1
00:00:00,040 --> 00:00:02,460
The following content is
provided under a Creative

2
00:00:02,460 --> 00:00:03,970
Commons license.

3
00:00:03,970 --> 00:00:06,910
Your support will help MIT
OpenCourseWare continue to

4
00:00:06,910 --> 00:00:10,660
offer high quality educational
resources for free.

5
00:00:10,660 --> 00:00:13,460
To make a donation or view
additional materials from

6
00:00:13,460 --> 00:00:17,390
hundreds of MIT courses, visit
MIT OpenCourseWare at

7
00:00:17,390 --> 00:00:18,640
ocw.mit.edu.

8
00:00:25,850 --> 00:00:29,610
PROFESSOR: So this is a really,
really, really, really

9
00:00:29,610 --> 00:00:35,780
stunning fact in a sense of we
talk about lots of forms of

10
00:00:35,780 --> 00:00:38,830
inequality and unfairness
in the world.

11
00:00:38,830 --> 00:00:43,074
But here's a fact that's
really quite,

12
00:00:43,074 --> 00:00:45,700
quite, quite striking.

13
00:00:45,700 --> 00:00:48,895
You start from the basic,
just the fact.

14
00:00:51,570 --> 00:00:56,200
It's the fact that if you look
at the ratio of men and women

15
00:00:56,200 --> 00:01:02,780
in the population, most
countries in the world have

16
00:01:02,780 --> 00:01:05,080
many more women than men.

17
00:01:05,080 --> 00:01:09,900
This is mostly because
women are just better

18
00:01:09,900 --> 00:01:14,460
designed, it turns out.

19
00:01:14,460 --> 00:01:20,830
When men get to the age of about
60, various bad things

20
00:01:20,830 --> 00:01:21,520
happen to them.

21
00:01:21,520 --> 00:01:26,410
In particular, their arteries
seem to harden.

22
00:01:26,410 --> 00:01:29,170
And lots of them die
from various

23
00:01:29,170 --> 00:01:31,460
cardiovascular diseases.

24
00:01:31,460 --> 00:01:38,390
So as a result, most countries
have many more women than men.

25
00:01:38,390 --> 00:01:45,020
So the ratios up there,
Europe has 1.05.

26
00:01:45,020 --> 00:01:54,190
Sub-Saharan Africa has 1.02.

27
00:01:54,190 --> 00:01:55,900
All of Latin America
is above 1.

28
00:01:55,900 --> 00:01:57,490
The US is about 1.

29
00:01:57,490 --> 00:02:02,340
And while it doesn't look like a
lot, 4%, we're talking about

30
00:02:02,340 --> 00:02:08,310
4% of 3 billion.

31
00:02:08,310 --> 00:02:11,690
So that's a fair number
of extra women

32
00:02:11,690 --> 00:02:12,570
we're talking about.

33
00:02:12,570 --> 00:02:15,620
The world has many more women
than men overall.

34
00:02:15,620 --> 00:02:21,310
But the one place where that's
not true is in this belt which

35
00:02:21,310 --> 00:02:25,780
goes from basically North
Africa to Korea.

36
00:02:25,780 --> 00:02:30,140
That's that one belt in the
world pretty much all the way.

37
00:02:30,140 --> 00:02:37,920
You can draw a belt starting
somewhere in Morocco and keep

38
00:02:37,920 --> 00:02:41,820
going east, about at
that latitude.

39
00:02:41,820 --> 00:02:44,780
And then go a bit north into
China, and you'll find that

40
00:02:44,780 --> 00:02:49,460
that's where they are about
4% or less women than men.

41
00:03:04,840 --> 00:03:06,690
These numbers are not
necessarily the worst numbers

42
00:03:06,690 --> 00:03:07,320
you're going to see.

43
00:03:07,320 --> 00:03:11,300
So I'll show you some more
striking numbers in a minute.

44
00:03:11,300 --> 00:03:17,590
I mean, just the fact that
there's 6% less women than men

45
00:03:17,590 --> 00:03:25,210
is striking, given that that
doesn't look like the

46
00:03:25,210 --> 00:03:26,460
biological model.

47
00:03:31,800 --> 00:03:36,720
So Amartya Sen, who was a Nobel
Prize winner at Harvard

48
00:03:36,720 --> 00:03:39,290
did this calculation, basically
very simple

49
00:03:39,290 --> 00:03:40,550
calculation.

50
00:03:40,550 --> 00:03:43,700
He said, look, Sub-Saharan
Africa is the

51
00:03:43,700 --> 00:03:45,080
poorest part of the world.

52
00:03:49,470 --> 00:03:54,680
So if there is some maybe
institutional reason why women

53
00:03:54,680 --> 00:03:57,310
die, so there might be reasons
why women die that have to do

54
00:03:57,310 --> 00:03:58,010
with poverty.

55
00:03:58,010 --> 00:04:01,640
What's an example of a reason,
which would hurt women more

56
00:04:01,640 --> 00:04:05,880
than men, reason why
poverty would hurt

57
00:04:05,880 --> 00:04:07,816
women more than men?

58
00:04:07,816 --> 00:04:09,160
AUDIENCE: During childbirth.

59
00:04:09,160 --> 00:04:11,260
PROFESSOR: Childbirth, the
obvious one is childbirth.

60
00:04:11,260 --> 00:04:11,750
Yeah?

61
00:04:11,750 --> 00:04:16,160
AUDIENCE: And then compounding
upon that, if women in poor

62
00:04:16,160 --> 00:04:19,426
situations have less access to
contraceptives and have more

63
00:04:19,426 --> 00:04:21,060
children, which increases
the lack [INAUDIBLE].

