E351: Why Most Family Offices Fail (And How a16z Fixed It)

What if the biggest edge in investing isn’t picking better assets—but structuring them better for taxes, incentives, and control? In this episode, I sit down with Michel, Founding CIO of a16z Perennial, to discuss how institutional investing frameworks translate to individual portfolios. Michel breaks down why most wealth management fails at true investment management, how misaligned incentives shape outcomes, and why scale, access, and structure matter more than traditional asset allocation. We also explore concentrated portfolios, tax alpha, and how psychology ultimately determines whether a strategy succeeds or fails.

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Many of these firms will charge you a relationship fee, some sort of flat fee based on your AUM. Fifty bibs, sixty bibs, no matter what you do. So let me ask you, if you made the same money doing something quite easy and highly profitable, or something very difficult and questionably profitable, which one would you pick?

• 0:18David Weisburd?

  One of the big trends in alternatives is the rise of a taxable investor. Why is that such a recent phenomenon? And why is asset management not focused on the taxable investor?

• 0:29Speaker 00

  A quarter of the total assets in the US markets is taxable. Let's say you and I are running our own asset management firm, and I told you, okay, we need a billion dollars to get to a reasonable level of profitability. And I'm offering you two paths to get there. One is find 10 institutional investors at 100 million each. Or two, let's sign 1,000 people that will write a $1 million check. And I don't even know where to go to find those. And if they do go through intermediaries like banks or eyes, they're fiercely guarding those clients. They're not going to sort of propose that those people invest in your fund.

• 1:02David Weisburd?

  You came from Jordan Park where you were CIO, and Mark and Ben gave you this interesting thought experiment of how would you build a multifamily office from scratch? What were your first thoughts?

• 1:13David Weisburd?

  So Michelle, you're the CIO of A16Z Perennial. We're here at Andreessen Horowitz, built this massive venture capital firm. Why take the time and maybe the attention to build a family office?

• 1:23Speaker 00

  Well, several reasons. I think reason number one is a deep commitment to the founder community and the idea that this firm helps founders in every dimension of their startup. So why not extend that also to their personal financial work? And therefore ostensibly let them focus more on their startup and less on these sort of mundane issues.

• 1:43David Weisburd?

  Addition through subtraction.

• 1:44Speaker 00

  Addition through subtraction. I like that. I like that. That's good. So that's one element. Another one is just a community building element. I think these people like to interact and be together. And so building a community is valuable. And who else to build a community than the largest venture community there is? And then third, they would look at to the LPs they have in their own big funds. And they would get, you know, the big pension funds, the some wealth funds. They have these very professional teams and they'd witness that. And then they turn and witness their wealth team. And they'd realize that the wealth team really weren't investors by training, right? They were more sort of service providers. And what they wanted was investment acumen. And so they figured they couldn't be the only ones that had this problem. And if they fixed that, it probably would be valuable for many people.

• 2:26David Weisburd?

  The idea that wealth management isn't necessarily the best investment management, not very controversial to me or my listeners, why build a multifamily office?

• 2:35Speaker 00

  Building a single family office is not as easy as you might think. So on the side of the principal, I think many of these principles, their dream is not to be the CEO of an asset management company, right? And manage people like me, right? So that's the first step. If you are really going to run a single family office properly, you're going to be the CEO of an asset management company. You're going to have to hire people, motivate them, give them a career path, et cetera, et cetera. And that's not something many of the founders or many people in general want to do, want, love to do. They'd rather focus on their startup or whatever they're doing, or their next company. So I think first, there's a desire element. And I think in fact, that's a lot of, I've seen a lot of single family offices start and then stop when the principal realized that their job is actually to be a manager, right?

• 3:18Speaker 00

  And their whole sort of idea was I got wealthy and now I'm suddenly a manager? Like, I wanted to detach myself from that, not implicate myself even more. So that's on the side of the, of the principal. And then on the side of the investment professional, why would I leave? If I'm really a top notch investor at a large asset manager, making millions of dollars a year, what's the appeal for me to go to a family office? It has to be that I've got growth opportunities, I can do more investing, interesting investing. I can maybe build a team, I can do things that are not, that are more difficult to do in a more constrained environment. And often, when they arrive at the family office, and I've seen again, many of these stories, it's not at all clear how much discretion they have or authority they have. And it sometimes takes years and years for the principal to get to a place where they're

• 4:02Speaker 00

  comfortable giving some leash to this person. And so as a result, you see often the CIOs come and go. You see these revolving doors because both sides have misinterpretations of what the job is and what it requires.

• 4:14David Weisburd?

  It's interesting because that also applies to endowments. So you might get a fresh pool of capital, you deploy it, and then you're in management mode for sometimes decades.

