All-In PodcastE136: Hacking the pod, Threads launches, Fed minutes, immigration, balloon farce, heart health
EVERY SPOKEN WORD
150 min read · 30,038 words- 0:00 – 5:34
We hacked the pod! Jcal is on vaca
- DSDavid Sacks
Let me get everything queued up here. How does J-Cal open the show? What does he say?
- BGBrad Gerstner
Hey, everybody.
- DSDavid Sacks
Hey, everybody. Yeah, do it, Chamath.
- CPChamath Palihapitiya
Hey, everybody. Hey, everybody. I'm Jason Calacanis. I'm the grifter with the mostest, the mostest that's the shortest, the shortest that's the fattest, and the fattest that's the dumbest.
- DSDavid Sacks
(laughs) Oh, God.
- CPChamath Palihapitiya
All in with you or me.
- DSDavid Sacks
He's not even here, you can't do that. It's awful.
- BGBrad Gerstner
Oh, my God.
- DFDavid Friedberg
With you or me. With you or me.
- DSDavid Sacks
That's just bullying, it's awful.
- DFDavid Friedberg
The best deals are at All In.
- JCJason Calacanis
I'm going all in. We'll let your winners ride.
- CPChamath Palihapitiya
Rain Man David Sachs. I'm going all in.
- BGBrad Gerstner
And they said we open sourced it to the fans, and they've just gone crazy with it.
- DFDavid Friedberg
Love you, SNL.
- JCJason Calacanis
A queen of quinoa.
- CPChamath Palihapitiya
I'm going all in.
- DSDavid Sacks
Great, great, great open, Chamath. Uh, welcome to the All In Pod.
- BGBrad Gerstner
Where is J-Cal? Where is J-Cal?
- DSDavid Sacks
He wanted the week off, and any time any of us take the week off, J-Cal says, "The show must go on," and he rings up Brad Gerstner and says, "Hey, we need a sub, come on in this week." This week, when J-Cal wanted the week off, he said, "Guys, we're all taking the week off." And Sachs said, "The show must go on." J-Cal refused to show up, his producer/editor refused to show up, so we are here solo hacking our way to All-In Pod episode 136. As your moderator today, Dave Friedberg, I am extraordinarily joyous and happy to bring you the first episode of the All-In Podcast without the hostess with the mostest, Jason Calacanis. Joining me today, Il Duce from Elba Island, Chamath Palihapitiya. Chamath, firing up tweet storms lately, taking over the Twitter, soon to take the threads by storm, I'm sure. Rain Man David Sachs, joining us from a curtain showroom in the south of wherever.
- BGBrad Gerstner
(laughs)
- DSDavid Sacks
And, from an attic in an old house, the man who manages $10 billion to generate 50% plus returns year to date, the one and only Brad Gerstner. Brad, welcome to the show. Great to have you.
- BGBrad Gerstner
Good to be here.
- DFDavid Friedberg
Whoop, whoop.
- BGBrad Gerstner
Appreciate the call half hour ago.
- DSDavid Sacks
Yeah, great. Thanks for-
- BGBrad Gerstner
Are we, are we even gonna be able to drop this episode? How are we gonna actually upload it to Apple Podcasts and all the rest?
- DSDavid Sacks
Okay. So here's the deal. I have the email login. I think I can get into the accounts.
- CPChamath Palihapitiya
(laughs)
- DSDavid Sacks
I h- I have... I'm gonna, I'm gonna like do the request for password reset on all the accounts.
- 5:34 – 17:15
Meta launches Threads
- DSDavid Sacks
well, let's kick it off. Um, first, uh, topic on the docket today, which I think is a great one, and Chamath, you've been tweeting a lot lately as we just talked about. I think that'll end up probably being our cold open. Um, but, uh, Zuck announced, uh, and Facebook, Meta announced the launch of Threads. Uh, Gerstner, big shareholder of Meta I think, right? Reasonable-
- BGBrad Gerstner
Yeah.
- DSDavid Sacks
Reasonably sized shareholder. Uh, this, uh, Zuck this morning announced 30 million downloads of the app overnight. Threads is Instagram/Facebook/Meta's-... competitor to Twitter. Looks almost like a clone. I mean, it is so similar in features, in interaction, in everything. Brad, maybe you can kick us off and talk a little bit about the importance of Threads and how this, um, is important to Meta, and then we can talk a little bit about the app itself.
- BGBrad Gerstner
Well, I mean, eh, the, the first thing is, if rumors are true, they developed this over a period of six to nine months with 20 people, okay? And so the-
- DSDavid Sacks
Wow.
- BGBrad Gerstner
... extraordinary pace that's now occurring with inside, uh, or inside Meta. We've heard from several people inside that said people are pumped. They're actually updating real-time, uh, user accounts. There are over 30 million users already on Instagram. And it just gets back to this culture of flattening the organization, speeding up the organization, and I think everybody's excited. You know, the best engineers in the world wanna see their products set free. And so, you know, I think Threads is really interesting for a couple of reasons. Number one, if we flash back to what went on in China with Toutiao and Doyan, remember that text-based social networks were where kind of it started there. And so it's a bit anomalous. Doyan, which is the TikTok equivalent in China, was started based on Toutiao, which was a text-based news feed. And so it doesn't surprise me to see the social graph being leveraged into a text-based, uh, feed. I think initially, they're seeding it obviously with all your friends so it's, it's, uh, it, it looks more like celebs and entertainment and it's m- my team is like, "It's easier to use. It's faster. It feels more free." Um, I think, you know, Adam Mosseri, he, he posted on Threads that he stayed up all night last night, um, so excited with the product launch. He and Zuck are responding to people actually live on Threads. Um, and so I, part of this just goes to the pulse of the company. Over the next four to six weeks, they're gonna have massive launches out of, uh, the AI side of the business.
