Baylor University's endowment is about $2.7 billion, and roughly 2.5% of it sits in Anthropic. It owns nothing in SpaceX and nothing in OpenAI.
The usual endowment complaint about venture is illiquidity. David Morehead's complaint is arithmetic: a fund that returns 15x after 18 years loses to a growth-equity manager who returns 3x in six years and does it three times.
"Students can't pay their tuition with returns. They have to pay with dollars."
Morehead has run Baylor's investment office since joining the university in 2011, and before that was a senior portfolio manager at several Chicago hedge funds, investing across corporate securities, distressed debt and public and private energy.
The full interview is covered here so you can skip it. 69 minutes of audio, 29 minutes of reading.
Here are the 19 principles that matter.
👤 Guest: David Morehead, Chief Investment Officer at Baylor University, who runs its roughly $2.7B endowment and previously managed money at several Chicago hedge funds
🎙️ Host: Harry Stebbings, founder of 20VC and a venture investor himself
📰 Published: 14 September 2026 on YouTube (20VC with Harry Stebbings)
🔴 YouTube | 🟢 Spotify | 🟣 Apple Podcasts | ⏱️ 1 hr 9 min | ✅ Time saved: 40 min
Key Takeaways
A 15x venture fund over 18 years is beaten by a 3x growth equity fund recycled three times
Three consecutive 3x funds over six years each compound to 27x
Baylor holds about 2.5% of the endowment in Anthropic and nothing in SpaceX or OpenAI
The office banned talking about a return without saying how long it took
5x over 30 years is a bad outcome; 5x in five months is a great one
Morehead bought software in early 2026 by calling family-business owners, not engineers
Their answer to replacing a working system with AI-written code was no
A data center's scarce input is no longer power, it is a permit
Baylor's data center sites are up 50% in six months
Baylor tells managers the dollar it wants in each portfolio company, not the percentage
$3 million per company, because a $400,000 exit means nothing to a $2.7 billion fund
A manager who changes strategy gets fired even if the new strategy is working
Private credit is the asset class he calls overhyped, because it prices like equity on the way down
1. Where the Tuition Went
Morehead opened on the business his endowment sits inside, because it is what sets the return the office has to earn.
Fewer American high school students are coming through, a consequence of the birth rate after the global financial crisis, and fewer high school students means fewer tuition dollars.
The second squeeze is international students, who are full-pay. The last couple of years have made it harder for them to come to the US, get the right visas and stay.
The result is competition for domestic students, and this year's incoming class — the class of 2030 — left a lot of schools short of their targets. Where the tuition does not come in, the revenue has to come from somewhere else.
His conclusion is that the money the endowment pays out gets more important from here, for a long time. "And so in this time and space, I think realistically for the next 10 or 15 years, the distributions that are coming off of endowment funds are going to be increasingly important."
Baylor's office has historically been good at the downside. He cited the first quarter of 2026, when the S&P was down 4% and Baylor was flat, and pointed back to the fourth quarter of 2018, the first quarter of 2016 and 2012.
The project of the last five years has been the other side of the distribution. Knowing the demographic problem was coming, the office reorganized to get better on the upside.
Stebbings asked whether a fund can do both. Morehead said finance faculty have laughed at him for trying. Baylor runs a value-centric, high-quality book on the equity side because equity beta is hard to control, and on hedging it with puts he said, "That's kind of like a money losing effort like over long periods of time."
The trade-off he is trying to escape is that a value book trails in a momentum market — and "the market's up 70% of the time." The answer he has been building for three to five years is convexity, structured so that Baylor is not paying a premium for the optionality on a normalized basis.
2. Funds of One
Asked what he meant by convexity, Morehead described going around the standard higher-education arrangement of handing the money to a list of commingled funds.
A commingled fund has one general partner and, in his framing, a hundred or a thousand limited partners. Every one of them gets the average risk-return profile the manager has to run to keep all of them satisfied.
Baylor's need at any moment may not be the average LP's need, so the office goes to the manager and asks for the same strategy run as a separate account for Baylor alone.
His words for the approach: "We need to like optimize our risk return profile better. If we give you a bunch of money, would you run the same strategy but do it just for us?"
The advantage is that Baylor can see into what is being added. His worked example is a manager who wants to add Nvidia: "We actually say no, we don't need more Nvidia." If the position would fill a hole instead, the answer runs the other way — Baylor will take the Nvidia and ask the manager to make it three times as big.
