E338: How I Invest $9 Billion into VC & Private Equity

0:00Chris, you run equity strategies at the Margaret A. Cargill Foundation. What part of the equity market keeps you up at night today? Being an investor or an allocator, you're kind of paranoid in general. So I would say everything, but to be more specific, I would say that AI risk is super topical. Obviously, aside from private credit, which I don't manage and is well publicized in the media. I think software related private equity could see an adjustment period. You know, the AI risk is real. I think CEOs of SaaS businesses should be very scared of being disrupted. The critical part here is making sure that if you're invested in private equity, that you have a very good general partner.

0:35On the other hand, OpenAI came out in fall of 2022. So it's not new. DeepSeek made their announcement in early 25. So hopefully good GPs are aware of risks to traditional SaaS. And depending on the asset, not all SaaS is going to be disrupted anytime soon. Think of like system of record, critical software, compliance software, highly sensitive data. The other thing is today there's going to be a lot of opportunities because everybody's aware of the AI risk. So I think there's going to be a lot of stuff that's coming through the pipe right now that could be quite interesting. And lastly, I would say top 10 tech companies in the U.S.

1:10today are, I think, on average, something like 35 years old. So they've gone through desktop, laptop, mobile, SaaS, cloud. And so not all these companies are going to disappear out of the blue. And on the other hand, it could be an incredible productivity boost that might help asset prices long term. There's a meta angle to this. Clayton Christensen famously wrote this book, Innovator's Dilemma, which talked about the disruption of cycles, how the next cycle kind of disrupts the incumbent. But what happened was that this generation of CEOs, Elon Musk, Mark Zuckerberg, Sergey Brin, they all have this book on their bookshelf.

1:46So they're all much more self-aware. Instead of being like the Kodak or the IBM of last generation, they're trying to actively disrupt themselves. I think that's a great point. I also wanted to mention, aside from the AI risk, I worry about a lot of what other industries worry about, like sovereign bond crisis, higher interest rate spike because of it. We have war going on. And I think the bigger one is just structurally higher inflation, which I think is with us for the rest of my career. That's a big change from the past 40 years. So absent some exogenous shock, I think we're in for structurally higher rates.

2:20I'm not saying we're going to go much higher, but I'm just saying I don't see us going back down below 2 % inflation in the next 10 to 15 years. The national debt also keeps me up at night. And thankfully, I have a lot of resources I could call up a lot of people. And it seems to be that the smart consensus view there is that the most likely scenario is not some austerity measure, but it's the increase of inflation over maybe a decade, maybe two decades, that's going to be able to allow the government to pay back its debt. What are your views on this? Yeah, I totally agree. I think David Rubinstein said something to this effect a couple years back is that he basically made the point of like, you cannot cut entitlements.

2:57You cannot raise taxes too much more. They're already very high in many states and federally. So the real way out of this is to slightly inflate. So, you know, 3%, 3.2%, 3.5 % inflation, which will significantly help over 20 year period. On the other hand, you know, you have a new Fed chairman coming in who, you know, I think generally is considered a hawk. And it'll be very interesting to see if this plays out. If I were to guess, I would say the new Fed chair is not someone who's going to tolerate three plus percentage inflation. Expert calls have always been one of the most powerful ways to build conviction.

3:32But today, investors are asked to cover more companies, move faster, and do it with leaner teams. With AlphaSense AI-led expert calls, their TGIS call service team sources experts based on your research criteria and lets the AI interviewer get to work. The magic is in the AI interviewer, purpose-built and knowledgeable-based information to conduct high-quality context-rich conversations on your behalf, acting as a trusted extension of your team. Then they take it one step further. Your call transcripts flow natively into your AlphaSense experience and become querible, searchable, and comparable, so your primary insights plug directly into earnings preps, digital work streams, and pitchbooks with zero tool switching.

4:12And with AlphaSense expert call services, the AI-led expert calls are just one option because we know the importance of a hybrid expert research approach. AI for coverage and efficiency. Humans for complexity and conviction. It's the institutional edge that scales research without scaling headcount. For hedge funds, that means validating thesis assumptions across dozens of experts before earnings instead of a handful. For private equity, it means faster pre-IOI scans and deeper commercial diligence. For investment banks and asset managers, it means pulling real operator perspectives straight into models and sector positioning without disconnected tools or manual handoffs.