• 4:21Speaker 00

  That's right. So your, your fund is fully deployed, you're the CIO, you've got nothing to do, you're not, you're not in the latest trends anymore, you're not honing your skills. And so having a small, static pool of assets, and by small, I mean a billion or less, call it. It becomes very difficult to keep a team. And I would argue that to do it properly in multi-asset global kind of team with all the ancillary functions around tax, you probably need about 20 people, I'd say at a minimum. And so if you're really going to do this, you know, equivalent to a real asset management firm. And so that requires quite a bit of assets and quite, it's quite a bit of headwind on your returns. So you need to have enough of a pool to do that. I'm not saying you shouldn't do it. I'm just saying that it's, you know, only large pools of assets should really attempt to do this.

• 5:03Speaker 00

  Now, you do see families that are smaller and that try to do this, but they'll do it in a specific asset class where they have expertise or knowledge. So maybe, you know, your smaller family, I don't mean really small, but your small billionaire family, maybe you're good at PE for whatever reason, your operating business, so you hire a PE person and you do PE in your space. That's fine, but that's different than building global multi-asset portfolio.

• 5:24David Weisburd?

  Right before we start recording, you said something that surprised me, which is asset management scales. So as you get more capital, you actually get an efficiency of scale. I've been led to believe size is the enemy of returns. How do you square those two ideas?

• 5:38Speaker 00

  A lot of LPs investors don't appreciate the fact that scale benefits them too. They think that scale only benefits the GP. But I think that it benefits the LP too in several ways. One is of course, the ability to sort of throw your weight around seed strategies, ask for concessions on terms, right? So there's that element. There's also the risk management, monitoring and reporting element, which becomes very, very difficult when you're dealing with dozens of portfolios that have completely different allocations, right? And as a result of having that difficulty, it's very hard to maneuver and control something when you don't have a good read of what it is, right? So you don't really know what your risk exposure is. You don't really know what factors you're exposed to or whatever.

• 6:20Speaker 00

  And so as a result, it's very difficult to do course corrections or be opportunistic or whatever. So having scale allows you to build all those systems and infrastructure that then enables the flexibility to do that stuff. So it's, I think it's valuable for both. I think it's a curve that maybe looks something like an upside down U. There's a point where you're so large. Right, that your minimum check size becomes really big. And so now you're in highly competitive deals. And so your sort of margin gets eroded. But there's definitely a sort of a scale optimum, probably in the, I don't know, 15, 20 billion dollar size.

• 6:52David Weisburd?

  Expert calls have always been one of the most powerful ways to build conviction. But today, investors are asked to cover more companies, move faster, and do it with leaner teams. With AlphaSense AI-led expert calls, their TGS 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 queryable, searchable, and comparable. So your primary insights plug directly into earnings prep, digital work streams, and pitchbooks with zero tool switching.

• 7:36David Weisburd?

  And 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. All of it lives inside the AlphaSense platform, trusted by 75% of the world's top hedge funds

• 8:19David Weisburd?

  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/HowIInvest. I want to go back to this concept of wealth management versus investment management. Maybe double click and maybe give very specific ways how there's a principal agent problem. Is it just on the fee and selling products? Or is it fundamentally something on a portfolio basis?

• 8:53Speaker 00

  So many of these firms will charge you a relationship fee, some sort of flat fee based on your AUM, 50 bps, 60 bps. No matter what you do. Okay, so let me ask you. If you made the same money doing something quite easy and highly profitable, same revenue, right? Doing something quite easy and highly profitable or something very difficult and questionably profitable, which one would you pick?

• 9:16David Weisburd?

  I'd probably do the easy, but my brain as such, I would figure out a way to systematize the easy and keep on going hard.

• 9:22Speaker 00

  But that then requires you to go hire a team of people who know how to do alternatives. We're back to that problem of how do you manage that team? How do you incent them and so on? So you're gonna have to have a scale and you're gonna have to have an investment DNA in the org. So many wealth manager firms don't have that. It's very hard to attract. Let's use a physician analogy, like you're trying to run a physician practice, but all the leaders don't know anything about medicine and don't care about it. If you're, would you join that? Or would you join something that's run by other physicians who appreciate your art and your craft? Right? So I think it's very difficult inside a wealth firm to even attract the right talent to start with, because you're gonna be sort of a fish out of water.

• 9:58David Weisburd?

  I kind of liken it a bit to these AI recruiting wars. There's probably a couple thousand, maybe 10,000 elite AI researchers in the world. And this is the pool of people. So they're being picked up. I think the same as investment management. And the reason I know that is when I meet with people, usually they talk about the same 50 to 100 people that are really good in the space. So there's a finite amount of resources and finite amount of talent. And if you don't have that DNA, you don't have the resources, you don't have incentives, you're not gonna recruit them. And also, if you recruit one of them, they're gonna leave. So you need a certain critical mass as well.

• 10:29Speaker 00

  Totally. You need critical mass, you need growth, you need ability to participate in the current trends and whatnot to keep these people engaged and interested. One day, maybe AI will replace all that. But right now, not yet, luckily.