- DSDavid Sacks
Yeah.
- BGBrad Gerstner
I expect a lot more support for their OpenAI model, um, or for their open, uh, LLM. I also think they're gonna launch a bunch of agents on WhatsApp and Instagram. So, what I see is just velocity and cycle time within the business improving. And then one final thing. In the age of AI, right, you want to collect as much data as you can from your users. And we know that Facebook is very heavy on video and pictures. But what they're light on is text. And so in a world where the most important thing on the internet to train your model is the s- is the word, right, now they're gonna collect a lot of words, a lot of conversations, a lot of sentiment among the users. Um, and, you know, maybe we'll come back to this. Uh, you know, I just wanna kick it back and get people's thoughts on the product, but, um, already, by my estimate, if these guys get to 100 million, which is, looks like they may be able to get to tonight, based upon the monetization that Twitter has, that's already a business that's doing something like two and a half billion dollars in revenue if they chose to monetize it, 1.2 or 1.5 billion in EBIT. Apply their current multiple to that. That's about a $20 billion increase in enterprise value built by, if rumors are true, 20 people over the course of six months.
- DSDavid Sacks
Yeah, it's pretty impressive. I remember in the early days of Friendster and MySpace and then even Facebook, these social networks launched and everyone thought it was gonna be a conversational system. And so much of the usage and the page views and the minutes spent ultimately accrued to photos and, uh, over time, video. Um, and now it's almost like this interesting reversion back to the origin that it's back to this conversational system. But these systems always seem to kind of evolve to, hey, just images are what win and what, uh, kind of gather up all the, the mind space. I mean, Chamath, having worked at Facebook, maybe you can share a little bit about your point of view on Threads as a product and, um, as an evolution away from social network on Facebook to Instagram and now to Threads.
- CPChamath Palihapitiya
I, I think what people don't understand is that the successful category winner in each of these categories needed to invent something de novo that nobody else had. In the case of Facebook, we invented photo tagging when it didn't exist, and then we invented the news feed when it didn't exist, and then there was a bunch of people that copied it, but it didn't matter because we had refined and owned that use case already. So, I think the real challenge for Facebook isn't, can 20 people copy Twitter? I mean, you know, Mastodon is a copy of Twitter. There's a bunch of... Parler's a copy of Twitter. Truth Social is a copy of Twitter. The challenge is going to be, can you invent some de novo feature that makes people actually want to use this? And, you know, h- rage quitting Twitter because you're not a fan of the product right now or Elon isn't a successful long-term use case, because all of those people will eventually come back. So, I don't know. I'm a little bit more skeptical of the whole thing. I think that right now, what you see is not even a full copy of the system. It's-
- DSDavid Sacks
Mm.
- CPChamath Palihapitiya
... you know, a 50, 60% copy with a lot less usage, and so as a result, a lot less traffic.
- DSDavid Sacks
Mm.
- CPChamath Palihapitiya
Um, but I think you need something new. You know, TikTok, the reason why it became successful was it was a fundamentally new use case relative to the alternatives, and I think that that captures people's imagination and mindshare. So, this is not what Threads does. And so inasmuch as it's a copy of something that is already an established behavior, I don't think it has very high chances of success.
- BGBrad Gerstner
But don't Stories and, and, and, and Reels undermine that argument? I mean, Reels was a copy of TikTok, and Stories was a copy of Snapchat, and they're both giant products today.
- CPChamath Palihapitiya
Um, I think that you can definitely copy features into an existing product with new distribution, but to invent a new product wholly from scratch, that's a carbon copy of something. But again,Reels attaches onto Instagram-
- BGBrad Gerstner
Right.
- CPChamath Palihapitiya
... which is a unique use case, right? And so, I think that, again, it goes back to you have these five or six established modes of social media that essentially are from large pieces of content to small pieces. That's probably the best organizing principle that we have. And then that's one axis, and the other axis is text, video, audio. And along that spectrum, you can basically compartmentalize all these things, and then there are about four of them that have real scale. And, and so in as much as you have established distribution, yes, you can copy a feature from somebody else and it will get used. But that's not what this is. So if this instead gets integrated Insta- Insta- Instagram, I think it's much more dangerous to Twitter than as a standalone product. As a standalone product, I tend to think it's DOA, for the most part.
- BGBrad Gerstner
Zacks-
- CPChamath Palihapitiya
Unless, again, it invents something totally new that we're missing.
- DSDavid Sacks
Zacks, you're an-
- BGBrad Gerstner
Ye-
- DSDavid Sacks
... advisor to Twitter. What do you advise Elon and, you know, how do you think about this as a competitive threat?
- BGBrad Gerstner
Yeah, I'm not, I'm not a formal advisor to Twitter.
- DSDavid Sacks
Okay.