4:49All of it lives inside the AlphaSense platform, trusted by 75 % of the world's top hedge funds alongside filings, broker research, news, and more than 240 ,000 expert call transcripts, turning raw conversations into comparable, auditable insight. Take advantage of AlphaSense AI-led expert calls now. The first to see wins, the rest follow. Learn more at alpha-sense.com slash how I invest. One of the unique things about AI and this tech revolution is that it's not limited to the technology space in that if you had a biotech revolution, biotech might be transformed, but it's not gonna transform SaaS companies.

5:30Talk to me about that. How do you look at the AI revolution and how it affects your entire equities portfolio? You mentioned biotech. It's already disrupting biotech. There's numerous examples of AI-led drug research, which is creating drugs much faster than they have in the past. It affects productivity everywhere. We're using it internally. I'm sure other companies and other foundations, environments, et cetera, are using it. So that's the kind of crazy impact of this, which is still unknown, right? As I mentioned, it might lead to a massive productivity boost. And similar to software, it's not a vertical, it's across every single thing, you know, it'll eventually be in old school industrials, maybe even utilities and things that are, you know, otherwise quite boring.

6:13It'll take some time. Again, as I mentioned earlier with the whole critical software type of stuff, I don't know if anyone that runs a nuclear power plant is going to switch to an AI model anytime soon. But over time, I think it can disrupt just about everything that we know in the economy. It's actually a very difficult question to answer, which is what will AI not disrupt? My first guess is plumbing or electrical work, you know, kind of behind a wall. You need a new light put in your house. I think it's gonna be a long time before they have an AI robot that's going to come into your house and do that work.

6:41Although I think at some point, maybe, maybe they will. I've seen a lot of very interesting roboticization applications that are coming. But I think that might be, you know, a decade plus off, if not longer. It's a famous Moravec's paradox, which is AI is going to struggle at things that humans could do, and humans will struggle at things that AI will do. MACP decided to put equities, public and privates, under one roof, under you. Talk to me about that decision, and how has that played out? We're a relatively young organization, and one of the benefits, and there were some negatives too, but one of the benefits of being young is that we were able to kind of whiteboard on how to organize.

7:14our CIO chose prior to my arrival, I had nothing to do with it, to organize us by risk assets. So equity, credit, and fixed income, and real assets. I was very highly attracted to the structure as it dovetailed well with a piece that I co-authored almost 15 years ago. It was entitled, Hedge Funds Are Not an Asset Class Implications for Institutional Portfolios. This basically was the concept that many hedge funds, among other concepts, but one of the concepts was many hedge funds have traditional beta embedded in their strategy. So they should reside in those risk assets, if you will. This makes for a much simpler strategic asset allocation as it can be uniform and singular.

7:54You don't have to have separate optimizations. For example, you have equity beta in perhaps three or four different asset classes. You have it in private equity, you have it in public equity, you have it in hedge fund equity, and you even have it in some credit portfolios, depending on what the strategy is. So running multiple strategic asset allocation optimizations in a vacuum, and each one of those is kind of clearly suboptimal. Moreover, equity is equity, and doing diligence on public, private, or hedged equity has significant overlap. There are differences, of course, on the periphery. Some short securities, some use leverage, some are illiquid.

8:27But in the end, it's all equity risk. I'd also note that I think a lot of other institutions haven't structured this way. And I think about why a lot, at least used to. And I think that very few institutions had large private equity portfolios in, let's say, the 1970s. So they added a private equity group when private equity became popular. Likewise, in the 90s, hedge funds became popular. So they added a hedge fund team, which totally made sense. And I think today, you'd say, well, why don't they restructure? I just think there's a lot of operational pain there. There's going to be turf warrants amongst people that run those different groups.

