Tariffs, Trump's Economic Endgame, Market Chaos, Bitcoin Reserve, CoreWeave IPO
2025-03-08 spoken.md · speaker-labeled ▶ watch ← E217 all episodes E219 →
Every number here is replayed from score_events — the same ledger the pool ranks on. Decay is what the 109 ideas nobody mentioned gave up this week; it applies only when an episode is processed.
Tier crossings
conviction thresholds crossed by this episode — 65 / 45 / 15 · ideas born here show where they landed. why 65 isn't always green
Kill dates that landed since E217
1 hit · 0 partial · 3 miss — windows that closed after 2025-03-01 and up to 2025-03-08, auto-scored against price data and never hand-set. verdict · R · α
| idea | verdict | R | α | closed |
|---|---|---|---|---|
| 🏛️ OpenAI's nonprofit-to-for-profit conversion becomes a tax and litigation overhang | MISS | +2.4% | -11.6 | 2025-03-08 |
| 📈 Peak Apple: a GDP-levered cyclical with no product optionality left | MISS | -40.7% | -54.7 | 2025-03-08 |
| 🪙 Spot ETF approvals extend past Bitcoin to Ethereum | MISS | -43.4% | -57.4 | 2025-03-08 |
| 🏛️ TikTok gets banned or force-divested and US platforms take the attention | HIT | +24.0% | +10.0 | 2025-03-08 |
Who moved the board
each voice's force on conviction this episode — supports and opposes, weighted exactly as the replay applied them · share = % of this episode's movement
What got argued (25 ideas)
ordered by how hard each idea moved · quotes are verbatim from the transcript, timestamps deep-link into the episode
Chamath's read of Bessent's Main-Street rhetoric and Trump saying he is not watching the tape: the administration is deliberately willing to let equities de-rate, because deflating asset prices deflates the margin-loan/wealth-effect consumption that is feeding inflation, and the resulting flight to quality out of stocks pushes the 10-year yield down exactly as Treasury has to roll roughly $10 trillion inside nine to twelve months - 100-150bp of savings measured in trillions of dollars. The tradeable expression is long duration with no bid under the equity market. Friedberg puts it at 60/40 that Trump is genuinely less sensitive to the market than in 1.0 and is listening to Bessent; Joe Lonsdale agrees lower rates are the target but wants to get there through AI productivity and spending cuts rather than by hitting assets; Jason does not believe Trump has stopped watching the stock market.
Well, if we are incentivized, if the government of America is incentivized to implement policies that crack the equity markets, it's actually really good in some ways. Number one is, if you deflate asset prices, you also deflate inflation.
So I think disinflation is very important, and I think we have to find some way to get there. Chamath may be right that it's worth hitting assets to get to disinflation, that's something that Scott could be working on because of all the debt
I do think that he is aware, and I would imagine the administration generally with Besant and others in kind of key leadership positions are trying to make the case that if we can get rates down, we have an opportunity to kind of refinance this $10 trillion that's coming due in the next 12 months and get ourselves into a kind of more sustainable financing position.
Friedberg's 2003 'speed doubler' analogy applied to the CoreWeave IPO: a neocloud rents access to scarce GPU capacity, but four or five hyperscalers are each spending roughly $80B of capex this year building exactly that capacity, so within a year or two the service gets bundled into GCP/AWS/Azure and the arbitrage closes the way broadband killed the dial-up accelerators. Customer concentration compounds it - about 60% of CoreWeave's revenue is Microsoft, which Satya Nadella has signalled may not be long-term usage. Chamath frames the equity separately as a pure bet on GPU useful life: the amortization schedule behind roughly $8B of debt assumes a long life, and if 10 years turns out to be 5 the business is 'deeply underwater' - though his net view of the company is complimentary. Joe Lonsdale takes the other side: the founders are commodity traders who locked down chips, data centres and power ahead of everyone.
It's really interesting what they did is they locked down the full supply of tons of data centers, tons of the power they needed, tons of the chips. They're very, very thoughtful. I mean, they're effectively, they're traders and they're very economic. And I think they have a lot of the stuff well.
And so I worry a little bit about a business like this, where there's four or five companies that are each doing 80 billion dollars of CapEx this year to create infrastructure that effectively starts to replace what these guys are effectively offering out as a service. ... If I was to do diligence on this business, that's where I would spend a lot of my time is like, guys, what's the capacity going to be in a year or two? Sort of like when broadband hit the Internet, you didn't need speed doublers anymore.
And what I would say is the other side of the tariff knife is that if there are markets, though, where you're making something that's fundamentally innovative, where you are but one maker, the problematic part of where tariffs really affect the US consumer is then the price of that product, of which there are no competitive alternatives, can go very, very high. And that's inflationary. And I think that that slows down consumption.
I buy LED lights in my greenhouse, and the price of the LED lights just went up by 25 percent. This week, I actually spoke with the CEO of an LED company, and I was like, why don't you guys make the LEDs here? And there starts to become a crossover point where it actually makes economic sense for the company to make the LEDs here instead of sourcing them from Asia. And there's 100,000 examples of this.