On how it has gone: "And it's actually worked like exceedingly well over the last two, three years."
3. Privates Exist to Make Money
Stebbings asked how Baylor thinks about portfolio construction from a blank sheet. Morehead said the office spends more time on that than on picking managers, which he acknowledged is unusual.
The order of operations is fixed: settle the private allocation before anything else, because the private book determines how much liquidity the whole fund has and how freely it can move between managers and strategies. Once set, it is very hard to change — secondaries and tweaks at the margin, and that is it.
Baylor's policy range for privates is 35% to 55%, and the target is about 45%. The 55% end is deliberately set so that a fall in public markets cannot push the private share high enough to force a sale.
"Like the number one thing to avoid is fraud, and the number two thing to avoid is forced selling."
In 2022, when technology slid, the private side reached about 51% or 52% — not enough to constrain allocation and nowhere near a forced sale.
Within the private book, the test is a single one. "The single reason that privates exist is to make money." Anything in there that will not keep up with the highest returns the private book can generate is being wound down — most of the real-asset holdings are running off and not being renewed.
What is left is venture, expansion and growth equity, and buyout. If capital is going to be locked up, he wants the highest return available for it.
On getting into the brand-name venture funds, Morehead was blunt that Baylor came later than the Ivy League schools and Stanford, and that it has not been for lack of trying: "It's just like when you knock on the door, they kind of like don't answer."
The category that has worked is the one his team already owns. "What I will say though, is that the ladies in our office have had exceptional, absolutely exceptional returns out of like the expansion growth equity category." The office has asked itself whether to move more dollars there.
4. Returns Need a Timeline
Asked whether he likes venture, Morehead said he does, and then made the objection that runs through the rest of the interview.
What bothers him is fund life. Historically venture funds ran 10 to 12 years; now he sees 15 and 18, "much to the chagrin of like all LPs." His reading is that general partner incentives are not aligned with the math that runs an endowment.
The arithmetic he gave: a 15x return over 15 to 18 years is beaten by a growth equity fund with a six-year weighted average life returning 3x, redeployed into another 3x, and then again — 27x over 18 years, which he put at better by a factor of two.
He understands why managers hold winners. Carrying a company another five years to turn a 3x into a 6x looks better in the marketing and helps raise the next fund. His position is that this is a business decision for the manager and not his problem: "I'm not optimizing for the best business for the GP."
The purpose he is optimizing for is the one that ends the argument. "Students can't pay their tuition with returns. They have to pay with dollars." He wants the largest pile of money, and the size of the pile is set by compounding.
So the thing he is buying is turnover, not multiple. "And so what we're really after is the velocity of capital, not just returns on capital." When the velocity starts flattening out, he wants to be out and into the next thing.
The office has made it a rule. "But I would say that there's a rule in our office that you're not allowed to talk about returns without also talking about time, because it's very common on the private side to just say everything like, well, you're up 2x, 3x, 5x, whatever."
His illustration: "If you're up 5x over 30 years, that's horrible. And if you're up 5x in five months, that's amazing."
5. 2.5% of Baylor in Anthropic
Stebbings put the obvious follow-up: if velocity is what matters, why do venture at all when growth equity delivers 3x in six years?
Morehead said the question gets batted around the office a lot, and that the answer is laddering — some managers return capital in one to three years, others in three to five, others in six to ten.
He was direct that venture is a diversification allocation for Baylor rather than the engine. Something in it can take off, and one thing has.
"We've had no exposure to OpenAI, but about 2.5% of the endowment is in Anthropic." He also said flatly: "We have no exposure to SpaceX."
Stebbings congratulated him and then told him to take the credit even though the position came through managers rather than the office. Morehead declined the advice — "That's not really how we roll at Baylor, but understood."
On what happens when a private position goes public, he said Baylor has no standing rule. It depends on the view of the name and the size of the position once it is listed: "We've sold shares before, we've also had shares before, we've also let shares run before."
6. Buying the Software Crash
Stebbings raised a story from one of Baylor's managers, Sean Barratt, who had told him Morehead thinks more like Charlie Munger than anyone he has met and that Morehead went deep on software when it was being sold off early in 2026 and then bought it. Morehead deflected the compliment — "Only because we're in the middle of the country, I think" — and then explained the process.