• 10:39David Weisburd?

  One of the big trends in alternatives is the rise of the taxable investor. Why is that such a recent phenomenon? And why is alternatives and why is asset management not focused on the taxable investor historically?

• 10:51Speaker 00

  I was just looking at these statistics yesterday just to refresh myself, right? Something like a quarter of the total assets in the US markets is taxable. So the vast majority is not taxable. That's one, right? So if you're running a firm, obviously, first place to go is the bigger pool, right? The bigger TAM. Second thing that's important also is, let's say you and I are running our own asset management firm and I told you, okay, we need a billion dollars to get to a reasonable level of profitability. And I'm offering you two paths to get there. One is find 10 institutional investors at 100 million each. And by the way, just by doing a simple Google search, I can identify 300 institutional investors that could cut those checks. Or two, let's find a thousand people that will write a $1 million check. And I don't even know where to go to find those. And if they do go through intermediaries like banks or eyes,

• 11:36Speaker 00

  they're fiercely guarding those clients. They're not going to sort of propose that those people invest in your fund, right? In fact, they will actively try to convince the clients not to do your fund, right? Because they're being paid based on how much AUM they're monitoring for you, right? In terms of relationship fee. So if I were to ask you that, I think everyone would answer, I'll go for 10 clients that I can easily identify. And I only have 10 relationships to manage, right? So there's that element. And of course, now that there's been this sort of explosion in, in, you know, families and taxable individuals that are of institutional size. There's now, of course, a lot more interest in this as a new growth market because that other market is fairly saturated, right?

• 12:13David Weisburd?

  So now all the endowments, pension funds, a lot of them to our, to our earlier conversation have deployed. They have picked their 15 to 25 managers and now they're only adding out the edges. And now the new capital coming on, you'd say the taxable investor, also there's obviously the sovereigns coming on, but taxable investors are one of those net new.

• 12:30Speaker 00

  They're the marginal dollar coming into the market. That's right.

• 12:33David Weisburd?

  You came, you came from Jordan Park where you were CIO and Mark and Ben gave you this interesting thought experiment of how would you build? A multifamily office from scratch. What were your first principles?

• 12:44Speaker 00

  The first principle was to get rid of all these principal Asian misalignments that are out there. Second principle was to my point earlier about wealth management firms. We want to be investor centric because what we're doing is giving you advice on your portfolio. And so to use a medical analogy again, if you walk into a hospital, a building that says hospital on the, on the door, you kind of expect there to be physicians inside, right? Same thing here. If it says asset manager or wealth management of which asset management is part of that, you'd expect asset management people to be inside of there. So many of the people on the team have had direct investing experience have been professional investors before being allocators and portfolio constructors, right? And what do I mean by professional investor? At some point they were paid on the P and L of their investments. That was all that was driving their compensation. And that discipline teaches you a lot of things around how to invest.

• 13:29Speaker 00

  It also gives you the flexibility to implement investments either through third party management or directly. If you don't have the direct experience, you're limiting yourself to only doing the fund to fund approach. Right. And that's okay, but it's an expensive way to do things. Right. On a look through basis, a typical endowment foundation portfolio, you're probably talking 400 basis points of fee drag. Right. So anything you can do direct, you're going to reduce your fee and ostensibly you can even do it more tax efficiently.

• 13:56David Weisburd?

  It's important to state the opposite of what you did. So if you bring in investment managers, doesn't that sound obvious? Wouldn't everybody do that? Well, the industry practice is you bring somebody with a book of business, with the relationships and you focus on who has the most clients, not who is the best investors. Those are kind of the two archetypes and this one has really dominated.

• 14:13Speaker 00

  Yep.

• 14:13David Weisburd?

  The last 40 years.

• 14:14Speaker 00

  That's right.

• 14:14David Weisburd?

  How do you look at a private and public portfolio for an individual? And how do you weight those?

• 14:19Speaker 00

  So that's another thing where perennial is very focused on being customized. Right. And so if I'm trying to, so I just, I just said a minute ago, scalability can benefit the LP, but there's some instances where scalability is solely for the benefit of the GP. One of those is running playbooks. So I look at you and I put you in some bucket. I'll say, you know what, David, you're like a middle risk kind of guy. So here's the standard profile I'm going to deploy for everyone just like you. And I'm not going to take the time to understand your tax situation, your personal trusts, your personal preferences, nothing like that. I'll just build kind of standard portfolio for you. Right. So we don't do that. We'll sit down and try to understand, and there's an element of psychology in this. Try and understand your risk profile, your preference, your goals.

• 15:01Speaker 00

  Like, do you want to have some sort of large philanthropic gift later in life?

• 15:05David Weisburd?

  And the psychology also takes root in the portfolio and in the returns. The strategy is only as good as its implementation.