Here's where tariffs make a lot of sense. If you have markets where there are domestic alternatives and or where things are fundamentally commodity, there's no reason why tariffs can't work to create the incentives to re-domicile economic productivity inside of the United States. That's a slam dunk, I think.
In China and Indonesia and all these other countries, they are just shitting over the environment as they make things. I think tariffs are very reasonable in that case. It's not fair to make it more expensive for us because we're doing it well, and then we outsource it for them to destroy things.
Chamath's collision of two Republican priorities: a FERC report has almost 91% of December's incremental US generation coming from renewables, and the IRA's tax-incentive/tax-equity provisions are a roughly $200B market underwriting that build. A gas turbine ordered today is a five-year wait and nuclear is 2035, so a total IRA repeal strips out the only near-term source of incremental electrons and forfeits the AI race. Ways and Means therefore has to put the energy-credit portion back and the repeal ends up narrow rather than total - Trump's own 'time to get surgical' tweet is the tell - which lifts the repeal overhang off the renewables complex.
But the other part of the IRA, this narrow part is what it did to reinforce tax incentives and tax equity, which is a $200 billion market that incentivizes that 90% of energy generation. So my point is that when you start to get into the details, when the house, ways and means community has to figure out what part to put back, this is going to be hard because it's like, hold on, if you got the whole thing, now all of a sudden you take 90% of the incremental energy generation incentives out of the market, you don't have enough electrons.
Let's take a million workers out of the consulting class around DC, out of the paper pusher class around DC, and let's actually deploy them to the actual productive economy.
Listen, Trump's negotiating, right? ... He's using this to negotiate. ... And I think it's a reasonable thing to use to negotiate and force them to do that.
Tariffs aren't being done in isolation. They're being done along with a coordinated policy effort to reduce income taxes and another policy effort to reduce government spending. So those are three actions, three legs on a stool. So tariffs, reduced income taxes, reduced government spending, and they are related to each other.
And to Chamath's point, this has all got to become predictable. You cannot put tariffs on off every week or else how does your friend, Dave, who wants to do the LED lighting where you're encouraging to do LED lighting, know if they should build a factory and invest 10 million in that?
The individual governments of Europe are vibrant and powerful. The European Union itself was created almost without any real teeth. And so what happened is the folks there started to pass inordinate numbers of laws. And that has made it really complicated to be a European company, a European citizen. And that has to get sorted out.
I have someone who works with me, their spouse, was a real estate agent, and they are having a really tough couple of years because the industry has spiked up. If you get the industry down again, there are so many different places in America where people start making money again, with the title companies, with brokers, there are so many things that happen with that.
because the way it's worked for 20 years, like EPROS, you could just raise $250 million this week. It's a great company.
I mean, to me, collectibles are not securities, but if it's a collectible, you got to disclaim that, look, this has no intrinsic value, right? That's a collectible. Think about a baseball card. A baseball card is a piece of cardboard. It has no intrinsic value, but the value comes from basically other collectors being willing to buy it from you.
I think there should also be some nuance that you could work on specifically with this group of being clear that when you have a ticker symbol associated with something, or you do charts associated with something, you know, it starts to smell like a duck, it looks like a duck, it's quacking like a duck, but then in the terms of services, it says, hey, this is a collectible.
We had a big contest four years ago, L3, Raytheon, Northrop. These guys, in and out of government for decades, they've gotten tens of billions of dollars to the same technology area. When we went head to head with them, we didn't just beat them by a little bit. We shot down the hardened drones nine and a half times farther away. Same size, same power, nine and a half times. Completely wiped the floor.
And it's a really important point also that we should mention, is that the last four years, the economy has looked OK. But part of that is because government's been hiring like mad.
So we have the reserve and we have the stockpile. So the reserve is just Bitcoin. The goal is long-term preservation. Think of it as like a digital Fort Knox for digital gold. We want to put the digital gold in there. We want to keep it secure. We never want to sell it.
The direct employees of federal agencies is a fraction of people that are employed indirectly by government spending. This is super important. Because as we all know, many government agencies write checks to large contractors, subcontractors, and third party service providers that do the work for them.
If it was equal to what Elon was doing, it'd be at least five trillion dollars. And I push a bunch of them. They say, Oh, the Congressional Budget Office and all these things. I'm sure there's some tough things there. But like, this is crazy. We need to see what these cuts need to be on Doge. And we need to cut five trillion. And none of these people have the balls to do that. It's not the political will to do it, is my interpretation.
There is a bill that passed the last Congress in the House, but Biden basically, he didn't veto it, but he basically, he and the Democrats stopped in the Senate, the Democrats did. It was called FIT 21 And it was authored by French Hill, who's now the chairman of the House Financial Services Committee. So we expect that he will be introducing a new version of his bill, probably in the next few weeks. I don't think I'm breaking any news by saying this. I think people expect it. And that's going to provide the framework for market structure.
