The Exchanges

Q So when you take a new paradigm like that and still layer it into somewhat of a traditional portfolio without it, how did you think about how to size your capital allocation to that strategy within the context of your portfolio?

A There's two bets here, right? First one is on the manager self and I've known Julia for 20 years and worked together. And the fact that we still work together is, is a testament to how her integrity and her ability to really be what she is, which is a, one of the best minds in investing that I've ever run across. And then the second piece is then we're going to see if we save some capital to seed products. So we started with sort of a five million dollar investment and it grew from there. So we now have Four strategies with Rosetta, and three of them have been very, very successful. One of them has been turned off.

AI assessment note: “we started with sort of a five million dollar investment and it grew from there”

Q How did you start thinking about that challenge of seven hundred million dollars? You can In theory, do anything with it.

A Yeah, I think they were long and loud when I got there, and they had a CIO who was very well-known, Lou Morel, sort of a great investor on the sector, long analysis, did a lot of work on mutual funds, and was basically buying and selling ETFs pretty effectively. But when he was wrong, he was really wrong. When he was right, he was really right. So the challenge is I came back and looked at the portfolio and said, where are the issues here? Let's start with sort of governance and the big picture. What does Wake need? Yeah. And that was the starting point. So I sat down with the, the trustees who, who hired me and said, okay, what are you willing to lose? That's the first question. And that was sort of interesting. They hadn't really thought about it that way. And looking at Wake Forest again, they had lost two, 28% in 2008 before I joined. And I looked at that portfolio and said, well, what did that mean? Like 20%, that's not too bad considering all the other schools that lost 29, 30, 31. You're actually in good company. What did that mean to Wake? Well, that meant that 10 years of fundraising got wiped out in four months. That meant that, ah, 14 kids couldn't come back to college. So what, what can you lose? What are you willing to lose? And from a standpoint of the university, you've got to think about fundraising, you've got to think about the endowment, you've got to think a…

AI assessment note: “Let's start with sort of governance and the big picture. What does Wake need?”

Q What are the issues involved with managing some of the money internally? It sounds like that's mostly on the kind of the hedging program of what you do.

A Yeah, I think it's a unique part. So when I got to Wake Forest in 2009, we had this sort of long beta portfolio and I went and said, okay, we want to hedge. So I call, I had just left Wilshire and I had negotiated every new ISDA because you had to. So I went to SocGen and Goldman Sachs and Morgan Stanley and said, look, I want a new ISDA. I just negotiated for four weeks with Wilshire. I want the same terms. And they laughed at me. You're not wheelchair anymore. You're, you're a hospital. You have a university with tuition. You're not gonna get the same terms. We'll, we'll let you post collateral every day. We'll post once a year, and we'll do it in New York and do it in Greensboro. That's not gonna work. So, you know, we went back and we went to one of our partners and said, how do we do this? And a manager who we trusted said, here's how we do it. And we asked, can we help? Can we, can we use your balance sheet? Can we use your trading desk? Can we use your collateral to manage this portfolio? And they said, Sure. Why not? Now that's become a big business for them. They've got hundreds of billions of dollars. We were the first. So I think that was one of the ideas we went and said, how can we leverage these relationships we have with trustees, with, with managers. And I think that was a big driver for us is we couldn't build it internally. We don't have the ability.

AI assessment note: “we couldn't build it internally. We don't have the ability.”

Q a large group of managers, it's actually hard to get concentration and best ideas. So how have you gone about thinking about what's the appropriate size and in the instances where you're kind of overly diversified because you are in all these different factors and you know, 55, 65 different managers. Do you look at co-invest? Do you try to find ways of getting at larger sizes of best ideas?

A Yeah, I think we, we have done co-invest, but it's hard for us. So we're not doing co-invest in equity positions. We have one, but more likely we'll co-invest in a CLO or a CDO, and we'll do it alongside the manager. We've also done co-invest in things like agriculture. So we try to find things that for us, we loved distressed investors, not distressed assets. So we try to find investors who have any issues, and we can be that capital. So agriculture's a good one. We haven't owned this almond farm in Australia. 25 year investment. Eight percent yield. We got a little bump when we bought at a discount. But who can own a 25 year investment? Very few investors. We have the benefit of time. We can own anything.

