Investing Billions - E87: Lessons from Investing in 1,400+ Startups - Jamie Biddle and Steve Kim

Episode Date: August 20, 2024

Jamie Biddle (Founding Partner and CEO) & Steve Kim (Partner, Investment Strategy and Risk Management) of Verdis Investment Management, sit down with David Weisburd to discuss how Verdis has managed t...o invest into 41 unicorns by indexing early-stage venture capital funds, how they came up with the venture index thesis, and the practical considerations of executing on their venture index strategy. They also discuss the purpose of venture in a diversified portfolio, the benefits of QSBS for taxable investors, and avoiding adverse selection in venture capital. The 10X Capital Podcast is part of the Turpentine podcast network. Learn more: turpentine.co -- X / Twitter: @dweisburd (David Weisburd) -- LinkedIn: Jamie Biddle: https://www.linkedin.com/in/jamescbiddle/ Steve Kim: https://www.linkedin.com/in/stevenrkim/ Verdis Investment Management: https://www.linkedin.com/company/verdis-investment-management/ David Weisburd: https://www.linkedin.com/in/dweisburd/  -- LINKS:  Verdis Investment Management: https://www.verdisinvestment.com/ -- Questions or topics you want us to discuss on The 10X Capital Podcast? Email us at david@10xcapital.com -- TIMESTAMPS: (0:00) Episode preview (1:18) Overview of Verdis (3:18) Shifting investment focus and tax considerations in asset selection (4:39) Verdis’s approach to venture capital (5:24) Indexing early-stage venture capital (7:22) Key factors influencing venture capital returns (8:35) The role and tax benefits of venture investments in a portfolio (11:02) Performance of the venture capital index strategy (13:23) Family office dynamics (14:26) Follow & subscribe to the 10X Capital Podcast (14:47) Unicorn capture rate and venture capital power laws (16:32) Comparing generalist and sector-specific funds (18:48) Loss ratios and execution of the venture index strategy (21:10) Developing and reflecting on the venture index thesis (25:45) Key learnings from Verdis & investing as a family office (26:52) Practical constraints on over-allocating to venture (29:39) Closing remarks

Transcript
Discussion (0)
Starting point is 00:00:00 There's 41 unicorns you guys have captured. Are there power laws to that? Are the top one or top three really driving those returns? There are. What's interesting about the power law is this notion of scale and variance. So some small percentage are going to be driving the bulk of your returns. Venture capital is interesting because it's probably, from a private market standpoint, the best asset class to develop an index for.
Starting point is 00:00:22 If you think of the S&P 500, historically, it's given you about a 9% return. Venture, it's historical return. They're about twice that number. As a taxable investor with a long-term horizon, there is no better asset class than Venture. And the earlier we can be, the better, because you get that long-term compounding. As Steve noted, it's a 17.5% CAGR since 1980. There's really no other asset class that performs like that. So it's really the perfect asset class for families. Tell me about how your strategy has performed since 2017. There's been about 161 unicorns since 2017. And between our first index and our second index, we've captured about 25 forward to chat. We've been going back
Starting point is 00:01:12 and forth. Excited to have you guys on. Welcome to 10X Capital Podcast. Thank you. It's great to be here. Yeah. Thanks for inviting us. It's great to have you guys both. So Jamie, let's start with you. So tell me a little bit about the story around Veritas. Veritas is a single family office. We're based just outside of Philadelphia. And we're managing capital for generations seven, eight, nine and 10 of a branch of the DuPont family. The DuPonts came to America in 1800. They came from France and started a gunpowder manufacturing business on the banks of the Brandywine River. But over the next 150 years, the family evolved and grew it into a large global industrial conglomerate. And by the 1950s, it was the largest market cap company in the world. And we built an endowment style investment program here.
Starting point is 00:01:55 Steve and I have been working together now for 25 years and managing the family capital for the last 20. You invest not only in venture capital, but other alternative asset classes. Tell me about the family office strategy at a high level. Yeah. So we, you know, we use an endowment approach. So we're allocators of capital to fund managers. We do almost no direct investing ourselves. It's a highly diversified portfolio across most asset classes. Very, very similar to a Yale style endowment model. I think the big difference really ties to Steve's background. As I mentioned, I went to Wharton, I'm kind of a deep value cashflow investor. But Steve is a computer scientist and a systems architect and a data scientist. So Steve brings a totally different perspective to thinking about investing and really influenced our thinking
Starting point is 00:02:34 about how we invest in asset classes, using a lot of data and data analytics to think about that process. So we evolved over time, over the last 20 years. I think in the beginning, we were very focused on picking managers, trying to drive alpha. But with a family that's been around 225 years, it's very difficult to avoid decaying to the average over time. And I think that's one of the things that we've learned in the process is that it's sort of inevitable in most of these asset classes that we will, as a long-term investor, decay to the beta. So our philosophy shifted from being more beta orientation and alpha generation. And so we moved away from picking and more towards better understanding the asset classes and the shapes of the distributions of the returns of asset classes, what the tails look like, how the asset classes work together, what their correlations are, and specifically how they correlate in different time periods. So our shift has become very much more high level, but more data analytic in our thinking. So we're doing fewer asset classes than we used to. Some asset classes that were part of the endowment model either don't work for us from an after-tax basis as a taxable investor, or because we actually don't like the shape of the distributions of returns. Once we understood them better,
Starting point is 00:03:38 we decided to move to different asset classes and avoid certain ones. So the portfolio has changed over time, but still remains pretty broadly diversified. Unlike Yale, you're a taxable investor, and you said that that plays into your asset allocation strategy. Give me some examples of asset classes that you don't pursue due to tax consequences. Yeah, the one that's hot right now that we see a lot of managers looking to raise capital from us is private credit of all sort of shapes and sizes. But private credit for us just doesn't really work as an asset class. They have a higher IRR, but its multiple is quite low. And we're serving us back short-term income gains. So from a taxable perspective, it's really inefficient. And we're not allowing our capital to compound for a very long period of time. And then we're getting hit
Starting point is 00:04:17 with short-term rates. So we juxtapose that with venture capital, where we have the longest duration. Our capital is compounding at the highest CAGR for the longest period of time without paying tax. So I sort of put them at opposite ends of the tax efficiency spectrum. So, Steve, Jamie mentioned that you're a computer scientist and you helped develop the venture strategy. Tell me about the Avertis venture strategy. Yeah, I think in a nutshell, you can describe it as an index. Venture capital is interesting because it's probably, from a private market standpoint, the best asset class to develop an index for. If you think of the S&P 500, historically, it's given you about a 9% return, about 18%
Starting point is 00:04:56 standard deviation, and a sharp of about 0.5. Venture, its historical returns are about twice that number. So you're talking about like 17.5%, 18%, a standard deviation of about 30%, and a sharp of about 0.58. So it's giving you twice the return with similar risk return profile as the S&P 500. It's really a great candidate for an index, for a private market index. And our first program in 2017 is the first iteration of that. But that's our strategy in a nutshell, is we want to index early stage venture capital. Easier said than done. How do you go about indexing venture capital? So that's a really good question. The focus is on trying to find the managers that best represent that index. And we need to make sure that we are covering around
Starting point is 00:05:39 20% of the startup, the early stage startup universe. So we're looking on a vintage year basis to get exposure to about 20% of all the startups that form. That equates in a three year investment to about 1,200 startups that we're investing in. And our cycles are about three years. We built our portfolios to invest in a three year period. And we're looking to find the managers that best represent that venture ecosystem and make sure we have that kind of coverage. Let's say that your end goal is to generate beta. Some would argue that if you're looking for average investments, you're going to actually get bottom 50% instead of just hitting on the 50% that you actually have to look for alpha. How do you balance the need to make sure that you actually get the beta versus being adversely selected? If you think about venture
Starting point is 00:06:22 capital, generally speaking, most of the investments aren't very good. And you can see that in the median return over time, the median return is somewhere around 6%. So when you think about it, most venture investments aren't very attractive. So only about 1% or 2% are really driving those returns. So in order to get adverse selection in the asset class, you need to predetermine what those 1% to 2% are. And we think that's just as difficult as finding those 1% to 2%. So it's a great asset class from the standpoint that you need to get a lot of diversified exposure to make sure that you have a reasonable probability getting into that 1% to 2%. But we're not worried about adverse selection because like I said, most of the investments in venture are not very attractive. So it's really about getting into the attractive
