O’Hara, Arbitrage on Wall Street


SPEAKER: This is a production
of Cornell University Library. MAUREEN O’HARA: Thank you all
for coming out this afternoon. And thank you for inviting me
to give a Chat in the Stacks. I think it’s a neat idea. And when I was asked, I
was delighted to come. I want to talk
about modern finance and what has been described to
me as an ethically challenged field. And it’s hard to
disagree in some ways. Banks and financial
service firms placed dead last yet again
in this year’s global ranking of trust in industries. A recent Harris poll found
68% of people surveyed disagreed that, in general,
people on Wall Street are as honest and
moral as other people. Bloomberg, in a poll that
just came out in July, found only 32% viewed Wall
Street banks favorably. We were delighted to see that
Congress came in below us at 29%. But insurance companies
were dead last at 26%. So those of us who are
in academic finance– and I also work in
the industry a bit– you tell yourself,
oh, that’s not true. But after a while, when you
started seeing poll after poll, you begin to wonder if there’s
not something out there. Here’s another one
I’ll just share. This is the Chicago
Trust Index, percentage of trust in financial
institutions. The most recent
survey fell back down. We’re at 27%, which,
needless to say, I think even Congress
might beat us on this one. So it’s disconcerting. And it’s one of the reasons
that I wrote the book. And there are really
two issues that I think are reflected in this. The first one is the one that
is a bit trickier to deal with. But I think most
people don’t actually understand what modern
finance is and how it works. And I think modern
finance, in many ways, really dates from the 1990s on. And as you think about what
we do in finance today, it’s not what people
in the past did. So the notion of a
loan and a mortgage and different sorts
of specific entities is not actually the way
modern finance works. So one of the goals
of the book was really to try and write at
a level that we’ll call the “educated layman”
could read and really understand what we do. When you read that Goldman
Sachs get sued for fraud and they used something
called a synthetic CDO, what in the world did they do? Most people really
can’t understand that. But actually, it’s
not that hard. And one of the things I
tried to do in the book was give people an ability to
be able to think about that. The second problem may be
the more important one, which is that I think
there’s too many finance practitioners who do not
understand or even recognize that there is an ethical
component in finance. And for that, I think
we in academic finance bear some of the
explanation because we teach finance like engineering. It’s a tool. So you want me to build
a contract that does x? No problem. I will build that contract. You want me to
build something that can get around this pesky little
law that says you can’t do x? No problem. I can build something
that’ll go right around it, and no problem. And if you think about
that for a minute, it’s like, wait a minute. Maybe this has led
to a culture that simply leads us to ask
whether I can do something, but not whether I
should do something. So I think we’ve begun, as the
finance academic community, to rethink this. I think the banks have begun to
rethink this because they’re up to $323 billion in fines
worldwide and $160-some billion in the US alone. So it’s gotten to be
an expensive hobby to avoid thinking about
the ethical issues. So you might ask, why? Why is this happening? And I just thought I’d
share a couple of– here’s a theory that
a lot of people like. This is the finance and
the bad people theory. This is by Judge Richard Posner. He argued that basically, firms
that have short-term capital– they have to
compete ferociously. They know that it’s
a jungle out there. And so as a result, the business
model of those sorts of firms, which he defines to be
the financial firms, attracts people who
have a taste for risk and attach a high
utility of money. And then he goes
on, as you can read, the complexity of
modern finance, the greed and gullibility of
individual financial consumers, difficulty so many ordinary
people have in understanding what’s going on here facilitate
financial sharp practices, enabling financial fraudsters
to skirt criminal sanctions. So when you first read this,
it’s like, well, I don’t know. And then the second
time you read it, as a finance
professor, he seems to equate people
who work in finance as financial fraudsters. And I think that
is a little harsh. So let me give you
an alternative view that I developed in this book. And it’s that arbitrage
plays a starring role. And I’m going to
explain arbitrage a bit more in a minute. But modern finance is not
really, as I mentioned, about traditional contracts. It’s about cash flows. So whether those
cash flows are coming via interest payments
on bonds, or they’re cash flows coming from
dividends on stock, or they’re cash flows coming
from mortgage payments, or they’re cash
flows coming any– I don’t care where the
cash flows are coming from. Those are the building
blocks of modern finance. And we don’t tend to look at
things as, let’s take these– think of this as taking these
cash flows and these cash flows. And we put them together. And maybe we added an option. And we added a
credit default swap. And we do this, we
do that, and voila. We create something. So you tell me the
contract you want, and I’ll create it using cash flows. The problem is, in the
process of doing so, the ethical dimensions get lost. And that’s part of what I want
to try and do in this book– is first make it
a little clearer about how modern finance
works and then talk about those ethical dimensions. And again, I particularly
like this little cartoon. I don’t know if any
of you know xkcd. Here we have two people
at a Mexican restaurant. They’re the ones
giving the chips away. If they don’t see the arbitrage
potential, sucks for them. In a deep sense,
society only functions because we generally avoid
taking these people out to dinner. I really like this, right? Stealing chips from
Mexican restaurants is really– there’s
no law against it. But it’s just not done. And I think there are
lots of things out there that fit into that category,