AI assessment note: “we have done co-invest, but it's hard for us.”

Q What was your favorite sports moment? Could be as a participant or a fan.

A Well, I think, um, That's a hard one, too. I think my, I would like to say my boys' sporting events is what I really would, I really enjoy going to their events, and they both play football, one at Williams College and one at the Christ School in Asheville, and I spend pretty much every weekend in the fall at their games, because I can, and I think, you know, I, I did play sports at Villanova, and you know, what I learned from playing, being a goalkeeper at Villanova was that when you win, you have nothing to do with it as a goalkeeper. When you lose, it's all your fault, So, I, I really can't point to those, but I think I would say the most, the event that I went to recently was the Villanova National Championship. It was kind of a lark last minute, but we had an absolute blast, and we saw, I thought we saw the best shot in college basketball with four seconds left, and then four seconds later, I saw the best shot in college basketball, but I think the funniest part of that story was that, um, you know, we, we partied way into the late night, and I, I got on the first, first thing smoking next morning, And I got upgraded because I travel a lot, and I sat in, and I was, I ran to the plane, sat in my seat, first on the plane, Villanova gear head to toe, and I'm going back to Charlotte. And I really hadn't thought much about it, and I'm sitting there, I'm still, you know, basking…

AI assessment note: “the event that I went to recently was the Villanova National Championship.”

Q Yeah. How often did these opportunities come up to these, Sort of co-invest or secondary opportunities come up for you?

A We get a lot of opportunities. We probably only do one a year. So we get quite a few. Some of them, the reality is we're humble enough to know that if we get the call, we're not the first call. So we have to really be, we know the manager, understand where we fall in the cap structure, understand where we fall in the call structure. If we get a call from a broker on this fund or a secondary or something like that, we know we're not the first call. I remember Bruce Zimmerman called me and said, I'm selling this part of my portfolio. You, Timko, are you interested? I said, who'd you call first? And I said, why are you selling it? Why do I want to own something you're selling, Bruce? And he laughed. He said, you know, you're right. So I think that's the issue for us is that, you know, we're, we know where we fit in the hierarchy and we have to be very careful who we'll do deals with. So I think from the standpoint of, you know, I think in all of these deals for us, it's transactional, right? So when we get transactional, we get worried. You know, I think when we look at the street and they want to do a deal with us, the first time we question, the first question we ask is how are we going to get screwed?

AI assessment note: “We get a lot of opportunities. We probably only do one a year.”

Q I will take you up on that. As long as you have the license. What, uh, what, do you have a favorite book or recent book that you've read that you've enjoyed?

A You know, in my bag right now I have, uh, The Undoing Project by Michael Lewis, which I really enjoyed. And it's become kind of a, I pull it out, but it's all, um, earmarked and lined in. And I find it's pretty interesting from an investment perspective as well about how they, how they look at the world. But I think one of the books that I read recently was JD Vance's Hillbilly Elgee. And, uh, you know, being a lower middle-class kid, my parents weren't addicted to drugs, and we didn't live in Appalachia, but, you know, having grandparents who were immigrants, and parents didn't go to college, and living in, you know, in Philadelphia, you know, it's a different upbringing. I found a lot of analogy to what he was doing, but a lot of it had to do with just hard work, the value of hard work, and the value of family. And I really enjoyed that book. You know, I think people who read it were depressed about it, and more about the Current political environment and how, you know, Trump won the Appalachian vote. But I also think there's a lot to be said about kind of hard work and, and that value of not feeling sorry for yourself and not blaming the government. And I think that comes through in that book.

AI assessment note: “in my bag right now I have, uh, The Undoing Project by Michael Lewis”

Q to drive it, which is the manager selection process. So Let's just start with, how do you think about what you're looking for in kind of the kind of manager in the portfolio? I mean, you started with a group of managers, so does it start with, hey, our model's telling us we like value now. Let's do a hunt for where we think value's most attractive in the world.