- BGBrad Gerstner
I was just a guy hanging around during the transition. No, look, I mean, I tend to agree with, with Chamath. The thing that Facebook did really well here is there's a one-clicks sign-up flow from Instagram, where you just click to download Threads, and then you can log in with your Instagram account and port over your bio and all of your data and your social graph. So it's super easy to get set up. But by the same token, you know, if they have 10 million or even 30 million sign-ups, that's really just a 3% conversion rate on the billion, you know, users that Instagram has. So a lot of people are just gonna m- make that click because they're curious. They wanna find out what Threads is. I wouldn't be surprised if they got 100 million users or 200 million users that way, just people, you know, clicking over from Instagram to see what this new thing is. The question is, what's gonna f- be the habit-forming behavior here? And the people who use Twitter are really addicted to using Twitter. It's where the conversation is. There is a strong network effect there, not just around the users and the social graph, but, like, the habit of daily usage. So, you know, how many of these people who are signing up are actually gonna go use it every day? Are they gonna take the time to copy over, to kind of copy-paste all of their Twitter posts over to this new medium? And then what about all the comments and replies? So I wouldn't just look at the sign-ups here. I think you have to look at, like, the amount of posting and the actual usage before you know that Twitter has a threat.
- DSDavid Sacks
For sure. Right. Did, uh, did any of you guys use it this morning or last night?
- BGBrad Gerstner
Yes.
No.
- DSDavid Sacks
What, I mean, what'd you think about the UX?
- 17:15 – 28:36
Chat AI interest in decline?
- BGBrad Gerstner
- DSDavid Sacks
So it's interesting because at the same time as Threads is launching to compete with Twitter, there's, uh, some really interesting data coming out showing declining usage of ChatGPT, uh, and interest level in Bard. So both if you look at Google Trends, as well as data coming out of SimilarWeb, which, uh, tracks, uh, usage across sites, um, ChatGPT usage seems to be falling, uh, from, uh, a peak in May. Uh, and it had a modest increase from March to April, and then an even less modest increase from April to May. And May to June, we're actually seeing a decline in usage. The question is, is this driven by educational usage? So a lot of kids were using ChatGPT to write essays and to use it in school, and that seemed to be a primary use case. Or does it speak to a more broad kind of challenge with ChatGPT really disrupting search and disrupting other ways that people are kind of accessing and, and, and browsing the internet, uh, that it, that as an interface, uh, maybe it's a little bit too challenged.... and it doesn't replace the, the simple two-word keyword click and click on the result. Uh, I'd love your guys' point of view on the product and the experience as well as why do we think that there's declining usage, uh, in ChatGPT.
- CPChamath Palihapitiya
Well, you had these products launch in a f- in a moment where there was an, in many ways, a, a usage vacuum. What I mean by that is if you look back over, like, the last 15 or 20 years, there were these waves that would create these layers of innovation that consumers would get infatuated with and try. (no audio) It actually came right, it came right on the heels of a pretty massive head fake, which was around VR, and that was supposed to be this next big tidal wave of consumer innovation, which turned out to be just a total fart in the wind.
- DSDavid Sacks
Mm-hmm.
- CPChamath Palihapitiya
And so, I think that there was a set-up here where consumers were, you know, hankering for something really interesting and unique and new and novel. A lot of people wrapped those labels around the chat versions of these LLMs, and so you had, again, this explosion of usage, but I think what we're going to find there is that there's some pretty useful use cases, but narrow, where these chat interfaces are very useful, and the usage will decay to that number. And that number is probably a fraction of what the peak was. Mean- so then people will get disillusioned, and the press will say this was, you know, another fad from Silicon Valley. But then I think the reality is that the real stuff, which is around enterprise software, healthcare, uh, the physical sciences, that's where the real AI leaps, I think, will have really momentous value. Those are still 18 to 24 to 36 months away from seeing the light of day, in, in terms of real products that, that actually work. So, um, yeah, again, you know, it's part of the hype cycle and we all kind of fell for it. The only winner here is NVIDIA. I think the loser he- here are most of the VCs who pumped in hundreds of millions to billions of dollars in rando stuff too early.
- DSDavid Sacks
(laughs)
- CPChamath Palihapitiya
Um, and the consumers, they tried it, they didn't like it, they moved on, they're waiting for the next thing.
- DSDavid Sacks
Zach's agree, disagree?
- BGBrad Gerstner
Well, I think that's, that's going a little bit too far. I mean, I, I think that, uh, there may be a couple things going on here. One is that the curiosity factor ha- may have been played out. I mean the novelty has worn off a little bit. I think there was a lot of people using it initially just to see what it could do, and hearing about it and wanting to test it out. So, I think people have sort of scratched that itch. Also, school is now out of session, so for all the people who are using it to do some kind of academic research, there's not a need to do that right now. Um, I, I, I do agree that the use cases that are most exciting to me are, are enterprise, you know, uses. And I agree with Chamath about that, and I think that is all still to come. I think in terms of the consumer, um, I, I don't think it's going away, but I think that, um, they're gonna have to improve the, the accuracy. They're gonna have to improve the performance, the speed of it, uh, maybe improve the interface, add some more features if they wanna g- get to the next level of usage.
I mean, think about, think about the product today. You know, 100 million downloads, the, the whole, the whole downtick here is explained by kids being out of school. I mean, it's down 10%, and, like, kids have to be more than 20% of the usage of this thing, I would think.
