1
00:00:00,090 --> 00:00:02,430
The following content is
provided under a Creative

2
00:00:02,430 --> 00:00:03,850
Commons license.

3
00:00:03,850 --> 00:00:06,060
Your support will help
MIT OpenCourseWare

4
00:00:06,060 --> 00:00:10,150
continue to offer high quality
educational resources for free.

5
00:00:10,150 --> 00:00:12,690
To make a donation or to
view additional materials

6
00:00:12,690 --> 00:00:16,620
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:16,620 --> 00:00:17,860
at ocw.mit.edu.

8
00:00:26,184 --> 00:00:27,600
PROFESSOR: What I
want to do today

9
00:00:27,600 --> 00:00:30,300
is to continue where
we left off last time

10
00:00:30,300 --> 00:00:37,080
in talking about the empirical
properties of stocks and bonds.

11
00:00:37,080 --> 00:00:39,660
I want you to
develop an intuition

12
00:00:39,660 --> 00:00:43,050
for how to think about markets.

13
00:00:43,050 --> 00:00:47,790
We've already done that over the
course of the last few lectures

14
00:00:47,790 --> 00:00:52,890
by looking at market
prices and understanding

15
00:00:52,890 --> 00:00:55,590
how to price them,
but I'd like you

16
00:00:55,590 --> 00:00:58,080
to get some kind of a
historical perspective

17
00:00:58,080 --> 00:01:01,260
now on specific asset classes.

18
00:01:01,260 --> 00:01:04,319
Because we're going to be
relying on market prices

19
00:01:04,319 --> 00:01:07,470
to make inferences about
other kinds of securities

20
00:01:07,470 --> 00:01:09,407
and other decisions
you're going to make.

21
00:01:09,407 --> 00:01:11,490
As I told you at the very
beginning of the course,

22
00:01:11,490 --> 00:01:14,520
we're going to rely on
markets for information,

23
00:01:14,520 --> 00:01:17,850
because it's the wisdom of
crowds that really gets us

24
00:01:17,850 --> 00:01:20,340
the information we
need in order to make

25
00:01:20,340 --> 00:01:22,000
good financial decisions.

26
00:01:22,000 --> 00:01:25,290
So I want to begin that
process of now giving you

27
00:01:25,290 --> 00:01:27,960
the intuition about
the wisdom of crowds

28
00:01:27,960 --> 00:01:32,160
by looking at the historical
performance of stocks

29
00:01:32,160 --> 00:01:33,097
and bonds.

30
00:01:33,097 --> 00:01:34,680
And then we're going
to talk about how

31
00:01:34,680 --> 00:01:38,760
to quantify risk
more analytically

32
00:01:38,760 --> 00:01:43,620
and put it all together in the
very basics of modern portfolio

33
00:01:43,620 --> 00:01:45,160
theory.

34
00:01:45,160 --> 00:01:48,180
So I want to start by
asking the question, first

35
00:01:48,180 --> 00:01:53,020
of all, what characterizes
US equity returns?

36
00:01:53,020 --> 00:01:57,690
How do we get our arms around
the behavior of that asset

37
00:01:57,690 --> 00:01:59,022
class?

38
00:01:59,022 --> 00:02:00,480
And the way I'm
going to do that is

39
00:02:00,480 --> 00:02:03,270
to give you some
performance statistics

40
00:02:03,270 --> 00:02:09,150
about the volatility, about
the average return, about how

41
00:02:09,150 --> 00:02:13,620
predictable they are,
and also patterns

42
00:02:13,620 --> 00:02:18,580
of returns across
different kinds of stocks.

43
00:02:18,580 --> 00:02:22,260
So we're going to look at
some empirical anomalies

44
00:02:22,260 --> 00:02:25,710
before actually turning to
the analytical work of trying

45
00:02:25,710 --> 00:02:29,880
to figure out how to make
sense of this from a more

46
00:02:29,880 --> 00:02:33,090
formal mathematical framework.

47
00:02:33,090 --> 00:02:36,300
Before I do that, let me ask
you to think about the following

48
00:02:36,300 --> 00:02:44,070
question, which is, if you are
designing a market for stocks,

49
00:02:44,070 --> 00:02:47,910
what properties would you
want that market to have?

50
00:02:47,910 --> 00:02:50,460
And I'm going to
argue that there

51
00:02:50,460 --> 00:02:53,340
are a few properties
that all of us, I think,

52
00:02:53,340 --> 00:02:58,390
can recognize as being good
properties for stock prices.

53
00:02:58,390 --> 00:03:02,730
So the first is that
stock market prices

54
00:03:02,730 --> 00:03:06,030
are random and unpredictable.

55
00:03:06,030 --> 00:03:08,460
Now, that might seem a
little counterintuitive,

56
00:03:08,460 --> 00:03:10,860
and certainly I think
you would acknowledge

57
00:03:10,860 --> 00:03:13,470
that over the last
several weeks markets have

58
00:03:13,470 --> 00:03:15,340
been supremely unpredictable.

59
00:03:15,340 --> 00:03:16,590
And that doesn't feel so good.

60
00:03:16,590 --> 00:03:19,270
It doesn't seem like
that's a good thing.

61
00:03:19,270 --> 00:03:21,240
But in a minute,
I'm going to try

62
00:03:21,240 --> 00:03:23,850
to make that a little
bit more clear by looking

63
00:03:23,850 --> 00:03:27,000
at the alternative
of predictable--

64
00:03:27,000 --> 00:03:29,050
or unpredictable,
which is predictable.

65
00:03:29,050 --> 00:03:30,900
So let me come
back to that point.

66
00:03:30,900 --> 00:03:33,780
The second property that
I think you'll agree

67
00:03:33,780 --> 00:03:35,670
is a reasonable one
for us to expect

68
00:03:35,670 --> 00:03:43,760
is that prices should react
quickly to new information.

69
00:03:43,760 --> 00:03:47,330
It should adjust
to new information

70
00:03:47,330 --> 00:03:50,180
really without
any kind of delay.

71
00:03:50,180 --> 00:03:55,280
And finally, we'd like to
see that investors shouldn't

72
00:03:55,280 --> 00:04:01,050
be able to earn abnormal returns
after you adjust for risk.

73
00:04:01,050 --> 00:04:03,170
So in other words,
once risk adjustment

74
00:04:03,170 --> 00:04:05,480
is taken into account,
there shouldn't

75
00:04:05,480 --> 00:04:08,360
be any additional
return left over.

76
00:04:08,360 --> 00:04:12,161
That's what we think of as
a well-functioning market.

77
00:04:12,161 --> 00:04:13,910
Another way of putting
it is that a market

78
00:04:13,910 --> 00:04:15,320
is highly competitive.

79
00:04:15,320 --> 00:04:19,040
It's hard to make
money in those markets.

80
00:04:19,040 --> 00:04:21,800
Now, they may not be markets
that you would enjoy trading

81
00:04:21,800 --> 00:04:23,460
in, but that's not the question.

82
00:04:23,460 --> 00:04:26,420
The question is, what
would be a good market,

83
00:04:26,420 --> 00:04:29,240
an efficient market?

84
00:04:29,240 --> 00:04:31,550
So let me talk about
predictability for a minute,

85
00:04:31,550 --> 00:04:33,950
because I said that it seems
a little counterintuitive

86
00:04:33,950 --> 00:04:37,650
that a good market is one
that's not predictable.

87
00:04:37,650 --> 00:04:41,240
So let's pretend that
this is the stock market.

88
00:04:41,240 --> 00:04:43,220
This is the S&P 500.

89
00:04:43,220 --> 00:04:48,200
That looks nice-- a
nice, regular curve.

90
00:04:48,200 --> 00:04:51,050
Anybody come up with
a prediction for this?

91
00:04:51,050 --> 00:04:53,870
How would you go about
predicting the behavior

92
00:04:53,870 --> 00:04:57,600
of this kind of a stock market?

93
00:04:57,600 --> 00:04:58,170
What's that?

94
00:04:58,170 --> 00:04:58,920
STUDENT: Cyclical.

95
00:04:58,920 --> 00:04:59,753
PROFESSOR: Cyclical.

96
00:04:59,753 --> 00:05:01,659
What kind of curve
would you fit to this?

97
00:05:01,659 --> 00:05:02,450
STUDENT: Sine wave.

98
00:05:02,450 --> 00:05:03,574
PROFESSOR: Yeah, sine wave.

99
00:05:03,574 --> 00:05:05,730
In fact, that's how
I generated this.

100
00:05:05,730 --> 00:05:09,420
I used a sine wave, and
then I add a little noise.

101
00:05:09,420 --> 00:05:14,130
Now, why might this not be
a good model for a market?

102
00:05:14,130 --> 00:05:17,520
If this were the stock
market, what would you do?

103
00:05:17,520 --> 00:05:18,227
Yeah?

104
00:05:18,227 --> 00:05:20,283
STUDENT: Everybody
would buy on the low

105
00:05:20,283 --> 00:05:22,800
and sell them high
[INAUDIBLE] the other end.

106
00:05:22,800 --> 00:05:23,910
PROFESSOR: Exactly.

107
00:05:23,910 --> 00:05:29,490
After a few of these cycles,
you sort of get the idea.

108
00:05:29,490 --> 00:05:32,610
And if you're down here, you're
going to think, well, gee,

109
00:05:32,610 --> 00:05:34,560
I think it's likely
to go back up so I'm

110
00:05:34,560 --> 00:05:36,272
going to buy a ton over here.

111
00:05:36,272 --> 00:05:38,730
And when you get right up there,
you'll say, gee, you know,

112
00:05:38,730 --> 00:05:40,877
I think it's time
for me to sell a ton.

113
00:05:40,877 --> 00:05:42,960
And you don't have to go
through too many of these

114
00:05:42,960 --> 00:05:45,600
before you get richer
than your wildest dreams.

115
00:05:45,600 --> 00:05:46,361
Yeah?

116
00:05:46,361 --> 00:05:49,798
STUDENT: I would think that
[INAUDIBLE] regular market

117
00:05:49,798 --> 00:05:53,480
like that is not
true, because as soon

118
00:05:53,480 --> 00:05:56,181
as you want it to raise,
it's going to collapse

119
00:05:56,181 --> 00:05:58,977
at 50, 10 points below.

120
00:05:58,977 --> 00:06:00,310
PROFESSOR: That's exactly right.

121
00:06:00,310 --> 00:06:03,030
So as soon as you
start doing this,

122
00:06:03,030 --> 00:06:07,620
as soon as you try to do this,
what happens to the pattern?

123
00:06:07,620 --> 00:06:09,982
The pattern
disappears-- exactly.

124
00:06:09,982 --> 00:06:11,440
You see, this is
one of the reasons

125
00:06:11,440 --> 00:06:14,770
why finance is a lot more
challenging than physics.

126
00:06:14,770 --> 00:06:18,019
In physics, if you try to drop
a ball in a gravitational field,

127
00:06:18,019 --> 00:06:19,560
it won't change its
mind and say gee,

128
00:06:19,560 --> 00:06:21,935
now I'm going to change the
gravitational constant on you

129
00:06:21,935 --> 00:06:23,860
just because you're testing me.

130
00:06:23,860 --> 00:06:26,050
But in financial
markets, the moment

131
00:06:26,050 --> 00:06:28,900
you try to take advantage
of this pattern, the pattern

132
00:06:28,900 --> 00:06:29,540
changes.

133
00:06:29,540 --> 00:06:31,390
In fact, the more
you try to take

134
00:06:31,390 --> 00:06:35,150
advantage of it, the more
quickly the pattern changes.

135
00:06:35,150 --> 00:06:38,230
In fact, if you do this a lot--

136
00:06:38,230 --> 00:06:43,060
if there are a lot of people
trying to predict patterns--

137
00:06:43,060 --> 00:06:45,700
then you know what you get?

138
00:06:45,700 --> 00:06:47,110
You get no pattern.

139
00:06:47,110 --> 00:06:48,730
You get randomness.

140
00:06:48,730 --> 00:06:53,950
That's the idea behind an
efficient market being random.

141
00:06:53,950 --> 00:06:57,430
If it were not random,
then that means

142
00:06:57,430 --> 00:06:59,620
that there aren't
enough people who

143
00:06:59,620 --> 00:07:02,740
are bothering to try
to forecast the price

144
00:07:02,740 --> 00:07:07,010
and incorporate
information into the price.

145
00:07:07,010 --> 00:07:10,180
Now, I said two things that
at first seemed different,

146
00:07:10,180 --> 00:07:13,480
but in fact they're opposite
sides of the same coin.

147
00:07:13,480 --> 00:07:17,050
When you are forecasting
market prices,

148
00:07:17,050 --> 00:07:18,250
you know what you're doing?

149
00:07:18,250 --> 00:07:21,130
You're actually
helping markets become

150
00:07:21,130 --> 00:07:25,970
more efficient by incorporating
information into that price.

151
00:07:25,970 --> 00:07:26,810
How do you do that?

152
00:07:26,810 --> 00:07:30,100
Well, if, for example,
you think that having

153
00:07:30,100 --> 00:07:33,760
a presidential election
will cause volatility

154
00:07:33,760 --> 00:07:36,520
to decline, then if
you know that there's

155
00:07:36,520 --> 00:07:38,440
a presidential
election coming up,

156
00:07:38,440 --> 00:07:42,370
you will start trading in a way
that will ultimately be betting

157
00:07:42,370 --> 00:07:44,170
on volatility declining.

158
00:07:44,170 --> 00:07:48,160
As you start that trading, you
force that volatility index

159
00:07:48,160 --> 00:07:49,700
to go down.

160
00:07:49,700 --> 00:07:51,760
So the fact that
you've got information

161
00:07:51,760 --> 00:07:54,664
and you think you
can forecast prices,

162
00:07:54,664 --> 00:07:56,080
when you use the
information, what

163
00:07:56,080 --> 00:07:57,610
does it mean to
use the information

164
00:07:57,610 --> 00:08:00,790
when you buy or sell
securities on the basis

165
00:08:00,790 --> 00:08:02,140
of that information?

166
00:08:02,140 --> 00:08:05,200
Then the price of the
security ultimately

167
00:08:05,200 --> 00:08:08,410
reflects the information, right?

168
00:08:08,410 --> 00:08:13,210
So an efficient market is one
where you don't have this.

169
00:08:13,210 --> 00:08:17,140
You don't have a very
strong predictability.

170
00:08:17,140 --> 00:08:20,800
If it is strongly
predictable, then most likely

171
00:08:20,800 --> 00:08:25,370
either the market is rigged,
or there aren't enough people

172
00:08:25,370 --> 00:08:29,780
that are trading in order to
make prices fully reflective

173
00:08:29,780 --> 00:08:32,299
of all available information.

174
00:08:32,299 --> 00:08:35,690
Now, this is the way
markets really look.

175
00:08:35,690 --> 00:08:39,020
These are random
walks with drift,

176
00:08:39,020 --> 00:08:42,860
drift meaning there's a positive
trend or, in some cases,

177
00:08:42,860 --> 00:08:44,570
a negative trend.

178
00:08:44,570 --> 00:08:48,540
But otherwise, it's
random around that trend.

179
00:08:48,540 --> 00:08:51,890
So you can't really
easily forecast it.

180
00:08:51,890 --> 00:08:55,280
And you can see that
prices go up, they go down.

181
00:08:55,280 --> 00:08:57,440
There are long periods
where they go up,

182
00:08:57,440 --> 00:09:00,230
but there are also, for other
stocks, long periods where

183
00:09:00,230 --> 00:09:01,310
they go down.

184
00:09:01,310 --> 00:09:04,200
And you don't know what's
going to happen next.

185
00:09:04,200 --> 00:09:09,276
This is a sign of a
very efficient market.

186
00:09:09,276 --> 00:09:13,100
A while ago, there
was an academic study

187
00:09:13,100 --> 00:09:15,560
that was done to try
to test for efficiency,

188
00:09:15,560 --> 00:09:20,540
and one of the tests was
that if the underlying price

189
00:09:20,540 --> 00:09:26,090
series was not very
volatile, that was

190
00:09:26,090 --> 00:09:29,024
considered an efficient market.

191
00:09:29,024 --> 00:09:30,440
But it turned out
that was roundly

192
00:09:30,440 --> 00:09:35,000
criticized because of the point
that just because a market is

193
00:09:35,000 --> 00:09:38,570
not volatile, it doesn't
mean that it's working well.

194
00:09:38,570 --> 00:09:40,170
And an example
was, at the time--

195
00:09:40,170 --> 00:09:43,490
this was, like, 20 or 30 years
ago-- the Chinese stock market,

196
00:09:43,490 --> 00:09:44,840
the Shanghai Stock Exchange.

197
00:09:44,840 --> 00:09:48,955
It was a relatively young
market, and at that time,

198
00:09:48,955 --> 00:09:50,830
there were only two
stocks that traded on it.

199
00:09:50,830 --> 00:09:54,350
It was the National Railroad
Company and the Bank of China.

200
00:09:54,350 --> 00:09:56,990
And at that time,
which is, again,

201
00:09:56,990 --> 00:10:00,740
about 15 or 20 years
ago, it was considered

202
00:10:00,740 --> 00:10:04,910
unpatriotic to sell the
security if you had bought it.

203
00:10:04,910 --> 00:10:07,310
So you could buy it, but you
weren't allowed to sell it.

204
00:10:07,310 --> 00:10:10,010
And so the price went
way up and up and up,

205
00:10:10,010 --> 00:10:13,910
and that's not an example
of an efficient market.

