[MUSIC] [APPLAUSE] >> Well, Mark welcome back to Stanford and from one Midwestern to another,
we're honored to have you here. >> Great, thank you, thanks everybody. >> So, I'd like to begin by exploring
your intense intellectual curiosity, recent wired article profiled your library
And instead of it being in your home, it's actually in the lobby of A16Z. So could you unpack for us why you put it in a place open to
visitors as opposed to your home? >> Yeah, well, so two reasons, one is
if you've been to other Venture Capital offices, they look a little
bit like insurance companies. It's all tombstones of long dead
companies that they sold and took public once upon a time, so. We just thought that was
kind of depressing, so we decided to do something different. So the message that we're trying
to send with the library, and it's a very deliberate message,
kind of goes as follows. Which is, the great thing about
the valley, especially in our time, is the sense of newness and
the sense of the future. And the sense of audacity that
there are radical new ideas and that there are new people that come
to the valley all the time and can pursue things that have
never been done before. So, very future oriented and
that is an enormous strength. Elon Musk talks about this,
kind of most vividly, he describes, he says you always want to think for
first principles. You want to kind of not take any
received wisdom from the past, instead you want to think from scratch and
he makes very good method assumptions. And the past may no longer be true and
so if you rethink things from scratch, you can reach different conclusions and do things differently than
could've be done before. And I think there's
a huge strength to that, I think there's also a big problem with
that, or it's an incomplete theory. Which is, there are also thousands of
years of history in which lots and lots of very smart people worked very
hard and ran all kinds of experiments on inventing new technologies,
or creating new businesses, or new ways to manage or
new ways to lead or all kinds of things. And they ran all these experiments, and they ran these experiments
throughout their entire lives. At some point, somebody put them down in
a book, and for very little money and for a few hours of time, you can literally learn from
somebody's accumulated experience. And I think there's just, there is so much more to learn from the past
than I think that we often realize. And so I think you could productively
spend literally all of your time, all your life, just reading the
experiences of great people that have come before, and I think you learn every time. >> So, speaking of history, one of the significant elements in that
library are the history of Hollywood. And I'm curious why that fascinates you,
and what parallels, if any, you might see to the development
in Silicon Valley. >> Yeah, so in a lot of ways,
Hollywood is very unlike the Valley. Probably the biggest
different is in the Valley, if you say that somebody's
startup is just all story. And no substance,
it's a very offensive thing to say. In Hollywood,
they take that as a huge compliment, because the whole point
is to tell a great story. And so they view that as a clean win, there's another big difference actually
which I find very, very instructive to think about which is people in the
valley often think our lives are harder. Or it's harder to start these companies or
it's hard to compete for the business. Hollywood is much more difficult, it's
a much more actually hardcore competitive, even vicious business environment. And so it's always good, I think, to have
perspective of people who have it harder than we do and
still manage to pull things off. I think that's true of the big
successes in Hollywood. The thing that they have
in common is we in tech and they in entertainment are the two
big original California industries. Right, I think between the two industries
we kind of embody the spirit of California and the spirit of the West in the US. Once upon a time there was a gold rush,
then we pulled out all the gold and we're like,
okay now we've gotta make some new gold. And we make it through technology and
they make it through storytelling. But very similar entrepreneur
personalities, very, lots and lots of people. Up here everybody's got a start up idea,
down there everybody's got a screen play. Lots and lots of people come from
all over the country, migrate in, almost everybody's from out of town. People have a real sense of what great is,
one of the things about the valley that I think is amazing is,
it's impossible to be in the valley and not have a sense of what
a great technology is. because you have your nose rubbed
in it all the time, it's like yup, there's Google, okay,
all right, that's the bar. In Hollywood, and for what they do, it's
the same thing, they made everything from Casa Blanca to Star Wars all the way
up to all the great movies today. And so it's just an incredibly
high quality, high caliber, high achieving culture. Incredibly energetic people and between
these two industries, California is the I think the sixth largest economy
in the world as a stand alone. If it were a stand alone country it would
be the sixth largest economy in the world. And I think we are the two
industries that power that. >> So as you think about the adventure
capitalists you take in a lot of information from a lot of different
companies, different trends. This seems to be a parallel
to how you approach learning taking in a wide variety of history and
other elements from books and stuff. How do you think about learning? How do you go about actually figuring out
what you want to focus on and, develop? >> Yeah, so there's part of what we do, so the way modern adventure firms work
is kind of strange and interesting. So we get inbound pitches, so we get inbound 2000 pitches a year,
from qualified referred entrepreneurs. So somebody who's kind of hitting
the basic criteria, have been able to network in the Valley, and
been able to connect with someone we know. A very qualified set of founders, so
we see 2000 a year, and so a lot of it is just meeting with all those founders,
and trying to survive the tsunami. I read, there's all these books, there's a
great book, I disagree with the book, but it's a very well-written
book called The Rise and Fall of American Growth by
an economist named Robert Gordon. And he says, basically, innovation is
over, and all the great ideas have been taken, and from here on out just
everything's going to get boring. There's no new fundamental technological
breakthroughs on the horizon. Some days it would be
a relief if that were true, because out here,
we're just drowning in it, right. It's just this constant
tsunami of new ideas and new thinking, and
new people with new thinking. And so a lot of it is just being in
that fire hose, that's part of it and then like I said, I try to compliment
that trying to be proactive. And so, one is trying to explore new areas
of science in particular and there's, I think, some very interesting
new areas of science right now. Many of which are happening right here at
Stanford, and then the other is history, I always go back to history. And I always go back to try to understand,
okay, when you had a fundamental, I'll give you
an example, self-driving cars, right. So self-driving cars are a fundamental
advance in computer technology. They're a fundamental advance
in the transportation industry. They're obviously going to be a very
big impact on the automotive industry, which is a very big global production
industry, consumer industry. It's like okay, well what happened, the thing about the impact
of self driving cars. Well let's start by understanding
what were the impact of cars, right. And there were very, very interesting
lessons from how cars rolled out that I think can be applied 100 years
later to figuring out some of the, both the issues and
the opportunities of self driving cars. And so I try to go back and look for the patterns that can kind of be more
helpful in predicting the future. >> Speaking of science, backstage you
mentioned that at this point in our history, 90% of the scientists that have
ever existed and engineers live right now. >> Yeah, living or working today. It's 90% of all scientists, 80 or 90%,
I think it's 90%, of all scientists and engineers, who have ever lived
are living and working today. >> So what does that imply for humanity,
going forward the next two decades, and even the next century? >> Well, this is why I'm such
a gigantic bull on future innovation. So there's two basic theories,
the Gordan theory is basically, it's a question of low-hanging fruit. And so electrical power was low-hanging
fruit, steam engine was low hanging fruit. Electromagnetic waves were low-hanging
fruit, and from here on out, everything gets much more complex. And so it's harder, you're putting more,
diminishing returns. You're climbing higher and
higher, and putting more and more effort in for less results. And actually, by the way, what Gordon
points to that actually justifies that point of view is Measured productivity
growth across the economy is quite low. Right?