and particularly in finance. And so as we think about this
role that arbitrage plays, I want to make it a little
bit clearer that arbitrage is more than just buy
low and sell high. So a classic
arbitrage is you have the same good trading in two
places at different prices. So if you remember back– some of you it will be easier,
some of your less easy– some undergraduate
class that you took. And you talked
about gold in London having one price and gold in
New York having another price. So you sell where
it’s expensive, and you buy where it’s cheap. And that’s an arbitrage. And when you have the same
good trading in two markets at different prices, then when
you sell it at the high price and buy it at the low
price, what have you done? You basically have gotten
something for nothing. You were able to arbitrage. And in the process, you bring
the market back in line. That’s certainly still true. That’s classic arbitrage. But it’s more than
that now because now what we’re going to do is– in modern finance– using the
tools of modern finance, which are things like swaps, which
are things like credit default swaps, which aren’t
actually swaps at all– they’re actually options. But using derivatives
and other sorts of ways that we can change cash
flows, what we can do is we can look at something
like a natural security. And we can build a
synthetic version. So I’ll give you an example. Suppose you wanted to hold–
you’re running a big insurance company, like TIAA. And you want to buy AAA bond as
an investment, nice and safe. great investment for
an insurance company. But what’s the problem? We only have two AAA
companies in the entire United States now. Back when [? Jean ?] was in
school and others, we had 180. But we have two, right? Those are Microsoft
and Johnson & Johnson. So if every insurance company
wants to buy AAA bonds, the yields will be negative. But here’s the good news. We can make more. We can synthetically
create a AAA bond by using a credit default
swap and a little bit of financial magic. So what you can do in this
new world of finance is great. It can do really neat things. It can create products
for a company like TIAA that allows us to better
hedge the risks over time so that when you retire, you’re
going to have this big lump sum of money or big– we hope–
annuity that you’ll take. So the tools of finance
today are fantastic. So what we can do is
build synthetic versions of the contract that you want. So where does arbitrage come in? I’m going to create
this synthetic IBM bond. And I have a real IBM bond. And arbitrage keeps the two
prices in the same place, just like the gold in London
and the gold in New York. So arbitrage is going
to be extraordinarily important in this world. We’re going to build
synthetic things. And I want to argue the
synthetic world is everywhere. But the notion of arbitrage
permeates finance. All of option pricing
theory is built on the concept of arbitrage. And when we teach our
students, we often are teaching them to look for
opportunities to arbitrage and make markets more
efficient and yourself richer in the process. So think about the
LEGO idea as we’re going to use these cash flows. All these little LEGO
pieces are cash flows. And we’re going to build
what it is that you want. But when you do that, it’s not
always easy to see the lines. For example, was
Goldman Sachs helping Greece get into the EU
via financial engineering unethical? It’s a really
interesting question. What they did was use
these rather complex swaps to be able to take debt
off their balance sheet and therefore get in
underneath the rules that said a country couldn’t have
more than this percent debt on their sovereign
balance sheet. Goldman came up with
a way to do that. Was that ethical or not? We can talk about that later. What’s that? AUDIENCE: [INAUDIBLE] MAUREEN O’HARA: Well, we’ll
talk about that one later. It’s an interesting question. And we’ll come
back because I know this is of particular interest. Here’s one. Was JP Morgan manipulating
the California energy market? Or were they optimizing against
an inefficient algorithm? If you talk to
most of my friends in economics and
finance, they’ll say, hey, if the rules
allow me to do it, so be it. If you don’t like the
outcome, change the rules. I think that’s missing
a lot of points. And we’re actually going to
look at the JP Morgan Chase in a minute to show you
what I do in the book because the book goes
through a variety of these. And we try and figure it out. Are HFT strategies designed to
take advantage of other HFTs ethical? Some of you may have read
Michael Lewis’s Flash Boys. Michael Lewis was
adamant that the HFTs are horrible and
should be banned from the face of the Earth. But what’s interesting is that
most finance professionals like myself think that what
he picked on in the book isn’t a problem at all. There are problems. He doesn’t know them. So we’re going to come
back and talk about them. But the point of this is
that some of these things are tricky. It’s not like if I said to you,
is it unethical go rob a bank? I think everybody
would go, yeah, I think that’s
probably not good. Is it unethical to sell a
widow, an orphan a security that’s worthless? You’d say, yeah, that’s
probably unethical. They’re easy ones. But once you get into the
world of natural securities and synthetic securities
and, tell you what we’ll do. We’ll transform this cash flow. And we’ll do that. And we’ll move it over here. And we’ll do that,
It can be really tricky to keep track
of what happened to the ethical guidelines. So what are we
going to try to do? We’re trying to
draw these lines. We’re going to help
little Dogbert here, who has this view
that the easy ones are to define between the
felons and the good people. The stuff that
we’re interested in is the gray area,
or the weasels. It’s not exactly illegal. I’m actually teaching our MBAs. And the phrase, it’s not exactly
illegal, should be, I think, a red flag. Maybe this is not
exactly the way you should be running
your business. So what’s our quest? Our quest is to eat all those
really good-looking cookies. But in the process,
we’re going to sort out the positive effects of when
finance actually generates something for nothing– when we actually
can make everybody better off using these tools
of finance, which we can. But we need to sort