206
00:10:13,910 --> 00:10:15,920
It was not at all volatile.

207
00:10:15,920 --> 00:10:18,980
But as a result, there
was no real information

208
00:10:18,980 --> 00:10:21,190
reflected in that price.

209
00:10:21,190 --> 00:10:21,972
Yeah?

210
00:10:21,972 --> 00:10:23,805
STUDENT: India is talking
about getting out.

211
00:10:23,805 --> 00:10:26,391
It looks like somebody's doing
one of those roller coaster

212
00:10:26,391 --> 00:10:28,224
rides that goes up and
down and up and down.

213
00:10:28,224 --> 00:10:30,680
So is volatility a sign of [?
inefficient ?] [INAUDIBLE]?

214
00:10:30,680 --> 00:10:33,680
PROFESSOR: Well, it's not
volatility, per se, but rather

215
00:10:33,680 --> 00:10:37,580
the combination of the
predictability per unit

216
00:10:37,580 --> 00:10:38,390
volatility.

217
00:10:38,390 --> 00:10:40,070
That's really what
you want to focus on.

218
00:10:40,070 --> 00:10:42,611
We're going to come back to that
when we talk about portfolio

219
00:10:42,611 --> 00:10:44,900
theory and look at this
trade-off between risk

220
00:10:44,900 --> 00:10:46,310
and expected return.

221
00:10:46,310 --> 00:10:49,050
But no, I wouldn't say that the
Indian market is inefficient.

222
00:10:49,050 --> 00:10:52,430
It's undergoing some
pretty significant changes,

223
00:10:52,430 --> 00:10:55,179
as is the US and
as is the world.

224
00:10:55,179 --> 00:10:56,720
But that's because
the global economy

225
00:10:56,720 --> 00:10:58,220
is contracting as
we know it because

226
00:10:58,220 --> 00:10:59,940
of this financial crisis.

227
00:10:59,940 --> 00:11:03,980
So I wouldn't characterize it
as inefficient at this point,

228
00:11:03,980 --> 00:11:06,230
but that remains to be seen.

229
00:11:06,230 --> 00:11:07,130
Yeah?

230
00:11:07,130 --> 00:11:10,190
STUDENT: [INAUDIBLE] down
the road [INAUDIBLE] market

231
00:11:10,190 --> 00:11:13,130
[INAUDIBLE]?

232
00:11:13,130 --> 00:11:16,700
PROFESSOR: There isn't any
hard and fast rule, no.

233
00:11:16,700 --> 00:11:19,820
But if you take a look
at the trade-off of risk

234
00:11:19,820 --> 00:11:22,730
to reward, in other
words that ratio

235
00:11:22,730 --> 00:11:24,920
of expected return
to volatility,

236
00:11:24,920 --> 00:11:26,810
you can come up
with rules of thumb

237
00:11:26,810 --> 00:11:29,240
that will give you a sense
of whether or not a market is

238
00:11:29,240 --> 00:11:30,360
efficient or inefficient.

239
00:11:30,360 --> 00:11:31,860
So we're going to
come back to that.

240
00:11:31,860 --> 00:11:33,734
In fact, we'll look at
that in just a minute.

241
00:11:33,734 --> 00:11:35,510
Let me turn to some
data now, and then we

242
00:11:35,510 --> 00:11:38,965
can see exactly what those
trade-offs look like.

243
00:11:38,965 --> 00:11:41,090
Now, what's going to be
interesting about this part

244
00:11:41,090 --> 00:11:44,190
of the talk is that when I
tell you about these numbers,

245
00:11:44,190 --> 00:11:46,790
these numbers are
based upon data from

246
00:11:46,790 --> 00:11:49,910
I think it was 1946 to 2001.

247
00:11:49,910 --> 00:11:53,870
In fact, a lot of the data
that has been collected

248
00:11:53,870 --> 00:11:57,140
over the last year is very,
very different from this,

249
00:11:57,140 --> 00:12:01,850
so it will be interesting
to sort of compare the two.

250
00:12:01,850 --> 00:12:04,430
So there are four
empirical facts

251
00:12:04,430 --> 00:12:08,310
that I want you to take with
you about the US stock market.

252
00:12:08,310 --> 00:12:13,040
The first is that interest
rates, in general,

253
00:12:13,040 --> 00:12:16,700
have been slightly positive
on average, but not by much.

254
00:12:16,700 --> 00:12:18,590
In other words, the
real interest rate,

255
00:12:18,590 --> 00:12:21,110
the nominal interest rate
minus inflation rate,

256
00:12:21,110 --> 00:12:26,880
has been pretty low over
the course of history.

257
00:12:26,880 --> 00:12:29,420
So the first fact
is that real rates

258
00:12:29,420 --> 00:12:31,610
have been slightly positive.

259
00:12:31,610 --> 00:12:35,150
So you can see, for example,
the average rate of return

260
00:12:35,150 --> 00:12:39,530
for the one-year T-bill
is about 38 basis

261
00:12:39,530 --> 00:12:42,314
points on a monthly basis.

262
00:12:42,314 --> 00:12:42,980
This is monthly.

263
00:12:42,980 --> 00:12:44,806
I haven't annualized it.

264
00:12:44,806 --> 00:12:47,690
But inflation over
that same period

265
00:12:47,690 --> 00:12:49,760
is about 32 basis points.

266
00:12:49,760 --> 00:12:51,260
So when you subtract
the two, you're

267
00:12:51,260 --> 00:12:54,020
going to get six basis
points on a monthly basis

268
00:12:54,020 --> 00:12:57,830
as the real rate of interest.

269
00:12:57,830 --> 00:13:02,490
On the other hand, if you take
a look at the stock market,

270
00:13:02,490 --> 00:13:05,500
which is represented
by VW stock index--

271
00:13:05,500 --> 00:13:08,570
VW doesn't mean
Volkswagen. VW stands

272
00:13:08,570 --> 00:13:10,280
for the value weighted index.

273
00:13:10,280 --> 00:13:13,010
It's an index of all
the stocks on the NYSE,

274
00:13:13,010 --> 00:13:14,840
Amex, and NASDAQ
weighted according

275
00:13:14,840 --> 00:13:19,370
to their outstanding
market capitalization.

276
00:13:19,370 --> 00:13:22,430
And you can see that
the valuated stock

277
00:13:22,430 --> 00:13:26,780
market over this period
is about 1% per month.

278
00:13:26,780 --> 00:13:29,720
The equal weighted
stock market, EW,

279
00:13:29,720 --> 00:13:32,450
is a little bit higher, 1.18.

280
00:13:32,450 --> 00:13:38,690
And Motorola over this period
had an expected rate of return

281
00:13:38,690 --> 00:13:42,750
of about 1.66% per month.

282
00:13:42,750 --> 00:13:47,570
So the return has been
higher for these indexes.

283
00:13:47,570 --> 00:13:51,530
And if you want to get a
sense of why that might be,

284
00:13:51,530 --> 00:13:53,270
take a look at the
next column, which

285
00:13:53,270 --> 00:13:54,740
is the standard deviation.

286
00:13:54,740 --> 00:13:59,510
This, remember, is a measure of
the riskiness of the security.

287
00:13:59,510 --> 00:14:01,700
It's a measure of
the fluctuations.

288
00:14:01,700 --> 00:14:03,890
And if you take a look
at the equal weighted

289
00:14:03,890 --> 00:14:07,140
and the valuated,
the volatility is--

290
00:14:07,140 --> 00:14:10,910
they're both a lot
higher than for T-bills.

291
00:14:10,910 --> 00:14:15,920
So this is one of the reasons
we got the idea in finance

292
00:14:15,920 --> 00:14:18,710
that there's a
risk-reward trade-off.

293
00:14:18,710 --> 00:14:22,430
The more risky, the higher
the expected rate of return.

294
00:14:22,430 --> 00:14:25,820
And if you look at Motorola,
the riskiness of Motorola

295
00:14:25,820 --> 00:14:29,690
is much larger than that of
any of the stock indexes.

296
00:14:29,690 --> 00:14:32,450
Instead of a 5% monthly
standard deviation,

297
00:14:32,450 --> 00:14:34,850
you're looking at
double, or 10%,

298
00:14:34,850 --> 00:14:36,169
the monthly standard deviation.

299
00:14:36,169 --> 00:14:37,460
But look at the rate of return.

300
00:14:37,460 --> 00:14:39,950
The rate of return is
commensurately higher.

301
00:14:39,950 --> 00:14:40,759
Yeah, Remi?

302
00:14:40,759 --> 00:14:43,693
STUDENT: Professor, why
[INAUDIBLE] standard deviation

303
00:14:43,693 --> 00:14:44,671
greater than zero?

304
00:14:44,671 --> 00:14:46,140
Isn't that [INAUDIBLE]?

305
00:14:46,140 --> 00:14:49,000
PROFESSOR: Oh, because it
fluctuates from month to month.

306
00:14:49,000 --> 00:14:51,960
So the idea is if you're buying
a T-bill and you're holding it,

307
00:14:51,960 --> 00:14:53,660
it still has price fluctuation.

308
00:14:57,970 --> 00:14:59,860
The other thing that
I want you to see

309
00:14:59,860 --> 00:15:05,600
is something along the lines
of the minimum and the maximum.

310
00:15:05,600 --> 00:15:07,810
This is another
way of representing

311
00:15:07,810 --> 00:15:10,870
the riskiness of the security.

312
00:15:10,870 --> 00:15:17,200
So T-bills are bounded
between 0.03 and 1.34 in terms

313
00:15:17,200 --> 00:15:18,220
of their return--

314
00:15:18,220 --> 00:15:20,500
very narrow band.

315
00:15:20,500 --> 00:15:24,100
Treasury notes, which
are 10-year instruments,

316
00:15:24,100 --> 00:15:27,790
obviously are going to be
swinging around much more, more

317
00:15:27,790 --> 00:15:29,210
than a one year T-bill.

318
00:15:29,210 --> 00:15:34,390
So the longer the maturity,
the longer the duration,

319
00:15:34,390 --> 00:15:37,390
the riskier is the instrument.

320
00:15:37,390 --> 00:15:40,840
But if you take a look
at the valuated return

321
00:15:40,840 --> 00:15:44,050
and the equal weighted
return and then Motorola,

322
00:15:44,050 --> 00:15:49,480
you can see progressively
more and more risk involved

323
00:15:49,480 --> 00:15:52,840
in these kinds of securities.

324
00:15:52,840 --> 00:15:56,320
Now, if you look at their
compound growth rates,

325
00:15:56,320 --> 00:16:00,130
you get what you pay for,
in the sense that you're

326
00:16:00,130 --> 00:16:03,220
looking at T-bills down here.

327
00:16:03,220 --> 00:16:06,010
So $1 invested in
one of these guys

328
00:16:06,010 --> 00:16:11,170
will give you maybe
$10 at the end of 2001,

329
00:16:11,170 --> 00:16:14,140
but you're looking at a
much, much larger return

330
00:16:14,140 --> 00:16:18,460
for either the valuated
or equal weighted indexes.

331
00:16:18,460 --> 00:16:22,360
More risk, more
expected return--

332
00:16:22,360 --> 00:16:25,660
that's the message that you get
from looking at the basic data

333
00:16:25,660 --> 00:16:27,280
here.

334
00:16:27,280 --> 00:16:29,200
Now, that's just to give
you a sense of where

335
00:16:29,200 --> 00:16:30,241
interest rates have been.

336
00:16:30,241 --> 00:16:32,560
I think we discussed this
when we did fixed income.

337
00:16:32,560 --> 00:16:34,900
Interest rates have really
been all over the map.

338
00:16:34,900 --> 00:16:36,670
There was a point
in our history,

339
00:16:36,670 --> 00:16:40,570
not that long ago, where
the short-term interest

340
00:16:40,570 --> 00:16:45,460
rate, the one year T-bill,
was something like 16% to 17%

341
00:16:45,460 --> 00:16:46,600
per year.

342
00:16:46,600 --> 00:16:48,640
That's the one-year T-bill.

343
00:16:48,640 --> 00:16:50,500
It's astonishing.

344
00:16:50,500 --> 00:16:51,970
But that was a
period where there

345
00:16:51,970 --> 00:16:56,170
was a large amount of
inflation in the United States.

346
00:16:56,170 --> 00:16:59,410
On the other hand, if you take a
look at the more recent period,

347
00:16:59,410 --> 00:17:01,828
interest rates have
been extremely low,

348
00:17:01,828 --> 00:17:04,119
and that's part of the reason
we're in a credit crisis,

349
00:17:04,119 --> 00:17:07,010
is because credit is very cheap.

350
00:17:07,010 --> 00:17:09,400
So you can get yourself
into a lot of trouble

351
00:17:09,400 --> 00:17:12,099
when it's relatively
easy for you to borrow,

352
00:17:12,099 --> 00:17:17,250
and it doesn't cost you that
much in terms of the payments.

353
00:17:17,250 --> 00:17:19,349
Now, let me show you what
the total returns look

354
00:17:19,349 --> 00:17:21,089
like for these
different asset classes.

355
00:17:21,089 --> 00:17:23,819
By total returns, I mean if you
bought one of these instruments

356
00:17:23,819 --> 00:17:25,770
and you held it
a month at a time

357
00:17:25,770 --> 00:17:29,280
and you computed the return
for holding that instrument.

358
00:17:29,280 --> 00:17:32,150
So for a bond, it includes
whatever coupons get paid,

359
00:17:32,150 --> 00:17:34,360
plus the price fluctuations.

360
00:17:34,360 --> 00:17:36,060
It's the total return.

361
00:17:36,060 --> 00:17:38,820
For stocks, it'll include
the dividends that got paid

362
00:17:38,820 --> 00:17:40,830
as well as the
price fluctuations.

363
00:17:40,830 --> 00:17:43,860
And I'm going to do this
on the exact same scale,

364
00:17:43,860 --> 00:17:47,280
from minus 25% to 25%.

365
00:17:47,280 --> 00:17:50,520
So these are monthly
returns now that I'm

366
00:17:50,520 --> 00:17:53,990
plotting from 1946 to 2001.

367
00:17:53,990 --> 00:17:58,200
And so you can see that
the total return for the US

368
00:17:58,200 --> 00:18:02,730
10-year bond actually has
different periods where,

369
00:18:02,730 --> 00:18:06,540
in some cases, it's not very
risky but in other cases

370
00:18:06,540 --> 00:18:08,250
it bounces around a great deal.

371
00:18:08,250 --> 00:18:11,190
When there's a fair amount
of interest rate uncertainty,

372
00:18:11,190 --> 00:18:14,250
you get a lot of volatility,
but when markets are not

373
00:18:14,250 --> 00:18:17,310
moving around that much
on the interest rate side,

374
00:18:17,310 --> 00:18:21,940
you get periods that
are relatively calm.

375
00:18:21,940 --> 00:18:23,208
Yes?

376
00:18:23,208 --> 00:18:26,701
STUDENT: In the previous
slide, that interest rate

377
00:18:26,701 --> 00:18:28,170
are the yield to
maturities, right?

378
00:18:28,170 --> 00:18:29,420
PROFESSOR: Yeah, that's right.

379
00:18:29,420 --> 00:18:30,300
That's right.

380
00:18:30,300 --> 00:18:30,860
Correct.

381
00:18:30,860 --> 00:18:34,480
These are yields to maturity.

382
00:18:34,480 --> 00:18:36,470
These are total returns, though.

383
00:18:36,470 --> 00:18:39,290
These are what you
get as an investor.

384
00:18:39,290 --> 00:18:42,920
This is what you get if you
hold a particular security

385
00:18:42,920 --> 00:18:44,240
to maturity.

386
00:18:44,240 --> 00:18:49,130
Basically, on a given day,
you will get these spot rates.

387
00:18:51,860 --> 00:18:53,396
You have a question?

388
00:18:53,396 --> 00:18:55,886
STUDENT: [INAUDIBLE] 10 year.

389
00:18:55,886 --> 00:19:00,368
So in '81 and '82,
you could buy any bond

390
00:19:00,368 --> 00:19:02,870
from the year '86 [INAUDIBLE].

391
00:19:02,870 --> 00:19:03,758
PROFESSOR: Yeah.

392
00:19:03,758 --> 00:19:08,638
STUDENT: So [INAUDIBLE] that
incorporates future [INAUDIBLE]

393
00:19:08,638 --> 00:19:09,620
interest rates, right?

394
00:19:09,620 --> 00:19:11,030
PROFESSOR: Well,
and also inflation.

395
00:19:11,030 --> 00:19:13,196
STUDENT: And that inflation--
was it people thinking

396
00:19:13,196 --> 00:19:18,454
that inflation was going to be
[INAUDIBLE] 10% of the year,

397
00:19:18,454 --> 00:19:19,430
the 10 years out?

398
00:19:19,430 --> 00:19:20,990
PROFESSOR: Either
that, or they felt

399
00:19:20,990 --> 00:19:24,405
that that, plus whatever
interest rate expectations,

400
00:19:24,405 --> 00:19:25,780
was going to be
what you're going

401
00:19:25,780 --> 00:19:27,040
to get over a 10-year period.

402
00:19:27,040 --> 00:19:28,190
Yep.

403
00:19:28,190 --> 00:19:30,412
That's right.

404
00:19:30,412 --> 00:19:33,875
STUDENT: And mentally, people
think that stock [INAUDIBLE]

405
00:19:33,875 --> 00:19:36,904
[? upon ?] [? death.
?] [INAUDIBLE].