So when Alibaba pushes Gwen this week, it was an exceptional model. It's probably one of the better open source, if not the best open source model out there. Deepseeks I think is also quite good. It doesn't get any press anymore. So I think we're starting to get to this place where there's such abundance, that actually people are just kind of like overwhelmed with all the choice and they don't know how to differentiate.
We know that AI needs a tremendous amount of power. And whatever you thought you knew, actually, you had to re-underwrite because what Elon has shown is actually you need to now create these mega clusters, 100,000 GPUs going to a million GPUs. All of the power forecasting that we have is miscast. It doesn't even account for this, number one. Number two, there are 35,000 applications into FERC to get approved to generate electricity. 35,000, that's going to meander through some administrative rigmarole. There is a five-year delay to get a gas turbine into America and online.
Episode digest
written during extraction and stored in data/extractions/ep218.json — the auditable source of truth, including everything market-adjacent that did not earn a capture
Two-hour tariff/macro episode with Joe Lonsdale (8VC) in the guest chair for the first 1h37m and Sacks dropping in for the last half hour on the strategic Bitcoin reserve. The new thing is Chamath's 'Trump's endgame' thesis, coined here as administration-engineers-equity-drawdown-to-refinance-debt-2025: Bessent's Main-Street clip plus Trump saying he isn't watching the tape means the White House will tolerate an equity de-rate, because deflating asset prices deflates wealth-effect consumption and drives a flight to quality that lowers the 10-year just as Treasury rolls ~$10T inside nine months. Friedberg puts it at 60/40, Joe Lonsdale wants to get to disinflation via AI productivity instead, Jason doesn't buy that Trump has stopped watching stocks. On tariffs themselves the panel split four ways and Friedberg REVERSED his own E215 leverage thesis - he now frames tariffs as one permanent leg of a tariffs/tax-cuts/spending-cuts stool, cites his own greenhouse LED bill going up 25%, and says 'there's a lot of inflation because of tariffs' - while Chamath repeated his tariffs-are-real, commodity-markets-only position and Joe stayed on the fentanyl-negotiation read. Chamath's IRA rant produced a new idea (91% of December's incremental generation was renewable, tax equity is a $200B market, gas turbines are five years out and nuclear is 2035, so the repeal has to get surgical). CoreWeave got a new bearish idea from Friedberg's 2003 'speed doubler' analogy - four or five hyperscalers each spending $80B of capex commoditize the neocloud in a year or two - with Chamath framing the equity as a pure GPU-useful-life bet (10 years vs 5 = 'deeply underwater') but net complimentary, and Joe on the bull side. Sacks stayed substantive on the reserve: Bitcoin as digital Fort Knox that is never sold, the stockpile as discretionary Treasury portfolio management, acquisition only if budget-neutral ('I'm not saying that we will'), a FIT21 successor from French Hill in weeks, and memecoins as disclosable collectibles with no intrinsic value. NOT captured: (a) no 'sovereign BTC reserve creates a persistent bid' idea, because Sacks explicitly declined to commit to any buying - the EO only authorises budget-neutral accumulation; (b) Chamath's European leg - EU defence borrowing blowing out European sovereign yields and 'severe fiscal pressure' - has no clean US-listed expression (unhedged intl-Treasury ETFs are FX-dominated) so it stayed out of an idea; his EU-overregulation claim went onto the E187 socialist-policy-turn idea instead; (c) Friedberg's multipolar-abundance/exit-NATO case, the Inner Mongolia thorium reserve (60,000 years of energy) and AI-driven rare-earth discovery 'many orders of magnitude' argument are all decade-framed - noted here, not coined, and the rare-earth-abundance point cuts against the critical-minerals scarcity trade; (d) the Jason/Joe/Friedberg Citizens United and super-PAC debate has no tradeable edge; (e) Chamath's own rate-path idea got only a strength-1 support because he walked the level from 2.5% to '3%, 3.8%, 4%' while keeping the direction; (f) CoreWeave IPO pricing above its Nov-2024 $23B secondary is a live counterexample to E207's blocked-exit-window thesis but nobody argued it, so no oppose was recorded. DIARIZATION: labels are clean for the whole show EXCEPT the final 4.5 minutes - from the turn at 1:59:59 to the end of the file, label 'Joe Lonsdale' is actually JASON (he had signed off at 1:37:31; the turn says 'I know Hester. I've had her on This Week in Startups') and label 'Chamath Palihapitiya' is actually SACKS (that turn says 'my role as call it innovation policy advisor'). The one Jason mention at 1:59:59 is re-attributed on that basis. Also three stray one-liners labelled David Sacks at 53:07/53:23/1:15:33 predate his 1:37:33 arrival, one 'David Friedberg' one-liner at 1:41:35 postdates his exit, and SPEAKER_3 at 1:05:26 is a played Scott Bessent clip - all quarantined, none capture-worthy. Conflicts of interest to note: Joe Lonsdale is talking his own book on defence (Epirus/Anduril/Palantir) and on Grok/cognition; Jason plugged Superhuman (his portfolio) and disclosed a public GOOGL position; Sacks is the sitting AI/crypto czar and every crypto capture carries sacks_policy.