64
00:04:21,060 --> 00:04:22,434
There will be complications
for [INAUDIBLE].

65
00:04:22,434 --> 00:04:25,950
PROFESSOR: So everything to do
with childbirth is reason why

66
00:04:25,950 --> 00:04:32,330
you might think that in poor
countries, women will have a

67
00:04:32,330 --> 00:04:34,990
disadvantage.

68
00:04:34,990 --> 00:04:37,360
You do see evidence of
the disadvantage.

69
00:04:37,360 --> 00:04:41,250
Sub-Saharan Africa has less
women than Europe.

70
00:04:41,250 --> 00:04:45,320
So you do see an example
of that disadvantage.

71
00:04:45,320 --> 00:04:49,100
It's about 2 percentage points
or 3 percentage points less

72
00:04:49,100 --> 00:04:55,320
women than Europe.

73
00:04:55,320 --> 00:04:58,110
If you think of Sub-Saharan
Africa as the part of the

74
00:04:58,110 --> 00:05:01,860
world which has the highest
maternal mortality rates.

75
00:05:01,860 --> 00:05:04,220
It's the part of the world
which has the highest

76
00:05:04,220 --> 00:05:05,180
fertility rates.

77
00:05:05,180 --> 00:05:09,460
Both to your point and your
point, if you take the part of

78
00:05:09,460 --> 00:05:12,770
the world where women are most
likely to die in childbirth,

79
00:05:12,770 --> 00:05:18,960
and the place where women have
the most children, both of

80
00:05:18,960 --> 00:05:20,590
those are highest in
Sub-Saharan Africa.

81
00:05:20,590 --> 00:05:22,620
So you could think of this
as being a reasonable

82
00:05:22,620 --> 00:05:25,350
approximation of the worst
case scenario.

83
00:05:25,350 --> 00:05:28,850
Take the worst case scenario
and then go to China.

84
00:05:28,850 --> 00:05:30,270
So that's the exercise.

85
00:05:30,270 --> 00:05:33,570
Let's take the worst
case scenario.

86
00:05:33,570 --> 00:05:40,440
Now go to China and ask, if
we applied the Sub-Saharan

87
00:05:40,440 --> 00:05:44,650
African ratio to the Chinese
population, how many more

88
00:05:44,650 --> 00:05:46,370
women would they have had?

89
00:05:46,370 --> 00:05:49,450
The answer is they would have
had 44 million more women.

90
00:05:49,450 --> 00:05:53,420
So 44 million more women
are missing.

91
00:05:53,420 --> 00:05:59,900
And you keep doing that, and
same order of magnitude in

92
00:05:59,900 --> 00:06:03,640
India, and then smaller
countries, so smaller numbers,

93
00:06:03,640 --> 00:06:05,850
but it all adds up toward
100 million.

94
00:06:05,850 --> 00:06:10,090
So you get about 100 million
missing women in the world by

95
00:06:10,090 --> 00:06:12,300
doing this simple exercise.

96
00:06:12,300 --> 00:06:18,690
And that's sort of prima
facie evidence.

97
00:06:18,690 --> 00:06:21,690
As we'll see, you have to think
harder about exactly

98
00:06:21,690 --> 00:06:22,950
what this evidence means.

99
00:06:22,950 --> 00:06:24,820
This is prima facie evidence
of something

100
00:06:24,820 --> 00:06:26,070
very bizarre happening.

101
00:06:29,450 --> 00:06:31,330
We'll see what's
going on here.

102
00:06:31,330 --> 00:06:34,150
But there's something very
troubling about this fact.

103
00:06:34,150 --> 00:06:38,640
Because it seems like somehow
populations are managing to

104
00:06:38,640 --> 00:06:42,970
get rid of women, quote unquote,
"losing women"

105
00:06:42,970 --> 00:06:46,120
through mechanisms which are
not the normal biological

106
00:06:46,120 --> 00:06:47,370
mechanisms.

107
00:06:50,090 --> 00:06:52,720
So that's the missing women.

108
00:06:52,720 --> 00:06:53,020
Yeah?

109
00:06:53,020 --> 00:06:56,120
AUDIENCE: So this order for
killing women in China, are

110
00:06:56,120 --> 00:06:57,790
they missing in rural China?

111
00:06:57,790 --> 00:06:59,700
Or is it just all
around China?

112
00:06:59,700 --> 00:07:00,925
PROFESSOR: I'll show
you in a minute.

113
00:07:00,925 --> 00:07:03,300
I'll show you all kinds of facts
about China, in fact.

114
00:07:08,400 --> 00:07:13,790
So before I come to that,
there's just some hypotheses

115
00:07:13,790 --> 00:07:14,970
you might have.

116
00:07:14,970 --> 00:07:21,370
One is, is development
the answer?

117
00:07:21,370 --> 00:07:33,640
Maybe there's some reason why
just being undeveloped for

118
00:07:33,640 --> 00:07:36,630
reasons we don't understand
translates into less women.

119
00:07:36,630 --> 00:07:41,950
Maybe we have a theory that
maybe it's the case that

120
00:07:41,950 --> 00:07:44,830
people are more traditional,
something.

121
00:07:44,830 --> 00:07:48,010
Now what makes this fact much
more worrying is that

122
00:07:48,010 --> 00:07:51,870
development is not making
this fact go away.

123
00:07:51,870 --> 00:07:58,940
That's sort of at least
within modern history.

