- Many economists view financial markets

as efficient with prices incorporating

all available information about future values.

Behavioral economists say that model

simply doesn't reflect how markets actually work.

So are markets efficient?

Welcome to The Big Question,

the monthly video series from Chicago Booth Review.

I'm Hal Weitzman and I'm joined by an expert panel.

Eugene Fama is the Robert R. McCormick

Distinguished Service Professor of Finance at Chicago Booth.

Well-known for his empirical analysis

of asset prices and for developing

the efficient market hypothesis,

he was the joint recipient

of the 2013 Nobel Prize in economic science,

and Richard Thaler is the Charles R. Walgreen

Distinguished Service Professor

of Behavioral Science and Economics at Chicago Booth.

He's director of Booth's center for decision research

and co-author of the bestseller, Nudge.

His most recent book is Misbehaving

The Making of Behavioural Economics.

Panel, welcome to The Big Question.

Gene Fama, let me start with you.

You came up with the efficient market hypothesis,

so tell us briefly, what is it?

- [Voiceover] Well, it's a very simple statement

that prices reflect all available information.

Testing the hypothesis turns out to be more difficult,

but it's a simple hypothesis in principle.

- Okay, Richard Thaler, do you agree that market prices

reflect all available information?

- Well.

Like Gene says, it's easier to say than to test

and I like to distinguish two aspects of it.

One is whether you can beat the market,

that's the one most people are most interested in,

and the other is whether prices are correct,

so if prices reflect all information,

then they should land on the right price

and we can separate those questions

because they're different.

- Okay, but the basic premise

about the containing information

is something you don't agree with then, it sounds like?

- It's almost impossible to test that hypothesis.

Except through the two questions that I've asked,

can you beat the market and are prices right?

- Okay, is that right, Gene Fama?

If you say that prices reflect all available information,

it necessarily means that a price is right

at any particular point in time?

- That's the statement of the hypothesis,

but it's a model, it's not completely true,

no models are completely true.

They're approximations to the world, the question is,

for what purposes are they a good approximation?

As far as I'm concerned,

they're a good approximation from almost every purpose.

I don't know any investors who shouldn't be able

if markets are efficient, for example,

and there are all kinds of test

with respect to the response of prices

to specific kinds of information

in which the hypothesis that prices adjust quickly

to information looks very good.

There are others that looks less good.

So, it's a model, it's not entirely always true,

but it's a good working model for most practical uses.

- Okay, is that right, is it a good working model?

- Well.

I think for the first part, can you beat the market?

I think Gene and I are in

virtually complete agreement

which is that's a good working hypothesis for any investor.

- Does that mean you should assume,

any individual investor should assume

that markets are efficient?

- Behave as if, would be the way you put it.

- Well, behave as if.

- Well, yeah, certainly,

there's evidence going back to

the thesis of Mike Jensen who was,

I guess, I mean, one of Gene's first students,

who was around at the same time,

who did a thesis on whether mutual funds, on average,

beat their benchmarks,

and after they account for their fees, they don't,

that was in the 60s, it's been updated a zillion times.

You can quibble about exactly how to do it,

but that's approximately true.

So.

Just investing based on fees

is not a dumb thing to do

regardless of the nuances that Gene and I might get into.

- Okay, but you talked earlier about,

in some cases, the model works,

in other cases, it works less well.

In your book, Richard Thaler,

you talk about the 1987 crash, Black Monday,

and you give that as an example of how prices are not right,

the efficient markets don't really work.

You say, "If prices are too variable,

"they're in some sense wrong.

"It's hard to argue the price at the close trading

"on Thursday, October 15th,

"and the price at the close of trading the following Monday,

"more than 25% lower, can both be rational measures

"of intrinsic value given the absence of news,"

but isn't the idea that efficient markets

are supposed to be unpredictable?

- Yes, but unpredictable doesn't mean rational.

I have a two year old granddaughter

who runs around like crazy,

and I defy anyone to

use a rational model to predict what she's gonna do next,

so she will be unpredictable,

but her behavior isn't well-captured by

a model of maximizing

anything other than what she calls fun.

- So, does the market behave

in the same way as your granddaughter?