406
00:19:36,904 --> 00:19:37,820
PROFESSOR: Apparently.

407
00:19:37,820 --> 00:19:40,770
And there was a time, in fact,
when the stock market did yield

408
00:19:40,770 --> 00:19:42,710
that for quite a bit of time.

409
00:19:42,710 --> 00:19:45,662
Absolutely.

410
00:19:45,662 --> 00:19:47,870
On the other hand, let me
turn it around and ask you,

411
00:19:47,870 --> 00:19:50,600
now that interest rates are at--

412
00:19:50,600 --> 00:19:54,410
I don't know, the 10-year I
think is at-- or the 30 year is

413
00:19:54,410 --> 00:19:56,880
at 4.17 this morning--

414
00:19:56,880 --> 00:19:59,600
do you think for the
next 40 years or 30 years

415
00:19:59,600 --> 00:20:04,560
that treasury bills are only
going to return 4% a year?

416
00:20:04,560 --> 00:20:06,780
Is that realistic?

417
00:20:06,780 --> 00:20:09,690
I mean, it's not so
easy to say, is it?

418
00:20:09,690 --> 00:20:12,300
When you're in the midst of
it, it's not so easy to say.

419
00:20:12,300 --> 00:20:14,010
Based upon historical
evidence, it

420
00:20:14,010 --> 00:20:16,170
seems crazy to think
that we could possibly

421
00:20:16,170 --> 00:20:19,590
be in such a low interest rate
environment over the next 30

422
00:20:19,590 --> 00:20:22,470
years, especially given
that we're printing money

423
00:20:22,470 --> 00:20:24,190
like it's going
out of style now.

424
00:20:24,190 --> 00:20:26,815
And we're going to be doing that
over the next couple of years.

425
00:20:26,815 --> 00:20:28,470
We've got to, because
somebody's got

426
00:20:28,470 --> 00:20:32,140
to pay for all of
these rescue packages.

427
00:20:32,140 --> 00:20:36,840
And so we basically
have to engage

428
00:20:36,840 --> 00:20:41,040
in some kind of inflationary
monetary and fiscal policy.

429
00:20:41,040 --> 00:20:44,790
But it's still saying
4.17 as of this morning.

430
00:20:44,790 --> 00:20:50,610
So the market does the best
it can, given the data,

431
00:20:50,610 --> 00:20:52,200
but it's hard to forecast.

432
00:20:52,200 --> 00:20:54,990
And we just said at the
beginning of this class

433
00:20:54,990 --> 00:20:57,120
that it better be
hard to forecast,

434
00:20:57,120 --> 00:20:59,040
because if it's not
hard to forecast,

435
00:20:59,040 --> 00:21:00,720
then something's wrong.

436
00:21:00,720 --> 00:21:02,820
Then it means that
it's not reflecting

437
00:21:02,820 --> 00:21:04,817
all available information.

438
00:21:04,817 --> 00:21:05,316
Yeah?

439
00:21:05,316 --> 00:21:08,238
STUDENT: When you showed that
graph in the previous slide

440
00:21:08,238 --> 00:21:10,809
of the stocks going way up,
I'm wondering how much of that

441
00:21:10,809 --> 00:21:13,225
is due to the fact that this
was just a really good period

442
00:21:13,225 --> 00:21:16,908
of the United States and if
you compare to other countries

443
00:21:16,908 --> 00:21:17,820
at other times.

444
00:21:17,820 --> 00:21:20,170
PROFESSOR: Yeah, that
is definitely a factor.

445
00:21:20,170 --> 00:21:22,800
So I'm not trying to
explain the numbers,

446
00:21:22,800 --> 00:21:24,480
and I'm not trying
to justify them.

447
00:21:24,480 --> 00:21:25,510
You're absolutely right.

448
00:21:25,510 --> 00:21:30,270
This is a very special country
in a very special time.

449
00:21:30,270 --> 00:21:33,490
So you can either thank your
good fortune that you're here,

450
00:21:33,490 --> 00:21:36,210
or you can argue that, well,
it's not going to persist

451
00:21:36,210 --> 00:21:40,782
and time to move to wherever.

452
00:21:40,782 --> 00:21:41,490
But I don't know.

453
00:21:41,490 --> 00:21:44,060
I don't know which that is.

454
00:21:44,060 --> 00:21:45,150
But it is very unusual.

455
00:21:45,150 --> 00:21:46,530
If you look at other
countries, there

456
00:21:46,530 --> 00:21:48,190
are other countries that
are having difficulties

457
00:21:48,190 --> 00:21:50,640
during this time period, but
there are other countries

458
00:21:50,640 --> 00:21:53,530
that are growing even faster.

459
00:21:53,530 --> 00:21:55,890
If you look at China
over the last 10 years,

460
00:21:55,890 --> 00:21:58,440
the growth rate of
the Chinese economy

461
00:21:58,440 --> 00:22:01,260
is double to triple
what the US economy.

462
00:22:01,260 --> 00:22:03,345
Now, it was a smaller
economy, but still

463
00:22:03,345 --> 00:22:04,845
over an extended
period of time it's

464
00:22:04,845 --> 00:22:07,380
got tremendous growth rate.

465
00:22:07,380 --> 00:22:09,540
So that really is
the challenge, is

466
00:22:09,540 --> 00:22:11,760
to try to understand
what's going

467
00:22:11,760 --> 00:22:17,190
on in the context of where we're
living and how we're living.

468
00:22:17,190 --> 00:22:19,410
So let me go through and
show you some more numbers,

469
00:22:19,410 --> 00:22:21,840
and then we can talk about
some of the interpretations.

470
00:22:21,840 --> 00:22:25,110
So this is the total
return for the US 10-year.

471
00:22:25,110 --> 00:22:26,880
You've get a sense
of the scale--

472
00:22:26,880 --> 00:22:28,710
minus 25 to 25.

473
00:22:28,710 --> 00:22:31,950
Now, this is the
return of the US stock

474
00:22:31,950 --> 00:22:35,400
market during that same
period using that same scale--

475
00:22:35,400 --> 00:22:36,990
more risky.

476
00:22:36,990 --> 00:22:41,020
So you can see the
difference in fluctuations.

477
00:22:41,020 --> 00:22:46,050
And if this weren't
exciting enough for you,

478
00:22:46,050 --> 00:22:48,650
this is Motorola.

479
00:22:48,650 --> 00:22:52,130
And there are many
stocks like Motorola.

480
00:22:52,130 --> 00:22:55,370
So when you invest in
an individual stock,

481
00:22:55,370 --> 00:22:58,160
you're getting not just the
fluctuations of the economy,

482
00:22:58,160 --> 00:23:00,530
but you're getting
the fluctuations that

483
00:23:00,530 --> 00:23:03,650
affect that specific company.

484
00:23:03,650 --> 00:23:07,700
So you should expect, if
you're taking on more risk,

485
00:23:07,700 --> 00:23:10,160
that you're going to be getting
a reward for these kinds

486
00:23:10,160 --> 00:23:12,140
of incredible bouncing around.

487
00:23:12,140 --> 00:23:13,340
And in fact, you do.

488
00:23:13,340 --> 00:23:16,790
The average return of
Motorola is quite a bit higher

489
00:23:16,790 --> 00:23:19,370
than that of the market.

490
00:23:19,370 --> 00:23:22,370
Now let me show you a little
bit about predictability.

491
00:23:22,370 --> 00:23:26,190
We talked about a market
that's random as one that

492
00:23:26,190 --> 00:23:29,000
is a good or efficient market.

493
00:23:29,000 --> 00:23:33,800
This plots the return
today versus tomorrow,

494
00:23:33,800 --> 00:23:36,680
or yesterday versus today if
you want to think about that--

495
00:23:36,680 --> 00:23:43,160
pairs of returns for the
aggregate stock market.

496
00:23:43,160 --> 00:23:48,000
And this is done
on a daily basis.

497
00:23:48,000 --> 00:23:53,940
Now, if you look at it on
a monthly basis for the S&P

498
00:23:53,940 --> 00:23:59,460
500 from 1926 to 1997,
you've got something that

499
00:23:59,460 --> 00:24:02,400
also looks kind of random--

500
00:24:02,400 --> 00:24:07,300
so not much predictability here,
not much predictability there.

501
00:24:07,300 --> 00:24:13,960
But suppose we were to graph
the return of General Motors

502
00:24:13,960 --> 00:24:16,460
against the S&P 500.

503
00:24:16,460 --> 00:24:19,450
Well, now, all of a sudden,
it looks like there's

504
00:24:19,450 --> 00:24:20,920
a little bit of a pattern--

505
00:24:20,920 --> 00:24:22,310
not totally random.

506
00:24:22,310 --> 00:24:26,140
Sort of looks like there
is kind of a line that goes

507
00:24:26,140 --> 00:24:30,830
through that scatter of points.

508
00:24:30,830 --> 00:24:33,690
There is a
relationship on a given

509
00:24:33,690 --> 00:24:38,850
day between General Motors
and the broad market index.

510
00:24:38,850 --> 00:24:43,230
But over the course of
two days or two months,

511
00:24:43,230 --> 00:24:46,860
there's very little
predictability.

512
00:24:46,860 --> 00:24:48,960
Now let me talk
about volatility.

513
00:24:48,960 --> 00:24:50,356
Your question?

514
00:24:50,356 --> 00:24:54,080
STUDENT: So on the last graph,
GM is included in there, right?

515
00:24:54,080 --> 00:24:55,080
PROFESSOR: That's right.

516
00:24:55,080 --> 00:24:58,290
GM is one of the
stocks in the S&P 500.

517
00:24:58,290 --> 00:25:01,676
STUDENT: So would you still
see that relationship if it was

518
00:25:01,676 --> 00:25:03,417
everything but GM in the S&P?

519
00:25:03,417 --> 00:25:04,500
PROFESSOR: Oh, absolutely.

520
00:25:04,500 --> 00:25:06,900
In other words, it's
not any one stock

521
00:25:06,900 --> 00:25:10,186
that gives this random
scatter of points.

522
00:25:10,186 --> 00:25:12,060
In other words, this
random scatter of points

523
00:25:12,060 --> 00:25:16,260
is really all 500 stocks
put into a portfolio.

524
00:25:16,260 --> 00:25:18,400
But remember, we're asking
a different question.

525
00:25:18,400 --> 00:25:21,030
This is a question about the
relationship between the S&P

526
00:25:21,030 --> 00:25:24,090
500 last month
versus this month.

527
00:25:24,090 --> 00:25:25,920
There's no real relationship.

528
00:25:25,920 --> 00:25:29,250
This, on the other
hand, is a question

529
00:25:29,250 --> 00:25:32,190
about the relationship
between S&P this month

530
00:25:32,190 --> 00:25:34,650
and GM this month,
the same month.

531
00:25:37,520 --> 00:25:42,140
Now, you're right that S&P has
GM as one of the components,

532
00:25:42,140 --> 00:25:43,940
but it's only one of them.

533
00:25:43,940 --> 00:25:47,000
If it weren't in there, you
would actually still see

534
00:25:47,000 --> 00:25:48,518
this kind of a relationship.

535
00:25:48,518 --> 00:25:49,922
Yeah?

536
00:25:49,922 --> 00:25:52,028
STUDENT: On slide
16, you showed us

537
00:25:52,028 --> 00:25:55,316
the one-year and
10-year interest rate.

538
00:25:55,316 --> 00:25:59,694
What does it mean when the
two figures cross each other?

539
00:25:59,694 --> 00:26:01,610
PROFESSOR: There's no
particular significance.

540
00:26:01,610 --> 00:26:05,150
It just means that that
one-year rate happens

541
00:26:05,150 --> 00:26:07,730
to be identical to
the 10-year rate,

542
00:26:07,730 --> 00:26:10,460
so people are just assuming that
that rate is going to continue

543
00:26:10,460 --> 00:26:12,380
over a period of time.

544
00:26:12,380 --> 00:26:14,390
So there's no particular
economic significance

545
00:26:14,390 --> 00:26:15,830
to when they cross.

546
00:26:15,830 --> 00:26:17,450
These are all
annualized, remember.

547
00:26:17,450 --> 00:26:20,972
So you're asking the question,
over a 10-year period what

548
00:26:20,972 --> 00:26:22,430
is the average
interest rate you're

549
00:26:22,430 --> 00:26:24,410
going to get paid by
the US Government,

550
00:26:24,410 --> 00:26:25,827
versus over a one
year period what

551
00:26:25,827 --> 00:26:28,118
is the interest rate you're
going to get paid by the US

552
00:26:28,118 --> 00:26:28,670
Government?

553
00:26:28,670 --> 00:26:29,170
That's all.

554
00:26:31,760 --> 00:26:34,810
Now let me talk
about volatility.

555
00:26:34,810 --> 00:26:38,340
These are monthly estimates
of US Stock Market

556
00:26:38,340 --> 00:26:40,800
daily volatility
from 1926 to 1997.

557
00:26:40,800 --> 00:26:44,850
So every month I've
got 20 days or 21 days.

558
00:26:44,850 --> 00:26:46,525
I'm going to take
the daily returns

559
00:26:46,525 --> 00:26:49,020
and calculate the
standard deviation

560
00:26:49,020 --> 00:26:53,190
for those daily returns, and I
plot it and it looks like this.

561
00:26:53,190 --> 00:26:55,620
Now, these are monthly, so
if you want to annualize it

562
00:26:55,620 --> 00:26:58,170
you have to multiply it
by the square root of 12

563
00:26:58,170 --> 00:27:00,690
to get the annualized
volatility.

564
00:27:00,690 --> 00:27:04,650
But the point of this is that
there are periods of time where

565
00:27:04,650 --> 00:27:07,200
the market is
extremely volatile,

566
00:27:07,200 --> 00:27:10,390
and there are periods of time
where it's relatively quiet.

567
00:27:10,390 --> 00:27:13,710
And if I would have extended
this to the most recent period,

568
00:27:13,710 --> 00:27:16,890
you would see that within
the last several weeks

569
00:27:16,890 --> 00:27:21,600
we had a spike that was about
as big as this spike over here.

570
00:27:21,600 --> 00:27:23,894
Anybody know what this spike is?

571
00:27:23,894 --> 00:27:24,810
You can sort of guess.

572
00:27:24,810 --> 00:27:27,240
Yeah, October 1987.

573
00:27:27,240 --> 00:27:29,910
That was the big stock
market crash that

574
00:27:29,910 --> 00:27:32,430
occurred during that month.

575
00:27:32,430 --> 00:27:35,490
Volatility shot way up during
that month-- lots of risk.

576
00:27:35,490 --> 00:27:37,680
And right now we're
in a situation where

577
00:27:37,680 --> 00:27:41,550
there is lots of risk as well.

578
00:27:41,550 --> 00:27:44,910
Now, I'm going to just
conclude my overview

579
00:27:44,910 --> 00:27:49,380
with a few interesting anomalies
that I want to tease you with

580
00:27:49,380 --> 00:27:51,120
to get you to think
a little bit more

581
00:27:51,120 --> 00:27:54,150
carefully about stock markets.

582
00:27:54,150 --> 00:27:55,680
What I'm going to
show you are just

583
00:27:55,680 --> 00:27:59,970
a bunch of factoids, factoids
meaning that they are

584
00:27:59,970 --> 00:28:02,110
empirical facts in the data.

585
00:28:02,110 --> 00:28:06,990
But if you change some of
the assumptions or the sample

586
00:28:06,990 --> 00:28:10,320
that you use, these
facts could change.

587
00:28:10,320 --> 00:28:13,140
So these are not
universal constants

588
00:28:13,140 --> 00:28:15,990
that, for some theoretical
reason, has to be true.

589
00:28:15,990 --> 00:28:18,870
This is just
properties of the data.

590
00:28:18,870 --> 00:28:21,390
What I've done here is
to take all the stocks

591
00:28:21,390 --> 00:28:27,420
in the NYSE, Amex, and
NASDAQ from 1964 to 2004.

592
00:28:27,420 --> 00:28:30,420
And on a monthly
basis, I'm going

593
00:28:30,420 --> 00:28:34,080
to rank them by
market capitalization,

594
00:28:34,080 --> 00:28:40,230
break them up into 10 equal
numbers of securities,

595
00:28:40,230 --> 00:28:46,680
and then average the
returns in those deciles

596
00:28:46,680 --> 00:28:49,410
and then compute it
over time and then

597
00:28:49,410 --> 00:28:53,220
average those decile returns.

598
00:28:53,220 --> 00:28:55,260
That sounds a
little complicated,

599
00:28:55,260 --> 00:28:56,850
but pretty straightforward.

600
00:28:56,850 --> 00:28:59,790
Let me tell you, for example,
what this means here.

601
00:28:59,790 --> 00:29:05,250
This particular bin
over here is the bin

602
00:29:05,250 --> 00:29:10,740
of all stocks that
have the largest market

603
00:29:10,740 --> 00:29:17,150
capitalizations among all
of the stocks in my sample.

604
00:29:17,150 --> 00:29:19,020
I've divided them up
into 10 equal groups.

605
00:29:19,020 --> 00:29:21,540
This is the largest
10th, and I'm

606
00:29:21,540 --> 00:29:25,680
going to compute the
returns monthly for that bin

607
00:29:25,680 --> 00:29:28,890
and then average that
across my entire sample.

608
00:29:28,890 --> 00:29:33,390
And I get an annual
return of close to 10%.

609
00:29:33,390 --> 00:29:34,950
This is annualized now.

610
00:29:34,950 --> 00:29:37,080
It's not monthly-- annualized.

