May 23, 2022 · 51m · allin
E81: All-In Summit: Bill Gurley & Brad Gerstner on markets, downturns & investment cycles
Brad Gerstner · 15m spoken Bill Gurley · 12m spoken Chamath Palihapitiya · 7m spoken Jason Calacanis · 5m spoken David Friedberg · 2m spoken David Sacks · 2m spoken
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
At the All-In Summit, venture capitalists Bill Gurley and Brad Gerstner join the All-In Podcast hosts to deliver an in-depth analysis of macroeconomic cycles, interest rates, startup valuation discipline, and venture capital dynamics following the tech downturn.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. The hosts hold 37.9% of the talking time here. How this is scored →
The hosts as informed peer 4.8 Guest teaching 5.3 Guest disagreement 2.6 The hosts pushing back 2.7
speaking balance: gold is the hosts, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 38:30 Brad Gerstner rejects paying 75x ARR
Brad forcefully rejects late-stage fund valuation practices, stating someone would need to pry dollars out of his hand with a crowbar before he paid 75x ARR.
Hardest push from the hosts ▶ 24:14 Friedberg pushes Gurley on VC valuation disciplineFriedberg directly challenges Gurley on why VCs refuse to underwrite lower valuations or enforce market-driven down rounds.
Biggest teaching moment ▶ 22:16 Brad Gerstner reveals software revenue realitiesBrad educates the panel by revealing that only 21 public software companies ever surpassed $2B in revenue, proving 100x ARR multiples were built on false assumptions.
The host holds their own ▶ 17:00 Chamath breaks down public SaaS multiplesChamath demonstrates host domain authority by citing Morgan Stanley data on compressed SaaS multiples (5.6x-8.5x) to prove private dry powder deployment assumptions are flawed.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | The hosts as informed peer | Guest teaching | Guest disagreement | The hosts pushing back | Why |
|---|---|---|---|---|---|---|
| Bill Gurley on Market Cycles and the Sawtooth Pattern | 2 | 5 | 1 | 1 | Bill Gurley explains Howard Marks' view of VC structure and details how market downturns follow an abrupt 'sawtooth' risk-off pattern rather than a smooth sine wave. Jason and Chamath mostly listen, offering light banter about company stadium naming rights. | |
| Brad Gerstner's Macro Presentation on Rates, Inflation, and CPI | 3 | 6 | 2 | 1 | Brad Gerstner delivers a macro data presentation using bank models to demonstrate that CPI is backward-looking and forward inflation is rolling over. Hosts interject with brief jokes, while Chamath notes that they previously flagged these economic shifts. | |
| VC Dynamics in Speculative Bubbles and the Upside Dilemma | 7 | 5 | 3 | 4 | Chamath actively drives the discussion on public market terminal valuations, noting that SaaS multiples have compressed to 5.6x-8.5x. Brad and Bill challenge the notion that $250B of VC dry powder will be deployed into high valuations. | |
| Shift from Pure Software to Physical Assets and Valuation Discipline | 6 | 6 | 2 | 3 | Chamath describes shifting fund allocations into hard assets like lithium mines while Brad Gerstner schools the panel with data showing only 21 public software companies exceed $2B in ARR. Bill Gurley emphasizes discounted cash flow and free cash flow over crude revenue multiples. | |
| Founder Power Dynamics, Governance, and Down Rounds | 4 | 6 | 3 | 4 | Friedberg presses Gurley on why VCs avoid down rounds and continually inflate valuations. Gurley reframes the dynamic, pointing out there is no VC club price-fixing and explaining how founder power expanded during the bull run. | |
| Liquidity Strategies: Distributing Stock vs. Holding Post-IPO | 6 | 5 | 2 | 3 | Chamath and Sacks openly admit to making massive post-IPO holding mistakes with Slack and Affirm stock. Brad Gerstner details Altimeter's strict LP distribution rules, emphasizing disciplined capital return. | |
| Media Portrayals vs. Realities of Tech Founders | 5 | 7 | 5 | 3 | Sacks asks about hedge funds gaslighting the market into 100x ARR valuations. Brad Gerstner forcefully attacks late-stage herd mentality and criticizes bank CEOs panicking on CNBC without building proper models. | |
| Early-Stage Advantages and Subsidized Unit Economics | 6 | 6 | 3 | 3 | Gurley highlights the post-correction calm as the best window for early-stage investing, warning that 49 out of 50 negative unit economic experiments fail. Chamath explains consumer surplus businesses created by massive capital subsidies. | |
| Market Predictions, Bill Gurley's Future, and Panel Conclusion | 4 | 2 | 2 | 2 | The panel discusses market outlooks and Gurley's step back from Benchmark into angel and public investing. Chamath closes the segment with a comedic impersonation of Bill Gurley at the poker table. |
Statements from this episode (21)
Gurley: VC market cycles resemble a sawtooth, not a sine curve
“It doesn't happen like a sine curve, which is what we all imagine when we think of a cyclical business. It's more like a sawtooth. Risk, risk on is a very slow process, and it, and it's reflexive, so it grows, and grows, and then risk off tends to be very abru…”
Gerstner: A 1% rate change drives a 15-20% multiple shift
“The iron law of investing is interest rates. A one percent change in rates leads to a 15 or 20% change in a multiple.”