Starting point is 00:07:03 investments and you have to pick those out ahead of time to be adversely selected into them. So it's as hard to pick as it is to pick against them. Exactly. It's just as difficult. But you do have a tilt. You've really dissected the numbers and you don't just blindly invest into venture funds. You have a geographical tilt. You have a strategy. Tell me about that. We spend a lot of time in the office thinking about material effects. And when we think about being data-driven, what we're looking at is what is the data actually telling us that is material
Starting point is 00:07:31 to our return outcomes? And a lot of time, a lot of investors that are data-driven are really focusing on factors that don't have a lot of material impact. And when we looked at venture capital, early stage venture capital, there were only two real factors that had material impacts on returns. One is stage, which we've talked about. We go as early as we can, which for us means pre-seed, seed, or early A. And the other one is geography. Geography has a very, very strong material impact on returns. If you look at the amount of outliers that come out of California, for example, it's 76% of them. And the expected value of an outlier in California is two to four times what it is in the rest of the country. So if you add, let's say 14% is New York and 76% is California, you're talking about the bulk of all the outliers are coming out of those two
Starting point is 00:08:22 geographies. So that's a huge material impact. So that's why we tilt our portfolio and stage is the other one. Stage has a material impact on returns. So that's the other factor we look at. And that's why we're very, very stage focused as well. Virtus has this contrarian view on the purpose of venture in a portfolio. Tell me about why your family invests in venture. What is the purpose of venture in your portfolio? Our view isn't necessarily contrarian in its role in the portfolio. It's probably more contrarian in how we implement it. But as a taxable investor with a long-term horizon, there is no better asset class than venture.
Starting point is 00:08:53 And the earlier we can be, the better because you get that long-term compounding. As Steve noted, it's a 17.5% CAGR since 1980. There's really no other asset class that performs like that. So we can lock as much of our capital up in an asset class that performs like that. So we can lock as much of our capital up in an asset class that's compounding at that rate, and we can defer our taxes for as long as possible, which we can in that asset class. But it's really the perfect asset class for families. The biggest challenge, of course, is its illiquidity. So we have to manage that tightly and carefully to make sure that we can meet the needs of the family and
Starting point is 00:09:22 have as much allocation to venture as we can. I think what's maybe a little bit different is Steve's approach and how we got around to indexing the asset class from a perspective probably originally as being pickers and thinking that we really had to be in the top 10 names in order to be in the asset class. And I think our perspective has dramatically changed over the last 20 years. QSBS, which is this tax treatment of early stage investments, which sometimes accounts for 100% of both federal and state, depending on the state, a tax break. How much does QSBS play into your strategy? And do you actively manage that? Yeah, it's really important. You know, as an allocator to fund managers, we don't actually control the actual election itself, right? So it's up to our managers to make that selection. What we do is we incorporate it in a side letter with most of our managers to request that they opt in when they can, because it's incredibly powerful to be able to take advantage of that.
Starting point is 00:10:11 So we're looking at a portfolio that for us is diversified across thousands and thousands of companies, right? So if you have preferential tax treatment across all of those startups and it rolls up to us, it's tremendously powerful. For large family offices that are looking to invest in venture and want to implement it at the scale that you're looking at, how does that actually, from an administrative standpoint, how does that get implemented? Do your fund managers send you back tax forms with QSBS designations? Yeah, that's exactly how it works, right? So our finance team is managing that and we're requesting that information. But we're broadly diversified across 30 plus managers. So Steve and his team are doing almost a dozen managers a year.
Starting point is 00:10:47 So the pace of allocation, I think we're probably the highest volume early stage venture investor in terms of number of managers. So it's a big operation. So I wanna get back from theory to practice. So you guys had a great theory started in 2017, but now you should have enough data
Starting point is 00:11:02 to start to really test the validity of your thesis. Tell me about how your strategy has performed since 2017. The strategy's performed really, really well as far as how we're correlated to the benchmark. So if you look at our 2017 results, it's roughly 1400 startups, 20 plus managers driving those startups. We're 98% correlated to the Burgess benchmark. So that's the benchmark we use to gauge what the market returns are. And Burgess is a good example of that in the sense that they use cash flows from LP, so it doesn't have any survivorship bias. So it's a strong benchmark. It goes way back to the late 70s. And so it's also% correlated. There's been about 161 unicorns since 2017. And between our first index and our second index, we've captured about 25% of those, or we have exposure to 25% of all of them.
