0:39

As a result a project where,

by the way, when I say project, I mean small, big, anything.

The whole company is like a project.

Project's risk is largely.

1:08

Across assets.

Why? Because everything unique

to the project is gone.

It gotten diversified.

Not because of the manager.

This is rare but say the important thing the manager think like they invest.

The investor is not worried about your project, if they were worried about your

project that would be exactly when all their money's in your project.

That's not the case.

They're worried about your project to the extent that it's related to the market and

look at the beauty of this.

How are all those relationships captured?

1:44

Ri, which in our case was Apple,

alpha plus beta i, market,

plus epsilon i.

I have replaced in the typical regression y on the left hand side by Ri.

2:44

You know it was just waiting for finance to happen.

So quick thing I want to move on now is given all this development,

what does it mean?

First, risk is defined, I told you, defined.

3:06

How easy, how easily measured, just a regression.

I must emphasize, though, that as soon as you round just to the ratio we have to

worry about how months of data, typically 60, did the anomaly you are trying to

measure, in this case apple, has not changed dramatically.

Still looks similar in the past to what your business is, which is orange, right.

So things like that.

And I'm not going to talk too much about that because that falls in the domain of

statistics but you have to do this intelligently.

There are statistical biases you have to worry about and so on.

So, if I want to measure beta properly, you need to know statistics.

And I am skipping that part and moving on to something that is extremely important.

Is how do I know use this risk measure, to figure out what.

I was interested in risk to figure out the return.

Because risk drives return, which then I will use to evaluate orange.

And why Apple's return?

For two reasons.

One again, Apple is comparable and second,

at this point, we have seen that debt is equal to 0.

So Apple's return on equity which is in this regression will also,

the risk of this equation is telling me the risk of the equity of Apple right?

That will also determine the return of equity of Apple.

But because there's no debt RA and RE for

Apple are the same and I'm in business because I am after RA.

Okay.

So let's see how do I develop that relationship?

6:30

The first part is the average market risk premium.

So what you want to figure out is, how much does the market give over and

above the risk free rate?

Remember the table we saw last week?

6:44

In America, the rate of return on average over the last 70, 80 years,

we, in books, tend to use about 7%.

I'm very wary of this, because remember, you're going to use this.

There's a lot of data backing it, 80 years.

But what did I tell you?

One of the biggest questions is why has this been so high?

So this is a little bit questionable, but used very commonly.

Many people do surveys of what it's likely to be in the future.

If you put all data together, this has to be a lower number and

it's a very powerful number.

Because, even if you change it to, say, 5%, what some people suppose,

it has a dramatic impact, because a 2% rate of return is a huge amount.

So you start off with the risk to create.

You add to it the risk premium on the market.

7:34

But the risk premium on the market is not what a project is.

What is the risk of your project?

It's your beta relative to the market.

So these two things multiplied together tells you the risk premium for Apple.

So let me give you some, let's move on, and get some intuition for this.

Because, I think, this is probably the most used equation other than

discounted value of, the stock prices, the discounted value of dividends.

Look at this equation linearly.

What is it showing?

First of all, the graph, what is the risk free asset here?

8:26

And the reason and the thing I like about it is it's very easy to measure.

Right?

It's the one thing I know about the future is how much will a government bond pay.

And why do I feel confident?

Because I believe the government will pay.

8:42

This is the risk premium which I will say should be somewhere between 5 and

7% and this is where it's very important to recognize which

number you want to use here and textbooks tend to use 7% I believe.

It is a good idea to use 5% as well, at a minimum.

Because remember, all the answers are not perfectly precise.

And what is there?

9:07

This is the risk of comparable equity.

Why?

It's almost always the risk of the equity, why?

Because equity trades, and even if there's debt in your business,

9:28

most countries debt is a private contract between the company and a bank.

So it doesn't trade.

In America you can get some information on debt as well.

But typically if you see a beta anywhere, and

we'll see some soon, it's about the equity.

Because equity trades and easy to calculate.

Now here is ironic,

which is more difficult in instrument to think about debt or equity?

Debt is a contract, simple to understand.

Equity, love and fresh air.