9:02They're going to have to go to their board and investment committee and see permission to restructure. There could be job losses. And so outside of a big catalyst internally, let's say a changeover of a CIO or maybe a sustained period of underperformance, I don't see other groups doing a reorg to somehow fit our model. And I'm sure there's opposing views to our model. You were the second hire at the foundation after the CIO. How did you go about building your investment strategy from scratch? It was clear from the start that we were going to build an endowment-like model. I say endowment-like because unlike endowments and some foundations, we had a sole donor.

9:36So we don't take in donations. We don't do fundraising like many peers in the ENF space. So in terms of risk, we're more conservative for a couple of reasons. First, we don't have fresh capital. We don't have fresh liquidity to work with or to help cushion the liquidity. Secondly, I think senior leadership here is committed to being a reliable donor to our grantees. So we take less risk in order to meet those commitments in both up and down markets. So aside from these differences, the strategy was to create kind of a lower risk endowment-like model. So we have much larger allocations to credit and fixed income and real assets than other endowment peers for sure, and some foundation peers as well.

10:11When we last chatted, you mentioned that you use both a quantitative and a qualitative mindset. Talk to me about that, and how do you fit that into the investment process? Until recently with the rise of AI, especially like super AI, the one that does cognitive thinking and has feelings, et cetera, I don't think that there's a model or a quantitative process that can pick the best managers. So I think the best allocators, and this is somewhat self-serving, you know, as I have a liberal arts degree and a MBA from a school that's known for its quantitative analysis. I'll give you an example. In venture capital, data rooms often are lacking hard data.

10:45And the one thing I always mention to staff is that we want to select general partners who lie a little bit to us rather than ones that lie a lot to us. I know that's very cynical, but I think it's hard to develop a quantitative model that could detect how accurate or how inaccurate a GP is being with us. If you work with models or if you have an optimizer, it's going to seek a solution and it might be a corner point solution. The famous example is mean variance optimization. So models love, you know, for example, in this case, a return series that are low vol and persistent excess returns. But the Madoff fraud, for instance, was exactly that.

11:19It was persistent returns and a low vol strategy. So you can't rely on a model just to pick that out because, of course, it's going to go there. That's what the model is designed to do. The other thing I'd say is like, you know, all models are false by definition. You know, I say that as a University of Chicago graduate school business graduate. You know, I'll give another example. I used to build model airplanes, you know, the plastic model airplanes like World War II airplanes when I was a child. But nobody would ever assume that you'd fly that over the Atlantic because it's a model. So models are extremely useful.

11:48Don't get me wrong. But one should always consider what is the logic of the model? how accurate is the data going into the model? Again, I mean, variance optimization relies on capital market assumptions. Capital market assumptions are frequently inaccurate. I've got ones in my drawer from 2010, 11, 12. They're all wrong. And so you have to be just highly aware that you're dealing with a model that's useful, but that is not a perfect solution. I'm like, you have an MBA from Tuck School from Dartmouth and a master in psychology from Harvard. And everyone always comments, oh, that should be really useful.

12:17But I've actually never met anyone do both degrees. I think it's just there's no dual program is my simple, my Occam's razor to why more people don't do that. But when it comes to this quantitative and qualitative mindset, if you're to distill the main takeaways from that, when you're diligencing a manager, what exactly are you looking quantitatively? What is in the spreadsheet and what are you looking that's not in the spreadsheet? The simplest things in the spreadsheet are, you know, are they producing excess returns? And what are those excess returns? You know, there's a lot of talk of alpha and then there's excess returns.

12:51I say excess returns because I know that there's alpha in there, hopefully. And there's probably some types of alternative beta or betas that make up that excess return. It could be timing. It could be sector selection, things that aren't classically considered alpha. That's something we're going to look at on a quantitative side. You know, maybe digging a little bit deeper is if it's an active equity manager, we're going to look for idiosyncratic stock selection. That's the highest. That's the purest form of alpha. And so that's got to be high. If it's a hedge fund, we're going to want to see if they're driving alpha or excess returns out of their shorts.

13:20I don't want to just pay to have them short the market if they lose money on their shorts all the time. So that's maybe some of the higher level research that we'll do. On the qualitative research, that's where it gets a lot more gray, a lot more foggy, if you will. We're looking for consistency and kind of what the strategy is in discussions with them. I'm looking for that consistency through the entire employee staff from partner down to, you know, the first level analyst. It's one of the reasons I like in in in person meetings in their offices is that you can kind of get them separated by themselves and ask kind of the same questions.