- DSDavid Sacks
Totally. Yeah.
- BGBrad Gerstner
Um, so I think you, you set that aside, but just think about how hard it is to use this product, right? If you, uh, y- most people, you have four million people paying 20 bucks a month. Could you imagine people paying 20 bucks a month to use Google? Then on top of that, try downloading plug-ins. It's still a pain in the ass, and if you don't have a plug-in, I think there are only 500,000 people using plug-in? If you don't have a plug-in, you have no new data past 2021, which makes it dead on arrival as a product. So I don't even think ... I think we're so early in this that to judge the chat interface based upon the product that exists today, I think's a huge mistake. Um, I think that what we're going to see is kids come back to school, you're gonna see the normal uptick in traffic. It probably hangs out around this 100 million, 150 million level in terms of usage. I think the next 10X for a chat-based interface at least as it, in so far as it concerns the consumer, is when we move to inf- from information retrieval to action. And you've heard Zuckerberg talk about this on the Lex Fridman podcast, you've heard, uh, uh, uh, you've heard Meta talk about, uh, uh, you know, this outside of that context, and then you've heard Mustafa talk about it from Pi, which is this idea that I say, "Hey, show me the five best hotels in Milan." Um, it shows you five hotels and then you say, "Book me, uh, the Cipriani for these dates." And it actually will book it directly with the hotel, engaging with an agent there. Now, Mustafa from Pi says this is months away. Zuckerberg alludes to the fact that they're gonna have action bots on WhatsApp and Instagram in the not too distant future. Um, I think all of the, you know, all the links exist in the world to do this today. So I would say look for that as the next 10X feature for consumer-facing chatbots, and then finally I'd just say, I was at the Snowflake Summit last week. There were 600 applications, so new startups, that applied, uh, for their startup contest. They were each using an application of ChatGPT built on top of Snowflake data. 600 new companies, and this is at, uh, you know, they just, they just launched their application layer. So there's a tremendous appetite for people to help enterprises.... right? Access that information and build much more seamless discovery on top of it.
I, I tried to use ChatGPT, the mobile app, on July 4th. Um, I was just trying to find a good quote to use in a, you know, by, like, an American patriot or f- founder, framer of the Constitution, somebody like that, uh, to use to tweet out for July 4th, and it just errored out on me. And it was, like, one of these error messages that you can tell an engineer wrote. It's a totally non-anticipated error. And I tried it, like, two or three different ways, a couple of different chat threads. Uh, it just didn't work. It... Completely inexplicable. So when you have those kinds of experiences, it makes you just want to go to Google, which is what I did.
Right.
You know, I actually went to Google to find what I was looking for. Um, I thought ChatGPT could do a better job because I thought it could help me think, find things that would be specifically appropriate for July 4th, so I... Th- I was curious to see what opinion it might have, um, but Google is just much more performant. So they have to, like, work out those kinks.
Yeah.
I mean, that's, that's pretty clear.
- DSDavid Sacks
I think, I think, like, natural language prediction as a capability is certainly here to sa- to stay, along with the agents and the action bots that can integrate behind the scenes to do things for you. The real question for me is, is this a winning interface? If I think back to u- un- user interfaces, user experience in the history of computing, we can kind of go back to the original terminal interface on DOS, and you would, you know, have, like, lines that you would type and, you know, you would kind of program the computer to do stuff. That was almost like the first real computing interface that, that people could use, and then there was Windows. And when Windows came along, it was a new type of interface, a new type of interaction model. Um, and then we had the icons that arose with the iPad, iPhone revolution, uh, and obviously, the internet, the browser, was a, was a new UX. And the browser provided hyperlinks, and in one window, you would click through and go to lots of different things. And then that browser interface eventually integrated images and video, and then the interface that, that we're all most used to lately that we all probably spend most of our time on is the scrolling interface, where there's infinite personalized content created for you, and you just keep scrolling, whether that's on Facebook or Twitter or Instagram or what have you, and that's where we spend a lot of our time now. And if you think about the transition, you're getting more for less, meaning you're getting more content, you're getting more output with less input. The challenge with the chat interface is that you're getting almost today, in its current iteration, a little bit less output because it's mostly textual, and it's requiring more input. You're having to write sentences and ask it stuff. Um, and so I think in... If you think about the evolution of user interfaces in computing, chat as an interface faces quite a bit of friction. The output of it, however, is so compelling in certain use cases, as Chamath points out, that there's certainly gonna be places where it's absolutely gonna replace the current modality, um, and then the backend of it can do incredible things that no other computing interface can do, uh, but it needs to have, you know, kind of a, a, a revision or, or rebuild on the front end for it to really work. That, that's my point of view on how I think about kind of the long trajectory of where we've all been trained, um, mentally with respect to computing interfaces and, and how this might kind of be challenged. I don't know if that-
- BGBrad Gerstner
I, I think that's the key, how, how, you know, four older guys have been trained. Watch how kids interact with their, their phones. First, they're, they're never on a computer. Yeah. They're always on their phones, and they're always talking to their phones, and it's always chat-based. And so I think we have a full-
- DSDavid Sacks
Right.
- BGBrad Gerstner
... new, new generation-
- DSDavid Sacks
Right.
- BGBrad Gerstner
... that, like, it's so native to the way-
- DSDavid Sacks
Right.