[LAUGH] Very tough, but

on the other hand risk of these to capture because of training.

Don't forget the importance of market.

Okay, so what does this say?

That the return that is expected, and please recognize

10:12

all r's in this are expected, because it is about the future.

You are going to use them from the past data except for

rf because you know it about the future and assure the government will pay.

Please remember the expected is very important you are trying to

project to the future what may happen, that's why I gave you an emphasis on.

Don't use 7% for all for r and minus rf so easily, because it's questionable.

11:10

Right?

What's the risk of the risk free asset?

So rf, its beta f has to be 0, by definition.

Why?

Because a government bond, remember I told you what the risk was?

Was 0 simply because I believed I will get 1,000 regardless of the state of

the world, the face value of 1,000.

So bf is 0.

So, I pick up the treasury build rates.

Suppose it's 4%.

Tell me which point I will have to find on the graph.

4%.

I botched here and I know beta is 0.

There's another guy whose beta I know.

And what is that?

12:12

Sol I have two points, one is this and

one is this, and I draw this red line right through.

You see how simple this is, and that's the simplicity I was talking about.

One little equation.

12:26

So let me ask you this, what should be the return on a risk free asset?

Suppose I found another project which I believed had no risk.

Unlikely, but let's find it.

What should its return be?

Well if its risk is 0,

I plug in 0 here, this whole thing cancels, I'm left with a risk free rate.

Does that make sense?

12:49

Absolute sense.

Let me give you one more intuition.

Let me ask you.

What is the risk of a portfolio, or sorry, let's not talk about a portfolio.

What is the risk of a project whose riskiness is the same as the market?

That is, the project moves one and one with the market, 1% up, 1% down.

Its beta is what?

13:25

Because what happens when you plug 1?

Rf and rf cancel.

Isn't this cool?

What it is saying?

It's saying you can predict what's going to happen for

two points and they all both make sense.

If your projects almost risk free, your discount rate should be risk free right.

Think about this,

if your project is bananas your discounter should reflect banana.

14:56

That's the premium in the market place that you [INAUDIBLE].

Is this measurable?

Sure.

If the first two are measurable I can go historically and

look at the difference over 80 years, or over earlier markets and so on.

And this could be 7% or 5% or many people believe maybe even less.

15:18

And this is the biggest research area.

One of the biggest research areas in finance,

is why has the risk premium been so high?

In the US, more importantly, will it be the same in the future.

15:47

The intuitive and simple idea is that people are risk averse.

They therefore hold portfolios.

If they hold portfolios the risk of

one thing will depend on its relationship with a bunch of other things.

That one thing in our case is Apple because we are trying to evaluate orange

comparable.

And what is everything else?

The whole marketplace.

16:11

Simple measure of risk, why?

Because not only is the idea simple, the measure of risk is very simple.

How do we measure it?

We have done it at length, I'm just highlighting everything.

You measure it by taking returns on Apple, a comparable for us, and

running a regression on markets.

16:39

I think this is even, so the idea is simple, the measure of risk is simple,

but the relationship between risk and return is just so simple and

intuitive, and that is what we talked about in CAPM.

It's linear, and

easy to measure.

Very easy to measure.

I mean I'm getting a little carried away here, but I understand.

Why? Because I can't think of anything

that could have been simpler than that and easily measurable, of course.

Is it perfect?

Answer is no.

17:22

Lot of research has shown, not surprisingly, that not all assets

fit this perfect relationship between when you measure the beta and

you measure the return.

Beta doesn't capture all the riskiness.

And that's not surprising, is it?

[LAUGH] For example, if I told you you can measure love.

Would you find one simple linear relationship if you did there's something

missing, am I right?

But the awesomeness of this is conceptually it's so

clean, and practically measuring is so clean too.

But don't expect it to do wonders every time, okay?

In a more detailed class,

we measure different ways of measuring risk and return.

But the profound simple fact is this.

If you hold a diversified portfolio,

you should just be looking at common elements across securities, not specific.

And that basic idea carries through all future developments.

I wanted to highlight that.

Don't expect it to be perfect.

But it's profoundly close to perfect.