Professor Robert Shiller:
Today I want to spend--The
title of today's lecture is:
The Universal Principle of Risk
Management,
Pooling and the Hedging of Risk.
What I'm really referring to is
what I think is the very
original, the deep concept that
underlies theoretical finance--I
wanted to get that first.
It really is probability theory
and the idea of spreading risk
through risk pooling.
So, this idea is an
intellectual construct that
appeared at a certain point in
history and it has had an
amazing number of applications
and finance is one of these.
Some of you--This incidentally
will be a more technical of my
lectures and it's a little bit
unfortunate that it comes early
in the semester.
For those of you who have had a
course in probability and
statistics, there will be
nothing new here.
Well, nothing in terms of the
math.
The probability theory is new.
Others though,
I want to tell you that it
doesn't--if you're shopping--I
had a student come by yesterday
and ask--he's a little rusty in
his math skills--if he should
take this course.
I said, "Well if you can
understand tomorrow's
lecture--that's today's
lecture--then you should have no
problem."
I want to start with the
concept of probability.
Do you know what a probability
is?
We attach a probability to an
event.
What is the probability that
the stock market will go up this
year?
I would say--my personal
probability is .45.
That's because I'm a bear
but--Do you know what that
means?
That 45 times out of 100 the
stock market will go up and the
other 55 times out of 100 it
will stay the same or go down.
That's a probability.
Now, you're familiar with that
concept, right?
If someone says the probability
is .55 or .45,
well you know what that means.
I want to emphasize that it
hasn't always been that way and
that probability is really a
concept that arose in the 1600s.
Before that,
nobody ever said that.
Ian Hacking,
who wrote a history of
probability theory,
searched through world
literature for any reference to
a probability and could find
none anywhere before 1600.
There was an intellectual leap
that occurred in the seventeenth
century and it became very
fashionable to talk in terms of
probabilities.
It spread throughout the
world--the idea of quoting
probabilities.
But it was--It's funny that
such a simple idea hadn't been
used before.
Hacking points out that the
word probability--or
probable--was already in the
English language.
In fact, Shakespeare used it,
but what do you think it meant?
He gives an example of a young
woman, who was describing a man
that she liked,
and she said,
I like him very much,
I find him very probable.
What do you think she means?
Can someone answer that?
Does anyone know Elizabethan
English well enough to tell me?
What is a probable young man?
I'm asking for an answer.
It sounds like people have no
idea.
Can anyone venture a guess?
No one wants to venture a guess?
Student: fertile?
Professor Robert
Shiller: That he can father
children?
I don't think that's what she
meant but maybe.
No, what apparently she meant
is trustworthy.
That's a very important quality
in a person I suppose.
So, if something is probable
you mean that you can trust it
and so probability means
trustworthiness.
You can see how they moved from
that definition of probability
to the current definition.
But Ian Hacking,
being a good historian,
thought that someone must have
had some concept of probability
going before,
even if they didn't quote it as
a number the way--it must have
been in their head or in their
idea.
He searched through world
literature to try to find some
use of the term that preceded
the 1600s and he concluded that
there were probably a number of
people who had the idea,
but they didn't publish it,
and it never became part of the
established literature partly
because,
he said, throughout human
history, there has been a love
of gambling and probability
theory is extremely useful if
you are a gambler.
Hacking believes that there
were many gambling theorists who
invented probability theory at
various times in history but
never wrote it down and kept it
as a secret.
He gives an example--I like
to--he gives an example from a
book that--or it's a
collection--I think,
a collection of epic poems
written in Sanskrit that goes
back--it was actually written
over a course of 1,000 years and
it was completed in the fourth
century.
Well, there's a story--there's
a long story in the Mahabarahta
about an emperor called Nala and
he had a wife named Damayanti
and he was a very pure and very
good person.
There was an evil demon called
Kali who hated Nala and wanted
to bring his downfall,
so he had to find a weakness of
Nala.
He found finally some,
even though Nala was so pure
and so perfect--he found one
weakness and that was gambling.
Nala couldn't resist the
opportunity to gamble;
so the evil demon seduced him
into gambling aggressively.
You know sometimes when you're
losing and you redouble and you
keep hoping to win back what
you've lost?
In a fit of gambling,
Nala finally gambled his entire
kingdom and lost--it's a
terrible story--and Nala then
had to leave the kingdom and his
wife.
They wandered for years.
He separated from her because
of dire necessity.
They were wandering in the
forests and he was in despair,
having lost everything.
But then he meets someone by
the name of--we have Nala and he
meets this man,
Rituparna, and this is where a
probability theory apparently
comes in.
