Professor Susan Athey: “The Future of Currency”

Professor Susan Athey: “The Future of Currency”


[MUSIC] Yes, today I’m going to talk
about the future of currency and I’m going to focus mostly on
cryptocurrency and blockchain innovations. One of the things that has become
interesting about talking about this subject over time is that,
when I first started speaking about it a general audience
would know almost nothing. Today almost everyone has an opinion about
it, which in some ways, makes it harder to start a conversation since people are
coming from very different perspectives. But today I’m going to sort of
start at the beginning, but hopefully even for those of you
in the audience who are experts, some of the perspectives
will get you thinking. And then we’ll have a discussion
starting around 10:15, where we can take it in all
sorts of different directions. So, I got interested in cryptic currency
and blockchain back in 2012 and 2013, I had some computer science
colleagues who were working on it. And then I also went to various
conferences, tech conferences where a lot of people in Silicon Valley
were starting to get interested in it. And you might hear someone say at
one of these sort of tech elite conferences something like, I was employee
number x double digits at Google and I have now put 20% of my wealth into
Bitcoin, this is a great thing. And my first reaction to that was,
you guys are bad people, you’re doing a ponzi scheme,
how could you be doing this? This is not good, but
when I started digging into it and really understanding it I got very excited
about the technology and it’s potential. Since then I joined in as an advisor for
Ripple and I’ll talk a little bit as I go through about what made me
initially excited about Ripple. And then joined the board a year later
before their series A and now Ripple is one of the more successful companies
in this space, which is really great. And that was co-founded by
Chris Larson who’s a GSB alum as well. So, I’ve had a really exciting
time with this space, and now I’m teaching a class to the MBAs. We were way over-enrolled and had about
100 people in the spring on this topic, the students here at Stanford are also
very excited about this space. There’s sort of an informal organization
of students that has about 700 people on their mailing list, so
it’s a really, really exciting area. At the same time there’s a pretty big gap
between sort of how many student projects come in from MBAs to
engineers whether they’re all excited about what to do with blockchain, versus the reality of what’s actually
going into commercial productions. So I’ll say, it’s sometimes hard for
me to even to communicate my schizophrenia on this topic which on
the one hand is like fascination and research and teaching and
leadership in the area. And on the other hand, for any particular person who’s thinking
about investing in a company or in a currency, there’s still a lot of
skeptics that needs to be kept into place. So hopefully we’ll get the balanced view
of all of these things as we go through. So, first of all I want to talk about
it’s actually even using the words, it’s kind of tricky in this area
whether you talk about crypto currency. Whether you talk about block chain, actually none of the words are really
perfect for describing what’s going on. So it’s more about the thing that you’re
doing rather than the terminology. But there’s really a bunch of different ways to think about
what cryptocurrency is. And one reason it can actually be
difficult to have a conversation around it is that everyone has a different
idea in their head and people start talking across purposes. So, one thing to think about with
cryptocurrency is it’s a way to store value. It’s sort of like digital gold and there’s
a group of people who are very strong on that use case and
that’s a use case that’s happening. Especially, if you’re a billionaire and you’ve sort of diversified in
all of these different assets. And you have your house in Newzealand and
you have your banker, why not put 20% of your asset into
something lioke Bitcoin and then the keys are actually storedin bankers around the
world and helps you kind protect yourself. But it’s actually sort of a cumbersome
way for a typical person to try to hedge, for security reasons and also there’s
a lot of technological impediments. As an asset class, I’ll probably won’t have time to
show you too much on the statistics. But it is kind of an interesting asset
class because it’s not that correlated with other traditional asset classes. And so it’s even in the short time and
you’re not worried about the world blowing up in some way, it does provide
sort of a differentiated type of asset. The thing that sort of got me most
interested in the beginning is the idea that it can be
used as a payment system. Now it’s going to turn out that it’s
pretty hard to make a new payment system, and if you’re in an economy that has
a pretty well functioning payment system, it’s really not going to
provide a lot of benefit. I’ve had students coming to my office for
15 years saying how they’re going to compete with the credit cards, that’s
just actually a very hard thing to do. The economics of credit cards
are pretty hard to enter, it’s not about the technology. So you could say the technology is
archaic, but it’s actually the economics that make it a difficult place to enter,
not the technological improvements. But, one of the places where things are
really broken are international payments. And so that’s a place where
we’re seeing adoption as well as real economic value from cryptocurrencies. And then the final thing, which is really
where actually most of the sort of engineering teams are really working,
is that it’s really a platform, a technology platform. So especially something like Ethereum
is built as sort of a software development system, and I’ll talk a little
bit more about how people are using that. But it’s really a technology
that’s a building block for some other kind of service, a database, some way to store information,
and that’s what a lot’s going on. If anybody’s interested more
in these application areas, because we have relatively
short time today. GSB lecturer Doug Galen has written
recently a very nice paper that surveyed all the Blockchain For Good initiatives. And it talks in-depth about
where the startups are, which ones are close to actually,
having a working product and so on. And interestingly the biggest
category he finds is healthcare and not finance, and it’s basic, there’s a lot of businesses coming up
with databases for health information. Where people can actually control other
people’s access to their own health information. And that information is stored in
a database that they can selectively grant access to. There’s a lot of variance on that idea but
that’s one of the more popular ideas, and there’s a lot of these companies because
every country in the world has their own system. And health data is sort of something
that’s kind of both valuable and somewhat broken in most economies. So, already hopefully just as this intro we can think about there’s not just one
thing here, there’s lots of things, so I’ll try to build up