Starting point is 00:11:52 And that's at the first check. So that's a pre-seed, seed, and early A. About half of those unicorns come from emerging managers, actual fund ones. So that's an interesting statistic as well. So we're running about around a 6% tracking error to the index. So it's interesting statistic as well. So we're running about around a 6% tracking error to the index. So it's right on the benchmark. I think I have an idea how to track S&P benchmarks, but how do you actually track technically yourself to Burgess index? Yeah. So we're subscribers to Burgess. They provide all the benchmarking information on the private
Starting point is 00:12:19 market side. So it's not just venture capital, but they also supply benchmarks to buy out private real estate, oil and gas, infrastructure, growth equity, different flavors of venture, whether it's not just venture capital, but they also supply benchmarks to buy out private real estate, oil and gas, infrastructure, growth equity, different flavors of venture, whether it's early or late stage venture as well. They get LP cash flows from underlying portfolios, and then they create that benchmark. The benchmark is released on a quarterly basis. So it gives you both time-weighted returns, TBPI, IRR, DPI, all the statistics you could possibly want across the private markets. And we use that on a quarterly basis to see how we're doing. Includes DPI, even though it's still early, you're able to somehow benchmark the DPI as well. We are. And if you look at the data, you can see that DPI is very, very correlated to TVPI
Starting point is 00:12:58 versus IRR. So IRR is a pretty noisy statistic, but TVPI is less noisy and it's really pretty indicative of what you're going to eventually get from a DPI perspective. So you can see over time that funds will correlate into their TVPI number. It's usually for early stage venture, it's probably out past year eight, closer to year 10. But as you approach year 10 and get to year 11 and 12, the DPIs and TVPIs converge pretty tightly. You have hundreds of family members that essentially invest within this vehicle. How do you balance the needs of those families? And how do you deal with different opinions for somebody that's managing such a large single family office? Well, thankfully, we're only managing one branch of the family. So
Starting point is 00:13:38 there are literally thousands of DuPonts. We're not responsible for all of them. Otherwise, it would really be complicated. But we also believe really strongly that you need to hire the very best people that you can and let them do their job. And so we have an investment committee. But other than myself, there's no family member that sits on it. It's all the professional staff and they make the investment decisions. And I think that allows us to recruit and retain the best people because the family isn't meddling in that decision making process. We're letting the professionals make the decisions. And you're part of the DuPont family? I am.
Starting point is 00:14:06 I'm a member of Generation 8. I was tasked with managing the family's capital. And so we decided to go down this approach to create a family office in this way about 20 years ago. It's been a good process. I think it's worked very well. A really important component is transparency and communication. So we do a lot of that with the family.
Starting point is 00:14:22 And I think the more we communicate, the more transparent we are, the more comfortable everybody is. Congratulations, 10X Capital podcast listeners. We have officially cracked the top 10 rankings in the United States for investing. Please help this podcast continue climbing up in the rankings by clicking the follow button above. This helps our podcast rank higher, which brings more revenue to the show and helps us bring in the very highest quality guests and to produce the very highest quality content. Thank you for your support. So Steve, you mentioned that you guys have been in 25% of the unicorns. First of all, tell me about the total universe of unicorns, total universe of investments. Tell me about a lay of the land. Since 2017, there's been
Starting point is 00:14:58 161. So there's been a lot out there. And I think the key to early stage returns is getting into those unicorns early. Again, our goal is to compound capital and the multiple is very, very important to us. So we want to get access to those unicorns as early as possible when the valuations are the lowest. And we want to grow with those companies for as long as we can to harvest those multiples later when those companies become, quote, outliers or winners. So that's the goal. We've been fortunate enough that, you know, our diversified portfolio has allowed us to capture, like I said, about 25% of those.
Starting point is 00:15:31 We're holding onto those. The 2017 index is still pretty early, so we still have some time left, but it's growing nicely. Those companies are growing nicely and we're optimistic that we'll meet our return objectives. I think the long-term capture rate is about 1% to 2%, Steve, if I'm about right. Yeah, it's about 1% to 2%. Yeah.