- BGBrad Gerstner
... that they interact with the world. And, you know, to be honest, I was sitting there this morning prepping for the, for the pod, and I had a brilliant conversation with ChatGPT about what happened with rates and inflation, et cetera, in 2000, 2001, 2002. It was interactive. I was using my voice. I wasn't typing anything. I thought it was terrific, way better than the experience I would have on Google. So again, I think we're early in the evolution of this, um, but, you know, I think that, uh, you know, for my kids, this is completely native.
- DSDavid Sacks
Yeah, I mean-
- BGBrad Gerstner
So it's great-
- CPChamath Palihapitiya
I think for kids, it's amazing. I just don't think kids will have 20, 30, 40 bucks a month to spend on this.
- BGBrad Gerstner
(laughs)
- CPChamath Palihapitiya
I mean, they... Some kids will. But that, that'll just further exacerbate the divide of the kids that can versus the kids that can't. But I don't think that that's what's gonna create a valuation framework for a multi-decabillion-dollar company. If you wanna s-
- 28:36 – 56:07
Fed meeting minutes, economic data and outlook, and EU risk
- CPChamath Palihapitiya
see a well-used education product, you know, Chegg's got something probably not near... You know, uh, that's probably as good, and that's a three or four billion dollar market cap company last time I checked, so...
- DSDavid Sacks
Yeah. Well, Brad, you mentioned that you were doing a little research for the pod talking about rates. Obviously, that was in anticipation of the discussion on, uh, the Federal Reserve, um, meeting minutes that came out a couple days ago. I'll read a few excerpts, uh, from the published minutes, uh, where, obviously, the, the Federal Reserve meets to discuss overnight rates, whether to raise rates, and, and their outlook for rates, um, driven in part by current economic data, including their view on inflation, their view on economic activity, et cetera. Reading from the notes directly, um, "Participants agreed that inflation was unacceptably high and noted that the data, including the CPI for May, indicated that declines in inflation had been slower than they had expected. Participants, uh, observed that although core goods inflation had moderated since the middle of last year, it had slowed less rapidly than expected in recent months despite data and reports from business contacts indicating that supply chain constraints had continued to ease." Um, in their discussion on the household sector, they noted that, "Consumer spending so far this year has been stronger than expected and that aggregate household wealth remained high, as equity and home prices had not declined much from their recent highs," and a few participants mentioned that while overall, the household sector still retained much of the excess savings it had accumulated during the pandemic, there were signs that consumers were facing increasingly tighter budgets given high inflation, especially for low-income households despite savings. Um, so Brad, maybe you can give us your quick read on the minutes, and, and they did say that they're not raising rates, but they do expect two more...... 25 basis point rate hikes later this year, maybe you can tell us what market data is indicating, is that actually the case? And is the Fed not giving a- a- a clean reading on economic activity and inflation as it's being published in other places than what the Fed is currently reading?
- BGBrad Gerstner
Well, I'm- I mean, I think the most interesting thing here, and perhaps the most non-consensus thing here is, the Fed kind of seems to be orchestrating a pretty soft landing, right? Nobody wants to hear it. The market fought it for the first half of the year. The Nasdaq was just up 39% in the first half, and most people didn't participate, so everybody wants to talk this down and say that inflation's still out of control. Chamath's been talking about inflation higher for longer, but notice he doesn't say, "That's a problem," or, "The economy therefore is going to crash." It's just like inflation's going to c- be a little stickier on the way down. Rates are gonna be a little bit higher for longer. I think when you look at this in totality, the reason they hit pause is because they've come a long way quickly. They know that things are starting to trend in the right direction, so their own forecast for CPI is that it comes down a lot by the end of the year. Goldman Sachs is now at 3.3%. And I know we have readings like true inflation that say the inflation today is actually lower than that, but you don't even need to get to the debate between true inflation and CPI. They're both trending a lot lower. But we got a really interesting reading this morning, because, like, why do we want to, you know, raise rates? Well, we want to cool off the economy. And so everybody's looking for that indication that the economy is cooling. The leading indicator for this is jobs. So we had a really hot ADP jobs report this morning, right? Over 450,000 new jobs created and everybody got nervous. The 10-year shot up to over 4%. The market turned down at the start of the day. And then we got the JOLTS report. Now Larry Summers has described JOLTS as a much more... that's job openings, okay? So this is more of a leading indicator, like how many of the job openings are we consuming? And that actually came in better than expected. So what it suggested is that more people took jobs than we had anticipated. And remember, JOLTS peaked closer to 12 million, and now we're at about 9.8 million. And so when I look at that, you know, and then lo- and look at the jobs report, ADP jobs report, over half of the new jobs created were in hospitality and leisure. So think about what was happening a year ago. Airlines were trying to hire people, restaurants were trying to hire people, hotels were trying to hire people, and nobody took the job for two reasons. Number one, they were afraid of COVID. But number two, they were getting stimmy checks from the government. They didn't need to work. Now what we see is people are starting to take jobs. Those show up as higher employment numbers in ADP but in lower JOLTS, right? So fewer job openings. To me, again, that probably speaks to a, you know, healthy trend, uh, in the economy. Um, but clearly the economy is not crashing. We had a lot of people who said that the economy was going to hit the skids in Q1 or Q2 of this year due to these higher rates. The economy's incredibly resilient, but if I had to describe, you know, the rate trajectory, the Fed has said, "Listen, we may have two more rate hikes as basically priced in," and if you look at real rates... So this is the- the interest rate that really acts as the brake on the economy. So this is effectively the 10-year versus future expected inflation. It's now at one of the highest levels we've had since 2008, 2009, back closing in on 2%. So, you know, it seems to me that the Fed's got its foot pretty strong on the brake. It's jawboning the economy, saying, "We'll do more if we have to," but the things that they're looking at, job openings are coming down, inflation is coming down. So I don't know. I think you have to leave open the possibility that the fat part of the distribution curve looks like a softer landing than most people wanna- wanna admit.
