The following content is
provided under a Creative Commons license. Your support will help
MIT OpenCourseWare continue to offer high quality
educational resources for free. To make a donation or to
view additional materials from hundreds of MIT courses,
visit MIT OpenCourseWare at ocw.mit.edu. ANDREW LO: Last
time when we met, we saw that the yield
curve was somewhere-- the short end was somewhere at
the 30 to 40 basis point level. And let's see where it is today. The yield on a three
month treasury bill, according to this, is at
71 to 72 basis points. So that's pretty good. That's better than
it was last week. There was a point, actually,
earlier this morning, that the yield curve was-- the short end was
slightly above 1%. But it's now come back down,
because of additional trading and demand for these securities. But that suggests that at least
the panic is not as severe as it was last week. Things are getting a bit better. And not surprisingly, the reason
they're getting a bit better, is because there's
more certainty now that something was
going to happen. When we met last, it seemed
as if there was a possibility that this wasn't going
to happen at all, that there was going to be some
breakdown between Democrats and Republicans, and that
there was an impasse. Fortunately, that got
resolved over the weekend. At least it seems to be. It's going to be voted
on as we speak actually. So hopefully, we'll
find out by the end of class or end of today
whether or not it happens. If it doesn't
happen, what do you think is going to happen
to the three month? Yeah, so you could actually
look at this as a thermometer. Check the temperature
of our economy. It's pretty amazing, isn't it? It tells you that financial
markets are very dynamic, and that you actually can
learn a lot from market prices. Again, are market
prices correct? No, there's no such
thing as correct. I want you to get away from
that notion of correct. There is a market
price that reflects the aggregate sentiment of the
economy and the participants on a given day, at a
given point in time, with a certain set
of market conditions. And then you have to
decide whether or not that set of prices is something
that you would like to use in your own calculations. So right now, these
are the prices that reflect what's
going on in the economy. By the way, at the
long end, last time we saw that two weeks ago, the
long end of the yield curve was pretty high, because
of concerns that there was going to be inflation. And then last week, we
saw that it went down. What is it now? Well, if you take a look at the
30 year, the yield is at 422. That's slightly
lower, not by much, but it's slightly lower
than what we saw last time. And certainly lower than
what it was two weeks ago. So the concerns about inflation,
while they're still there, at least from the
data here it looks like they're a little bit less. So are people right today
and were wrong last week? Who knows. The point is that this reflects
what the current market sentiment is. And so at every point in time,
when you look at market prices, what you're getting is a
window on current expectations and current information, and you
have to make the best of that. Any other questions? Yup? STUDENT: I just have
sort of two questions. One is that, when the three
months treasuries are so high, we said it was just a
couple basis points, why wouldn't you
just short those? Because don't you
have a [INAUDIBLE],, they can't go above 0. So you have a couple
basis points downside, and [INAUDIBLE]
basis points upside. ANDREW LO: That's right. You could have shorted them. Andy, do you want to answer? ANDY: I'm not sure I agree
that you can short them. ANDREW LO: OK, why not? ANDY: Because going
short that means that you want to
borrow money at 3 basis points for three months, but
you're not the US Government. And no on will allow
you to do that. ANDREW LO: Well, it would be
hard to borrow the securities and then sell them, right? And unfortunately, you can't
manufacture the pieces of paper the way the US Government can. It's kind of hard
to do the printing press in just the same way. In fact, I think it
may even be illegal. But you're right that if-- it's such a low
level, what you would like to be able to
do is you'd like to be able to issue that stuff. And by the way,
the US Government did take the opportunity to
issue some paper last week to take advantage of this. Because it's a great
way to do it, right? You borrow money at
virtually zero interest rate because you are
the US Government, and all you need
to do is print up these wonderful certificates. But I think the issue
is exactly right. If you wanted to
short it, you've got to be able to borrow it from
somebody else and then short, and they have to let
you borrow it from them at appropriate premium. So there's a risk
and a price for that. But if you could do it, it
was a pretty good trade. On the other hand, think
about what you're saying. What you're saying
is that you would like to be able to allow
people who want liquidity to have liquidity. You would like to provide them
with that kind of a liquidity. If everybody is panicking
and wanting liquidity, then that might be
a very good strategy because when markets
calm down, eventually, you will do quite well. In effect, that's
what the US Government is hoping to do with
this so-called bailout package, which is
what I mentioned last week that bailout is
probably not the right term. It's a rescue
package undoubtedly. But whether or not it's
a bailout or a very savvy investment depends
simply on the price-- on the price that
you can get it at, and the price that you
ultimately sell it for. So that remains to be seen. Other questions? Yep? STUDENT: I don't know the
details of the [INAUDIBLE],, but I'm wondering,
this crises is based on the whole economy is
leveraged on some assets that are not really working
or are worth less than they were supposed to. And I wonder, at the
end, would the people that have credit, but
bad credit, suffer? Will they save
their homes or not? I don't see--
because the only way I see for this to be corrected
is to go to [INAUDIBLE].. There's a lot of people
leveraged that cannot pay so how will this get to-- Am I explaining myself? ANDREW LO: I think so. I think so. I think you're expressing
the same kind of concern and confusion that the
American public has expressed at the bailout package. Because it doesn't
seem like the bailout is really applying to the
ultimate root cause of this, which is the home owners. The politicians would say
that you're bailing out Wall Street when
you should really be bailing out Main Street. Let me hold off on answering
that, because it turns out that this Thursday, October
2nd, from 5:30 to 7:00, the Sloan School
will be organizing a panel discussion
of the bailout, as well as the root causes
of some of these issues. So rather than take up
any more class time, let me defer that question
to that Thursday panel, and then I'd be happy to
talk about it afterwards. But I'd rather make
sure that we stay on track with our
curriculum and just use this as an illustration. But let me give you the
short answer to the question. The short answer
is that the idea is that you have to deal with
the current crisis right now. So it's sort of like having a
patient come into the emergency room and they're bleeding
out, and it turns out that the reason they're bleeding
out is they've abused drugs and they've done all
sorts of bad things to their diet and health. Now, at that time,
you probably don't want to give a lecture
on good nutrition and the dangers of
recreational pharmaceuticals. You've got to stop the bleeding. And then, over the course of
the next few weeks and months, you try to rehabilitate
the patient. So what the package is
meant to do, first of all, is to stop the bleeding. And then, over time, we're
going to have to address exactly the issues that you raised. And that's part of what the