He located his own edge outside technology. "I would say, like, if we had an edge, I would say that we're pretty good on human behavior." He is not an engineer, and said much of what comes out of Silicon Valley is over his head.
The research was phone calls. The prevailing view was that software was dead, going to zero, and about to be replaced by code written on the fly, so he rang friends who run private family businesses of three to 500 people and asked whether they would tear out their customer systems. The answer was not in a million years.
His example is a friend who runs what he believes is the world's only vertically integrated potpourri maker: "He knows everything that there is to know about that, but he is not going to tear out key important parts of what makes his business run behind the scenes on some unproven thing."
The standard he thinks the sellers ignored is accuracy. He recalled the chief executive of Salesforce saying six or eight months earlier that the best AI would get to is 93% right, which he called phenomenal and possibly better than a lot of people — but "the issue of software is 100% right." If the books have to reconcile, 93% does not do it.
His forward view inverts the bear case: in vertical industries, software becomes the delivery mechanism for AI, because the trust already built with the software vendor is what gets the AI bought. And, he said, the software companies are not going to sit still and let a $20 billion or $50 billion business go to zero.
Stebbings pressed on the rarity of an institution acting rather than delegating, given software was down 50% to 60% from October of 2025. Morehead's caution was about what exactly he was buying: "The trouble is I'm not sure what's the baby and I'm not sure what's the bathwater."
So he bought through the manager rather than around him. He gave Barratt more money with an instruction: go through the list and own the names least likely to be interdicted by AI. His own contribution was the behavioral call; the manager's was knowing the ground.
Asked why he intervenes at all, Morehead said allocating is the job, and reached for the Berkshire comparison Stebbings had brought up — Warren Buffett and Charlie Munger deciding whether the incremental dollar goes to Burlington Northern or to the energy business.
7. Who Sets the Price
Stebbings asked whether the venture portfolio works as an intelligence channel on AI adoption and new technology. Morehead said it runs the other way, and then defended public markets against the charge of being a casino.
"I actually learn a lot from the public side managers." What he distrusts is the spun-up timeline: in 2016 the industry promised autonomous cars within three years. "Yeah, we're 10 years on. And what do we have, like 5,000 cars on the road? Like, please."
On whether public markets have become a casino, his answer was no. "The public markets are the big leagues. There's millions of people making decisions on dollars every single day for every single company."
The contrast is how a private mark gets set: three people in a room agree on a number and that resets the price for everybody. He said he has literally sat in those conversations: "I'll give you $50 million at that price. OK, fine."
He does not claim public prices are right. On SpaceX at what the market has decided is a $1.8 trillion business, and on Elon Musk carrying a premium of his own, his line was that the crowd's verdict is not a proof of value: "Yeah, that doesn't mean that they're right. It just means that it incorporates all available information, which does not happen on the private side."
"They can be irrational because they are governed by people. The difference is that there are tens of millions of people trading on that information, whereas on the private side, there's like three."
Asked whether he trusts the marks his managers send him, Morehead said the office works hard to keep them conservative, and explained why from his trading background. A trading book is priced every day so that psychology cannot drift: if a position is carried at $30 million, is worth $10 million and someone bids $20 million, the manager will refuse a bid that is a premium to fair value because taking it books a loss.
"And so pricing is just a way to make sure that you are psychologically aligned to the reality of the market."
His evidence that Baylor's marks run conservative is the gain in the six to nine months before a company is taken out. He put Baylor's average at 60% to 90%, against a market figure he gave as 30% to 50%.
He has to vouch for those valuations to the regents and the administration, and said he feels comfortable that Baylor's private marks are more sane than average.
8. Why Growth Equity Wins
Asked whether venture's dominance of the biggest companies in the world should change how he allocates, Morehead said no, and named the category he would rather own.
Growth equity is Baylor's largest private allocation, and the reason is the return timeline plus the loss rate. "The return timeline, there's also fewer zeros."
The mechanics are arithmetic: with fewer zeros, the winners do not have to carry the failures, which is what lifts the fund return. He put the growth equity book at annualizing about 30%, against a bogey of eight or nine percent.
On bad vintages — Stebbings raised 2021 and 2022 for venture and private equity — Morehead said it comes with the territory. Baylor sets the private allocation, splits it across private equity, expansion capital and venture in chosen sectors, and hands it to the team; the portfolio carries an expected return hurdle it has to clear.
