VC investment right now is focused on generating press and quick returns, rather than on creating real value, according to Chamath Palihapitiya, founder of VC firm Social Capital.
Venture capitalists have the classic definition of the “principal-agent problem,” as they never invest their own money, Palihapitiya said. “You’re there to get credit, [get] a TechCrunch article, to get a press release.” It fosters a culture where people get promoted for “making good decks… not creating value.”
Speaking at CB Insights’ Innovation Summit Palihapitiya criticized VCs on their unwillingness to invest in “hard things.” Tech-focused funds in particular seem to have a narrow focus. “I find that the funds that are not founder-driven are the most open. Ones farthest from tech are the most open. At the center are dopes,” Palihapitiya said.
He stressed the importance of healthcare initiatives like oncology and diabetes therapeutics, that would take a longer time to see returns. “We will get paid a lot, but it will take longer,” he said.
But the challenges with areas like healthcare isn’t just the time it takes to develop a product fit for the market. It also requires working with regulatory bodies like the FDA, something Palihapitiya sees investors as unwilling to do.
“Mature and grow up and realize that we are doing serious things … that require working with serious people.”
Robin Wigglesworth, Reporter, Financial Times: I’d like to hear a little bit more about Social Capital. I mean, you’ve compared it more to something like Alphabet in it’s sort of investing strategy in the classic VC. I’d like to find out why that is, and how that differs, and why you think that’s potentially a better model.
Chamath: Yeah. I’ll answer it in two ways. Well, the practical reality is there’s like, I mean, a very simple phrase which is that “Those that can do, do. Those that can’t, teach.” And in investing, I think that also applies, which is at the end of the day, there’s really only a handful of people who can truly drive outsize returns over very long periods of time. And to the extent that you can, you should be leveraging your own capital in that strategy as much as possible.
And so, Social Capital was meant to prove something that I had been doing on the side even while I was at Facebook, which was investing. And I was investing across all asset classes, public equities. As an angel investor, I was investing off my own balance sheet, which wasn’t that large at the time. And I was having enough success where I said to myself at some point in mid-2011, “What do I wanna do with my life?” And the first thing that I said was I wanted to better myself, because I had never really had the true founder experience. And what’s great about building something like this is that, you know, in the investment business, if you do it well, you can see that intellectual curiosity and act across a whole bunch of asset classes, and I was relatively unafraid of that challenge.
And then the second thing was just more a practical desire to work on things that I felt needed to get built in the world, which should exist but I didn’t see existing, and those were sort of like the structural systems that society needs to operate well. And I think that what has happened sort of like in the last 20 to 30 years, we’ve perverted what the definition of capitalism has been to such a degree that the things that should be done aren’t done. And instead what happens is, like, a very quick reactive, you know, process where you seek out the dopamine rush of trying to find the cheaper, faster, better thing all the time, and what that leaves in its wake is a bunch of, you know, unfulfilled expectations and promises. It also leaves in its wake a lot of societal discontent. And ultimately when that stuff compounds to a degree, it boils over.
And you could see the front end of that where Facebook, at the time, where I was, was constructively solving some of it, but there were so many other parts of the world that were just completely broken. And so I wanted to try to merge these two things together, which was take a big bet on myself. I got exceptionally lucky, to be completely honest, at working at a place like that at the seniority level that I did, because it gave me a massive balance sheet. But I wanted to try to then go and allocate that in ways that I thought could compound ridiculously well over decades in areas that substantively mattered and were important to me and what hopefully will turn out to be the world.
Robin: But how does Social Capital differ from a classic VC then? I mean, you sound like you’re not…
Chamath: Well, there’s a part of it…
Robin: …a huge fan of that model.
Chamath: Yeah, I think VCs are pretty…
Robin: Go on. Let it rip.
Chamath: …worthless. I mean, they’re okay at…well, yeah.
Robin: Was it a worthless, I think…because that was…
Chamath: Yeah, I mean, here’s what I don’t like about that model. The first thing is they have the classic definition of a principal-agent problem. They generally never invest any of their own money, and they, over the last 25 to 30 years…again, talking about, like, capitalism has been perverted. You know, society’s definition of value has also been perverted. People celebrate, you know, fancy degrees from fancy schools. They celebrate hierarchical, calcified, ossified organizations, you know, where you get promoted for being political, and, you know, creating good decks and not necessarily creating value. And what venture has become is a place to celebrate those people. And over 25 to 30 years, we’ve moved so far away from the people that originally started that business, which were these outlier individuals who took bets on other outlier individuals.
So, I’m not a big fan of the business because I think it suffers from this principal-agent conflict. Like if you are about to be an LP in a venture capital firm or in anything, the first question you should ask yourself of the person that’s allocating the money is, “How much of your money is at risk with my money? And what does that represent as a percentage of your net worth?” And when you hear, “1%,” “2%,” “3%,” you might as well just go away, because that’s a job to them. And so how do you ever expect them to do anything really, really good? So when the going gets tough, the tough gets going. And why that’s bad is then you also starve capital from things that are hard, things that take a long time. Because there is no immediate quick feedback loop that pays off what you’re really looking for, right? So, if you’re not writing huge checks, you’re not really there to make a lot of money, you’re not really there to fundamentally change the world, you’re there to get credit. You’re there to get a TechCrunch, you know, article. You’re there to get a press release. And that, to me, is what’s problematic in that industry.
So, how do we look at the problem? The first thing is what I would say is the future of what we are trying to build is something that says, “Wherever technology will either create positive disruption or create some kind of negative implication, we want to understand.” The minute that you say that, two things happen. The first is your team construction is totally different. So, there are no financial people at Social Capital. There are product managers and engineers that allocate capital as a by-product of really good decision-making, and those people needed to have done something really important at Facebook, at Google, at Apple. You name it. We cherry pick great, fantastic people that understand the technology landscape. That’s the first thing.