proposal was trying to do. That's why it took them
time to put it together. It's easy to figure out
how to stop the bleeding. Money will stop the bleeding. But the problem is that
throwing money, good money, at bad assets is not necessarily
the long run solution. You have to figure out what
the ultimate causes are dealing with foreclosures, dealing
with all of these very complex securities, figuring
out how to value them, coming up with proper
insurance agreements to be able to create stability
across the entire market. And that's what the various
aspects of the bailout package are designed to do. So we'll talk about
it on Thursday, and I would encourage
all of you who are interested to
come to that session. We've got a number of economists
and accounting faculty and other folks who are
going to be there to present. You'll get a notice about that
probably later this afternoon. STUDENT: One more thing. On the Wachovia deal, what's
going to happen with the bank? Is it going to
continue the same? ANDREW LO: Well, obviously
that's a work in progress. It looks like most of
the units of Wachovia will be sold off to
Citigroup, but there are a few units of Wachovia,
including AG Edwards, which is a broker dealer,
and Evergreen, which is another broker dealer,
that will remain separate and will be freestanding. So that will not be
acquired by Citigroup. But apart from that, all
the other units of Wachovia will be taken on by
Citigroup, and that there will be a backstop provided
by the FDIC in case the losses exceed
more than $40 billion. So Citigroup will
be able to take that onto its balance sheets. And in exchange
for taking on all of these bad debts
and other problems, Citigroup gets the
retail access to all of the various
different channels that Wachovia has set up. So now, Citigroup
has the ability to compete head-to-head with
Merrill Lynch having been acquired by Bank of America. Whereas before, they wouldn't
have been able to do that. So you see, this is what
I was saying last time, that with every kind of crisis,
with every kind of dislocation, there are opportunities
that are created. And so when you have
one door closing, three other doors
open for opportunities that can be taken advantage of. And by the way, let we
mention, this is also true for your careers. You might be discouraged
about financial services. I would argue just the opposite. Right now, all of you are
at an excellent position as first year students,
because first of all, you're here in school waiting out
the passage of the storm. And when the storm
passes, believe me, there are going to be
tons of opportunities. In fact, typically the
largest growth period for jobs is not at business cycle
peaks, but its exactly after these kinds of
troughs that occur. So within the next
6 to 12 months, there's going to be tons
of career opportunities. In fact, for those
of you who are interested in going to
the New York Banking Day, and you really should if
you're interested in a career in finance. My guess is if you visit
Goldman Sachs, Morgan Stanley, as difficult as a
set of circumstances they're in right now, my
guess is every single one of these firms will be hiring. And the reason they're
going to be hiring is because they want to take
advantage of the opportunity to cut costs and to
hire younger, more energetic employees to
be able to really beef up their future generations
of human capital. So they're going to be
making an investment in that. So I think that's a good example
of how it's true that you're going to have consolidation. So now, after
this, there's going to be three major money center
banks, JP Morgan Chase, Bank of America, and Citigroup,
which is astonishing because just a few months ago
there were quite a few others. So the landscape has changed. But the competitive
landscape changing means that opportunities
get created along the way. STUDENT: I was just wondering
from your point of view. Why is it better to have the
banking industry consolidated into three buckets? In that, wouldn't
it have been better to let Wachovia fail and
let the regional [INAUDIBLE] pick up the slack instead of
now having literally JP, B of A, and CitiGroup dominate
the entire landscape and be in a position to
monopolize [INAUDIBLE] going forward. ANDREW LO: Well, so that's
an interesting thought, letting Wachovia fail. Obviously, you're not
a Wachovia customer. I think that what's
happening right now is that there's a great
deal of sensitivity, not only on the part of
Wall Street, but regulators, to stem the tide of
mass financial panic. We talked a bit
about that last time. The reason that regulators
and the government sprang into action was not
because Lehman went under, or AIG went under, or
any of these other large organizations. The reason that finally got
them over the edge of moving to do something substantial
is because the reserve fund, a retail money market
fund, broke the buck. And if that happens on a regular
basis beyond the reserve fund, you will have a very, very
significant financial market dislocation. It turns out that Wachovia is
part of that retail network. And if you let
Wachovia fail, you risk igniting further problems
in that retail sector. Citigroup is perfectly
happy to take them over and are able to
given their balance sheet-- are able to manage
that without any problem. So that seemed like
an ideal solution from everybody's perspective. Because if you allow
Wachovia to fail, remember, the FDIC is on the hook to pay
all the depositors their FDIC deposit insurance up to
$100,000 per name, per account. That could be a very substantial
number by letting the bank fail and by having all of its
value completely lost. This way, they actually
preserve a fair amount of value, because as an ongoing
concern, Wachovia has quite a lot of good business. So it actually is the
cost minimizing solution, but at the same time
it also preserves the current fragile integrity
of financial markets at least until the bailout fund is set. My guess is that in about three
or four weeks, if we have banks that end up not being able
to make their commitments, they are going to
be allowed to fail. Because at that
point, those failures won't jeopardize the
entire financial system, they'll be dealt with by
this bailout organization. So I think that
that's the logic. Yeah, last question,
let's move on. STUDENT: Is that
the same thought process as freezing Washington
Mutual's failure [INAUDIBLE].. ANDREW LO: Well, that's right. But the difference there is
that Washington Mutual has much bigger exposure to
these subprime loans, and so I think in that
case, there really wasn't much of a choice. And very much so transferring
the business units that are able to be moved
over JP Morgan Chase would make a fair bit of sense. So there's a lot of
consolidation going on in this industry, but
once again, consolidation, while it seems like
it's a big upheaval, and it is for the people that
are at these organizations, it's very disruptive,
the fact is that these kind
of disruptions are part and parcel of how
businesses grow and develop and morph over time. In fact, if you went
back to the 1960s, and you looked at a Wall
Street Journal on microfiche-- I happened to do
that just because I was looking for a particular
citation at one point-- if you look at the
advertisements in the 1960s or even 15 years ago, you
look at the advertisements in the Wall Street
Journal in those days, there are names of financial
institutions that you've never heard of, that were really
big institutions back then. So it's rare that we
have institutions that survive for 50, 75, 100 years. It's part and parcel of
how businesses develop. And the key is to focus on the
process by which businesses change. So when we start talking
about equity evaluation, we're going to see that by
looking at income statements and balance sheets together,
we can see not only what's a good business and
what a bad business, we can also see how
businesses evolve over time. And it's that
