Add preferred source
Baylor University endowment chief David Morehead is breaking from the venture industry’s consensus in a way that should make every LP and GP pay attention. In a rare, candid interview on 20VC, the man overseeing roughly $2.7 billion argues that the venture capital industry’s drift toward 15-to-18-year fund lives has created a structural misalignment between what GPs need and what endowments actually require. The math, he says, is brutal: a 15x fund over 18 years is mathematically inferior to redeploying capital through three sequential 3x growth equity funds, which compounds to 27x in the same period. This “velocity of capital” framework drives everything from Baylor’s preference for growth equity over venture, to its use of “fund of one” structures that let it customize risk exposure, to its dollar-based position sizing that demands every investment actually matter to the bottom line. Morehead also defends his contrarian bet on software, arguing AI’s 93% accuracy ceiling means incumbent providers become the delivery mechanism for AI rather than its victims — a call he made by phoning owners of mid-sized family businesses, not by reading code. For LPs, the interview is a blueprint for applying compounding math to a private portfolio. For GPs, it is a warning that the smartest allocators are doing the arithmetic on fund duration and will vote with their capital.
Key Elements
The Enrollment Cliff and Why Privates Exist
David Morehead runs money for an institution facing a slow-moving demographic problem. Fewer American high school graduates — a hangover from the Great Financial Crisis — combined with a harder visa environment for international students means the incoming class of 2030 missed enrollment targets at schools across the country.
The consequence is simple, and it shapes every decision in Baylor’s portfolio. Endowment distributions become increasingly important revenue. Morehead expects this pressure to hold “realistically for the next 10 or 15 years.”
Baylor’s annual distribution rate is about 5%, and the dollar amount keeps rising — because it pays for scholarships and professorships. That creates an uncompromising mandate. As Morehead puts it:
“The single reason that privates exist is to make money. Period. End of story.”
That principle has led to a hard choice. Baylor is winding down and not renewing its real asset allocations. If you’re going to lock up capital in illiquid private markets, Morehead reasons, it has to be in the categories with the highest return potential: venture capital, expansion and growth equity, and buyout. Everything else is dead weight.
“Students can’t pay their tuition with returns. They have to pay with dollars. So, I’m expressly interested in creating the largest pile of money.”
The Compounding Math That GPs Don’t Want LPs to Do
The sharpest argument in the conversation — and the one most likely to make venture fund managers uncomfortable — is about fund duration. Typical venture funds that were historically 10 to 12 years have stretched to 15 to 18 years.
Morehead is “a little perplexed” by this. And he’s not sure the incentives are aligned.
The worked example is devastating in its simplicity. A phenomenal 15x outcome over 15 to 18 years sounds like a home run. But compare it to a growth equity fund with a six-year weighted average life returning 3x. Redeploy that into another six-year, 3x fund, then do it again. Over 18 years, that compounds to 27x — roughly double the vaunted 15x venture outcome.
| Strategy | Return Profile | Terminal Multiple (18 Years) |
|---|---|---|
| Long-duration venture fund | 15x over 15-18 years | 15x |
| Sequential growth equity (3x per 6 years) | 3x, redeployed 3 times | 27x |
The culprit, Morehead argues, is GP marketing incentives. Holding a winner another five years to turn a 3x into a 6x looks better on the pitch deck for the next fundraise. It is a rational business decision for the GP. It is not, however, aligned with the endowment’s math.
“I’m not optimizing for the best business for the GP. I’m trying to optimize for the biggest pile of money for our students.”
There‘s a rule in the Baylor office: you are not allowed to talk about returns without also talking about time. A 5x over 30 years is horrible. A 5x in five months is amazing. This single discipline explains why Morehead prefers growth equity over venture: the winners don’t have to cover for as many zeros, and the manager book annualizes around 30%, easily clearing Baylor’s roughly 8% bogey.