Starting point is 00:15:49 Yeah, we're capturing around 3%. So I don't think we entirely know why. We're capturing a little bit higher. It could well be the tilt that Steve has put into the portfolio around geography and other things, but we're capturing a little bit higher rate than we had anticipated. And if I'm doing my math correctly, there's 41 unicorns you guys have captured. Are there power laws to that? Are the top one or top three really driving those returns? There are. What's interesting about the power law is this notion of scale and variance. So you're going to get that at any scale. So whether it's unicorns or decacorns, you're going to see a power law there. So some small percentage are going to be driving the bulk of your returns, even at whatever scale you decide to use. So yes, that answer is that you have a
Starting point is 00:16:29 handful of companies that are driving those returns. Is crypto not the perfect market for power lines for the asymmetry and how much of your portfolio is in crypto? We don't do sector funds. So we're generalists. So if you think about our crypto exposure, it's relatively small because we don't invest in crypto specific funds. I think crypto is a great example of the power law, but AI is a great example of the power law. Could be space is a great example of the power law. There's certainly biotech is a great example of the power law. Innovation in general is power law driven. And what's interesting is when you look across asset classes, I think asset classes, if you're a long-term investor and you're trying to compound capital, it's geometric.
Starting point is 00:17:06 So investing, I think, in general is power law driven. Now you can quibble with, is it heavy right skewed and not really technically a power law? You can certainly have those discussions. But anything that is multiplying capital over an extended period of time with reasonable volatility is going to develop really, really long right tails. And that's just the nature of investing. Innovation and venture capital, because the volatility is so high and the outcomes can be so massive, I think is just a really souped up example of that.
Starting point is 00:17:36 And why are you guys leaning in on generalist funds? Why not do sector funds as well? The easy answer for that is we want to front run sectors. So if you look at most of the sectors that have been developed and you look at the handful of companies that are really the winners in those sectors, they've been discovered by generalists. So what happens is a generalist discovers Notion or discovers Coinbase or you name the sector. And then once that sector becomes hot, then a bunch of sector funds break into that sector and make the sector hot. What we want to do is front run those sectors before they become hot. And it's quite easy to see if you take the top three companies out of sectors, those sectors are no longer really attractive. They kind of collapse as a sector. We want to be in those three or four companies that make sectors hot sectors before they become hot sectors. So that means that we want to invest in generalists. And the other part is that it's really, really hard to build a hyper-diversified portfolio picking sectors.
Starting point is 00:18:29 You want to be in those companies before the sector is even named. Exactly. And that's why part of, you know, we want that to happen for sectors, but that's why we go so early in venture too. And we want to get into those companies before they become the hot companies, right? So we're trying to front run all of it. Do you worry at all about loss ratios or are you just completely like, you know, give me the power law? We're pretty power law. So we worry about loss ratios in other asset classes.
Starting point is 00:18:52 Venture capital isn't one of them. So, and that, I think Jamie's point there is kind of how we think about it is we look at the distributions of these asset classes and we understand what they look like, what the averages look like and the other parameters that drive these distributions and we optimize our strategy based on that. And if you look at venture, it's very power law driven, which means that loss ratios don't really matter.
Starting point is 00:19:12 Jamie, you've been able to see Steve execute the strategy from the CIO position. What has been most surprising about the actual execution of this beta venture strategy? That it worked. As advertised? Or what did you expect and what did you not expect beta venture strategy? That it worked. As advertised or what did you expect and what did you not expect about the strategy? So I'm slightly kidding, Steve, but, you know, I'm a deep value cashflow guy, right? So my whole career has been about picking and, you know, I started out as a venture capitalist and, you know, my orientation was around, look, you wanted to be operationally focused. You needed people who had access to networks and capabilities to improve businesses and drive their growth.
Starting point is 00:19:48 This notion that the outcome is random was very hard for me to get my head around. And it's not how we think about any other asset class. We do honestly believe that some skill matters, although that's a constant debate internally. What is skill? What is luck? Can we determine skill? How long does it take to determine skill? But, you know, in venture, the idea that it looks random in the data and that we should accept the randomness and invest accordingly, you know, was a big hurdle for me to get over. And so, you know, we sent Steve sort of back to crunch the numbers over and over again, right? Kept trying to come up with questions, you know, that would poke a hole in it. Like, all right, what have I not thought of to ask Steve to go back and look at again?
Starting point is 00:20:25 He probably ran, sliced and diced the numbers and the data over and over again, as many permutations as we could come up with. And the data is what the data is. It was pretty compelling that this looks random and that your capture rate is about 1% to 2%, an early stage venture. And as we were kind of going along this journey, a few other family offices we know became kind of interested in what we were doing, wanted to see if they could participate with us. So we ended up growing the network. We're now up to 16 families who are working together to implement this approach. So it's now a significant amount of capital with a lot of families on a global basis, all pursuing this idea that, you know, Steve came up with a decade ago. So we had our own journey to get our heads around this.