evolution that we hope to try to bring across
to you in this course. I want to show you how it
is that you can understand the dynamics of changes
in business conditions, because that really is, I think,
the key to a lot of what you can use in your own careers. I know there are more questions. But let me hold off on those
and start on the lecture today and then we can cover those a
little bit later on after we've made some progress. So this is a continuation
of last lecture where we were talking about
convexity and duration as two measures of the
riskiness of a bond portfolio. And I concluded last lecture
by talking about the fact that if you think about
a bond as a function of the underlying yield, then
you can use an approximation result that says that the bond
price, as a function of yield, is approximately
going to be given by a linear function
of its duration and a quadratic function
of its convexity. So we have an
approximation that says that the price of
the bond at a yield y prime, is going to be equal
to the price of the bond at a yield y multiplied by this
linear quadratic expression. And really, the
purpose of this is just to give you a way of
thinking about how changes in the fluctuations
of a bond portfolio, as well as the curvature
of that bond portfolio, will affect its value and
therefore its riskiness. These are just two measures
that will allow you to capture the risk of a bond portfolio. So I have a numerical
example here that you can take a
look at and work out, and you can see how good
that approximation is. This is an approximate result
that the price at a yield of 8% is going to be given as
a function of the price of the bond and a
yield of 6% multiplied by this linear
quadratic expression. And the actual result
of the bond price, now that we have high speed
digital computers that can calculate all of this
at a moment's notice, you can see the difference. It differs basically
by about a penny. A penny it's not a big deal. But when you're dealing with
billions of dollars actually, a penny is a pretty
significant amount. So what you want to
do is to make sure that you use the right
formula to calculate it. I wouldn't argue that you should
use convexity and duration to do any kind of
bond pricing analysis, but for a quick and dirty
method for getting intuition about how risky a bond portfolio
is, the two questions you ought to ask somebody is,
what's the duration and what's the convexity. And what those two numbers, you
can develop a kind of intuition for how the bond
price is going to move in response to underlying
changes in the yield curve. And right now, we
see that the yields are changing pretty rapidly. The Treasury yield curve,
at least at the short end, is bouncing around
depending on what happens every day in Washington. And so if you have that sense
of short term yields changing, by looking at
convexity and duration you can get a sense of how
sensitive your portfolio might be to those kinds of exposures. The last topic I'm going to
take on is now corporate bonds. Up until this point, the
only thing that we focused on has been default free
securities, namely government securities,
because governments can always print money and
therefore they can always make good on the claim that
they will pay you a face value of $1,000 in 27 years. There's no risk that they can't
run those printing presses. What I want to turn
to now is risky debt, and in particular
I want to point out that risky debt is fundamentally
different in the sense that there is a chance that
you don't get paid back. So one of the most
significant concerns of pricing corporate
bonds is default risk. And the market has
created its own mechanism for trying to get a sense of
what the default risk really is. Namely, credit ratings. These are ratings put out
by a variety of services. The services that
are most popular are Moody's, S&P, and Fitch. And these services do
analyzes on various companies, and then they issue reports,
and ultimately ratings, on those companies. They'll say this company
is rated AAA, AAA being the highest category. And I've listed the
different ratings categories for the three
different agencies here so you can get a sense
of how they compare. Typically, these
ratings are grouped into two categories, investment
grade and non-investment grade. And really, the
difference is just the nature of the default
risk or the speculativeness of the default probability. Bonds that are below
investment grade have a higher default rate,
and bonds that are supposedly investment grade
are ones that are appropriate for prudent and
conservative investments. STUDENT: Do you mind
maximizing the slide? It's a little hard
to read back here. ANDREW LO: Oh, sorry about that. Thank you. Yeah, that's better. So investment grade for Moody's
is AAA, high quality is AA, upper medium quality is A,
and then medium grade is BAA, and then anything
below BAA is considered non-investment grade. Now, the one thing you have to
keep in mind about fixed income securities is that
apart from some of the more esoteric strategies
that we talked about last time like fixed income arbitrage,
this idea of taking a bunch of bonds and figuring
out which ones are mispriced and trading them, apart
from those strategies, most people invest in
bonds not because they want exciting returns. If you want exciting
returns, you put your money in the stock market or real
state or private equity or other kinds of
exciting ventures. Bonds are supposed to be boring. You put your money in, and five
years later you get your money out with a little extra. That's what bonds
are supposed to do. And it wasn't until the 1970s,
when the era of junk bonds came on the scene, 70s and 80s,
with Michael Milken and Drexel, Burnham, Lambert,
that you really had a very different face
of fixed income markets. By and large, fixed income
markets dwarf equity markets. But the reason that
they're so large is because most people use
them as a kind of a safe haven. And as you get
riskier and riskier, it starts to look less like
bonds and more like equity. In fact, if you think about
the bankruptcy process, if you've got a risky corporate
bond, you're the bond holder, and the company
declares bankruptcy, they can't pay your interest
payments that are due to you, when they declare
bankruptcy, then at least from a
theoretical perspective, you the bondholder now
become equity holders. You own the assets. Because they can't
pay you, so they're obligated to give you
control of their company. So as bonds become
more risky, they start to look more and more
not like debt, but like equity. That is, the returns
are random and you don't know what you're going to get. It's sort of a
surprise every day. It's the gift that
keeps on giving. But for the most part, investors
that are invested in bonds aren't looking for that. We're looking for safe returns. And they're looking
for the highest yield that is a safe return. So investment grade
is the category that typically pension
funds, endowments, and other relatively
conservative institutions look to. Within that category,
they would like to get as much
yield as possible. So which of these
different grades do you think offers
the highest yield? Why is that? Yes, you're right. Why is that? What's the logic for that? STUDENT: [INAUDIBLE] Exactly. Given that it's lower
rated, that means it's got a higher
probability of default, you've got to pay investors a
little bit of extra for them to bear that risk. Simple as that. So the reason that there are
multiple categories, even in investment
grade, is that there are different levels
of risk aversion that investors want to take on. Some investors are
highly risk averse, and for the very, very
risk averse investors, they're going to take on AAA. And for those that are a
little bit more adventurous, they'll take on lower grade. And for those hedge funds, who
are looking for lots of risk and lots of return,
they're the ones that are dealing in the