13:53Similar to what a police detective would do in a crime investigation. And so those are kind of the softer things of like hearing consistency across what they're doing. And then on the, you know, away from just the manager meetings themselves, et cetera, is, you know, referencing, particularly off reference sheet references, which you have to do a lot of work there and kind of triangulate and figure out who should I talk to who's not on this list, who probably knows this person from the past or maybe their current firm left or, you know, situations such as that. It's funny. You use this analogy of a detective.

14:27Alex Edelson, who's been on the podcast three times, he's arguably the very best person I've ever met in terms of references. And he got his training as a deposition lawyer. So he knows exactly how to frame questions, how to ask questions and references, which are simultaneously probably the most boring part of manager selection. And also probably the most alpha is gained in terms of picking managers. That would be a great background. I'm sure he's superlative at doing that. One of the things I was going to mention is I took a course, it was years ago, by a group of ex-CIA counterintelligence people.

15:01And they basically did this whole study on what people say and how they act, meaning like what do they do with their arms and they call them anchor points when they're lying. And they gave many, many examples and they showed many examples of famous people who were in either depositions or maybe just speaking with the press. And they're clearly lying. Everyone knows now because the stories have come out. And it was really enlightening to see something where you can actually begin. It's not totally foolproof. And it's not just that they say one thing that you know they're lying. But you can discern or kind of trace together dots if they're doing a lot of these different behavioral traits when they're speaking to you.

15:35It's interesting about references, which I obsess over on the podcast. I just think they're so critical, is that you could have two institutional investors sitting in on the same conversation. And they have a completely different qualitative read on the reference from each other. Two very experienced people. That's a good point. I really like the diligence with one of my colleagues. And the reason is, you know, people, you know, I hear things that maybe they hear or they don't hear, or maybe I get it backwards sometimes. So I think it's really helpful to have more than one touch point when you're doing diligence and maybe multiple touch points.

16:06I say that multiple touch points mean in private equity, I much prefer to meet the general partner prior to a fundraise, you know, a year in advance, two years in advance. You meet, you know, meet with them two, three, four times prior to the fundraise. because in the fundraise, there's at least two things that happen. One, the book is marked up in advance of the fundraise, almost always. And secondly, it's a beauty pageant at that point. Everything is framed in the best light. And if you get in there a year or two before, and I always ask what assets are doing well, and then give me at least one example of something that's not doing well, and what's the learning point from that.

16:41You get much better dialogue prior to the fundraise when they're not in kind of sales pitch mode. you're also measuring the slope where were they a year ago how have they evolved over the last year or ideally two three years you mentioned that you like to go to the manager's office and meet them in their office why do you like to do that first we want to establish that they have an office i know that sounds ridiculous but i think it's important uh secondly i want or that they're even a human yes true um secondly i want to meet with partners and employees below the partner level um you know maybe a famous example of this was during pandemic when they did virtual annual meetings and i was convinced first of all it was just the three maybe four head partners that spoke secondly i was convinced it was the script that the general council had reviewed so it was very sterile and maybe not enlightening at times.

17:37It's much easier to meet with maybe the, you know, first level analyst when you're in their office. I also start diligence, you know, at the front door. You know, what is the mood of the office when you walk in there? You can sense mood if you look around and kind of see are people, you know, having a chat and in a good mood? How are you greeted at the door? I think that's an important one. I think you can get a sense of culture from kind of the minute you walk in and certainly after you spend a few hours. I also start, I mentioned this earlier, I start with a uniform set of questions so I can get answers that are hopefully consistent and ensure that I have conversations in isolation.

18:07And I'm sorry if that sounds paranoid, but I think many investment managers are, you know, they're investment managers, Wall Street, whatever you want to call it, are motivated by most many, not most, many are motivated by fees. And performances can be an afterthought. And so we're trying to figure out, are they high quality and are they really motivated for performance, not just the management fee? In many ways, diligence is just the operationalization of paranoia, figuring out whether what you're being told is the truth. That's brilliant. I'm going to use that for future. I want to double click on going to the office.