And so, right, because basically there's two arguments, critics of technology
basically make two arguments. One is there's no innovation,
there's no productivity growth. And the other is, there's so much innovation that it's
going to destroy all the jobs. Right, and what's important to realize
is they're diametrically opposed arguments, right? They literally, factually,
logically can't both be true. Either you have super low productivity
growth and nothing changes, right, on one side of the argument, or you have
super high productivity growth through technological change and
everything changes. But they can't be reconciled,
they can't both be true. However, there is one way they
actually can both be true, which is where things
get really interesting. Which is they vary by industry, right? And I actually think this is the part
of his argument that I agree with, which is that there are industries
like technology, consumer electronics, media, retail as examples, where there's extremely rapid productivity
growth in technological adoption. And then there are other
industries like healthcare, education, law, government where there's
a little technological adoption, a very little productivity improvement. And so one of the things we try to
do is try kind of navigate through the differences industry by industry and
try to kind of understand how to apply the technology particularly the areas
that where it hasn't been applied before. >> To the point of jobs in our economy, one of the main criticisms of self-driving
cars in the future is there's 5 million truckers all around America that
the role of the economy may diminish. We see companies like Facebook and
Twitter compared to their equivalent market capitalization companies in the
past decades having far fewer employees. So are you of the mindset that
the employment opportunities will actually increase as technology proliferates,
and why do you think that? >> Also there's an argument from
logic which is sort of the doctrine of free trade, specialization of labor, it's kind of well established in
economic theory why that would be true. Then there's observation by history, which as we're sitting here today after
300 years of technological change and there are more jobs in the world
than ever before, right? There are more jobs in the US
today than ever before, there are more jobs around
the world than ever before. There are more people who are employed. By the way, income levels have
never been higher, right? For all of our issues of stagnation of
income, overall income levels across the entire population of
the world are at an all-time high. And so, if technological change were
going to cause elimination of jobs, one presumes we would have seen it by now. >> Yeah.
>> Right, again, the counter argument to that is but you're
not taking an account like some dramatic breakthrough in artificial intelligence. At which point I start accusing my
opponents in this argument of just hand-waving. If it's so easy to make a super genius
AI robot, why don't you give it a shot? We are looking to try to fund it, in fact come in and pitch us,
we will happily invest. It may be the last
investment we ever make, but it will be a very successful one. >> [LAUGH]
>> So, we're totally on board with that. >> Yeah. >> So, there's sort of
the long run historical view, which I think is pretty well established. Then the questions get very specific. And these are some of the questions that
have come up in this election cycle, which is okay, the 5 million people
involved in driving, one form or another are transportation jobs in the US,
transportation delivery services, trucking, things like that. And you're like, wow,
5 million jobs seems like a lot of jobs, and it is a lot of jobs. This is where I think
people underestimate. Even with our low rate of
productivity growth today, people underestimate the rate of change
in the economy that happens anyway. And the thing that happens there is that
what gets reported are always headline numbers and the headline numbers
are always the net numbers. And so any given month, they'll report. It just came out, I think, two days ago. It was last month or I forget, 170,000 new
jobs or something like that in the US. And so you're like, jeez,
you're not creating very many jobs. How can you possibly create 5 million? So that's the net number,
it's not the gross number, right? So the gross number is much higher. Both in terms of jobs destroyed,
and jobs created. And the top line number there,
is every year in the US, on average, about 21 million jobs are destroyed, and
about 24.5 million jobs are created, for a net add over time of about
3.5 million jobs a year. And so the real answer to how do you
replace 5 million jobs is we already replaced that in less than a quarter. Today, just this quarter today, just this
calendar quarter with the technology we already have, we're already going to
create more than 5 million jobs. 5 million gross new jobs. And so change is happening in
the economy all the time, right? And by the way, it's happening whether or
not we elect Hillary or Trump, it's happening whether or
not we have steel here, somewhere else, it's happening whether or
not Silicon Valley rises or falls. The change is just going to keep coming. >> Yeah. >> And then to me that gets back to
our day job and what I think is so important to focus on which is okay. So, then it's saying okay,
the world is going to change, the jobs are going to change. How do we set people up to be able
to take advantage of the change? Right?