non-investment grade issues. Those are the ones where you
have relatively large returns, 15% or 20% returns,
you didn't think you can get a return of
15% to 20% for bonds, but you can if
there is a 5% or 10% chance that you
won't get anything. So when you do get
paid, you get paid well, but you don't always get paid. So that's the
categories that are developed by the various
different ratings institutions. And once you get a rating,
that allows you to approach investors and say,
OK, this is what I'm looking to get
for my corporate bond, and what I'm hoping to get is
commensurate with the risks that we're bearing. Here's a little history of the
yields on Moody's BAA bonds minus the US 10
year treasury yield. So this spread tells
you what the difference is between a very
safe asset and a BAA asset, which in this
category, is just above non-investment grade. So it's the lowest grade
that you can get and still be passing. This is sort of like
the 65 or something of junior high school
and high school. So that spread between
BAA and US treasuries is an indication of the risk
premium implicit in the default potential of a BAA bond. And look at how it's changed. In the 1930s, this spread was
about 7 and 1/2 percentage points. That's a big spread
by today's standards. Now of course, by
today's standards, literally today,
things are different, and we may be getting
up there soon. But let's take a
look at where we were at least where the data
ended, which is back in 2005. At the end of this
dataset, the credit spread was maybe 1 and 1/2 to 2%. That's at a near historic low. Now, you can see that there
are a little bit of a blip every once in a while. December. 1987, this is after the stock
market crash of October 87. You see a big blip going up. And September of 1998
after LTCM, that goes up. And then of course credit
spreads widen over here. September 11 happened,
2001, over here. And so credit spreads got
as high as something like 3- 3 1/2% percent, and now,
prior to what's happened over the last several weeks, credit
spreads were at a close to all time low. What does it mean when credit
spreads are really low? What does that tell you? What does it say? Yeah. STUDENT: [INAUDIBLE] ANDREW LO: Right. That's one interpretation,
that the market is perceiving the default
risk as not as significant as it used to be. Another way of
interpreting that is that the investment
population is less concerned about the default
risk than back in the 1930s. Not surprisingly. Something did
happen in the 1930s that was kind of significant. What was that? The crash of 29, and
then the depression that led from that crash. So that tells you that at
least at the end of 2005, beginning in 2006, people were
less risk averse, at least on paper what this shows. What else does it tell you
about the probability of credit? STUDENT: [INAUDIBLE] ANDREW LO: Exactly. Lots of money. Another way of interpreting this
is that there's lots of money out there. Lots of money willing
to be lent out to all sorts of risky
ventures without much in the way of expectation
that they should get paid a much larger premium. So those two interpretations
are likely to be both true. That is, the
population of investors did seem less risk averse, and
there is empirical evidence to support that. But on top of that,
it also suggests that there's tons of
money out there being lent to various different
projects, and because there's so much money, there's such an
increase in supply of funds, the extra premium that is
commanded by those funds could not be that great, simply
because of the competition to supply funds to these
various risky ventures. So if you wanted
to do a startup, the time to have
done it was in 2006, because you would have gotten
great deals since there was so much capital out there. Now that's changed. But part of the
reason it's changed, part of the reason that we're
in the current financial difficulties that we're
in, is because there was too much money
chasing too few genuinely good opportunities. And so we're seeing
now the after effects of some of those
poorer investments in those opportunities. So this kind of
credit spread picture can give you a sense of
the dynamics of money flows within the economy,
and definitely worth keeping track of. Now, there are a number of
things that are in that spread, in that premium. Obviously, there's an
expected default loss, but there's also tax effects. There's also some other kind
of systematic risk premium that has to do with
aggregate risk exposure. And a variety of
other academic studies have been done to
decompose that spread into different components. Graphically, you can see
that if you look at-- if you take a look at the
composition of that premium, you can show that part of it
is due to default, part of it is due to the riskiness of
the particular investment, and then the other part is
simply the default free. That's the part that we've
studied up until today. So the other two parts, the
other extra risk premium, is really decomposed into a
default risk premium, but also a market risk premium. That is, just general riskiness
and price fluctuation. People don't like
that kind of risk, and they're going to
have to be compensated for that risk
irrespective of default. Just the fact that
prices move around will require you to reward
investors for holding these kind of instruments. And in the slides, I give you
some citations for studies on how you might go
about decomposing those kind of risk premiums. So you can take a look
at that on your own. But the last topic
that I want to turn to, in just a few minutes
today, before we move on to the pricing
of equity securities. The last topic I want to
turn to is directly related to the problem of
subprime mortgages. I promised you that I
would touch upon this. I'm not going to go through
it in detail, because this is the kind of material
that we will go through in other sessions on the
current financial crisis. But I want to at least
tell you about one aspect of bond markets
that's been really important over the last 10 years. And that is, securitization. Now, when you want to
issue a risky bond, as a corporation or
even as an individual, you have to deal with a
counterparty, a bank typically. Banks were the traditional
means of borrowing and lending for most of the 20th century,
and up until the last 10 years. But about 10 years ago, an
innovation was really created. Actually, it wasn't
created 10 years ago but it really took
off 10 years ago, where instead of borrowing
from financial institutions like banks, you were able
to tap into the borrowing power of financial markets. This is what's often
called disintermediation. Banks are considered
intermediaries. They serve as a conduit between
us, the retail investor, and financial markets or
other counter-parties. They stand in the middle. They take money from
us, put it in deposits, they take those deposits,
lend it out to corporations, and they take money
from corporations, and bring it into their
bank, and lend it to us in the form of mortgage
payments-- mortgages, so that we can buy our house. About 10 years
ago, intermediation started to unwind because of
innovations in securitization. The idea being that we are going
to instead of dealing directly with banks, tap into the
power of financial markets in borrowing and lending. And so I want to give you an
example of how that works. Something that I went
over in the Pro Seminar. But for those of you
who didn't attend, I want to show it to you because
it's such an important idea. And this is an idea
that is best done through a very simple
numerical example. So in about 10 or
15 minutes, I'm going to illustrate
to all of you the nature of problems in
the subprime mortgage market. That's all it'll take. To get to the bottom
of it could take years. But at least to understand
what's going on, I'm going to do this
very simple example. Suppose that I have a bond,
which is a risky bond. It's an IOU that pays $1,000