- DSDavid Sacks
Yeah, so Chamath, Gerstner is predicting a soft landing. Do you agree? I mean, you've said we're gonna see rates stay high for a very long period of time.
- CPChamath Palihapitiya
Yeah, I mean, I think I've said this for the last few weeks, so I'll just say it again. But I think that it's, um... there's no hard landing, right? That's the... Hard lending was sort of the Druckenmiller thing that- that I think that, you know, I said I think it was a few weeks ago. It's very difficult to see a hard landing when China stimulates, um, just because of their natural gravitational pull, um, in the world economy. And, you know, we can talk about that again, because it just looks like China's in super, super, super trouble. Meanwhile, I- I sent this little image to you, and Friedberg, you may just want to throw it up here just to look at, but, you know, when you look at inflation, we've done the best job of the developed nations in getting this thing under control. And so, um, I think now it's just about making sure that we adequately contract the money supply and making sure that we have bullets in the chamber in case things get bad. What do I mean? When you look at this chart, the thing that I think about is, I'm glad rates are almost at 6%. Why? Because when you look at the UK, Italy, Germany, most of the euro area, France, Japan, and then, you know, uh, China, it's nonexistent, but China is a different issue. The- the reality is that, you know, a pretty bad set of economic circumstances could actually touch the Western developed nations if you start to see some of these countries rip rates higher to like 7, 8, and 9%. The UK, i- it seems like, I don't know what you guys think, it's definitely going to seven. And it could go to eight. I mean, it could be a very bad situation for the UK. They're in a lot of trouble. And so I think it's important that the Federal Reserve have tools, so by continuing to constrict the money supply, have rates that are relatively elevated maybe a quarter or two longer than they need to, it gives them a lot of positive optionality if they need to step in if something bad really happens. And so, um, I generally think they've done a very good job.
- BGBrad Gerstner
I mean, Cham- Chama-
- DSDavid Sacks
And I think... Go ahead.
- BGBrad Gerstner
Chamath, not to interrupt. To your point, what's really interesting is the symmetry with what happened in 2000. So, I think, I think we had all agreed that the last time we saw the zaniness that we saw in 2021 in the ZIRP environment was probably like, 1999, 2000 era. Well, in 2000, remember, we had this huge asset bubble, and the Fed raised rates from 5.5 to 6% in 2000, okay? In the first half of 2000. And they were saying that they're gonna stay high for long. And remember, in 2001, we entered a recession, and we ended 2001- uh, one with rates all the way down at 1.75%. Those are the bullets in the chamber that you're talking about. That then pulled us out of the recession, um, you know, and some people blame them that they caused that recep- recession. But, m- you know, when you look back from this horrible perspec-
- DSDavid Sacks
No, in this case it looks like- Yeah, in this case it looks like the recession is really not of our making, but, you know, we may actually have a soft landing. But the economy could actually contract, because when you look at these Western economies turning over, it's- it's not a good situation. I think the Eurozone, quite honestly, is probably a quarter into a recession already. So, you know, we have six or seven of the really important countries probably in a recession, the UK in really big trouble, the US relatively keeping things, you know, in a pretty good place. China, e- economically, I don't know where the demand is gonna come from. You know, on the same time that their, that their internal demand is imploding, they're, uh, creating all these export controls that I think are not going to actually help them solve the problem, because it just accelerates Western economies' desire to de-lever from China. So, the whole setup, I think, is very complicated, one, but the US looks really good, frankly, and, uh-
- BGBrad Gerstner
Well, I, I, yeah, look, I don't think we're out of the woods quite yet. Um, so, so Druckenmiller, by the way, said that he was predicting a hard landing in the second half of the year. We're just starting the second half of the year, so he's still got six months to be proven correct. Um, I find his analysis compelling, or I did at the time I heard it. Uh, but think about our situation. We still have an inverted yield curve. It's the most inverted it's been. I think this is the longest it's been inverted. Now the Fed is signaling that we're gonna get two or three more quarter point rate hikes, so it's about to become even more inverted. And what that does is it puts incredible pressure on the banking system, because the whole banking business model is to borrow short, which is, uh, from depositors, and lend long. And if short rates are higher than long rates, that whole business model doesn't work. So, either they can't engage in lending activities, meaning there's a credit crunch, or, uh, deposits just leave the system. And that- that's happening too. So, any business out there that's dependent on credit, the real estate industry or auto industry, I mean, any industry that depends on loans and credit, they're gonna continue to be hammered by this. So-
- DSDavid Sacks
Yeah, I mean, to, to your point, Sax-
- BGBrad Gerstner
Go ahead.