out those activities from the activities that
lead to the opposite outcome, where the financiers take all
the gains and society pays the cost. And that’s really what I think
finance has to strive to do. So what’s the book do? Well, for those of
you who don’t maybe know modern finance,
in about two chapters, it tries to tell you how
modern finance works. And as I tell friends of
mine who are not in finance in any way, shape, or form– I say, look, if you
read those chapters and you start getting a
little lost, just keep going. Ignore the chapter for now. You’ll come back to it later. And head into the
rest of the book. So the first part is, how
does modern finance work? And it explains to
you what swaps are and all sorts of things. And then it sets
out some frameworks for evaluating the ethical
limits of arbitrage. And I think this is tricky
in that we don’t normally think of ethical– buying
low and selling high is not a moral decision. But using arbitrage to get
around legal rules, using arbitrage to be able to
take advantage of someone because it’s so
complex they have no idea what you’re doing– that has a moral dimension. And then what we’re
going to do is– the rest of the book
ventures into the gray. And so it’s a series
of chapters that looks at essentially arbitraging
the complexity, arbitraging for deception, arbitraging
for a variety of things. There’s a whole bunch, almost
like little vignettes and case studies, including
Goldman in Greece. And here I’ll have
to make a confession. I’m a big football fan. It just shows you what kind
of patience and forbearance I have. I’m a Bills fan. I know. And it’s not been easy, has it? [INTERPOSING VOICES] MAUREEN O’HARA: I
don’t like hockey. So anyway, one of my favorite
things on TV in the old days was when you watched
Monday Night Football. They had those little segments
called You Make the Call. And they’d show this play. And they say, well, did
this violate the rules? And I loved that part. So I wrote this part of
the book thinking of that. You make the call. I tell you what happened and
then set out what happened. And then you make the call. Did this cross the line or not? And then I offer my
thoughts on whether it did or not because
again, if it’s easy, you don’t need to ponder it. And so that’s part
of the challenge. And then at the
end of the book, we try and emerge out of the
fog and talk about how to make finance more ethical. So that’s the
overview of the book. And I just thought
to give you a taste of what we’re doing today. I’m not going to explain
the basic workings of modern finance
because I have found that’s a little dry
in a setting here. I’m actually not going
to spend a lot of time on developing the
ethical frameworks, although I found
this really fun when I was writing the book
because not everybody looks at the world the same way. But there’s a surprising
amount of unanimity across a wide range
of ethical frameworks. And so I was trying to
bring that out in the book. And then what I am going
to spend some time on today is I want to let
you make the call and explain a little bit about
some of these ethical issues and where they emerge. And then we’ll briefly conclude. And then we’ll open it up
and talk about whatever people want to talk about. So before I get
there, though, I do want to talk a
little bit about why. And again, why do we seem to
have these problems in finance? And I really don’t
like the crappy people work in finance theory. It may be true. But I don’t believe it because
I’ve worked in a lot of places. And most of the people
I know are great. And they really don’t
view themselves that way. But why does it happen now? Why didn’t it just happen
in the 1920s or whatever? Maybe it did. It might have. It might have. AUDIENCE: Wasn’t there
the crash of 19-something? MAUREEN O’HARA: There was. But I think some of
the ethical issues we see today are
pretty unique to today. But let me talk about why
I think today is different. One thing is that
almost everything today takes place in markets. The largest lender in the United
States now for mortgage loans is Quicken Loans. You do that on the web. You don’t meet anybody. You don’t shake hands
across a table anymore. So the nation of
markets is important. Almost everything
we do in finance now operates in an
impersonal market. The other challenge we
have is that a lot of what happens in finance is complex. So you have what we call
“delegated behavior agency problems.” The quant who restructures
the financial product is not the trader who
interacts with the client or the senior guy who put
the whole thing in place in the first place. And the question is,
who in that group is responsible for the
ethical dimensions? And I think in practice,
the answer is none of them. The other problem are complexity
of products in corporate form. Goldman Sachs has 946
subsidiaries in tax havens alone. If you look at the structure of
a major financial institution, there are thousands
of subsidiaries. It’s extraordinarily complex. And these things often
go across markets. So as you try and sort
through who’s doing what, it gets pretty tricky. AUDIENCE: Excuse me, can
you explain “subsidiary”? What do you mean by that? MAUREEN O’HARA: So
you have a company. And then they have
other companies. They have other divisions that
are set up, in many cases, as separate companies. So Johnson & Johnson, for
example, has 286 subsidiaries. Some of them make this product. Some of them make that product. So you have a corporate
structure up here. And then you have all these
other little companies down here. And can you, as the
corporate structure, make sure that every one of
these little companies that you own at Johnson & Johnson
are behaving the way Johnson & Johnson wants? So in a bank, they have
thousands of those. And a lot of them
are located in– it creates a great
management challenge. And as a management
professor, it’s something I worry about a lot. And then there’s in personality. You never see anybody. Statistical victims
always seem a lot less compelling than real victims. So for a moment, let’s
talk about, does it matter? So I’m going to give
you something here you guys can think about. Isn’t he cute? This is a mouse
experiment that was run. And this is an experiment
about how people change when they operate in markets. So here’s the experiment. He’s awfully cute. So participants get