9. What Cash Is Worth
Stebbings asked what the endowment has to produce in liquidity each year. Morehead split the answer in two.
The objective requirement is the payout, about 5% a year, and the dollar amount rises as the endowment does. That is what funds scholarships and professorships.
The subjective requirement is opportunity. He runs new analysts through it as a question: what are the odds we find something, anywhere, that goes up 20% in the next four years? The answer is very high, so holding cash is worth 5% a year on top of whatever the cash itself yields.
"So if cash is earning 3.5% plus 5%, so opportunity costs, cash is worth 8.5%." Anything Baylor can find above that hurdle gets done.
The cash balance is therefore a readout of the opportunity set, not a view. In the 2017 to 2019 stretch the office could not find enough to do and let cash build, going into the pandemic at 15% to 16% cash. Today the balance is low, because the office keeps finding things annualizing 20% to 30%.
Stebbings quoted Rudyard Kipling at him on keeping your head, and said discipline is hardest when momentum takes over.
10. Never All In
Asked for an allocation mistake, Morehead could not name a specific one, and instead described the lesson every trader learns and the mechanism Baylor built to survive it.
Everyone walks into the seat thinking it looks easy, and then loses money, sometimes a lot, sometimes for a long time. "I never want to be all in."
So the buying into a decline is mechanical rather than discretionary, which is the point — it takes the psychology out. Baylor thinks in 10% increments and sets liquidity up to be able to allocate every 10 percentage points down. Zero to 10% is normal and gets ignored.
"The way that I approach it with young analysts, I'm like, if something's on sale for 10 percent, do you rush out to the store to buy it?" They say no. At 20% they think about it, at 30% and 40% they buy.
Stebbings made it concrete against his own record, naming Wix and monday.com, where he could not separate the good from the bad, did nothing, and watched them fall another 10% to 30%. Morehead's answer was that he never draws a line: "We're like down 20, maybe I'm 20 percent in. Down 30, I'm another 20 percent in. Down 40, I'm another 20 percent in."
The cost is real. Baylor never gets fully invested before a rebound and leaves money on the table. The benefit is that it is never in the position of loving something, being down on it, and having no capacity left.
When they were buying software in February and March, the discipline was the same: down 50% or 60% is no argument against another 20% or 30% of downside.
Asked how to tell conviction from stubbornness when a position is under water, Morehead said Baylor does not trade individual names, so the work is on the manager's psychology. In the first part of the year he was on the phone with Barratt every day for four weeks, trading articles and news at all hours and pushing him on what he would do if each name fell another 20%.
The effect of that pressure is concentration, which is where the portfolio ends up anyway in a real drawdown.
Concentration does not worry him, because of what an endowment actually owns. "So like we own everything from like sunscreen to helium to like tech," plus consumer goods, business-to-business software companies he has never heard of, and real estate development projects. He said it is always funny to him when people compare an endowment portfolio with the S&P 500 — the endowment is far more diverse.
11. Hiring Out of Undergrad
Asked what his peers do that he finds strange, Morehead turned it around and described the thing Baylor does that almost nobody its size does.
Baylor hires its investment team almost exclusively out of the undergraduate ranks. He limited the claim to endowments of Baylor's size — "So we're about 2.7 billion. 14 months ago, we were 2.2 billion," and a couple of years before that, 1.4 billion.
The reason is geography. The office is in Waco, 100 miles from Dallas and 100 miles from Austin. "It'd be really difficult to pull somebody from LA or New York to Waco and say, like, I need you to be here for 10 years." An undergraduate has already chosen the school and the area, and the office screens hard for that.
The cost is the part he says he missed going in: for five or six years the new hires are levelling up, and during that time the work sits on him. He got the stable team he wanted and forgot he would be wearing all the hats while it formed.
He would not call his peers wrong, only differently exposed: hiring mid-career professionals removes the carrying cost but adds turnover risk. His own stability is concrete — he has worked with one colleague for almost 16 years and the next hire has been there 11.
"It's like longevity begets returns."
Stebbings asked whether the endowment model's incentives are broken, given a fund-of-funds manager earns carry and an endowment CIO does not. Morehead said the mechanism is not wrong, it just requires people who are missional.