611
00:29:40,520 --> 00:29:43,940
If I now compute that
same rate of return

612
00:29:43,940 --> 00:29:47,360
for the smallest
stocks, the stocks

613
00:29:47,360 --> 00:29:50,870
that are the smallest
part of my sample--

614
00:29:50,870 --> 00:29:52,910
that's down here--

615
00:29:52,910 --> 00:29:57,260
I get a return of something
like 15% per year.

616
00:29:59,960 --> 00:30:04,730
15% versus 10%, that's
a huge difference.

617
00:30:04,730 --> 00:30:07,260
That's like a 6 percentage
point difference.

618
00:30:07,260 --> 00:30:12,280
No, it's more like 9%
versus 15%-plus something.

619
00:30:12,280 --> 00:30:15,600
Five percentage points,
500 basis points a year--

620
00:30:15,600 --> 00:30:20,630
that's the gap between small
stocks and large stocks.

621
00:30:20,630 --> 00:30:26,510
This is known as the size
effect or the size premium.

622
00:30:26,510 --> 00:30:30,020
Small stocks seem to do
better than large stocks.

623
00:30:30,020 --> 00:30:32,930
Now, there are lots of stories
you could tell about why.

624
00:30:32,930 --> 00:30:34,580
I'm not going to
tell those stories,

625
00:30:34,580 --> 00:30:37,760
but I just want you
to see the data.

626
00:30:37,760 --> 00:30:40,160
Now, you want to know
what's weird about this?

627
00:30:40,160 --> 00:30:41,660
What's weird about
this picture is

628
00:30:41,660 --> 00:30:46,260
that I'm going to change
the sample ever so slightly.

629
00:30:46,260 --> 00:30:48,710
What I'm going to
do is I'm going

630
00:30:48,710 --> 00:30:55,700
to compute this graph only
for the month of January

631
00:30:55,700 --> 00:30:59,495
and then for all the other
months outside of January.

632
00:31:01,744 --> 00:31:03,035
And I'll show you what happens.

633
00:31:07,120 --> 00:31:13,690
The yellow bars-- those
are the January returns,

634
00:31:13,690 --> 00:31:18,850
and the blue bars are all
of the non-January returns.

635
00:31:18,850 --> 00:31:22,060
So I don't really think there's
that much of a size effect

636
00:31:22,060 --> 00:31:24,610
once you delete the Januarys.

637
00:31:24,610 --> 00:31:27,700
You get a little bit of a
difference between the biggest

638
00:31:27,700 --> 00:31:29,590
and the smallest,
but the difference

639
00:31:29,590 --> 00:31:34,640
that we're talking
about now is very small.

640
00:31:34,640 --> 00:31:37,820
But look at the January effect.

641
00:31:37,820 --> 00:31:39,830
That's big.

642
00:31:39,830 --> 00:31:46,040
And it turns out that this
seems to be a phenomenon that

643
00:31:46,040 --> 00:31:50,120
has become a little bit
less pronounced recently,

644
00:31:50,120 --> 00:31:54,650
but for many years it was
quite strong and reliable.

645
00:31:54,650 --> 00:31:58,760
And people actually traded
on this particular pattern,

646
00:31:58,760 --> 00:32:04,160
buying stocks in January,
holding them in December,

647
00:32:04,160 --> 00:32:08,300
and holding them until January,
or doing a spread where you

648
00:32:08,300 --> 00:32:10,370
basically tried to go long--

649
00:32:10,370 --> 00:32:16,190
small stocks in December
and large stocks in December

650
00:32:16,190 --> 00:32:17,180
and hold that spread.

651
00:32:17,180 --> 00:32:18,620
And then it widened.

652
00:32:18,620 --> 00:32:22,670
So this is a really
puzzling anomaly.

653
00:32:22,670 --> 00:32:25,144
Again, you can come up
with explanations for this.

654
00:32:25,144 --> 00:32:26,810
I'm not going to tell
you what they are.

655
00:32:26,810 --> 00:32:30,920
You can think about them and
possibly even trade on them

656
00:32:30,920 --> 00:32:32,240
over the next month or so.

657
00:32:34,940 --> 00:32:38,630
Now, the second anomaly that
I want you to be aware of--

658
00:32:38,630 --> 00:32:41,270
and this, depending
on who you speak to,

659
00:32:41,270 --> 00:32:43,010
may not be considered
an anomaly.

660
00:32:43,010 --> 00:32:44,480
For example, Warren
Buffett would

661
00:32:44,480 --> 00:32:48,890
call this genius and fact.

662
00:32:48,890 --> 00:32:50,990
This is the value premium.

663
00:32:50,990 --> 00:32:55,130
What this suggests is that
there are certain stocks that,

664
00:32:55,130 --> 00:32:59,150
for whatever reason, are
just simply undervalued

665
00:32:59,150 --> 00:33:01,340
systematically--
or, alternatively,

666
00:33:01,340 --> 00:33:05,720
other stocks that are
systematically overvalued.

667
00:33:05,720 --> 00:33:09,230
The value premium is where
you take the same universe

668
00:33:09,230 --> 00:33:11,330
of stocks that I showed
you before, but instead

669
00:33:11,330 --> 00:33:13,280
of sorting according to
market capitalization,

670
00:33:13,280 --> 00:33:15,810
you sort it according to
another characteristic.

671
00:33:15,810 --> 00:33:19,340
And the characteristic is
the price-equity ratio,

672
00:33:19,340 --> 00:33:22,340
or the reverse of that is
what you usually hear about,

673
00:33:22,340 --> 00:33:23,720
book-to-market--

674
00:33:23,720 --> 00:33:26,610
book value divided
by market value.

675
00:33:26,610 --> 00:33:29,480
Now, you all know what
that difference is by now.

676
00:33:29,480 --> 00:33:33,500
Book value is the
value of the company

677
00:33:33,500 --> 00:33:38,990
as it was initially formed
and as it accrues either cash

678
00:33:38,990 --> 00:33:44,240
or profitability, but based
upon its accounting book value.

679
00:33:44,240 --> 00:33:47,240
Whereas the market value
is what the market thinks

680
00:33:47,240 --> 00:33:48,830
the company is worth.

681
00:33:48,830 --> 00:33:50,990
And in many cases,
with technology stocks

682
00:33:50,990 --> 00:33:53,090
and other growth
stocks, the price

683
00:33:53,090 --> 00:33:58,700
gets way, way ahead of the
value of the company's assets.

684
00:33:58,700 --> 00:34:00,680
And so those are
situations where

685
00:34:00,680 --> 00:34:04,640
you've got a very large
price-to-book ratio.

686
00:34:04,640 --> 00:34:06,944
If you have a very large
price-to-book ratio,

687
00:34:06,944 --> 00:34:08,360
that means that
you're going to be

688
00:34:08,360 --> 00:34:12,080
on the right hand
of this spectrum.

689
00:34:12,080 --> 00:34:14,360
And that says,
according to this chart,

690
00:34:14,360 --> 00:34:18,989
that the expected rate of
return is generally pretty low.

691
00:34:18,989 --> 00:34:21,770
On the other hand, low
book-to-market or high

692
00:34:21,770 --> 00:34:23,750
book-to-price--

693
00:34:23,750 --> 00:34:26,639
sorry, the other way around--

694
00:34:26,639 --> 00:34:29,900
high book-to-market,
low price-to-book

695
00:34:29,900 --> 00:34:32,460
is on the left-hand side.

696
00:34:32,460 --> 00:34:35,420
These are what Warren
Buffett and Graham and Dodd

697
00:34:35,420 --> 00:34:37,040
would call "value stocks."

698
00:34:37,040 --> 00:34:41,270
These are stocks where you've
got lots of good book value,

699
00:34:41,270 --> 00:34:43,850
but somehow the market
doesn't really appreciate it.

700
00:34:43,850 --> 00:34:47,489
And so the price is low
relative to the book value.

701
00:34:47,489 --> 00:34:51,050
In other words, the
price-to-book ratio is low.

702
00:34:51,050 --> 00:34:52,880
And look at the returns there.

703
00:34:52,880 --> 00:34:55,370
The value premium
is the difference

704
00:34:55,370 --> 00:35:00,350
between the high price-to-book
and the low price-to-book,

705
00:35:00,350 --> 00:35:04,610
and you're looking at a premium
of about 600 to 700 basis

706
00:35:04,610 --> 00:35:07,010
points, on an annualized basis.

707
00:35:07,010 --> 00:35:11,840
That's a big difference,
because in both of these bins

708
00:35:11,840 --> 00:35:16,310
it turns out the risks are
actually roughly comparable.

709
00:35:16,310 --> 00:35:18,890
So it's not as if the
stocks on the left

710
00:35:18,890 --> 00:35:21,680
are way more risky than
the stocks on the right.

711
00:35:21,680 --> 00:35:25,190
That actually is true of the
market capitalization effect.

712
00:35:25,190 --> 00:35:27,890
Small stocks actually
have higher volatility

713
00:35:27,890 --> 00:35:29,630
than large stocks.

714
00:35:29,630 --> 00:35:32,480
But that's not true
of value and growth.

715
00:35:32,480 --> 00:35:36,570
So that's another puzzle, or,
depending on who you talk to,

716
00:35:36,570 --> 00:35:39,640
this is a way to invest.

717
00:35:39,640 --> 00:35:44,300
Momentum-- this is something
that academics discovered maybe

718
00:35:44,300 --> 00:35:48,260
10 or 15 years ago, which
is also really anomalous.

719
00:35:48,260 --> 00:35:55,760
By momentum, we mean
simply last year's return.

720
00:35:55,760 --> 00:36:00,470
If that's positive, does it
tend to persist over the next 12

721
00:36:00,470 --> 00:36:01,650
months?

722
00:36:01,650 --> 00:36:07,040
So the stocks with low
momentum on the left-hand side

723
00:36:07,040 --> 00:36:12,590
seem to do worse than the
stocks that have high momentum.

724
00:36:12,590 --> 00:36:15,380
So the momentum effect
seems to be really strong.

725
00:36:15,380 --> 00:36:17,180
And again, look at
that difference.

726
00:36:17,180 --> 00:36:20,510
That difference is something
on the order of a 15% spread,

727
00:36:20,510 --> 00:36:21,770
if not more.

728
00:36:21,770 --> 00:36:24,080
It's a very, very big spread.

729
00:36:24,080 --> 00:36:26,660
So this might lead you to
try to construct a trading

730
00:36:26,660 --> 00:36:28,890
strategy based upon this.

731
00:36:28,890 --> 00:36:30,710
And there are a bunch
of other anomalies

732
00:36:30,710 --> 00:36:33,240
that have been reported in
the academic literature.

733
00:36:33,240 --> 00:36:36,290
In fact, for a while,
certain academic journals

734
00:36:36,290 --> 00:36:39,980
were accused of never meeting an
anomaly that they didn't love,

735
00:36:39,980 --> 00:36:42,980
because they just kept
publishing one after another.

736
00:36:42,980 --> 00:36:48,590
And in a way, you have to be
a bit skeptical about this,

737
00:36:48,590 --> 00:36:51,410
because there are so many
different ways of looking

738
00:36:51,410 --> 00:36:53,660
at stocks, so many
characteristics.

739
00:36:53,660 --> 00:36:59,360
And you know that in a sample
of 100 random variables,

740
00:36:59,360 --> 00:37:05,030
5% of them are going to be
statistically significant, even

741
00:37:05,030 --> 00:37:09,410
if none of them are in
terms of being significantly

742
00:37:09,410 --> 00:37:11,032
different from zero.

743
00:37:11,032 --> 00:37:13,490
So you've got to take these
anomalies with a grain of salt,

744
00:37:13,490 --> 00:37:15,230
but what I've
presented to you are

745
00:37:15,230 --> 00:37:17,000
the ones that seem
to be the most

746
00:37:17,000 --> 00:37:19,640
persistent, the ones that
people spin stories about,

747
00:37:19,640 --> 00:37:22,640
the ones that people
construct mutual funds around.

748
00:37:22,640 --> 00:37:25,580
And so you'll have to think a
little bit about whether or not

749
00:37:25,580 --> 00:37:27,710
you believe any of
these anomalies,

750
00:37:27,710 --> 00:37:29,930
but I wanted to make you
aware that they're there.

751
00:37:29,930 --> 00:37:34,430
And if you take
15433, investments,

752
00:37:34,430 --> 00:37:36,882
you'd actually end up spending
a fair bit of time digging

753
00:37:36,882 --> 00:37:38,840
through each one of these
to see whether or not

754
00:37:38,840 --> 00:37:40,730
there's something in
there that you could

755
00:37:40,730 --> 00:37:42,320
use for investment purposes.

756
00:37:45,850 --> 00:37:49,750
The last thing I want to mention
with this introductory lecture

757
00:37:49,750 --> 00:37:52,620
is mutual funds.

758
00:37:52,620 --> 00:37:58,050
These anomalies were
obviously very, very exciting

759
00:37:58,050 --> 00:38:00,810
from the perspective of
active portfolio management,

760
00:38:00,810 --> 00:38:02,951
because once you identify
one of these anomalies

761
00:38:02,951 --> 00:38:04,950
you could argue, I want
to take advantage of it.

762
00:38:04,950 --> 00:38:07,710
Of course, then the argument
that was raised earlier

763
00:38:07,710 --> 00:38:08,550
comes into play.

764
00:38:08,550 --> 00:38:10,216
If you're going to
take advantage of it,

765
00:38:10,216 --> 00:38:12,160
isn't that going to disappear?

766
00:38:12,160 --> 00:38:15,300
And the answer is, in
general, yes, it will,

767
00:38:15,300 --> 00:38:16,410
but it may take a while.

768
00:38:16,410 --> 00:38:20,010
And along the way,
you'll do quite well.

769
00:38:20,010 --> 00:38:22,920
So the question was
asked, well, if that's

770
00:38:22,920 --> 00:38:25,140
the case, if there are
all these anomalies

771
00:38:25,140 --> 00:38:27,530
and if you can take
advantage of them,

772
00:38:27,530 --> 00:38:30,270
well, then mutual
fund managers ought

773
00:38:30,270 --> 00:38:34,560
to be able to outperform
simple buy and hold strategies.

774
00:38:34,560 --> 00:38:38,880
Because they can take
advantage of these anomalies.

775
00:38:38,880 --> 00:38:44,340
If you do a histogram
of mutual fund returns

776
00:38:44,340 --> 00:38:48,240
that are in excess
of their risks--

777
00:38:48,240 --> 00:38:52,080
so if you make some kind
of a simple risk adjustment

778
00:38:52,080 --> 00:38:55,530
and you look at the mutual
funds' additional value added

779
00:38:55,530 --> 00:38:58,470
above and beyond those
risk adjustments--

780
00:38:58,470 --> 00:39:01,650
those excess returns are
given by this histogram.

781
00:39:01,650 --> 00:39:07,950
For data from 1972 to 1991,
the histogram of excess returns

782
00:39:07,950 --> 00:39:11,700
has basically this
kind of a distribution.

783
00:39:11,700 --> 00:39:15,540
You've got some positives,
you've got some negatives,

784
00:39:15,540 --> 00:39:17,760
you've got more
negatives than positives,

785
00:39:17,760 --> 00:39:22,720
and on average it's
actually less than zero.

786
00:39:22,720 --> 00:39:28,780
Mutual funds net of fees are
actually losing money for you

787
00:39:28,780 --> 00:39:29,440
on average.

788
00:39:29,440 --> 00:39:33,640
That was the conclusion by these
academics as of a few years

789
00:39:33,640 --> 00:39:34,300
ago.

790
00:39:34,300 --> 00:39:35,442
Yeah?

791
00:39:35,442 --> 00:39:38,117
STUDENT: In defense
of the big ones,

792
00:39:38,117 --> 00:39:40,820
isn't part of the
purpose to lower

793
00:39:40,820 --> 00:39:43,850
your risk of your portfolio
to lower the volatility?

794
00:39:43,850 --> 00:39:49,310
For example, you might be able
to get the return plus zero.

795
00:39:49,310 --> 00:39:51,289
But then the next
year you might be

796
00:39:51,289 --> 00:39:52,705
able to get the
return minus five.

797
00:39:52,705 --> 00:39:55,130
Well, I think
mutual fund managers

798
00:39:55,130 --> 00:40:01,360
aim to get you steady terms
that are [? being ?] exactly

799
00:40:01,360 --> 00:40:02,620
the term, plus zero.

800
00:40:02,620 --> 00:40:05,260
PROFESSOR: Well, so first of
all, that's not necessarily

801
00:40:05,260 --> 00:40:06,850
the objective of
every mutual fund

802
00:40:06,850 --> 00:40:09,760
right there are mutual
funds that are not

803
00:40:09,760 --> 00:40:11,260
trying to give you
lower volatility,

804
00:40:11,260 --> 00:40:13,180
but rather they're
trying to give you access

805
00:40:13,180 --> 00:40:16,960
to broader investment
vehicles and instruments.

806
00:40:16,960 --> 00:40:19,570
So the original index
fund that was set up

807
00:40:19,570 --> 00:40:22,210
by Wells Fargo in the 1970s--

808
00:40:22,210 --> 00:40:24,820
the purpose was to
allow an investor

809
00:40:24,820 --> 00:40:29,530
to get access to 100 securities
without having to actually go

810
00:40:29,530 --> 00:40:31,589
out and buy 100 securities.

811
00:40:31,589 --> 00:40:33,130
STUDENT: Thereby
diversifying, right?