Starting point is 00:21:02 It's not easy to accept. The more picking oriented you are, which I confess is my orientation, the harder it is, you know, to get your head around this. Steve, how did you originally come up with this thesis and what methodologies did you use? Yeah, so we actually started on the public market side. There's a lot more data available when we started on the public market side. So we could start thinking about what the distributions look like, what those shapes look like. And a lot of public market data that we looked at, we could go way beyond thinking about just the mean and standard deviation. We could look at the size of the tails and whether there's a skew. So that got us oriented towards what we should look in the data and what that
Starting point is 00:21:39 means for the asset class. And then recently, probably about the time when we got started, there was just a lot more private market data available. And it recently, probably about the time when we got started, there was just a lot more private market data available. And it's exponentially grown. So we were able to apply that same lens that we use on the public market side to the private market side. We could actually look at quarterly returns, right? We could look at returns across a very, very large data set. A good way to think about that is we're very distribution driven, which when you think about distributions and the long-term impact of distributions, it's an amalgamation of all those factors. That distribution kind of gives
Starting point is 00:22:11 you a sense of the interplay of all those factors to drive asset class returns instead of isolating yourself to one single factor. So it gives us tremendous information on what the returns look like, what the strategy should be. So that's kind of our approach when we say that we're data-driven versus a lot of other data-driven VCs that are very much focused on certain factors. It is interesting to think about the persistence of the strategy. There's several factors there. One is not everybody has access to the same data. Even if they do, the minimums to get into these funds are so high oftentimes that you need tens, if not hundreds
Starting point is 00:22:45 of millions of dollars to execute on a broad diversified strategy. You also need to have the relationship capital. You need the reputation to access it. You can't just go into your website, go into your fidelity and buy those. And then perhaps most importantly is you need the patience. You need a 12 to 14 year outlook, especially in year four or five, where you've gotten $0 in and you've kind of kept on deploying, right? So there's a psychological, informational, as well as relationship kind of barriers to replicating the strategy. I'd add operationally too, and give a shout out to our CFO, Kevin Gaffey. Kevin came to us from Princeton's endowment about 20 years ago. Kevin and his team are processing 2,000 capital calls and distributions a year. So the amount of back office activity that's required to implement the strategy is enormous. And it's why it made sense
Starting point is 00:23:28 for this syndicate of families to work together, because it just didn't make sense for everybody to try to replicate the strategy uniquely. What do you guys wish that you knew before you started this program in 2017? For us, one of the things we try not to do is like kind of try to predict things, predict outcomes. I think that's, we spend a lot of time trying not to do that. I think the advent of data and what it allowed us to do and the power of it, we probably didn't quite understand at the time. Like, you know, the question that was really at the forefront was, okay, private market data is starting to become available.
Starting point is 00:24:00 Is that valuable? Can you really make decisions based on that data in the private markets? It's way different than public markets. Like you said, the time horizons are longer. Liquidity is still an issue. Certainly in the private markets, access is different. So can you actually leverage that data that you're getting on the private markets? I think at the time we started our first index, I think that was a big question mark. You know, I think it goes back to something I touched on earlier, just the orientation, someone who's been trained to select investments, this idea of looking for better data took us a while to get there and the data took us there. So I wish
Starting point is 00:24:34 that we had stopped trying to be better than average, you know, 20 years ago and instead spent as much time as we do today looking for better averages. But that's kind of how I would assess our approach today is we're much more focused on finding the better averages. It's a lot harder than you would think because we have so many built in biases around the way we've been trained and how we've invested for years. So to move away from asset classes that have been sort of ingrained in our psyche as allocators that we should be in because the data is just telling you it's not behaving the way you want it to. It's not just looking at means and medians and even the tails, but looking at the correlations and most importantly, how these asset classes and when they're correlated with each other. A great example is commercial real estate. It turns out that's not adding a lot of diversification benefit to our portfolio at all.
Starting point is 00:25:16 And in fact, it's averaging down our average over time. So it's caused us to move away from commercial real estate allocation. That was a long process to kind of mentally get there. So I think that's been one of the biggest challenges and one of the biggest learnings is let the data take you there and be willing, as Steve says, hold your opinions loosely. It's hard to unlearn things. It certainly is. And it's really hard to hold your opinions loosely. We say that a lot, but it's hard in practice. That's for sure. What would you like our audience to know about you, about Virtus and anything else you'd like to shine a light on? One of the things that we try to do is we try to be transparent and share our data and information and let other people decide whether it's valuable or not.