Rituparna tells Nala that he
knows the science of gambling
and he will teach it to Nala,
but that it has to be done by
whispering it in his ear because
it's a deep and extreme secret.
Nala is skeptical.
How does Rituparna know how to
gamble?
So Rituparna tries to prove to
him his abilities and he says,
see that tree there,
I can estimate how many leaves
there are on that tree by
counting leaves on one branch.
Rituparna looked at one branch
and estimated the number of
leaves on the tree,
but Nala was skeptical.
He stayed up all night and
counted every leaf on the tree
and it came very close to what
Rituparna said;
so he--the next
morning--believed Rituparna.
Now this is interesting,
Hacking says,
because it shows that sampling
theory was part of Nala's
theory.
You don't have to count all the
leaves on the tree,
you can take a sample and you
count that and then you
multiply.
Anyway, the story ends and Nala
goes back and is now armed with
probability theory,
we assume.
He goes back and gambles again,
but he has nothing left to
wager except his wife;
so he puts her and gambles her.
But remember,
now he knows what he's doing
and so he really wasn't gambling
his wife--he was really a very
pure and honorable man.
So he won back the entire
kingdom and that's the ending.
Anyway, that shows that I think
probability theory does have a
long history,
but--it not being an
intellectual discipline--it
didn't really inform a
generation of finance theory.
When you don't have a theory,
then you don't have a way to be
rigorous.
So, it was in the 1600s that
probability theory started to
get written down as a theory and
many things then happened in
that century that,
I think, are precursors both to
finance and insurance.
One was in the 1600s when
people started constructing life
tables.
What is a life table?
It's a table showing the
probability of dying at each
age, for each age and sex.
That's what you need to know if
you're going to do life
insurance.
So, they started to do
collecting of data on mortality
and they developed something
called actuarial science,
which is estimating the
probability of people living.
That then became the basis for
insurance.
Actually, insurance goes back
to ancient Rome in some form.
In ancient Rome they had
something called burial
insurance.
You could buy a policy that
protected you against your
family not having the money to
bury you if you died.
In ancient culture people
worried a great deal about being
properly buried,
so that's an interesting
concept.
They were selling that in
ancient Rome;
but you might think,
but why just for burial?
Why don't you make it into
full-blown life insurance?
You kind of wonder why they
didn't.
I think maybe it's because they
didn't have the concepts down.
In Renaissance Italy they
started writing insurance
policies--I read one of the
insurance policies,
it's in the Journal of Risk and
Insurance--and they translate a
Renaissance insurance policy and
it's very hard to understand
what this policy was saying.
I guess they didn't have our
language, they didn't--they were
intuitively halfway there but
they couldn't express it,
so I think the industry didn't
get really started.
I think it was the invention of
probability theory that really
started it and that's why I
think theory is very important
in finance.
Some people date fire insurance
with the fire of London in 1666.
The whole city burned down,
practically,
in a terrible fire and fire
insurance started to proliferate
right after that in London.
But you know,
you kind of wonder if that's a
good example for fire insurance
because if the whole city burns
down,
then insurance companies would
go bankrupt anyway,
right?
London insurance companies
would because the whole concept
of insurance is pooling of
independent probabilities.
Nonetheless,
that was the beginning.
We're also going to recognize,
however, that insurance got a
slow start because--I believe it
is because--people could not
understand the concept of
probability.
They didn't have the concept
firmly in mind.
There are lots of aspects to it.
In order to understand
probability, you have to take
things as coming from a random
event and people don't clearly
have that in their mind from an
intuitive standpoint.
They have maybe a sense that I
can influence events by willing
or wishing and if I think
that--if I have kind of a
mystical side to me,
then probabilities don't have a
clear meaning.
It has been shown that even
today people seem to think that.
They don't really take,
at an intuitive level,
probabilities as objective.
For example,
if you ask people how much they
would be willing to bet on a
coin toss,
they will typically bet more if
they can toss the coin or they
will bet more if the coin hasn't
been tossed yet.
It could have been already
tossed and concealed.
Why would that be?
It might be that there's just
some intuitive sense that I
can--I don't know--I have some
magical forces in me and I can
change things.
The idea of probability theory
is that no, you can't change
things, there are all these
objective laws of probability
out there that guide everything.
Most languages around the world
have a different word for luck
and risk--or luck and fortune.
Luck seems to mean something
about you: like I'm a lucky
person.
I don't know what that
means--like God or the gods
favor me and so I'm lucky or
this is my lucky day.
Probability theory is really a
movement away from that.
We then have a mathematically
rigorous discipline.
Now, I'm going to go through
some of the terms of probability
and--this will be review for
many of you,
but it will be something that
we're going to use in the--So
I'll use the symbol P or
I can sometimes write it out as
prob to represent a
probability.