Why Baylor Uses “Funds of One” Instead of Commingled Vehicles
The mechanism for fixing the misalignment is a structure most retail investors never see: the “fund of one.” Instead of joining a commingled fund where the GP manages money for a hundred LPs and delivers an average risk-return profile, Baylor goes directly to the GP and asks for the same strategy run just for them — with a look or call into what goes into the portfolio.
The Nvidia example makes it concrete. If Baylor already holds plenty of Nvidia and the next marginal manager wants to add it, Baylor can decline. Conversely, if Baylor holds none and wants the exposure, it can ask the manager to make that position three times larger. The result is a portfolio back-fit to Baylor’s specific needs, not the average LP’s.
Morehead says this approach has worked “exceedingly well” over the last two to three years. It is the structural solution to the velocity problem — a way to optimize for dollars, not for headline multiples.
Position Sizing in Dollars, Not Percentages
The same dollar-focused logic governs how much Baylor commits to each underlying company. Morehead describes receiving a GP note about a company that sold for a 7x return. Baylor’s share was $400,000. His reaction: “who cares?” A 7x on $400,000 is meaningless to a $2.7 billion endowment.
So Baylor now starts with the end in mind: roughly $2.5 to $3 million per company. If a manager will hold 10 companies, Baylor allocates about $30 million. If a company goes 5x, Baylor gets $15 million back — “that’s enough to matter.” The discipline is mechanical, and it keeps the office from celebrating wins that don’t move the needle.
Buying the Software Selloff with a Behavioral Edge, Not an Engineering Edge
The most vivid case study involves the early-2026 software drawdown, when the sector fell 50% to 60% from its October 2025 peak. The question was whether AI would simply replace enterprise software — a thesis that had sent valuations spiraling.
Morehead’s edge, he admits, is not engineering. “Much of the stuff that comes out of Silicon Valley is over my head,” he says. His edge is human behavior — knowing how people think and make decisions.
So he tested the thesis with phone calls. He called friends who run 300-to-500-person private family businesses and asked a simple question: would your son-in-law vibe-coding something lead you to tear out your CRM? The answer, universally, was “not in a million years.”
He cites a friend who runs what he believes is the only vertically integrated potpourri maker in the world — someone who knows everything about that business and will not rip out key systems for an unproven tool. He also invokes a comment attributed to the CEO of Salesforce roughly six to eight months prior: the best AI would be is 93% right.
“The best that AI was going to be is like 93%, which is phenomenal and might be better than a lot of people, but the issue is software is 100% right.”
Books have to match up. In niche vertical industries, trust compounds. The SaaS companies are not going to let $20 to $50 billion of market value go to zero without responding.
“I actually think that in some of these vertical industries that software is actually going to be the delivery mechanism for AI.”
The action followed the analysis. Morehead was on the phone with his software manager, Shawn, roughly every day for four weeks during the first part of 2026, trading names, articles, quotes, and news stories at all hours. The instruction was not to buy the dip indiscriminately, but to identify the names least likely to be interdicted by AI and concentrate there.
The allocation framework, however, kept emotion out. Baylor treats 0% to 10% market declines as normal for an infinite-life portfolio and doesn’t care. The office thinks in 10% increments, with liquidity set up to allocate at every 10 percentage points down. The cost is never being fully invested before a rebound. The benefit is never being the investor who loves a down asset and lacks the cash to buy more.
“I never want to be all-in. Things can always get worse.”
The Public Markets Are the Big Leagues
Morehead is blunt about price discovery. On the private side, as he describes it, “three people get in a room and say, ‘Hey, I think the value is X,’ and they’re like, ‘I’ll fund it at that,’ and that resets the whole price.” Public markets can be irrational because they’re governed by people, but tens of millions of people trade on that information. On the private side, there are three.
That conviction shapes how Baylor pushes managers to mark their books. From his trading background, Morehead knows that mismarked assets distort decisions. If you believe an asset is worth $30 million when it’s actually worth $10 million, and someone offers $20 million, you’re liable to refuse — even though it’s a premium to actual value.