if it pays off at all. So the face value of
this bond is $1,000. But this is a risky
bond in the sense that it pays off $1,000
with a certain probability, and it pays off nothing
with another probability, let's say 90 10. So the simple expected
value of this is $900. And so you might think
that that should be a proxy for the market price. And in fact, that would be
a pretty good approximation. Now, right there is
an interesting insight into the pricing of risky bonds. Because if we have a security
that has $1,000 face value, but it's got a
probability of default, and you compute the
current value as $900, then that gives you an
implicit yield for the bond. What yield is it? It's whatever number, one
point something multiply by 900 gives you $1,000. So the very fact
that it defaults, now allows us to compute a
yield without reference to the time value of money. The time value of money
can add an additional piece into this calculation. I decide to ignore it
just for simplicity. But suppose the
interest rate was 5%, the risk free interest rate
was 5%, then what I might do is to say, OK, $900 is what I
expect to get out of the bond. I'm going to take that
$900 and discount it back a year by 1.05, and
that will give me a number such that
when I compute the yield on that number
relative to $1,000, it will have the total
yield of this bond. 5% of which is the risk free
part, and the other part is the default part. But I want to keep
the example simple. So let's just assume that the
risk free rate of interest is zero 0. So I've got my bond that
pays off $1,000 next period with probability 90%. So the expected value is 0.9
times 1,000 plus 0.10 times nothing. $900 for this bond. Now let's suppose that I have
not just one of these bonds, but I have two of them. And they're absolutely
identical in every respect. They're just two
of the same bonds. For each of the bonds, you
might think that it's not that easy to find a buyer. And you're right, because a 10%
default rate is pretty risky. In a minute, I'll
show you how risky when we look at the default
rates, historical default rates, of bonds with various
different credit ratings. But right now,
with a 10% rating, this bond would be
rated below BAA. It would be below
investment grade. So you're not going to get a lot
of people that want to buy one. In fact, we can auction
it off in this class right now to figure
out what the price is. My guess is that
I may not even get $900 for that in this class,
given your current mood and liquidity issues. But I'm going to show
you some magic that will make this incredibly
interesting to a large number of investors,
including all of you. I'm going to take
these two bonds and put them together
in a portfolio. Now, what exactly
does that mean? So far I've said nothing. I've drawn a circle
around the bonds. By portfolio, I
mean that I'm going to create an entity, a
corporation, whose sole purpose it is to buy these two bonds. And therefore, the
fortunes of the corporation are tied not to the performance
of any one or two bonds, but to the performance of the
collective portfolio of bonds. So that's what I mean when
I say, form a portfolio. I mean, consider a
single entity that will hold both of these bonds. And let's assume, for
the sake of argument, that the default of these
two bonds is uncorrelated. In fact, I'm going
to assume that these are two separate coin flips,
and they're different coins, they have the same probability
of coming up heads 90%, 10% tails, but they're
different coins. They have nothing to
do with each other. So they're independent. Whether or not one
bond fails has nothing to do with the other bond. When I put this
into a portfolio, how does the portfolio
behave looking at it as a single entity? There are three
possible outcomes. Actually, there are four,
but only three of them are really distinct. Both bonds will pay off,
or both bonds will default, or one bond pays off
and the other defaults. Those are the only three
outcomes that are possible. And the payoffs
and probabilities, assuming that they are separate
and independent coin tosses, is given in that table. $2000 if they both pay off. And the probability of
that is 81%, 0.9 times 0.9. And I'm multiplying, because I'm
assuming they're independent. The probability that they both
don't pay off, in which case my portfolio is worth
nothing, is 1%, 10% times 10%. And then whatever's left
over is in the middle. That is, there's a
chance that one of them pays off but the other
one doesn't, and then the portfolio is worth
$1,000, and there's an 18% chance of that. So here's the stroke of genius. The stroke of genius
is to say, I've got these two securities
that are not particularly popular on their own. What I'm going to do is to
stick them in a portfolio, and then I'm going to issue
two new pieces of paper, each with $1,000 face value. So they're just like
the old pieces of paper, but there's one difference. They have different priority. Meaning there is a
senior piece of paper and there is a junior
piece of paper. The senior piece of
paper gets paid first, and the junior paper only
gets paid if and when the senior paper gets paid. So I'm going to issue
two pieces of paper. The blue is the senior
and the red is the junior. The senior paper I'm going
to call the senior tranche. Tranche, I believe is the
French word for trench, which seems much more
appropriate today than it did before. We're digging our
own trenches here. The senior paper
is going to have first dibs on that portfolio. And the junior
paper will only get paid after the senior
paper gets paid. And so let's see what
happens with that. Remember, they're both
$1,000 face values. So on paper, I've
done nothing in terms of creating or destroying the
total claims on the asset pool. The portfolio has claims
at $2000 of face value, and my new securities has
claims on $2000 at face value. But all I've done is to change
the order, the priority, of the payout. Here's the table. I have three values for my
portfolio, $2000, $1,000, and nothing. Now let's see what happens
to each of those two claims, the senior claim
and the junior claim, of my new securities
that I've issued The senior claim
gets paid $1,000 if both bonds in my
portfolio pay off. But the senior claim
also gets paid $1,000 if only one of those
bonds pays off. So two out of the
three outcomes are good news for the senior debt. And in the third case, where
both of them don't pay off, then the senior
paper is out of luck. Now, the junior paper
is exactly the reverse. The junior paper only will get
paid if both bonds pay off, because in that case,
the senior guy gets paid and then there's money
left for the junior guy. In the latter two cases,
if only one bond fails or both bonds fail, then the
junior claim gets paid nothing. So what I've done is to
take two identical bonds, and I've created two
non-identical claims, such that one is a lot
safer than the other. How much safer? Well, the bond that is senior
has a 1% chance of default, 1%, because both bonds have to fail
before the senior guy doesn't get paid. 1%, what was it before? It was 10% for both bonds. But because I stuck
it in a portfolio, and I changed the
priority of payouts, the senior claim now
looks a lot safer. But that's not a free lunch,
because the junior claim is a heck of a lot riskier. The junior claim now loses
money 19% of the time. It used to be the case
that one of these bonds had a 10% default rate,
but the junior claim has a 19% default rate, 18%
plus 1% from those two outcomes. As long as investors
know the structure, nobody's getting a good
deal or a bad deal. There's no cheating going on. We explain this to
investors, so you all see these probabilities. And now, let's calculate
what the expected values are for the payouts. The expected values--
before I do that, let me comment on default rates. So a 1% default rate
seems like a small number. And a 19% default rate