• 15:11Speaker 00

  It's only as good as implementation. And if it doesn't meet the goal of the client, and I, there's an interesting analogy there to do with endowments and foundations, then it's useless.

• 15:21David Weisburd?

  Earlier in your career, you spent time at McKenna, which was a spinoff from Stanford's Endowment. What did you learn at McKenna?

• 15:27Speaker 00

  So before McKenna, I was in a hedge fund. And before that, I was doing, serving private equity clients. So I was much more of a direct investor. And then McKenna was my first sort of foray into, you know, being an allocator. And I was what they call there the strategist, which was sort of the number two to the CIO. This was the team that spun out of Stanford. Right. And so I got to see what a top notch endowment does, what professional high end manager selection does, and all the good things around how a portfolio should be managed, monitored, risk management, all those kinds of things. And then, and then from there, then I went more to the family office world. And, and you know, the difference is quite stark in terms of the asset management quality, but also in terms of the attention to tax and individual goals, right?

• 16:09Speaker 00

  Where the Endowment Foundation world is a very specific goal in mind, whereas individuals have much more varying, you know, goals.

• 16:19David Weisburd?

  You said you got to see what excellence looks like at Mecana and what excellent asset portfolio and construction is. Double click on that. What does it mean to be elite when it comes to portfolio construction?

• 16:30Speaker 00

  It wouldn't necessarily say it's portfolio construction as much as it is having access to the top managers. If you look across all the top endowments, you know, I could rattle off a dozen managers that every single top endowment and foundation is in, right? It's, I think there's broad agreement.

• 16:46David Weisburd?

  Consensus.

• 16:46Speaker 00

  Consensus, group think, if you don't want to be around what a sort of standard portfolio should look like.

• 16:54Speaker 00

  And then people sort of tweak on the edges around that in the Endowment Foundation world. And it's reflected in the returns. If you look at the returns, they're all clustered, three, five, 10 year, eight, nine, 10%. They're all pretty tightly clustered. Ex ante, if you didn't know what I just said about, you know, kind of this sort of consensus, 10 different portfolios, you'd expect quite a bit of dispersion. And in fact, they're quite tightly clustered, right? And between that eight and 10% roughly spread across them, right? If you assume that the returns across the different endowments are sort of more or less uniformly distributed, that means every 50 bips across that 2% is like a quartile, right? So eight to eight and a half, the bottom quartile, eight and a half to nine, you're now at the middle.

• 17:33David Weisburd?

  This is over 10 year periods.

• 17:34Speaker 00

  Over 10 year periods. And then the other interesting thing, right, for endowments is the compensation structure is a sort of self reflective compensation structure because you're paid based on your performance relative to other endowments. And since they're all super tightly clustered, effectively what you're doing is you're telling one endowment manager, you will get a bigger bonus if you manage to move your return. Cause 50 bips is your median and top quartile, something like that. I'm making a lot of assumptions here, but, and so what does that mean in terms of portfolio difference across the, it's a small tweak, right? You're not going to have this massively different asset allocation. That's why when you look at their asset allocations, they're not that different. You don't have someone with 50% venture. Or it's pretty rare if you do, right? And I would argue again, for individuals, it's a different story.

• 18:18Speaker 00

  For endowments, right? What's the goal? The goal is to meet your payout or five, six percent. Hopefully not compromise the corpus, right? So maybe you keep the dollar amount steady or maybe even keep the real value of it steady. So if you assume like a three percent inflation, you need to get an eight or nine percent return on the portfolio.

• 18:35David Weisburd?

  Because you're distributing five percent.

• 18:37Speaker 00

  Distributing five, you got three of inflation. Lo and behold, they're generating right that. So it's not when, when people say, oh, endowments care about returns. They care about risk managed returns, right? It's more of a risk management exercise. It's not a maximize return.

• 18:51David Weisburd?

  And the reason they're not doing 50% venture and trying to be top quartiles because there's also downside risk. You lose your job.

• 18:56Speaker 00

  You lose your job.

• 18:57David Weisburd?

  They're trying to optimize just enough to get their bonus, but not necessarily enough to lose their job. And of course, all this is subconscious.

• 19:03Speaker 00

  Subconscious? Okay. I don't, okay.

• 19:06David Weisburd?

  Before I get in trouble.

• 19:08Speaker 00

  It's 100% subconscious. That's right. But, but to your, to your point, right? So as a result, you get these very similar portfolios and the exercise, a risk management exercise, right? It's not a profit maximization exercise.

• 19:20David Weisburd?

  Begs the question for a taxable investor. I'm not asking you to tell me about funds or investments, but in terms of on a portfolio level, what's a realistic growth portfolio? What should that return over 10, 20 year period?

• 19:31Speaker 00

  But this is where the psychology and the desire of the client comes into play. Right? So you could have an individual tell you, and I think this is a valid asset allocation, tell you, look, I only need 10% of my balance sheet to live the rest of my life, me and my family. So I'll keep that stuff liquid, low risk. Everything else I wanted in the highest returning asset class. Maybe that's a venture. So there you go. 90, 10. That's a valid asset allocation for an individual. You would never see an endowment or foundation do that because obviously it's a volatile portfolio.

• 20:01David Weisburd?

  And the reason it's valid is because over, the assumption there is that over a long enough time period, you're not saying put it 90% into one fund or one vintage. You're saying maybe over six.

• 20:10Speaker 00

  Oh, yeah, no, no, no. Very systematic, multi-strategic, multi-vintage, all that stuff for sure.

• 20:16David Weisburd?

  So the philosophy is if something is the best returning and you're fine with the volatility, you put as much into this best returning as a class and then you diversify within that.

• 20:22Speaker 00

  What's really interesting, by the way, is.

• 20:24David Weisburd?

  It's pretty simple.

• 20:25Speaker 00

  It's pretty simple, right? And what's really interesting is when the first optimization tools started coming out. People would put in private equity and they would say, okay, I assume it's going to outperform listed equity by 600 base points. They typed that into the assumption and then they would look at the observed volatility, which is different than the actual volatility. And they'd say, oh, it's got lower volatility than the stock market. So, and then the computer would say, guess what? Put all your money in that. And I would always say, I don't need a computer to tell me that, right? If it's got a higher return and a lower risk, we don't need any optimization engine or spreadsheet to know that's where you put all your money, right? So then what people would do is then they put an artificial constraint. They say, well, for liquidity reasons, we can only put 20% in PE.

• 21:04David Weisburd?

  Or we have to be diversified.

• 21:05Speaker 00

  Right. Low and behold, the computer comes access 20%. And then people would point to that and go, look, the computer said 20%, right? But actually, what it was, was them constraining the output to say what they wanted to say, right? It was always kind of, kind of comical.

• 21:16David Weisburd?

  Do you have any clients in this 90-10 portfolio or some, some combination of that? And if so, and if not, why not?

• 21:22Speaker 00

  Yes, for sure. And I would say most, most, I would imagine many asset management owners, asset management firm owners would have the vast majority of their wealth in whatever it is they do. You're a head, you own a hedge fund, you're gonna have 70, 80, 90% of your wealth in your hedge fund. You own a PE, whatever it is, right? And you, and by the way, you're not paying fees as you're the owner, right? And so on and so forth. So it makes it even that much more compelling to put, you know, a large amount. Back to my other point, that depends on the psychology of the person. If you're more of a risk managed person, and you're worried more about downside, then maybe you build more of an entirely stock portfolio.

• 21:55David Weisburd?

  So I want to get to the psychology in a bit, but it feels intuitively that that's crazy. Is that just an intuition that comes from the marketing and the narrative?

• 22:04Speaker 00

  It's called dogma. It's, it's, we've all been trained to react as if that we're crazy to sort of do everything.

• 22:11David Weisburd?

  My second order question is, if you are 90% of that asset class, how risky is that across different vectors, especially if you've done well? So how risk, can you do that in a diversified manner, in a conservative manner?

• 22:26Speaker 00

  I hesitate to call such a portfolio conservative.

• 22:30David Weisburd?

  But why not?

• 22:31Speaker 00

  Well, I would say it's very different than putting all your money in one stock. Right? So I think people analogize that to saying I'm putting all my money in Apple or whatever. Pick your stock. It's different here because you're going to have, you know, hundreds of positions. Right? And so, yeah, it's, there's less idiosyncratic risk on any one company doing well. So there it's risk mitigated. Nonetheless, you're a sort of growth equity risk factor. If you want to use factors is going to be very high. Right? So you're going to have a high, you know, correlation and to the equity markets. You know, it's lagged and hard to measure. You have high volatility, but again, if you've taken that money out of your mind as something that's not required for your lifetime, then why not do that? Right? So I'm not advocating that it's the right allocation for everyone, but I'm saying it's not an invalid allocation either.

• 23:17David Weisburd?

  It's a great thought experiment. I'm actually in over 500 startups. And the way that I look at it, I'd like you to correct my thinking is that it's a baby S and P 500. So I'm not like in 500 fintech companies. I'm not in 500 consumer. I'm very diversified. Now I'm very illiquid. It's another issue, but I'm very diversified within that pool of assets and I kind of think about it as some sort of index of private companies across stages.

• 23:39Speaker 00

  I would argue that you're trying to maximize your total end value wealth. Right? Instead of trying to sort of keep some steady sort of return profile that mitigates inflation.

• 23:49David Weisburd?

  It looks really difficult for about six, seven years. It's a very difficult ride, which I think, which gets us to the second part of the equation, which is if you're psychologically built for that.

• 24:01Speaker 00

  And not everyone is.

• 24:01David Weisburd?

  Is there even a way to figure out some, whether somebody's psychologically built for that before you construct a portfolio?