How do we have change work for people? How do we expand opportunities? Why education is so important. That's why financial services
actually is so important. That's why it's so important to be able
to raise money for new businesses. That's why it's so important for people to be able to move from company
to company to be able to get new jobs. And then the conversation I think we ought
to be having is the next hour's worth of conversation on all the things that we
could do to have more people have access to all the opportunity that the new
technology in many cases is creating. >> Yeah, you mentioned Robert Gordon. Who else is in your intellectual diet
that you turn to when you want to think about deep issues like change and
funding in our economy? >> So I have a little mental
model of Peter Thiel. I have a little simulation of Peter Thiel. He lives on my shoulder right here. >> [LAUGH]
>> And I argue with him all day long. >> Who wins? >> I always win. >> [LAUGH]
>> Face to face, it's slightly more challenging. So, I'm a big believer,
goes back to history. Charlie Munger talks a lot about
being able to learn from the past. He always says, the way he describes it,
basically he calls it a build of mental model of a person and
you want to kind of try to understand. And it's hard because the people we're
doing this for are very smart people. And so you have to try to simulate people
who [LAUGH] in a lot of cases are smarter than you are. We want to build a mental model, and
that could be a mental model of, in modern times, Steve Jobs or Jeff Bezos
or somebody, Larry Page, Elon Musk. We have mental model of somebody from
the past, Thomas Edison, J.P. Morgan. It can be a model of
just somebody you admire, a great philosopher or religious figure. And you want to kind of construct
a model of how they think. And be able to be very objective and fair where you can kind of think
things through from their standpoint. And then you kind of you have
your own view on things and then you try to run through kind of in
your head what you know of them and so okay here are the conclusions
that they would reach. And if you put enough time into that, you start to be able to have these
conversations I find sort of with yourself. People might look at you funny
while it's happening but you get to kind of engage in this dialogue
and so I'm a big believer in that. The three people in the valley who I
do that with most often, my shoulder, Peter, Elon Musk and Larry Page. The three of them I find pushed the
boundaries of what technology can do and what Silicon Valley can do
further than anybody else. I think they're the three most
audacious people, I think, probably who have ever
worked in the Valley. >> Yeah. >> And then you've got your
models of the great founders. Steve Jobs, obvious, and many other
folks who built amazing companies as entrepreneurs and then the new one for
us is the great CEO's, right? So Andy Grove being kind
of our base case for that. So, and whenever we talk about management
in our firm, we're always trying to index back on okay, what would Andy Grove
say if he were sitting here today. Good news that Andy is that he was
very forceful in expressing his views. And so he's on the record with
on most of the important topics. And as a very good kind of reference
point for thinking about those issues. >> What I'm struck by is you love
to engage with people who may disagree with you? >> Yeah. >> What's the broader lesson here, for those of us here in business school
who are thinking about these things? >> Yeah, so I just think, and again,
this is the kind of thing where, this way. Adventure Capital,
[LAUGH] venture capital. Venture capital is a funny business. Let me say why I feel so strongly on this. So in venture capital, there's two
kinds of mistakes you can make. There's a mistake of omission. Or there's a mistake of commission. Which is mistake most people
make because I make a mistake. I make a decision, I invest in a company,
I lose all my money. It's the mistake everybody
kind of thinks about. And then there's the mistake of omission. Which is Mark Zuckerberg walks in the door
at venture capital firm XYZ in 2004. And, you know,
like what is this little kid doing? And this idea's crazy. And Fredster proved that
this could never work. And this is ridiculous, which is
what he got told by a lot of people. In the venture capital business, every highly successful VC has made
mistakes of omission, really big ones. Of companies that they had the chance to
invest in, they should've invested in, they didn't invest in. It turns out the mistakes of Co-Mission,
they matter, but they don't scar you for life. >> [LAUGH]
>> They just fade into history. The mistakes of omission, right? It's a asymmetric payoff. When these companies work, you know,
they get a thousand extra. If you lose, you lose 1 x, if you win,
we win 1000 extra, 10,000x. The other is,
the companies succeed, right? The companies you passed on succeed,
and they torture you, right? >> [LAUGH]
>> Because, you know. >> So, who's tortured you recently? >> No comment
>> [LAUGH] >> I can't even bring myself to talk about it. You pull up Tech Meme in the morning or
whatever your choice of thing is, and you're just like, shit, that one again. >> [LAUGH]
>> And so they just torment the hell out of you. And so, it's a humbling job. I think the same basic principle
applies to, but much more broadly, I think the same principle, especially for
people kind of in this ecosystem, I think the same principle applies to
even people, that mentality applies to people with kind of what you consider
to be more normal careers, right? So, as you think about having
a career over 50 years, you're going to make basically
a sequence of bets, right? And you're going to make a sequence of
bets of the places you choose to go and the people you choose to work with. And you're going to
screw some of those up, and you're going to make a series
of bets and decisions for places you don't want to go and
people you don't want to work with. And you're going to make those
mistakes on both sides, and you're going to have the exact
same feeling on the other side. When you had the chance to be
employee number four at Google and you didn't take it, right? That sticks with you for a while. >> [LAUGH]
>> That's a big one, rhat one tends to wear on you. And so, I think that mentality, we call
it the slugging percentage mentality. Which is basically, take the bat,
lose 1x, don't take the bat, possibly miss on 1,000x. That mentality is a very
powerful mentality. If you're going to do that, you have to
be kind of ruthlessly open-minded, right? And then this is sort of my view,
when we make mistakes of omission, why do we do it? And I think it's almost always because we
have some theory for why something's not going to work, and you try to
investigate that theory, but you know, the problem with human nature, you develop
a theory and you tend to want to prove it. A confirmation bias, they call it. And so, you develop an idea,
and then you look for all the evidence that supports it, and you
ignore all the evidence that disproves it, and so you get locked into your ideas,
right? And then by the way,
pattern matching works against you, right? Because things that didn't work
in the past, might work now. And this is the problem with history,
this is why Elon focuses so much for first principles. Just because like MySpace, for
example, didn't reach Facebook levels, the scale didn't mean that
Facebook wouldn't be able to. So, you have to be ruthlessly open-minded, you have to be constantly willing
to re-examine your assumptions. And to do that you have to try to figure
out a way to not get emotionally tied to your beliefs. So as I always like to say, there's
this software term called sandbox, and there's the idea that you can
basically run code on bare metal, you can run code on the chip, or
you can run code on a sandbox. The idea of a sandbox is, it's a contained
environment, where if the code goes bad, if it's Malware, or
