So

in that underlying mental model, the excess returns can stay more or less forever, right? More capital coming in won't bid down the returns much. I

think that's right. I mean, you know, the returns have been pretty good, at least for the better venture capital funds. And here, you know, we may want to come back to that caveat. But, you know, going back to the beginning of the story in the 60s with Arthur Rock, the pioneer of venture capital, right through to today, you do see good returns. I mean, Sequoia, which I regard as the top venture capital partnership in Silicon Valley, has generated returns of about 12 times investors' money this century. That's 12x, that's

1,200%. It's pretty good. So why isn't that skill replicable if the returns to replicating it are so high?

Well, I think there's a sort of inefficient markets thinking. There is a sort of notion that you can assume away skill and that if skill is good, you could just make more of it. And that would mean that you would compete the returns down. I think some of the skill is so far off in the tail of the distribution that it's quite hard to replicate it. I mean, a really good venture capitalist combines technical knowledge of what he or she is investing in, whether that's biotechnology or computer science, plus business feeling, plus the skill of networking with people and putting teams together, and a phenomenal energy, you know, because you've got to be out there getting up in the morning for one breakfast with one potential person you might invest in and then doing 14 cups of coffee before you go to bed and being that wired, you still sleep. You know, it is tough to find enough people like that. And, you know, maybe we're going to discover that the last four or five years of boom have sucked in so much talent that it will be competed away. That could be true, but so far it hasn't happened. What

do you think of the view that in recent years, there's been a huge consumer retail tech boom, basically FANG stocks, right? And when that is over, it might be over now. The excess returns to VC will go away. If you look at venture capital for biotech, which has been hammered lately as we're speaking here, late January 2022, and maybe venture capital is a limited model for one period of time, and otherwise it just does okay. True or false? False.

I say that because in a cyclical sense, you might be right. But I think there's a deep structural shift, which is really important. And that is that intangible capital has become more and more important in our economy. And the nature of intangible capital is that it's hard to measure it in financial reports. And so to understand whether a particular software investment, for example, is worth a huge amount or really nothing, you kind of need to understand what that software development within the company is doing. And so you need to be hands-on. You need to have the technical skills to evaluate that software project. And the more that intangible capital rises as a share of new GDP creation, the more this venture style hands-on expert investing is going to be valuable.

Your explanation, if I understand it, to me seems to suggest that venture capital for biotech won't work very well. So you're portraying it as something that's very, very hard to do, a very limited skill. So you're going to be wrong a lot of times. So that means the times you're right, the product has to be scalable very rapidly. But in biotech, there's regulators, right? You often need a sales force. It's not scalable in the way that, say, LinkedIn or Netflix are scalable. So doesn't that mean VC will just stay limited to a very small area, those things that are super rapidly scalable? Or if you think it's pretty easy to pick winners, then you have to think the rents get exhausted. Yeah.

Well, I mean, this is a version of actually a wider debate, which goes beyond biotech, which is sort of the claim that venture capital is really only good for software projects, that software can be scaled very, very fast. There are network effects once you get product market fit. And you don't need much capital. The marginal cost of serving one more customer is pretty much zero once you've built the Google search engine. And so people will argue, look, that's all that venture capitalists do. That is for sure historically inaccurate. I mean, there's a long history of venture investing, which was more about hardware than software. You know, it was about Apple computer. It was about UUNet building the pipes of the Internet. It was about compact computers. It was about semiconductor firms and so forth. So we know the software version of this argument, that venture capital only does cheap to scale software. That's just wrong. Now, biotech, I agree, is a tougher example. It's more regulated. And historically, you know, there have been cases early on like Genentech, which went public in 1980 with the first artificial insulin, huge, huge venture return really set Kleiner Perkins on the way to dominance over the next two decades in Silicon Valley in terms of venture investing. But that's an outlier. And it's true that healthcare investing has been tougher and less profitable. And a lot of companies have withdrawn from that. But I think the past isn't necessarily a guide to the future. And the fun thing about venture is that you have to be watching to see where basic science is generating innovation that can be commercialized. And the thing that didn't necessarily do that for a long time might be the thing that does it in the future. And now you've had a bunch of innovations from gene sequencing to CRISPR, which makes it faster and easier to develop new drugs. those drugs can have huge consequences. I mean, Moderna, the coronavirus vaccine for Moderna is a good example of this. And I think that it would be wrong to rule out the idea that even a pretty capital intensive, regulation intensive sector like biotech, you know, I think it might still work.

But what then in your mental model does limit the size of venture capital? Because as a percentage of entire capital flows, it's pretty tiny, right? Much smaller than private equity. What at the margin makes venture capital not work? You're

right, it's small. And the paradox is that the impact is big. I mean, just a couple of numbers on that. Fewer than 1% of companies that get formed every year receive venture capital backing. But if you look at the year since 1995, half of all the companies that go public got venture backing and three quarters of the market cap from those companies derived from venture backed companies. So tiny share get the money, less than 1%, but three quarters of the market cap as a result. I mean, that's the first point, just because it's small doesn't mean it has low impact. I think at the limit, to answer your question directly, there are things which are either so capital intensive, like building a new semiconductor fab, where you're really in the billions from the get-go, where venture doesn't do that. It's about- But why not?

Capital's not that scarce, right?

Yeah. I mean, venture's advantage maybe helps to answer why sometimes it has a disadvantage. So one of the advantages is the idea of stage-by financing. So you give the company a bit of money, and then if it fails early, it fails cheaply because you haven't given it a huge check right from early on. And something which is going to definitely take an enormous amount of money up front, like a new semiconductor fab. In terms of comparative advantage, venture conceivably could raise that amount of money and could conceivably go finance that. It just wouldn't be the natural sweet spot. And it would be better to leave that to, TSMC or Samsung or some established maker of semiconductors. As

you know, Sequoia has changed its rules so that it can now hold investments for much longer term periods than had been the case. If you extend this to its logical conclusion, you could imagine a future 10, 15 years from now where hedge funds, VC firms, a lot of other financial intermediaries are all blended mixes of doing varying combinations of similar things. 10 or 15 years from now, what do you think will be the unique feature of venture capital? Or do you think everything will be a blend? That's

a great question. I mean, I would say that, you know, the useful definition of venture capital is that it is an early stage venture, an adventure, in fact. And when you get to investing in companies, which are more than about $500 million in market cap, that's a different thing. That's growth equity. You know, I didn't call that venture anymore. And the reason I picked 500 million as a number is that when Amazon went public in the late 90s, its public market cap when it IPO'd was between 400 and 500 million. And now what's happened is that the IPO point has been delayed because you've got this ability to raise late stage growth capital. And more and more checks are being written, 100 million, 200 million, 300 million, into companies that are worth 1 billion, 2 billion, 10 billion. And, you know, that is being grafted into traditional venture partnerships like Sequoia Capital. So I write a lot about Sequoia. And, you know, one of the amazing things is how much franchise risk they'd be willing to take. They were a traditional early stage investment shop in 2000. And then they grafted on this growth equity business. They grafted on a hedge fund business. They built an endowment fund on top of that. And now, as you say, they've got permanent capital. So they're going to end up with multiple business lines, a bit like Goldman Sachs has multiple business lines. But I think the venture capital portion of Sequoia's business will remain the early stage part. Even

after the pandemic, it's striking to me how much venture capital remains concentrated in the Bay Area. The major deals are mostly done there. What's your mental model for that? Well,

I think sort of the lesson of Silicon Valley is that agglomeration effects or plustering effects are actually even bigger than economics has traditionally explained. In other economics, when I, you know, just reading when I was writing my book about, you know, sort of economic geography and that whole literature, and the traditional story about why a cluster is a productive thing is that you get better matching between the skills of workers and the needs of companies when you have a very deep labor market. So if you're talking about coding and a company wants to hire a particular database engineer in a particular kind of database software, in Silicon Valley, there will be the precise type of engineer that you want. Whereas if you're in a less deep pool of labor, you won't find that. And the same argument goes for suppliers. If you want to have a supplier that provides a particular kind of bespoke semiconductor, you're more likely to find that supplier locally and be able to sort of go visit them if you're in Silicon Valley

than elsewhere. But that would be why the startups are clustered. Why are the venture capital deals clustered? And furthermore, in a funny way, they're anti-clustered. Like they're not all in New York, which is our major financial center by a long ways. San Francisco is not that, right? So it's a mix of anti-clustering and then some extreme clustering for this set of deals that have some particular features. Yeah.

Well, I mean, I think that shows you why venture capital is fundamentally different to the other kinds of capital that are going on the East Coast and in Greenwich, New York and Greenwich and so forth. I mean, it's just a totally different thing. I always like to say the kind of post-war archetypical financial companies on the East Coast, the archetypes were prudential and fidelity, and their names tell you a lot about their attitude to risk, right? They were about stewarding capital and not losing it. This kind of high-risk power law return, you know, grand slam business that emerged in venture capital on the West Coast is utterly different. So it's not surprising that we would have an anti-cluster, a different center for that kind of financing. But I think it clusters because this is a business where people syndicate into each other's deals, right? So one venture capitalist will lead the series A around financing a company, and then another one will lead the series B and another one, the series C. Often a single round like the series C will have multiple investors in it. And it's just economical for the entrepreneurs to be able to visit a whole bunch of different VCs and pitch to 10 of them without having to get on a plane and fly around. So I think there are clustering effects. And also just sort of the circulation of ideas, people and money within the clusters, which is something that venture capitalists facilitate, works better when it's concentrated geographically. And that's why, you know, at least until the last 10 years, when perhaps we've gotten to the with Zoom and remote work and so forth, where this is less true than it used to be. But I think the agglomeration effects were really very, very powerful. And venture capital needed to be clustered to be most effective. And that's why you got the Silicon Valley dominance. When

is venture capital more effective versus when is angel investing more effective? Angel investing wasn't

really a factor until the mid-late 1990s. Once it became a factor, and that happened because Silicon Valley had reached a point of maturity where there were enough rich, exited entrepreneurs who had made money starting their own companies and then wanted to turn around and fund some younger people who looked a bit like them. That's what happened with Google. Google could raise a million dollars in 1998 before going to any venture capitalists because angels had become a thing. One of them was Jeff Bezos of Amazon. Another was Andy Beschlstein, who had started Sun Microsystems sometime before. So once you get angels, then I think entrepreneurs are going to want to use them because they're kind of friendly. They're typically people who know a lot technically, and that you relate to, you look up to when you're a founder. And so I think angels are useful for the first round of money, the first amount of mentoring. And then when you get a bit more further along in your company, you want the series A, which means more money, a more formal company structuring, a more sort of tougher, clearer sense of what your next milestone is that you have to reach. Otherwise, you'll be shut down. Then you go to the proper VC.

True or false? Good CEOs make good VCs.

Not always true. So as a generalization, I would say false. I mean, the premise of Andreesen Horowitz, the venture partnership set up in 2009, was that to be a good venture capitalist, you had to be an entrepreneur. And you had to have run a company and understood what a startup is like from the inside. And that just turned out to be false. Andreesen Horowitz has now given up the rule that to become a partner at the company, you have to have been a former founder of a company yourself, because it just isn't the best way necessarily. It's one way you can choose a good venture capitalist, but it's not the only way by any means. Why

has Mike Moritz been so good at VC? Wow,

that's a great question. I think he was willing to do two things. it came to company founders, he had the ability to enlarge their sense of themselves. He could intuit what they were doing. He could get on people's wavelength and not only understand them, but sort of understand them better than they understood themselves and see more potential in their project than even they saw when they looked at it. So, you know, the first example of this is Yahoo, where, you know, Jerry Yang and David Filo were, you know, in their sort of porter cabin on the Stanford campus. And they thought they were building a hobby type thing. And they were sort of proud of Yahoo as a directory of the emerging Internet. But, you know, Moritz showed up, listened, understood and said, this is the new Apple. You are going to make something with a quirky name. Apple was quirky. Yahoo is quirky. You're going to have a brand and you're going to be the face of a new phase in tech history. So he enlarged their sense of themselves. And he was just great at delivering that kind of call to greatness speech. He sat down at PayPal with one of the founders of PayPal, who was resisting the idea of a merger with the Elon Musk rival, which was called X.com. And he said to him, listen, you know, if you do this merger, I will never sell stock in the company and you will build something that makes history in the valley. And that sort of call to greatness, you know, was inspiring. So that's one thing, that EQ to get the best out of people and make them be even more ambitious than they already were. The other thing Mike Morris did is he risked the franchise of Sequoia. You know, he did a bit what I was talking about before, he was willing to go into China, which was a whole different challenge, go through a period when he had to fire the co-founder of Sequoia China because it wasn't working out, but stick with it. And if you ask the question, what is the top venture capital company in China? The answer is Sequoia China. It's the same as the answer to the question, what's the top venture capital company in Silicon Valley? It's Sequoia. But

that's a bit anti-clustering, that point, right? It

is. Yes, yes. So China requires a whole, and in a way, it's a kind of confirmation. You're going to say I'm twisting the argument here. But China, of course, is its own ecosystem when it comes to technology, because the government erects these barriers, which makes it tough for US companies to compete there. And so Chinese giants have dominated Chinese tech. But they've been funded, at least in the early phase of the digital economy in China, almost always by American venture capital companies. So Sina, Sohu, NetEase, the early sort of web directories in China, all funded by Americans. Then you've got Baidu, Alibaba, Tencent, Ctrip, all of these companies backed by Western VCs. And so what it shows you is that just in the same way that in Silicon Valley, the VCs came in and built this cluster, which had great circulation of ideas and people and money around the ecosystem. So too, they repeated the same trick in China. And they built a new, it's not really quite as tight of a cluster, because it's split between Hong Kong and Shanghai and even Hangzhou and Beijing. So it's not like Silicon Valley, it's multiple cities. But it is a kind of China cluster unto itself. But interestingly, built with the same Silicon Valley DNA, the same lawyers, incorporation in the Cayman Islands for companies, use of American style, employee stock options, which were just not a thing in China until American Silicon Valley lawyers showed up in China and explained to people how stock options worked.