• 24:06Speaker 00

  You know, it's really interesting because the, the more I do this, the more I realize that part of my job is kind of psychology, whether I'm dealing with institutional or individual. Right? Institutionally, it's board psychology. Right? There's always the one person on the board that's the doomsayer. And there's the one person that's the unabashed optimist. And you have to sort of try to balance the group, right? So that they don't do anything too extreme.

• 24:27David Weisburd?

  Is that, is that a strong counterbalancing force?

• 24:31Speaker 00

  I see it a lot on these boards.

• 24:32David Weisburd?

  Is that positive or would you rather have two optimists, two doomsayers?

• 24:37Speaker 00

  I wouldn't want to have too much of any one thing, I guess, because then you sort of tilt the portfolio in a way that the group may not agree with. There's also an element, again, psychology, that some people are more vocal than others, and they tend to dominate the discussion. So board dynamics is its own thing. I remember, you know, meeting with boards during, for example, the global financial crisis, right? And there'd be a very vocal person on one of these boards who was a doomsayer. And he's like, well, next, you know, so and so who's a genius told me the next stop for the S and P is 200, 300, another 50% drawdown from here. And he would say this with great conviction, and then the board wouldn't rebalance. Right? And so that's what I sort of labeled that in my head. It's not a polite term, but it's like board dysfunction. Right? And so the way I would immunize boards against that is I would have everyone agree

• 25:19Speaker 00

  that we rebalance as a, as a base agreement, unless someone vocally tries to block the rebalance. I tried to flip it on its head. Instead of having everyone agree to a rebalance, you'd have to have someone disagree with it. So that's a psychological. Opt out versus opt in? Yeah. But that's psychology. It's not, has nothing to do with investment management. Right? And with individuals, a little different because you're working with one person or maybe two, because a couple, or they might have a couple of advisors as well. So it's a little less trying to sort of get a group to agree to something. And there's therefore, you got to try to spend more time trying to understand the person's outlook and risk. And often, they come already with a portfolio that tells you a lot. So they're coming to you with a portfolio that's 70% bonds is a pretty clear sign they're not a risk. It's their behavior.

• 26:04Speaker 00

  Right. You come to me with your portfolio of 500 startups, right? What am I going to be doing? I'm trying to get you to tamp down on risk. Yes. Right? As my wife does. As your wife does. As she should. As she should. And hopefully, as she's somewhat successful. Yes. And then, and then the other way around. If someone came to me with 70% bonds, it'd say, you know, it's okay to take a little risk, right? Especially if you have a large balance sheet, right? So a lot of this is trying to get people to accept the path. To a better risk.

• 26:28David Weisburd?

  If this is psychology, maybe you're my therapist. So let's start, let's start with this 90%. Let's not call it model portfolio. This is called is a crazy man's portfolio. How would you start to build on it and introduce new asset classes to?

• 26:41Speaker 00

  You know, the issue with the 90 10 type portfolios that you have a very high illiquidity. Right. And so the 10% you do have hopefully is highly liquid and low risk because otherwise if that has a drawdown, now you have a problem with putting food on the table, which I think is probably what your wife worries about.

• 26:55David Weisburd?

  Yeah.

• 26:57Speaker 00

  But as you have, so the idea is that hopefully you can bring that a little bit down over time and introduce some other asset classes that have interesting return profiles, but also have really good tax characteristics, right? As an individual.

• 27:09David Weisburd?

  Like real estate.

• 27:10Speaker 00

  Like real estate, but real estate structured in the right way.

• 27:13David Weisburd?

  What does that mean?

• 27:14Speaker 00

  So you as an individual can take advantage of depreciation, depletion, cash out, refi and so on, so forth. And that has to happen inside a partnership type structure. Even a REIT structure does not transfer all the great tax benefits that real assets and real estate have to an individual.

• 27:31David Weisburd?

  What would these magical structures be called?

• 27:34Speaker 00

  There's nothing magical about it. It'd be an evergreen type fund, right? Where you as the investor have to keep control on the disposition of the asset, which puts you at odds with most of the managers in the space who they want to sell because that's how they crystallize their carry. So the way you align them is you pay them out of the operating cash flow of the asset. So now they're aligned.

• 27:55David Weisburd?

  This is this tax away real estate investing. I had a spin out from David Kattler's family office that's really focused on this space. I've never really seen anyone else on it.

• 28:02Speaker 00

  Well, we do that. We do that here and so.

• 28:04David Weisburd?

  And that's, that's with ingrown talent. That's with A16Z perennial.