The evidence suggests Baylor’s managers are more conservative than average. In the six to nine months prior to an asset being taken out, Baylor’s average gain is roughly 60% to 90%, versus a market average of 30% to 50%. That gap implies Baylor’s marks are saner than the industry norm.
Team Building, and the $1B-to-$5B Inflection
Baylor does something few endowments its size do: it hires almost exclusively from undergraduate ranks. The reason is geographic. The office is in Waco, Texas — 100 miles from Dallas, 100 miles from Austin. Pulling a mid-career professional from Los Angeles or New York and getting them to stay for a decade is a hard sell.
The cost is real: for the next five or six years, senior staff pour time into training that person while wearing all sorts of hats. The benefit is stability. Morehead has worked with one colleague for almost 16 years; the next hire has been there 11. Across the industry, he observes, longevity begets returns.
Baylor is currently in the middle of what Morehead calls the $1 billion-to-$5 billion transition — a major inflection point involving how to grow a team, what systems to set up, and how to systematize into a self-perpetuating office while retaining the creativity to do new things. He did not appreciate how much decision-making happens in this range until living it, and he expects Baylor to get out the other side in two to three years.
The Forward Views: Biotech, Data Centers, and Europe
Morehead’s most-excited area is biotech, which he thinks will be more impactful over the next 10 years than the last 10 or 20. The science, he says, is less correlated with markets, and scientists are actually solving diseases rather than treating symptoms. Baylor has substantial biotech exposure and is actively considering whether to have more.
On data centers, he is watching a permitting bottleneck in real time. The most valuable attribute for a data center used to be power, then land, then powered land. Now, it’s permitted powered land. Data center sites in Baylor’s book are up 50% from six months earlier. The pushback is local: permitting boards are governed by citizens putting signs on front lawns. Officials who want reelection say no.
That bottleneck creates an asymmetry Baylor is positioned for. Enough projects are failing to proceed that power companies are coming to those who do have permits and offering power sooner than expected.
On Europe, he’s bearish — permitting is worse, the defense structure is strained, Russia is a factor, and the region is behind on AI. But that doesn’t preclude allocating to European managers. Baylor has long-short managers in Europe precisely because some companies will win and some will lose. Some of its bigger macro hedges are on European indices. And it holds a permitted data center site in the UK worth a lot of money simply because it has a permit.
What Sophisticated LPs Actually Want
For allocators and GPs alike, the interview serves as a signal of where institutional capital is heading. The demand for fund-of-one structures is a direct response to the average-LP trap. The willingness to fire managers who change their stripes — the baseball analogy is instructive: a third baseman who wanders to second base gets fired, “full stop,” regardless of returns — is a warning that fit matters more than raw numbers. And the math on duration is relentless: if your fund is a 15-to-18-year vehicle whose incentive structure rewards holding winners for the next fundraise, the largest and most sophisticated endowments are doing the compounding arithmetic and will allocate accordingly.
The paradox Morehead leaves unresolved is his own. He concedes his edge is human behavior and public-market information processing, not technology — yet his largest single position is Anthropic at roughly 2.5% of the endowment, and the most consequential macro force in the book is AI capex and data-center permitting, areas where he explicitly defers to managers and disclaims expertise. That tension — between behavioral conviction and technological deference — may be the most honest reflection of what it means to allocate institutional capital in 2026.
Full content available at:Why Velocity of Cashback is the Most Important Thing | David Morehead, Baylor University CIO
- Kioxia Weighs US ADR Listing to Raise at Least $10 Billion, Earliest Debut Next Year
- US Chip Stocks Plunge Across the Board; Nasdaq Opens Down More Than 1%
- Google-affiliated AI startup Discovery Loop reportedly launches new funding round, targeting $50 billion valuation
- Anthropic CEO Warns Rogue AI Agents Could Seize the Internet Within Months
- Anthropic Posts Adjusted Operating Profit for Second Straight Quarter, Reportedly Selects Nasdaq for Listing
Once added, BigGo Finance appears first in Google Search Top Stories, so you get the broadest, most up-to-the-minute, and most comprehensive global financial news first.