it's an AI that wakes up or whatever, you can nuke the sandbox and
nothing bad happens. I think we need to run ideas in sandboxes. Beliefs should be run in sandboxes. As much as possible,
you shouldn't self identify with beliefs. We should treat them all as
sort of more abstract objects. And be willing to pull them in, think
about them, and then put them back on the shelf, as opposed to saying,
this is part of my identity. And it's a very hard thing. It's almost a zen-like,
you have to take the ego out of ideas, which is a very hard thing to do. >> Absolutely. You never went to business school. You're in an audience full of folks who,
obvious, are in business school. What do you hope that
we leave here knowing? Both about the Valley and technology
that you learned from actual experience? >> Yeah, so the big thing,
it all sounds so easy, from a distance. >> [LAUGH]
>> It all sounds so straight forward, like the case makes it all so clear. So, strategy is important. Strategy is thinking about like, okay what are the decisions that should
be made over time, is very important. I like to come at it, particularly,
in terms of running something, like running a company. Making strategic decisions
that affect your company. The strategic decisions sound
more obvious than they are, and one is there's just a lot of
context at the time that's missing. Even in the histories
it's usually missing. The other thing I think, the biggest thing
I find newly minted MBAs have a hard time wrapping their heads around is,
the abstract idea of what the right thing to do is in a business
situation is all well and good, but as an actual manager in an actual
business, you have other constraints. In particular,
you have organizational constraints. You have constraints as to what
your actual organization can do. And those constraints have to
do with your employee base, how well your company's organized and run,
the existing commitments you've made to those employees, the expectations of your
investors, the expectations of your board. Andy Grove had this sort of way the he
described it, which is basically he said, as the CEO's in place, basically on day one a CEO has complete
latitude to do whatever they want. On day two that's no longer true, because you've started to
make commitments to people. You've started to tell the board,
okay here's my plan, I'm going to do x. You told investors,
here's my plan, I'm going to do x. You've told employees, here's what
we're going to do, here's the plan. So then as a CEO, you get 6 months in, 12
months in, 18 months in, 24 months in, and you change your view on strategy. And you go back to all the people and
you tell them we're going to do something different and they all say, but you just
got done telling us the opposite last week, or last month, or last quarter,
like what's wrong with you? And so, basically, what happens is
the people who take the idealized approach of shifting strategies when it seems
logical, typically end up getting fired. They literally end up getting either fired
explicitly, their board will fire them or their investors will fire them, because
they appear that they're lurching all over the place and
whip-lashing the organization. Or they get fired,
they just get abandoned, right? Like the employees and the investors all
get a vote, and they get to leave and they quit. I'll have this conversation sometimes
with founders where I'm like, that's all well and good, but like look around you, you're going to
be sitting here by yourself, right? How do you feel about that? With this big new decision that
you think you're going to make. And so, the real world
constraints are very intense, and they're very context specific to that
company, at that moment in time, and kind of how it got to where it got to. Now, Andy's response to that
was very crisp and solid, which was basically,
he literally said this. He's like, well okay, if the next
guy is going to come in here and he's going to do all these decisions,
why don't we just go outside and come back in through the door and
we'll just go ahead and we'll just breach all the old commitments
and we'll do the clearly right thing. And so, you're the leader, you're
responsible for it, the answer has to be, you have to be able to
work your way through it. You have to be able to break
the glass to be able to do that. It's just that's much harder and
riskier than it seems from the outside. >> Speaking of context in leadership,
you ran a very successful company with technical employees, and now you run a
very successful VC with a lot of partners who were successful entrepreneurs
in their own right. What are the differences in leadership
you have to display in both those environments that you've
learned to bounce, I guess? >> Yeah, so the big thing,
Siri is weighting in. Siri has opinions on management
here in the front row. So, the big thing as an operator, what I
always found as an operator, it's like a 90, 10 thing, where you're spending
90% of your time making decisions. Like 90% of the time you're sitting
there and customers, and issues, and you're getting employee product reviews,
and employee feedback. And you're making hiring decisions, firing
decisions, market expansion decisions, financial decisions, and so it's just an
unending series of decisions all day long. And then you get maybe 10% of
your time to sit and think. If you're lucky and
if you're disciplined you carve out 10%. You kind of think and read, and
get caught up on kind of your own. Your own theories on things. And you kind of have to do that, because these markets tend to
be brutally competitive, and if you just sit around thinking,
your company's not going to go anywhere. As an investor, and anybody who's
been an investor will tell you, if you take that approach to investing,
you'll blow yourself to bits, right? If your approach to
investing is 90% action, 10% thought, right, then you're like
every other schmo in the stock market. You're just churning through stocks and
you're going to, ultimately, as a stock market investor, you're going to do terribly because
you're going to churn your portfolio. As a VC, you're just going to
blow yourself to bits. You'll do the first ten deals and then
that's it and then they'll all fail and you're done, your career's over. And so a biased transaction as
an investor is a very dangerous thing, generally a very bad thing. And so, the great VCs and the great,
by the way, great VCs, the great public market managers, the great hedge fund
managers, anybody with long term horizon. They seem to spend 90% of their time
thinking, thinking and arguing. There's thinking and then there's the
actual process of all the people around you having to try to convince
them of your crazy idea so the arguing part ends up dominating. So 90% of your time arguing and
10% of your time, or maybe even less, making decisions. One of the things we ask ourselves is, how many decisions a year do
we make that actually matter? And it's probably 20? It's probably the new investments we make,
which is about 20 a year and that's probably it. And there's lots of things we advise on
and help with, but the decisions we make, it's probably two a month, at max. And so, it's a radically
different balance of time. By the way, if you've in an operating job,
it sounds very appealing because you don't have to sit there all day long and
be involved in all this activity and you get all this time to think and read
and talk to people and it sounds great. Except a lot of operators
are highly action-oriented and their entire success in their career
has come from taking action right? Crisply and aggressively. And so a lot of operators can't make
the jump to ever being an investor. because it's just too frustrating, right? Six months in, they're just
like crawling the walls, right? They'll go home at the end of the day and
they're literally, I didn't do anything today, right? This day was a total write off,
I made no decisions, right? And that's true of about 28 out