"So like I wake up every morning motivated by sending some sophomore in high school to Baylor that hasn't even thought about college yet or some seventh or eighth grader who doesn't know if they're going to go to college and they're thinking about baseball scores from the prior night." Nobody on his team, he said, is motivated by the kitchen remodel.
12. AI and Human Thinking
Stebbings asked whether AI worries him as a threat to education. Morehead narrowed the question.
"I worry about the impact of AI on human thinking." He cited studies coming out of MIT and similar institutions suggesting students who use AI for everything show less brain function, while saying he does not know how sound they are.
His analogy is physical: sitting in a chair all day atrophies muscle, and the same logic applies to reasoning.
The distinction he draws is between the tool and the abdication. "Animals don't think, humans think. But if you abdicate your responsibility for thinking, it's not clear that humans do that either."
He expects education to figure out how to use it well. In his framing the problem is human discipline rather than technology.
13. The Endowment Tax
Asked how he advises peers hit by the endowment tax, Morehead said he envies them.
Baylor is not subject to it, because its endowment per student is too small.
He has run the conversation with Baylor's president, Linda Livingstone, in his head: the bad news is an endowment tax bill, the good news is that the endowment is three times bigger than the last time they spoke. "So I would love to have to pay the endowment tax because the endowment was bigger."
Asked whether he plays the comparison game, he answered with four words: "Comparison is the thief of joy."
14. Changing the Engine
Stebbings put Baylor's full-year 2025 return of 9.4% to him and asked whether he compares side by side or runs his own race. Morehead said both, and then explained the number.
Every school has different priorities. Baylor's is raising the endowment on a per-student basis.
He conceded the year was disappointing on a relative basis and gave two reasons. Baylor raised its annual commitment pace into privates by about 60% to 70% in 2020, 2021 and the years after, so the private book has been working through what he called a second J curve. A separate asset class the office had allocated to was flat and is starting to inflect up.
This fiscal year is the first without the J-curve drag and the first with returns from both the fund-of-one category and that other allocation. His newest analyst summarized it back to him as trying to change the engine while the car was moving, and he agreed: "We were trying to put a new, bigger engine in the car while it was still going down the highway. And we did it."
On the current year, his estimate was specific and did not depend on the headline private names: "I think we'll be 18 and a half, 19 percent this year without any SpaceX or Cerebras or anything like that." He said the next couple of years look good structurally.
15. $3M in Every Company
Asked about position sizing on the private side, Morehead described a change the office made after a note from a manager left him unimpressed.
The prompt was a good outcome that did nothing. A portfolio company sold at a 7x return, and Baylor's share of it was about $400,000. "It was like a 7X return. And I'm like, okay, great. What does that mean to us? And they're like, well, we'll get back like $400,000. And I'm like, what? Who cares?"
So Baylor now sizes from the bottom up, starting with the dollars it wants in each underlying company rather than the size of the fund. "But basically what we're saying is we want $3 million to be in each underlying company." Ten companies on the platform means a $30 million commitment.
Stebbings offered the conventional version — a $200 million fund, a $20 million commitment, 10% ownership per company, 1% exposure per company — and Morehead said Baylor does the same math in dollars. Eight, 10 or 12 companies at $2.5 million to $3 million each.
The test is materiality: at $3 million a company, a 5x gets $15 million back, and that is enough to matter to the fund. He was explicit that the office is not dictating portfolio construction to managers, only doing its own arithmetic.
16. Stay on Second Base
Stebbings asked whether he would rather a manager did what they said they would do or played the game in front of them, citing Bill Gurley's version of the argument. Morehead's answer was unambiguous.
He wants managers to do what they said. His analogy is a baseball general manager who hired for positions: "But if I ever walk out on the field and I have two second baseman and no third baseman, the third baseman is getting fired, like full stop, right?" The returns do not change that, which is why the office spends more time on asset allocation than on individual managers.
Stebbings tested it with the strongest case: markets moved, he moved with them, and he is making Baylor money. Morehead did not blink — "If it doesn't fit what we're trying to accomplish, we won't re-up." What he wants first is the conversation before the change.
His question in that conversation would be about evidence: why should you be able to do this when there is no data suggesting you are good at it?
