[MUSIC]

So what causes investors to trade?

Well one of the things that can cause investors to trade is overconfidence.

So I think I have a lot of great insights into how stocks will perform,

so that causes me to buy, buy, and sell them, or

I underestimate the risk associated with investments.

So because of this overconfidence, maybe this leads to trading a lot,

but it turns out ex-post not to be good on based on good information.

So at the end of the day, all I end up is making the brokerage house risk.

So like then the key question is to think about and

let's bring Le Penseur, in here whenever we have a question.

Is there a group of investors that one can identify as more likely to be

overconfident than not, and can you do this a priori?

So by a priori, we mean ahead of time, can you identify a certain group of

investors as more likely to be overconfident?

So let's think about that and then let's discuss some possibilities.

So how to identify overconfidence.

So here you have Napoleon leaving retreat from Moscow in 1812,

a very devastating excursion.

Kind of taking half a million, 600,000 french troops in to conquer Russia,

then the retreat where, depending what you read,

he comes out with 10,000 to 100,000 troops, decimated army.

So certainly overconfident in terms of the prospects of conquering Russia in

the 1800s here.

So can we think of certain personality attributes or physical characteristics,

that could maybe give a clue that this person's more likely to be overconfident?

So what are some potential proxies for overconfidence?

One is perhaps gender.

So Barber and Odean have done some work where they hypothesize that males when it

comes to making investment decisions, maybe more overconfident than females.

So that would mean males, you would think are trading more.

But they don't have anything to show for it at the end,

in terms of better performance.

All they have are the higher trading costs.

And then also hypothesis that tall people might be more overconfident.

Now in this study here done by Addoum, Korniotis, and Kumar,

they don't have data on the portfolio performance.

They just have data looking at like, are these people more or

less likely to invest in stocks, for example?

And taller people do seem to take on more risk in the portfolio.

But it's hard to call that if that's necessarily overconfidence,

because we need to have the benchmark about how are these investments doing?

But I thought, given I'm kind of six feet three inches myself,

I was kind of intrigued by this, are this tall people more confident or not?

So when we look at Napoleon, look at these two measures,

he would have like one of the two, but not both of them.

I thought just going back in time, like is there anything to this kind of height,

causing people to be more confident and then maybe perhaps overconfident.

We'll just go back to the beginning of our country here in the US.

You have George Washington, he kind of towers among all his compatriots here.

If you go to the Signers Hall in Philadelphia right across from

Independence Hall, you see these bronze statues of all the founding fathers that

signed the Declaration of Independence.

I looked at George Washington, he's exactly my same height.

So I think it's six feet three inches, he just towers over all the others,

like James Madison.

It seems like he's over half a foot taller than him.

So maybe one of the reasons he was elected president, besides being the great

winning general, is that he just physically towers over all the people.

And I also like this picture with this kind of bizarre to his left,

almost body double here.

David Humphrey, who is like his aide-de-camp.

So like the second tallest person in the picture is this almost clone

kind of dressed exactly the same here.

So, maybe there is something to this hype.

And there's also kind of an interesting Malcolm Gladwell,

noted that in the US population, 15% of all men are 6 feet or over.

But among CEOs of Fortune 500 companies, 60% almost,

58% to be precise, are over 6 feet.

And if you look at 6 feet 2 inches or more, that's only 4% of a population.

But it's 30% among CEOs.

So kind of the logic here is that when kids are young, particularly boys,

they're always encouraged or given compliments just because they're tall.

Like are you really in second grade?

It looks like, given your height, you could be a fourth grader, okay.

Maybe they're better at sports like basketball, so that kind of builds some

confidence that maybe would spill over to the investment realm.

So for the height, maybe that's leading to confidence,

maybe it's leading to overconfidence.

Hard to measure that without kind of having good measures, or

look at the performance of portfolios.

So instead, we're going to focus on kind of the gender issue, and

we're going to look into this Barber Odean study.

And their hypothesis is that males are more overconfident than females.

And I'll let you kind of decide if you think that's reasonable or not.

I'm just throwing that out there as a hypothesis

based on the work of Barber and Odean.