A Yeah, I think it's a little more for us, a little bit different. I think, you know, our job is to basically separate talent from luck. So, and we want to find managers basically we can do more than one thing with. So you find us more with multi-strat managers, or managers that have multiple products. Because if we're going to do all this work we do, and background checks, and all the stuff we have to do, for a twenty-five million dollar biotech long-only manager, it's not a real good use of our time. We have 13 people, and it's really just not, it's too niche-y for us. And it only gives us one exposure in the model. So, we find ourselves really trying to focus on, trying to find managers that can do more than one thing for us. And it could be a separate account, it could be an overlay, it could be other things, other products, but you find that with only 60 relationships, 65 relationships, we have managers that do a lot of different things for us. Another thing we try to do is also try to find best idea accounts, so we have a couple separate accounts where we just give them a broad mandate and say, give us your best ideas. We're very fortunate, you know, we've got a West Coast manager who I really like, Scott Minard at Guggenheim, who I think is a bon savant. And we basically let him go and do what he wants to do. So I think that's one of the things we try to focus on is separa…

AI assessment note: “we want to find managers basically we can do more than one thing with”

Q As you went from Wake to Verger, your outsourced CIO, could you talk a little bit about what worked, what surprised you positively, what surprised you negatively?

A Yeah, I think it's like chewing glass, Dan. I'm starting to like chase my own blood. Um, but you know, it, it's really an entrepreneurial opportunity, but within a university, you still have minutes, and you still have boards, and you still have constituents that want to do it the Roberts Rule way, and, and that was surprising to me. You know, I kind of want to set off and do this entrepreneurial and, and do it the way we wanted to do it, and that was not the case, which is totally understandable, but it was difficult to get through that. The other thing I didn't really understand was sort of the, the, the need to go to every constituent. So every board member had to understand why we're doing this and, and buy into the model as well as the reason to do this. Then you go down and, okay, everyone in accounting, do they understand why you're doing this? Everyone in HR, do they understand why you're doing this? You know, that, that piece of it was really important. And then the other piece of it can only was how do you name it? So demon deacon management was the first name. And we, we, I went around to all these other outsourcers, Chris Bittman at Perella and Alice Handy. And Spider, and so what did you guys do wrong? What would you do differently? And one of them said, just don't name it Wake Forest Asset Management. And I went to NC State, Libby George, who had money with UNC. A…

AI assessment note: “you still have constituents that want to do it the Roberts Rule way, and, and that was surprising”

Q a demo at ridgeline.ai. And now, back to the show. So that's a very different frame for the reasons you discussed than the Yale model or the typical endowment model. How did that translate into action? So you, you say we understand risk. There's a certain amount of risk tolerance that we have at Wake. What do you then do in terms of framing the structure of the investment pool?

A So Ted's a great question. So two things we did first. First thing we did was I asked the committee when I started, how much do you pay in fees? And nobody knew. So I went and looked at every manager and said, here's what we pay in fees on each manager. And, you know, some of the numbers were very big, but some of those managers were doing really, really well. But incentive fees were big. I mean, I have no problem paying incentive fees. I mean, they're doing a good job. So we looked at that and said, okay, who are we paying money to? And that's the first thing, that fee part of it. The second part of it is where's, where are the returns coming from within these managers? So we sort of did what we call reverse optimization, ran all, every manager through this factor analysis and said, okay, what's driving their return? Is it momentum? Is it growth? Is it value? Is it spread? Illiquidity? Duration? And you find interesting things. Your real estate managers. Spread and duration. You wouldn't think that. That's a fixed income concept, but spread and duration. Cap rates and, you know, how long the rents are. We like student housing, because you can change it every year. You know, do you want a long-term lease based on cap rates of four? Probably not. So I think there's, there was a view on us that we wanted to focus on What were the drivers of return in the portfolio? And let's try …

AI assessment note: “two things we did first. First thing we did was I asked the committee”

Q you lose, and you had the frame of 2008, that doesn't incorporate time, right? That's one year, 28% drawdown. Now it turned out if you rolled that forward to 2009, you didn't know it would happen at the time, but you get a lot of that back. So how did you conclude that portion of that conversation of, of what was Wake willing to lose and over how much time?