And the second is we are functionally agnostic about where we allocate the capital. So, we do seed investing. We incubate businesses. We’ll invest in traditional Series A. We do growth. We have a huge hedge fund. And all of that is because that is a tactic of a strategy, and what we try to do is become really good strategists.
And then the question is, “How do you do that?” And the way that we do that is not dissimilar to the team that I was fortunate to be a part of at Facebook, which was very data-oriented. And what I mean by that is, look, when you look at sort of public institutions, public institutions have a way of normalizing their measurements. Scott mentioned this earlier. Stock price is a way, right, but mostly cap financials are a way. Right? There’s a very, very simple way to take an auto industry company, an insurance company, and a health care company, and say, “Which one is better?” And you can make those trade-offs.
But how do you, for example, trade off Airbnb with Pinterest with a Series A startup that I just invested that’s, you know, building a machine-learning ASIC? How would you do that? And the thing that you have to do is look past cash flows, because oftentimes they don’t exist, and look underneath the hood in how these companies really operate. And what that requires is a deep technical understanding of how products get built, what is a leading indicator of success, what is a lagging indicator, and putting those together in a complete picture of company formation.
And so what we’ve done is we’ve taken an agnostic view of capital allocation, we pooled together 35 incredible product managers, engineers, people who know how to think about building really disruptive things, and we’ve collected enormous amounts of data from third-party sources, as well as all of our portfolio companies, as well as all these companies that want to work with us, and we use that to create a mosaic that allows us to make reasonably intelligent, thus far, and highly productive capital allocating decisions.
Robin: But, I mean, just to defend, I guess, kind of traditional VCs here, I mean, it’s interesting that you don’t have all these financial types in the company, but other VCs are also stuffed to the gills full of great product people, smart people who understand industry, that understand data and love it and know it as well.
Chamath: I disagree with you.
Robin: You don’t think so.
Chamath: Give me an example.
Robin: I don’t know. Andreessen Horowitz?
Chamath: What’s the explicit example?
Robin: Well, on the company side…
Robin: …or the product people that they have?
Chamath: Yeah. Like, what decisions then manifest?
Robin: Investment in successful companies that go on to do interesting things.
Chamath: Okay. So, that’s one. Name another one.
Robin: Anybody want to present their firm here?
Chamath: I mean, there was $56 billion allocated last year, $60 billion something before that, $60 billion something before that. So, it’s one thing to cherry pick an example of a model. And, by the way, I would agree with you. I think what Andreessen has done is taken somewhat of the same approach, slightly different. They have a very specific view and reticence and conservatism around capital allocation. They don’t want to go broad. But in the thing that they do do, they are excellent, because they have taken product people and engineers, and they’ve created this envelope of trust and brand with entrepreneurs. So, I agree with you. That’s clearly a model that works, and I think it’s because they’re speaking the same language. And what I think should be the future are more firms that get started, more pools of capital that get allocated that are driven by those kinds of people, folks like Marc and Ben, folks like myself, folks like Joe Lonsdale, company formers, company builders. Because I think that it allows you to take a distribution of risks that I suspect on balance are better and broader than financially oriented investors.
Robin: You also spoke about some of the problems that you want to solve, hard problems versus soft problems. I mean, I’ve heard lots of people say that a lot of stuff that people try and solve are sort of problems that middle-class white dudes have. Be interesting to hear what you think are the hard problems…
Chamath: I think it’s more…
Robin: …what do you get excited about?
Chamath: Yeah, it’s more…
Robin: What do you want to fix?
Chamath: It’s more nuanced than that. Look, I have had the honor of working on 3 of the top 15 companies or products ever built on the consumer internet: Winamp, AIM and ICQ, Facebook. That’s a pretty reasonable résumé for me to judge other consumer internet businesses. We’ve done very few consumer internet investments. And as a practical thing, when I put on my investing hat, the reason is not because I can’t get access to those deals, but it’s because I just don’t think that those are the challenges of our time. And I also think that’s not where most of the money will be made.
So, for example, there are markets like health care. Because the combination of costs and regulatory tailwinds, there are massive monopolistic tendencies at play where you can build huge bulky cash flow creative businesses that solve meaningful problems. And so we’ve done that. We’ve spent enormous amounts of time and money trying to get those things off the ground. I’ll give you two examples.
My family is riddled with diabetes. Seventeen aunts and uncles, 11 of them have it. My dad died of complications of it. He had it for 30 years. Where I was born, Sri Lanka, the entire country is like, it’s either going to become Singapore or literally it’s going to become Jamaica. And the difference will be can they manage chronic disease? And I use that example, by the way, because, you know, in the 1950s and ’60s, those two countries, if you didn’t know, had the exact same GDP. So, it’s an interesting example for developing countries.
So, in any event, I started a diabetes business. It has taken us six and a half years. The ability to raise capital is so difficult outside of us and a couple of other people, yet six and a half years later, we announced just on Monday at J.P. Morgan a massive deal with Novo Nordisk. We now manage a million people with diabetes, which is not a huge number, unfortunately, because that disease is gonna touch 400 million people globally. But this is a business that’s going to generate tens of billions of dollars over time, and I feel really proud of that. It will make me billions of dollars. That’s okay, because we will use it for other things.
Robin: That’s okay, yeah.
Chamath: But it happened because we had the courage to do diabetes. And when I would send that company out on Sand Hill Road, it was always, “No, no, no. Too hard, takes too long. Why would you wanna work through the FDA?” You don’t have a choice but to work through the FDA, you know. And when you get cease and desisted from the FDA, you just gotta work with them. You can’t just disrupt around them. You know, that kind of, like, language just doesn’t work.