- DSDavid Sacks
... I'll just read the minute commentary on this. "The meeting did include a discussion on the stress on the banking sector and, uh, economic activity in general because of decreased lending. Participants generally noted that banking stresses had receded, and conditions in the banking sector were much improved since March. Participants generally continued to judge that a tightening in credit conditions spurred by banking sector stress earlier in the year would likely weigh further on economic activity, but the extent remained uncertain." So, obviously a wait and see. And then they mentioned that, "Credit conditions have not appeared to have tightened significantly beyond what will be expected in response to the monetary policy actions taken since early last year." So, at this point, saying it's a wait and see does not seem to have overstretched in terms of credit tightening at this stage, but it could.
- BGBrad Gerstner
Well, yeah, I mean, so here's the thing, is that the Fed engaged in an extraordinary intervention a few months ago to save the banking system with the Bank Term Funding Program, right? Where they basically said to all the banks that, "If you have US treasuries..." And I think it also applied to mortgage-backed securities as well. "We will, we will basically take all of those assets and give you par value. We'll give you 100 cents on the dollar-"
- DSDavid Sacks
Yeah.
- BGBrad Gerstner
... um, "in exchange for..." I guess you have to pay, you, the bank, would have to pay the, um, the one-year, uh, I think it was the one, the 12-month interest rate with like 10 base points or something like that. So, if you're sitting on bonds that have gone down 20 or 30% in value, you can go get all that money back by, you know, presenting them to the Fed.
- DSDavid Sacks
Mm-hmm.
- BGBrad Gerstner
And then you just have to, you know, you have to then take the 100 cents on the dollar you get and make sure you loan it out at a higher than the one-year interest rate, which is around, I guess, 5%.
- DSDavid Sacks
Right.
- BGBrad Gerstner
Which you can still do in residential, right? Because, uh, mortgages are still something like over 7%. So, the Fed has provided a lot of liquidity to the banks through this, uh, BTFP. Uh, it hasn't helped the commercial real estate guys, I don't think. That's why there, there're still, I think, huge problems in the commercial real estate sector. But the Fed did engage in a huge intervention to save the banking system, and they may not be done with that yet.
- DSDavid Sacks
You guys, I, I-
- BGBrad Gerstner
So, I just wonder-
- DSDavid Sacks
Yeah.
- BGBrad Gerstner
... you know, I just wonder if the F- if you didn't have all of these distortions, what would the real shape of the economy be? You know, another one on the fiscal side-
- DSDavid Sacks
I mean, you're-
- BGBrad Gerstner
... on, so on the fiscal side was the whole, um, Biden's energy bill.
- DSDavid Sacks
Yeah.
- BGBrad Gerstner
Which was supposed to be 350 billion, but as our energy investor friends are saying, it's probably gonna be more like a trillion when they add it all up. I mean-
- DSDavid Sacks
No, but that's honestly... Hold on. Tho- Those guys are being hyperbolic, and they're just stupid and wrong, 'cause I know which friends they are, and I, and I know them to be mostly stupid and mostly wrong most of the time. Look, the good thing to know about all of that- Oh, my God, I'm gonna, I'm gonna get a text after this.
- CPChamath Palihapitiya
Well, yeah, you should-
- 56:07 – 1:21:48
Florida Senate Bill 1718, jobs, and immigration debate
- BGBrad Gerstner
- DSDavid Sacks
Okay. I want to talk a little bit about jobs. Uh, the Florida State, uh, Senate Bill 1718 went into effect on July 1st. This bill now requires that any company with more than 25 employees use the E-Verify system to verify the legal immigration status of their workers, of their employees. And this will create, um, a- a real impact on businesses that employ more than 800,000 illegal immigrants in the State of Florida, uh, to do a lot of work and a lot of labor. It's unclear how many of those 800,000 illegal immigrants that are working in Florida today are working at a firm with more than 25 employees, but the impact is gonna range from the construction industry, uh, to farm labor, um, and a lot of other manual labor, um, sectors, uh, in a state that has 2.3% unemployment. Uh, DeSantis has made this, um, you know, uh- uh- an important speaking topic, uh, when he's spoken publicly. He did quite a bit of press around the passage of this bill as a way to counter Biden's "open immigration policy." Sax, um, I know that you've talked about DeSantis in the past. Do you have a read on the impact that this immigration policy will have? Do you agree with it? Would love your thoughts.
- BGBrad Gerstner
I mean, I don't think this is a- a jobs bill. This is a immigration bill or it's an immigration issue. And look, I don't know what the impact on Florida is gonna be from this. I don't trust a single story that's out of... I guess there's just a story in the Wall Street Journal about it. I mean, this feels like a campaign story. So, um, I wanna see more stories before I reach a determination on what the impact's gonna be. But politically, do I think this is smart? Yeah, because the border is the number one issue in the country right now. I don't think you guys are as clued into to like how on fire the country is about what's going on at the border.
- DSDavid Sacks
At our fundraiser, RFK talked about the- the border 'cause he went there. There's a bunch of really good clips online. I don't care who you are. You can't listen to that accounting. It's just an accounting, right? So anybody could do it. So it doesn't matter what your political, um, perspective is on RFK. Anybody who listens to that accounting can't be anything but shocked. I was shocked.