The public-equity version is cleaner. Some managers say they cannot time cash and stay fully invested; others use cash as an allocation tool between 0% and 15%. Both are benchmarked the same way. A manager who said they are always fully invested and then turns up 10% in cash is fired, because "I don't want to be the guinea pig."
He drew the line where the strategy genuinely changes, not where a category label does. A manager who backed companies with product-market fit and now wants to fund two people in a garage is a different manager. Moving from Series B to late Series A is not.
Stebbings said he had expected to disagree and found they were aligned. Morehead's framing of the split is pre-data and post-data investing: selling the story, or showing the money. Some people are great at each, and the job is backing them for what they are great at.
On the convention that an LP must commit to three funds to judge a manager, Morehead did not push back. With one fund, and even two, there is not enough data to decide. He also conceded the shape of Baylor's own skill: strong where there is data to analyze, weaker where the pitch is a vision.
17. When Size Breaks a Check
Asked what he would do differently with unlimited money and Harvard's balance sheet, Morehead said probably nothing — and then described the problem he is trying to see coming.
Scale makes the job harder, and he has been asking other CIOs at what point an endowment has to change how it invests. He said it is very top of mind for Baylor and something a lot of allocators struggle with.
The data points he has gathered: Notre Dame is at $20 billion and expected to hit the wall at $10 billion or $15 billion and has not. He suspects the break sits somewhere between Notre Dame and the largest funds.
The mechanism is materiality, and he did the math out loud on a benchmark-quality outcome. "When you have $40 billion or $60 billion, and you can allocate $20 million to a fund, even if you're up like a real lot, doesn't move the needle as much as it used to."
His example was a $20 million check into a fund that returns 50x — another eBay — which returns a billion dollars and is still 2% of a $40 billion or $50 billion endowment. Stebbings joked about the Christmas card.
That, he said, is what the large multi-strategy venture platforms are built for. "Yeah, the platforms win. Like, just don't do those checks. Just give me 300 million bucks."
He does not love the billion-dollar-plus funds. He understands why they exist, but thinks the law of large numbers makes it harder to earn a return over the required period that pays for the risk.
18. Permits, Power and Europe
Stebbings observed that the conversation had been unusually unlike the AI-maximalist norm on the show. Morehead agreed and then gave the concrete evidence he sees in his own book.
The pushback on AI is showing up at the data center, and he sees it because the data centers are being built where he lives, not in Silicon Valley. He is invested in them, and he is watching a national and international resistance build.
The scarce input has shifted twice. "If we go back like five, six years, it used to just be land. And then it was powered land. And now it's actually permitted powered land."
The objections are power prices and water, particularly in arid states such as Texas and Arizona. He said solving water would go a long way with the average person, but power dispatch remains supply constrained, so household power prices go up until that is fixed over the next five, six or seven years.
The price effect is already in Baylor's portfolio: "So like literally we have this situation in our book where our data center sites are up 50 percent from where they were like six months ago."
Asked how many built data centers never get running, he declined to give a number he could not support — and then gave the tell. "But I will say that enough are not happening, that the power companies are coming to those who do have permits and saying, we can get you power sooner than we thought. That's literally happening."
The bottleneck is permitting, and he dated it precisely: it did not exist six months ago. The reason is electoral — "Because the permitting boards are governed by the citizenry and the citizenry is putting signs up in everybody's front lawn saying, we don't want this." Officials who want re-election say no.
On China, he said the constraint does not bind there because the state builds where it needs to. The harder market is Britain: "It's like by far a much bigger issue in the UK than it is in the US on the permitting front." Baylor owns a permitted UK site: "So we have a permitted data center site in the UK and it's worth a lot of money simply because we have a permit."
Asked if that makes him bullish on Europe, he said no, and listed the reasons without ranking them: "Yeah, because of all of it, because the defense structure of it, because of Russia, because of behind on AI, because, because, because."
It does not stop him allocating there. Baylor has money with long-short managers in Europe precisely because he expects winners and losers, and some of the fund's bigger macro hedges are on European indices.
19. The Quick Fire Round
Stebbings closed with short prompts.
What he has changed his mind on in 12 months: software, which the office leaned into hard. At about the same time Baylor took energy exposure off, with the arrival of the US-Iran war and the situation in the Strait of Hormuz, when crude went north of $100. In March and April the office added to private equity sponsors.
The overhyped asset class: "Private credit because it's easy." He said plainly that he is not in it and does not understand it.