It is always a number that lies
between zero and one,
or between 0% and 100%.
"Percent" means divided by 100
in Latin, so 100% is one.
If the probability is zero that
means the event can't happen.
If the probability is one,
it means that it's certain to
happen.
If the probability is--Can
everyone see this from over
there?
I can probably move this or
can't I?
Yes, I can.
Now, can you now--you're the
most disadvantaged person and
you can see it,
right?
So that's the basic idea.
One of the first principles of
probability is the idea of
independence.
The idea is that probability
measures the likelihood of some
outcome.
Let's say the outcome of an
experiment, like tossing a coin.
You might say the probability
that you toss a coin and it
comes up heads is a half,
because it's equally likely to
be heads and tails.
Independent experiments are
experiments that occur without
relation to each other.
If you toss a coin twice and
the first experiment doesn't
influence the second,
we say they're independent and
there's no relation between the
two.
One of the first principles of
probability theory is called the
multiplication rule.
That says that if you have
independent probabilities,
then the probability of two
events is equal to the product
of their probabilities.
So, the Prob(A and
B) = Prob(A)*Prob(B).
That wouldn't hold if
they're not independent.
The theory of insurance is that
ideally an insurance company
wants to insure independent
events.
Ideally, life insurance is
insuring people--or fire
insurance is insuring
people--against independent
events;
so it's not the fire of London.
It's the problem that sometimes
people knock over an oil lamp in
their home and they burn their
own house down.
It's not going to burn any
other houses down since it's
just completely independent of
anything else.
So, the probability that the
whole city burns down is
infinitesimally small,
right?
This will generalize to
probability of A and
B and C equals the
probability of A times
the probability of B
times the probability of
C and so on.
If the probability is 1 in
1,000 that a house burns down
and there are 1,000 houses,
then the probability that they
all burn down is 1/1000 to the
1000th power,
which is virtually zero.
So insurance companies
then–Basically,
if they write a lot of
policies, then they have
virtually no risk.
That is the fundamental idea
that may seem simple and obvious
to you, but it certainly wasn't
back when the idea first came
up.
Incidentally,
we have a problem set,
which I want you to start today
and it will be due not in a week
this time,
because we have Martin Luther
King Day coming up,
but it will be due the Monday
following that.
If you follow through from the
independent theory,
there's one of the basic
relations in probability
theory--it's called the binomial
distribution.
I'm not going to spend a whole
lot of time on this but it gives
the probability of x
successes in n trials or,
in the case of insurance
x, if you're insuring
against an accident,
then the probability that
you'll get x accidents
and n trials.
The binomial distribution gives
the probability as a function of
x and it's given by the
formula where P is the
probability of the accident:
P^(X) (1-P)^(N-X) [n!/(n-x)!].
That is the formula that
insurance companies use when
they have independent
probabilities,
to estimate the likelihood of
having a certain number of
accidents.
They're concerned with having
too many accidents,
which might exhaust their
reserves.
An insurance company has
reserves and it has enough
reserves to cover them for a
certain number of accidents.
It uses the binomial
distribution to calculate the
probability of getting any
specific number of accidents.
So, that is the binomial
distribution.
I'm not going to expand on this
because I can't get into--This
is not a course in probability
theory but I'm hopeful that you
can see the formula and you can
apply it.
Any questions?
Is this clear enough?
Can you read my handwriting?
Another important concept in
probability theory that we will
use a lot is expected value,
the mean, or average--those are
all roughly interchangeable
concepts.
We have expected value,
mean or average.
We can define it in a couple of
different ways depending on
whether we're talking about
sample mean or population mean.
The basic definition--the
expected value of some random
variable x--E(x)--I guess
I should have said that a random
variable is a quantity that
takes on value.
If you have an experiment and
the outcome of the experiment is
a number, then a random variable
is the number that comes from
the experiment.
For example,
the experiment could be tossing
a coin;
I will call the outcome
heads the number one,
and I'll call the outcome
tails the number zero,
so I've just defined a random
variable.
You have discrete random
variables, like the one I just
defined, or there are
also--which take on only a
finite number of values--and we
have continuous random variables
that can take on any number of
values along a continuum.
Another experiment would be to
mix two chemicals together and
put a thermometer in and measure
the temperature.
That's another invention of the
1600s, by the way--the
thermometer.
And they learned that
concept--perfectly natural to
us--temperature.
But it was a new idea in the
1600s.
So anyway, that's continuous,
right?
When you mix two chemicals
together, it could be any
number, there's an infinite
number of possible numbers and
that would be continuous.
For discrete random variables,
we can define the expected
value, or µ_x
--that's the Greek letter
mu--as the summation i =
1 to infinity of.