I’ll give you another example. We started a precision cancer business, you know. And again, the same situation. The difficulty in trying to get other people to have a shared perspective other than strategically oriented corporate VCs, who, by the way, I think are, like, a huge part of the success pattern for Silicon Valley in the future, which hopefully we get a chance to talk about, because they have a very different mandate than these financially oriented investors that I spend a lot of time with. And they can then support it. So, this precision cancer business, you know, we found funding from a lot of the big blues. We found funding from Safeguard. And that was critical oxygen for us, because, you know, even when I’m writing $25 million, $30 million checks, it’s still not enough for some of these problems, because they are fundamentally hard. They take an amount of investment in people that is just different than an eight-person team trying to build an app on the internet. This is fucking cancer, okay? It’s hard. And I mean, like, if you build a company, you see the investment that’s required to solve really meaningful long-term problems.
And so that’s just lost, I think, sometimes in Silicon Valley. And so, for us, it’s just become a rallying cry about how to do things. We will get paid a lot, but it will take longer, and we’ll go through these valleys where we are just…it just feels sometimes really lonely, and then we just have to have the courage to keep doubling down if we believe that it’s right. But that goes back to instrumenting and building the company and helping them build it in a way where we can understand the underlying data so that we can see these leading indicators that tell us success is in the offing.
Robin: Is it more like moonshot projects? And so you want to build something really big?
Chamath: Well, it’s a balance. And the thing is, like, big and hard and moonshot are not always the same thing, you know. Then this goes back to, like, how do you take that view of trying to create a better world but also overlay it with a very practical reality of trying to make lots of money and make sure that you never run out of money, right? Because if we go away, which I think some VCs would love, because I don’t think they find us very challenging, you know…and look, I publish all this data. I use a lot of the CB Insights data actually. Because you can easily rank us all up now. And the data is clear. It’s us in benchmark. The data is indisputable. And then you can go back and say the CB Insights’ data is garbage.
You’re here because apparently you think it’s not. Who knows? But, to me, it proves that what we’re doing is right. And also I think it speaks to the future of venture, which is venture specifically is at this point in time where one of two things has to happen if you’re a venture firm: you have to get really big and know what the hell you’re doing with deep sophistication, or stay really small and really, really precise. So, in many ways, you have the benchmark over here, right? Five, six guys, unbelievable practitioners of their craft, building a hugely successful impactful business, but $200 million, $300 million, $400 million, $500 million at a time. That’s good for some people, and I think they should go do that.
Or then there’s what we’re doing, which is multi-asset class, everything from seed to eventually private equity, credit, public hedge fund, but the overlay is tech. And then you can take down billions of dollars of capital, tens of billions of dollars of capital, and then you have market pricing power and the ability to help really shape outcomes. And that’s gonna be necessary in some of these really hard markets where you can’t just write $5 million, $6 million dollars checks and hope these companies, you know. You have to go and sit down with Ford and say, “Let’s create a program together to solve autonomy.” You have to sit down with, you know, the big HMOs and say, “Hey, you know what? You wanna do something in precision medicine? Let’s reinvent the entire stack together. You put up a couple of hundred. I’ll put up a couple of hundred. Let’s go do it together.”
Robin: But science can be, I mean, it’s certainly a huge advance, but isn’t the end of the road…there’s a danger then you just lose focus, and you start writing bigger and bigger checks because you have more and more money and…
Chamath: So, you’re speaking…
Robin: …you spread your attention.
Chamath: So, you are speaking to the biggest thing that happens when people increase their capital allocation capability…
Robin: Yeah. Everybody says…
Chamath: …which is as they…
Robin: …they maintain focus, and they generally don’t.
Chamath: No. No, because they’re risk-averse, and so what they do is, say, spread it over N number of things, but your attention doesn’t spread over N number of things equally. So, then it goes back to the same question that I asked earlier. What I’ll tell you is, like, I am not afraid. I got immensely lucky. I grew up on welfare. I didn’t grow up rich. So, to go back to that means fucking nothing to me. And so, for me, it’s like you don’t need to do 50 things. Now you can do 10 things. But now instead of 10 million on 10 things, I’m willing to put up a billion on 10 things. Why not? If it’s really working, just keep doing it. And that’s where you see the best of the best, the true best of the best, the Warren Buffetts of the world, you know, the Seth Klarmans of the world. What they show you is even as you amplify capital, you have to have the precise mental framework to continue to be concentrated. And that is, again, where the emotional and psychological mettle of an investor is tested and most people fail.
Robin: Yeah. You know, it’s something I see a lot. I mean, the Klarmans are sort of very rare.
Chamath: And look, and Seth keeps 40% in cash and a $30 billion hedge fund. How do you do that? Because he knows what he’s doing. You know, Seth has a very famous quote, which is, “You don’t pay me to hold cash. You pay me to know when to hold cash.” It’s such a nuanced statement, but so powerful. It’s hard to do this well, you know. I think it was Munger that said, “It’s not supposed to be easy. If you think it’s easy, you’re stupid.” He’s right. Anything that you’re doing, where you work, it’s hard. It’s not easy. What I do is hard. It’s not easy. And people try to boil it down to these, like, simple things, and, like, simple frameworks. And unfortunately, what that is like, you stay at this first level of thinking, and you don’t really figure out the behavioral biases that are required to truly build something durable. And then when push comes to shove, and, you know, you’re lucky enough to scale up capital or whatever, you capitulate and you do what everybody else does, and then you’re just like everybody else and you die. And you get disrupted by somebody else who comes along, and we’ll see if we can do it.
Robin: To quote Larkin, “It’s the circle of life.”
Robin: You mentioned Joe Lonsdale. I remember I talked to him last year. It was interesting. And he said that, I can’t remember the exact wording, but if he saw another payment company or a pay-to-pay lender, he would throw up. They are there being that there are certain areas that just generate a lot of attention and buzz.
Chamath: They’re horrendous.
Chamath: They’re horrendous.
Robin: …you know, what do you think are the really boring areas that just don’t excite you, or you just think, you know, are wildly overrated?
Chamath: I’ll answer your question in a slightly different way, which is I think that there are some really interesting businesses that are misunderstood as technology companies, and they’re not. So, all lending is not a technology company. I don’t care what you say to me. You’re never gonna convince me. Lending is about cost of capital. And you got confused when interest rates were at zero, and you’ll be less confused when interest rates are at 4% and your risk hurdle all of a sudden changes, and you didn’t even know what risk hurdle meant. So, I think that there are these businesses, and I agree with Joe, that are just, like, payments businesses. You know, payments businesses are, like, people love them for whatever reason. And I look at them, and I’m like, “Well, when you make basis points, in order for basis points to equal anything reasonable, then the top line has to be hundreds of billions approaching trillions of dollars.” The last time I checked, that’s really hard. Trillions of anything is really, really hard. Of anything. Trillions of peanuts, go get me a trillion peanuts. You couldn’t do it. It’s hard. So, how the hell are you gonna get a trillion dollars of transactions?
So, yeah. I think that the thing is there is like this superficial sexiness of raising venture of trying to be a technology company, and I think we lose sight of what’s really important. Venture is an asset class. At the end of the day, money is money. You should build a company the best way that you think you know how. It should come from wherever you think is the best, the people that help you the most and get out of your way.
The other thing is that, you know, venture specifically, the way it’s constructed, comes with all these reverse implied odds that I think are really negative, and they force you to build companies in very specific ways that in many ways can be counterproductive to really achieving something substantial.
Robin: So, on the venture side, with corporate venture capital as well, you’re excited about that, or you thought was gonna be a far bigger part of a lot of the…
Chamath: No, they’re really, really important. And the reason is because we now need to spend time in hard things, and hard things are dominated by established legacy durable businesses. That’s why they’re hard, right? They were able to stand the test of time. They solve things. But they are now gonna face an existential crisis. And the way that I describe this is we’ve had people…you know, like when people see your results and your portfolio, what happens is investors wanna come through the door and, you know, they wanna get to know you. And I always start the conversations with the following. And these guys have huge pools of money. Like I’ll talk about a sovereign wealth fund that was just in. They have $680 billion of assets. And I said, “You should be prepared to understand that every asset you have is either already impaired or will be impaired, and what you haven’t uncovered yet is the technology that’s gonna do it.”
And I think corporates face many of the same challenges. So, if you look at the automotive industry, right, and you look at all of these corporate balance sheets, and you look at all these corporations that have been producing and manufacturing really complicated things in a highly complex, you know, supply chain environments at scale for decades and generating hundreds of billions of dollars of top-line revenue, etc., they have a very specific skill, but they’re clearly under threat. The best thing for them to do is to not have some grandiose vision that they present at the board and say, “We’re gonna solve this problem ourselves.” That’s bullshit. It doesn’t happen. No great engineer’s ever gonna go and work for GM or Ford. That’s not gonna happen. But what they can do is they can find those pools of people in Silicon Valley that can be those magnets, and then you partner, and you use your shared balance sheet, and that’s where you can create some very innovative ideas and ways of allocating capital that don’t exist today. It happens a lot in private equity, but it doesn’t happen enough in venture. And the reason is because, again, most VCs have a very brutal definition of their job, and they’re, at the end of the day, not managing their own money. They’re managing somebody else’s money, so their propensity to take risk is nonexistent, right?
So, corporates, as an example, can find a handful of folks. So, Lonsdale, I think, is an excellent example. I think Andreessen actually with their market development team is a pretty good example where you sit down and you say, “What are our true strategic problems?” And then you go and have some of us be the actual tip of the spear that goes and help you solve it. We actually just did this with Ford. So, Ford has a big autonomy problem. There is a repeat entrepreneur that made us a lot of money in 2013 and ’14, and he’s like, “I’m gonna start up basically like the software layer that sits on top of a lot of these ADAS systems.” I’ll send Ford. It’s great. And what they got was they got a right to buy, and we’ll see if they exercise that right, and we negotiated a framework that we felt was fair. But I would do that 50 times in a row.
Robin: And so how much could individual entrepreneurs or VCs or technology go to some of these big legacy companies and say, “I think you have a problem here,” and actually sort of tailor-make and be sort of an outsider and pitch solutions to these companies?
Chamath: I think the issue for us is, like, you know, it seems very byzantine on the outside looking in, and so there’s no really clear way in which to engage a company with that.
Robin: Yeah, like who do you talk to?
Chamath: Who do you talk to, how do you start those conversations. So, I think it actually has to be the opposite. Like yesterday, we had a massive insurance company in, and they have just…I mean, an enormous balance sheet. Like these numbers I hear, and I just think, like, where do they keep the money, you know? Is it like in an BofA account? Like, what do they do?
Robin: Yeah, exactly.
Chamath: It’s crazy. And so now, like, the discussion has quickly transitioned to, like, “Okay. Let’s have a really strategic discussion about, like, how we can use our collective balance sheets to solve some important problems for you that I care about.”
Robin: So, they would go to a VC firm or a company of people and say, “We have this problem. Do you know anybody that can help us solve it?”
Chamath: No, in this case what happened was one of our LPs, he’s a very well-known investor himself, and he brokered this thing. And he said, “Go talk to Chamath from that team, and you guys figure out a solution to use this balance sheet with his balance sheet and go solve some problems.” And the way that we framed it is, like, “Okay. What do they care about?” You know, they care about figuring out different ways of repricing risk. They care about different ways of process automation. They care about understanding what all these gobbledygook around big data and machine learning are. What I said was, “Well, here are some of the things that we’re doing. And I could be compelled to learn about those things, but let’s do it together in a way that makes sense where we share the risk.” And those can be one-off things. Those can be broad balance sheet-type discussions.
So, for us, we have to be positioned, like any other smart, strategic flexible company, with really smart people who are just trying to build a big business ourselves. And so when we are too brutally defined in what we do, $5 million Series As, it limits those discussions. And instead what happens is the corporate VC is rendered as the last mile check writer, you know, who often does deals, frankly, that are the crappiest and in really bad structures. And I think that’s just, like, unproductive. Because, again, if you go back to, like, they are actually been solving really big, important problems. We, together, have a much better chance of solving some of these really big, important problems.
My experience, for example, in building our health care portfolio with the corporate VCs we’ve partnered with has been unbelievable, because we need them to exist. These provider networks, these paragroups, I mean, they are hugely important. For us, it’s been a really constructive experience. And I think there’s a trail of breadcrumbs to a much brighter future on how that should work together.
Robin: So is that just under the drug companies and the pharma companies, they just have a bigger R&D, so it’s more in their culture? I mean, which big companies do you think are doing a big job of this, and which basically have you went and seen and you thought were underwhelming, put it that way?
Chamath: I find that the funds that are not founder-driven are the most open. And I also find the ones that are the furthest away from tech are the most open-minded. So, at the intersection of non-founder driven, least technology are actually some of the brightest, most thoughtful capable people I’ve interacted with. At the center are all these dopes that have conflated luck and skill who think they know what they’re doing, and they are just, like, a huge waste of time.
Robin: It’s like Buffett’s skilled coin flippers.
Chamath: I mean, it’s just like they don’t realize…it’s like they are immensely lucky to work at some of the companies that they do, and there’s these huge massive secular tailwinds that they didn’t create, and frankly, nobody in that company knows how were created, and then they confuse that with their own skill of having built something. And that becomes very dangerous, because then when push comes to shove and you wanna solve a problem, their initial reaction is, “We can solve it. We got this. I got it.” And you can just, honestly, like, I think to myself…half the time what I do is I end up then on the public side, we’ll short them. Honestly, we’ve made an enormous amount of money just doing that.
Robin: Yeah, in fact, well, I did want to talk to you about it, because that’s very much in my world. The hedge fund you set up, it’s a really fascinating…how do you bleed some ideas and concepts between the two then? When you talk, is it like, “God, I’m gonna short this guy.”
Chamath: It’s so integrally important, because, like, you need to have, again, a complete view of the ecosystem. And in order to do that, you need to understand and spend time with these big public companies. What are the roadmaps? What are they planning? And what that allows you to do is build things that you know can be complementary, but also when you see things that you know are being built that we are a part of that will dramatically impair what they are doing, you can have some really constructive discussions, and/or…so that’s if you wanna stay in the long term. You can, frankly, get on the other side of that.
So, it is just a hugely important part of what we do, because it forces us to really think. These are all companies. They’re just in a spectrum at a point in time: Day 1 of a company, Day 10,000 of a company. And everything is just a very elegant composition of those dots in between, and it’s our job to try to figure out what things are we looking at depending on what day in its life that company is in. That’s a wonderful challenge, because it forces ourselves to not conflate luck and skill. Because it just is constantly changing the way our team has to reevaluate an understanding of the future, and then use that to understand the risk. And that’s very important for us, because we will hopefully then not take ourselves too seriously and think we’ve solved it.
Robin: But do you ever worry that if you’re gonna go into…like somebody’s coming in to see you, the VC, they should then worry about you, the VC, telling you, the hedge fund, to short the crap out of them?
Chamath: Well, I mean, public companies don’t come to the VC.
Robin: No, no. I guess not.
Chamath: But, like, these young companies, they love it. Because we’re like, “Oh, we’ll tell you exactly what we are seeing, and we’ll help you.” And the big companies that we work with, the things that we have found is, like, you know, there are some big companies that just want our help because that data-driven incisive point of view of having built, for example, a Facebook user base to a billion…and folks on my team that have, like, you know, helped build a lot of the infrastructure at Google, helped figure out, you know, the ads ranking, you know, algorithms at Instagram. I mean, like, that’s invaluable stuff, you know. And we’re very clear what we want to do. And we are also really honorable. There is a public company, in my opinion, shittily run, that is in a space that I know quite well in social media who are active where we spent a lot of time with them trying to figure out, “Can we constructively help you rebuild this company because we really believe in it?” And when they said, “We’re not ready to do it,” and we were like, “Wow. This is a really bad sign,” we just left it alone.
Robin: I think I can guess the company.
Chamath: I’m not gonna say what it is. Flash the screen. Flash the blue screen. Yeah, so we’re not there to be mercenary about it. But we try to put our best foot forward and help people across the spectrum. But when, like, ego overrides good decision-making, I’m not gonna just sit there and not take an opportunity to just pound them into the ground.
Robin: Yeah. Exactly. Well, on an entirely unrelated note, I did see the pounding you did a while ago when we talked about Twitter.
Chamath: Oh, that’s a great one. Let’s talk about that.
Robin: Yeah. We’ll take this to Twitter.
Chamath: Yeah, yeah. Let’s talk about Twitter.
Robin: And you talked about…and I thought it was really pinpointing the idea that they just have a very big infrastructure and cost base compared to what they are. Look, as a journalist, I love Twitter, all right?
Chamath: That’s a technical structure that is just fundamentally broken.
Robin: I mean, can it be fixed?
Chamath: Oh, yeah.
Robin: I mean, do you think it’s…
Chamath: It will take six months.
Robin: …fundamentally broken though? I mean, is it just…
Robin: …a cost base issue that they…
Chamath: No, no, no. No. Unfortunately, there’s probably too many chiefs, not enough Indians. That was a cheap joke. That was a cheap joke. I’m so sorry. There needs to be a broad set of really, really empowered people. Probably the organization needs to be fewer so that there’s more white space for really great talented people to operate. But I would think the first thing that needs to get done is to completely figure out how to rewrite the infrastructure of that company. It just doesn’t make sense that a company with one-ninth of the user base of Facebook spends one-third the amount of money for a business where you can only post in 140 characters, versus Facebook, you can literally now stream HD live times 1.8 billion people. So, you don’t have to be a genius to figure out that something’s wrong.
At the same time, I think that if you were to actually call AWS or GCP or Azure and say, “Help me build this infrastructure in the cloud,” I suspect that one of those companies would probably pay you to have the ability to say, “Yeah. You know what? I just migrated Twitter to my cloud.” And so then you get the leverage of those thousands of brilliant engineers and product managers working on your behalf to undo what is some unfortunate technical debt. So, the idea that one wouldn’t do that, to me, it’s just a vestige of a kind of decision-making that really scares me, which, again, I go back to is if there’s a common theme in what I think about all the time, it’s we should all have ego, but we should not be ego-driven in critical decision-making, and we should be unpacking our behavioral biases constantly and finding ways of mitigating them. And I think those are examples where there is that screaming red flags of things that are structural that just provide a very poor future forecast for that company’s ability. And even if you were to fix it, if you don’t fix that psychological tendency, that framework for decision-making, it’s gonna fail, right?
So, I love Twitter as well. I think it’s exceptional. And in a world now where, you know, we are really living in a state of either uninformation and misinformation, that channel is crucial. It’s crucial for how we think about things. And so it needs to be a vibrant, ongoing business, but I think it needs to fix itself. And there are people here that are willing to help them, but they just seem unwilling to want to help.
Robin: Do we have some questions?
Cameron McCurdy, CB Insights, Host: Yeah.
Robin: Yes, I’m sure we do.
Cameron: We have some questions from the app.
Chamath: It’s the voice of God. Where are these people asking questions?
Cameron: So, you’ve talked a bit about regulation and regulatory bodies like the FDA. Do you think that government entities are laggard in the industries they’re trying to regulate? Or is this something that startups should be working alongside with no matter what?
Chamath: I started in a place where I had the same Silicon Valley rhetoric as everyone else, which is regulation is terrible and we should just work around it. After five and a half years and a billion dollars of money invested, what I would say is that was a really immature perspective. The reality is that these organizations exist for a reason, and that they provide belts and suspenders that are, frankly, necessary. And the reason I say that now is after having gone through that gauntlet with a couple of our companies a few times, I am so glad that that exists at a very practical reason, because we are that much more likely to build de facto standards and monopolies. Because I know that nobody else is gonna take the brain damage to go through what we just went through. So that’s a very, kind of, like, very practical, maybe perspective.
The other thing that I would say is regulation has existed for years in all the markets that we’ve operated in other than consumer internet. And I just think that we amplified what the consumer internet was like, which was this freewheeling wild wild west where nothing mattered. You could do whatever you wanted. You know, I remember in 2003 or 2004, the team that I was a part of at Winamp, we released some open source software on AOL servers. And that software then became the root cause of BearShare, LimeWire, all these peer-to-peer headless networks that effectively destroyed many content businesses. And it just seemed so cool at the time. Now, I look back on it, and I’m like, you know, I’m glad I went through those experiences. But in any other market, if something like that had happened, we would have been prosecuted and put in jail.
And so I just think that we have to mature and grow up and realize that we’re doing serious things. We have to do serious things. Serious things require working with serious people in a serious way. And so I’m at a point now where I’m like, “That stuff is great. Just buckle down. Spend the time. It’s not about the quickest, fastest thing. It’s about the right thing, and let’s just go.” And what will happen is what we will filter out are all of the charlatan, fly-by-night people who really don’t have the wherewithal and the courage to go through that gauntlet.
Cameron: So, one of the running themes seems to be companies that are public typically tend to take a shorter-term sort of outlook, do a quarterly earnings. You’ve talked in the past about companies like Amazon and Workday who are sort of generating more long-term value in public markets. What are some other companies you think do a good job at that, and maybe how do they do that?
Chamath: I think the best example of a company that does that is Tesla. The other best company that does that, I would say, is Google. And let me just talk about that for a second. It’s a real litmus test on a CEO’s ability to tell a narrative that creates an envelope of trust. It’s not dissimilar to how a CEO has to manage his or her employees at a company if you’re gonna build a very successful company. You have to tell a story that is both inspirational but practical and believable. And when you can do that to your employees, what happens is you have high levels of retention, typically high levels of employee satisfaction and morale, and people execute. That’s not dissimilar to working with Wall Street and basically telling a narrative that is relatively believable, and then backing it up with enough data points along the way.
Robin: In that order as well.
Chamath: Absolutely. And so I think, like, you know, when we talk about sort of what’s happening right now, there’s too many of these private companies or public companies, these small tech companies, that are, frankly, not really durable. And I think we’re just not willing to have the hard conversation about that, which is that there are some good businesses. Are they great businesses? No. Will they ever be great businesses? No. And so, like, we just need to admit that.
But Google, Tesla, Workday, Amazon are amazing. Like, for example, like, you know, Tesla, Tesla got battered in the press because Elon didn’t hit his 2016 ship goals, except he committed to shipping 80,000 and he shipped 72,000 or 73,000 cars. It’s like that’s just incredible. And, like, in the grand scheme of things, in 5 years from now, will those 6000 or 7000 units really matter? No, because he will sell them in ’17, you know. And by 2021, the guy is gonna basically drive the energy independence of the United States. So, he deserves some white space to operate. The quarterly perturbations don’t truly matter.
Robin: Yeah. But that’s just sort of the curse. And we talked about that with Scott, I mean, the quarterly capitalism and the pressures that brings. But, you know, most people should just learn to look through it and ignore it.
Chamath: I think that most people should probably be running quarterly businesses, unless they are not. And if they’re not, they need to tell that story and just be open and transparent about it and build the credibility to tell a broad narrative and then back it up with facts. Again, it’s like stay small and really tight or go really, really big. And, again, it goes back to the reverse implied odds of raising all this money and creating all these false expectations and turning out to be a middleman business that’s actually gonna have compressing margins over time. And there’s a lot of them that are like that.
I think right now, you know, what’s great about, like, the last few months is for the first time in probably, like, 17 years, Silicon Valley is forced to basically explore and introspect around all this rhetoric. And the rhetoric was just dripping in this, like, self-righteousness. And I think for the first time, we’re realizing, oh my God, actually all of that was just a crock of bullshit. And what’s good about that is we can now become like everybody else. And I think what we need to internalize is we are as much stewards of the status quo as everybody else. And that’s probably a very healthy place to be, actually, for the long-term fluidity of what we’re trying to do.
Cameron: Great. So, you talked a little bit about the difference between founder-led companies and companies that are not. Looking at the mix of public tech companies like Amazon, Apple, Google, Facebook, and Microsoft, how do you think they rank against each other? And do you think the founder-led aspect played a role?
Chamath: I think the most important company in the world is Amazon. I think it’s the most natural monopoly in the world. I took a lot of flak in 2013, ’14. You know, I sold all my Facebook stock, which is a lot of stock, and I moved it into Amazon. And I did it for myself because I spent a year and a half thinking, studying, analyzing, modeling, modeling, modeling, going to every meeting I could, and I just came to this conclusion. There is a handful of exceptionally good companies, but there is always one company that is the best. It is the best. It is the best. Like THE, like capital T-H-E best.
Robin: Go Bezos.
Chamath: Because it does something that is just so fundamentally utilitarian and it does it just incrementally better every day, and they’ve had to learn how to be good. That is a really hard thing, to learn how to be good. It’s not as if you have an inherently amazing business that builds on top of itself that is then handed to you that your job is to then monetize and not screw up. It doesn’t mean you can’t build hugely valuable businesses. It doesn’t mean they’re not great. They are fantastically run companies, but it’s just very different than trying and starting and living on single-digit margins and grinding for 20 years.
So, to me, like that is the single best company in the world. And then the next most important company after that is Google, and the reason is because whatever happens in AI will be dictated by them. They are just such an order of magnitude further ahead. And the reason they are such an order of magnitude ahead is that they have really understood the problem of AI, which is that the problem exists at the absolute lowest levels of the stack in silicon and the absolute levels at the top of the stack in the training data that drives the day-to-day inference that makes your products great.
And so just to double-click on that, about two years ago, Google started to talk about this thing called TPU in their quarterly releases. My spidey senses went up. I’m like, “Why would they talk about some little chip?” And what you know now is if you gather together enough public data, that chip, which is their custom machine learning ASIC, custom AI ASIC that they built from scratch, is less than 10% of all the silicon in their data center but more than 50% of their compute. Actually, sorry, 40% of their compute. You may not know what that means, but you should write it down and ask your CTO why that’s important.
That is hugely important. There is nobody else doing it. Everybody here is being schlepped CPUs and then video GPUs and some gobbledygook crap and some reasonably interesting software sold by startups. None of that matters. Google’s TPU matters, because it is the future of how AI and compute will be brought to a massive scale. And then immediately once you deploy that at that level of scale, what do you realize? It’s all about the acquisition of data, and it’s about how you use that data to build training models, and then to use it to create great inference. Inference meaning at the point of interaction with your customer, the decision that’s made in real time based on the models that exist. That is the game, guys. It’s not about training. It’s about using training to do inference, and it’s about the silicon that allows you to do it mega, mega scale. They are the only company in the world that solved it.
Knowing that, what do we do? We plucked the team that founded it and we started the NexGen chip with them. So, I feel really good about what we’re doing and hopefully bringing that to the masses. So, actually what we did was we actually then…who do we go to? We went to Tesla, we went to Amazon, and we went to Facebook. And we said, “We’re building an ASIC. Here’s what it looks like. Here’s the team that did it. Let’s figure out a way to work together.” We haven’t figured anything out yet, but the point is, that’s the kind of stuff that you should be doing, you know. And that’s why working with corporations now matter because these big companies will continue to get bigger and bigger, and they’re gonna be able to consolidate, and it’s because they will have a very flexible view of what their balance sheet is meant to do. Their balance sheet is not meant to drive ROI. It’s meant to make sure they survive. And the folks that figure that out will survive, and the folks that don’t will get disrupted.
Robin: It’s interesting that idea of always standing on the shoulders of giants as well.
Chamath: I mean, if you look at enterprise software, you know, like if you’re a Salesforce, my God, what an incredible business. Why? Because they figured out the sales channel. And why is that critical? Because all of these small companies that sell software hit these ceilings, and these big companies with entrenched go to markets will be able to consolidate the entire enterprise software space. And that’s why you’ve seen multiples contract over, like, the last five or six months. And knowing that, because that was a public market phenomenon, of course it’s impacted how we think about the risk management of our private portfolio. But imagine you’re just a VC schlepping $5 million Series A checks, you don’t even think about that.
So, there’s a lot of stuff happening, but these big companies are really important. But the two that really matter for very different reasons, one is basically just the safest business in the world that I think can just compound for me, you know, over my hurdle, Amazon. And the most interesting canary in the coal mine on the true future of artificial intelligence, at the absolute lowest level of stack in AI, at the highest level of stack in inference, is Google. And if you wanna experience that experience, by the way, I would tell you just buy the new Google phone and just live in it. They can only deploy that bloody silicon to ads, search quality, and, like, you know, the photos app on the Pixel. I mean, that’s how much in-demand that stuff is. Like your fucking spidey sense should be going off by now.
Robin: I’m gonna have a look at that.
Cameron: We’re just gonna…
Robin: I wanna run out and Google some stuff quickly after this talk. Do we have any more questions over there?