• 28:08Speaker 00

  So this is back to my earlier point. If you hire a team with people that had direct investing experience before they were doing portfolio construction allocation, you've now liberated yourself from only having to go to third party managers. Right. So you can do a lot of this directly with actually in some cases, the personal assets of GPs of real estate firms, things like that. And so you can align yourself with them very carefully. Everyone gets paid out of the operating cash flow. There's no more pressure to sell the asset. Right. And you're able to underwrite the asset because you have the right person on the team. So we go out and hire seasoned investment professionals, people with 20 years experience doing this in the asset class. So that we are not. Purely beholden to third-party managers, right? You should have the flexibility to decide whether you pay a third-party manager

• 28:52Speaker 00

  or whether you do something yourself. I'm not saying that either one, you should never just sort of preemptively limit yourself to just one of them.

• 29:01David Weisburd?

  And if you do have a direct team, does that limit you from manager selection? Or is that just a skill that you could learn very quickly?

• 29:07Speaker 00

  Well, there are people who've done both. They're rare, and that's why it's taken us a while to get our team to 20.

• 29:13David Weisburd?

  So you hire people like that?

• 29:14Speaker 00

  We hire people like that that have worn both hats. And so then the question is, do we have the in-house skill to do something ourselves? Great. We don't? Let's go pay somebody. But I don't want to ex ante by my very team setup, my fund setup, to limit myself to one of those, right? And the entire industry is that way, right? So you go to McKenna or other firms, any endowment firm, I'll source the L firm, it's all 100% third-party managers, right? And then you go to a hedge fund or a PE fund or whatever, and of course, it's going to be 100% direct. Very few people try to do the hybrid. And I would argue that as an individual or even an endowment or foundation, when you're building your own thing, you would like to do a blend.

• 29:57David Weisburd?

  It sounds reasonable, but wouldn't going direct always be better?

• 30:01Speaker 00

  There's a point where you would have to have such a large team to be able to sort of address every single idiosyncrasy within an asset class that, I mean, I suppose you could always go direct, but you would need a very, very large team.

• 30:12David Weisburd?

  It's just some of these are such niches, and you need a certain amount of capital to make it work. The same scaling issues that we talked about.

• 30:17Speaker 00

  That's right. That's right. So pick your spots where you go direct. Save yourself, in the case of real estate, a gross to net spread of 400 to 500 basis points. Maybe you cut that in half. In other places where you have to pay someone because it's a niche, it's a specialty, you don't have it, fine. But don't preemptively, before you even wake up in the morning, limit yourself to one of the two.

• 30:36David Weisburd?

  What other asset classes do you like as a taxable investor? Real estate, venture, anything else?

• 30:42Speaker 00

  As you know, there are a lot of hedge funds out there that have launched these taxable wear products, right?

• 30:46David Weisburd?

  None of them will go on the podcast because it's political football.

• 30:49Speaker 00

  I will not name names, but a simple Google search or Claude's search will tell you that. But the sort of strategies that they use are accessible to experts. The way you do this is with swaps. And the sort of weird idiosyncratic tax treatment of how these swaps are treated when it comes to tax.

• 31:09David Weisburd?

  And you do them together with banks?

• 31:11Speaker 00

  You do them with banks and you swap in and out of certain exposures and this allows you to generate losses or short-term losses, long-term gains, things like that. And can mitigate the tax characteristic of income, which normally is, you know, the worst tax rate you can pay. So it's, you know, I heard you say at the beginning that you cared, that you believed in structural alpha. This is structural alpha. You're structuring your fund, you're structuring your investments. We deductible cost for you as the investor. So all the things you can do inside the fund to structure them result in the fact that you don't even have to be a great real estate investor to outperform the average real estate fund. I love that. If I can, I'm obviously I'm-

• 31:53David Weisburd?

  It's idiot-proof. It's like the Warren Buffett investing principle. Invest into great businesses that even an idiot could run if you had to.

• 32:00Speaker 00

  My, my hope and my belief is that my teammates are not just checking. And I think they'll do better than average, but let's sort of back away from that and let's say they're just average. Right. By the way, we structured things. We're going to outperform on, on, on an after-tax basis.

• 32:15David Weisburd?

  And when these venture capitalists or some people call us crazy people come with you, these venture portfolios, is there a way to play defense on that part of their portfolio? Or do you just take it as illiquid capital and build idiosyncratically around it?

• 32:27Speaker 00

  Build defense. What do you mean by build defense?

• 32:31David Weisburd?

  You mentioned a little bit about it, which is if 90% of your money is illiquid, 10% of it should be highly liquid, even if you take away returns. Let's just say there's a direct trade-off between liquidity and you need to play defense. You need to have liquidity there because 90% is illiquid. So how do you think about building portfolios around venture portfolios?

• 32:47Speaker 00

  Going back to what I said earlier, right? Liquidity and, and counter-correlation, right? So if you can get things that are not correlated to the venture, and I'm going to sound repetitive, but for example, certain credit strategies, real estate strategies are not correlated to the broad equity market beta that venture is. That can actually be better than just buying fixed income, which has basically no correlation to anything.

• 33:09David Weisburd?