124
00:07:58,940 --> 00:08:02,320
We don't know what happened
100 years ago.

125
00:08:02,320 --> 00:08:07,380
But right now, development is
not making this fact go away.

126
00:08:07,380 --> 00:08:11,090
If you look at the sex ratio in
India, it's not changing.

127
00:08:11,090 --> 00:08:15,250
It's in fact pretty
much steady.

128
00:08:15,250 --> 00:08:16,770
I'll show you numbers
in a minute.

129
00:08:16,770 --> 00:08:20,950
If you look at the sex ratio
in China, it's falling.

130
00:08:20,950 --> 00:08:24,740
It's less and less women
every year in China.

131
00:08:24,740 --> 00:08:29,450
And if you look at the richest
parts of China versus the

132
00:08:29,450 --> 00:08:33,640
poorest part to your question,
you see a negative

133
00:08:33,640 --> 00:08:34,260
correlation.

134
00:08:34,260 --> 00:08:36,419
The riches parts of China
are the ones which

135
00:08:36,419 --> 00:08:38,010
have the least women.

136
00:08:38,010 --> 00:08:40,066
Yeah?

137
00:08:40,066 --> 00:08:42,561
AUDIENCE: Do we know anything
about missing men?

138
00:08:42,561 --> 00:08:46,553
Maybe there's some component
where there's just a general

139
00:08:46,553 --> 00:08:49,048
[INAUDIBLE] for children in
some parts of the world.

140
00:08:49,048 --> 00:08:51,044
And the number of children
in total are missing,

141
00:08:51,044 --> 00:08:54,510
not just the girls.

142
00:08:54,510 --> 00:08:56,650
PROFESSOR: That's
a good question.

143
00:08:56,650 --> 00:09:00,380
I think that's what people do
when they say, the infant

144
00:09:00,380 --> 00:09:03,290
mortality rate is too
high or something.

145
00:09:03,290 --> 00:09:08,800
And I think that that's
also a concern.

146
00:09:08,800 --> 00:09:09,910
It's not that there
aren't other

147
00:09:09,910 --> 00:09:11,160
concerns because of this.

148
00:09:16,170 --> 00:09:19,790
A lot of people would say that
you shouldn't just look at GDP

149
00:09:19,790 --> 00:09:21,230
to measure welfare.

150
00:09:21,230 --> 00:09:22,880
Emphasize infant mortality
rates.

151
00:09:22,880 --> 00:09:25,360
Infant mortality rates is
exactly the question you're

152
00:09:25,360 --> 00:09:27,570
asking, missing boys
and girls.

153
00:09:30,320 --> 00:09:33,900
That number is very high
in Sub-Saharan Africa.

154
00:09:33,900 --> 00:09:37,280
Sub-Saharan Africa has the
highest infant mortality rate

155
00:09:37,280 --> 00:09:39,040
in the world by far.

156
00:09:39,040 --> 00:09:42,490
China has now infant mortality
rates which are approaching

157
00:09:42,490 --> 00:09:45,410
that in the US.

158
00:09:45,410 --> 00:09:52,440
So infant morality rates by
itself, by now it's not a huge

159
00:09:52,440 --> 00:09:53,770
problem in China.

160
00:09:53,770 --> 00:09:58,330
China is down to the range
where it's very, very

161
00:09:58,330 --> 00:09:59,760
close to the US.

162
00:09:59,760 --> 00:10:02,497
I think it has infant mortality
rates of like 25 our

163
00:10:02,497 --> 00:10:04,890
of 1,000, 1,000 live births.

164
00:10:04,890 --> 00:10:07,660
US has 18 or 14 or something.

165
00:10:07,660 --> 00:10:12,700
It's really in the range, where
Sub-Saharan Africa,

166
00:10:12,700 --> 00:10:17,520
Sierra Leone has 140.

167
00:10:17,520 --> 00:10:20,260
So if you want to see where
infant mortality is a problem,

168
00:10:20,260 --> 00:10:22,655
that's not where missing
women is a problem.

169
00:10:22,655 --> 00:10:24,270
These are different problems.

170
00:10:24,270 --> 00:10:25,870
It's not to say you're
not right.

171
00:10:25,870 --> 00:10:26,740
That's another problem.

172
00:10:26,740 --> 00:10:28,190
But it's a different one.

173
00:10:31,410 --> 00:10:39,300
So this is sort of just
to show you that fact.

174
00:10:47,180 --> 00:10:50,770
So what I showed you so far
was the levels, how

175
00:10:50,770 --> 00:10:53,830
many women per men.

176
00:10:53,830 --> 00:11:00,170
This is, in 2005, what's
the future.

177
00:11:00,170 --> 00:11:04,920
Future's much more scary
than the present.

178
00:11:04,920 --> 00:11:06,170
The present is what--

179
00:11:10,710 --> 00:11:13,440
there was like 95 girls
per 100 boys.

180
00:11:13,440 --> 00:11:17,440
So that's about 105 boys
per 100 girls.

181
00:11:17,440 --> 00:11:18,990
That's the current.

182
00:11:18,990 --> 00:11:22,555
The future, look at the number
for Eastern Asia.

183
00:11:25,220 --> 00:11:33,680
That number is 120 boys
for 100 girls.

184
00:11:33,680 --> 00:11:36,980
It's a completely different
order of magnitude.

185
00:11:36,980 --> 00:11:38,590
Think of the future.

186
00:11:38,590 --> 00:11:41,942
These are the kids who are
being born under five.