[P(x=x_i) times
(x_i)].
I have it down that there might
be an infinite number of
possible values for the random
variable x.
In the case of the coin toss,
there are only two,
but I'm saying in general there
could be an infinite number.
But they're accountable and we
can list all possible values
when they're discrete and form a
probability weighted average of
the outcomes.
That's called the expected
value.
People also call that the mean
or the average.
But, note that this is based on
theory.
These are probabilities.
In order to compute using this
formula you have to know the
true probabilities.
There's another formula that
applies for a continuous random
variables and it's the same idea
except that--I'll also call it
µ_x,
except that it's an integral.
We have the integral from minus
infinity to plus infinity of
F(x)*x*dx,
and that's really--you see it's
the same thing because an
integral is analogous to a
summation.
Those are the two population
definitions.
F(x) is the continuous
probability distribution for
x.
That's different when you have
continuous values--you don't
have P (x =
x_i) because it's
always zero.
The probability that the
temperature is exactly 100°
is zero because it could be
100.0001°
or something else and there's
an infinite number of
possibilities.
We have instead what's called a
probability density when we have
continuous random variables.
You're not going to need to
know a lot about this for this
course, but this is--I wanted to
get the basic ideas down.
These are called population
measures because they refer to
the whole population of possible
outcomes and they measure the
probabilities.
It's the truth,
but there are also sample
means.
When you get--this is
Rituparna, counting the leaves
on a tree--you can estimate,
from a sample,
the population expected values.
The population mean is often
written "x-bar."
If you have a sample with
n observations,
it's the summation i = 1
to n of
x_i/n--that's
the average.
You know that formula, right?
You count n leaves--you
count the number of leaves.
You have n branches on
the tree and you count the
number of leaves and sum them
up.
One would be--I'm having a
little trouble putting this into
the Rituparna story,
but you see the idea.
You know the average, I assume.
That's the most elementary
concept and you could use it to
estimate either a discreet or
continuous expected value.
In finance, there's often
reference to another kind of
average, which I want to refer
you to and which,
in the Jeremy Siegel book,
a lot is made of this.
The other kind of average is
called the geometric average.
We'll call that--I'll only show
the sample version of it G(x) =
the product i = 1 to
n of (x_i
)^(1/n).
Does everyone--Can you see that?
Instead of summing them and
dividing by M,
I multiply them all together
and take the n^(th) root of
them.
This is called the geometric
average and it's used only for
positive numbers.
So, if you have any negative
numbers you'd have a problem,
right?
If you had one negative number
in it, then the product would be
a negative number and,
if you took a root of that,
then you might get an imaginary
number.
We don't want to use it in that
case.
There's an appendix to one of
the chapters in Jeremy Siegel's
book where he says that one of
the most important applications
of this theory is to measure how
successful an investor is.
Suppose someone is managing
money.
Have they done well?
If so, you would say,
"Well, they've been investing
money over a number of different
years.
Let's take the average over all
the different years."
Suppose someone has been
investing money for n
years and x_i
is the return on the investment
in a given year.
What is their average
performance?
The natural thing to do would
be to average them up,
right?
But Jeremy says that maybe
that's not a very good thing to
do.
What he says you should do
instead is to take the geometric
average of gross returns.
The return on an investment is
how much you made from the
investment as a percent of the
money invested.
The gross return is the return
plus one.
The worst you can ever do
investing is lose all of your
investment--lose 100%.
If we add one to the return,
then you've got a number that's
never negative and we can then
use geometric returns.
Jeremy Siegel says that in
finance we should be using
geometric and not arithmetic
averages.
Why is that?
Well I'll tell you in very
simple terms,
I think.
Suppose someone is investing
your money and he announces,
I have had very good returns.
I have invested and I've
produced 20% a year for nine out
of the last ten years.
You think that's great,
but what about the last year.
The guy says,
"Oh I lost 100% in that year."
You might say,
"Alright, that's good."
I would add up 20% a year for
nine years and than put in a
zero–no,
120 because it's gross return
for nine years--and put in a
zero for one year.
Maybe that doesn't look bad,
right?
But think about it,
if you were investing your
money with someone like that,
what did you end up with?
You ended up with nothing.
If they have one year when they
lose everything,
it doesn't matter how much they
made in the other years.
Jeremy says in the text that
the geometric return is always
lower than the arithmetic return
unless all the numbers are the
same.
It's a less optimistic version.
So, we should use that,
but people in finance resist
using that because it's a lower
number and when you're
advertising your return you want
to make it look as big as
possible.
We also need some measure
of--We've been talking here
about measures of central
tendency only and in finance we
need,
as well, measures of
dispersion, which is how much
something varies.