seems like a large number. Well, let's take a look at
the empirical evidence given debt ratings. These are historical
default rates for bonds from 1920 to 1999. So I've got almost an
80 year record of bonds that have been issued
by corporations and that have been stamped by
Moody's with their ratings. And the different
bars correspond to how long the bonds have been
issued and are outstanding. Because obviously, the
longer the bond is out there, the more likely it is
that it will default. So you have to separate them
by the years out in the market. If you take a look at a
five year period, that's the bars all the way
to the extreme left of each rating category, you see
that for a five year periods, the default rate
of AAA securities is well, well below 1%. It's measured in basis
points, that probability. If you wait 20
years for AAA bonds, the default rate goes
up to maybe 2% or 3%. So that means that when
AAA bonds are issued and you wait for 20 years
to see what happens to them, it's a very, very small group
that ends up defaulting, if they have a AAA rating. On average, AAA bonds default
maybe 1% of the time or less. On the other hand,
if you take a look at below investment
grade-- so BAA is just at the borderline
of investment grade. And if you take a look
at the default rates here, lots higher. But by lots higher, we're
talking about 5% to 10%, 5% to 10%. So based upon these
categorizations, we can now rate our own bonds,
what I just decided to issue with these securities, right? The senior tranche is
rated AAA, and what would you rate the junior tranche? BA maybe. BAA at the best, but
probably more like BA. So the senior tranche
looks pretty good, and the junior tranche
looks pretty bad, but you know what their
ratings are, and therefore, go ahead and price
them accordingly. So let's do that. If you price them
accordingly, what happens is that the senior tranche
gets priced at $990, and the junior tranche gets
priced at 810, 990 and 810. Now, this is very different
from what the price was before. The price for both
bonds was 900, right? The expected value
of the payout. And the expected value
of each bond is 900. When you add them
up, you get 1,800. Here, when you add up these
two bonds, you also get 1,800. So I have neither created
nor destroyed value. All I've done is to
reallocate that value. I've given the senior
bondholders lower default risk and therefore higher likelihood
of getting their money back, therefore a lower
yield is necessary in order to sell that bond. On the other hand,
that extra benefit that we've given to
the senior claimants comes from the junior
claimants, and therefore, they get a lower price,
or they have to be given a higher yield, to entice
them to bear that kind of risk. Now, why is this such
a stroke of genius? It's because what we've done is
take two identical securities that nobody was
particularly excited about, and we've created, by this
securitization process, we've created two other
securities that actually a number of communities
are very excited about. For example, the pension
funds, endowments, foundations, all of the very
conservative investors that want a very
boring bond portfolio. No excitement, no headline risk. They just want their money
back with a reasonable rate of return. They've got it with
the senior tranche. And by the way, if they're even
nervous about this very, very safe structure, let's insure it. Let's get a large,
stable insurance company, oh I don't know, maybe
AIG, and let's get them to insure that these won't
default. Because if they do, AIG will pay an extra
premium on top of that. So then they're called
super senior securities, super senior tranche securities,
because they've got guarantees on top of the
securitization features. Pension funds love this. They bought this in
large, large quantities. Now, what about the
so-called toxic waste that the junior tranche? Well, we know it's toxic waste. We've priced it as if
it were toxic waste. And so those investors that are
looking for 15% or 20% returns, that are not looking
for boring, safe assets, they will go for
the junior tranche. Namely, the hedge funds. And as many of you
know, hedge funds have grown by leaps and bounds. Over just the last
10 years, they've increased their assets
under management by a factor of 10 to 20. And that money is
looking for a home. And boring, safe
investments that earn 6%, 7%, 8%, that's
not for hedge funds. They are looking
for 15, 20, 30 40%, and they get that with
that junior tranche. That's why the market, over the
last 10 years, has exploded. It's twofold. It's because money has
come in from pension funds for the senior debt. Money has come in from the hedge
funds for that junior debt. And together, they brought
much, much more money into this business
than ever before. And the question is, how
do you take that money and push it out to people. Well, it turns out that because
housing markets are going up, that was a perfect
way to get these this money out to investors. Yes? STUDENT: This model is based on
the probability of the Moody's. I have a question. Is there any way investors
can estimate this probability by themselves instead
of relying on Standard and Poor's or Moody's? ANDREW LO: It's very
difficult for investors to estimate the
probabilities themselves, because they don't have
access to the same data that Moody's and
S&P and Fitch do. Typically when you
estimate the probabilities, you need data in terms of
the underlying portfolio and the riskiness. As a typical investor, certainly
as a pension fund investor, you would not be given access. And even if you
were given access, you don't have the staff that
can actually analyze this. STUDENT: So if Moody's
or Standard and Poor's made a mistake, then every
pension fund or other investors will made the same mistake. ANDREW LO: The mistakes
can carry over. That's right. There is no way for
pension funds, endowments and foundations, or
retail investors, to do their own kind
of due diligence. They're relying on those
managers of the pension funds to do that. And by the way, it wasn't just
pension funds that did this. There was some mutual
funds that did this too. And not only mutual funds,
but there were also some money market funds that did this. Now, money market funds you
say, why would money market funds get invested in this? Well remember,
money market funds are supposed to be putting money
into short term, fixed income instruments. Well, these could be
short term, and these are fixed income instruments. And if you add some
insurance on top of that, they're even safer,
on paper anyway, than many of the other
traditional instruments. So you can see now how a
wonderful idea, and this really is wonderful because
it dramatically increased the risk bearing
capacity of the economy. And by the way, it made a
lot of people better off. So right now, we're in the
midst of a financial crisis and we're focusing
on the negatives. But let's not be
too quick to forget that these kinds of
securitization processes brought in huge amounts of
money that ultimately went to homeowners to be able to
buy homes that they otherwise couldn't have afforded, and
maybe you would argue they shouldn't have afforded,
but there are still many, many homeowners out there
that have subprime loans that are paying their
mortgage payments, that are perfectly happy
living in their atoms, and otherwise couldn't have
afforded it without it. And Moreover, there
are a whole host of individuals that made tons of
money because of the real state boom and because they
were able to leverage using these kinds of funds. STUDENT: In this example, the
ones who [INAUDIBLE] are us. Is that like Lehman Bothers? Are those the companies-- ANDREW LO: That's right. So an example would be some
of the investment banks, as well as some of
the commercial banks. Now, in a minute, I'm
going to tell you what went wrong with all of this. So far, the story is great. This is really an innovation
in financial engineering, because by securitization,
by repackaging, we've done nothing dishonest. We've told people
exactly what we're doing. We've given them transparency. And we've given the safer
asset to the community that wants safer assets. And we've given the
very exciting assets to those who want
the exciting assets. Now, where does this go wrong? Question before we do that. STUDENT: The assumption
that you made in this is that they are not correlated? Isn't there more likeliness of
correlation between the two? ANDREW LO: So that
there's always somebody that's ready to spoil
the party for the rest of us. You're absolutely right. That's where the story
gets interesting. I've assumed that these
two bonds are uncorrelated. What if that
assumption is wrong? In fact, what happens if not
only are they uncorrelated, but what happens if the bonds
are perfectly correlated? Let's work that out. That's a numerical example
that's not hard to do. If the bonds are
perfectly correlated, that means they default
at the same time and they pay off at the
same time, then instead of three outcomes, we
only have two outcomes. Either we get paid
$2,000 in the portfolio, or we get paid nothing
in the portfolio. Now, what happens to the
senior and junior tranches? Well, now, the senior
tranche, the tranche that was AAA, the tranche
that had less than 1% probability of
default, the tranche that was supposed to be so
safe that all sorts of very conservative institutions
could take it on, that tranche has now increased in
riskiness by a factor of 10. The probability of default
has gone from 1% to 10%. And the junior
tranche, the tranche that was supposed
to be toxic waste, and that had a 90%
default rate, now it looks incredibly good,
because it's gone up in terms of its quality. It's gone down in terms
of its default probability from 19% to 10%. So if you look at
the pricing, now the pricing of these two
things, of course, is $900. If they're perfectly
correlated, then securitization does nothing. All you've done is to
take two pieces of paper and slice them up into two
identical pieces of paper. That's what happens if
they're perfectly correlated. So now, why would they
become perfectly correlated? Well, this has to do with what
happened in the housing market. When the housing
market turned down, as it did shortly
after June of 2006, that created a huge dislocation
in these credit markets, because what was
uncorrelated all of a sudden became highly correlated. It's as if an
insurance company that was insuring property and
casualty across the country, all of a sudden experienced
earthquakes in every one of the 50 states all at once. An insurance company
cannot withstand that kind of an event, unless of
course it's prepared for it. And earthquake
insurers prepare for it by insuring not just earthquakes
but hurricanes, fires, and other natural
disasters, which rarely come all at the same
time and all in the same place. We weren't prepared for this. The people that sold these
securities, that held them, weren't prepared. In fact, I skipped over a
quote at the very beginning of this section. This is a quote that appeared
in The Economist magazine, anonymously, and it
was the Chief Risk Officer of a major
financial institution. And the risk manager wrote in
the first part of the article, "Like most banks,
we owned a portfolio of different tranches
of collateralized debt obligations" that's what the
securitized set of obligations are called, "which are packages
of asset-backed securities. Our business and
risk strategy was to buy pools of
assets, mainly bonds, warehouse them on our own
balance sheet" meaning put them in a portfolio in our company,
"and structure them into CDOs and finally distribute
them to end investors." Issue the pieces of paper
to the different investors. "We were most eager to sell the
non-investment grade tranches," the toxic waste, "and
our risk approvals were conditional on
reducing these to zero." So they were very,
very careful to get rid of all that toxic waste. "We will allow positions
however of the top rated AAA and super-M senior
(even better than AAA) tranches to be held on
our own balance-sheets as the default was deemed
to be well protected by all the lower
tranches, which would have to absorb any prior losses." "in May of 2005 we
held AAA tranches, expecting them to rise in value,
and sold non-investment grade tranches, expecting
them to go down." They were long the
seniors, short the juniors. That's a strategy. "From a risk-management
point of view, this was perfect: have a long
position in the low risk asset, and have a short one in
the higher-risk asset. But the reverse happened
of what we had expected: AAA tranches went down in
price and non-investment grade tranches went up,
resulting in losses as we mark the positions to market." And then the risk manager,
this Chief Risk Officer of a major financial
institution, said the following. "This was entirely
counter-intuitive. Explanations of why
this had happened were confusing and focused on
complicated cross-correlations between tranches. In essence it turned
out that there had been a short squeeze in
non-investment grade tranches, driving up prices
and generally selling of all the more senior
[INAUDIBLE] even the very best ones." He still doesn't get it. The numerical example
that I just showed you explains what happened. What happened is
the correlations, that were assumed to be zero,
turned out not to be zero. And when things change, when
the correlations change, that changes the risk. And when you change the risk,
it changes the valuation because the markets
are not stupid. People realize, wow, I assumed
they were uncorrelated, but now these things
are very correlated. I better recalculate my model
and see what it tells me. And it tells me that AAA
is not AAA any longer. And it tells me that the
BA is actually now BAA. So what he experienced is
what every major financial institution dealing
with this stuff experienced over the
last couple of years. It's that kind of
a double whammy, because of the
default rates changing in a way that was never expected
given the historical behavior of the US housing market. Yeah, Beta. STUDENT: [INAUDIBLE] ANDREW LO: It was typically
through the banks. So the banks actually arrange
with the insurance companies to insure those assets. And then they would sell
it to the end investor. The end investor wasn't
the one engaging in these. Although, under
certain circumstances, certain pension
funds were so risk averse that they
ultimately ended up buying extra insurance in the
form of credit default swaps on these kinds of contracts
with other counter-parties. So in many cases, some of
the insurance companies actually did have relationships
with the end investors, as well as with the
investment banks. Yeah, Maria. STUDENT: [INAUDIBLE] ANDREW LO: They do. They do downgrade these. Absolutely. And in fact, not only do
they downgrade the bonds, but they also downgrade the
equity of the companies that are issuing the bonds. So for example,
AIG was downgraded because there was
concern of whether or not it could meet its obligations. And because of that
downgrade, that triggered a bunch of
other transactions. STUDENT: And the other question
is, are they independent, and are they really
objective when they are-- ANDREW LO: Well, so those
are two separate questions. Are they independent
and are they objective? Yes, they are independent,
strictly speaking, in the sense that S&P and Moody's
and Fitch are not owned by any of the companies
that are being rated, number one. Are they objective? That's a different
question because remember that S&P, Moody's and
Fitch are businesses, and businesses generally
try to make money. And in order to make money,
you have to get revenues. And in order to
get revenues, you have to have lots of customers. And so the question
is, did they ultimately end up giving ratings
out too easily that they shouldn't
have because they wanted to get more business? I don't know the answer
to that, but there is going to be a lot of
people, particularly lawyers, asking those questions
in the coming months. So the rating agencies
have definitely been under fire by a number
of different organizations. I don't know where
that's going to come out and I don't know the details
of how they actually conducted the ratings, but there
is definitely an issue because what is AAA should
not default more than 1% or 2% over the life of
the particular loan. And clearly, with
these securities, they've defaulted at
a much higher rate. STUDENT: [INAUDIBLE] ANDREW LO: Yes. That's right. They only relied
on three companies. And actually, it's
very difficult to start a rating
agency now, because the regulatory
authorities require certain kinds of
standards to be met that are virtually impossible
for a startup to be able to do. So you're right, that
investors relied on this, and they ultimately
were badly misled. But the argument that S&P,
Moody's and Fitch would make is that, we were doing
the best we could, we looked at historical
default rates of mortgages, and we made very
conservative assumptions. In fact, if you assume that
they were zero correlation, but instead you tried
to be conservative and you said OK, the
correlation maybe is not zero, but let's make it,
oh I don't know, 25%. Even though historically,
the correlation is maybe much, much
less than that. If you just used an artificial
number like 25% or 30%, you would still not have
had the kind of dislocation that we saw over the
last couple of years, because the correlation
actually has gone much, much higher than
that, particularly for the subprime
mortgages, as you know, because the housing
market's turned down. And a lot of this was
triggered by this decline in housing, which has been
a very, very sharp decline. And over the last 30
years, the housing market in the United States has really
never gone down by more than 1% or 2% in a year. Never mind going down 10%
over the last 12 months. That's a really big
shock to the system. Yeah? STUDENT: [INAUDIBLE] ANDREW LO: Well, in fact they
have done that in the sense that they've actually chopped
up these kinds of security into five different tranches. And they've done it,
they've spread it out very, very broadly. That's how the US
housing market was able to grow as quickly as it
has over the last 10 years. It's because they brought
in huge amounts of money, unprecedented amounts of
money, through this mechanism. And all of the
investors invested based upon these
ratings, as well as their sense of how secure
these markets were. And in each case, there is
going to be dislocation, other than perhaps
in the middle tranche where it hits exactly
right and you don't get any kind of dislocation. But that's not the
biggest tranche. The biggest tranche was by
far the most senior one, because that's the one that has
the largest amounts of money waiting to be invested. STUDENT: [INAUDIBLE] ANDREW LO: Oh yes, absolutely. Yes, absolutely. Hedge fund managers have
profited greatly from this, because they bought the toxic
waste that nobody else wanted and then the value has
gone up dramatically because of these kind of
increases in default rates. Because they were priced to be
much worse than they ultimately ended up being. So it's absolutely the
case that the money has not disappeared into thin air. It's gone from the
senior to the junior. It's a wealth transfer in a way. Sorry, Ken. STUDENT: Just a comment
on why Moody's maybe rated these things the way they did. At least in my experience,
what would happen was the structuring
teams would meet with people who were going
to rate these securities and explain to them,
hey, this is what we did, this is why it makes sense,
and essentially convince them. True earlier why the ANDREW LO: Oh sure. There's a lot of
research that goes on. In other words, Moody's,
S&P and Fitch doesn't just decide based upon how they
feel that day whether is should get AAA or not. They do a fair
amount of research, and they go through the
details of the portfolio, they look at the
seniority of the claims, they look at the
legal documents, they look at the
historical record, they go back and go
back 30, 40, 50 years and take a look at the data. So that's why I
said, they actually have a case for making
the ratings as they did. How they could have
gone so far wrong is a question that
we're going to debate for the next several years. And ultimately, I
think we're going to learn that we need to make
our models more sophisticated. We need to have parameters
that are time varying. We need to have a
different approach to how we do
quantitative analysis for these kinds of markets. But that is an open
question that I think will have to be
examined in much more depth. STUDENT: [INAUDIBLE] ANDREW LO: Oh, in the current
situation can somebody make money? Absolutely. Absolutely. This is why I was saying at
the very start of this crisis that times of crisis are
also times of opportunity. You can absolutely make money,
because these securities now are priced all over the place. Some way to high,
some way to low. And if you understand
these models better than the next
person, you will make money. One of the largest payouts
that has occurred in hedge fund history occurred last year
to a hedge fund manager in New York named John Paulson. I think he was paid-- it's in
the Wall Street Journal so you can look it up-- I think he was paid something
like $3 or $4 billion. Was it $4 billion? That was his take
home pay last year. That was on his W-2. That was not wealth,
that was income. And he did it by betting
on certain movements in these markets, including
these kinds of securities. So there's a lot of
money to be made. There's a lot of
opportunity out there. But it requires an edge. So you really have to
spend some time trying to understand these securities. And what we've done today, this
relatively simple analysis, is an analysis that apparently
eluded this Chief Risk Officer. Because they're focusing at
a very, very detailed level on models that are
probably not as relevant for the
macroscopic picture. Yeah? STUDENT: If no one
is able to determine the right price
for these things, how is the government going
to use the $700 billion to buy these things? ANDREW LO: Well, that's
a great question. That's one of the reasons
why there's so much debate. It's because the view is that
if the very best minds on Wall Street couldn't get
this right, what makes us think that the
Treasury can get it right, which is a little scary. I agree. There are a couple
of things that are being done to address that. One is that as part
of the proposal, they plan to set up an
advisory board of people that are in the industry, seasoned
veterans that are engaged in these transactions,
to help the government price these things. That's one. The second approach
is that they plan to engage in equity ownership
as a possible outcome for this. So in other words,
it's a bailout if you buy for $100
what is really worth 60. But it's not nearly
as much of a bailout if you buy for 50 what's 60. And it could actually be
quite profitable if not only do buy for 50 what's
worth 60, but you also get to own 80% of the
company in the process, which is kind of like the deal
that AIG has struck, and not that different from
some of the discussions that Warren Buffett has
had with Goldman Sachs. So the idea behind
the current proposal is that there will be
additional protections to allow the
government to benefit from the upside of
these securities, and to be able to get the
expertise needed to price them. And there are other protections
that would require industry to pay up for additional
insurance on these portfolios, and ultimately to
allow legislation to recoup some of the losses,
if at the end of 5 or 10 years the government ends
up losing money on these kind of transactions. Let me stop here, and
we'll see you on Wednesday. We'll talk about common stock.