812
00:40:33,130 --> 00:40:34,810
PROFESSOR: Right,
by diversifying.

813
00:40:34,810 --> 00:40:37,210
But it doesn't lower
the volatility,

814
00:40:37,210 --> 00:40:39,514
except through diversification.

815
00:40:39,514 --> 00:40:40,180
So you're right.

816
00:40:40,180 --> 00:40:43,630
Diversification will lower
it, versus buying Motorola.

817
00:40:43,630 --> 00:40:45,400
But the question
is, how does this

818
00:40:45,400 --> 00:40:50,860
do versus buying 100
stocks, or rather buying

819
00:40:50,860 --> 00:40:54,790
a mutual fund like
Vanguard, the Vanguard 500

820
00:40:54,790 --> 00:40:56,470
Index Trust, where
you're not trying

821
00:40:56,470 --> 00:41:01,150
to outperform the market you're
trying to match the market?

822
00:41:01,150 --> 00:41:04,120
And the argument is that the
mutual funds that have been

823
00:41:04,120 --> 00:41:09,944
trying to beat the market,
on average they don't.

824
00:41:09,944 --> 00:41:11,150
They don't beat the market.

825
00:41:11,150 --> 00:41:12,650
Some of them do;
some of them don't.

826
00:41:12,650 --> 00:41:14,950
But as a group, they
don't add any extra value.

827
00:41:14,950 --> 00:41:17,180
That's the argument
that was made.

828
00:41:17,180 --> 00:41:19,610
Now, that's not to say that
there aren't good mutual funds

829
00:41:19,610 --> 00:41:21,460
and there aren't
bad mutual funds.

830
00:41:21,460 --> 00:41:22,600
There may be.

831
00:41:22,600 --> 00:41:26,350
So somewhere in here is
Peter Lynch's Magellan Fund--

832
00:41:26,350 --> 00:41:29,800
terrific fund, very
talented portfolio manager.

833
00:41:29,800 --> 00:41:31,600
But on the other
hand, if you can't

834
00:41:31,600 --> 00:41:35,770
tell in advance who is going
to be the next Peter Lynch

835
00:41:35,770 --> 00:41:37,780
and who's going to be the next--

836
00:41:37,780 --> 00:41:41,020
I don't know who; I won't
cast any aspersions.

837
00:41:41,020 --> 00:41:44,110
But if you can't tell in
advance who's going to do bad,

838
00:41:44,110 --> 00:41:48,530
then you're essentially throwing
a dart at this histogram.

839
00:41:48,530 --> 00:41:50,960
You may be lucky and you'll
get on the right side,

840
00:41:50,960 --> 00:41:53,270
or you may be unlucky
and hit the left side.

841
00:41:53,270 --> 00:41:56,270
But on average, you
should do better

842
00:41:56,270 --> 00:41:58,950
by putting your money
in a passive index fund.

843
00:41:58,950 --> 00:42:02,300
Now, that's the
argument of Vanguard

844
00:42:02,300 --> 00:42:06,374
and all of the passive
investment vehicles.

845
00:42:06,374 --> 00:42:08,540
I'm not going to take a
stand on that, because we're

846
00:42:08,540 --> 00:42:10,373
going to come back and
talk a bit about that

847
00:42:10,373 --> 00:42:11,480
at the end of the course.

848
00:42:11,480 --> 00:42:13,105
And then as part of
investments, you're

849
00:42:13,105 --> 00:42:14,400
going to re-look into that.

850
00:42:14,400 --> 00:42:17,390
I just want you to get a feeling
for the data that's out there.

851
00:42:17,390 --> 00:42:18,980
And the data that's
out there says

852
00:42:18,980 --> 00:42:21,350
it's very hard to
tell whether or not

853
00:42:21,350 --> 00:42:26,020
mutual funds, as an aggregate,
are adding any value.

854
00:42:26,020 --> 00:42:28,900
By the way, you realize
that there are actually

855
00:42:28,900 --> 00:42:33,220
more mutual funds out there
than there are stocks.

856
00:42:33,220 --> 00:42:34,730
You know that?

857
00:42:34,730 --> 00:42:37,130
Yeah, there are about
10,000 mutual funds.

858
00:42:37,130 --> 00:42:39,200
There are about 8,000
stocks out there,

859
00:42:39,200 --> 00:42:42,520
including the penny stocks
and pink sheet stocks.

860
00:42:42,520 --> 00:42:44,650
There are probably only
4,000 or 5,000 stocks

861
00:42:44,650 --> 00:42:48,700
that you would actually ever
invest in as a retail investor

862
00:42:48,700 --> 00:42:49,940
yourself.

863
00:42:49,940 --> 00:42:51,880
And so the number
of mutual funds

864
00:42:51,880 --> 00:42:54,910
far exceeds the
number of stocks.

865
00:42:54,910 --> 00:42:57,550
The way that mutual
fund managers

866
00:42:57,550 --> 00:42:59,800
justify that is by
saying, look, there

867
00:42:59,800 --> 00:43:02,800
are 31 flavors of
Baskin Robbins and so we

868
00:43:02,800 --> 00:43:04,600
want to provide
investors with lots

869
00:43:04,600 --> 00:43:06,520
of different possibilities.

870
00:43:06,520 --> 00:43:08,320
Not everybody wants the S&P 500.

871
00:43:08,320 --> 00:43:10,180
Some people want the S&P 100.

872
00:43:10,180 --> 00:43:12,340
Some people want the S&P 250.

873
00:43:12,340 --> 00:43:14,770
Some people want the S&P 385.

874
00:43:14,770 --> 00:43:18,400
And so I'm going to construct a
fund for every clientele that's

875
00:43:18,400 --> 00:43:19,440
out there.

876
00:43:19,440 --> 00:43:21,400
That's a legitimate
argument, as long

877
00:43:21,400 --> 00:43:24,970
as investors understand that
when you buy into a mutual fund

878
00:43:24,970 --> 00:43:28,420
you're buying something that
may cost you more than if you

879
00:43:28,420 --> 00:43:32,710
try to do this on your own.

880
00:43:32,710 --> 00:43:35,140
So the key points that
I want you to take away

881
00:43:35,140 --> 00:43:40,210
is that the average return on
US stocks from 1926 to 2004

882
00:43:40,210 --> 00:43:42,250
was 11.2%.

883
00:43:42,250 --> 00:43:46,330
Now that's considered the
good old days, so no more.

884
00:43:46,330 --> 00:43:49,340
The average risk
premium was about 8%--

885
00:43:49,340 --> 00:43:50,860
again, the good old days.

886
00:43:50,860 --> 00:43:53,530
That's probably not going
to happen for a while.

887
00:43:53,530 --> 00:43:55,900
Stocks are quite risky.

888
00:43:55,900 --> 00:43:59,380
Standard deviation of returns
for the market is about 16%

889
00:43:59,380 --> 00:44:01,150
annually.

890
00:44:01,150 --> 00:44:02,950
That isn't risky anymore.

891
00:44:02,950 --> 00:44:05,200
That, again, is
the good old days.

892
00:44:05,200 --> 00:44:10,930
The market today, using the
implied volatility of the VIX

893
00:44:10,930 --> 00:44:15,070
index, the implied volatility
of S&P at the money options,

894
00:44:15,070 --> 00:44:18,580
was about 49%.

895
00:44:18,580 --> 00:44:24,670
So the annual forward-looking
S&P 500 stock market volatility

896
00:44:24,670 --> 00:44:30,276
right now is about 49%, which
by the way is down from 80%

897
00:44:30,276 --> 00:44:32,210
a couple of weeks ago.

898
00:44:32,210 --> 00:44:35,500
So as I predicted,
volatility was

899
00:44:35,500 --> 00:44:39,130
going to decline once
the election was clear.

900
00:44:39,130 --> 00:44:41,890
That eliminated a
piece of uncertainty,

901
00:44:41,890 --> 00:44:43,480
but there is still
a remaining piece,

902
00:44:43,480 --> 00:44:45,650
which is what's going to
happen to our economy.

903
00:44:45,650 --> 00:44:49,660
That's why the volatility is at
49% versus a historical average

904
00:44:49,660 --> 00:44:51,220
of 16% to 20%.

905
00:44:54,280 --> 00:44:57,790
Stocks on an individual
basis are clearly

906
00:44:57,790 --> 00:45:02,250
much more risky than as a group,
so you're absolutely right,

907
00:45:02,250 --> 00:45:04,450
Remi, that when you
put it into a portfolio

908
00:45:04,450 --> 00:45:06,400
you reduce the risk.

909
00:45:06,400 --> 00:45:11,380
And so the S&P 500 is a lot
less risky than Motorola.

910
00:45:11,380 --> 00:45:15,110
Also, stocks tend to
move together over time.

911
00:45:15,110 --> 00:45:17,270
Over time, from one
day to the next,

912
00:45:17,270 --> 00:45:19,100
there is very
little relationship,

913
00:45:19,100 --> 00:45:23,510
but on a given day stocks
tend to move together

914
00:45:23,510 --> 00:45:26,510
in groups, General
Motors seemingly

915
00:45:26,510 --> 00:45:31,040
to be correlated with the
S&P as well as other stocks.

916
00:45:31,040 --> 00:45:33,470
And obviously, market
volatility changes

917
00:45:33,470 --> 00:45:36,890
over time, and
financial ratios that

918
00:45:36,890 --> 00:45:39,830
can be used to create these
different kinds of bins

919
00:45:39,830 --> 00:45:44,010
for sorting stocks and
constructing these anomalies--

920
00:45:44,010 --> 00:45:48,230
they actually do seem to have
some kind of predictive value.

921
00:45:48,230 --> 00:45:50,720
Why, we don't know
in many cases,

922
00:45:50,720 --> 00:45:52,490
but the anomalies are there.

923
00:45:52,490 --> 00:45:56,720
And so that's something
to be aware of.

924
00:45:56,720 --> 00:45:59,650
Other questions?

925
00:45:59,650 --> 00:46:04,940
Now that you have a
feel for the data,

926
00:46:04,940 --> 00:46:07,740
I want to take a step
back and ask the question,

927
00:46:07,740 --> 00:46:09,830
how do we make use
of this information

928
00:46:09,830 --> 00:46:13,580
in a way that can help
the typical investor

929
00:46:13,580 --> 00:46:15,710
and also help the
individual trying

930
00:46:15,710 --> 00:46:18,350
to decide on a corporate
financial decision?

931
00:46:18,350 --> 00:46:22,490
How do we use these kinds
of empirical insights

932
00:46:22,490 --> 00:46:23,840
in our theory?

933
00:46:23,840 --> 00:46:28,820
So to do that, I'm going to
turn now to lectures 13 and 14

934
00:46:28,820 --> 00:46:36,410
and focus on how to measure
risk and return in a more

935
00:46:36,410 --> 00:46:38,930
systematic way and
then incorporate that

936
00:46:38,930 --> 00:46:41,390
into portfolios.

937
00:46:41,390 --> 00:46:44,270
So we're going to talk
a bit about motivating

938
00:46:44,270 --> 00:46:47,660
the idea behind
portfolio analysis

939
00:46:47,660 --> 00:46:50,810
and then use some of
the theoretical concepts

940
00:46:50,810 --> 00:46:54,110
that we introduced last
time, like mean, variance,

941
00:46:54,110 --> 00:46:58,070
and covariances and use
it to piece together

942
00:46:58,070 --> 00:47:00,960
a good portfolio.

943
00:47:00,960 --> 00:47:05,060
So the motivation for what
we're trying to do now

944
00:47:05,060 --> 00:47:10,070
is to figure out how
to combine securities

945
00:47:10,070 --> 00:47:13,760
into a group that will
have attractive properties.

946
00:47:13,760 --> 00:47:16,790
If you're an investor or a
corporate financial manager,

947
00:47:16,790 --> 00:47:20,220
this is a decision that you've
got to do almost every day.

948
00:47:20,220 --> 00:47:22,700
For example, all of you have
already made that decision

949
00:47:22,700 --> 00:47:24,890
today, whether you
know it or not.

950
00:47:24,890 --> 00:47:27,770
Because if you didn't do
anything between yesterday

951
00:47:27,770 --> 00:47:30,110
and today to rebalance
your portfolio,

952
00:47:30,110 --> 00:47:34,730
you've made an affirmative
decision to let the debt ride.

953
00:47:34,730 --> 00:47:35,480
Let it ride.

954
00:47:35,480 --> 00:47:37,430
You're going to be
taking another roll

955
00:47:37,430 --> 00:47:40,907
of the dice at 4:00
today, and let's hope

956
00:47:40,907 --> 00:47:41,990
it turns out well for you.

957
00:47:41,990 --> 00:47:43,550
But you've made a decision.

958
00:47:43,550 --> 00:47:46,160
Every day you don't do
anything with your portfolio,

959
00:47:46,160 --> 00:47:48,150
you are making a decision.

960
00:47:48,150 --> 00:47:49,730
So what we want to
do is to see if we

961
00:47:49,730 --> 00:47:51,980
can think about making that
decision a little bit more

962
00:47:51,980 --> 00:47:53,330
systematically.

963
00:47:53,330 --> 00:47:56,250
To do that, I want to define
what I mean by a portfolio.

964
00:47:56,250 --> 00:47:59,450
So the definition,
in very simple terms,

965
00:47:59,450 --> 00:48:05,480
is just a specific weighting or
combination of securities, such

966
00:48:05,480 --> 00:48:07,820
that the weights add up to one.

967
00:48:07,820 --> 00:48:10,072
It's just a way to
divide up the pie.

968
00:48:10,072 --> 00:48:11,780
If you've got a certain
amount of wealth,

969
00:48:11,780 --> 00:48:14,660
you're going to allocate it
among different securities

970
00:48:14,660 --> 00:48:16,280
into a portfolio.

971
00:48:16,280 --> 00:48:20,180
A portfolio is defined as a set
of weights of those securities

972
00:48:20,180 --> 00:48:22,730
where the weights add up to one.

973
00:48:22,730 --> 00:48:24,680
Now, this is sort
of the framework

974
00:48:24,680 --> 00:48:27,710
that we use in terms
of the notation.

975
00:48:27,710 --> 00:48:31,280
So omega, the
Greek letter omega,

976
00:48:31,280 --> 00:48:33,800
when you write it
as a vector, it

977
00:48:33,800 --> 00:48:38,100
denotes the set of weights
for your portfolio.

978
00:48:38,100 --> 00:48:41,750
So if you've got n securities,
then this vector omega, which

979
00:48:41,750 --> 00:48:45,210
is equal to omega one,
comma, omega two, dot, dot,

980
00:48:45,210 --> 00:48:49,800
dot, omega n, that
is a portfolio.

981
00:48:49,800 --> 00:48:52,490
And if you want to be clear
about how to define it,

982
00:48:52,490 --> 00:48:55,550
it's simply the number
of shares of security

983
00:48:55,550 --> 00:48:59,480
I multiplied by
that price, divided

984
00:48:59,480 --> 00:49:04,820
by the sum of all of the
values of your securities

985
00:49:04,820 --> 00:49:06,980
in your entire collection.

986
00:49:09,570 --> 00:49:10,670
Any questions about this?

987
00:49:10,670 --> 00:49:12,770
Very basic stuff, but
it's important to get

988
00:49:12,770 --> 00:49:15,820
it right upfront.

989
00:49:15,820 --> 00:49:17,910
Now, by the way,
the number of shares

990
00:49:17,910 --> 00:49:22,940
NI, I'm going to let
that be a real number,

991
00:49:22,940 --> 00:49:28,660
meaning it could be zero, it
could be five, it could be 500,

992
00:49:28,660 --> 00:49:31,300
it could be minus 200.

993
00:49:31,300 --> 00:49:34,780
Minus 200 means that
you have short-sold

994
00:49:34,780 --> 00:49:37,810
200 shares of that security.

995
00:49:37,810 --> 00:49:41,020
So these weights,
they all sum to one,

996
00:49:41,020 --> 00:49:43,480
but they don't have to
be all non-negative.

997
00:49:43,480 --> 00:49:45,370
Some of these weights
could be zero.

998
00:49:45,370 --> 00:49:49,450
Some of these weights
could be negative.

999
00:49:49,450 --> 00:49:52,249
If some of these weights
are negative, then

1000
00:49:52,249 --> 00:49:54,290
what do you know about
some of the other weights?

1001
00:49:57,540 --> 00:49:59,540
Well, yes, they have to
be positive because they

1002
00:49:59,540 --> 00:50:00,456
have to add up to one.

1003
00:50:00,456 --> 00:50:03,647
But tell me more
about the weights.

1004
00:50:03,647 --> 00:50:04,730
STUDENT: Greater than one.

1005
00:50:04,730 --> 00:50:05,660
PROFESSOR: Right.

1006
00:50:05,660 --> 00:50:08,510
Because in order
for it to add up

1007
00:50:08,510 --> 00:50:11,520
to one, if some of these
things are less than zero, then

1008
00:50:11,520 --> 00:50:13,520
some of these other things
are greater than one.

1009
00:50:13,520 --> 00:50:16,490
What does it mean for a
security to be greater than one,

1010
00:50:16,490 --> 00:50:19,687
a weight to be greater than one?

1011
00:50:19,687 --> 00:50:21,094
Does that make sense?

1012
00:50:25,567 --> 00:50:26,650
What's the interpretation?

1013
00:50:26,650 --> 00:50:27,240
Yeah, Lucas?

1014
00:50:27,240 --> 00:50:28,410
STUDENT: Does it mean
you're leveraged?