to decide between– this is one in Germany. And the lab that
had all these mice had used them for various
research projects. But those research
projects are gone now. They’re over. And so now what do
we do with the mice? So they set up this
little experiment. And in the first
experimental part of this, they had a group of people,
let’s say all of you. And they offered each
person the following choice. You can have 10 euros,
but one of the little mice is going to be killed. Or you can forego
getting your 10 euros, and the mouse will be spared. So they’re going to
do that treatment. Then they’re going to do
a second experiment, where they take half the room. And they say that you’re
going to be the sellers. And you’re going
to be the buyers. And so we’re going to match
up each seller with a buyer. And we’re going to give the
seller the property rights to the mouse. So each seller has
the right to decide, if they will, the
owner of the mouse. If you and your buyer can agree
on how to split the 20 euros, you get 20 euros. And the mouse is killed. Or if you guys agree that
you don’t want the money, then the mouse will be spared. Does everybody see
the difference? In the first one, each
individual gets to decide. And in the second one, the buyer
and the seller together decide. Here’s some interesting results. When individuals were given
this choice, 45% of them took the euros. And the mouse was toast. But on the other hand,
almost 55% spared the mouse. When you put those same
people in a market setting, the mouse is toast. 72% of the time, that
mouse hits the dust. In fact, to get individuals to
kill the mouse at the same rate that they’ll do it in a market,
you had to pay them 47.5 euros. Now, what does that mean? I admit that mice
are not necessarily the same thing as
people or trading and various other things. But I think one of
things it points out is that once you get
into market settings, the immediacy of some of
these issues seem to fade. And in our case here, the
poor mice hit the dust. So what are the ethical
limits of arbitrage? A lot of people say,
well, let’s just rely on the legal boundaries
to determine when we cross the line. But that’s probably
not a good idea because we can simply
arbitrage around them. That’s what modern
finance can do. You tell me what
the rule is, I’ll create a way, using my cash
flow approach, to get around it. So you don’t have to go
back as far as Aristotle. And this is the only
thing I’m going to talk about– the ethical limits. I’m going to let you guys
decide on some of these things. Aristotle pointed out that
every action had technique and prudence and that
every technical action has a moral component. And so one way to
think about it is, every arbitrage, which
is a technical action, has a moral component. And that should
really be kept in mind as we think about these. So we’re going to run out of
time because we’re supposed to keep this to 35, 40 minutes. So let me just look at
a couple of examples of the modern things
that happened in markets. And the HFT ones are pretty
easy that I’m going to show you, I think. Michael Lewis in Flash
Boys didn’t like the fact that you could design
an algorithm that would use machine
learning to try and see if you could predict where
the larger orders were going to trade and step
in front of them. So in Michael
Lewis’s book, think about the problem of a large
institution who’s going to trade, say, 100,000 shares. In our current market structure,
we have 13 different stock exchanges. So typically what happens is
you chop them up in the order. And you start sending
them off to the exchanges. But they’re not all
right next to each other. So the orders take
a little bit longer to get to some down the road– like milliseconds,
but still longer. The HFT guys write
algorithms to watch for patterns in
the exchanges that have the shortest distance
to go or the lowest latency. And then based on
that, they run ahead and put in orders in
the ones that are ahead in front of what they think
are the orders that are coming. Is that unethical? What’s interesting is almost
everyone I know in finance says no. People have watched
markets for years. If you’re a market
watcher and you say, every time I see the market
go over 60 three times a day I buy, that’s kind
of the same idea of, I’m training my machines
to watch for a pattern. And then I’m going to trade. But you might disagree. Michael Lewis does. But here are some things that
are a bit more challenging. Here’s what they do. So this is an algorithm that’s
been written to take advantage of another algorithm. So see those blue dots? That’s an order that has
been placed by a broker dealer for a client. So this client wants to buy. And so you see the market opens. And over here,
you’re going to see– those green dots
are also orders. Those are being put
in by a machine. So there’s an HFT
machine algorithm that’s been written
to put it in order and cancel it, put it an order
and cancel it, put it an order and cancel it. That all happens
within microseconds. So if you look at
those little lines, when you see three
lines, there’s an order. It gets canceled instantly. And then another order
is put one tick above, another a tick above that. You can see as you
go along at 30.08– this is at 9:30 in the
morning, eight seconds– the blue order gets put. That’s an order to buy. There isn’t a seller
out there right now. So that order is
going to sit there. But this quote dangler keeps
putting orders in and canceling them, putting them in
and canceling them. What’s he trying to do? He’s trying to fool
the algorithm that has sent in the blue order
into thinking that there is a lot of interest out there. And they’re trying
to raise the quote. And so he won’t be able to buy. And so the blue orders
stay there for a while. But then that second
algorithm gets sucked into thinking there
actually is someone out there. And you can see that the
quote dangler is basically– there’s no trades actually
taking place on this yet. These are all just
orders in the books. You can see he takes the
price all the way up there. This is three minutes later. And actually, the blue
guy doesn’t ever trade. He pulled his
order, and he quit. So the green guy didn’t succeed. What he was trying