Central tendency is a measure
of the center of a probability
distribution of the--Central
tendency is a measure--Variance
is a measure of how much things
change from one observation to
another.
We have variance and it's often
represented by σ²,
that's the Greek letter sigma,
lower case, squared.
Or, especially when talking
about estimates of the variance,
we sometimes say S²
or we say standard
deviation².
The standard deviation is the
square root of the variance.
For population variance,
the variance of some random
variable x is defined as
the summation i = 1 to infinity
of the Prob (x =
x_i) times
(x_i -
µ_x)^(2).
So mu is the mean--we just
defined it of x--that's
the expectation of x or
also E(x),
so it's the probability
weighted average of the squared
deviations from the mean.
If it moves a lot--either way
from the mean--then this number
squared is a big number.
The more x moves,
the bigger the variance is.
There's also another variance
measure, which we use in the
sample--or also Var is used
sometimes--and this is
∑².
There's also another variance
measure, which is for the
sample.
When we have n
observations it's just the
summation i = 1 to n of
(x - x bar)²/n.
That is the sample variance.
Some people will divide by
n–1.
I suppose I would accept either
answer.
I'm just keeping it simple here.
They divide by n-1 to
make it an unbiased estimator of
the population variance;
but I'm just going to show it
in a simple way here.
So you see what it is--it's a
measure of how much x
deviates from the mean;
but it's squared.
It weights big deviations a lot
because the square of a big
number is really big.
So, that's the variance.
So, that completes central
tendency and dispersion.
We're going to be talking about
these in finance in regards to
returns because--generally the
idea here is that we want high
returns.
We want a high expected value
of returns, but we don't like
variance.
Expected value is good and
variance is bad because that's
risk;
that's uncertainty.
That's what this whole theory
is about: how to get a lot of
expected return without getting
a lot of risk.
Another concept that's very
basic here is covariance.
Covariance is a measure of how
much two variables move
together.
Covariance is--we'll call
it--now we have two random
variables, so I'll just talk
about it in a sample term.
It's the summation i = 1
to n of [(x –
x-bar) times (y –
y-bar)]/n.
So x is the deviation
for the i-subscript,
meaning we have a separate
x_i and
y_i for each
observation.
So we're talking about an
experiment when you
generate--Each experiment
generates both an x and a
y observation and we know
when x is high,
y also tends to be high,
or whether it's the other way
around.
If they tend to move together,
when x is high and
y is high together at the
same time,
then the covariance will tend
to be a positive number.
If when x is low,
y also tends to be low,
then this will be negative
number and so will this,
so their product is positive.
A positive covariance means
that the two move together.
A negative covariance means
that they tend to move opposite
each other.
If x is high relative to
x-bar--this is
positive--then y tends to
be low relative to its mean
y-bar and this is
negative.
So the product would be
negative.
If you get a lot of negative
products, that makes the
covariance negative.
Then I want to move to
correlation.
So this is a measure--it's a
scaled covariance.
We tend to use the Greek letter
rho.
If you were to use Excel,
it would be correl or
sometimes I say corr.
That's the correlation.
This number always lies between
-1 and +1.
It is defined as rho=
[cov(x_iy
_i)/S_x
S_y]
That's the correlation
coefficient.
That has kind of almost entered
the English language in the
sense that you'll see it quoted
occasionally in newspapers.
I don't know how much you're
used to it--Where would you see
that?
They would say there is a low
correlation between SAT scores
and grade point averages in
college, or maybe it's a high
correlation.
Does anyone know what it is?
But you could estimate the
corr--it's probably positive.
I bet it's way below one,
but it has some correlation,
so maybe it's .3.
That would mean that people who
have high SAT scores tend to get
higher grades.
If it were negative--it's very
unlikely that it's negative--it
couldn't be negative.
It couldn't be that people who
have high SAT scores tend to do
poorly in college.
If you quantify how much they
relate, then you could look at
the correlation.
I want to move to regression.
This is another concept that is
very basic to statistics,
but it has particular use in
finance, so I'll give you a
financial example.
The concept of regression goes
back to the mathematician Gauss,
who talked about fitting a line
through a scatter of points.
Let's draw a line through a
scatter of points here.
I want to put down on this axis
the return on the stock market
and on this axis I want to put
the return on one company,
let's say Microsoft.
I'm going to have each
observation as a year.
I shouldn't put down a name of
a company because I can't
reproduce this diagram for
Microsoft.
Let's not say Microsoft,
let's say Shiller,
Inc.
There's no such company,
so I can be completely
hypothetical.
Let's put zero here because
these are not gross returns
these are returns,
so they're often negative.