of 30 days kind of by definition. And so I'd say most, I'm sorry, most
investors can't make the jump operator, because they would turn
into a little puddle. >> [LAUGH]
>> Of plasma under their desk. Most operators can't become investors, because they can't deal with
the frustration of not taking the action. And so we're trying to straddle
it somewhere in the middle. This is the speech I give when
we only have former founders or CEOs as GPs at our firm. So we have all been through this
transition, and this is the little speech that I give every time, and they all
tell me, yep, you got it, I understand. And then six months later,
they're like, I see. >> So you mentioned crazy ideas. What are some of those crazy ideas that
you guys are banting around in the A16Z offices that you think will define
the future of the Silicon Valley and tech here? >> Yeah, the way we sort of go is we call
ourselves extremists thinking about ideas. Generally, smart people in the industry
are going to have figured out everything that's going to happen the next five or
ten years. It's one of the things like,
[LAUGH] Peter, my little argument with my shoulder Peter,
Peter talks a lot about secrets. And he talks about secrets being
something that you know or believe that other people either
don't know or don't believe. And I actually think like over a five or
ten year period, there actually aren't very many secrets. Most of the good ideas
are already kind of obvious. And by the way, they've probably been
tried before and they probably failed, but the time is going to come
where they're going to work. And so we in the industry,
the iPad is a spectacular new idea, except it wasn't a new idea. Apple had something called
the Newton 20 years earlier, but it was basically the same thing. It failed then, it worked in 2009. So tablet computing was not a new idea, it just happened to take
20 years to get it to work. But for that entire 20 year period,
it did look like a good new idea. So most of the good ideas
are kind of out there. They're circulating or, by the way, they're running in labs,
over here in the engineering department. They're already up and running as prototypes and
grad students are working on them. So that part of it is not, and we swim in
that world and trying to understand it and all that, but it's not that there are huge undiscovered
ideas with a short term timeframe. So I think we spend a lot more of our time
on the idea side, trying to think, okay, ten plus years out, right? Sort of take the extreme view, right,
of okay, it's sort of a [INAUDIBLE] of the venture philosophy, which is,
forget whether or not it will work. Ask the question about
what if it did work, and then kind of push that question
out as far as you can, right? So I mentioned self-driving cars. So let's just assume for a moment
that self-driving cars actually work. Okay, what are the consequences of that? Well, for example, one consequence is, potentially, cars changed our
idea of geography, right? Before cars, everybody used
to have to live in the city. Cars created the idea of a suburb, right? because you could actually commute. We all sit here 80 years later, wishing that nobody had thought of
that because commutes are horrible. The number one correlator to job
satisfaction is commute time, right? And so, you see this every day. You're sitting there in traffic,
sitting there next to, and so everybody has the same epiphany. It's like, it's eight in the morning,
I'm sitting there, and I'm like, okay, I'm sitting, there's car over
there with one person in it, there's a car over here
with one person in it. And I'm sitting here along. Like somebody should really
come up with a car pooling app. Every morning, right, every driver has
that idea and we all still commute, right, in our cars. And said we all hate it, and
so, and it's all wasted time. And we have people at our companies now
who commute two or three hours a day, and it's just completely wasted time. I mean,
now they can play Pokemon Go, right, while they drive, but
that has other issues. So then he said, okay, self-driving cars. Well, self-driving cars,
maybe you can reclaim all that time. Maybe you can reclaim all
the commute time, right? And so maybe all of a sudden you can have
the idea that maybe an hour long commute is actually a big perk, right? Because instead of driving,
instead of having to sit and focus and merge through traffic, what if your
car is a rolling living room, right? And what if you get to spend that
hour playing with your kid or reading the news or watching TV or
actual working, right? Because you don't have
to worry about driving. Or what if you had
different kinds of cars? What if some cars are rolling offices? What if I can take a nap, I could sleep. I can sleep for six hours at home,
get in the car, sleep for another hour and a half on my way to work,
and be all set, right? And so in that version of the world,
all of a sudden, maybe suburbs, like maybe now we go to exurbs. Maybe now geography becomes actually
much more protractible, and we can have these urban
environments get much, much larger. And then we're trying to think,
okay, then back from that, what would be the consequences of that? What would be the consequences of that
in terms of how these company get built? What would be the infrastructure that
has to get built to support that kind of thing? What are the kind of early signals that
kind of show that that kind of thing is either starting to happen or not. And then try to chart at least some
view of how the future will unfold. And so, yeah. >> So, you mentioned a lot of us have
great ideas we want to execute on. You've also noted that capital is
abundant and opportunities are scarce. And I think venture capitals are only
a $50 billion industry amongst a multi-trillion dollar economy. What do you make of the different
investment choices being made by larger companies vice what's happening here? >> Yeah, so this is the big thing,
I think, about tech. [LAUGH] l So, once upon a time,
we had a tech bubble. Some of you may have noticed or
read about it. It was followed by
a catastrophic crash in 2000. It was an actual bubble,
the numbers are very clear on that. There was about four years of nuclear
winter, where very little happened, and then starting in 2004,
tech started to work again. And between 2004 and today, it's just been
constant commentary of bubble, bubble, new bubble. Tech bubble 2.0,
all the way up and to the right, as tech has worked constant
screaming of bubble. Up to and including the chairman
of the Federal Reserve. Janet Yellen has developed
a theory that tech's in a bubble, which we really appreciate out here. >> [LAUGH]
>> I think she might have slightly larger issues to deal with. And by the way, with imminent
predictions of a crash, right? Which hasn't happened, right? And it's every six months, imminent
predictions of a crash, no crash, and it might actually be
starting to give up on this. Which ironically would probably
be the sign of an actual bubble. >> [LAUGH]
>> In any event, the result of all that is
exactly what you said. So all of new tech in the US,
all of new tech venture capital, growth, investing, all of what we would consider
to be what Silicon Valley does. Total capital in is about
$50 billion a year. And again, sounds like a big number,
is a big number on an absolute basis. It's a drop in the bucket from
a national or global standpoint. So the US economy, the US economy is,
I don't know, like $6 trillion or something like that and
then just look at financial assets. I think the number is now it's like
$14 trillion of negative yielding government debt in the world. So bonds that people have to pay for
the privilege of holding and then just even looking
in corporate America. This year alone corporate America will
discourage more than a trillion dollars of cash back to shareholders. Let's just think about it this way, like a
trillion dollars in cash was coming out of the Fortune 500 that literally can't
figure out anything to do with the cash other than give it back,