to do was induce him to bet against himself. And then when he
gets high enough, the green guy will
trade against him. This is called being
a quote dangler. This should be illegal, but
it’s almost impossible to catch. But it’s clearly
across the line. This guy’s manipulating
the market. He is simply trying to fool
you into trading against him. Here’s another type of strategy. This is called a “momentum
ignition strategy.” So again, this is all
done by computers. So what’s happening here? The blue lines are
orders that are being submitted and canceled,
submitted and canceled, and submitted and canceled. And what they’re
trying to do is– and actually, in this case,
these are tiny little trades, like trades of a share. So you can see
what they’re trying to do is they’re trying to
move the price up and down and up and down and up and
down, getting wider and wider. And why are they doing that? Because there are
people who put what are called “stop
orders” in the book. And when the price hits
the stop order, when– suppose you want to
protect yourself. You own IBM. You put a stop order in at 60. IBM is trading at 70 right now. So unless the price hits 60,
that order won’t execute. And so what they’re
trying to do is they’re trying to find
the stops in the book. And here they succeed. So you can see at some
point, all of a sudden the price just falls through the
floor because what they’ve done is they triggered
all the stop orders. And what are they trying to do? They’re trying to
buy at the low point. So is it ethical to write
algorithms to do that? I don’t think so. Is it ethical to
write an algorithm to try and guess
where people are going and try and go there first? I think so. But others may disagree. But these markets are tricky. We’re going to do one more. And then we’re
going to do Chase. And then we’ll move
on for a minute. Remember, this guy became
famous because some people say he contributed
to the flash crash. But he’s doing something
much like that first diagram I showed you. He used a layering
algorithm in futures. And what that means is he
put lots of sell orders at prices three, four, and
five ticks above the price. So he’s trying to give you the
idea there’s a lot of depth out there. But what’s interesting
about his algorithm is that it includes
code that said, if the price ever gets
close to these things, cancel the orders. Now, to write an
algorithm that says, cancel if my order
can ever execute, has got to be unethical. It should be illegal. So in this brave
new world of the HFT and the world of modern finance,
the kind of behaviors you see are really remarkable. Let me give you one that’s
kind of my favorite. And I talk about this
one with the MBAs because I think we train
them to do exactly this. So JP Morgan Chase became the
owner of 28 outdated power generating plants when
they took over Bear Stearns in the crisis. So some of you may remember
Bear Stearns failed. And the Fed got JP Morgan
Chase to take them over. But Bear Stearns was a big
player in energy finance. So all of a sudden,
JP Morgan Chase is now the proud owner of 28
electric generating plants. And they’re all outdated. And they don’t make any money. So how do we make
them profitable? Well, what they realized was,
we could invest in those plants, spend lots of money, bring them
up to date, and go that route. But they don’t want to do that. Instead, they
realized that the way we trade energy in the United
States is really complicated. But it involves an auction. And it’s an auction that is
the world’s most complicated auction involving– there’s a day-ahead
auction, and then there’s the day-of auction, and
all kinds of things. And it’s run by a group called
CAISO, at least in California. And that stands for the
California Independent Power Authority or something
along those lines. And CAISO runs these
auctions in a way to try and make
sure that there’s going to be enough
electricity in California. But some days it’s really hot. And everybody turns on
their air conditioning. And so we’re going to need
a lot more electricity. And here’s the kicker– electricity can’t be stored. So in order to come up
with more electricity, we’re going to have to induce
some of these old power plants to ramp up and start
producing electricity. And to do that, since
they’re expensive to run, we’re going to have to
have compensatory payments. So JP Morgan realized that– let’s not think
about the problem of generating electricity. Let’s think about the problem
of selling it in the auction. And so they developed
bidding strategies to try and gather all
these compensatory payments and, if possible, not
to actually ever have to sell the electricity. They came up with 11 strategies. And all of them were within
the rules of the auction. So what would they do? Well, here’s an example. Basically, remember,
why is this arbitraged? They’re arbitraging
the algorithm. They don’t care about
the electricity market. They’re arbitraging
the algorithm. So what would they do? Well, the way this market
works is you submit a bid in what’s called the
day-ahead market– so say Monday. So they submit a
bid on Monday to be willing to produce electricity
between the hours of 11:00 PM and 12:00 AM on Tuesday. And they’ll sell it for
minus 30 a megawatt hour. And you say, minus 30? But the rules of
the auction were designed to allow wind
farms, who get subsidies from the government,
to submit bids. And so for them,
sometimes it’s better to just sell even at minus 30. So they bid minus 30. And their bid is accepted. So that means the
next day, they’re going to be producing this
electricity at night from 11:00 to midnight. And they’re going to obviously
not make any money at minus 30. But why would they do this? Well, here’s the rule. It turns out that because power
plants can’t come up and down overnight or, for that matter,
instantaneously, there’s something called ramp
up and ramp down rules. So the rules are that
you have to allow a power plant to operate for
three hours at a stretch. So once their bid got
accepted for minus 30, the next day, they bid
to provide electricity from midnight until
2:00 in the morning. Only now they want to be
paid $999 per megawatt hour. Now, what’s the normal
bid price that you get at this time in the morning? About $15 an hour. But this bid has to be
accepted because that bid was accepted because of