from the very beginning. I’m going to focus on the currency aspect,
both that’s sort of the title of the talk, but also I think it’s the one that’s
the easiest to start wrapping your heads around. And once you’ve got those basics you
can then think more about these broader applications. One other comment though about
this whole application area Is it, I’ve seen near hundreds of
projects come across my desk. I feel like a lot of those
conversations where, it’s my advisor, where the students come back later and
tell me where they ended up. There’s a meme on Twitter one of
these little flowcharts that says do You Need A Blockchain?. And so you’re expecting a whole
bunch of different things, and then there’s just one arrow down and
it says no. [LAUGH] And at some level that’s
a pretty good approximation for what a typical project comes up at. Nonetheless, I think it’s still
incredibly interesting to think about where we are going to go. And it just says that when somebody
gives you a shiny new hammer, you don’t necessarily,
immediately find the nails and there is a lot of hammers looking for
nails. But overtime, as our understanding in
these technologies emerge, then when this new cases emerge, they do fit, it
will be more like here is my problem, and there is a technology
that might actually fit. But it’s much harder to start a business
with a good business idea, and then for the technology rather than saying,
there is a shiny new object, let me start a business where
that technology might be useful. And I think that’s sort
of the stage we’re at. And that’s what really came out of my
cryptocurrency class in the spring as well that most even the best VC’s
didn’t really have a lot of real, solid examples going on. Okay so now let’s really spend some
time getting into what this thing is building up from the beginning. So here’s a picture of a ledger, ledgers
have been used for a very long time. And sometimes your Econ 101 course
you’ll hear about the early bagger economies where like somebody
trades an orange for an apple. But actually that was never
a super common way to do business, because it is kind of hard to have an
orange and apple at exactly the same time. And so through millennia,
people have kept track of who has what and had some sorts of ledger systems. And they’ve been everything from stones
to marks on wood before we had paper. To keep track of, okay, you gave somebody,
you gave something and then later on, you should get something back. And so ledgers of course are really
a fundamental way to understand money as money is a way to make
it simpler than having ledger. Because if I just give you a dollar, I
don’t have to go look up on a ledger, but it sort of serves the same purpose. You give me something now,
I give you a dollar. Later on you can use that dollar for
something else. So ledgers are old, ledgers are powerful,
it really is useful to have ledgers. And the problem is that when we have
gone to the digital economy, the cash, which sort of made it simpler
than keeping track of a ledger, doesn’t work in the digital world. I can’t beam you a dollar. And our financial systems haven’t
really caught up to be fully digital so that we can move money sort of as
easily as we move information. So I’m going to focus on Bitcoin
just to the sort of, kind of, the biggest and first big innovator here. So what Bitcoin is at it’s
core is just a big ledger. It’s a big spreadsheet, okay? So wow, how could people be so
excited about a big spreadsheet? But that is what it is, a big spreadsheet. So imagine you’ll be convenient if
you’re trying to interact with people digitally to keep tract of who has what. Suppose you’re across countries and
it’s sort of hard to really move money. You can just have a big Google Docs,
big Google spreadsheet, and you could have people only authorized to make entries in
the spreadsheet if they had a password. And we can put a little bit of logic on
top of that which might say that your entry gets rejected, if you don’t have
another entry in the spreadsheet. So we can have a big
spreadsheet that says and forget about the initial conditions just
for a moment, we’ll come back to that. But suppose the spreadsheet starts and
says, I have a Bitcoin, you have a Bitcoin,
someone else has a Bitcoin. Then essentially what you would want is, I
can login if I have the right credentials and say I’m going to make an entry that
says I give a Bitcoin to someone else, and we’ll reject that entry
if I don’t have a Bitcoin. But if I have a Bitcoin and I logged in, I can make an entry that says I
gave it to someone else, okay? This would be a convenient thing to do to,
if we, didn’t have,
we’re not all in the same banking system. It would be nice just to have a big
spreadsheet to keep track of who has what. And that’s what Bitcoin is. That’s all it is, is a big spreadsheet. If it says that you have something,
then you have a password. You can log in and make another entry
that says you give it to someone else. That’s really the only
functionality of this technology. Okay so
why would you be excited about that? Well if you imagine that I was keeping
$50 billion worth of stuff on a big Google spreadsheet,
we might have some security concerns. Okay, and if I said I’m going to spin
up a server in my house on campus and I’m going to keep the copy of
the spreadsheet on my server. Well, I would be like a magnet for
attacks from all over the world. And of course, I want people to be able to
access this, so I would have to be sort of open to the Internet in some way,
which would create a security nightmare. So the idea of Bitcoin
is to get this product, the product is a very simple product,
which is just a spreadsheet where people can,
from all over the world, make entries. But I’m going to find a way
to make it secure, okay? And I’ll come back to the security,
that’s, of course, one of the really interesting parts. But for now, let’s just imagine we have
this product, I’m going to tell you more about what you would do with the product,
and then we’ll come back to how it works. Just to get a sense of
exactly what happens, again there’s a big spreadsheet and
it’s going to have addresses on it. These are pseudonymous addresses and in
fact people can have multiple addresses. So if I have an address that’s mine,
the entry that would come to me at first would be somebody else sending me a
Bitcoin and making an entry on the ledger. Then if the ledger says I have one and
I have my password, what I would do is I would
sit at my computer, log in, say, I want to make a new
entry authenticate. The new entry will be a Bitcoin goes
from my address to someone else’s. And so when I say log in
what I mean is that I have the credentials associated with that
address that allows me to make an entry. And as long as the spreadsheet says
yes this address received those coins in the past, it will accept the new entry
to send them to someone else, okay. So for this sounds kind of fun. Sort of like trading beany babies or
something. I have a beany bay,
I send you a beany baby, something you send a beany
baby to someone else. Of course this only is useful
if there’s some value in having these entries on the ledger. But I want to point out that that value
does not derive from the ledger itself. So we’re going to be able to have
markets where we buy and sell entries on the spreadsheet, but those markets
are just regular old financial markets. They’re not part of the basic protocol, the basic protocol is
just the spreadsheet. So if you want to see like how you access
it this is the screen shot of one of the early wallets and the early days
of Bitcoin most people would sort of of manage their own Bitcoins. And so the idea was that on your own
computer, you would get have a little piece of software that would actually
download the entire history of the ledger, and then you would keep a copy of it and
actually you would join the network by participating,
by keeping copies on your computer. Then the software would look
through the whole ledger and just show you the things
associated with you. And so
it would simplify the whole history, and say, these are the transactions
associated with addresses in your wallet. And so you would put in a password to
access this, and then you could do things like send and receive, and keep track
of contacts who would be addresses, and then make little transactions,
either receiving or sending. So when this actually goes out,
what is a block chain? It’s actually a sequence of little spreadsheets that are new
additions to the ledger. And so the block,
this is basically all of it, this is what happens when
you make a new entry. And so, there’s going to be a block, and
then it’s going to have a transaction, a fee, a size, the From and the To. So that’s basically all
the data in the protocol. So now let’s start thinking
about what you do with this and why it would even be useful. So I think the first thing I
thought about as an economist, and those of you from the finance sector will
presumably had the same initial reaction is, why would you
introduce a new currency? We have the Euro zone to avoid
having lots of currencies. Any of you that work in
international businesses know that hedging exchange rate risk is a problem. It’s not something that you
would choose to take on. And generally, for consumers, you wouldn’t want them to take
exchange rate risk if they didn’t need to. So if somebody’s going to pay their
rent and their taxes in dollars, you would generally not
want to hold another currency, unless you were using it as
some sort of hedging strategy. But it wouldn’t be something
that you would choose to do. You wouldn’t just take volatility for
no reason. So the first I said, okay,
this sounds really cool, I really like this technology for
keeping track of who has what. But why do it in this thing called
the BitCoin, why not do it in dollars? So why not have a bit spreadsheet
that says who has which dollars. And the key insight for that is that actually we can’t put
dollars on a spreadsheet directly. So just imagine we have this
spreadsheet technology. We could certainly put a message on
it that says, you sent me $1, and I could make another method that says,
I sent you $1. But what would that really mean,
because I didn’t beam you a dollar. There’s a message on a spreadsheet
that says I sent you $1, but I can’t actually send you
$1 on a spreadsheet, okay? I could have messages that
say I promise to give you $1, but I can’t actually
send $1 on a spreadsheet. And that’s really core to understanding
what could possibly be new and different from a sort of currency or
asset perspective, is that the definition of a BitCoin
is the entry on the spreadsheet. So if I authenticate sending a BitCoin
to you, you have the BitCoin. It’s not an IOU for a BitCoin,
it’s not a promise to obey a contract and give you a BitCoin,
it’s actually the BitCoin. The entire definition of the asset
is the entry on the spreadsheet. Well, if I put $1 on a spreadsheet,
because dollars are not digital, I can’t actually send you $1
literally on the spreadsheet. And of course those you from finance
will be familiar that in some sense, like when you’re moving money
internationally, you don’t think about dollars and euros, and actually,
those dollars and euros live in a bank. So there’s like Citibank dollars or
JP Morgan dollars, because if you were
moving around the world, the way that you go through dollars
is to hold accounts in a bank. And if you go from country to country, you
might go from a smaller bank to Citibank. Then Citibank would make an internal
ledger transfer themselves to a foreign bank,
where they have an account. And then you would make another
transfer from that foreign account to a smaller bank in that country. And so what we really have when
we look digitally, are IOUs. When you open up Bank of America in
the morning and look at your balance, you say, I have digital money. Well, you don’t actually
have digital money, you have an IOU from Bank of America. Bank of America promises that if you
send a transfer, they’ll make a back-end transfer, but they actually cannot promise
you to do whatever you want with it. You said, I want to send it to
North Korea, they would block it. If you sent a bunch of transactions for
$9,999 over and over again, they would block it, okay? And we’ve seen various politicians get
caught up with that type of problem. So there’s an IOU that is
subject to the rules and regulations of a particular country,
and so I can hit send, but
I don’t know if the money will go. And in fact, internationally, about 5%
of wires fail, the money gets lost, so you hit send but
it doesn’t actually receive. So any companies that have a big supply
chain will have a whole little department of failed wire transfer chaser-downers
to deal with that problem. So an asset like this, of course it’s
an interesting asset and it works for most purposes, but it is
fundamentally different from an asset where when I make the entry on
the spreadsheet, it’s done, okay? It’s good and bad. If you’re a consumer and you type
in the wrong address in Bitcoin and you hit send, it is gone, okay? There is no 1-800-Bitcoin to reverse it.>>[LAUGH]
>>You made the entry, the entry is permanent. It is there, it is there forever. If I sent a transaction to Silk Road when
I was a college student and everybody knew what the Silk Road address is, that
is also permanently on the record, okay? So if somebody could associate
that Bitcoin address with me, then forever you would know that I was
sending money to Silk Road when I was 21. So these things are permanent, irrevocable
and actually on the Bitcoin ledger public. And in fact in some of my
research a few years ago, I went in and analyzed the block chain. And you could see what people were doing
and how many people were gambling and how many people were going to Silk Road. You could figure out what time zone
they were in by when they made their transactions. And for people who had ever posted
on the web their Bitcoin address, you could figure out them and you could
figure out their friends and so on. So it’s actually in some ways
a pretty unsafe way to do things, because you don’t usually
value your financial privacy. And you have no recourse if
you type in the wrong number. Also, if you lose your password,
this is a problem. So in the early days,
there was a Big Bang Theory episode about this that kind of formalized
a bunch of my friends’ experience. I had lots of computer science friends and early people toying around
with Bitcoin in 2011, 12. The things were worth a cent or
two cents or ten cents. And then they get up to $1, $10, then in last December they are $18,000. And you’re like, wait, what happened to
that laptop, what happened to my password? Imagine trying to go back and find
something that was on a laptop in 2012. And there were a lot of sad people. Because you can look at the blockchain,
and you can say, those Bitcoins are mine, but you can never get them back. So there are lots of people who
are like multimillionaires on paper, but just can’t get the coins. So now those are like all
of the negatives, but of course there’s also some positives. So if you think about it in a business
context, wouldn’t it be nice sometimes to make a transaction and ten minutes later
have it be done, and there’s no rollback? Because then you can make another
transaction on top of that, and another one on top of that. And you could have computer
programs do those transactions. And you would have this
certainty that they took place. So that is just a different characteristic
of these assets that cannot by kind of regulation cannot be provided by
the existing financial system for consumer protection anti money laundering. All that stuff. So we’ll not allow a bank to
really offer that service. So it’s just different, okay. So now let’s talk a little bit about
what you would do with this and also that’ll bring up the idea of how
these things are actually valued. So suppose my kid was
doing a Skype lesson. Actually one of my kids now is taking
Python over Skype which is a very convenient thing to do,
and you can get a teacher. In Silicon Valley, it’s hard to
hire someone to do anything for less than $100 an hour. But you can get someone who lives
somewhere else who’s a nice, starving computer science student
who can teach Python to my child for a much more reasonable price. So you’re doing some sort of Skype lesson. But suppose the teacher is
actually somewhere else. Or maybe you’re learning a language,
Japanese or Mandarin. And you’re tutor is in another country. You would like to pay them right away. If they’re in another country, if I wanted to send them directly
a peer to peer wire transfer, that might cost me 50 or $60, which is
large, relative the the cost of a lesson. Now, platforms and so
on will try to make that easier for you. But even actually like Air BnB and
Uber and so on. Are actually very upset about the state of
the current financial system, because it’s so expensive for them to facilitate
these payments around the world. So there are actually some of the big
movers and pushing for improvements. So if I wanted to do
this just peer to peer. If the recipient actually was
understood the way that BitCoin works, I could do the following thing. I would open up an account on a company,
say, like Coinbase or Bitstamp, and these are companies, they are not part of
Bitcoin, they’re just startup companies. And what the service they provide is
a place where you can send currency, and then do trading on their system. And so they just have simple software that
creates an order book in every currency. So I transfer in my dollars
using an ACH transfer or a wire transfer to get my dollars there. Once my dollars are there,
then I can just place a limit order or a market order to say hey
I want to buy a BitCoin. And some people want to sell BitCoins and
they have nice little visualizations, in fact the software for this is so easy that I talked to a startup who
setup an exchange over a weekend. There’s basically just
off the shelf software. This is not rocket science
to do the really basic part, of course security’s another problem but
the basic software is easy. So then if I say I
want to buy some Bitcoin. What will happen is that exchange will
then find the place where the bids and ask cross, make the exchange and
suddenly I’ll be the owner of a Bitcoin. Now what I can do with that
Bitcoin then is, at that point, the exchange is still holding the Bitcoin
itself, so they’re still in custody. If the exchange went out of business
right that minute, I’d be out of luck. And that used to happen quite a bit. Now there are some more well established
companies with governance and real engineers and things like that. So one of these more established ones I’m
not real worried about them going out of business in the next five minutes. So then I have the Bitcoin and I can then send a transfer to
the receiver’s Bitcoin wallet. And that can take place
in about ten minutes. So I say, please send. Then their wallet would receive that, and
then we could just get off the Skype call. They’ve been paid, we’re done, okay? So it’s a very nice feature that you
really cannot get across borders anywhere else. Now what do they do once they have it? Well, they could take the Bitcoin and buy
a cup of coffee or buy something online. But if they wanted to actually
change it back into yen, then they could transfer it,
send the Bitcoin onto an exchange, and it might be a Japanese exchange,
which would receive the Bitcoin. Then they could place an order to sell, the Japanese exchange would
give give them some yen. And then they could use the Japanese
internal country transfer systems to put it into their bank account,
okay? So this is how you could do this,
and this stuff happens all the time. It’s a very common thing to do, especially
for these small value transactions. People use it to pay programmers in
Venezuela or all sorts of other use cases. Now, there’s a few problems with this. First of all, this is kind of high
friction, this is not like simple consumer product, you’re exposed to exchange
rate volatility along the way. Now that exchange rate volatility might
not much be within a day although like over Christmas and I was caught up in
this as well because I was holding and I was just trying to sell. The price might move 5% or
10% in the course of a day. So you had to be like be there at your
computer trying to get this step done and not be like lazy about it. But generally on a given day
it doesn’t move that much and you also have to worry about these
exchanges going out of business. One thing that doesn’t matter is
the level of the exchange rate. So this thing works as well when Bitcoin
is a $100 a Bitcoin as it is 18,000 a Bitcoin. So I’m just getting in and getting out. All those exchange rates
will move in concert, I don’t really care how high it is. What I do need is transaction volume. Because in the early days when it was not
that popular the spreads would be high, and you would lose a lot
on the bid-ask spread. Today there might be Jjst even ripple
which I’m going to talk about later. We were having lots of $4 billion days so
$4 billion of transaction volume over the holidays and then more recently its
been about a half a billion dollars a day. So that’s more than a lot of stocks and
so that means these are actually fairly liquid relative to stocks and
that’s important for making this work. Nonetheless, this is really not a great
consumer product and the way this would end up getting mass marketed would be
that a company would do this for you. And the consumer really wouldn’t see it. So the consumer would have an app and just
say I want to send money to Japan, and the app would implement
this on the back end. And there are a number of products all
over the world that try to do that. Now there’s a few additional concepts
that become interesting here. Is once you understand all of that,
you realize that actually these exchanges are kind of a big source
of friction and a source of risk. So what you might want to do instead
is have a decentralized exchange where this is all done in software. So that all of the transactions
go one after the other. And so
V1 of this was actually Ripple in 2013. So this is what their
software looked like then. They built a sort of new version of
Bitcoin that was similar in some ways and different in others. But one of the ways that it was different
was is it actually built into the protocol the ability to have market
makers make bids and asks. As well as a path finding algorithm that
would say if I want to send something to Japan, it’s going to find
a sequence of trades I can make, to get me from the dollars
to the yen automatically. So it’s the same thing as this
previous picture except for it all happens at once through software. So there might be a market maker
making a market between XRP, their currency in dollars, there might be another market maker
making an offer between yen and an XRP. They would plug into an API and those
market makers would commit that if someone took their bids and offers,
then the transaction would go through. They wouldn’t have to manually approve,
there would just be a standing offer. And so you could just take it and
execute that in three seconds. So this is what the view one of ripple
then was a differentiating thing and I thought at the time,
this was a really a big improvement. Because I was generally thinking consumers
would not want to hold a cryptocurrency themselves, so if I wanted to think of
utility this would be a better way. So and now there’s a bunch of v2’s
of this coming out now called decentralized exchanges
bit on other protocols and platforms are trying to
implement this idea. The interesting thing that happened
with Ripple is that they went out and tried to sell this to banks and
said hey banks, you might want to use a system like
this to move money across borders. And some banks were interested, but
actually a lot of banks said hey, we’re a little worried about this cryptocurrency
thing, it’s got regulatory risks and it’s going to take some time to figure
it out But we love this idea, so could you do this for us but
without the cryptocurrency? And then you actually realize that, wow,
yeah, why didn’t we do this before? We should’ve done this without
the cryptocurrency before? And you might ask,
why don’t we do it today? And the answer is,
it’s not about the technology. It’s about, we all use Swift, and we’re
all stuck, and we have software systems that are used to an old way of doing
things, and we just haven’t advanced. So my economist friends say,
well, couldn’t we just fix that? And I’m like, yeah, they are. [LAUGH] That’s what we’re doing. So Ripple introduced in something
called an interledger protocol, which basically takes this idea
of a decentralized exchange. And it’s a protocol that basically, will
say, if I want to get money from one into the other,
there might be a bunch of middle men, market makers and people who
have accounts at multiple banks, they’re all going to make
offers onto this protocol. And then if I want to get money from A to
B, we’re going to have a protocol that ensures the money instantly goes from A to
B, but it never gets stuck in the middle. And that’s a really nice thing because,
in today’s system, you tend to have delays and
have things get stuck in the middle. So the bottom line of how this is getting
used is that banks are basically plugging in, there’s a big group in the Middle
East, there’s also a big group in Asia, a big group of Japanese
banks have just joined. So a lot of smaller countries
are the early adopters. And they’re setting up their own networks
so they don’t have to go through Swift. Because Swift, some are archaic and it goes slowly,
their messaging doesn’t work very well. And then the smaller countries also
feel like they are disadvantaged because it’s sort of everything is going
through a few large US institutions. So in terms of the future of currency,
in some sense part of the role of the US dollar has been to be a hub of
currency in a hub and spoke system. But with these new technologies, the smaller countries don’t
have to go through that hub. Anybody who flies an airline through
a hub knows that you pay a higher price. And it can be slow, and the airlines kind of exact monopoly
power over you when they control a hub. And the smaller banks and countries feel the same way about their
relationship with the large US banks, where they have to pay markups and
face delays as they go over that system. So they’ll be able to basically create
a peer-to-peer network that allows them to move money instantly among themselves
without having to go through that. Now a set of those countries
are also using the cryptocurrency. Right now it’s mostly
remittance providers. So there’s a company called Qualix
that is doing it US to Mexico. And there’s an association of 1,400
US credit unions that’s doing it to help the credit unions access
international markets. And so it’s kind of coming up from bottom
from the smaller player who aren’t well served. All right, so let me now say a couple
of quick words about mining just so that you won’t walk away not
understanding how this is. So this all started with this
anonymous white paper, Bitcoin, a peer-to-peer electronic cash system. And basically what they wanted to
accomplish was a worldwide network of what we call weekly identified parties to reach
consensus on the contents of the ledger. So you might ask, why would you want this? And I think actually a lot of us would
still ask that if you were using a commercial implementation. But let’s just take for a second what they wanted to accomplish
was sort of intellectually mind boggling. We said before we got these copies of the
ledger, we want to decentralize them, but I’ve got to provide security somehow. So how are you going to do that? And their idea was to allow people to
join the network when you had no idea who they were. No way to know, no way to verify. Anybody can join. College kids, people in any part of
the world can join this network and can help keep track of what the truth is,
okay? If you’re going to ask that
anybody in the world can join and keep track of what the truth is, you’re going to have to figure out
some way to make it costly to join. And that is part of what this
this this whole system does. So the Nakamoto consensus said that we
can let anybody in the world join and keep track of the ledger. But if they disagree, we’re going to use majority voting
to say which one is the right one. So if I can join the network, and say,
I’m keeping a copy of the ledger, my temptation would be put to put an entry
that says everybody just gave all their bit coins to me. That would be fun. And if I can keep a copy of the ledger,
why wouldn’t I do that? Well, the answer is that everybody has
to vote, majority rule, over whether that’s the right thing, and nobody else
would want to approve that transaction. So because I wouldn’t have authorized it, I wouldn’t have followed
the rules to do that. So it’s voting, and
you’re going to have a majority rule. Now, the immediate thing you think of was,
wait, if it’s voting, why don’t I just set up 100,000
computers and have them all vote for me? Okay, and that is in fact possible. Now, the way that you avoid
that are a couple of things. First of all,
if I set up a 100,000 computers and made them all give the money to me, then
the prices and the markets might crash. So I might not really get
anything out of that. And that is one of the key
things that’s keeping it secure, it’s actually an economic incentive. If I tried to break the system by
creating a lot of computers and giving all the Bitcoin to me,
I wouldn’t be able to sell them because nobody would want to buy them
because I just broke the system. So it’s really an economic
incentive model. It’s not a cryptographic breakthrough. It is economic incentives. That’s very interesting about this. But the second way that they do this
is they make it expensive to join. So you can’t just join and
create the ledger. You have to actually pay a fee, okay? But the problem is we’re
all over the world. We just said we don’t have an
international banking system that really works particularly well. And these people,
we don’t even know who they are, so how could I get money from them? So I want them to pay to join the network,
but I don’t have a great way
to collect cash from them. So the solution is you have to
burn money to join the network. That’s the solution. Now, if I literally had you burn
cash,you might worry about, how do I know that you
really burned the cash? I could doctor a video of me burning
cash and say hey, I’m burning cash, let me join the network, but
that might be hard to verify. So they had to figure out a way
to burn cash to join the network in a way you could tell, unambiguously,
that you burned the cash, okay? So what’s the solution to that? They set up this very hard math problem
that’s using cryptography, basically. Where, to solve the math problem, the only
way you can solve it is to keep guessing. So every ten minutes, this system
is going to pose a question, and then all around the world,
people guess the answer to it. And so you can then use electricity and
computing power to keep generating guesses and guesses and guesses and guesses, and
the only way you’d ever get a right answer is if on average,
is if you burned a lot of energy. So the barrier to entry, the analogue of
burning money is to burn electricity. And the way it’s credible is that I
only reward you if you actually solve the problem, and on average, you can only
do that if you burn a lot of electricity. So that is the system. Bunch of people burning
money all over the world for the right to vote,
which is the right system. But I just want to I re-emphasize, that is an economic breakthrough,
not actually a cryptography breakup. And then what happened over time is that
actually, and chips were invented whose only thing they’re good at is to solve
these problems, just for Bitcoin. So the whole industry developing
chips just for Bitcoin. Then there are the server farms
that are now very streamlined and efficient that buy these chips,
make the server farms, and then solve these math problems,
and try to get the reward. So the people responded to the economic
incentives to create mining farms. Very quickly, and it’s a bit of an arms
race because you actually can’t be profitable at it unless you’re using
cheap energy and the latest chips. Otherwise, you burn more energy
than the rewards that you get. So now, we have a distributed ledger
with a bunch of people burning money all keeping copies of it, and we’ve
achieved the security we were looking for. Now, it’s turned out
because of the economics, there’s been a lot of
concentration in that. It’s economies of scale. So it’s better to be a big network. And so,
we do actually have more concentration. And people still worry about the control
of a small number of groups. And in addition, this money burning
has really gotten quiet expensive. So it didn’t seem like a big deal when
you were sort of a bunch of people playing around with a science project. But right now, BitCoin consensus costs
about $10,000 per minute, and accounts for about 0.1% of world energy consumption. So it’s actually very
expensive to burn that energy. And then, people are out there
thinking about alternatives. Are there ways to get all these benefits
without having to burn so much money? It ranges from trying to do something
productive with your electricity, instead of solving this
useless math problem. To other things like proof of stake, where you get to vote if you have
some commitment to the ecosystem. And then, a third solution is to
actually not just let anybody join. So and that sort of thing, I’m burning
0.1% of US energy consumption just because I wanted to have the feature that anybody
in the world could join anonymously. Maybe if I restricted entry a little bit,
I wouldn’t have to do that. So finally, I’ll just mention that one
of the cool things you can do with BitCoin is to create software
that moves this money. So in a typical escrow system for
buying a house, you would have a central
party who keeps everything, checks if certain conditions are met,
and distributes the funds. Well, Smart Escrow is a simple
example of a smart contract. It’s a piece of software
that can hold on to funds, check with the state of the world,
was there a hurricane, did the stock price go up,
did something else change? And as a result of that, in a stake
contingent way, distribute the assets. So I could hedge a risk around the world, I could place a bet with anybody without
having to trust them, know them, or pay a bank to act as a middleman,
just by running up a piece of software. And that software is available at scale,
they’re platforms for creating that software. So people can create these
types of contracts, and allowing new things to happen. And that’s one of the big
areas of innovation. So I think I will stop here and
open up for questions. And yeah, we can take it any direction. There’s lots of places we haven’t gone yet about the value of the currency
of everything else. So let’s open up to questions now. And we got mics coming around. So up here.>>Can you talk about the future
of energy consumption and whether that’s a risk for
our climate change?>>Yes.>>Can you talk a little about the future
of energy consumption in all this data mining? Yes, so I mean,
I think it’s a big problem. And at some level, if you said, is a country going to try to
use BitCoin for their national currency? Are we really going to see
something like at scale? I think that this energy consumption,
money burning thing, is just a fatal defect from having it
really get big as a transaction mechanism. And so, when you talk to countries, some countries are thinking about
getting rid of cash entirely. England has advanced in that. Some of the Scandinavian
countries are thinking about that. And they have thought about
can we use this technology to have a fiat currency but just not