1015
00:50:28,410 --> 00:50:30,160
PROFESSOR: It means
that you're leveraged,

1016
00:50:30,160 --> 00:50:32,430
that's right-- leveraged
meaning that you're buying

1017
00:50:32,430 --> 00:50:36,390
more than you have money.

1018
00:50:36,390 --> 00:50:38,770
Where are you getting
that money from?

1019
00:50:38,770 --> 00:50:43,644
You're basically getting
a loan, but who's

1020
00:50:43,644 --> 00:50:44,560
loaning you the money?

1021
00:50:47,130 --> 00:50:48,150
Andy?

1022
00:50:48,150 --> 00:50:49,823
STUDENT: Well, the
stocks that you

1023
00:50:49,823 --> 00:50:51,239
sold short gave
you the extra cash

1024
00:50:51,239 --> 00:50:52,650
that you can buy long stock.

1025
00:50:52,650 --> 00:50:53,650
PROFESSOR: That's right.

1026
00:50:53,650 --> 00:50:59,940
So nobody loaned you money, but
somebody loaned you something.

1027
00:50:59,940 --> 00:51:01,680
They loaned you a stock.

1028
00:51:01,680 --> 00:51:05,160
So this simple little
framework already

1029
00:51:05,160 --> 00:51:07,210
has given us one
interesting insight,

1030
00:51:07,210 --> 00:51:12,000
which is that when we short sell
a stock and buy another one,

1031
00:51:12,000 --> 00:51:16,610
we're actually getting
a loan from somebody

1032
00:51:16,610 --> 00:51:19,880
who's lending the stock
to us that we've sold.

1033
00:51:19,880 --> 00:51:23,600
We've taken that cash and we're
putting it into another stock.

1034
00:51:23,600 --> 00:51:27,320
So we're getting a
loan of one security

1035
00:51:27,320 --> 00:51:29,300
and turning around
and using those funds

1036
00:51:29,300 --> 00:51:33,140
to buy another security.

1037
00:51:33,140 --> 00:51:36,740
That's a new transaction
as far as we're concerned,

1038
00:51:36,740 --> 00:51:40,130
but it's one that's going to be
very important in how we think

1039
00:51:40,130 --> 00:51:42,420
about portfolio construction.

1040
00:51:42,420 --> 00:51:45,110
If you didn't understand
that, go back and take a look

1041
00:51:45,110 --> 00:51:48,140
at these nodes and try to
work out a numerical example

1042
00:51:48,140 --> 00:51:49,350
for yourself.

1043
00:51:49,350 --> 00:51:52,040
And if you still don't
understand it, ask again

1044
00:51:52,040 --> 00:51:54,790
next time or ask the
TA during recitation.

1045
00:51:54,790 --> 00:51:58,076
It is a very important point.

1046
00:51:58,076 --> 00:51:59,450
Now, there is a
case that I'm not

1047
00:51:59,450 --> 00:52:00,920
going to talk
about in this class

1048
00:52:00,920 --> 00:52:04,340
where the weights can
actually sum to zero.

1049
00:52:04,340 --> 00:52:07,340
I don't want to talk about
that because that's a much more

1050
00:52:07,340 --> 00:52:11,660
complex situation,
where basically you have

1051
00:52:11,660 --> 00:52:14,040
a portfolio with no money down.

1052
00:52:14,040 --> 00:52:16,250
This is sort of like
the arbitrage portfolios

1053
00:52:16,250 --> 00:52:19,280
that I described to you
in earlier settings.

1054
00:52:19,280 --> 00:52:22,250
We're not going to analyze
that in the context of stocks,

1055
00:52:22,250 --> 00:52:24,710
but there are a
very large number

1056
00:52:24,710 --> 00:52:27,170
of hedge fund strategies
that are based

1057
00:52:27,170 --> 00:52:29,747
upon just these portfolios.

1058
00:52:29,747 --> 00:52:31,580
And so this will be a
very important concept

1059
00:52:31,580 --> 00:52:35,106
that you'll cover in
433, but we're not

1060
00:52:35,106 --> 00:52:36,230
going to talk about it now.

1061
00:52:36,230 --> 00:52:37,813
I just want to make
you aware of that,

1062
00:52:37,813 --> 00:52:41,700
that you can have the weights
summing not to one but actually

1063
00:52:41,700 --> 00:52:42,200
to zero.

1064
00:52:44,850 --> 00:52:47,390
Now, the assumption that
I'm making implicitly

1065
00:52:47,390 --> 00:52:49,670
is that the
portfolio weights are

1066
00:52:49,670 --> 00:52:54,680
summarizing everything there is
to know about your investment.

1067
00:52:54,680 --> 00:52:56,780
So once you know the
portfolio weights

1068
00:52:56,780 --> 00:52:58,640
and you know the
stocks, then you

1069
00:52:58,640 --> 00:53:01,820
know what your
portfolio is about.

1070
00:53:01,820 --> 00:53:06,110
So as an example, you
have an investment account

1071
00:53:06,110 --> 00:53:09,230
with $100,000, and you've
got three stocks in there--

1072
00:53:09,230 --> 00:53:14,240
200 shares of A, 1,000
shares of B, 700 shares of C,

1073
00:53:14,240 --> 00:53:18,410
so that your portfolio is
summarized by the weights

1074
00:53:18,410 --> 00:53:22,220
10%, 60%, and 30%.

1075
00:53:22,220 --> 00:53:25,600
So from now on, we're not
going to worry about prices

1076
00:53:25,600 --> 00:53:28,050
and shares anymore.

1077
00:53:28,050 --> 00:53:31,190
We're going to focus
just on portfolio weights

1078
00:53:31,190 --> 00:53:35,150
and the returns of your
securities multiplied

1079
00:53:35,150 --> 00:53:36,869
by those weights.

1080
00:53:36,869 --> 00:53:37,910
It's just simplification.

1081
00:53:37,910 --> 00:53:39,258
Yeah, Megan?

1082
00:53:39,258 --> 00:53:42,716
STUDENT: I'm just wondering what
you think about a 130-30 type

1083
00:53:42,716 --> 00:53:45,690
portfolio in the context
of the last slide.

1084
00:53:45,690 --> 00:53:46,940
PROFESSOR: Yeah, sure.

1085
00:53:46,940 --> 00:53:48,590
So you're already
asking a question

1086
00:53:48,590 --> 00:53:51,710
that's quite a bit more advanced
than what we're going to cover.

1087
00:53:51,710 --> 00:53:53,635
What's a 130-30 portfolio?

1088
00:53:53,635 --> 00:53:55,010
Can you explain
that, or have you

1089
00:53:55,010 --> 00:53:57,308
just heard that in the news?

1090
00:53:57,308 --> 00:54:00,079
STUDENT: I understand, I
guess, the basic concept,

1091
00:54:00,079 --> 00:54:04,154
which is you're 130%
long and are 30% short,

1092
00:54:04,154 --> 00:54:06,452
so you're headed back to
a net exposure of zero.

1093
00:54:06,452 --> 00:54:07,160
PROFESSOR: Right.

1094
00:54:07,160 --> 00:54:08,180
That's right, exactly.

1095
00:54:08,180 --> 00:54:10,130
So let me describe
a product that's

1096
00:54:10,130 --> 00:54:13,350
out there that's been developed
just over the last few years.

1097
00:54:13,350 --> 00:54:18,860
It's called a 130-30
portfolio, and what 130-30

1098
00:54:18,860 --> 00:54:27,490
stands for is 130%
long and 30% short.

1099
00:54:27,490 --> 00:54:31,030
Now, you can have
a 120-20 portfolio

1100
00:54:31,030 --> 00:54:34,210
or a 180-80 portfolio.

1101
00:54:34,210 --> 00:54:39,160
But the idea is that you have
weights that add up to 100%,

1102
00:54:39,160 --> 00:54:45,220
but the long positions
are no greater than 130%.

1103
00:54:45,220 --> 00:54:50,180
And the short positions
are no less than minus 30%.

1104
00:54:50,180 --> 00:54:52,950
Now, the reason that this is an
interesting product is that--

1105
00:54:52,950 --> 00:54:54,741
I have to give you a
little bit of history.

1106
00:54:54,741 --> 00:54:57,250
This is getting a
little out of our area,

1107
00:54:57,250 --> 00:55:01,650
but I'll give you a
preview of Investments 433.

1108
00:55:01,650 --> 00:55:04,320
Typically, when
institutional investors

1109
00:55:04,320 --> 00:55:07,330
like pension funds or mutual
funds, when they invest,

1110
00:55:07,330 --> 00:55:10,061
they are not allowed
to short sell.

1111
00:55:10,061 --> 00:55:11,560
This has nothing
to do with the SEC.

1112
00:55:11,560 --> 00:55:13,290
It has nothing to do with law.

1113
00:55:13,290 --> 00:55:15,690
It has to do with the
particular entities that

1114
00:55:15,690 --> 00:55:19,065
are investing, because
short selling was viewed way

1115
00:55:19,065 --> 00:55:22,800
back as being a very risky
endeavor because you could lose

1116
00:55:22,800 --> 00:55:24,054
everything and more.

1117
00:55:24,054 --> 00:55:25,470
There's unlimited
amounts that you

1118
00:55:25,470 --> 00:55:28,560
could lose because your short
selling a stock can go way up

1119
00:55:28,560 --> 00:55:31,380
and you could lose
tremendous amounts.

1120
00:55:31,380 --> 00:55:34,500
So mutual funds
were originally not

1121
00:55:34,500 --> 00:55:37,120
allowed to short sell at all.

1122
00:55:37,120 --> 00:55:40,020
So for a mutual fund,
the portfolio weights

1123
00:55:40,020 --> 00:55:42,750
were restricted to
be non-negative,

1124
00:55:42,750 --> 00:55:47,160
and certain pension funds were
not allowed to short sell.

1125
00:55:47,160 --> 00:55:50,610
So if you were a pension
plan or a state university

1126
00:55:50,610 --> 00:55:54,390
and you gave your money
to investment manager xyz,

1127
00:55:54,390 --> 00:55:58,890
that investment manager would
be required not to short sell

1128
00:55:58,890 --> 00:56:02,380
for your portfolio.

1129
00:56:02,380 --> 00:56:05,350
It was discovered over
the last several years

1130
00:56:05,350 --> 00:56:09,580
that this kind of
constraint artificially

1131
00:56:09,580 --> 00:56:12,190
dampened the return
of a portfolio--

1132
00:56:12,190 --> 00:56:14,770
not surprisingly, because
when the market goes down,

1133
00:56:14,770 --> 00:56:18,160
if you're long-only
you're going down with it.

1134
00:56:18,160 --> 00:56:20,200
But if you have a
short position, then

1135
00:56:20,200 --> 00:56:24,040
at least the shorts would
be able to buffer some

1136
00:56:24,040 --> 00:56:25,737
of the losses on the long side.

1137
00:56:25,737 --> 00:56:27,820
So institutions have started
getting more and more

1138
00:56:27,820 --> 00:56:30,032
sophisticated,
thanks to hedge funds

1139
00:56:30,032 --> 00:56:31,990
pushing them into this
area because, of course,

1140
00:56:31,990 --> 00:56:33,310
hedge funds can do anything.

1141
00:56:33,310 --> 00:56:36,580
They can short, they can long,
they can go sideways, whatever.

1142
00:56:36,580 --> 00:56:40,720
So hedge funds led
the pack by saying,

1143
00:56:40,720 --> 00:56:42,520
we're going to
actually short sell

1144
00:56:42,520 --> 00:56:44,440
some of your long-only
portfolio to help

1145
00:56:44,440 --> 00:56:47,910
you get a little bit of
extra return on the downside.

1146
00:56:47,910 --> 00:56:50,440
And so pretty soon,
institutional investors

1147
00:56:50,440 --> 00:56:54,040
said, well, I actually
like the idea of you short

1148
00:56:54,040 --> 00:56:56,702
selling a little bit but I don't
want you to do it too much.

1149
00:56:56,702 --> 00:56:58,660
Because I don't really
know what the risks are.

1150
00:56:58,660 --> 00:56:59,830
I'm new to this.

1151
00:56:59,830 --> 00:57:02,830
I don't want to get the
risks out of control,

1152
00:57:02,830 --> 00:57:05,770
so I'm going to limit how
much you can short sell.

1153
00:57:05,770 --> 00:57:08,320
The limit of how much
you can short sell

1154
00:57:08,320 --> 00:57:14,230
imposes a limit on how much
more long you can go than 100%.

1155
00:57:14,230 --> 00:57:16,315
Just like we said, if
something's negative,

1156
00:57:16,315 --> 00:57:18,690
then some of the weights have
got to be greater than one,

1157
00:57:18,690 --> 00:57:20,710
or they got to add up
to greater than one.

1158
00:57:20,710 --> 00:57:23,770
So when you have
that situation where

1159
00:57:23,770 --> 00:57:28,450
you have a limit on the total
negative position you can have,

1160
00:57:28,450 --> 00:57:32,830
that limit puts a symmetric
upper bound beyond the number

1161
00:57:32,830 --> 00:57:36,250
one of what you can go long.

1162
00:57:36,250 --> 00:57:40,390
And so for reasons that are
probably a little bit too far

1163
00:57:40,390 --> 00:57:43,420
afield to get into
here, 130-30 seems

1164
00:57:43,420 --> 00:57:46,510
to be a bit of a sweet spot
for managers out there.

1165
00:57:46,510 --> 00:57:50,170
So they said, we will
limit our short positions

1166
00:57:50,170 --> 00:57:53,200
to no more than
30% of the capital.

1167
00:57:53,200 --> 00:57:55,930
So you give us $100
million to manage,

1168
00:57:55,930 --> 00:57:59,980
we will take no more than
$30 million of shorts.

1169
00:57:59,980 --> 00:58:03,580
And therefore, we will take
no more than $130 million

1170
00:58:03,580 --> 00:58:04,840
of longs.

1171
00:58:04,840 --> 00:58:08,540
But when you add them up,
you still get back to 100%.

1172
00:58:08,540 --> 00:58:09,370
So that's 130-30.

1173
00:58:09,370 --> 00:58:11,650
It's very popular,
and it's something

1174
00:58:11,650 --> 00:58:14,230
that is likely to grow,
particularly given

1175
00:58:14,230 --> 00:58:15,580
this current market environment.

1176
00:58:15,580 --> 00:58:20,390
Because 130-30 has done better
than the S&P, not surprisingly,

1177
00:58:20,390 --> 00:58:24,780
because of that
30% short position.

1178
00:58:24,780 --> 00:58:28,560
So here's the example
of your own portfolio

1179
00:58:28,560 --> 00:58:29,880
and how you get those weights.

1180
00:58:29,880 --> 00:58:32,500
That's pretty straightforward.

1181
00:58:32,500 --> 00:58:35,130
Here's another example
where you've now

1182
00:58:35,130 --> 00:58:39,210
got some short positions
here, and the short positions

1183
00:58:39,210 --> 00:58:40,920
are not from short selling.

1184
00:58:40,920 --> 00:58:43,710
But the short positions
are in riskless bonds.

1185
00:58:43,710 --> 00:58:47,320
In other words, now instead
of shorting a stock,

1186
00:58:47,320 --> 00:58:50,860
you're shorting a bond or
selling a bond or borrowing

1187
00:58:50,860 --> 00:58:54,770
money from a broker
and getting leverage.

1188
00:58:54,770 --> 00:58:57,430
So this is leverage where
the security that you're

1189
00:58:57,430 --> 00:59:01,150
levering up is using bonds,
and you're levering up

1190
00:59:01,150 --> 00:59:02,890
the equity positions.

1191
00:59:02,890 --> 00:59:05,920
So your portfolio
weights look like this.

1192
00:59:05,920 --> 00:59:09,580
Your equity positions,
when you add up the equity

1193
00:59:09,580 --> 00:59:14,840
positions, those portfolio
weights, you get 200%.

1194
00:59:14,840 --> 00:59:20,410
But your riskless bonds,
you shorted $50,000.

1195
00:59:20,410 --> 00:59:23,960
You're borrowing
$50,000 from the broker,

1196
00:59:23,960 --> 00:59:25,930
so you've got minus 100%.

1197
00:59:25,930 --> 00:59:30,080
When you add those two, you
get back 100%, or in this case,

1198
00:59:30,080 --> 00:59:31,270
$50,000.

1199
00:59:31,270 --> 00:59:34,450
So you start out
with $50,000 cash

1200
00:59:34,450 --> 00:59:37,900
and then you buy
$100,000 worth of stocks

1201
00:59:37,900 --> 00:59:42,590
by borrowing an extra
$50,000 from your broker.

1202
00:59:42,590 --> 00:59:43,315
Yeah?

1203
00:59:43,315 --> 00:59:44,898
STUDENT: I understand
the [? degree ?]

1204
00:59:44,898 --> 00:59:47,752
of the portfolio [INAUDIBLE],
but I thought, fundamentally,

1205
00:59:47,752 --> 00:59:50,670
it's to have different
risks and to manage that.

1206
00:59:50,670 --> 00:59:51,330
PROFESSOR: Yes.

1207
00:59:51,330 --> 00:59:52,704
STUDENT: So when
you are actually

1208
00:59:52,704 --> 00:59:55,574
structuring portfolios, do you
have these risks [INAUDIBLE]?

1209
00:59:55,574 --> 00:59:58,919
Because I would just be
willing to do them separately.

1210
00:59:58,919 --> 01:00:01,210
There is no reason for you
to have them as [INAUDIBLE].