the contiguous ramp up, ramp down rule. So these are the kinds of
strategies that JP Morgan has come up with. We’ll put in a bid for minus 30. And now you’re going to have
to accept our bid for 999 because the rules of the
auction say you’ve got to do it. So they had 11 of
these strategies. And they started making
money hand over fist. So what do you think? Did this cross the line? Anybody think it didn’t? Well, JP Morgan
didn’t think it did. However, the problem is, the
regulator thought it did. And the Federal Energy
Regulatory Commission charged them with
market manipulation, arguing that they interfered
with and distorted the well-functioning
markets in CAISO. Now, I think– not everyone–
but most people looking at this go, are you kidding me? And people say,
well, you know, they made the market better
by revealing this flaw. And the argument I
would make is, yeah, but no one else was trying
to distort the market. I mean, this is a problem. They raised electricity
costs for everybody else. JP Morgan’s the only one who
benefits from this behavior. So these are the
kinds of questions. Personally, I think we do
train our finance students how to arbitrage in
exactly this way. And what is
particularly scary here is that the regulator, after JP
Morgan developed the first two strategies, said, don’t do it. And so they go, OK,
I’ll shut that one down. And then they
develop another one. And then they
develop another one and another one and another
one until finally they get them for manipulation. AUDIENCE: [INAUDIBLE]
from what– MAUREEN O’HARA: Enron did? AUDIENCE: Yes. MAUREEN O’HARA: Yeah. One of the reasons
FRC was able to go after them for this behavior
was after Enron, which also manipulated everything–
after Enron, they changed the
rule about what is manipulation in energy markets. And so they have a
much broader rule. Enron played every
game in the book. But they thought
they had stopped that until JP Morgan came in. And this was just last year. So I’m almost out of time. So these are just some
examples of the sorts of things that we look at in the book and
think about where exactly does the weasel zone start. I don’t think finance
is unstoppable. That is, I don’t think that it’s
the case that you can’t stop finance, that you can create
a law that we’ll always get around. But I think you have to think
more carefully about how are you going to make
it a source of good. You can try and change
the culture in finance. Some of you may know the Dutch
were so mad after the crisis because they had to bail out
two of their major banks. Now all 90,000
Dutch bankers have to swear an oath
that says they won’t take people’s money and
misbehave and things like that. Will that make Dutch
bankers better bankers? I don’t know. That’s one approach. I think you change
the way you regulate. If you tell me
what the rule is, I can tell you how
to get around it because I can build a synthetic
way to get around that rule. I think we need to change
from a world of trying to be very specific towards
a world of standards. And that’s why JP
Morgan got caught. Because JP Morgan got caught
because the rule about what is manipulation was
so broad that even though every strategy
they used was technically within the rules, the
overall intent was to manipulate the market. And they got them. And this flips the old rule that
says, use standards of people are trustworthy and
rules otherwise. I want to argue
that you don’t want to use rules in a modern market
because the minute you tell me what the rule is, I can
build my way around it. You also have to
recognize the importance of market acceptance. Goldman Sachs was trying to
get into the Islamic finance market. And the market simply
wouldn’t do it. So the market can have a role. And finally, I’d
say that I think if we have more focus on ethics
by regulators, journalists, boards, managers, protesters– go occupy Wall Street– and even finance
professors, we’ll have greater awareness,
discussion, debate, and hopefully change. So let me end there with this,
go back to our little thing. As a start, don’t be this guy. Don’t be the guy
who is arbitraging the Mexican restaurant. So thank you. [APPLAUSE] So– Yep? AUDIENCE: Just to finish
the JP Morgan one, the ratepayers really
ended up transferring value to JP Morgan? MAUREEN O’HARA: That’s correct. AUDIENCE: Thank you. MAUREEN O’HARA: That’s–
I mean, that’s basically because [? Caso ?]
is a nonprofit. So they cover their costs by
raising the electricity rates to cover– AUDIENCE: JP Morgan was
cheating their own customers. MAUREEN O’HARA:
Some of them, yeah, but not the ones who
were their shareholders. I know you want to talk
about Greece, right? AUDIENCE: Yeah, well two things. One is about Greece and the
role that Goldman Sachs played. And the second– if
I can ask a second, you don’t have to answer it– is you mentioned how colleges
are teaching how to go around. Is that what you– Did I hear that? MAUREEN O’HARA: What we
teach people how to do is how to synthetically
build things, right. So if I can teach– and we do– I mean, modern
finance is all about saying, tell me what kind of cash
flows and payments do you want, and I’ll figure out a
way to structure things to do that, right? And so because
everything’s being done with complex editions
of derivatives and everything else, they usually
don’t fall on the law because the laws
were written for sort of the traditional contracts. So we don’t teach
people to be criminals. What we teach people tools
that can allow them to create alternatives that are not
explicitly prohibited by law, and that’s the problem. And we don’t tell people think
about this before you do it, right. Think about whether
or not building a synthetic alternative to
something that’s illegal is exactly the right idea. So that’s the second question. You want to go back to Greece? So here’s the issue with Greece. What did Greece do, right. So Greece has lots of
debt and they want to get into the eurozone, but the
eurozone has rules that say you cannot have more than a certain
percentage of debt to GDP. So what does Goldman do? Goldman structures something
called an elongated swap, which is a type of derivative
that basically turns the debt from
their balance sheet into a stream of
derivative payments. Now, why does that work? It works because Eurostat, who’s