Really, they threw it out as a hypothesis, I'm just reporting the results.

So don't send me any angry emails if you disagree with this hypothesis here.

The study was called Boys Will Be Boys.

So in this analysis, Barbar and Odean are using this brokerage data set for

this large sample of US households over the period 91 to 96.

And we can break kind of individuals into four categories here.

Single women versus single men is kind of the interesting comparison.

We also have married women, married men, that's more complicated because presumably

that could be a joint account and it happens to have the wife's name on it,

the husband's name on it, and probably likely there are some.

And you expect some influence of men on women and

women on men in the married group.

So let's really focus on single women versus single men,

and the first thing simply to look at is, who trades the most?

And again, this is based on this brokerage data set from 1991 to 1996.

So you see single men, their monthly turnover is about 7% per year,

or about 85% on an annual basis.

So about 85% of their portfolio is flipped over in the course of a year.

For single women, 4% per month,

that would mean about 50% of their portfolio is turned over in a year.

So men are definitely trading more than women.

And you see, married men trading more than married women, but not the same amount.

So marriage is having this kind

of effect in causing the married men to trade less than the single men, okay.

So now let's look at the returns of the different groups here by gender,

we're looking here gross returns before trading cost.

So we see single men, they're outperforming their benchmark,

it’s a two-factor model.

So we're controlling for the CAPM market risk and we're controlling for

the size factor here.

So you see, men are outperforming by about 0.1% per month, or

about 1% on an annual basis, underperforming their benchmark.

Single females about four basis points, or about half a percent on an annual basis.

So this really aren't gigantic amounts, and isn't a really big difference,

but you know already this is bad news for single men, right?

They're trading a lot, and

looking at returns before trading cost, they aren't performing that well.

So once we look at the returns after taking into account the trading cost,

commissions to brokers, paying the bid as spread.

We see women are underperforming by about 14 basis points per month or

between 1 and a half and 2 percentage points on an annual basis.

But single men about double that, okay, underperforming by about 29 basis

points per month, or 3 and a half to 4 percentage points on an annual basis.

So you're looking, who do you want to be your mutual fund manager,

who do you want to be your financial adviser?

Maybe the single woman here, right?

Kind of the single woman and the single man,

neither outperforming their benchmark, given the risk of their stock investments.

But at least the single women are trading less, so

therefore, they have a lower, a less negative net return.

So kind of consistent with males being overconfident, trading more.

But not having anything to show for it at the end, except writing out

a bigger check to the brokerage house where they do their trades.

So overconfidence, what does it emanate from?

It might emanate from peer effects, okay?

So not surprisingly, word of mouth,

peer effects have strong influence on household portfolio decision.

So I have several papers that look at the role neighbors have on a given

household's decision of whether to own stocks and what stocks to own.

And using some careful identification strategies, can find that neighbor and

peers have very strong effects on your own financial decisions.

So just to kind of put it in perspective,

the effect of neighbors on your portfolio decisions is about

the same as the effect of your parents on your portfolio decision.

One portfolio decision I looked at in particular was the simple decision,

do you own stock or not?

The effect of neighbors on your choice to own stock or not.

It's about the same magnitude as to effect parents on that choice.

So I really like these three scenes here, they show various kind of

gossip happening here.

Now, one of the things to take into account is when someone's telling you

a story, particularly when it involves that person,

it's likely going to be positively bias.

Like people will tell you the great experiences they've had or

the great stock trades they've made.

But they probably won't tell you so much the bad experiences they had, or

the stocks they invested that went down in value.

So given that kind of bias in the information you collect from your peers,

it may make it seem like wow, it's easy to invest.

It's easy to pick kind of good stocks, okay?

So I think censored memories and

gossip can lead to overconfidence in investing, okay.

There's a bias that we likely forget bad performers, mainly because they disappear.

So we know Warren Buffett, we know his track record.

We don't know all the other thousands of investors that also

implemented strategies that didn't work out so well.

Because they just said they're gone, right?

They did poorly, so they disappear.

So all we hear about is Warren Buffett.

And it kind of suggests maybe it's not so

difficult to invest because we don't hear from all the people who failed.