- Well, sometimes.

I don't think anyone thinks

that the value of the world economy fell 25% that day.

Nothing happened.

- So, if markets were efficient,

there would be a certain bounded level of volatility?

So, if--

- Well, in the absence...

It's not a day when World War III was declared.

- But it was a time where people were talking about,

perhaps an oncoming recession

which turned out not to happen,

so in hindsight, this was a big mistake,

but it needn't have been.

So, in hindsight, every price is wrong.

- Yeah.

- That's 20/20 hindsight.

- That's 20/20 hindsight,

but what I would say is merely

the big fluctuations that entire week.

Two of the biggest up days in history occurred that week

and three of the biggest down days

and nothing was happening

other than the fact that people were talking about

how markets were going up and down

like crazy all over the world,

so that's one...

indirect way that we can measure market efficiency.

(coughs)

This was essentially the approach

that was pioneered by Bob Shiller,

who Gene shared the Nobel Prize with,

and his argument was prices fluctuate too much

to be explained by a rational process.

- Right, and Gene Fama, is that right?

There's a certainly level at which prices

just fluctuate too much?

- Well, there's a test for that

and the test says that

when we look at longer periods of time,

the variants of the price changes

should not grow like the length of the time period.

If there is all of this temporary variation in prices,

that's not rational and, in fact, that does not indicate

that there's temporary variation in prices,

so you gotta kind of come up with a different test.

Shiller's model was based on the proposition

that there's no variation through time in expected returns,

but we know there is a ton of variation in expected returns,

so that kind of branch of testing, people lost interest in

because you really can't come to any conclusions.

A straight test of whether

there's temporary variation in prices says,

"No, there isn't, you can't identify it"

and another test which would be,

is there too much variation in expected returns

to be attributed to rational behavior?

Well, now you have to define what you mean by that

and that's terribly difficult.

- But did augment then that investors

are constantly changing the view

of the expected future value of the shares?

- My view is that risk aversion

moves dramatically through time.

In particular, it's very high during bad periods

and it's lower during good periods,

and that affects that pricing of assets

and then the expected returns you expect.

- And so, every time there's a stock market crash,

people come to you, are they,

"Rubbish the idea of efficient markets,"

and point to this massive volatility

and say, "Does that prove that efficient markets are wrong?"

And then--

- Doesn't prove that at all.

- [Hal] What it proves is what?

Is that risk changes a lot very quickly?

- Well, the information changes a lot through time.

- Dick Thaler, let me turn to you.

Bubbles.

Gene Fama famously does not allow

use of the word bubble, but I'm gonna use it.

Do bubbles exist?

How do we define bubbles?

- Well, I think it's hard to say.

I'm gonna present two examples.

One which is,

I think, more convincing than the other.

The first one is of a graph that I think you'll show

the viewers,

it was produced by

Gene's son-in-law, John Cochrane,

and it's a graph of house prices

over a very long period of time in the US,

and what it shows is that, for a long period of time,

house prices were roughly 20 times rental prices,

and then starting around 2000, they go up,

depending on which measure you use,

they go up a lot or they go up a really lot,

and then they go back after the financial crisis.

I can't use this graph to convince Gene

that markets are inefficient.

- True.

Because the graph stops too soon.

- Well, yeah, but, you know,

we're not gonna live long enough to--

- No, no, no, I mean, if you continue the graph,

it goes back to the peak.

- Well, but--

- So, what's the bubble, the down, the up,

the subsequent down--

- Well, okay.

So, once again, we're in agreement which is that

studying data like this,

it's impossible to know for sure

whether something's a bubble

and there are hints in that graph

that prices seem to have diverged

from a level that had existed for a very long time,

they went up and then they went down.

Was this because of irrational exuberance

on Greenspan's phrase?

What we do know is that in the markets

like Vegas and Scottsdale and south Florida

where prices were going up the most,

expectations of future price appreciation

were also the highest

and that could be rational, but

I'm skeptical of that and of course, in hindsight,

it was wrong, but let me present another example

which is amusing at least.

Have I told you about this one?

The Cuba fund? - No.

No.

- Okay, this is good.

So, there's a closed-end mutual fund