187
00:11:41,942 --> 00:11:45,930
So at point, if this is what
stays steady, we are going to

188
00:11:45,930 --> 00:11:51,830
see numbers more like
120 than 105.

189
00:11:51,830 --> 00:11:56,640
Right now there are 105 boys
to 100 girls in China.

190
00:11:56,640 --> 00:12:01,340
And that number is going to go
up to closer to 120 as this

191
00:12:01,340 --> 00:12:04,510
cohort ages.

192
00:12:04,510 --> 00:12:09,505
So that's just backing up the
first claim I made, which is

193
00:12:09,505 --> 00:12:10,880
it's not getting better.

194
00:12:10,880 --> 00:12:14,230
If you look at the number
for India, it's 108.

195
00:12:14,230 --> 00:12:21,960
108 is bigger than the
current ratio of

196
00:12:21,960 --> 00:12:23,630
men to women in India.

197
00:12:23,630 --> 00:12:27,155
So current ratio is about
106, and that's 108.

198
00:12:27,155 --> 00:12:30,200
So all of these countries,
which are getting richer,

199
00:12:30,200 --> 00:12:35,250
their ratio is getting worse.

200
00:12:35,250 --> 00:12:37,660
Do you see what this is, how it
this is different from the

201
00:12:37,660 --> 00:12:38,910
one I showed you before?

202
00:12:41,480 --> 00:12:44,320
This is the ratio among
under fives.

203
00:12:44,320 --> 00:12:48,120
So these are the future of the
population, whereas I showed

204
00:12:48,120 --> 00:12:50,200
you the present of
the population.

205
00:12:50,200 --> 00:12:51,450
Future's worse, not better.

206
00:12:57,880 --> 00:13:01,725
Korea has the same kind of
numbers, which are worse.

207
00:13:05,970 --> 00:13:11,710
Interestingly, Afghanistan and
Nepal, which are relatively

208
00:13:11,710 --> 00:13:15,110
poor countries, are not getting
rich very fast,

209
00:13:15,110 --> 00:13:17,550
actually you don't see
much of a change.

210
00:13:17,550 --> 00:13:18,800
They're about where they were.

211
00:13:21,880 --> 00:13:24,780
Under five is about what they
are in the adult population.

212
00:13:24,780 --> 00:13:26,590
So that suggests that they
aren't getting worse.

213
00:13:26,590 --> 00:13:28,255
It's the richer countries
that are getting

214
00:13:28,255 --> 00:13:29,630
richer are getting worse.

215
00:13:29,630 --> 00:13:30,091
Yep?

216
00:13:30,091 --> 00:13:31,935
AUDIENCE: But wouldn't you
expect this number to be

217
00:13:31,935 --> 00:13:36,300
higher than the overall
population because women are

218
00:13:36,300 --> 00:13:40,620
more likely to live longer
and more [INAUDIBLE].

219
00:13:40,620 --> 00:13:41,760
PROFESSOR: That's
absolutely true.

220
00:13:41,760 --> 00:13:45,700
So 121 go to 120.

221
00:13:45,700 --> 00:13:47,930
120 is going to go to 115.

222
00:13:47,930 --> 00:13:49,300
That's about the difference.

223
00:13:49,300 --> 00:13:52,030
So 120 is going to
do 115 of 14.

224
00:13:52,030 --> 00:13:57,840
But 114 is a lot
worse than 106.

225
00:13:57,840 --> 00:14:00,332
All of the trends are bad.

226
00:14:00,332 --> 00:14:02,940
Just to understand, this
is a big problem.

227
00:14:02,940 --> 00:14:06,270
And it's not one that's getting
solved on its own as

228
00:14:06,270 --> 00:14:09,490
far as I can tell.

229
00:14:09,490 --> 00:14:11,530
So here's another way
to look at it.

230
00:14:11,530 --> 00:14:13,455
So this is India and China.

231
00:14:16,640 --> 00:14:18,600
Give yourself a few minutes
to look at this.

232
00:14:18,600 --> 00:14:21,910
So the darkest areas are
the worst areas.

233
00:14:36,080 --> 00:14:39,590
Does anybody know which are
the richer areas of China?

234
00:14:47,988 --> 00:14:49,964
AUDIENCE: The black areas?

235
00:14:49,964 --> 00:14:53,516
PROFESSOR: Yeah, it's pretty
much exactly the same.

236
00:14:53,516 --> 00:14:58,040
I think Shanghai is that gray
area right there, which is

237
00:14:58,040 --> 00:15:00,040
actually doing surprisingly
well.

238
00:15:03,770 --> 00:15:08,000
The really poor areas of China
are those very light

239
00:15:08,000 --> 00:15:11,550
areas in the west.

240
00:15:11,550 --> 00:15:13,310
That's the Gobi Desert, the

241
00:15:13,310 --> 00:15:16,480
Taklamakan Desert, the Chinkiang.

242
00:15:16,480 --> 00:15:20,890
All the areas with large Muslim

243
00:15:20,890 --> 00:15:23,140
populations are in the east.

244
00:15:23,140 --> 00:15:25,880
These are the poorest parts of
China, and they're the ones

245
00:15:25,880 --> 00:15:29,020
which are doing by
far the best.

246
00:15:29,020 --> 00:15:31,580
The areas that are being really
badly are the most

247
00:15:31,580 --> 00:15:32,510
prosperous.

248
00:15:32,510 --> 00:15:39,210
The dark areas in the southeast,
that's like

249
00:15:39,210 --> 00:15:43,350
Guangdong, and Guangzhou, and
all these really fast-growing,

250
00:15:43,350 --> 00:15:45,820
wealthy provinces of China,
that's the southeast.

251
00:15:51,590 --> 00:15:53,470
Yeah?

252
00:15:53,470 --> 00:15:57,782
AUDIENCE: So I'm not really sure
exactly how [INAUDIBLE],

253
00:15:57,782 --> 00:16:01,439
but the fact that there is
[INAUDIBLE] in the number of

254
00:16:01,439 --> 00:16:03,510
kids that couples can have.

255
00:16:03,510 --> 00:16:07,304
You know, people might be
deciding on purpose to have

256
00:16:07,304 --> 00:16:10,208
boys because they want boys
for whatever reason.

257
00:16:10,208 --> 00:16:13,959
And maybe the people who make
those decision are people to

258
00:16:13,959 --> 00:16:17,960
have more money, because to have
abortions is expensive.

259
00:16:17,960 --> 00:16:19,880
PROFESSOR: I think that has
a lot to do with it.

260
00:16:19,880 --> 00:16:20,940
We'll come back to it.

261
00:16:20,940 --> 00:16:26,400
You're perfectly right that a
lot of this ease has something

262
00:16:26,400 --> 00:16:29,656
to do with people's active
choices also.

263
00:16:29,656 --> 00:16:32,150
I'll come back to that
in a little.

264
00:16:37,880 --> 00:16:41,990
As China gets even richer,
because this is the trend, and

265
00:16:41,990 --> 00:16:44,780
it's a worrying trend,
let's say.

266
00:16:44,780 --> 00:16:47,060
Now look at India.

267
00:16:47,060 --> 00:16:54,383
Again, do you know what those
relatively pale areas are?

268
00:16:54,383 --> 00:16:55,830
AUDIENCE: South India?

269
00:16:55,830 --> 00:17:00,210
PROFESSOR: The South India,
which is relatively prosperous

270
00:17:00,210 --> 00:17:06,430
and well-educated, and East
India which is the poorest and

271
00:17:06,430 --> 00:17:07,680
the most backward
part of India.

272
00:17:17,700 --> 00:17:23,334
Richest state in India, large
state in India, is

273
00:17:23,334 --> 00:17:29,270
Maharashtra, which is kind
of the striped state.

274
00:17:29,270 --> 00:17:34,180
Let's see, on the west coast,
if you go from the south on

275
00:17:34,180 --> 00:17:36,940
the west coast, the first
striped one, that's

276
00:17:36,940 --> 00:17:37,390
Maharashtra.

277
00:17:37,390 --> 00:17:38,990
That's the richest.

278
00:17:38,990 --> 00:17:45,530
The second richest large state
is Punjab And Punjab is that

279
00:17:45,530 --> 00:17:49,520
piece of black that's
right in the north.

280
00:17:49,520 --> 00:17:54,330
So that's literally the second
richest, and still recently

281
00:17:54,330 --> 00:17:57,580
the richest state in India,
is that area in black.

282
00:17:57,580 --> 00:17:59,420
So again, the same pattern--

283
00:17:59,420 --> 00:18:03,640
rich states have worse outcomes,
not better outcomes.

284
00:18:03,640 --> 00:18:11,210
And there's no evidence of a
positive correlation between

285
00:18:11,210 --> 00:18:16,246
income and the gender ratio.

286
00:18:16,246 --> 00:18:20,190
AUDIENCE: Is Punjab the only
state in India that has a

287
00:18:20,190 --> 00:18:23,148
different religion than the
rest of the others?

288
00:18:23,148 --> 00:18:24,630
Or would that be a reason?

289
00:18:24,630 --> 00:18:25,880
PROFESSOR: Maybe.

290
00:18:30,320 --> 00:18:32,370
What that is is there's really
two states there.

291
00:18:32,370 --> 00:18:34,060
There is Punjab and Haryana.

292
00:18:34,060 --> 00:18:36,690
They used to be both
a part of Punjab.

293
00:18:36,690 --> 00:18:39,120
Now they've been split into
two states for some years.

294
00:18:39,120 --> 00:18:43,720
But one of those states
is Sheik majority.

295
00:18:43,720 --> 00:18:45,500
Sheik is a particular
other religion.

296
00:18:45,500 --> 00:18:47,580
The other is Hindu majority.

297
00:18:47,580 --> 00:18:48,450
There's no difference.

298
00:18:48,450 --> 00:18:50,750
Actually, Haryana's slightly
worse that Punjab.

299
00:18:50,750 --> 00:18:54,500
Delhi City is one of the worst
places, actually, in India.

300
00:18:54,500 --> 00:18:56,950
So that's one of the absolutely
riches places.

301
00:18:56,950 --> 00:19:01,920
So basically no evidence of a
positive correlation between

302
00:19:01,920 --> 00:19:05,440
income and the gender ratio--

303
00:19:05,440 --> 00:19:09,660
gender ratio, if anything, gets
worse as you get richer,

304
00:19:09,660 --> 00:19:12,770
for reasons that may very
be close to what you're

305
00:19:12,770 --> 00:19:13,270
suggesting.

306
00:19:13,270 --> 00:19:17,080
But it's clear that you get this
rather dismal pattern.

307
00:19:19,800 --> 00:19:24,900
Here's something else, another
fact about sort of China.

308
00:19:24,900 --> 00:19:27,580
And it's sort of a very long
term view of China.

309
00:19:32,310 --> 00:19:35,680
This starts in 1920.

310
00:19:35,680 --> 00:19:47,120
And you can see that basically
from the 1920s the excess sex

311
00:19:47,120 --> 00:19:54,380
ratio, which is in the five-year
birth cohort.