anything productive to do with it. So they give it back,
dividends and buyback. So a trillion dollars comes out and
then 50 billion, basically of that goes back into tech. Where does the other 950 billion go,
right? Into negative yielding bonds. And so which is what, to me, it's like
okay, why are interest rates so low? I think interest rates are so
low for a fundamental reason, which is the world is awash in capital. Basically, the 20th century worked. The 20th century created enormous amounts
of wealth all around the world and that process continues. And now that money is all out there,
sitting in these giant pools. Primarily, people's retirement savings. These huge pensions or these big sovereign
wealth funds that basically represent people's future retirement. And that money, all needs to earn a return
of 6 or 7 or 8% per year to pay for everybody's old age. And that money is seeking out
investment opportunities and it can put 50 billion into tech, and
then where else is it going to go? And so
I not don't think there's a tech bubble, I think there's a massive tech bust. I think we're running a critical shortage
of new technology in the world and I think you can view through a technological lens
in productivity growth, but I think you can also view that on a financial lens,
which is there literally are not. We as a society cannot come up with enough
places to productively deploy capital and this is the thing, like this just seems completely
bizarre like do we have any problems? Yes, we do. We've got massive education problems,
massive healthcare problems. There are big problems all over
the way still to this day. [LAUGH] Funny problems are funny. We use to have a massive
global hunger problem. Now of course,
we have a massive global obesity problem. Good news, people aren't starving
to death as much anymore. Bad news, now everybody's going to die
of diabetes and heart attacks, but like it's still a problem. So now,
we have a big problem on the other side. And so, we've got the money. Through most of recorded history,
you'd have these problems and you wouldn't be able to
find an answer to them. So we've got the money,
we can fund the answers. We've got the problems,
we know what we have to go do. It's the process in the middle
that's kind of all screwed up. And so, that's where I just think like
a tax on Silicon Valley like being bad at innovation or
whatever is kind of beside the point. Even if you're right, even if the valley,
even if the critics are all right and a tax on Silicon Valley is a joke and it's
all just a bunch of photo sharing apps and self-driving cars will never happen and
we're also it's like okay, so then what? If like if we suck, okay, fine. Like who else is going to do it? By the way, the Chinese are going for it. But again there, the dollar there's a big
tech investment happening in China. But again, the absolute dollar is just not
that relative to even the Chinese economy, much less our economy. And so, I think it's a tech depression
more than it's a tech bubble. >> Is that risk aversion on the parts of
investors or is it rational behavior? >> Well, it's a combination. It's a bunch of factors. I think like for sure, we've been on a risk reverse equity
market since the 2000 crash. If you look at the history of crashes and what happens after people
fight the last war. And so, what we know from
2000 is that stocks crash. So, everyone's constantly fearful. The headlines are always about
the market's about to crash, like all the way up. And then since 2008, on top of that,
we now have a massive aversion to debt. We're incredibly scared about debt. We're scared about real estate. It's really interesting. You talk to like our investors or
some of these big, like the Stanford endowment
here manages $25 billion. And if you talk to folks like the guys
who run the Stanford endowment not them specifically, but
that class of people. It's just like, they literally,
they go through. Here are all the things we can
invest in like public equities, terrible performance, 2008 crash horrible
bonds, we just had a financial crisis. Bonds are largely unsafe and
don't yield anything, anyway. Real estate, blew up in 2008. That can blow up at any moment. Commodities, it looked like oil
was going to go to the Moon. Now, looks like the world's
awash in an oil glut. The Saudis are dumping on
the market at any price. So, you can't invest in commodities. Hedge funds,
activist hedge funds were working. And now, they don't work any more and
they're blowing up. Venture capitals, we hear good things,
but everybody says, it's a joke. So maybe we can only put a small
amount of money in it, anyway. And so, they just literally
like a process of elimination. It's like, okay,
where is the money going to go? And so it's this weirdly perverse like
it's like I get really pessimistic about this, because I'm like the transmission
mechanism is not working properly. Like somehow, there's something
between the money, the problem and then the transmission mechanism to
connect the money to the problem that involves ideas, people,
companies, technology, management. Something in there is broken,
but I also get very optimistic. because it's like boy, if we could figure
out how to get that transmission mechanism working better,
we could go do much bigger things. And that,
by the way is a very big reason for optimism in the valley as you do have
a lot of founders in the valley now who are going after much bigger problems
that people used to go after. Like it did not use to be the case that
the valley would go after ideas like just with the huge effort, it's not going on to
buffer different areas of healthcare or transportation as an example or
even space. With SpaceX. It was just inconceivable that those
ideas would have been tried 20 years ago. And so, there are a group of people who
do have this view that we can now go do these things. >> I could sit here for hours. Unfortunately, I have time for one last
question before Q&A from the audience and this is deeply personal to me. I know you have a 20-month old son,
I have a 22-month old son. How should I raise him in the next five, ten years to be relevant
in the 21st century? >> [LAUGH]
>> Can I get back to you in [LAUGH] 20 or 25 years once we find out
how it went on our end? Look, I think curiosity. I mean, I just think curiosity. I think it's a cliche,
but I think it's true, which is just whatever the assumptions
are about how the world's working today. So many kids just get brought up,
I think Douglass Adams had this old line. He said, whatever existed before you were
born is just like ancient and irrelevant. Whatever happens when you are,
whatever happens, new technologies, new ideas happen when you're an adult. When you're a young adult are exciting and
fresh and whenever ideas that appear past the age of 50 are against
the holy order of things and it's this natural kind of human kind of view of
like it's weird, history is very long. We show up a bunch of stuff has happened. We draw a bunch of conclusions from that,
a bunch of other stuff happens. We have a relatively narrow
window to have an impact on that. And so I think that curiosity just is
really, really critical curiosity and an ability to learn. Frankly, everyone has
an opinion on education. My opinion is our education system is
still dominated by industrial age theory of we're creating cogs for a machine. I mean, luckily at places like Stanford,
there are exceptions to that, but a lot of K through 12 a lot of that is
fundamentally to get people to work on farms or in factories. And if you trace the ideas
behind education, people sitting in rows in a classroom and
somebody that's a senior authority figure sitting in front lecturing and
the whole thing. That's very kind of 1890
kind of era philosophy. Frederic, what's his name? It's mass production. >> Yeah. >> Applied to education. And so I think that the thing
with kids is to try to get them To the point where curiosity and