Another factor is that people will focus on what other people tell them,

not take into account the kind of bias nature of the news they may be hearing.

So Bob Shiller had a great quote in that regard.

That others have made a lot of money appears to many people as the most

persuasive evidence that outweighs even the most carefully reasoned

argument, okay?

So forget what Scott was telling you in the lecture.

Forget what the Nobel Prize winner, Bob Schiller, is kind of telling you.

The fact that your buddy said, hey, I bought this stock and it tripled in a year

suggests to you, that wow, trading stocks is a great way to make money.

But what you don't take into account,

is your buddy is not telling you the four times he or she lost money on trades.

It's just kind of telling you the good stuff, okay?

So Kaustia and Knupfer actually do an empirical analysis to see,

is there this censored information transferal from peers to individuals?

So they actually look at the number of new investors in an area, and

they see how the number of new stock investors is related to the past

performance of existing investors in that area.

And they're looking at data in Finland, and it turns out in Finland,

there's almost 2,700, 2,649 zip codes.

How can Finland have so many zip codes, right?

So it's like very small zip codes, right, given Finland as a whole,

isn't the largest country.

They're looking at stock investment decisions over 1995 to 2002.

So key to the analysis that they're examining,

whether potential new investors respond to both positive and

negative returns earned last month by their neighbors.

So if my story or my thoughts are true,

that people talk about their good experience investing, but not their bad.

We'd expect that, hey, if the stock returns of neighbors were good last month,

they'll tell their neighbors and

people will be more likely to invest in the stock market.

But if returns of neighbors were bad last month, the neighbors don't say anything.

So we don't see it making it less likely that people will invest in the stock

market, because they just don't spread the negative news.

And what's great about this data set that Kaustia and

Knupfer have, is they have the data

on the stock portfolio performance of all these individuals in Finland.

And they have the data down to kind of where they know the zip

code where people live.

So they can relate your decision to invest in stocks

to the performance of your neighbors in the same zip code, okay?

So let's look at a results here, and we won't focus on the magnitudes too much.

I just want to focus on the direction of the effect.

So first we're just looking at, what's the entry rate,

what's a likelihood of a people who aren't currently investing the stock

market starting to invest in stocks?

And how is that related to the return earned by

existing investors in your neighborhood, in your zip code, last month?

And these here, we see this 1.056 in column 1,

1.042 in column 2, it means there's a positive relationship.

Higher return of neighbors last month,

more new entrants into the stock market the following month.

And these stars below it indicate that's a statistically significant result.

Now what's interesting is when the authors break this out into the two components,

okay?

What if returns were positive?

So this max is going to be the maximum of the neighborhood return in zero.

So this is showing the effect of what happens to having a positive return,

versus what happens if the return is negative.

So the minimum variable here is the minimum of the negative return and zero.

So we see this relationship that was presented in panel C,

where higher neighborhood return leads to more new investors,

that's totally driven by the positive results.

So if you have a positive return of your neighbors, more new investors next month.

But if you have a negative return, okay,

this negative return doesn't cause there to be less investors.

So there's an asymmetry,

when you break the return into a positive versus negative.

If the neighbors did well last month,

more likely new people invest in the stock market.

But if the neighbors do poorly, you don't see,

then it's less likely to have new people investing in the stock market.

And that's consistent with the notion, people will spread their good fortune, but

they keep quiet about their bad fortune.

So then you only hear stories about people's success in trading stocks,

you don't hear the stories about where they fail.

Therefore, you think it's easy to trade stocks, therefore,

you're more overconfident, then you trade more as well, okay?

So when we talked about this overconfidence at the beginning of

the video, we thought maybe male predicts being overconfident, maybe height, okay.

If that's the case,

I must be like the most overconfident person in the world here.

But I think peer effects, okay, may also play a role.

Because of the censored nature of gossip, and

then also the censored nature of memories.

We remember Warren Buffett, but we forget all the other investors that tried

different strategies that didn't work out so well.

[SOUND]

3-2.1. Overconfidence

Investments II: Lessons and Applications for Investors

与讲师见面