312
00:19:54,380 --> 00:19:59,380
So people who were born in that
year, what's the fraction

313
00:19:59,380 --> 00:20:01,280
of boys versus girls?

314
00:20:01,280 --> 00:20:09,320
So it used to be about 7%
more boys than girls.

315
00:20:09,320 --> 00:20:14,830
Then it went up to about 16%
during the years when there's

316
00:20:14,830 --> 00:20:16,730
a lot of political
instability.

317
00:20:16,730 --> 00:20:20,180
So if you think of those years
as Japanese invasion, before

318
00:20:20,180 --> 00:20:22,990
that and after that, in a sense,
China is in civil war.

319
00:20:26,150 --> 00:20:28,910
What are called the warlords
are fighting each other

320
00:20:28,910 --> 00:20:29,780
through China.

321
00:20:29,780 --> 00:20:33,430
So during wars, you see this
ratio going way up.

322
00:20:33,430 --> 00:20:35,470
Then it comes in down, down,
down, down, down, down, down,

323
00:20:35,470 --> 00:20:41,820
down, till about 1976
or something.

324
00:20:41,820 --> 00:20:47,310
Now right around 1976 something
else happens.

325
00:20:47,310 --> 00:20:50,060
China introduces what's called
the one child policy.

326
00:20:50,060 --> 00:20:55,180
The one child policy is one
where at least in the richer

327
00:20:55,180 --> 00:20:58,850
areas, where it was the only
place it was enforced,

328
00:20:58,850 --> 00:21:02,146
families were allowed to have
only one child at most.

329
00:21:05,410 --> 00:21:10,350
Then that policy gets
implemented in the mid '70s.

330
00:21:10,350 --> 00:21:11,950
And you see what's
happened since.

331
00:21:11,950 --> 00:21:15,220
What's striking is it doesn't
just go up and stay up.

332
00:21:15,220 --> 00:21:18,180
It keeps going up.

333
00:21:18,180 --> 00:21:20,810
So there's something else going
on on top of the fact

334
00:21:20,810 --> 00:21:23,030
that the one child policy
gets instituted.

335
00:21:23,030 --> 00:21:25,760
In fact, over time, the one
child policy has been relaxed.

336
00:21:29,670 --> 00:21:32,550
And this reflects some
of the tests of

337
00:21:32,550 --> 00:21:33,920
the population clearly.

338
00:21:33,920 --> 00:21:38,070
Families which had one daughter
in some areas were

339
00:21:38,070 --> 00:21:43,250
allowed to have a son or have a
second child more recently.

340
00:21:43,250 --> 00:21:46,100
So there's some attempt to relax
the policy in exactly

341
00:21:46,100 --> 00:21:48,390
the direction that suggests
that there's a

342
00:21:48,390 --> 00:21:51,150
preference for boys.

343
00:21:51,150 --> 00:21:55,360
But with all that, the
trend has been up.

344
00:21:55,360 --> 00:21:56,240
Yep?

345
00:21:56,240 --> 00:21:58,165
AUDIENCE: Do you know what's
happened to those policies?

346
00:21:58,165 --> 00:21:59,985
Were they implemented,
[INAUDIBLE] relaxation?

347
00:22:04,000 --> 00:22:05,860
PROFESSOR: A bunch to the
relaxations happen.

348
00:22:05,860 --> 00:22:08,810
You see they're flattening
in about 1982.

349
00:22:08,810 --> 00:22:10,060
That's the first relaxation.

350
00:22:12,500 --> 00:22:17,680
So you do see a flattening,
but the trend is bad.

351
00:22:17,680 --> 00:22:19,025
Yeah?

352
00:22:19,025 --> 00:22:21,110
AUDIENCE: Even if it's being
relaxed, just being relaxed,

353
00:22:21,110 --> 00:22:26,074
if the first kid that you have
is a girl, and then in that

354
00:22:26,074 --> 00:22:27,493
case, you might then
try to have a boy?

355
00:22:27,493 --> 00:22:30,325
But if they had the boy right
away, maybe they'd just stop

356
00:22:30,325 --> 00:22:31,500
having kids.

357
00:22:31,500 --> 00:22:33,290
PROFESSOR: Absolutely.

358
00:22:33,290 --> 00:22:36,060
Actually, let's take it.

359
00:22:36,060 --> 00:22:42,220
So could this be generated
by a stopping rule?

360
00:22:42,220 --> 00:22:45,980
So imagine that families have a
stopping rule which says, we

361
00:22:45,980 --> 00:22:54,810
will stop when we have
the first boy.

362
00:22:54,810 --> 00:22:56,410
This is their stopping rule.

363
00:22:56,410 --> 00:22:57,930
This is a plausible
stopping rule.

364
00:22:57,930 --> 00:23:05,450
My mother has one brother
and three sisters.

365
00:23:05,450 --> 00:23:07,390
The brother's the youngest.

366
00:23:07,390 --> 00:23:10,640
So that's not an implausible
stopping rule.

367
00:23:10,640 --> 00:23:15,250
My mother's only brother has
three daughters and once son.

368
00:23:15,250 --> 00:23:17,280
The son is the youngest.

369
00:23:17,280 --> 00:23:19,660
It's not an implausible
stopping rule.

370
00:23:19,660 --> 00:23:26,240
However, imagine we had
this stopping rule.

371
00:23:26,240 --> 00:23:30,410
But in any but, the probability
of boy and girl

372
00:23:30,410 --> 00:23:32,790
would equal.