learning is good and exciting and fun. Right, and it sounds obvious, and yet,
for most people, that never happens. >> Yeah, well, let me turn it over
to the audience for some questions. >> Hi, Mark. This is Sam, 2015 alumni. I've seen the ecosystem for
the startup VC. Have changed, evolved significantly
over the past ten years, in terms more entrepreneurship,
educational resources, cheaper resources for start-ups,
and the more incubator startups. So, that probably make the competition,
amount of VC's more serious nowadays. So, I'm curious about what's your view
about this ecosystem over the next five, ten years? And what matters most to to in this world. Is it your vision, your PR marketing, or your disciplines on the research and
due diligence? >> Yeah, so the big thing I think,
from what you said, that I agree with you that the startup
ecosystem has changed a lot. And generally, in a very positive way. The big thing that's happened, as you
said, the big thing that's happened is the education is so
much more widely available. There are now so many more people, both
here in the valley and around the world, who now know how to start a company. And that, again, maybe sounds obvious,
but when I arrived in the valley in 1994, I went to the book store, and I tried to find a book on
how to start a tech company. And that was a long and
lonely trip through the bookshelves. >> [LAUGH]
>> There was very little. Luckily, Andy's book was there, but Andy's book was about how to run Intel,
which, it was a monopoly. I couldn't really get a lot of benefit
out of that for my startup at the time. And so, the resources just
simply didn't exist, right? And today you've got, it's everything. A lot of it comes out of Standford. Just the CS-183 course at Stanford
that various people have taught, Peter taught, Simon has taught. And then the fact that those classes are
not just accessible to people at Stanford, they're accessible to people
all around the world. And there are people all around
the world watching, right? Those videos all the time,
learning how to start companies. And so, the information's out, and you have these new ecosystem kind
of participants, like Y combinator, in particular, which has had
this big impact in the valley. Y Combinator is now producing 240
startups a year in the Bay Area out of their current program, and
they're planning to ramp that up. Ironically, for venture capital,
this has all been a tremendous blessing, because this is steel flow. And so, the more startups there are, the pond that we fish out
of is very well stocked. So, we are super, super thrilled
that all these things are happening, because we get to evaluate a far larger
number of startups per year than used to be the case. I think the competition is actually
much more intense between the startups. And you actually see that,
how many you'll see that, where there will be a company that will be super hot in
a YC class that we'll be doing an idea, and then they'll be the belle of the ball,
and they'll raise money, and it's all bad. And then six months later,
in the next class, there'll be another company
doing the exact same thing. [LAUGH] And now they're the hot company,
and the previous company's, what the hell? And then they compete. And then they both come to raise
venture capital a year later. And then we, as VCs,
get to pick between the two. And so that's, at least so far,
it's been really great for VC. If you're thinking about
starting a company, therefore, really stresses an idea, and again, I'll use my Peter here on
the shoulder, thing channel this through. Something Peter says I think is really
relevant, which is, a lot of founders think about what it takes to get
the second person into the company, the second founder or the second engineer,
or the third person or the fourth person. He says that's no longer the challenge. Everyone wants to start a company, and so,
everybody can get together two or three or four people to start a company. He said the challenge is how do
you get engineer number 20, right? Because engineer number 20, right? If they're in this ecosystem, right? They could go start their own company. And so, how can you convince
somebody to join your company's, essentially, you're number 20 with
a tiny percentage of equity, right? And not being in charge, as compared
to starting their own company, and getting the equity and being in charge. Now, one of the things to do is you
tell them they're going to have the same problem. But then they respond to you that
you also have that problem, and then you sit there and
stare at each other awkwardly. >> [LAUGH]
>> So, that's the thing. So, what you have right now is you just
have a very large number of seed funded companies that are experiments. And they're experiments on their ideas,
but more importantly, they're experiments on who has the will
and the drive and the horsepower, right? And the salesmanship and everything
else that goes into recruiting, right? And I think actually Sam Altman,
again, I talked to the other day. He said, if you could wave a magic wand, you would combine a lot of these companies
together to get critical mass, but mergers between startups almost never
happen for a variety of reasons. And so, it makes this sort of acute, this recruiting problem
is a very acute problem. By the way, the temptation is to think
about it as a recruiting problem, which is we have to get
really good at recruiting. But it's not really that as much as it's
like, okay, what about my company is going to be so spectacular, and so special,
and so unusual, and so distinct and differentiated that it's going to be
able to easily hire engineer number 20? Away from starting his or her own company. And so, to me,
it basically translates into basically, you just have to get better. You have to be able to compete, and you have to win in this
really brutal initial battle. The good news is the companies coming out
of that sort of churning kind of, sort of snakepit competition, the companies
that come out of that are really strong, because they had to be to come through it. But it is tougher, and it is the other side of how easy
it's become to start the companies. >> [SOUND] Thanks, Mark, for your time. I have a question about
the international investing. So, I see some on top VCs
like Saquioya Perkins, they went to China market many years ago. But in the contrast, I think, and just
pretty much focusing on the US market. I want to your point of view
of expending investing in those eMarkets like China India, South East Asia,
just like to hear about your view. >> Yeah, so, it's very,
it's extremely tempting. Because they're entrepreneurial
opportunities and great companies all over the world. Increasingly distributed,
right, all over the world. Literally, and so it's very tempting. But I'll tell you the typical problem, and then I'll tell you the theoretical
question that comes out the other side. So, the problem is, to the extent that venture capital
investing is like we understand it today, which is it's a very hands on process
of really deeply understanding. It's a people business. It's a process of very deeply
understanding the people you are working with. Deeply understanding the founders, deeply understanding the executives
in the company, the engineers, the management team, the culture. You both have to evaluate the company and
also to help work with the company. It has always been a hands on
business like that, right? When it's worked. So, if it continues to be
a hands on business like that, then there's a problem with
geographical remoteness, right? Which is,
if I'm not present in another geography, do I really know the people well
enough to make those decisions? So, then, what a bunch of firms
are trying to do is they then staff local teams, right? But then you have this fundamental problem
where you have this selection problem, where if that local team is really good,
they can easily leave and go run their own firm,
which is actually the exactly same problem I talked about with startups,