  Is more or less venture, you should think about it as correlated to equities and you could use that principle?

• 33:15Speaker 00

  I think so. So when you observe venture or PE portfolios volatility, right? Because the marks are only quarterly and because there's only one audit in market year, the volatility you observe is lower.

• 33:28Speaker 00

  Then the underlying volatility inside the actual assets. So what we do is we benchmark venture and some private equity to a small cap index that we actually lever to incorporate the high inherent volatility in those portfolios. But you know, a lot of people equate volatility with risk. By the way, this is a broader discussion than just venture. To me, if you have a very large balance sheet, volatility is actually a friend because volatility means there's going to be a sell off at some point, it's going to be a blue light sale at some point. And if your portfolio is flexible enough and built properly, you need to be able to move quickly and take advantage of the blue light sale. So if you're a long-term investor with a large balance sheet, and this could be an endowment, a foundation, or very large family, volatility is your friend because it provides you these occasional rifle shots at buying things on a cheap.

• 34:14Speaker 00

  Is venture capital the best asset class in the world?

• 34:18Speaker 00

  Is venture capital the best asset class in the world? That's a loaded question. It's the broadest dispersion asset class, right? So if you look at worst manager to best manager, I'm not telling you anything new, right? The dispersion is really wide, right? So whereas if you look at, I don't know, fixed income managers, right? There's hardly any dispersion. So manager selection, you know, is very important in this case, right? And if you take- Can you select in a way that makes it the best asset class without being lucky? I hope we can, and so far we've been able to, but it's difficult. It's difficult to guarantee that. I mean, obviously, we're not in the business of trying to guarantee those kinds of things, right? Certainly not in this business. Yeah, yeah, yeah. What's something that you've changed your mind on in the last 12 to 24 months?

• 34:58Speaker 00

  One has been the realization that tax alpha is by far the easiest and best alpha to generate for individuals. And as a result, when you're benchmarking the performance of a taxable individual, the appropriate benchmark is not the index. Tax loss harvested, tax managed index investing has become such table stakes that to me, that is the benchmark. And that's the benchmark against which you have to assess other investments. And that benchmark, already just the index, as we well know, you know, 90% of active managers don't beat the index on an after-fee basis. On an after-fee tax-adjusted basis, I don't know if there's anyone.

• 35:38David Weisburd?

  The endowments might benchmark S and P 500 MSCI taxable investors. You argue should benchmark a tax loss harvested, maybe S and P 500 MSCI as well.

• 35:48Speaker 00

  Correct. And those things are generating anywhere from 100 to 500 basis points of tax alpha, maybe even more, depending if you lever them a lot, you could generate.

• 35:56David Weisburd?

  And that assumes that you need the tax alpha.

• 35:59Speaker 00

  If you're taxable investor, you probably need the tax alpha. The beauty of it is that those taxes you can build up, the tax losses, you can carry them forward and build them up. So you can build up a pool of these things that you use for the day when you realize the gain somewhere.

• 36:11David Weisburd?

  Michelle, if you could go back to 2000 and you had just gotten your PhD from Stanford, what is one timeless piece of advice you'd love to have given yourself at that time that would have either accelerated your career or helped you avoid cost of mistakes?

• 36:22Speaker 00

  This might sound naive. At the time, I thought that

• 36:29Speaker 00

  if someone was appointed a leader, it meant that they had great leadership qualities. And what I've come to realize is that leadership is earned. It's not bequeathed and, and great leaders become recognized because of how they incent and empower their team. And I've been lucky enough in my career to work with some great leaders. So when I like to say, when I come out of meetings with some of the leaders that I've worked with, who were, grew into being leaders because of their skill.

• 37:00Speaker 00

  I like to say, come out, I could like sort of climb Mount Everest, you know? And then unfortunately, there are a number of, you know, leaders where you don't get that feeling when you come out of meetings. So that was a learning that was hard for me to sort of stomach to realize.

• 37:16David Weisburd?

  A lot of entrepreneurs say their biggest growth came from their biggest challenges. Has that, has that happened in your career at all?

• 37:23Speaker 00

  A hundred percent. So I'd say the most humbling job you can have is working in a hedge fund, where you're so obsessively focused on a specific company, a specific stock, whatever it might be, a specific security, you think you know that, you spent six months studying it, you know everything about everything, and yet you end up being down. So you get a giant frying pan, smashed on your head, you very quickly learn humility. I call it market therapy, back to the psychology theme. And after you've had a bit of market therapy, you're much more careful about how you deploy capital, the statements you make, the promises you make to people about what the portfolio may or may not do.

• 38:02David Weisburd?

  Michelle, this has been truly a masterclass. Thanks so much for staying down.

• 38:05Speaker 00

  Yeah, thank you. Thanks for having me.

• 38:07David Weisburd?

  If you found this conversation valuable, please click Follow How I Invest so that you don't miss the next episode with the world's top investors.