ABOUT THIS EPISODE
Miles Dieffenbach's path into venture started with getting cut from the Pittsburgh Steelers.
A team captain at Penn State, Miles signed with the Steelers as a priority free agent, played four preseason games, and was released - at 310 pounds, eating 7,000 calories a day, and paying for his own food for the first time in his life. He took the first job he could get, at CBRE, then moved to a real estate private equity firm, his first real look at private markets. When he saw a job posting at the Carnegie Mellon endowment, he didn't know what an endowment did. He applied anyway, got the job, and has been there for close to a decade.
Today he is managing director of investments at CMU, where he oversees a roughly $4 billion endowment with close to a quarter allocated to venture - a notably larger share than at many peer institutions. And he has run the math on where the asset class is headed. With fund sizes and the asset base up 10 to 15X over the past 15 years, he expects the beta of venture to be much worse going forward, and past performance to be a far weaker forward indicator than it used to be.
Nick and Beezer dig into how CMU underwrites every fund to a 4X net target, benchmarked against a public market equivalent that has compounded at 17 to 18% net since 1998; why a 2 and 20 fee structure on what is really large-scale growth equity can strip away the alpha; and why CMU is largely not active with multi-stage firms today. Miles also explains how he evaluates GPs, starting with whether they have their own cash in the fund, why it's nearly impossible to manage out AI exposure in any diversified portfolio, how secondaries work differently for seed funds and multi-stage firms, and why he believes true early-stage venture will always have a protected moat - and remains the innovation engine of the United States.
Quotes
“Are you putting your own cash into the fund? Which is, venture as an asset class seems pretty allergic to that concept. Would it hurt for you if this fund didn't do well? Would you experience the pain that we do, or are you just getting a free call option on our money? And if it hits, you win. If it doesn't, you've collected $30 million in management fees.”
“Venture has changed as an asset class from its scale perspective so drastically over a short period of time that it's very hard to look at those returns from 10 years ago and then pattern match that to ‘we're gonna expect those returns today.’ That's the really hard part of an LP job.”
Time Stamps
[01:02] Meet Miles Dieffenbach: From the Steelers to the CMU Endowment
[04:58] Points on the Scoreboard: Can You Hide in Venture?
[07:53] Why Venture's Beta Will Be Worse Going Forward
[10:07] How CMU Picks: Fat Right Tails and Entry Ownership Math
[10:50] The Anthropic Math: Why a Huge Outcome Won't Return a Mega Fund
[12:28] Underwriting to a 4X Net: The Public Market Benchmark
[13:56] Are LPs Subsidizing Bad Returns?
[16:30] 2 and 20 on the Qs: How Fees Strip Away the Alpha
[17:31] Long Feedback Loops and How AI Saved the Day
[19:09] Why Sovereign Wealth Funds Still Pay the Fees
[21:06] Missionaries vs. Mercenaries: Finding Great Pickers in AI
[22:25] GP Commitment and the Free Call Option
[23:42] Why CMU Won't Back a Manager for Just a Fund or Two
[25:41] When CMU Doesn't Re-Up
[26:46] Bigger Exits, but Not $20 Trillion IPOs
[27:39] Nowhere to Run: Managing AI Exposure at the Endowment Level
[30:34] Can Venture Be Bubbly While Public Markets Reprice Risk?
[32:32] Secondaries: Seed Funds vs. Multi-Stage Firms
[34:31] Liquidity, Opportunity Funds, and Conflicts of Interest
[37:02] Why Early-Stage Venture Still Has a Moat
[39:14] Venture as America's Innovation Engine
Links
Connect with the guest and hosts on LinkedIn!
Learn more about:
