Emerging innovations in fintech | Michael Casey and Tricia Kemp

Emerging innovations in fintech | Michael Casey and Tricia Kemp


Now I want to take
it from machine learning to our next topic in
the financial technologies. We have two excellent
speakers here with us that will tell us
all about digital currency and fintech. Tricia Kemp is a General
Partner at Oakland Healthcare and Fintech. She has over 15 years of
experience in investing in financial technologies. And before that, she worked
11 years in big companies. Michael Casey is the Senior
Advisor of the Digital Currency Initiative here
at the Media Lab. He was working as a journalist
for over two decades in financial technologies, for
within others, the Financial Times. Also, he has literally
written the book about crypto-technologies. So I look forward to this
panel, and would like to ask the speakers on stage. [APPLAUSE] OK, so we are the gap between
the refreshments afterwards, the classic last line. So we’re going to make
this a nice brief chat. We’ve got some rolling
that we have to do. Thanks very much, David,
for that introduction. Yeah, so I’m Michael Casey. My world is defined
by blockchain. That’s a word I think you
might have heard once today. Obviously, a bunch of you
probably know what this is. But often it is a technology
that is still really quite nascent and so I
do find that we’re talking to crowds
sometimes that don’t really understand what it is. But that’s going to be a
theme we’ll deal with here. You know, I personally
think that it is a– it needs to be thought
about as the next wave of profound disruption in the
architecture, if you like, the infrastructure of
our digital economy, basically, because it just
changes the framework. And we’ll get to that
in a little while. But fintech is kind of
like the overarching theme in many respects, and it’s
quite a quite a buzzword at the moment. It seems to be that
everybody wants in on it. But Tricia, give
us a sense of what the investment environment is. Sure. So I’m a General Partner
with the firm Oak HC/FT, and we’ve done maybe
25 fintech investments over the last decade, let’s say. Fintech investing– and the
term fintech can be different definitions of what is fintech– but it has exploded over the
last three to five years. So 2015, 2016, both saw
numbers like $9 billion in each of those years, a little
bit off fourth-quarter ’16 in US and Europe. And that’s about 10 times what
fintech investment was in 2010, 2011, just so you understand. So it’s greatly exploded. About 40% of those
investments have been in both the lending portion
and in the payments portion. So they have sort of been
focused in certain areas, but the amount of money that has
gone in has greatly expanded. Couple other interesting
facts to think about when you
think about fintech, is the valuations
have remained frothy and are just starting to fall. So we kind of track
how many down rounds. We can pull it all
up in PitchBook. And general tech companies
2016, I think the number was 19% were down rounds. 81% were up rounds. Fintech, it’s
still well sub 10%, so they’re all still having
higher and higher valuations. And with something
like health care IT, the portion of our
business is 30% down round. So the valuations have kind
of started tracking down. So again, this is just
kind of all interesting. It’s still frothier
than tech in general, even though it attracts
a lot of tech investors. But our expectation is,
it’s going to start to fall. I mean, one other interesting
fact that we look at is– and again, is just
of interest if you’re thinking about a
fintech company, is strategic exit most
likely is the highest percentage of exits. So if you actually kind of
track it, 27% of companies sell for under $50 million. Of the other 73%,
the vast majority, a percentage like 89%, sell
between $200 and $600 million and sell at a strategic
sale to a big legacy player. So this is very different
than a direct to consumer Uber, or Gilt, or
whatever it might be. They are our strategic sales to
legacy infrastructure players in the $200 to
$600 million range. So that’s kind of how the
market has looked over let’s call it the last five years. I mean, it’s interesting
the froth is there, because I think that
we can call that hype, or we can call it a sense that
something big is happening, right? So I think that you
get a wave of interest in something when the
word is out that we are doing something big. And I think in terms of
fintech, I try not to hype, but I do sell. This is the first time,
really, in 500 years that we’ve really just
tried to tackle and truly disrupt money, right? So it’s a big deal. So inevitably, the enthusiasm
in the story and people like me who write
books about it, we sort of help
to breed the hype. But as a result,
where is the value? Where are you guys
looking to invest within this kind of rather,
like, big-vision picture? But there’s actually
going to have to be some real-life
opportunities coming to market the next two or three years. Where are they? Well, so we invest in
largely B2B companies. So we’re not doing the next
direct-to-the-consumer robo adviser, like a Betterment
or a Wealthfront, or trying to come up with
the next mobile bank. So we think it’s far
less risky, and actually easier because of what I said
earlier about valuations, to kind of look at
the world in terms of all the financial-services
ecosystems. So we kind of frame it, if
you think of all the stacks– you know, banks,
payment processors, insurance companies, mortgage
processors, asset management companies, capital
markets companies– they all have ecosystems
in and around them. And they all have stacks,
from customer acquisition, to onboarding, to pricing,
to risk management, to account management,
to profiling, to claims processing. They all have kind of
legacy infrastructures. Dramatic change is happening. And why is dramatic
change happening? We all know cloud, AI that we
just all heard about, data. And data being
used for AI is one of the most compelling cases
within the financial services industry. You know, mobile,
all those stacks have to have now mobile
and instant uses. So all those new
technologies are being brought to bear up
and down those stacks. So what we like to
call this, is kind of enabling the
ecosystems, right? So we think one large change
and one large trend that’s going to happen in ’17
through the next decade, is you’re going to
see everything from– think of your insurance carrier. We all see the Geico and
the Progressive commercials. But besides that,
you get 8 1/2″ by 11″ pieces of paper and envelopes
in the mail once a year. And you’re priced once a
year, and your engagement is once a year. I mean, there’s no
mobile engagement, there’s no online pricing. There’s no real-time pricing
and pulling in data on you. There is no take the
photo of the fender bender and send it in, and get your
claims processed immediately. So you can kind of
go stack-by-stack across all those. And there has been money put
into payments and into banking, but you’re going to have
solutions up and down those stacks being
brought to bear by cloud-based solutions, data. “Stack” is a word
that I’ve suddenly realized as I’ve left
the journalism world and came into the tech world. It’s a word I love,
because it allows me now to think about in a business,
and the architecture of where we build applications
in an interesting way. So I think really– again, a little biased, but
I think of the blockchain actually as the ultimate
underlying stack, right? Yeah. It is the bottom stack. Because what we’re doing here– you know, for those of
you who may or may not know much about
it, we’re grappling with a fundamental problem. So there’s this core
issue that is– again, 500 years we’ve had this
centralized trust model whereby banks, and for that
matter other institutions, have acted as the intermediaries
of our trust problems. We’ve relied upon
these institutions to represent us
when we haven’t been able to exchange in a
peer-to-peer way for whatever reason we don’t trust. So money is the most
important application of that. We trust each other to
transfer cash with a stranger, because we’ve got this
one-bearer instrument that we all understand is the
conveyance of the value that’s being conveyed there. But once we move into
a remote environment, how does the world know that
the dollar that I’m spending I’m not double spending
somewhere else? The monetary system
requires some understanding of how that exchange happens. And therefore, we have to
invest the trust, if you will, in a bank. And this goes back
for 500 years. So now, what
blockchain does, which is the underlying technology
that enabled Bitcoin, is to create a decentralized
architecture, a distributed system in which
that trust can now be deferred to a
consensus mechanism that’s driven by cryptography and
a bunch of other things– there is some game
theory involved. But it has proven itself. Bitcoin is actually the first
use case, if you like, of this. And as much as it’s had its
hiccups as a digital currency people didn’t think–
were like, well, why do I want something that’s– I’ve got my dollars,
and I think of Bitcoin as a digital currency
and that’s all it is. In many respects it’s
the first use case. And it proved itself, because
it hasn’t been hacked. It has not been destroyed. The ledger is complete, even in
this decentralized environment. So once you start to
recognize that you can see it as this important
foundational layer. Marc Andreessen when he was
sort of first confronted with Bitcoin said
something like, wow, this is the distributed trust
architecture that we always wanted when we were first
rolling out the internet– that this was the
stuff that would truly allow peer-to-peer exchange. And if we’d had it,
maybe we wouldn’t have these behemoths like
Amazon and others that end up recentralizing control. So there’s something quite
profoundly opportunistic about this moment, but
it’s very, very early days. And so building that
architecture is– I think some people
compare it to the internet. It’s something like maybe
the early ’90s, right? We still haven’t got to– maybe we’re
grappling with email, but we certainly don’t
have the World Wide Web, or we don’t have
mobile smartphones. And we certainly don’t have
Uber and everything else yet. Those equivalent
things don’t yet exist. But what’s also
interesting is, because I think of this recognition that
the disruption could be so big and the opportunities are so
large we are rushing forward with attempts to build
applications on top of it. And you heard the presentation
from [INAUDIBLE] earlier, which is a really
interesting idea, that you use this immutable
ledger now to be able to sort of provably
transfer title and exchange mortgages in ways that cut out
all of that reconciliation. But the question is,
what architecture are you going to build that on? And there’s some interesting
ways around that. There’s private
blockchains, which aren’t quite as disruptive as
these decentralized models. But all of that is coming along. And I think that it’s going to
be very interesting to see how this interplay between
what are ultimately like fintech applications and
the rollout of the underlying infrastructure is. I mean, the
opportunities are huge. At the Media Lab, we’re
looking at things like– there’s a project we’re
doing that Mark’s actually involved in– Mark Weber– the
warehouse receipts. And so this idea
that you could really bring into our financial system
individuals and entities who have been locked out of it,
because that centralized trust architecture is
difficult to get into. There’s huge
requirements to do so. And the case of warehouse
receipts of those farmers that we’re looking
at in Mexico, who will get a receipt
for the grains that they submit to
a warehouse, which they should be able to use to
then get a loan to be financed. But because there’s no way to
prove that this hasn’t been double spent– that this same
receipt hasn’t been repledged three or four times– they get locked
out of financing. Banks don’t want
to finance them. So what if we place that
into this immutable record that nobody can change,
this decentralized trust architecture? You start to see
the possibilities. So these are the things that
we’re exploring in this realm. And I think it’s very
exciting, but as I say, we’re kind of at the very early
TCP/IP stage of development. But nonetheless,
in fintech itself some of them being built
on centralized systems, at this stage at least. There’s a lot going on. What are some of the
other big trends? I would just add to
what you’re saying. So blockchain has tremendous
promise, as we all know, in [? trends ?] use cases. There are all these
other things– use cases– that may or
may not use blockchain that within financial
services need to be fixed. You know, I gave the example
of insurance earlier. I’ll give an example of
identity authentication. You know, when you show
up to open a bank account. We all know you
have to show up, you have to bring like
a utility bill. You have to bring two forms
of where my address is. All this sort of stuff that
is just 30-year-old technology that needs to be changed. Invoicing, CFOs–
CFOs receive payments in seven different ways,
including dog and pony. And they send them out
eight different ways and it’s all
electronic AR and AP. It’s all very old,
antiquated is what you’d say. So blockchain has
tremendous promise, but there are also use cases
and solutions and services that are needed kind of
up and down the stacks. We can talk about the
digitization of payments, which everybody is familiar with. And I don’t know if you know
the numbers, but consumer payments– and this is going
to be approximately. But let’s say 2005, 41%– and you guys all look
too young, so you’re not remembering all this– but
41% of consumer payments were digital. You know, 60%, 59%
were cash and check. By 2015, that had
completely flipped. And it’s close to 77%
digital, 30% cash and check. B2B payments are still
well over 50% via check. So US companies send
around– they’re 70% of the check volume
in the United States. They are sending around
checks in something like 55% of the cases. So you’re going to have
the continued digitisation of consumer payments. And many people think
2017 is when we’re all going to start putting
our leather wallet away and start using our
phone as our full wallet because we all want
one-touch purchase. We’ll see if that
happens, but many people think it’s going to happen. But you’re going to
have this continuation of the digitisation, not only
up and down these stacks, but across payments. Then, across B2B payments,
which is a much larger volume than consumer payments. So digital digitization
of continued payments, which brings up what I kind
of say a third area of lots of promise in 1718
which is what I’d call regtech, regulatory technology. So it’s fraud, it’s risk, it’s
identity, it’s compliance, it’s know your customer. I don’t know if anybody saw,
there was a Wall Street Journal article recently. HSBC has 8,000 people in a
tower in Long Island City, New York and they’re
literally doing manually– I’m going to call it manually–
doing background checks. You go open up an account,
they are googling your name and making sure you’re not the
axe murderer, or the terrorist, or whatever it is. I mean, that whole
world of compliance is now somewhere between 15% to
18% of banks’ day-to-day costs. And as everything
becomes more digitized, and as we all have more and
more devices, those needs, and what your identity is, and
the negative history, which is your background check,
and your positive history is that Patricia Kemp
paid this utility bill for this many years
at such and such address, is all going to start to be
organized in a fashion such that it becomes easier. I’m glad you raised
the identity. That’s one of the
areas that we are very interested in as
a potential blockchain application. And one of the reasons
why is because you start to think about–
well, it’s a couple things. One is that compliance problem
could be better resolved by the transparency of the data
that resides in the blockchain. So in a real-time
way, I can start to look at where my flows are. I might anonymize that data,
because we don’t necessarily want that data to be
publicly identifying people. But at the same time,
if I can sort of– we talked about machine learning
and big data earlier on– if there is a sort of a big data
analysis of what a blockchain flow system looks
like, you can start to manage nodes rather
than individuals. And that’s an interesting
way to think about this, because what obviously– India is facing this
massive problem right now with this demonetization
they’re calling it. They’re sort of
moving away from cash. And it’s a massive
backlash, partly because people depend upon
cash as an anonymous form of exchange. And there is some actually
kind of human right almost to this
capacity to exchange. And so Bitcoin is
very interesting, because it is digital cash. It’s not a new way to
do digital banking, it’s literally digital cash. The anonymity that
obviously gets it in trouble on the one hand,
is a very important feature as well, because it allows
for fluidity of commerce. So once you start to think
about that architecture, you then think of, how might
we apply compliance to that? What other things could
do with this information? And there’s all sorts of
interesting applications around KYC, know your
customer concepts, or digital identity concepts
that don’t necessarily identify people. So knowing your
customer is not knowing their name and their
social security number and what the utility
bill is, but it’s knowing that this
particular entity, whoever it is that you’re
dealing with, has this provable record
of transactions that makes them a good
credit or a bad credit, or a high risk or a low risk. And you have this sort of
more automated interaction between financial institutions
and the customers. And then you start to see, OK,
so we could have potentially a safe world where
these flows are managed in a way that doesn’t
undermine the financial system. And we can identify risk
from a kind of a [? patent ?] perspective, but also
protects anonymity and privacy in these things, which
are becoming increasingly important. And hopefully that
is the sort of thing that really reduces that
massive compliance risk. Because by the way,
that compliance problem is creating a global problem,
which is this derisking issue. There are countries
all over the world that have had remittance
corridors shut down, because banks are
terrified about being sued for somehow unwittingly
funding a money launderer. Or worse, like a
terrorist operation. Well, I mean, I don’t
know if you saw it, but Western Union was fined,
it must be a month ago now, like $600 million– not an insignificant
number– because they are AML, and in money
laundering techniques were not appropriate. And the answer is to put up all
the more compliance offices, or shut down operations
with whoever they see risky, which either way is going
to cause the same thing. Because that just
adds to cost, which means the barriers to entry
to the financial system are higher. So we have a duty
for the sake, as far as I’m concerned, to
the global economy to resolve this problem. And I do think this is
where a lot of the growing opportunities lie. Yeah. No, and it’s a little bit
related to the last panel, where is data, and then good
use of data, AI, applicable? Financial services
is a great use case. I mean, just as we all now walk
through the security things at the airport, 99.9% of us
who are being double checked before they let us open
our Merrill Lynch account, and it’s only the 0.1%
that are in trouble. Lots AI and lots
of data can be used to make all of this
a lot smoother. It’s a terrific
match of use case. But a lot of it,
also, in my mind, is leading us eventually to
a complete reorganization of what we consider to be
a financial institution or a bank. I mean, what do you– how long before we
start to think– I love asking these questions. Will banks be the same as they
are now in 10 years, 20 years time? What’s the kind of glide path
for this from that big picture perspective? When do banks die, I
suppose is the question. And I’m kno wishing
that upon the banks, although I quietly am. But no. So I don’t know
the answer, and I guess our investment
time horizon is kind of 5 to 10 years. And what we’re trying to do
is get them efficient enough so that they can compete. I mean, it is very interesting. And Michael mentioned this,
the Indian demonetization. I mean, it’s as if somebody
in the United States said, OK, every $5, $10, and $20 bill
is worth zero December 31st. And all merchants
stopped accepting it, because they’re trying
to dramatically move that country to being
completely digital payments. And they announced it– oh,
they announced it the night of the US election,
November 9th. And they said, whatever
value of rupees as of 12/31 are no longer of
value, they’re zero. And so you had this
enormous, massive chaos. So what’s my point? My point is, that does something
like that in the United States– does this
digitisation continue, right? Does a concept of a bank how
many people walk into a bank if you no longer have to walk
into the bank with your utility bill to show who you
are, then what happens? You are going to
see dramatic change. You’re obviously going
to see dramatic change. I would counter, Michael,
that it’s people’s money. So I have always said,
it sounds great to have Wealthfront and
Betterment and Robin Hood and all these sort of things. That all sounds great
when you have $5,000. If you have whatever
number of dollars, you’re handing it to Vanguard,
or you’re handing it to Chase, or you’re putting it someplace. You know, this is
people’s money. So I don’t think it’s quite
the same as get rid of taxis and get into Ubers. I think there is a difference
to this, because there is a sense of security. Heavily regulated
environment– I mean, the company has to be heavily
regulated and licensed. So I don’t think it quite
shifts like other industries, because– Risk of losing your fortune
or losing your life, to that previous
panel, are big checks on how far this
technology can go. But at the same time,
there’s a mountain that’s slowly starting to move. And I think it’s going to be
interesting to see how it goes. We’re going to call it quits,
because we actually both have to try to make it to
transport to get us out of here and further south. So, sorry there’s no
questions, but hope that was interesting
and useful to you all. And thanks very much. Thank you. –for joining us. [APPLAUSE]

Author:

Leave a Reply

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