373
00:23:32,790 --> 00:23:37,180
What would be the population
sex ratio?

374
00:23:37,180 --> 00:23:38,644
AUDIENCE: Still the same.

375
00:23:38,644 --> 00:23:39,620
AUDIENCE: [INAUDIBLE]
boys, right?

376
00:23:39,620 --> 00:23:41,572
No, you'd have more women.

377
00:23:41,572 --> 00:23:43,524
AUDIENCE: No, it's half.

378
00:23:43,524 --> 00:23:44,988
PROFESSOR: Which one?

379
00:23:44,988 --> 00:23:48,892
AUDIENCE: If you keep having
children when you have women,

380
00:23:48,892 --> 00:23:52,198
then you stop as soon as you
have one boy, you'd have more

381
00:23:52,198 --> 00:23:55,174
women, or you would
expect more women.

382
00:23:55,174 --> 00:23:56,662
AUDIENCE: [INAUDIBLE].

383
00:23:56,662 --> 00:23:57,912
AUDIENCE: [INAUDIBLE].

384
00:24:00,134 --> 00:24:01,622
You stop immediately.

385
00:24:01,622 --> 00:24:04,102
You stop immediately if
you have one son.

386
00:24:04,102 --> 00:24:07,078
[INAUDIBLE].

387
00:24:07,078 --> 00:24:09,558
There might be a family where
there are more girls

388
00:24:09,558 --> 00:24:10,808
[INAUDIBLE].

389
00:24:13,285 --> 00:24:17,650
But most families
will stop there.

390
00:24:17,650 --> 00:24:20,090
AUDIENCE: [INAUDIBLE].

391
00:24:20,090 --> 00:24:23,818
PROFESSOR: So do you think
they'll be more [INAUDIBLE]?

392
00:24:23,818 --> 00:24:26,806
AUDIENCE: I think it pretty
much evens out.

393
00:24:26,806 --> 00:24:29,046
AUDIENCE: I think there will be
more boys, because if you

394
00:24:29,046 --> 00:24:30,292
have a boy, you just stop.

395
00:24:30,292 --> 00:24:31,786
So then that family
just has a boy.

396
00:24:31,786 --> 00:24:34,774
And if you have a girl and then
you try again, if you

397
00:24:34,774 --> 00:24:37,762
have a boy, then right now you
have one family with one boy

398
00:24:37,762 --> 00:24:39,256
and another family with
a girl and a boy.

399
00:24:39,256 --> 00:24:41,248
But there's still more
boys than girls.

400
00:24:41,248 --> 00:24:43,490
So whenever you get to
a boy, you stop.

401
00:24:43,490 --> 00:24:45,378
PROFESSOR: You want to
come back to that?

402
00:24:45,378 --> 00:24:48,035
AUDIENCE: Yeah, and so if we
just do an expected values,

403
00:24:48,035 --> 00:24:49,826
and you get half.

404
00:24:49,826 --> 00:24:52,316
Half the cases you'll
have one boy.

405
00:24:52,316 --> 00:24:54,310
What are the cases
of girl and boy?

406
00:24:54,310 --> 00:24:55,470
PROFESSOR: You're
exactly right.

407
00:24:55,470 --> 00:24:57,640
The answer is it doesn't
change anything.

408
00:24:57,640 --> 00:24:58,900
And it's very simple to see.

409
00:24:58,900 --> 00:25:00,000
The reason why it
doesn't change

410
00:25:00,000 --> 00:25:01,570
anything is very simple.

411
00:25:01,570 --> 00:25:06,930
Which is that every but has a
50% chance of having a boy.

412
00:25:06,930 --> 00:25:08,340
So just take the number buts.

413
00:25:08,340 --> 00:25:09,620
Forget about families.

414
00:25:09,620 --> 00:25:11,460
There's a certain
number of buts.

415
00:25:11,460 --> 00:25:13,890
Let's say 50 million buts.

416
00:25:13,890 --> 00:25:16,680
25 million of those
buts will be boys.

417
00:25:16,680 --> 00:25:18,580
And 25 million of those
buts with be girls.

418
00:25:18,580 --> 00:25:20,800
It doesn't matter which
family it happens.

419
00:25:20,800 --> 00:25:24,220
Every but is a separate,
independent event and has a

420
00:25:24,220 --> 00:25:26,080
50% chance of being a boy.

421
00:25:26,080 --> 00:25:30,010
So paradoxical as it sounds,
stopping girls don't change

422
00:25:30,010 --> 00:25:31,000
the sex ratio.

423
00:25:31,000 --> 00:25:32,750
They change the distribution.

424
00:25:32,750 --> 00:25:36,930
Now here's what might be
going on, which is more

425
00:25:36,930 --> 00:25:38,306
complication.

426
00:25:38,306 --> 00:25:38,804
Yeah?

427
00:25:38,804 --> 00:25:42,790
AUDIENCE: But again, but the
relaxation is now there.

428
00:25:42,790 --> 00:25:44,242
Like, it's completely relaxed.

429
00:25:44,242 --> 00:25:45,520
So then you can have it.

430
00:25:45,520 --> 00:25:48,150
So it won't be true if you can
have as many kids as possible.

431
00:25:48,150 --> 00:25:50,185
This is going to make a
difference, though, if you

432
00:25:50,185 --> 00:25:51,910
have to stall--

433
00:25:51,910 --> 00:25:55,930
PROFESSOR: As long as you don't
determine the sex of the

434
00:25:55,930 --> 00:26:01,260
child directly, it
doesn't matter.