- BGBrad Gerstner
Yeah. I mean, what he described is- what- what RFK described is he went to the- the border at Yuma and there were literally holes in the wall, uh, you know, f- from Trump's wall. And literally the building materials to finish the wall and plug those holes are sitting there on the ground and the Biden administration refuses to use them and to plug the hole because they don't want to give any credit to Trump presumably. Um, I mean, literally that's how petty it is. Or it's actually their policy. They actually like having an open border. So something like seven million illegals have come in through the southern border since Biden's been in office. And we don't-
- DSDavid Sacks
And they're not- and they're not- they're not from Central America, it turns out. These folks are from, uh, Eastern Europe, um, and Africa predominantly, many men of military age. Um, and it's an entire business that's run by the cartels. And as he describes it, it's- it's really just plainly shocking. Can I ask you guys, Sax-
- BGBrad Gerstner
If one of our- if one of our enemies wanted to get thousands of sleeper agents into the country, it would have been very easy for them to do this.
- DSDavid Sacks
Okay. Can I- can I just-
- BGBrad Gerstner
Think about that for a second.
- DSDavid Sacks
Um, yeah. I get it. Yeah, can I pivot away from the border policy question to one of labor? We obviously, as evidenced by the jobs report from ADP this morning, remain in a very tight labor market in the US. And we as a group often talk about H-1Bs and the importance of, um, a- allowing educated immigrants into this country, uh, to meet our- uh, our knowledge workforce needs. But there obviously is a pretty sizable manual labor workforce need in this country and a very tight labor market to fulfill that need, ranging from construction where we have, you know, deep, um, and significant aspirations as a country to improve infrastructure, uh, to farm labor where we have, you know, the largest agricultural export market in the world. And many of these industries rely on low cost labor for those businesses to meet their economic objectives, to be able to be profitable. And we simply may not have enough of a labor force of legal immigrants or legal, um, citizens or residents of the US, uh, to meet those obligations. What is the answer, Sax? Is it that we are supposed to take anyone that's illegal in the country and make it impossible for them to work here? And is that not gonna have a very adverse effect on these important, um, segments of our economy that today rely on, um, that labor force, uh, that is here illegally? And I'm not- I'm not sta- uh-
- BGBrad Gerstner
Listen, I think- I think you could-
- DSDavid Sacks
I'm not stating an opinion. I'm asking the, uh, like the important here.
- BGBrad Gerstner
I get what you're saying. I think w- we can start to resolve that issue when you solve the border crisis. The problem is the average American doesn't want to hear anything about, you know, quote unquote comprehensive solutions to the border crisis or immigration or whether there should be-
- DSDavid Sacks
But what do you want? Like- like for ex- I- I- I'm not talking about politics. I'm not talking about... I'm just talking about your personal point of view.
- BGBrad Gerstner
My personal point of view is we need to seal the borders so we stop having this problem.
- DSDavid Sacks
Okay, but what about the work?
- BGBrad Gerstner
And then after-
- DSDavid Sacks
What about the labor? Yeah.
- BGBrad Gerstner
Yeah, let me... I- I- I'll take a crack at it for you, Gurb.
- DSDavid Sacks
Okay. I'm just trying to, I'm trying to diagnose the difference between the, the, the border policy question and the real economic quest- I know, uh, the California ag sector very well. There's an important migrant visa, um, that the entire farming economy depends on into California, and, um, these are illegal immigrants that are legally allowed to work in the farming sector, and it allows farms to be able to do the work they need to do because there is absolutely no labor to do that work without this immigrant pop- population.
- BGBrad Gerstner
Of-
- DSDavid Sacks
And there- there's other elements of this that are critical across the economy. Sorry, Brad, go ahead.
- BGBrad Gerstner
Yeah, so why, why can't we have a common sense migrant, you know, worker program or worker visa, you know, predicated on that, right? The problem is, I think where, you know, most people are in agreement that if you have a relatively secure border, right, then you design good policies, we should have a policy for people who want to come and work and earn some money and, uh, you know, we, we have job openings in this country, and then return home, right, or go through a normal process, right, to get in line to citizenship. But the problem is people cutting the line, you know, skipping the system, and given the social safety nets that states on the border have lined up, and we as a, a, a, you know, a nation have lined up, that's just unsustainable in a world where we're already, whatever, $35 trillion in debt. So, I think these things can co-exist. You can treat labor humanely. You can create a safe harbor for labor to come to the country. But that doesn't mean that you should just have holes in the wall and anybody who wants-
- DSDavid Sacks
Right.
- BGBrad Gerstner
... to come here can come here in whatever capacity they want.
- DSDavid Sacks
Chamath, assuming that the border is closed and we fix that problem, what do you think the right solution is to the millions of illegal immigrants that are residing in this country today?
- CPChamath Palihapitiya
Well, I'll, I'll answer that, but let me, let me ask you a question. Do you think we should seal the border?
- DSDavid Sacks
Yeah, for sure.
- CPChamath Palihapitiya
And why do you think we should seal the border?
- DSDavid Sacks
So that we have a system that we can use to manage the flow of labor and decide what's appropriate for our country, and we can actually have a conversation about what qualifies someone to come across the border and to immigrate into the US, and I-
Episode duration: 1:38:20
Install uListen for AI-powered chat & search across the full episode — Get Full Transcript
Transcript of episode JD-b-Zl7yso
Get more out of YouTube videos.
High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.
Add to Chrome