His objection is the payoff shape. "Well, I think effectively what is happening is that you have credit exposure in companies that looks and acts a lot like equity to the downside, but you don't have upside equity returns." Baylor prefers the equity.
The endowment he most admires: "Brown, without question." He credited Jane Dietze and her team, said their 10-year returns are better than Baylor's while Baylor's may be better over five, and called them real investors who do things that take courage.
The fund he is not in and would most like to be: Benchmark. Stebbings said the same.
What he is most excited about: biotech over the next 10 years, which he expects to matter more than it has in the last 10 or 20. He was headed to a biotech conference that week and another in October, and the office is weighing whether it should own more.
His reason is the science rather than the market. "But what scientists are doing these days and actually like solving diseases as opposed to simply treating symptoms is extraordinary." He also noted the science is less correlated with markets.
Personally, what he is looking forward to is the office itself getting through the inflection between $1 billion and $5 billion, which he called a major one and said he did not appreciate until he was in the middle of it. The questions are how to grow a team, what systems let you keep track of everything at $5 billion or $10 billion, and how to systematize without losing the creativity that got the office there.
Bonus Insights
Stebbings opened by saying the reason for the episode is that venture investors rarely hear from the CIOs who fund them, and that the part of the industry needs more transparency.
Morehead said the office spends more time on portfolio construction than on manager selection, which he described as unique in the space. It is the sentence the whole interview runs on.
Stebbings twice described himself as the wrong person to be asking these questions — "I'm a low IQ individual after all" and "I'm dumb as rocks" — while pushing on velocity of capital, falling knives and the permitting bottleneck.
Morehead on being compared to Charlie Munger: "Only because we're in the middle of the country, I think."
Baylor's growth is recent and fast: about $1.4 billion a few years ago, $2.2 billion 14 months before the interview, and about $2.7 billion at the time of recording.
Asked what he would do with Harvard's balance sheet, Morehead's first answer was that he would not do anything differently, and his second was respect for the people trying: "So, hats off to Narv and like what his team is trying to do."
On being asked whether he would send a Christmas card for a billion-dollar outcome, he laughed it off: "I mean, like a billion dollars is great. But like, you see the point."
He signed off in the register he kept all hour: "We're down here in Central Texas trying to do a good job."
Morehead's bottom line is that an endowment is trying to build the biggest possible pile of dollars for students, that this makes the speed capital comes back more important than the multiple it comes back at, and that a venture industry stretching fund lives to 18 years is optimizing for its own business rather than his.
Products, Companies & Tools Mentioned
Anthropic (About 2.5% of Baylor's endowment, held through managers rather than direct; his single named venture position)
SpaceX and OpenAI (Baylor owns neither; SpaceX also his example of a public-style valuation, at what the market has decided is a $1.8 trillion business)
Nvidia (His worked example of how a separately managed account lets Baylor say no to more of a position it already has, or ask for three times as much)
Salesforce (He recalled its chief executive saying the best AI would get to is 93% right — his argument for why accounting software survives)
Benchmark (The fund he would most like to be in, and his example of how even a 50x outcome fails to move a $40 billion endowment)
Sequoia Capital, Founders Fund and Andreessen Horowitz (The brand names Stebbings raised; Morehead said Baylor came late and the door does not open, and that the platforms are built to take $300 million checks)
Cerebras (Named alongside SpaceX as the kind of headline private position Baylor's 18.5% to 19% year does not depend on)
Brown University's endowment (The peer office he most respects, under Jane Dietze; better 10-year returns than Baylor's)
Notre Dame, Harvard Management Company and UTIMCO (His reference points for where scale starts to change how an endowment can invest)
Thoma Bravo and Medallia (Named by Stebbings when he asked about bad 2021 and 2022 vintages)
Wix and monday.com (Stebbings' own example of falling knives he could not call, both down sharply before falling further)
eBay (The 50x fund outcome in his materiality math)
Burlington Northern Santa Fe (Part of the Berkshire allocation analogy for what an allocator's job actually is)
Books & Resources Mentioned
"If—" – Rudyard Kipling (Stebbings invoked it on keeping your head when everyone else is losing theirs, as the hard part of a mechanical buying rule)
If this was worth your time, send it to someone closer to the industry than you are.
Get the latest market chatter as it happens: