“In 2014 we ought to be at the point where everybody can get basic shelter, transportation, education and other services without bankrupting the country.” — Ben Horowitz
“It’s basically prime-time now, and in the next five years, to think about every business, every industry and every field and say, how can we reinvent it?” — Marc Andreessen
Around the firm we use “software eats the world” as a guide for how technology will impact every industry and every person — from education to healthcare and government. As a16z marks its five-year anniversary, we turned to the fellow who came up with the thesis, Marc Andreessen, to explain how it’s rippling around the globe. And to break down all the implications for building companies and managing people we tapped a16z’s cofounder Ben Horowitz. With both “a” and “z” in the room, this segment takes stock of the technology industry and startup ecosystem – from entrepreneurs to investors, and where it’s all headed next. Part one of two.
Michael: Welcome to the a16z podcast. I’m Michael Copeland and we have with us today both A and Z, Marc Andreessen and Ben Horowitz. Welcome, guys.
Ben: All right. Thank you, Michael.
Michael: Right about now it’s been five years since the firm was founded, since you guys founded the firm. We thought it was a good time not to look back so much, but to sort of take stock and talk about some tech trends that you’re seeing now and then spin things forward. So, one of the pillars that this firm really rests itself on is “software eats the world.” And Marc, just to refresh us, we use it as shorthand around here, but what is the thesis? Explain a little bit and then let’s get into where we are in that thesis.
Marc: So, yeah, the thesis is that the computer itself is only about 75 years old, right? The computer was originally invented to crack codes during World War II, and then got adopted by governments and big businesses in the ’60s and ’70s. And then the PC arrived and then rich people in the developed world, US, Europe, were able to get PCs in the ’80s and ’90s. Leading up to basically about five years ago, you lived in a world in which maybe a billion, billion and a half people had PCs. And then some of those PCs were on the Internet. And then at some point all those PCs finally got on the Internet. That’s all fine and good, but that sort of meant that most people in the world didn’t have access to computers and most of the people in the world weren’t on the Internet up until a few years ago.
And then starting in 2007, with the iPhone, the smartphone comes out. The smartphone finally packages computers in a form where everybody on the planet can have one. And this is the thing I’ve been talking a lot about that people occasionally get angry about, because they think it won’t possibly come true, but I’m convinced it’s going to come true, which is I think everybody on the planet by the end of the decade is going to have a smartphone. Effectively everybody. Almost everybody. Including in places where it’s still hard to get electricity or water, people are going to have smartphones. In fact, the big thing this year is the rise of the $35 smartphone in countries like India and Pakistan that are just exploding in just giant volume.
So, I think we’re going to live in a world by the end of the decade in which there’s five, six, seven billion smartphones in people’s hands, which means five, six, seven billion people in the world connected to the Internet with what we would consider modern tools and technologies and access. And so, that’s a world we’ve never lived in, right? That’s a world where everybody’s connected. That’s a world where everybody has access to the technology that we take for granted today. So in that world, you can start to think… I think you can think very profound thoughts about… like what happens in many different fields and businesses and industries as a consequence of that?
So, what happens in the media business is an example – what happens when everybody has the ability to consume news online? What happens in the TV business, when everybody has the ability to get video streams straight to their cellphone? What happens to retail when everybody can engage in e-commerce? What happens to education when everybody can take classes online through their smartphone? And so, I think the next five years is basically prime time to think about every business, every industry, every field and say, “Well, how can we reinvent it now knowing that software can basically play such an important role in everything.”
Michael: So, where are we? And what are you guys seeing in terms of the verticals that are being eaten? And then I want to get at what it’s like to be eaten, too. But what are the verticals that sort of… “fallen” is the wrong word, but fell first to “software eats the world” and what comes next?
Marc: Well, I would characterize it as building. We tend to be glass-half-full people. We think there’s huge opportunities that flow from this. Media’s been an obvious one. Financial services, software’s taking over at a very fast pace. And then retail, e-commerce has been a profound revolution over the last 10 years and that will continue. I think the three big ones that we think of a lot about for the next five or 10 years are… and they’re really, really big domains. These are long-term projects, but healthcare, education and then ultimately government.
Michael: So, those three that you mentioned have been the most intractable, it seems, and we’ve been promised… there’s electronic medical records. Obama’s got a great website, but that’s not software eating government. Like why now, and what makes it different?
Marc: Why now? Because you can. Because it’s possible. Because it can be done. Because everybody’s going to have a smartphone. Because everybody’s going to have to be online. These are very, very big fields. Very big entrenched interests. Lots of regulation. Like there’s nothing easy about it. But because they’re so big also makes them so important, right? So, I’ll just give you my perspective, which is healthcare and education, one of the things you always look at is you’re like, “Well, healthcare and education is something that you want everybody on the planet to have access to.”
Michael Copeland: Right.
Marc: But at the same time, if you just try to do that with conventional ways of delivering those services, you’ll bankrupt everything, right? You look at the US fiscal situation over the next 50 years and you look at this massive debt. A very large amount of it is geared towards healthcare, paying for people’s healthcare. And a lot of it then is geared towards education and this huge crisis we have around things like student loan debt. And so, healthcare and education suffer from something that economists call “Baumol’s cost disease” after an economist named Baumol. And the cost disease basically is technology steadily drives down prices in sectors like media and in financial services.
But technology is not yet driving down costs in healthcare and education but it should. And so, if you can bring revolutionary technology into those fields, you could basically break the back of the so-called “cost disease.” And then you can bring the same cost reduction you see in other fields into those fields, which then means you can really open up access. And that’s the critical thing. We need to get every kid on the planet access to what we consider today to be a top-end Ivy League education. We need to get every person on the planet to have access to what we would consider to be a modern healthcare capability. The only way to do that is to apply technology. Otherwise, you’d bankrupt the planet trying to do it with the current techniques.
Michael: Ben, I want to bring you into the conversation. “Software eats the world” is great if you’re on the side of the coding keyboard where you’re out there and you feel empowered to go after healthcare or government or whatever else it is. But what if you’re on the other side, and you’re being eaten. How do you view that, and then how do we bring those folks along?
Ben: Well, I think in the first place, the misunderstanding of it is that those jobs get eaten and there are just fewer jobs. I think the history of the country, and the history of the world, has been the jobs 50 years from now are not going to be the same jobs that we’re doing today, although we had a very long time when most of the jobs were farming. And thankfully, most of the jobs are no longer farming, because farming is hard work.
Ben: But most of the jobs today aren’t jobs that were around before. I mean, there were all kinds of jobs, like for example delivering ice was a big job. Making vinyl records used to be an important job. There were a lot of jobs that are no longer here. So the challenge isn’t the absolute number of jobs, but rather the transition from one set of jobs to another. And this is where I think I would just add to what Marc is saying that, like it’s really important to apply technology to bring down the cost of things so that you can make policy decisions to create both a safety net and education system that enables people to not starve to death or hit the streets or become homeless or really lose their dignity in the meanwhile, but then also be in position where they can go be productive again as quickly as possible. And that thing is not really affordable if all the money is going to “technology un-enabled” education and health care.
Ben: And so they kind of go together. I think that if software doesn’t eat the things that the government pays for, then we run into trouble. And you could imagine completely government funded awesome education enabled by things like Oculus, where you could put kids from anywhere in the very best classrooms in the world as though they were actually there.
Michael: Oculus, to be clear, virtual reality goggles would allow you to do what?
Ben: Well, basically, enter a virtual classroom with virtual other kids and virtual teacher, which might not sound that great if you’re at Harvard, but it sounds awfully good if you’re in an innercity school that’s very dangerous to go to with teachers that maybe no longer that interested in teaching for all kinds of reasons. And so by applying technology to the things that cost most, you can both improve the services and reduce the cost a lot, which also frees up money for things like a really robust social safety net, which basically gets you and you ought to be… in 2014, we ought to be at the point where everybody can get basic shelter, transportation, education and other services without bankrupting the country.
Michael: Does that social safety net come from private industry, via taxes? I mean, is it government run? Who pays for it?
Ben: Yes. I think the social safety net, ideally the money would come from the government and would come from taxes. And I think it would be smart as a taxpayer to do that. But it only works if there’s real technology behind it. Because otherwise… right now we can’t even afford healthcare. And so without the application of technology, we’re in this halfway point where software is eating lots of jobs or transitioning lots of jobs. But yet the big government services have not yet been transformed.
Michael: I see. We talk about this transition point. Marc, you mentioned billions of cellphones in people’s hands, or smartphones, I should say. Computers essentially in people’s hands. What is our evidence that we’re at this transition point. I want to talk a little bit about accelerating innovation in your careers, how the present looks different than maybe it did in the mid, late ’90s or early 2000s even?
Marc: Yeah. Well, things look really, really different. So, there’s sort of this running thing in the Valley that, like it’s another bubble. People have been calling it another bubble now. This is year 10 and people are calling it a new bubble. It started in 2005, so maybe at some point they’ll be right. This thing is like this is the ’90s all over again? The thing that has changed so much is the industry and the world are so much further along than they were in the ’90s. We built Netscape the company. We were out there doing it in the ’90s. We built Netscape. We had at one point over 90% market share for all people on the Internet using our browser. And it turned out there were 50 million people on the Internet. 50 million people around the world were on the Internet.
By the way, 25 million of them were on AOL, so to say that they were on the Internet would be a slightly generous statement. The other 25 million were directly online but that was it. You could have 90% market share and you’d have less than 50 million users. And that was it. That was all that we could do. By the way, those 50 million people were all coming in, for the most part, over dial-up. So they’re coming in very slow. They’re making new TV shows in which the period in which you had these old PCs, they now look retro.
Times have changed in a freaking hurry. And so the technology world was a much smaller place and the number of people who had access to technology was just a much smaller place. Fast-forward to today. More than 2 billion people online, rising very fast. So, market expansion. We think about things in terms of market sizes, market expansion on the order of 40, 50, 60-fold over the course of 15 years, which is just absolutely amazing. Which is why you have these companies like Google and Facebook now that have more than a billion users.
Marc: I like to work with Mark Zuckerberg at Facebook, and I like to tease him with like what a special position Facebook is in, because Facebook passed a billion users a year or two ago. And it’s like, how many businesses or human organizations or institutions in history have ever had a billion anything? And it’s like… it’s basically the Catholic Church and the Coca-Cola company. And that’s basically been it.
Marc: And then like Google and Facebook are number two and three… three and four. And then WhatsApp is going to do it. And then there’s going to be a whole bunch of others. iPhone is going to do it, and Android is going to do it. And so the world has really opened up and expanded very fast.
Michael: We invest in all kinds of things here. Software based mostly, but drones, BitCoin, 3D, virtual reality, Oculus. How do you guys keep up? If things are different and the pace is different, how do you guys personally keep up? And then how should the rest of us think about keeping up, if even that’s the right thing to think about?
Ben: Well, I think it’s like a lot of things that are extremely complex if you don’t have some unification understanding of what’s going on then it does get very complicated. But one of the tenets when we started the firm was we were only going to invest in things that we understood. And that meant things where software was a core intellectual property. And if you look at what’s happening across all of those things, they’re different manifestations of software becoming a more powerful force in the world due to all the things that Marc talked about. The improvement in the underlying platforms. The improvement in the programming languages and connectivity and all these things.
And so whether it’s software eating money or software eating entertainment, it’s all comes from the core root phenomenon. And so these things are relatively easy to understand quickly. Now some of them… the details of every one of them is complex, but the nice thing is it’s our favorite thing to do in the world, so it’s a great job in that way.
Michael: We’ve talk about software eating certain industries? One could imagine mostly software based or digital-based industries. But it’s moving into the physical world. Where do you see that most starkly and does it look different as software eats manufacturing, robotics, et cetera?
Marc: Yes. Basically, I think the key assumption I make is every physical thing is going to become smart over the next 10 or 20 years. This one’s going to take time, but it will happen. So basically every physical thing, whether it’s a toy or a light bulb or a doorknob or a building or your desk or whatever — take your pick, your car. They’re all going to become smart. By smart, I mean they’re all going to have chips built into them. Everything is going to have a chip in it.
Everything is going to be on the Internet. Everything’s going to be connected. Everything is going to have data. Everything is going to be responsive to human beings. And by the way, I mean like everything. One of the most… a very exciting trend that we’re not yet invested in, but I’m very excited about is I think… I’m convinced in the future pills are all going to have chips in them. So for example, like every pill you take is going to have a biodegradable chip in it.
Michael: We’ve been talking about “smart refrigerators” for a long time. Why do I want my refrigerator to be smart?
Marc: Because you want your refrigerator to know, for example, when the food in it has spoiled so that you don’t get salmonella. Very fundamental. You might like to know… you’re at the grocery store, you might like to know what you’ve actually got, what you actually need to re-order. And it’s just been the nature… the criticism of this stuff is always so interesting, because basically these products come out and they’re basically experiments. So people basically try different kinds of things. Some of them work and some of them don’t.
And there’s always this criticism of like, “Oh, the designers of these products didn’t think everything through,” as if there’s some magic process where you can send a bunch of geniuses off to a remote island and they could design everything perfectly and then build it right the first time. But that never happens. What ends up happening there are tons and tons and tons of experiments, right? Out of which you finally then get the things that are going to go broad.
And so, people experimented with how to bring computers to consumers for 10 years before the IBM PC went broad. People experimented. The smartphone concept is not new. It goes back to 1997. It took 10 years to get to the iPhone… to the actual product people want. The Internet was in existence for 30 years before we made the web browser and finally figured out how to make the Internet accessible to everybody. So that’s why I say this part of it… sort of “Internet of things” concept or making the real world smart is a long-term thing. It doesn’t just instantaneously happen. But it will happen. And it will happen through the process and experimentation that we see happening right now.
Michael: Let’s switch gears a little bit. You guys started the firm now five years ago in the teeth of the worst recession since the Depression. Is that fair to say?
Marc: It was the worst economic crisis for venture capital in 40 years is what everybody told us.
Ben Yeah, that’s what they said.
Michael: So it seemed like a good idea at the time. And turns out it was. The venture capital industry has been shrinking ever since. So, describe for us where we are in the VC industry. And then I want to get at how building companies and venture capital relate to each other today and as you see it going forward.
Ben: Yes. The shrinking in the number of the firms has really been a wonderful thing and very healthy for both the industry and very good for entrepreneurs. One of the aspects of venture capital that bothered us a great deal when we started the firm was that it was an industry that really was in the dark. And what I mean by “in the dark,” it was completely invisible to what was going on, what was behind these firms and so forth. And like anything that’s in the dark, that’s where the nasty business happens in terms of everything from kind of bad things growing in it to ethics and everything.
Whereas things in the sunlight, that’s when everybody behaves and is at their best. I mean, it was just the nature of how it evolved and so forth. It was a relatively shadowy industry. And as a result, it was very difficult for an entrepreneur to figure out what was what at a very basic level. And so there were these phenomenal firms like Sequoia or Benchmark or Greylock, but for an entrepreneur to really understand the nuances between the ones that were really great to work with or just people who simply had money and had no idea how a startup should get built was really impossible to tell.
And I think one of the contributions that we made was just the way we went about things, partly because there was counter-programming, is we put a lot of light on it. And that also, combined with what was going on with social networking, what was going on with the Internet and going on with real entrepreneurial community made entrepreneurs really aware of what they should expect and what they should want from a venture capital firm. And most venture capital firms, as you would expect, just didn’t stand that test.
Then at the same time, a new tier of venture capital emerged, angel investing or seed investing. And this enabled, for the experimental things, where you really just needed capital and maybe a little bit of advice, there was a whole new functionality at that layer. So, you had the landscape bifurcate into seed investing, where you need a little money and a little help.
Ben: Or venture capital investing, where you need a lot of money and a lot of help. The whole middle, like we don’t know how to help and we don’t have that much money, like that big set of hundreds of venture capital firms became no longer interesting. And so, the result of the landscape, I think, is it’s a way better time to be an entrepreneur now raising money than it’s ever been before. It’s easier to raise money in small amounts. And then if you raise money in large amounts, you get likely a much better investor than you would have got in the past.
Michael: We’re seeing large rounds raised, but I want to have you guys answer the question of why do, if I can… myself and three of my friends and virtually no money and just a credit card and Amazon Web Services, if I can build a company and start on my way, why do I need then traditional venture capital, even at the high end, like that big money? Why is that still necessary?
Marc: Cheaper to start. More expensive to grow. So, it goes back to what we were talking about before. So the market sizes are much larger. So, the prize is bigger. The market is larger and it takes more money to be able to build a company in the market. And so, yeah, you can start a lot of these companies with three laptops and three lattes a day and you’re off to the races. Three kids living on ramen noodles. But at some point, you need to build the company. And you need to build the company that is then going to go take the market. And the market is big. And the market is global and you’re going to need a company around that.
You’re going to need a sales force. You’re going to need marketing. You’re going to need a big development organization. You’re going to need customer service, customer support. You’re going to have a big recruiting campaign. You’re going to need partnerships. You’re going to need expansion capital. All these things kick in. And so these companies almost never stay small. It’s extremely rare that you’d see any company that does anything big in tech that doesn’t end up raising…doesn’t end up, number one, raising money and then number two, if you’re going to raise money, raise money from venture capital.
Michael Copeland: You mentioned “global.” I understand global markets, venture capital has done very poorly as a global enterprise. Is that going to change? Do you see that changing at all?
Ben: Well, I think it will change, but not in the way a lot of people are expecting. And to some extent, it is changing already in the sense that some really large companies are coming out in China right now. Like really big. Alibaba is a monster in a good way. So, it will change. I think that people expected that, “Oh, well, if we just focus on building software companies like Silicon Valley, we can be like Silicon Valley.” And that’s been largely mistaken that you can apply policy and tax breaks and even have a good university, but Silicon Valley is a network effect. And if you’re competing with a Silicon Valley company without the level of engineering talent, without the level of executive talent, it’s still rather hard.
Marc had a really good… I can’t remember. I think it was a series of tweets rather than posts. But a really good series of tweets which just said, “Look, there should be 50 Silicon Valleys, not one.” Or actually, that was an article. Fifty Silicon Valleys, not one, but they should not be Silicon Valleys. They should be doing things in other areas of technology. Other areas of innovation. And some of that can be facilitated rather than tax free zones, more like regulatory free zones, where things that are very difficult to build here can be built elsewhere. And the world will certainly be a better place if there are lots of new companies coming from lots of places all over the globe. And I think that Silicon Valley set such a great example of what’s possible that that’s going to happen.
Michael: Do you see evidence already, Marc, of those kind of zones and Silicon Valleys as it were?
Marc: Yeah. Well, there’s two ways to come at it. One is on the zone side. So, there is actually a long history of this sort of concept of a special economic zone. And actually, countries like Hong Kong at one point and actually China’s a very aggressive user of special economic zones. Actually also the prime minister of Japan, Abe, is now proposing special economic zones in Japan to be able to restart growth. And so, you certainly have a long history with that approach. And then you have examples today where that thing is playing out.
One very interesting example is South Korea. South Korea has famously been much more liberal in terms of legalizing new stem cell based health projects, which look… stem cell research for regenerative medicine looks incredibly exciting. And looks like we might be able to grow new organs and do all kinds of things. That research has historically been difficult to do in the US for regulatory reasons. And so, you see more examples of that. But I’ll tell you the other thing you see is just an explosion of entrepreneurial energy and enthusiasm all over the world. You just see… and we see it all over the place. You see it all throughout the US. Like 20 cities in the US now.
And then you see it all over the world. The most vivid example is a good friend of mine, Chris Schroeder, wrote an entire book over the last three years called “Startup Rising,” which is the story of high tech entrepreneurship in the Middle East. He traveled all through Jordan and Syria and Egypt and all these countries, going through all these dramatic changes. And in every single country, he found high tech entrepreneurs and startups, building unbelievable new companies and products. And most significantly, he actually just got… there’s like 15 visas a year for business travelers to Iran. And so he just went to Iran after he wrote the book and he just came back. It turns out in Tehran, there’s a thriving underground high tech startup scene.
Marc: And there’s all these entrepreneurs who want to start all these online businesses. If you want to meet… a 22-year-old who’s incredibly frustrated by the fact that his entire country has been embargoed, he or she in the Tehran high tech scene. So even in countries where you’re most concerned from a geopolitical standpoint, there’s enormous energy and often youth energy towards building these businesses.
Michael: As a firm, do you need to tap into that directly or indirectly? Or is there… how do you plan on that going forward, if you do at all?
Ben: One thing about investing is it’s important to keep in mind that you don’t want to try to do every opportunity. Every firm doesn’t have to be in every deal. And if you try to, unlike running, say, an enterprise software company, where you absolutely do want to get every customer, in investing you want to stick to the things that you’re best suited to do, where you’re the very best firm possible.
And for us, that’s primarily software based things and primarily Silicon Valley, just because this is what we know. We understand the culture. We understand the technology. And look, different cultures are different. The world may be getting flatter, but it’s certainly not flat in terms of culture and in terms of motivation, in terms of what stock options mean and the difference between beer drinking countries and wine drinking countries. And you have to know all these things to invest effectively globally. So it’s not… we can have a very good business investing in Silicon Valley. And then whether we expand that or where we expand that, I think, is TBD, but it’s not critical for survival in the way it would be if we were, say, a software company.
Marc: I think I’d say a lot of our companies actually bear directly on this, right? And so companies like Facebook and Twitter as platforms for global communication. Entrepreneurs are communicating on Facebook and Twitter every single day in incredibly beneficial ways. Or another example is a startup that we’ve seed funded called “Teleport” that’s making it much easier for engineers to consider relocating, like where to locate a new company or how to get talent to be able to move to a new location.
Or for that matter, Airbnb, which makes it… or Lyft, which make it much easier for people to travel and be able to temporarily live in other places as they go around to where the different opportunities are. So, one of the things we view ourselves as doing is funding the platform companies that are building this world in which there’s going to be a lot more entrepreneurship.
Michael: That’s part one of the Ben and Marc podcast. Check out part two, where the discussion enters the realm of disruption theory and how that helps the firm decide what not to invest in, as well as how large companies innovate and what Ben and Marc wish they had been told as young entrepreneurs and more.
The a16z Podcast discusses the most important ideas within technology with the people building it. Each episode aims to put listeners ahead of the curve, covering topics like AI, energy, genomics, space, and more.