Starting point is 00:25:53 But I think that's one of the things that we try to do is try to give back as much as we can, share as much of our analysis and as much of our data and as much of our learnings as we can. You know, single family offices are really taxable investors need to think differently than other institutional investors. What works for Yale doesn't really work necessarily for everybody. And you need to be focused on after tax returns and liquidity. There are things you can take advantage of. You know, our distribution payout rate to the family is much lower than Yale's distribution rate or a pension.
Starting point is 00:26:23 You know, we have a strong network in the pension community, and those guys have to pay out a much higher percentage of assets every year. So there are some benefits to being taxable and being families. You have a tremendous amount of flexibility about what you can do. But there are some constraints, particularly taxes and liquidity, depending on the needs of the family. So I think it's important to recognize you're a little bit different. And there's a huge network of other families and investors out there. So we've really enjoyed getting to know other families around the world and working with them and sharing ideas. We're big on transparency and sharing. We learn as much from others as we hope we're giving back. Given your low payout, why not put 50, 60, 70% in the strategy? Why not go all
Starting point is 00:26:57 in? What are the constraints on over-allocating in this house class? Yeah, it goes back to the comment about taxes. So the largest beneficiary of our distributions is the federal government. So we pay more in taxes than the family gets paid in dividends every year. So Uncle Sam is our biggest beneficiary. So we have to manage the cost of the distribution of the family and then the taxes, plus the operations of the office. So we have a fixed payout ratio. Now, it's low, which allows us to really take as much illiquidity as we can. But it's not zero. You mentioned PrintCo. Andy Golden, CIO of PrintCo, told me that DPI is the most overrated thing in the startup
Starting point is 00:27:33 community, in the venture community. What are your thoughts on that? I think you hit the nail on the head. If you're a venture investor, you need to really be patient because the asset class works because you're compounding the capital in high growth companies for a very, very long period of time. That's really why the asset class is attractive. And that's really the only reason you should invest in the asset class, to be honest. And if you are very, very focused on getting cash flow back early, it really neutralizes those
Starting point is 00:28:00 benefits. And there's a lot of asset classes that allow you to get your capital back early. We just talked about one in private credit is one. Even buyout has a shorter duration than venture capital. So that's another asset class that gives you cash flow back earlier. Private real estate is another one. So you don't have to invest in all the private asset classes. Just pick and choose the ones that make sense for you and the strategy that makes sense for you. So if you want early capital back, you want capital back early, especially early stage ventures, probably not the asset class. So I agree with that statement. I think the benefit of the asset class is the long lockup, is the compounding capital for a long period of time in high growth companies.
Starting point is 00:28:37 And if that's not for you, that's OK. And you've modeled thousands of funds in terms of your data. A lot of people are very skeptical of TVPI and some don't even think it's a number that should be considered. How accurate is TVPI, a predictor of DPI in your research? One way I would look at that is let's say that a particular GP is wrong in their mark by 10%. That's really in the noise. You're holding that asset for 12 years. It's compounding at 30% if you're lucky a year. These factors it on whether the mark is perfectly accurate or not.
Starting point is 00:29:11 Nevermind that they're being marked on a quarterly basis and there's a lot of movement there in the first place, right? But even if you're off by some percentage, it's just, it's completely in the noise. I mean, if you're going to worry about your venture returns, that's not what you should be worried about. You should be worried about getting the long-term average of the asset class. That's what you should be worrying about because whether a manager is marked at 5% below or above what it should be, it's going to get worked out when it finally exits. And that exit multiple is going to dwarf whatever that was at that one point in time. Absolutely. Well, this has been really illuminating.
Starting point is 00:29:42 I've learned a lot. Thanks for jumping on the podcast and I look forward to seeing you down soon. Great. Thanks so much. Yeah. Thank you for having us. Thanks for listening to the audio version of this podcast. Come on over to 10X Cabal Podcast on YouTube by typing in 10X Cabal Podcast into youtube.com and clicking the subscribe button. On the YouTube version of this podcast, you could see the graphs, visuals, and key takeaways that accompany every episode.

There aren't comments yet for this episode. Click on any sentence in the transcript to leave a comment.