thanks again. Thanks for joining Dave Twardowski and myself as we discussed direct indexing. It’s going to be interesting to see how this develops over the next several years, and how it’s implemented by investors and institutions alike. Hope you enjoyed the show and look forward to the next time. Thanks so much for joining me.

About Dave

Dave Twardowski is Vice President and Head of Investment Strategy at Vise, a technology-powered investment manager that drives portfolio customization and delivers personalized investment solutions. He leverages a background in engineering and financial expertise to develop innovative computational solutions.

Prior to Vise, Dave built financial science instruments as Vice President of Investment Strategies at Avantis and Vice President of Strategy Research at Dimensional Fund Advisors. He has also worked in the engineering industry as Lead Engineering Consultant for General Electric Aviation and as an adjunct staff member at the Institute for Defense Analyses. 

Dave holds a BS in Mechanical Engineering from the University of Rhode Island, a MS in Mechanical Engineering from the University of Michigan, and a PhD in Engineering Sciences from Dartmouth College.