the group that wrote the rules, does not count derivatives
in counting up the debt– AUDIENCE: Off balance sheet. MAUREEN O’HARA:
Off balance sheet. So what you did– exactly way to put
it– is you took something that was
on your balance sheet and you got to take it off by
turning it into a derivative. Now it’s not on
your balance sheet, so technically, you
don’t have that debt and you’re able to
qualify for entrance. Now the debt still there. It’s just you’re going to
be paying it down the road and it was done by a derivative. Now was that ethical or not? So here’s two– AUDIENCE: [INAUDIBLE]
suggested it? MAUREEN O’HARA: Well,
the government of Greece hired Goldman Sachs to help
and come up with a solution. So let’s start– Goldman
didn’t say to Greece you know, hey, I have an idea. Why don’t you sneak in. I mean, so the
government of Greece hires Goldman and
says help us, right. One of the reasons that
they thought to do this– and again, let’s put– was that Italy was running
into similar problems about four years before. They hired JP Morgan Chase. And JP Morgan Chase did
something very similar for the government
of Italy, all right. So first thing you
say is it wasn’t the first time this was done,
because it was done for Italy already. The second thing here that
is important to understand is that the European
agency had been asked about the
treatment of derivatives and recognizing that
being able to do this allowed people to get
around these rules, but they decided not
to change the rule. This is before
Goldman helped Greece. So the regulators knew that you
could do this with the rules. They opted to leave
the rules in place. Goldman wasn’t the first
bank to have done it at the bequest of the country. And so is this illegal? Rather, it’s certainly
not illegal, right. It’s absolutely
within the rules. Was it unethical? I have to admit I come
down on the side of saying, you know what? If the regulators all knew
that you could do this and approved it, if it
had been done for Italy, if the government of Greece
hired Goldman to do it for them, it’s hard to say that
it’s necessarily unethical. AUDIENCE: Who were
the regulators? MAUREEN O’HARA:
Well the regulators are part of the eurozone, right. So the eurozone, the
government of the eurozone, are the people for
the European Union. So all the European
Union regulators. Now the argument you might
make that it’s unethical is but the people in
Greece have ended up having to pay a
tremendous price. But it was their elected leaders
who put them into this mess, right. So that’s why I think these
issues are interesting. Was it ethical? Was it unethical? I think it was not
necessarily unethical. It certainly wasn’t illegal. But the outcome has
been terrible, so– But, you know, other
reasonable people can disagree. People can say no,
you should have known that you shouldn’t do this. But I think you
should think about it. AUDIENCE: With regard
to the crisis, the most recent financial
crisis, do you think that the lack of regulations
and or ethics would have been mitigated if there weren’t
the moral hazard created by the implicit backing of
the government for the banks? MAUREEN O’HARA: Yeah. That’s a tricky
one and, you know, we definitely have
a challenge, right. I mean, what’s
interesting is that most of the subprime
mortgages that will fail were not actually insured. They were generated by some of
the major Wall Street banks, but they weren’t
insured at the time. I think there’s no question
that the insurance and the too big to fail subsidy tend to gave
the banks a hubris about what they could get away with. But here’s the
problem, you’re always going to have to have
some sort of insurance in a banking system, right. An uninsured banking
system is too unstable. So what do I think
should happen. Well, you got to
ask yourself, why did only one person
go to jail, right. I mean, whether these
things were ethical or not, many of them are
completely illegal, but nobody goes to jail. After the savings
and loan crisis, more than 600
bankers went to jail. But after this, one. So I think you have
laws, or maybe you need to write your laws better. And you need to throw
some people in jail. I think that would help a lot. AUDIENCE: And it would
get some attention. MAUREEN O’HARA: And it
would get some attention. AUDIENCE: Why do you
think it didn’t happen? MAUREEN O’HARA: You know. I don’t know. It didn’t. Yeah. AUDIENCE: You mentioned machine
learning earlier on very briefly. I have a question on that. So basically now
there’s like, an army of people out there trying to
find that magical alpha using machine learning,
and when they find it they don’t necessarily
know what that alpha is. So in that context, how do you
discuss these ethical issue if you don’t even know
why you get that alpha. MAUREEN O’HARA: Well, I think
that’s an interesting question. I don’t think there’s anything
wrong with machine learning and I don’t think there’s
anything wrong with trying to find alpha, right. I mean, markets are markets. In every market, there’s
a buyer and a seller. And after every trade there’s
someone who, looking back, goes, oops, I wish I hadn’t
done that, or good me, I did. So let’s be real clear. I think markets are great. When you’re using
machine learning, I think there is
an ethical issue, but it’s not looking for alpha. It’s when you have these
programs that are built– When you build an algorithm
and the algorithm says, OK, I’m using my
machines and they’re telling me all these signals. And then when I
get these signals, I’m going to either
buy or sell, all right. There’s nothing
wrong necessarily, but you have to be
very careful about how you build these algorithms. You also have to be
very careful about who you let use your pipes
into these systems, because you can build
an algorithm that can generate these kind of
self-perpetuating price cycles, right. Because, for example, an
algorithm that says the more the price drops, the more
I’m going to sell, right. That’s a terrible algorithm
because the more the price drops, the more you’re selling. That causes the price
to drop even more. So suppose you’ve come
up with a strategy that says I’m going to
build an algorithm that says whenever I
see a price drop, we’re going to start selling. And then the more I can