have cash, have it be truly digital? But none of them are seriously thinking
about I’m going to have this kind of proof of work energy burning
thing to run a national system. They would instead sort of limit entry. Like manually,
you would have a distributed system, but you kind of have to have
permission to join. And then, you don’t have to burn
the money in this same kind a way. Yeah, is there [INAUDIBLE]?>>So my understanding from the regulatory
standpoint is the central banks will tolerate this as long as it’s marginal. But the idea that you can move
very large sums of money, which ultimately is fiat money, across borders is not something they’re
going to tolerate when it gets to scale.>>So that’s a great question. So there’s, I think, it’s really
important to separate two use cases. So one is the consumer use case,
like over Skype. And that, I think, is very threatening
to high inflation countries that have a lot of capital controls. And basically, I would go short any
country that is dependent on capital controls by the way, because you can’t
put this genie back in the bottle. This technology is out there and there
are privacy preserving versions of it. So essentially, you just cannot keep your capital in
your country at scale in the future. But the concern about that is one
reason they would clamp down. Of course, other things are sort
of any money laundering and so on. Now, the business, this is one reason
I’m more interested in the business to business case,
which is sort of what Ripple is doing. Because in that case, nothing changes
in terms of the regulatory framework. The consumer never touches it. The consumer is still going through
the same financial institutions, and the financial institutions
are just doing it on the back end. And for that use case, actually, the central banks of smaller countries
are incredibly enthusiastic. They’re actually pushing
their countries forward. Because a matter of national policy, having your entire country dependent on
paying markups to CitiBank is a bad thing. And you also subject to regulatory risk. The US can change regulations
which can make it harder for certain countries to move money
where they want to move it. So from that perspective,
getting this sort of peer to peer kind of regional networks, the Central Banks
are actually incredibly enthusiastic. What they want, I think, is they’re
sort of happy to go to a new technology, they just want to make sure that
they still have the normal gateways. And so that, I think,
is a much more promising avenue. Nonetheless, they’re still thinking
more about the technological solution. And only some of them have really
embraced cryptocurrency as a conduit, as a hub currency. Yeah.>>So in your opinion,
which cryptocurrency has actually tried to solve
the flaws of BitCoin? Or which is the most advanced? That’s part one. Part two is that is there a better
model available than burning energy that can cure of that, of the
fundamental flaw in all these mechanisms?>>So this is a very deep question, and you could read thousands of pages on
Reddit of people arguing about this. So I will say a few kind of comments, but
I cannot claim to have all the answers. So when I first started talking about
alternative cryptocurrencies in, say, 2013, and I was involved with Ripple very
early, there was a whole set of people, including some of our famous tech leaders
in Silicon Valley, who would stand up and say it’s scale economies, the best thing
doesn’t always win, Bitcoin was first. Bitcoin, Bitcoin, Bitcoin. Everything is going to be Bitcoin,
these other things are all going to die. I would stand up and quietly raise my hand
and say, well, if you’re thinking about Enterprise software, it’s not all just the
Betamax and VHS and Netscape versus IE. In fact, in Enterprise software, you typically have
a lot of different vertical solutions serving different parts of the market
with different combinations of features. And so, my thought was that we will see
different systems for different purposes. And in fact,
that is sort of how that’s evolved. So I don’t think there’s
one size fits all. So Ripple wants to move hundreds
of billions of dollars, or trillions of dollars,
between banks internationally. So they need a highly
scalable system that, and this proof-of-work stuff is
just a complete non-starter. So they have in some sense controlling
who keeps track of the ledgers. And they also have software that allows
you to just move things without even touching the cryptocurrency. So you wouldn’t even really call
it a cryptocurrency solution. For that use case,
that is what’s taking off. And the other things are not taking off so
much. For developers, you might, you might
want a cryptocurrency that actually has More ability to write software
embedded in the protocol. And so ethereum has gone a lot in that
direction in terms of making it easy for people to build software
on top of the ethereum. So etherium is almost thinking of itself
like an operating system, really. And so that’s good for that. There are coins that
preserve financial privacy. So there are coins that actually do
everything that Bitcoin does, and there’s still a big ledger but nobody
from the outside can look at it, and reconstruct the network
of who sent what to whom. If you’re trying to evade capital
controls or do money laundering, that’s your currency. That looks great. And of course, consumers might have
an interest in financial privacy even if they’re doing legitimate things, because
if you can see on the network that you hold a lot of cryptocurrency you
are actually vulnerable to kidnapping and all sorts of other things. And you’re more vulnerable with crypto
because the money can move across borders so fast. So you’re at risk of theft
in a way that you’re not for the money that’s stored in the bank. So I would just say that
they are different, and I don’t think it’s
going to be one size fits all. I don’t think there’s going to be 1000 but
they may be 15 or 10, that actually each saw
the different use case better. And the biggest problem right now is
we actually have very little product market fit. There’s hardly any actual use cases where
this thing is really doing a great job. So, trying to make it one size fits all
just makes that problem harder rather than easier. But this is a very controversial thing and
to top it off, there’s all these Twitter bots that are trying to make
people buy and sell currencies. Just like in the election, we have these
Twitter bots that create a division and tried to make everybody hate each other. Well, the crypto space has all these bots
that try to make everybody hate each other and troll each other. So it’s really just kind of a wild west, in terms of the conversation
on that topic.>>We keep hearing about tokens
all the time these days. Where do tokens fit in to all of this?>>The Bitcoin is the token for
that ecosystem and XRP is the token for the ledger. I think one of the ways people are just
kind of say you have a token economy and since I’m actually really
a market place expert and this crypto thing is
a relatively new thing. But some, I’m on the boards of
leaning clubs and Expedia and Rover which are marketplaces. So I got a lot of people coming