it also applies to venture capital firms. Say we had a really good
team on the ground in China, how can I convince them to stay
with us as opposed to just? Because they could very easily go down
the street open up their own shop, raise money. There's plenty of investors in China
who want to invest in venture capital. And so, if they're good,
they leave and start their own firm. If they're bad, they stay working for me. >> [LAUGH]
>> Which has its own issues. And so,
most of those experiments have not worked. It is very striking, Sequoia in
particular, has worked really well. And they have built a very good team, and they figured out a very
good dynamic on that. But it is a very tough question,
because you have to really dig into, okay, why did it work for them, and
why has it not worked for so many others, if you want to think about
doing that yourself. And so, that's a live topic, or
question, in the venture ecosystem. Now, there's the assumption,
go back to first principles, the assumption underneath that,is that
venture capital is a hands on business. And should be a hands on business,
the way I've described it, right? And again, if you look at history, if you
look at investment banks 100 years ago, they were very hands on like this. If you look at private equity firms in the
70s, when they were a boutique business, they were very hands on like this. At one point stock investing
was probably a lot like this. These days, in a lot of fields
it's not hands on anymore, now it's fully become computerized,
systematized and computerized. And so, the counterargument to my whole
theory on this is I'm describing venture capital the old-fashioned way. It shouldn't be so based on people, it
should be done through algorithms, right? You should be able to do
quantitative venture capital, right? And so you should be able to either like
open up a crowd funding marketplace, and let the world compete with ideas, and to be able to raise capital,
which is what AngieList is trying to do. Or you should, there's various firms now
trying to do quantitative venture capital. You should basically go gather all
the data, you can look at everybody's college transcripts, you can look
at everybody's past career records, their LinkedIn connections, or
whatever data you want to look at. And you can assemble a quantitative
view to predict success or failure, and invest that way. People are trying,
I would say we haven't yet gotten to the point where we think
we've figured out how to do that. Again, I fantasize that somebody
is going to come in and pinch us on building that as a company and
we can fund it, and then we can retire. I think, that question of is
it really going to continue to be that hands on is the question
underneath the question that you asked. I think it's an open question. >> Hi, my name's Sam Jackson. A question about the question
of capital wash. So, what advice would you have for people who have big ideas that
might be very capital intensive. And a couple of years
ago I asked Elon Musk, what was his advice if someone
started a new SpaceX or Tesla, and he suggested that I become
an internet billionaire first. >> [LAUGH]
>> So, I was wondering if you had any other,
maybe, easier ideas. >> [LAUGH]
>> That is how he did it. So, it is actually striking,
it is striking. The guys who are most prone to be like,
yeah, this software stuff is useless and stupid, and we should be going
to space travel instead, they all started doing the useless and
stupid software stuff, so that is true. I would like to tell you, again,
I'll word this as a question. I'll pose an open question. Which we're trying to think through, and I think a lot of people
are trying to think through. Here's how I think about this which is,
okay. There's a well known corporate investing
model which is GE or Honeywell. You go to one of those places or Lockheed,
and they've got their big programs, and you either work the bureaucracy
to do your program or not, and that's where most of these big
capital intensive things happen. And then there's the venture
capital model, and the venture capital model is sort
of staging the capital over time. It's milestone based investing, but
the milestones are kind of undefined, and then you're going to a new
investor in every round. And so you've kind of got this constant
financing risk, which says well, what if I can't raise
the next round of money? And then you've got the problem with these
capital intensive projects where it's like well, I know it's going to take a half
a billion dollars to get to market. If I go raise $20 million from VCs today, I have this massive funding risk of
$480 million in the future, right? And then from your standpoint, can I even recruit the team to
be able to build the product? Because I can't even tell them
that the money will be there. And then Andrisa stood on stage and told me about the $14 billion of
negative yielding government debt, and why can't I get somebody to give me
the $500 billion to do the thing. And so, the idea that's floating
around is sort of an idea to borrow an idea from actually a large
public infrastructure. And private infrastructure gets built,
which is project finance, which is basically like
the way a dam gets built. Which is basically the way a lot of
airplane projects get run, new airplanes, is basically,
like the Silicon Valley model, we're going to have a new idea,
we're also going to have milestones. Unlike the Silicon Valley model,
we're going to have a very, very crisp understanding of what
those milestones are, right? We're going to have a program management
office that's at a level of sophistication that a typically valley
start-up doesn't have. And we're going to chart out, in detail,
how that money is going to get spent, what the milestones are along the way, how the organization's
going to get built to do that. And it's going to be precise,
we're going to know week by week what all the work is, and all the parts,
and how they fit together. And we're going to try to get to
a much higher level of predictability to build the thing than a typical
Valley start-up would have. And then in theory, you could then
raise money, and this is sort of where there could be a new asset class,
and call it like, tech project finance. Where you could basically say, okay you
could go and you could raise $500 million, and the $500 million wouldn't
show up in your bank account. Instead enough money to get to
the first milestone would show up, and then there'd be a checkpoint, and then if
you hit the checkpoint you get the money, right, and so the money unlocks over time. And so, then you could think about
having a sort of tech project finance venture fund, where you could have
like 20 projects like that start. And then you could assume that like
10 of them fail along the way and get liquidated, right? And just get shut down before
most of the money is gone in. And maybe 10 of them work and then the fund is big enough to
fund 10 to completion, but not 20. And you kind of play it
like a venture portfolio. So, that idea is kind of
floating out there. And by the way,
this is what I'm describing. This is kind of how Google X works
a little bit, and this is how some of the advanced RNB projects, and
some of big value companies, operate. And so, that idea's out there, but it doesn't fully exist in
the form that I just described. And I think one of the really interesting
questions for the future of tech is, are we collectively going to figure
out a way to do that, right? because if we can't we can't
figure out a way to do that, then the project's you're describing
either have to be these flukes, right? You have a once in a generation Elon Musk,
Internet Billionaire, who decides to build a space company and
a car company at the same time. because who would do that? >> [LAUGH]
>> Or it's just going to be big companies doing all this stuff, right,
which would be a depressing answer. Mark, thank you so much for your time. We appreciate you being here. >> Good, good,
thanks everybody, thank you. >> [APPLAUSE]