Gerstner: Used car prices will be lower by May 2023
“For two months in a row, we've had sequential declines. You tell me, is the price of the used car this time next year higher than 29 or lower than 29? It's going to be lower because we're destroying demand by raising interest rates.”
Gerstner: $15 trillion of household net worth destroyed in five months
“We destroyed 15 trillion. You said this on pod 80. We destroyed 15 trillion of household net worth in the last five months.”
Gurley: VC returns depend heavily on peak market cycle performance
“What I realized was that the IRR numbers and the ROI numbers on the venture capital category were heavily dependent on performance in the hottest That's part of the cycle”
Friedberg: $250B in committed VC capital remains unallocated
“There's two hundred and fifty billion dollars of committed capital unallocated into companies.”
Gerstner: Venture capital investments from 2020-2022 will produce lousy returns
“I think the vintage of the last 18 months will be lousy. So the capital deployed over the last 18 months won't have a lot of return.”
Gerstner: Tech valuations will not return to 2021 peaks without zero rates
“Pretend you never saw them, not in venture, not in the stock market, because that is a delusional place to think we're getting back there. We're not, unless we have another pandemic or a nuclear war and rates go to zero, and then we have bigger problems.”
Chamath: Only 30 SaaS companies globally are growing revenue over 50%
“There's 30 companies in the SaaS index that grow, but only 30 in the entire world that grow above 50%.”
Gerstner: VCs will only deploy 25-30% of dry powder over three years
“25 or 30%. Depends on price adjustment.”
Gerstner: Only 21 public software companies have over $2B in revenue
“And public software companies over two billion in revenue. We got to 21. Ok. There are only 21 that are worth more than twenty-five billion dollars in all the public markets of all the millions of software companies that have been started.”
Gerstner: Startups valued at 100x ARR will likely never reach those prices again
“If you were slapping a hundred X ARR revenue, or multiple, on a company doing fifty million in revenue, It's highly likely that they will never see that price again, whatever you paid for that asset, because the dilution and the deceleration and their growth r…”
Sacks: Craft's $100M mistake was holding Affirm stock post-IPO
“In our second fund, we had about a hundred and twenty million of a firm stock, and we were sitting on it because we believe in the company and still do, and we're still sitting on it. And that was like a hundred million dollar mistake.”
Gerstner: Altimeter distributed over $6B to LPs in 2021
“Last year, we distributed over six billion dollars, which was more than all the venture we raised in our first five funds.”
Gurley: Benchmark distributes public stock over 3 to 6 quarters post-lockup
“We typically distribute over three to six quarters following the lockup release.”
Gurley: Super Pumped TV series was inaccurate and fabricated scenes
“Super Pump was not accurate just because they made up a lot of scenes. Like, Drummond wasn't very active at all, but he's in a lot of the scenes, so a lot of them were made up.”
Gerstner: Altimeter will not join startup deals at 75x ARR
“If somebody calls me up tomorrow and says, Hey, Tiger's doing this deal at 75 times ARR. Do you want to do it? They would have to pry the dollar out of my fucking hand with a crowbar. I'm not risking my money or my partner's money.”
Gerstner: Post-COVID inflation will roll over and long-term trends will resume
“I think the much more likely explanation is that the trends that existed for 20 years still exist. They were interrupted temporarily by us saving ourselves from a catastrophe with COVID. The transient thing that everybody has come to make fun of, right? Transi…”
Gerstner: Short everything if 10-year yield hits 7% and inflation stays at 5%
“If you told me you thought inflation was going to be five percent for the next decade and the tenure was going to seven percent, I would say short every tech company in the market. No, no, no. Short everything in the market and don't invest a dollar in venture…”
Gurley: 49 out of 50 startups using negative unit economics fail
“I think if 50 entrepreneurs try that trick, 49 are gonna auger in.”
Gerstner: Tech growth stocks will be higher by May 2023
“We will be higher for growth stocks this time next year, but we may very well get there by way of lower and potentially meaningfully lower because the counterfactual to the hyperinflation argument is not, you can't deliver the counterfactual for at least four …”