into my office pitching me these like marketplace economies
where the ideas that you give your medical data then somebody
is going to pay you in a token. You hold the token, now you want to
get other people to join the systems. So your token will get more valuable and
then that will grow or the tokens might also replace like frequent fliers Miles or
other type of ideas. So there’s all these ideas about sort of
businesses that are built on top of of blocked chains with tokens that are
specific to the business and then there becomes a little economy in those tokens
and everybody wants to make it grow and they’re all invested in
the success of the business. In some level that’s how Bitcoin started. So you can’t say that the idea doesn’t
make any sense, and it can make sense, but the hard part with all of those
things is just getting them started. They all face a chicken and egg problem,
and unless you have a good business, those tokens are just nothing. But it’s a very hot topic for
entrepreneurs.>>Yeah, I’m still really confused about
the whole voting business when it comes to all the different ledgers,
the different versions. I mean what actually happens if my
computer actually keeps a ledger and a new transaction comes in,
how does my computer either accept or reject that transaction as
a legitimate transaction?>>This is kind of confusing because
they’re sort of in the equilibrium and out of equilibrium. So, when the system is working then
there’s just software that you would run and it would say,
here’s a new batch of transactions. It would check a set of rules. So, was this person authorized
to make the transaction? Did they have the Bitcoin and so on? And so in that state of the world, the
software just checks a set of things and then accepts or rejects it. So, there’s no problem. What makes it more confusing is if
you think about some sort of out of equilibrium behavior, which could take a whole bunch of
different forums but they all would have the same idea of putting things on
the ledger that didn’t follow the rules. And so that’s the case that
where other people would say, hey my copy doesn’t look like your copy. And if they’re following
the rules they would just say, well I don’t know what happened but you
must have made a mistake we’re going to do the same thing as everybody else. I checked and I agree, and he checked and
he agrees, all good, we just keep going. So these attacks are the things
that are sort of, could take a variety of different forms.>>Could you speak a little bit
more about blockchain for good? Specifically vchain for
vaccine authentication or other healthcare applications?>>Sure, again, a whole variety of them,
a lot of them are nascent. I think one idea that I think
does have legs is trying to follow supply chains
in complex settings. So if you’re the Gates Foundation and
you’re trying to move money around Africa, you give the money, and there’s this
hole leakage along the way, and a lot of stuff is in digital, and
there’s a lot of corruption, so it’s hard to make sure it
got where it was going. So one thing you can imagine happening,
and it’s certainly being discussed is that, some big institution
whether it’s like the World Bank, or the Gates Foundation, or someone else,
that’s really big could sort of force their supply chain to track funds
through a distributed ledger. And that might be useful, because the people you’re dealing
with don’t actually have identities. You actually have to use suppliers
who you really can’t verify. And the countries themselves don’t have
good legal institutions and so on. So that’s a good use of a blockchain. If it’s just funds, you could just see
them going from one person to the other, and they would be dispersed,
split up as you go along. So big block grant here and then it’s
dispersed and dispersed and dispersed but you would have a. A ledger that keeps track
of that until the very end. Another way that you can do
that is with physical goods. And there you get into this challenge of
how do you keep track of this physical good, but the really simple version
of the idea is, there’s some kind of bar code that’s really embedded in
the good that you can’t take out easily. And you scan it at the beginning and then
as it moves along it’s scanned again, and every time it moves from party to party
that triggers funds being released as well as a record of what happened. And, again, we already have that,
IT systems to do that today. It works great inside a firm, but
it doesn’t work so well if you’re going across countries in Africa or
someplace that doesn’t have good systems. So, those are some of the application that
might get donors to give more money or make things more efficient. And then other types of things, the vaccines,
maybe there’s counterfeit vaccines. So you might want to scan
the vaccine at the factory and see that this is an authentic vaccine. And then at the end you
could tell that it was real. And people have also talked about it for
just counterfeit luxury goods in China. Now they’ve been versions of those
businesses that don’t involve blockchains, and when you really dig into it,
you might say I don’t really need, I don’t need the full Bitcoin,
I just need the database, I just need the database to access. So the questions still arise of
what’s the best way to do that? What’s the very best technological
system to manage a supply chain?>>Do you have any estimates
of how much illicit versus illegal transactions are going on
in the the cryptocurrency space.>>Yeah, so various people have analyzed
this at various points in time so my own analysis is a bit dated now
because it was mostly in 2015 and 2016. So by far the biggest thing that people
are doing with these things is day trading and holding them. So I was actually excited to look
at all the stuff people were doing, but most people are not
just doing anything. They bought them, they hold, and then
the price went up and the sold them, or they’re just trading them a lot. So most of the volume is just on and
off exchanges, and they’re not actually doing all these great
things that we imagine they would do. Now of course all that liquidity
can make it cheaper for someone to build a real business. And so like the US to Mexico, the fact
that there’s all these day traders makes it really cheap for a remittance provider
to get in and out of XRP going to Mexico, the fact that there’s
a lot of trading volume. But that’s the main thing they’re doing. And then among the things where you could
identify what they were doing, yes, Silk Road was big, gambling was big but
in some sense, that’s just sort of like, in the early days, there’s just
not really much to do with it. You want to just have fun. So you’re just kind of doing things for
fun. You’re buying coffee, because it’s cool,
not because you really need to do. So it’s hard to kind of draw inferences
from that very nascent time period. So right now really like people
aren’t using it at scale. Period. But these illicit used case are one
use case that’s where there’s not a better alternative today. And that’s partly why that’s
like an actual use case. Thanks so much everybody for coming.>>[APPLAUSE] [MUSIC]

Leave a Reply

Your email address will not be published. Required fields are marked *