18:43So is there one office culture that outproduces others? Is a happy office better? Or are there like chip on the shoulder offices? What's a predictive office culture that leads to success? That's a good question. I don't know if I know, but I'll say this. In places where we've made mistakes, not always, but many times, it's bad culture. Partners leave or people below the partner level leave. Turnover is always bad. Yeah, exactly. Turnover is bad, especially in a private fund. And so I do want a high-performing office that maybe has a little bit of edge to it, but I also want to see a good culture.

19:23And I could go into things like this concept of radical candor, which is a book by Kim Scott. And it talks about basically this concept that I'm going to say what I need to say to you because I care about you and I care about our performance or work as a firm. And I think that's kind of critical. So it's a bit of a balance. It's like I don't want everybody all chummy and all friends and no focus on making sure that the hard work is done and that's a high performance up and out culture. I actually quite prefer up and out cultures. But the up and out culture can also have a good culture to it where people are open, honest, communication is good.

19:59And so those are the things you're looking for. So it's a little bit of a mix, if you will. It's kind of like the seven dwarfs. They're working, but they're happy. It's kind of like a jolly, jolly working culture. So as a subset of building out the equity strategy, you had to build out your venture program. What were your first principles when you went about building a venture program? I am going to take it beyond venture. But I would say that a couple of things. When we came in, the public side was heavily indexed. So to some extent, we could just start working on privates. Plus, privates takes a lot longer to build.

20:28It takes longer. Funds aren't open all the time. They're closed-end funds by definition. So we spent most of our time in privates starting out. If I think back and think, what should I have done differently? I wished, and we tried, but I wish we had delayed and taken our time, I think for two reasons. First, I think we would have made marginally better picks by doing more comprehensive market mapping, for instance, and just taking our time. Secondly, I think we would have had more flexibility. So let's say co-investment, direct investment, specific secondaries. Keep in mind, we were building this portfolio during a very frothy period, you know, 2015 through 2019.

21:05And my colleague and I were thinking at the time, let's go slow. We're building a portfolio for an entity that will exist in perpetuity. So why do we want to rush forward in three to five years. And the reason we, I wouldn't say we rushed forward, but we moved with alacrity, is that there were some technical items in our reporting related to benchmarking, and it was painful from a performance perspective to go slow. These are technical items though. The other thing I'd stress is, that's a minor regret. The portfolio is, I think, in great shape, but it's doing well. But when I think back, I think this was an area of improvement that we could have, that may have been improved.

21:48A couple of best practices that I found on that. One is, this was originally popularized by Founders Fund. They didn't let any of their GPs make investment in the first year. So you had to see, it's impossible. You come into venture, you meet a venture capitalist. By definition, they're the 0.1 % of person you'll ever meet. You think, wow, this is awesome. And you invest in the first three funds because they're all 0.1%. You're actually right. What you don't know is that you're actually looking for the 0.01 % because everyone's 0.1%. you want to be in the top 10 % of that. So there's this principle of you have to see 100, 200, 300 managers before you really make that first investment.

22:22And you could operationalize that. The second, perhaps like a very qualitative approach to it that I've seen some foundations and family offices do is start by investing into secondaries, because they get exposure to early vintages. So you get more vintage diversification, you minimize the J curve. So you get to start to show how venture works. It's very difficult feedback cycle when you invest and you're like, okay, just trust me for 14 years. This is going to work well. So there's a behavioral aspect to that. And the best way that I found people kind of institutionalize this patience is North Dakota Land Trust.

22:57I have the CIO, Frank McHale. And before they invest, they actually find a way to capture the index of that portfolio, whether with an evergreen fund or with some other mechanism. So they're always exposed to, let's say, venture. And then they only invest if they could beat that index. So it allows... Dell PCs with Intel inside are built for the moments that matter. For the moments you plan and the ones you don't. Built for the busy days that turn into all night study sessions. The moment you're working from a cafe and realize every outlet's taken. The times you're deep in your flow and the absolute last thing you need is an auto update throwing off your momentum.