A Well, you don't look at 2008 in, in context, and the endowment's not a series of one-year returns, as you know well. You go back and look at, okay, 2008, and 2098, and 94. So what did you do in all these periods where you had these drawdowns? And if you're down 20% every five years, here's the math. Because you're spending four dollars for one dollar you bring in. You're not Yale. You're not getting a hundred million dollar gifts. You're getting a million dollar gifts. So you're spending fifty million dollars every year, And you're bringing in 14. It's a great number, but it's not enough if you have five, 20% of the clients every five years. So you do the Monte Carlo simulation. You say, okay, just so you know, in 2150, that year you will be zero. That's the math. That's the easy math. And that's okay. That will not be on your, your, your, this committee will not have that problem. But somebody will. So how do you protect the purchasing power of this endowment if you keep losing 20% every five years? And that was the key. And I think, you know, the, the running endowment's very contextual. It depends on the nature of the institution. It depends on the history, its ambitions, its financial resources. Wake is not Yale. It's not Harvard. It's not Georgetown. It's not Villanova. So they have to have their own mentality, their own, their own work. And I think, you know, that the big…

AI assessment note: “So you do the Monte Carlo simulation. You say, okay, just so you know”

Q Yeah. So this is sort of an interesting one, right? So neural network strategy. Why don't we just start with how do you think about how that fits into your portfolio?

A Well, you know, it's interesting. Modern portfolio theory is not very modern, right? So we sit around and think about what's the next thing that we should be focused on and learning about it and trying to do some more work on it. And I think it's the ideation around This concept came from thinking about what was going on in the world, thinking about computer power, thinking about what we're seeing our venture capital managers investing in these great companies that were going parabolic, looking at things like SAS networks and all of these incredible technologies. But what we kept seeing, the common theme of all of those was the idea of computing power. Has this made these things possible? But what was not changing was asset allocation. And we started looking around and trying to find other investors that were looking at this, and everyone was talking about machine learning, but in the concept of Go or chess, but not in a way that was able to invest in a real fashion. So as we looked at other Managers in the space that were looking at this, it was sort of the big hedge funds, but they weren't able to share what they were doing because it was such an edge they felt, so we weren't getting a lot of feedback. So we sort of thought about this in a more general fashion, and then Julie and Angelo came around, and they want to do something different. And I think when we look at our port…

AI assessment note: “it was sort of intellectually interesting for us to think about this as like a skunk works”

Q I want to get crystal clear on, on understanding what this model looks like. So let's start with You run this through your optimizer, and you come up with a bunch of factors that add up to a hundred in exposures. You mentioned 20% growth. What do the other buckets look like?

A It's interesting, because one of the big buckets that obviously we, as an endowment, have the ability to take advantage of is illiquidity. So illiquidity is a big driver of return, at least in the model. So you find that's a unique one that you have to Illiquity is different, right? It's private credit. It's private equity. It's some hedge funds. So I think that, that was one of the things that's the blurriest. So if you think about the mathematics behind it, high art squared, hard to apples to apples implement. You know, I think one of the areas, for example, right now, spread is not a big play based on where spreads are, right? So we're so tight in some areas that, you know, we don't, our models not want a lot of exposure to spread. I think when you think about momentum, That's a, that's a part of the portfolio that this year has done well for us. So we can look at long short, we can look at, so what you've seen in our portfolios broadly has been a shift from long only to long short. Now, we're probably early in that, hopefully.

AI assessment note: “one of the big buckets that obviously we, as an endowment, have the ability to take advantage of is illiquidity.”

Q So this, this question of optimizing factors and weighing factors comes up, right? You have this assets are not debate of whether you can time factors and smart beta. The factors themselves are a similar group of factors. How do you model what the factors are, and then how do you create a framework for WAKE that says, these are the exposures we want to these factors?

A Yeah, it's, it's a blurry lens. It's not much better than what you have now. So I think you still have to proxy the factors, right? So you can do certain things, growth versus value, you can look at different, different proxies, but I think the big key that the value of it is not constraining it. And I will tell you that the last couple years for us have been a challenge because The model says it wants growth. So you look at GDP growth and say, where do you want it? You're not buying in the U.S. Because there's no growth in the U.S. You're buying two percent growth, first investing in China and Russia, and for us, China and Russia, we can't do it because we believe in the rule of law. So we're not going to invest in those areas in a big way. So we end up finding our, we're doing more emerging markets, more frontier markets, and it's worked fine recently, but for the last eight years, if you weren't in the S&P, you weren't doing well. So I think, you know, the factor model is a blurry lens. It's not, it's, We end up finding our portfolios look a lot different, but our returns look very much the same.

AI assessment note: “I think you still have to proxy the factors, right? So you can do certain things”