Cameron: Yeah, we’re just gonna ask two really quick ones.
Cameron: So, it’s a follow-up question that there seems to be that, you know, Apple seems to be one of the companies that you haven’t spoken about in that mix.
Cameron: Apple. You know, what’s your outlook on them? How do you think they’re sort of approaching all the things you just talked about?
Chamath: I didn’t even hear the question. I just heard Apple. You want me to just tell you what I think about Apple?
Cameron: Pretty much.
Robin: All right.
Chamath: If I was a betting man, which I am, here will be the arc of Apple’s performance. Apple will follow Microsoft’s trajectory, an absolutely iconic founder who completely transformed industries, who handed the business off to someone that they fundamentally trusted, who tried their best, who did their job of being a steward, but probably, and, again, this goes back to psychology, is too affected by the legacy of that person who was their mentor for so long. And probably, and I don’t know this, probably is psychologically thinking constantly, “What would that person who made me want me to do in this situation? How do I honor his legacy?” And I think that’s an incredibly difficult position for any of us to be in.
But then when that leadership then transitions to that next generation who is less vested in the past, what you have is what Satya has done at Microsoft, which has been nothing short of unbelievable. And I suspect that at some point, Apple will become much more aggressive and much more daring in taking their brand and their capital to really reshape future markets. And I think that, to me, is the tragic thing that you see happening today is a company whose moment in the sun, I think, is gonna wane for a little bit as they try all kinds of small things. Because the risk of doing the big thing is that it will be compared to a big thing that Steve Jobs would have done, and it will look like it was just not as good, and it will freeze people. And I think you see that. I personally think that.
So, do I own it? Nope. Would I own it? No. Would I own it in the future? Yes. But you’re gonna have to see some meaningful change. You know, I’ve said this many times, but it should have tried to buy Tesla, you know. It should have tried to buy Netflix. It should try to buy Spotify. By the way, you could buy all those companies at 2X their prevailing market cap, and they’d still have $50 billion of cash.
Cameron: Last question.
Chamath: It’s crazy.
Chamath: I mean, isn’t that insane?
Chamath: So, the fact that it doesn’t happen is just tragic.
Robin: Well, it’s not that big a gamble. Like you say, if it doesn’t pan out, then people are gonna want Steve Jobs to do that.
Chamath: And what’s so sad about that is that it’s fundamentally not a big gamble, you know? These guys, they borrow money at, like, a negative interest rate. You know, they generate more cash than any company in the world. How big of a gamble is it? It’s not that big of a gamble. It’s just unpacking the psychological bias that drives a framework of decision-making that we’ve now seen over some number of years is probably leaving at least the people at the tip of the spear on the technology innovation cycle unfulfilled.
Robin: But how much is that just that they have this culture that “We build it. We don’t buy it.” No one was buying Netflix and some of these companies 10, 20 years ago.
Chamath: You know what? I think that’s such a false trade-off. I think there should be one culture, which is win. We fucking win. And then when we’re done winning, we win some more. And then when we do that, we have winning for dessert and then we go to bed in a blanket of winning.
Robin: Yeah. That’s the corporate motto.
Chamath: That’s what it’s about. It’s about winning. Those are artificial tactics. This is what I’m saying. We only do Series As. No. It’s about winning, you know. Buffett started out owning all kinds of businesses and learned about insurance. He learned about a market and then owned that market because he wanted to win. And then he bought Burlington Northern Santa Fe, because he just wanted to keep winning. Then he did Heinz and ketchup, and then he did…I mean, he’s a winner. You want to be around winners, right? I mean, do I really? But it’s like you want to win. Like a company’s job is to win on behalf of their shareholders.
Cameron: The last question, which I guess is related to that, is in the, you know, latter half of last year, you talked a lot about the election. What is your outlook on the next four years? What do you think is the biggest thing that’s going to change, at least as is relevant to your businesses?
Chamath: Here are the two things that I hope happen. The first is that we have a really honest conversation about how we are informed as a society. And don’t fall into the trap of, like, oh, it was fake news that won or lost the election. I don’t wanna have that conversation. I think what I would like to say is we are more segregated than we’ve ever been from a mind-set perspective, and a philosophy, and an outlook on the future, and what’s right and wrong. There has never been more segregation, intellectual segregation, in society, and I think that we really need to unpack that as a whole and figure out how to solve it.
And so that, to me, comes with the obvious question of, like, if you have too much power aggregated into two media companies that control basically how we all consume information, that probably has to change, and it probably has to change with your behavior in order to anoint other winners that allow a consumption and a model of payment and an economic vibrancy that allows people like you and many other great people like you to live in that world and tell the truth. So, that’s at this level.
And then at this level, which is more practical, I think it’s actually going to be the markets will rip. I think there are going to be hundreds of billions of dollars that flood the U.S. system. I think that there will be nominal job creation. I think the real problem will be what will that hangover feel like if it doesn’t work. But there will be a short-term reflexive thing that happens, which is, you know, an incentive, economically, that the presidency will create to just plow hundreds of billions of dollars back into the United States. But the worrisome thing for me is smart people will arbitrage it. The question is will there be good artifacts after that. Because the smart people will always be there to arbitrage markets, you know. The best example of that is, like, the Rothschilds, you know. How do you build a dynasty, you know? He doesn’t care who…what was the quote from Nathan Rothschild? He was like, “I don’t care on whose head the crown,” you know, “sits on which the sunset never sets. I care who controls the British money supply. He who controls the British money supply controls the British Empire.” And that’s the unfortunate place we are today and how capitalism has been perverted. So…
Robin: He’s also very good with data and information flow as well. Carrier pigeons was his secret.
Robin: He was the first high-frequency trader.
Chamath: Yeah. Happy new year, everybody.
Robin: Yes, exactly.
Chamath: Here’s to a fantastic 2017. Real honestly, I’m wishing for the best for everybody.