Suppose that in a given
year--and say this is minus five
and this is plus five,
this is minus five and this is
plus five--Suppose that in the
first year in our sample,
the company Shiller, Inc.
and the market both did 5%.
That puts a point right there
at five and five.
In another year,
however, the stock market lost
5% and Shiller,
Inc.
lost 7%.
We would have a point,
say, down here at five and
seven.
This could be 1979,
this could be 1980,
and we keep adding points so we
have a whole scatter of points.
It's probably upward sloping,
right?
Probably when the overall stock
market does well so does
Shiller, Inc.
What Gauss did was said,
let's fit a line through the
point--the scatter of
points--and that's called the
regression line.
He chose the line so that--this
is Gauss--he chose the line to
minimize the sum of squared
distances of the points from the
lines.
So these distances are the
lengths of these line segments.
To get the best fitting line,
you find the line that
minimizes the sum of squared
distances.
That's called the regression
line and the intercept is called
alpha--there's
alpha.
And the slope is called
beta.
That may be a familiar enough
concept to you,
but in the context of finance
this is a major concept.
The way I've written it,
the beta of Shiller,
Inc.
is the slope of this line.
The alpha is just the
intercept of this curve.
We can also do this with excess
returns.
I will get to this later,
where I have the return minus
the interest rate on this axis
and the market return minus the
interest rate on this axis.
In that case,
alpha is a measure of
how much Shiller,
Inc.
outperforms.
We'll come back to this,
but beta of the stock is
a measure of how much it moves
with the market and the
alpha of a stock is how
much it outperforms the market.
We'll have to come back to
that--these are basic concepts.
I want to--another concept--I
guess I've just been implicit in
what I have--There's a
distribution called the normal
distribution and that is--I'm
sure you've heard of this,
right?
If you have a distribution that
looks like this--it's
bell-shaped--this is x
and--I have to make it look
symmetric which I may not be
able to do that well--this is
f(x),
the normal distribution.
f(x) = [1/(√
(2π)σ)]
times e to minus
[(x-µ)^(2) / 2σ].
It's a famous formula,
which is due to Gauss again.
We often assume in finance that
random variables,
such as returns,
are normally distributed.
This is called the normal
distribution or the Gaussian
distribution--it's a continuous
distribution.
I think you've heard of this,
right?
This is high school raw
material.
But I want to emphasize that
there are also other bell-shaped
curves.
This is the most famous
bell-shaped curve,
but there are other ones with
different mathematics.
A particular interest in
finance is fat-tailed
alternatives.
It could be that a random
distribution--I don't have
colored chalk here I don't
think,
so I will use a dash line to
represent the fat-tailed
distribution.
Suppose the distribution looks
like this.
Then I have to try to do that
on the other side,
as symmetrically as I can.
These are the tails of the
distribution;
this is the right tail and this
is the left tail.
You can see that the dash
distribution I drew has more out
in the tails,
so we call it fat-tailed.
This refers to random variables
that have fat-tailed
distributions--random variables
that occasionally give you
really big outcomes.
You have a chance of being way
out here with a fat-tailed
distribution.
It's a very important
observation in finance that
returns on a lot of speculative
assets have fat-tailed
distributions.
That means that you can go
through twenty years of a career
on Wall Street and all you've
observed is observations in the
central region.
So you feel that you know
pretty well how things behave;
but then, all of a sudden,
there's something way out here.
This would be good luck if you
were long and now suddenly you
got a huge return that you would
not have thought was possible
since you've never seen it
before.
But you can also have an
incredibly bad return.
This complicates finance
because it means that you never
know.
You never have enough
experience to get through all
these things.
It's a big complication in
finance.
My friend Nassim Talib has just
written a book about it
called--maybe I'll talk about
that--called The Black
Swan.
It's about how so many plans in
finance are messed up by rare
events that suddenly appear out
of nowhere.
He called it The Black
Swan because if you look at
swans, they're always white.
You've never seen a black swan.
So, you end up going through
life assuming that there are no
black swans.
But, in fact,
there are and you might finally
see one.
You don't want to predicate
making complicated gambles under
the assumption that they don't
exist.
Talib, who's a Wall Street
professional,
talks about these black swans
as being the real story of
finance.
Now.
I want to move away from
statistics and talk about
present values,
which is another concept in
finance that is fundamental.
And so, let me--And then this
will conclude today's lecture.
What is a present value?
This isn't really statistics
anymore, but it's a concept that
I want to include in this
lecture.
People in business often have
claims on future money,
not money today.
For example,
I may have someone who promises
to pay me $1 in one year or in
two years or three years.
The present value is what
that's worth today.
I may have an "IOU" from
someone or I may own a bond from
someone that promises to pay me
something in a year or two
years.
According to a time-honored
tradition in finance,
it says that it's a promise to
pay $1, but it's not worth $1
today.
It must be worth less than $1.
What you could do hundreds of
years ago--and can still do it
today--was go to a bank and
present this bond or IOU and
say,
"What will you give me for it?"
The bank will discount it.
Sometimes we say "present
discounted value."
The banker will say,
"Well you have $1 a year from
now, but that's a year from now,
so I won't give you $1 now.
I'll give you the present
discounted value for it."
Now, I'm going to abstract from
risk.
Let's assume that we know that
this thing is going to be paid,
so it's a matter of simple
time.
Of course, the banker isn't
going to give you $1 for
something that is paying $1 in a
year because the banker knows
that $1 could be invested at the
interest rate.
Let's say the interest rate is
r and that would be a
number like .05,
which is 5%,
which is five divided by one
hundred.
Then the present value of
$1--The present PDV or PV of $1
= $1/(1+r).
That's because the banker is
thinking, if I have this amount
right now and I invest it for
one year, then what do I have.
I have (1 +
r)*(1/1+r).
It's $1, so that works out
exactly right.
You have to discount something
that's one period in the future
by dividing it by 1+r.
This is the present value of $1
in one time period,
which I'm taking to be a year.
It doesn't have to be a year.
The interest rate has units of
time, so I have to specify a
time period over which I'm
measuring an interest rate.
Typically it's a year.
If it's a one-year interest
rate, the time period is one
year, and the present value of
$1 in one time period is given
by this: the present value of $1
in n periods is
1/(1+r)^(n) and that's
all there is to this.
I want to talk about valuing
streams of payments.
Suppose someone has a contract
that promises to pay an amount
each period over a number of
years.
We have formulas for these
present values and these
formulas are well known.
I'm just going to go through
them rather quickly here.
The simplest thing is something
called a consol or perpetuity.
A perpetuity is an asset or a
contract that pays a fixed
amount of money each time
period, forever.
We call them consols because,
in the early 1700s,
the British Government issued
what they called consols or
consolidated debt of the British
Crown that paid a certain amount
of pound sterling every six
months forever.
You may say,
what audacity for the British
Government to promise to pay
anything forever.
Will they be around forever?
Well as far as you're
concerned, it's as good as
forever, right?
Maybe someday the
British--United
Kingdom--something will happen
to it, it will fall apart or
change;
but that is so distant in the
future that we can disregard
that, so we'll take that as
forever.
Anyway, the government might
buy them back too,
so who cares if it isn't
forever.
Let's just talk about it as
forever.
Let's say this thing pays one
pound a period forever.
What is the present value of
that?
Well, the first--each payment
we'll call a coupon--so it pays
one pound one year from now.
Let's say it's one year just to
simplify things.
It pays another pound two years
from now, it pays another pound
three years from now.
The present value is equal
to--remember it starts one year
from now under assumption--we
could do it differently but I'm
assuming one year now.
The present value is
1/(1+r) for the first
year;
plus for the second year it's
1/(1+r)²;
for the third year it's
1/(1+r)³;
and that goes on forever.
That's an infinite series and
you know how to sum that,
I think.
I'll tell you what it is:
it's 1/r,
or it would be
£1/r .
Generally, if it pays c
dollars for every period,
the present value is
c/r.
That's the formula for the
present value of a consol.
That's one of the most basic
formulas in finance.
The interesting thing is that
it means that the value of
consol moves inversely to the
interest rate.
The British Government issued
those consols in the early 1700s
and, while they were refinanced
in the late nineteenth century,
they're still there.
If you want to go out and buy
one, you can get on your laptop
right after this lecture and buy
one of them.
Then you've got something that
will pay you something forever.
But you're going to know that
the value of that in the market
moves opposite with interest
rates.
So, if interest rates go down,
the value goes up;
if interest rates go up,
the value of your investment
goes down.
Another formula is--what if the
consol doesn't pay--I'm sorry,
the next thing is a growing
consol.
I'm calling it a growing consol
even though the British consols
didn't grow.
Let's say that the British
Government didn't say that
they'll pay one pound per year,
but it'll be one pound the
first year, then it will grow at
the rate g and it will
eventually be infinitely large.
You get one pound the first
year, you get 1+g pounds
the second year,
etc., (1+g)²
the third year and so on.
The present value of
this--suppose it pays--let's say
it pays c pounds each
year, so it would be c
times this.
It would be c times
(1+g)³
in the third year,
etc., Then the present value is
equal to c/(r-
g)--that's the formula
for the value of a growing
console.
g has to be less than
r for this to make sense
because if g--if it's
growing faster than the rate of
interest,
then this infinite series will
not converge and the value would
be infinite.
You might ask,
"Well then how does that make
sense?"
What if the British Government
promised to pay 10% more each
year, how would the market value
that?
The formula doesn't have a
number.
I'll tell you why it doesn't
have a number:
the British Government will
never promise to pay you 10%
more each year because they
can't do it.
And, the market wouldn't
believe them because you can't
grow every year faster than the
interest rate.
Now that's one of the most
basic lessons,
you can't do it.
One more thing that I think
would be relevant to
the--there's also the annuity
formula.
This is a formula that applies
to--what if an asset pays a
fixed amount every period and
then stops?
That's called an annuity.
An annuity pays c
dollars starting in t =
1,2, 3, and n is the last
period, then it stops.
A good example of an annuity is
a mortgage on a house.
When you buy a house,
you borrow the money and you
pay it back in fixed--it would
usually be monthly,
but let's say annual payments.
You pay every year a fixed
amount on your house to the
mortgage originator and then
after so many--n is 30
years,
typically--you would then have
paid it off.
It used to be that mortgages
had what's called a balloon
payment at the end.
This means that you would have
to pay extra money at the end;
but they decided that people
have trouble doing that.
It's much better to pay a fixed
payment and then you're done.
Otherwise, if you ask them to
pay more at the end,
then a lot of people won't have
the money.
We now have annuity mortgages.
What is the present value of an
annuity?
That is, the present value of
an annuity is equal to the
amount--what did I
say--c*{1 –
[1/(1+r)]^(n)
}/r.
That is the present value of an
annuity.
I wanted to say one more thing
because I realize that you have
to--your first problem set will
cover this--is to talk about the
concept that applies probability
theory to Economics.
That is expected utility theory.
Then I'll conclude with this.
In Economics,
it is assumed that people have
a utility function,
which represents how happy they
are with an outcome--we
typically take that as U,
If I have a monetary outcome,
then I have a certain amount of
money, x dollars.
How happy I am with x
dollars is called U(x).
This, I think you've gotten from
other economics courses--we have
something called diminishing
marginal utility.
The idea is that for any amount
of money--if this x is
the amount of money that I
receive--utility as the function
of the amount of money I receive
is downwardly-concave.
The exact shape of the curve is
subject to discussion,
but the point of diminishing
marginal utility is that,
as you get more and more money,
the increment in utility for
each extra dollar diminishes.
Usually we say it never goes
down, we don't have it going
down, cross that out.
That would be where having more
money makes you less happy.
That may actually work that
way, but our theory says no,
you always want more.
It's always upward sloping,
but it may, after awhile,
you get close to satiation
where you've got enough.
Incidentally,
I mentioned this last time--I
was talking about--I was
philosophizing about wealth and
I asked what are you going to do
with a billion dollars.
We have many billionaires in
this country and I think that
the only thing you have to do
with it is philanthropy.
They have to give it away
because they are essentially
satiated.
Because, as I said,
you can only drive one car at a
time and if you've got ten of
them in the garage,
then it doesn't really do you
much good.
You can't do it;
you can't enjoy all ten of them.
It's important--that's one
reason why we want policies that
encourage equality of
incomes--not necessarily
equality,
but reasonable
equality--because the people
with very low wealth have a very
high marginal utility of income
and people with very high wealth
have very little.
So, if you take from the rich
and give to the poor you make
people happier.
We're not going to do that in a
Robin Hood way;
but in finance we're going to
do that in a systematic way
through risk management.
We're going to be taking away
from lucky--you think of
yourself as randomly on any
point of this.
You don't want--you know that
you'd like to take money away
from yourself in the
high-outcome years and give it
to yourself in the low-income
years.
What finance theory is based
on--and much of economics is
based on--the idea that people
want to maximize the expected
utility of their wealth.
Since this is a concave
function, it's not just the
expected value.
To calculate the expected
utility of your wealth,
you might also have to look at
the expected return,
or the geometric expected
return, or the standard
deviation.
Or you might have to look at
the fat tail.
There are so many different
aspects that we can get into and
this underlying theory motivates
a lot of what we do.
But it's not a complete theory
until we specify the utility
function.
Of course, we will also be
talking about behavioral finance
in this course and we'll,
at times, be saying that the
utility function concept isn't
always right--the idea that
people are actually maximizing
expected utility might not be
entirely accurate.
But, in terms of the basic
theory, that's the core concept.
I have one question on the
problem set that asks you to
think about how you would handle
a decision: whether to gamble,
based on efficient--based on
expected utility theory.
That's a little bit of a tricky
question but--So,
do the best you can on it and
think--try to think about what
this kind of theory would imply
for gambling behavior.
I will see you on Friday.
That's two days from now in
this room.