1211
01:00:01,210 --> 01:00:02,470
PROFESSOR: Well, the
reason that I have them

1212
01:00:02,470 --> 01:00:04,480
here is I want to show
you exactly how you

1213
01:00:04,480 --> 01:00:06,700
would compute the
portfolio weights

1214
01:00:06,700 --> 01:00:09,080
of your entire portfolio.

1215
01:00:09,080 --> 01:00:12,730
So basically, what this is
is your equity portfolio,

1216
01:00:12,730 --> 01:00:17,552
but in addition, your fixed
income position's added in.

1217
01:00:17,552 --> 01:00:19,510
I mean, your whole
portfolio could be anything.

1218
01:00:19,510 --> 01:00:22,530
It could be stocks, bonds,
options, currencies,

1219
01:00:22,530 --> 01:00:24,270
real state.

1220
01:00:24,270 --> 01:00:27,915
So I'm just including all of
this in the portfolio itself.

1221
01:00:27,915 --> 01:00:31,875
STUDENT: So basically, it is to
try to handle different risks--

1222
01:00:31,875 --> 01:00:34,850
I mean, different stocks or
bonds or whatever they have,

1223
01:00:34,850 --> 01:00:35,590
various risks?

1224
01:00:35,590 --> 01:00:37,850
PROFESSOR: Yes, absolutely,
and you are doing that.

1225
01:00:37,850 --> 01:00:39,710
Stocks A, B, and C
have different risks,

1226
01:00:39,710 --> 01:00:41,467
as does the bond.

1227
01:00:41,467 --> 01:00:43,550
And so you're mixing and
matching and putting them

1228
01:00:43,550 --> 01:00:45,050
together into what
hopefully will

1229
01:00:45,050 --> 01:00:46,340
be an attractive portfolio.

1230
01:00:49,150 --> 01:00:53,200
Now, we mentioned before
that when you get a mortgage,

1231
01:00:53,200 --> 01:00:54,910
that's leverage, too.

1232
01:00:54,910 --> 01:00:58,990
So this is an example
of a situation where

1233
01:00:58,990 --> 01:01:02,350
you buy a home for
$500,000, but you only

1234
01:01:02,350 --> 01:01:06,300
have a $100,000 payment.

1235
01:01:06,300 --> 01:01:09,360
Your equity in the
home is only $100,000.

1236
01:01:09,360 --> 01:01:12,150
The bank has loaned
you $400,000.

1237
01:01:12,150 --> 01:01:15,480
Your leverage ratio
is 5:1, so if you

1238
01:01:15,480 --> 01:01:17,250
were to look at your
portfolio weights

1239
01:01:17,250 --> 01:01:26,830
it would be 500% house minus
400% bank or bonds or mortgage.

1240
01:01:26,830 --> 01:01:29,080
That's very high leverage.

1241
01:01:29,080 --> 01:01:33,190
And in that case, when
you're leveraged five to one,

1242
01:01:33,190 --> 01:01:37,900
if the house price goes down by
something like, I don't know,

1243
01:01:37,900 --> 01:01:44,905
2%, you've lost 10% of
the value of your home--

1244
01:01:44,905 --> 01:01:47,520
or the value of
your equity, rather.

1245
01:01:47,520 --> 01:01:54,030
If the house price declines by
15%, that's really bad news.

1246
01:01:54,030 --> 01:01:56,640
So leverage is a
two-edged sword.

1247
01:01:56,640 --> 01:01:59,310
When things are working
well, it gives you a boost.

1248
01:01:59,310 --> 01:02:00,990
When things are
not working well,

1249
01:02:00,990 --> 01:02:03,730
it can hurt you on
the downside as well.

1250
01:02:06,510 --> 01:02:09,110
Here's another example
where you've got

1251
01:02:09,110 --> 01:02:11,660
a zero net investment strategy.

1252
01:02:11,660 --> 01:02:13,250
You can work that
out for yourself.

1253
01:02:13,250 --> 01:02:15,710
This is a little bit trickier
because the portfolio weights

1254
01:02:15,710 --> 01:02:16,901
now add up to zero.

1255
01:02:16,901 --> 01:02:19,400
You've got to think a little
bit about what it means to have

1256
01:02:19,400 --> 01:02:21,177
portfolio weights at all.

1257
01:02:21,177 --> 01:02:22,760
So I'll leave that
for you to look at.

1258
01:02:22,760 --> 01:02:24,135
That's something
that, as I said,

1259
01:02:24,135 --> 01:02:27,650
we won't cover this
course in great detail.

1260
01:02:27,650 --> 01:02:29,750
So now motivation--
what we're trying

1261
01:02:29,750 --> 01:02:32,990
to do now that you know the
basic language of portfolio

1262
01:02:32,990 --> 01:02:36,350
weights and how to manipulate
them to some degree,

1263
01:02:36,350 --> 01:02:40,520
I want to ask the question,
why bother with the portfolio?

1264
01:02:40,520 --> 01:02:42,380
Well, we've already got
a couple of comments

1265
01:02:42,380 --> 01:02:43,790
about why you want a portfolio.

1266
01:02:43,790 --> 01:02:46,310
You want to have stocks with
different kinds of risks

1267
01:02:46,310 --> 01:02:49,000
so you have diversification.

1268
01:02:49,000 --> 01:02:51,970
But there's another approach,
and the other approach

1269
01:02:51,970 --> 01:02:55,720
is championed by Warren Buffett.

1270
01:02:55,720 --> 01:02:59,330
Warren Buffett has criticized
this idea of diversification,

1271
01:02:59,330 --> 01:03:01,780
not putting all your eggs
in one basket, by saying,

1272
01:03:01,780 --> 01:03:03,880
you should put all
your eggs in one basket

1273
01:03:03,880 --> 01:03:06,670
and then simply just watch
that basket really carefully.

1274
01:03:06,670 --> 01:03:08,020
Isn't that better?

1275
01:03:08,020 --> 01:03:11,290
Well, that sounds
good, but what if it's

1276
01:03:11,290 --> 01:03:12,910
the case that you
don't really know

1277
01:03:12,910 --> 01:03:15,540
how to pick the right basket?

1278
01:03:15,540 --> 01:03:18,460
And so therefore, whatever
basket you're watching

1279
01:03:18,460 --> 01:03:20,470
may not be
particularly attractive

1280
01:03:20,470 --> 01:03:22,750
because you picked
the wrong basket.

1281
01:03:22,750 --> 01:03:26,290
So that's really the idea
behind portfolio theory.

1282
01:03:26,290 --> 01:03:28,390
It's that not all of
us are Warren Buffett.

1283
01:03:28,390 --> 01:03:31,000
Not all of us want to
become Warren Buffetts.

1284
01:03:31,000 --> 01:03:35,890
We want to have a relatively
systematic approach to making

1285
01:03:35,890 --> 01:03:37,270
a good investment decision.

1286
01:03:37,270 --> 01:03:39,100
We don't want to try
to beat the market.

1287
01:03:39,100 --> 01:03:41,110
We want to figure out
whether we can come up

1288
01:03:41,110 --> 01:03:45,640
with a responsible and
attractive way of investing

1289
01:03:45,640 --> 01:03:49,000
that has some kind of
economic logic to it.

1290
01:03:49,000 --> 01:03:52,160
So the point is that we don't
know which stock is best,

1291
01:03:52,160 --> 01:03:56,260
and so we don't want to pick
just one stock like Motorola.

1292
01:03:56,260 --> 01:03:58,900
Because there are periods
where Motorola looks fantastic

1293
01:03:58,900 --> 01:04:01,180
and periods where
Motorola looks horrible.

1294
01:04:01,180 --> 01:04:03,760
So we want to be able to
pick a portfolio that's

1295
01:04:03,760 --> 01:04:05,690
got good characteristics.

1296
01:04:05,690 --> 01:04:09,530
So diversification is
one way to do that.

1297
01:04:09,530 --> 01:04:12,490
It's to basically
spread your risk

1298
01:04:12,490 --> 01:04:19,570
across a number of securities,
and portfolios can do that.

1299
01:04:19,570 --> 01:04:24,930
But at the same time, they
can also create focused bets.

1300
01:04:24,930 --> 01:04:27,810
So it's not just
the case that you

1301
01:04:27,810 --> 01:04:31,830
have to buy every possible stock
there is out there in order

1302
01:04:31,830 --> 01:04:33,090
to diversify.

1303
01:04:33,090 --> 01:04:35,850
For example, you
may have information

1304
01:04:35,850 --> 01:04:40,110
or you may have conviction that
information technology is going

1305
01:04:40,110 --> 01:04:42,390
to do really well over
the next couple of years

1306
01:04:42,390 --> 01:04:43,950
because somebody's
got to figure out

1307
01:04:43,950 --> 01:04:48,360
how to process all of these
bad loans and problem banks.

1308
01:04:48,360 --> 01:04:52,080
And IT is going to
ultimately be the solution.

1309
01:04:52,080 --> 01:04:56,010
Well, if that's the case,
you can make a bet on IT

1310
01:04:56,010 --> 01:05:00,270
without having to make a bet
on any one firm or one stock.

1311
01:05:00,270 --> 01:05:03,540
The way you do that is to
form a portfolio of stocks

1312
01:05:03,540 --> 01:05:06,040
that are all in the IT sector.

1313
01:05:06,040 --> 01:05:09,040
And so you get diversification,
but at the same time,

1314
01:05:09,040 --> 01:05:10,590
you're able to make
a bet in an area

1315
01:05:10,590 --> 01:05:14,310
where you think you have
particular expertise.

1316
01:05:14,310 --> 01:05:18,810
And finally, portfolios
can customize and manage

1317
01:05:18,810 --> 01:05:20,760
your own personal
risk-reward tradeoffs.

1318
01:05:20,760 --> 01:05:23,670
So for some of you,
you want a lot of risk,

1319
01:05:23,670 --> 01:05:27,000
you want it concentrated in
a small number of industries,

1320
01:05:27,000 --> 01:05:31,530
and you want to do it with
relatively small priced stocks.

1321
01:05:31,530 --> 01:05:32,550
You can do that.

1322
01:05:32,550 --> 01:05:35,400
Somebody else might have a very
different set of preferences.

1323
01:05:35,400 --> 01:05:39,960
Portfolios allow you to tailor
the risk-reward tradeoffs

1324
01:05:39,960 --> 01:05:42,630
to your particular preferences.

1325
01:05:42,630 --> 01:05:45,210
So now we have a
motivation for portfolios.

1326
01:05:45,210 --> 01:05:48,780
Then the next question is,
that sounds great; now tell me,

1327
01:05:48,780 --> 01:05:52,410
how do I construct one
of these good portfolios?

1328
01:05:52,410 --> 01:05:54,436
And in order to
answer that question,

1329
01:05:54,436 --> 01:05:56,060
I've got to tell you
what "good" means,

1330
01:05:56,060 --> 01:05:58,200
or you've got to tell
me what "good" means.

1331
01:05:58,200 --> 01:06:03,090
So typically, what we say
about a good portfolio

1332
01:06:03,090 --> 01:06:11,400
is it's a portfolio that
has high mean and low risk.

1333
01:06:11,400 --> 01:06:13,080
That's what "good" means.

1334
01:06:13,080 --> 01:06:16,560
There are two characteristics
that we tend to focus on

1335
01:06:16,560 --> 01:06:19,530
for purely statistical reasons.

1336
01:06:19,530 --> 01:06:22,060
It's because those
are easy to compute,

1337
01:06:22,060 --> 01:06:24,270
and they are the
first two statistics

1338
01:06:24,270 --> 01:06:25,740
that one would
look at when you're

1339
01:06:25,740 --> 01:06:27,900
looking at an
investment-- the mean

1340
01:06:27,900 --> 01:06:29,920
and the standard deviation.

1341
01:06:29,920 --> 01:06:32,670
So you might think that,
naturally, it would make sense

1342
01:06:32,670 --> 01:06:35,430
to pick a portfolio
that's got high mean

1343
01:06:35,430 --> 01:06:37,770
and low standard deviation.

1344
01:06:37,770 --> 01:06:39,162
That's an assumption.

1345
01:06:39,162 --> 01:06:40,870
In other words, we're
assuming that we're

1346
01:06:40,870 --> 01:06:44,100
going to measure risk
by standard deviation,

1347
01:06:44,100 --> 01:06:46,600
and we're assuming that
we're measuring return

1348
01:06:46,600 --> 01:06:49,390
by the actual expected
rate of return.

1349
01:06:49,390 --> 01:06:51,740
For certain investors,
those are not appropriate.

1350
01:06:51,740 --> 01:06:53,281
For example, there
are some investors

1351
01:06:53,281 --> 01:06:58,542
that are really keen on
socially aware investing

1352
01:06:58,542 --> 01:07:00,250
so they don't want to
invest in companies

1353
01:07:00,250 --> 01:07:02,027
that pollute the environment.

1354
01:07:02,027 --> 01:07:03,610
They don't want to
invest in companies

1355
01:07:03,610 --> 01:07:08,200
that engage in nonunion
workers, or they

1356
01:07:08,200 --> 01:07:09,790
don't want to
invest in companies

1357
01:07:09,790 --> 01:07:15,590
that happen to be exploiting
labor in unregulated markets.

1358
01:07:15,590 --> 01:07:18,460
Those are examples of
non-pecuniary characteristics

1359
01:07:18,460 --> 01:07:21,190
that figure into this
choice of stocks.

1360
01:07:21,190 --> 01:07:22,910
We're going to
abstract from that.

1361
01:07:22,910 --> 01:07:25,660
So for our purposes, the
characteristics that we're

1362
01:07:25,660 --> 01:07:27,790
going to look at
for a good portfolio

1363
01:07:27,790 --> 01:07:31,840
is, does it have a high
return, does it have low risk?

1364
01:07:31,840 --> 01:07:33,910
And the way we're
going to measure risk

1365
01:07:33,910 --> 01:07:38,750
is in terms of the volatility
or standard deviation.

1366
01:07:38,750 --> 01:07:42,820
Now, here, again, there's lots
of ways of measuring risk.

1367
01:07:42,820 --> 01:07:47,820
We can measure by the
upper quartile, the 5% loss

1368
01:07:47,820 --> 01:07:51,370
or spread, but in fact,
what we're going to use

1369
01:07:51,370 --> 01:07:53,890
is this standard
deviation measure.

1370
01:07:53,890 --> 01:07:56,080
For symmetric distributions
like the normal,

1371
01:07:56,080 --> 01:07:59,800
it turns out that that's
not a bad measure,

1372
01:07:59,800 --> 01:08:03,310
but some people have argued
that by looking at spread

1373
01:08:03,310 --> 01:08:07,240
you're confusing the
upside with the downside.

1374
01:08:07,240 --> 01:08:11,780
Nobody has any problem
with upside risk or upside

1375
01:08:11,780 --> 01:08:15,217
distribution.

1376
01:08:15,217 --> 01:08:17,050
I haven't run across
anybody that said, gee,

1377
01:08:17,050 --> 01:08:19,340
this year I'm really
making too much money

1378
01:08:19,340 --> 01:08:20,979
and that's just
not a good thing.

1379
01:08:20,979 --> 01:08:23,290
If you meet anybody
like that, introduce me.

1380
01:08:23,290 --> 01:08:26,200
I'll help them out
with their problem.

1381
01:08:26,200 --> 01:08:29,560
But the point is that for
a symmetric distribution,

1382
01:08:29,560 --> 01:08:34,450
it doesn't matter, and in
more advanced approaches

1383
01:08:34,450 --> 01:08:38,080
to investments people have
used one-sided measures.

1384
01:08:38,080 --> 01:08:40,326
But we're not going to
do that in this course.

1385
01:08:40,326 --> 01:08:42,700
So we're going to focus on
variance or standard deviation

1386
01:08:42,700 --> 01:08:46,160
as the measure of risk.

1387
01:08:46,160 --> 01:08:48,080
And the assumptions
that I'm going

1388
01:08:48,080 --> 01:08:50,990
to make for the
remainder of the course

1389
01:08:50,990 --> 01:08:59,760
is that all investors like
higher mean and all investors

1390
01:08:59,760 --> 01:09:03,840
like higher variance.

1391
01:09:03,840 --> 01:09:08,040
Now, that's a really
reasonable assumption,

1392
01:09:08,040 --> 01:09:10,800
but you could
challenge it if you

1393
01:09:10,800 --> 01:09:14,609
wanted to argue that investors
care about other things.

1394
01:09:14,609 --> 01:09:18,120
So just be aware that I'm
making an approximation,

1395
01:09:18,120 --> 01:09:21,479
and the approximation is exactly
that, that mean and variance

1396
01:09:21,479 --> 01:09:25,710
are the only things that
our prototypical investor is

1397
01:09:25,710 --> 01:09:28,670
going to care about.

1398
01:09:28,670 --> 01:09:31,280
So now, we actually are
pretty close to being

1399
01:09:31,280 --> 01:09:33,529
able to come up with an
answer to the question, what's

1400
01:09:33,529 --> 01:09:37,818
a good portfolio and
how do we pick stocks.

1401
01:09:37,818 --> 01:09:39,859
One of the things that
we're going to answer over

1402
01:09:39,859 --> 01:09:41,240
the course of the
next few slides

1403
01:09:41,240 --> 01:09:45,620
is, how much does
a stock contribute

1404
01:09:45,620 --> 01:09:50,460
to the risk and the expected
return of a portfolio?