get the price to drop, I’m going to sell more and more
and more and more and more. And then as soon as I drive the
price down to a certain level, then I’ll buy. That’s unethical, right. That is completely unethical
because that’s just saying I’m going to manipulate
prices and take advantage of it. What isn’t unethical
is to say, I’m going to build an
algorithm that tries to look at patterns in volume. And when I see that volume
peaks up earlier in the day than it normally does, I’m going
to interpret that as there’s probably been good
news about this stock, because volume and news
are often correlated. So I’m going to buy
on volume and I’m going to sell when
the market’s quiet. That I don’t think is
unethical because you have a model in
your mind that says, I think stocks that
have more information are stocks I want to buy
and et cetera, et cetera. So that’s the difference I see. And I’ll tell you one
thing that broker dealer firm’s worry about. You build algorithms
that actually end up feeding off of each
other, because everybody builds their own algorithm
and they don’t actually ever check to see whether
my algorithm is actually going to create your
algorithm, and so that’s where I think the ethics comes in. So keep building your
machine learning, but just don’t take
the market down. Yeah. AUDIENCE: The thing
that surprised me when I read the flashpoints
was how rapidly you could turn these sales and so forth. What do you think
of a rule that says if you put a buy
order in the market that you can’t cancel it for 30
seconds or something like that. Or if you actually own the
stock you can’t sell it again for some reasonable– 30 seconds isn’t
very long, but it’s enough to defeat some
of these algorithms where people are paying
about $10 million to get a fiber network
that gives them– MAUREEN O’HARA: It won’t defeat
them and I’ll tell you why. All right. Suppose that you
have a rule that says if you put an
order in the book– you’re willing to sell,
for example, at 90 and there isn’t anybody willing
to buy at that price yet. So your order’s sitting in
the book and under your rule, it has to sit there
for 30 seconds. All right. Well suppose over that interval
the price blipped up to 94. Well, you don’t want to sell
at 90 now that it’s at 94. So what you’re going
to do in the minute that that 30 seconds is up
you’re going to try and cancel. But guess what? They’re faster than you are. They have already put in
an order to buy at your 90, which is going to happen as
soon as that 30 seconds is up and before you can cancel. So part of the
challenge in this world is that when you
put in rules that say orders have to
have a particular life, you expose the person who puts
it into adverse selection, because you’re not fast
enough and you never will be. So the New York Stock Exchange
has just introduced an order just like you want. And you have to
leave it there for– it’s like half a second or
something, which sounds like, ridiculous but the only people
using it are retail traders, and they all get
taken advantage of. No institution would ever do
that, because they know just what’s going to happen to them. So unless you’re going to have
your own fiber optic cable and you’re going to co-locate
in the New York Stock Exchange, I guarantee you the guy
who goes against your order when it’s against you is
going to take advantage of you all day long. Yeah. So it’s a bit of a challenge. AUDIENCE: Time for
one more maybe. AUDIENCE: You raised the point
about the Islamic markets. Could you explain that
just a little bit. Was it the bank–
nobody was interested? There was no interest. It that what that– MAUREEN O’HARA: So the Islamic
markets are really interesting. And I don’t claim to be a
particular Islamic expert, but there’s a variety of
features in Islamic markets. You can’t, for example,
charge interest, right. And there’s– you know,
Goldman Sachs wanted to set up something that was
akin to a bond– and the bond that would
pay interest– but you can’t pay interest. So they had structured this
very complicated transaction that basically resulted in
something paying interest. But because it had all
these moving pieces, each piece was compliant. And they actually got
a group of mullahs to sign off on
this thing, right. But there’s another
piece of Islamic finance that says that you should be
raising the money for causes that somehow benefit Islam. And Goldman was going to use
the money for other stuff. And so even though they had
these mullahs who signed off on it, there were
a variety of things where the rest of the
Islamic market said, I don’t care what they said. This is not a Sharia
compliant contract. And so Goldman had
to withdraw it. And then they went back
to the drawing board and met two years
later, they came back with something that
was a little bit closer to what is allowed
in Islamic finance and that actually worked. So there is no– You know, it is interesting
that religious authorities have kind of signed off. There’s no law against
these sorts of contracts, but they couldn’t sell it. And I think that’s something
that we underestimate that I think the market– you know, people– We can all look back– I know you both can– on
firms like Bankers Trust that always kind of skirted the law. And there was a big lawsuit
involving Bankers Trust and Proctor and Gamble about
a contract that Bankers Trust and Procter and Gamble– And then the court sided
with Bankers Trust, but that was the end
because the market sided with Procter and
Gamble, and nobody would deal with Bankers Trust. AUDIENCE: Arthur Anderson
does their accounts for them. MAUREEN O’HARA: Right. So anyway, thank
you all very much. It was a lot of fun. Everyone go be ethical. This has been a production of
Cornell University Library.

One comment on “O’Hara, Arbitrage on Wall Street”

  1. Michael T says:

    Talk about ethics and you get a handful of views. Talk about using arbitrage to make money, regardless of ethics and people milking it get 100k+ views. Shame.

    This talk is where there is true value. The ethos of what we are doing to make money- what value are we truly contributing into the economic system is being overlooked by how easy it is to re-sell something at a higher price.

    Thank you for this lecture.

Leave a Reply

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