1405
01:09:50,460 --> 01:09:53,750
So if you're thinking about
investing in a new stock,

1406
01:09:53,750 --> 01:09:58,682
it's like inviting
somebody into your club.

1407
01:09:58,682 --> 01:10:00,140
You want to ask,
well, what are you

1408
01:10:00,140 --> 01:10:01,717
going to contribute to my club?

1409
01:10:01,717 --> 01:10:03,800
What are you going to
contribute to the portfolio?

1410
01:10:03,800 --> 01:10:05,990
What will you add to
what I already have?

1411
01:10:05,990 --> 01:10:08,480
Are you going to help me
with my expected return?

1412
01:10:08,480 --> 01:10:10,910
Are you going to help
me lower my risk?

1413
01:10:10,910 --> 01:10:13,556
And if the answer is no to both
of them, then I don't want you.

1414
01:10:13,556 --> 01:10:15,180
You're not going to
do anything for me.

1415
01:10:15,180 --> 01:10:16,760
Why should you be
in my portfolio?

1416
01:10:16,760 --> 01:10:18,260
So that's the kind
of argument we're

1417
01:10:18,260 --> 01:10:22,940
going to make to be able to
construct a good portfolio.

1418
01:10:22,940 --> 01:10:26,550
So let's get a little bit
more specific about that.

1419
01:10:26,550 --> 01:10:29,930
Here's a graph, and you're
going to see this a lot.

1420
01:10:29,930 --> 01:10:32,390
This graph is going
to be one that we use

1421
01:10:32,390 --> 01:10:34,810
for all of portfolio analysis.

1422
01:10:34,810 --> 01:10:38,240
It's where we plot on
two-dimensional space

1423
01:10:38,240 --> 01:10:41,780
the average return
of the stock as well

1424
01:10:41,780 --> 01:10:44,540
as its risk, where
risk is now being

1425
01:10:44,540 --> 01:10:47,720
measured by standard deviation.

1426
01:10:47,720 --> 01:10:51,500
So I've got five
assets plotted here.

1427
01:10:51,500 --> 01:10:55,070
Merck is one and General
Motors is another one.

1428
01:10:55,070 --> 01:10:57,440
Motorola is a third,
McDonald's a fourth,

1429
01:10:57,440 --> 01:11:01,600
and I've got T-bills down
there on the lower left.

1430
01:11:01,600 --> 01:11:06,130
This gives you a sense of the
different trade-offs there are.

1431
01:11:06,130 --> 01:11:11,350
Clearly General Motors is
lower risk that Motorola,

1432
01:11:11,350 --> 01:11:13,920
but it's also lower return.

1433
01:11:13,920 --> 01:11:17,220
And McDonald's is
definitely going

1434
01:11:17,220 --> 01:11:24,960
to be higher risk than Merck,
but notice that McDonald's is

1435
01:11:24,960 --> 01:11:28,450
also lower return than Merck.

1436
01:11:28,450 --> 01:11:34,410
So at least in this setting,
nobody in their right mind--

1437
01:11:34,410 --> 01:11:36,480
by that I mean, no
rational investor--

1438
01:11:36,480 --> 01:11:41,130
would ever want to hold
McDonald's over Merck--

1439
01:11:41,130 --> 01:11:44,160
by our assumption.

1440
01:11:44,160 --> 01:11:50,250
We're assuming that investors
like expected return

1441
01:11:50,250 --> 01:11:53,160
and they don't like risk.

1442
01:11:53,160 --> 01:11:53,660
Question?

1443
01:11:53,660 --> 01:11:54,508
Yeah?

1444
01:11:54,508 --> 01:11:57,992
STUDENT: This [INAUDIBLE]
think that McDonald's will

1445
01:11:57,992 --> 01:11:58,617
perform better.

1446
01:11:58,617 --> 01:11:59,992
PROFESSOR: Exactly,
that's right.

1447
01:11:59,992 --> 01:12:01,926
So I was waiting for
somebody to say that.

1448
01:12:01,926 --> 01:12:03,800
Warren Buffett would
say that's the stupidest

1449
01:12:03,800 --> 01:12:07,790
thing I've ever heard,
because all you're doing

1450
01:12:07,790 --> 01:12:10,850
is plotting history
on this chart.

1451
01:12:10,850 --> 01:12:13,610
And this tells you nothing
about what might happen

1452
01:12:13,610 --> 01:12:16,610
over the next 12, 24 months.

1453
01:12:16,610 --> 01:12:20,090
It could be that health
care is going to just become

1454
01:12:20,090 --> 01:12:22,430
a real problem,
pharmaceutical companies

1455
01:12:22,430 --> 01:12:24,080
are going to get
battered because

1456
01:12:24,080 --> 01:12:25,730
of the Democratic
administration.

1457
01:12:25,730 --> 01:12:28,400
That's going to force
them to reduce the prices.

1458
01:12:28,400 --> 01:12:30,185
And so over the next
six to 12 months,

1459
01:12:30,185 --> 01:12:32,060
the only thing that
people will be able to do

1460
01:12:32,060 --> 01:12:35,300
is to go to their neighborhood
McDonald's and just enjoy

1461
01:12:35,300 --> 01:12:37,760
a nice hamburger and
complain about what's

1462
01:12:37,760 --> 01:12:40,940
been going on with the
pharmaceutical industry.

1463
01:12:40,940 --> 01:12:42,890
In that case, McDonald's
is a great bet

1464
01:12:42,890 --> 01:12:45,740
and Merck is a
terrible investment.

1465
01:12:45,740 --> 01:12:47,740
We abstract from all of that.

1466
01:12:47,740 --> 01:12:50,930
We are not in the business
of forecasting stock returns.

1467
01:12:50,930 --> 01:12:51,530
Why?

1468
01:12:51,530 --> 01:12:53,660
Because I just showed you
in the previous set of

1469
01:12:53,660 --> 01:12:56,360
slides that it's
hard to forecast.

1470
01:12:56,360 --> 01:12:58,970
In fact, you told me that
in an efficient market

1471
01:12:58,970 --> 01:13:00,710
it's actually hard
to tell what's going

1472
01:13:00,710 --> 01:13:01,990
to happen with these stocks.

1473
01:13:01,990 --> 01:13:03,500
And if you could
tell, then people

1474
01:13:03,500 --> 01:13:05,490
are going to start
using that information,

1475
01:13:05,490 --> 01:13:07,370
and then the
information is worthless

1476
01:13:07,370 --> 01:13:10,190
because it'll have already
been taken into account.

1477
01:13:10,190 --> 01:13:13,750
So you see, this
is a very important

1478
01:13:13,750 --> 01:13:17,950
philosophical difference between
Warren Buffett and academics.

1479
01:13:17,950 --> 01:13:19,840
Warren Buffett
believes that there

1480
01:13:19,840 --> 01:13:23,350
are systematic mispricings
out there that can be

1481
01:13:23,350 --> 01:13:26,400
found and taken advantage of.

1482
01:13:26,400 --> 01:13:29,400
Academic finance, as
of the 1960s and 70s

1483
01:13:29,400 --> 01:13:32,380
when this theory was developed,
started from the point

1484
01:13:32,380 --> 01:13:35,140
that you just came
to very quickly,

1485
01:13:35,140 --> 01:13:38,200
which is that there are
no patterns in the data.

1486
01:13:38,200 --> 01:13:41,660
If there were, someone
would have already done it--

1487
01:13:41,660 --> 01:13:44,390
which, by the way, Warren
Buffett would answer by saying,

1488
01:13:44,390 --> 01:13:46,850
you know what, that sounds
like a joke about the economist

1489
01:13:46,850 --> 01:13:49,460
walking down the road,
sees a $100 bill,

1490
01:13:49,460 --> 01:13:50,377
and walks right by it.

1491
01:13:50,377 --> 01:13:51,959
And when somebody
says, why didn't you

1492
01:13:51,959 --> 01:13:54,440
pick up the $100 bill, they
said, well, if it were real,

1493
01:13:54,440 --> 01:13:58,100
someone would have
already picked it up.

1494
01:13:58,100 --> 01:14:00,230
I mean, that's the argument
that we made together.

1495
01:14:00,230 --> 01:14:02,990
We made that argument, that if
there was a pattern somebody

1496
01:14:02,990 --> 01:14:04,948
would take advantage of
it and then the pattern

1497
01:14:04,948 --> 01:14:06,860
can't be there.

1498
01:14:06,860 --> 01:14:08,360
And then, again,
Warren Buffet would

1499
01:14:08,360 --> 01:14:10,235
say, that's the stupidest
thing I ever heard,

1500
01:14:10,235 --> 01:14:12,890
because in fact, I've done
it, I saw the patterns,

1501
01:14:12,890 --> 01:14:15,290
I took advantage of it,
and I have a bit more money

1502
01:14:15,290 --> 01:14:18,680
than you do, so there.

1503
01:14:18,680 --> 01:14:19,690
Who do you believe?

1504
01:14:19,690 --> 01:14:22,940
Well, it's kind of hard to
argue with a $40-something

1505
01:14:22,940 --> 01:14:24,860
billionaire.

1506
01:14:24,860 --> 01:14:28,010
I think that's his
wealth, $40 billion.

1507
01:14:28,010 --> 01:14:31,370
But that's not the
perspective of this analysis.

1508
01:14:31,370 --> 01:14:32,296
Mike?

1509
01:14:32,296 --> 01:14:34,797
STUDENT: Well, let's say
the expected return was

1510
01:14:34,797 --> 01:14:36,380
you had perfect
information and that's

1511
01:14:36,380 --> 01:14:38,215
what was going to be a
perfect crystal ball.

1512
01:14:38,215 --> 01:14:40,922
It would still be irrational
to buy McDonald's versus

1513
01:14:40,922 --> 01:14:44,978
Merck, so you'd short
McDonald's and long Merck

1514
01:14:44,978 --> 01:14:46,310
until the returns became equal.

1515
01:14:46,310 --> 01:14:46,820
PROFESSOR: Exactly.

1516
01:14:46,820 --> 01:14:49,010
So I'm going to talk about
that for a little while,

1517
01:14:49,010 --> 01:14:49,730
but you're right.

1518
01:14:49,730 --> 01:14:52,790
So if you could short,
then what you'd want to do

1519
01:14:52,790 --> 01:14:53,990
is exactly what you said.

1520
01:14:53,990 --> 01:14:59,540
You want to basically long the
low risk, high yield asset,

1521
01:14:59,540 --> 01:15:03,350
short the high risk, low
yield asset, make that spread,

1522
01:15:03,350 --> 01:15:05,599
and make it as
riskless as you can

1523
01:15:05,599 --> 01:15:06,890
buy including other securities.

1524
01:15:06,890 --> 01:15:07,715
Yes?

1525
01:15:07,715 --> 01:15:09,715
STUDENT: Yeah, but
eventually it would collapse,

1526
01:15:09,715 --> 01:15:11,090
and then the
relationship would--

1527
01:15:11,090 --> 01:15:12,339
PROFESSOR: And then it should.

1528
01:15:12,339 --> 01:15:12,920
That's right.

1529
01:15:12,920 --> 01:15:16,460
So the argument that
economists would make

1530
01:15:16,460 --> 01:15:21,540
is that this picture is
the equilibrium of where

1531
01:15:21,540 --> 01:15:26,580
these returns should be, given
what the market determines

1532
01:15:26,580 --> 01:15:31,030
their fair rates of return
are, relative to their risks.

1533
01:15:31,030 --> 01:15:33,960
So that's, again, a very big
philosophical difference.

1534
01:15:33,960 --> 01:15:36,720
An economist would say,
all of these securities

1535
01:15:36,720 --> 01:15:38,940
are exactly where
they should be.

1536
01:15:38,940 --> 01:15:43,560
And they may change over
time, but at any point in time

1537
01:15:43,560 --> 01:15:45,240
they are where they should be.

1538
01:15:45,240 --> 01:15:48,540
Supply equals demand,
market's clear,

1539
01:15:48,540 --> 01:15:53,200
everything is equilibrium,
and our decision

1540
01:15:53,200 --> 01:15:56,020
is simply to figure
out what to make

1541
01:15:56,020 --> 01:15:59,230
of the portfolio of
these securities.

1542
01:15:59,230 --> 01:16:02,750
What is the best portfolio
of these securities?

1543
01:16:02,750 --> 01:16:05,426
So I'm just warning you, this
is a philosophical departure

1544
01:16:05,426 --> 01:16:07,300
from what you're used
to thinking and reading

1545
01:16:07,300 --> 01:16:08,050
in the newspapers.

1546
01:16:08,050 --> 01:16:09,741
Because the newspapers
would say, well,

1547
01:16:09,741 --> 01:16:11,740
let's take a look at the
earnings at McDonald's.

1548
01:16:11,740 --> 01:16:12,790
Let's take a look at Merck.

1549
01:16:12,790 --> 01:16:14,020
Let's talk to the
macroeconomists

1550
01:16:14,020 --> 01:16:15,760
and see what's going to happen
over the next 12 months.

1551
01:16:15,760 --> 01:16:16,980
Let's talk to the
earnings analysts

1552
01:16:16,980 --> 01:16:18,896
and see whether they
forecast higher earnings,

1553
01:16:18,896 --> 01:16:20,020
lower earnings.

1554
01:16:20,020 --> 01:16:22,900
The whole point of the academic
infrastructure that we set up

1555
01:16:22,900 --> 01:16:26,110
is that you can't
predict these things.

1556
01:16:26,110 --> 01:16:28,360
And if you believe that,
then basically Warren Buffett

1557
01:16:28,360 --> 01:16:31,660
is just one really lucky guy.

1558
01:16:31,660 --> 01:16:37,450
So I'm going to have to justify
this academic position to you.

1559
01:16:37,450 --> 01:16:39,987
And I won't do that until
the end of the semester,

1560
01:16:39,987 --> 01:16:42,320
because first of all, I have
a lot of material to cover.

1561
01:16:42,320 --> 01:16:45,280
And I want to cover all of the
material in the basic form,

1562
01:16:45,280 --> 01:16:48,280
and then in the end I'm going
to give you a sense of where

1563
01:16:48,280 --> 01:16:49,930
things really stand.

1564
01:16:49,930 --> 01:16:50,900
It's a fiction.

1565
01:16:50,900 --> 01:16:54,710
It's a fiction that you
can't forecast stock prices,

1566
01:16:54,710 --> 01:16:58,150
but it's a fiction that actually
is pretty close to reality

1567
01:16:58,150 --> 01:17:00,730
for 99% of the public.

1568
01:17:00,730 --> 01:17:03,560
Now, you guys are not
99% of the public.

1569
01:17:03,560 --> 01:17:05,770
But for the people
that will someday

1570
01:17:05,770 --> 01:17:08,950
be your clients
or your investors,

1571
01:17:08,950 --> 01:17:12,010
it will be true, that
the typical individual

1572
01:17:12,010 --> 01:17:15,850
has no hope of being able to
out-forecast Warren Buffett.

1573
01:17:15,850 --> 01:17:17,860
And if you can't
out-forecast somebody,

1574
01:17:17,860 --> 01:17:20,320
then you may as well
assume that they're random

1575
01:17:20,320 --> 01:17:23,590
and they're perfectly priced.

1576
01:17:23,590 --> 01:17:25,330
And then you still
have the problem, OK,

1577
01:17:25,330 --> 01:17:27,667
if you assume that,
then what do you do?

1578
01:17:27,667 --> 01:17:29,500
That's all we're going
to try to figure out.

1579
01:17:29,500 --> 01:17:33,890
I'm going to tell you what
you do with portfolio theory.

1580
01:17:33,890 --> 01:17:36,580
Now, since we're
almost out of time,

1581
01:17:36,580 --> 01:17:38,330
I want to just tell
you where we're going.

1582
01:17:38,330 --> 01:17:40,580
What we're going to do
is look at this graph

1583
01:17:40,580 --> 01:17:45,230
and ask the question,
what do people want?

1584
01:17:45,230 --> 01:17:48,590
They want higher return,
they want to go north,

1585
01:17:48,590 --> 01:17:50,120
and they want lower risk.

1586
01:17:50,120 --> 01:17:52,220
They want to go west.

1587
01:17:52,220 --> 01:17:54,350
So the northwest is
where we're going

1588
01:17:54,350 --> 01:17:57,110
to be heading in this
graph, and the question

1589
01:17:57,110 --> 01:17:58,970
is, how can we get there?

1590
01:17:58,970 --> 01:18:01,820
How can we get as
northwest as possible

1591
01:18:01,820 --> 01:18:03,140
using these securities?

1592
01:18:03,140 --> 01:18:05,090
And the answer will
shock you, I think,

1593
01:18:05,090 --> 01:18:07,570
because you're going
to see that by doing

1594
01:18:07,570 --> 01:18:10,190
a very simple little bit
of high school algebra

1595
01:18:10,190 --> 01:18:12,740
we can actually create
a portfolio that

1596
01:18:12,740 --> 01:18:15,020
beats all of these things.

1597
01:18:15,020 --> 01:18:17,600
That is, if you
didn't know anything

1598
01:18:17,600 --> 01:18:21,170
about portfolio theory, you
would be severely worse off,

1599
01:18:21,170 --> 01:18:24,119
because you'd be stuck having to
be on one of these five points.

1600
01:18:24,119 --> 01:18:25,910
And if you knew a little
bit of high school

1601
01:18:25,910 --> 01:18:29,550
algebra and some finance, you
could actually do a lot better.

1602
01:18:29,550 --> 01:18:32,230
So we'll see that on Monday.