435
00:26:01,260 --> 00:26:04,860
Every but is an independent
event with a 50% chance of

436
00:26:04,860 --> 00:26:05,780
being a boy.

437
00:26:05,780 --> 00:26:08,570
So take the number of bus,
multiply it by 50%, you're

438
00:26:08,570 --> 00:26:09,660
going to get the number.

439
00:26:09,660 --> 00:26:10,670
So it makes no difference.

440
00:26:10,670 --> 00:26:15,000
Now let's think of one more
step in the [INAUDIBLE].

441
00:26:15,000 --> 00:26:20,420
Which families would a lot
of girls grow up in?

442
00:26:20,420 --> 00:26:22,440
Would they have more children
or less children?

443
00:26:22,440 --> 00:26:25,580
Would girls grow up on average
in a family with more children

444
00:26:25,580 --> 00:26:26,356
or less children?

445
00:26:26,356 --> 00:26:27,170
AUDIENCE: More children.

446
00:26:27,170 --> 00:26:28,490
PROFESSOR: More children.

447
00:26:28,490 --> 00:26:30,680
So if you think that families
that have more children have

448
00:26:30,680 --> 00:26:37,710
less resources, then you have
a stopping rule, then girls

449
00:26:37,710 --> 00:26:41,110
would be growing up in families
with more children,

450
00:26:41,110 --> 00:26:43,550
and therefore less resources.

451
00:26:43,550 --> 00:26:51,880
And therefore, a stopping rule
can have an effect on child

452
00:26:51,880 --> 00:26:58,070
outcomes, because it leads to
girls all living in large

453
00:26:58,070 --> 00:27:00,970
families and boys all living
in small families.

454
00:27:00,970 --> 00:27:05,480
So boys get all the attention,
the parent, the resources.

455
00:27:05,480 --> 00:27:12,510
The seven girls and one boy
in the girl families, and

456
00:27:12,510 --> 00:27:15,340
therefore the girls end up with
getting less resources.

457
00:27:15,340 --> 00:27:17,860
So that's one possible reason
why you might get this

458
00:27:17,860 --> 00:27:20,400
outcome, because it has
something to do with the

459
00:27:20,400 --> 00:27:22,210
stopping rule, but
not directly.

460
00:27:22,210 --> 00:27:25,450
It has to do with the stopping
rule then generates an

461
00:27:25,450 --> 00:27:27,180
allocation of girls
across families.

462
00:27:27,180 --> 00:27:29,160
Yeah?

463
00:27:29,160 --> 00:27:30,206
AUDIENCE: So that's also true.

464
00:27:30,206 --> 00:27:35,175
But If you think of about even
with the relaxation, imagine a

465
00:27:35,175 --> 00:27:39,840
family in a rich area where
everything's more expensive.

466
00:27:39,840 --> 00:27:42,910
Let's say even if it was a
little more relaxed, they

467
00:27:42,910 --> 00:27:44,990
still limit themselves to say,
OK, we're going to have a

468
00:27:44,990 --> 00:27:46,750
maximum of two kids.

469
00:27:46,750 --> 00:27:51,200
So let's say, yeah, maybe the
first kid that they have is a

470
00:27:51,200 --> 00:27:55,780
girl, but then they say, the
probability of having a girl

471
00:27:55,780 --> 00:27:59,130
for the next one, maybe it's not
really 50%, because they

472
00:27:59,130 --> 00:28:03,130
say, well, we're only going to
have this kid if it's a boy.

473
00:28:03,130 --> 00:28:05,000
PROFESSOR: Yeah, so
what you're saying

474
00:28:05,000 --> 00:28:06,145
is completely right.

475
00:28:06,145 --> 00:28:11,220
The only way you get away from
50-50 is by deliberately or at

476
00:28:11,220 --> 00:28:15,100
least doing something to the
girl that makes her not live.

477
00:28:15,100 --> 00:28:18,330
That's the only way
you get there.

478
00:28:18,330 --> 00:28:22,470
If the girls who are born
live, then you can't

479
00:28:22,470 --> 00:28:23,530
get away from this.

480
00:28:23,530 --> 00:28:25,820
It doesn't matter what your
stopping rule is.

481
00:28:25,820 --> 00:28:28,900
Even if the government makes
complicated policies, the

482
00:28:28,900 --> 00:28:30,390
policies don't change that.

483
00:28:30,390 --> 00:28:32,690
Every but is either
a boy or girl.

484
00:28:36,730 --> 00:28:40,880
So here's the last fact I wanted
you to see, which is

485
00:28:40,880 --> 00:28:43,190
that if you look at these sex
ratios in India, they're

486
00:28:43,190 --> 00:28:44,090
getting worse.

487
00:28:44,090 --> 00:28:48,140
They got worse for a while, now
they have stabilized at a

488
00:28:48,140 --> 00:28:49,460
low number.

489
00:28:49,460 --> 00:28:54,710
So in 1900, there were 972 girls
for boys, men for women,

490
00:28:54,710 --> 00:28:56,260
women for men.

491
00:28:56,260 --> 00:28:59,960
Now there are 933.

492
00:28:59,960 --> 00:29:05,680
In India, you don't see a trend
after 1971 or 1981.

493
00:29:05,680 --> 00:29:13,430
But the historical trend
is very negative.

494
00:29:16,210 --> 00:29:20,480
So that brings us to sort
of the question.

495
00:29:20,480 --> 00:29:23,020
Before we get to policy, in
a sense we need to know

496
00:29:23,020 --> 00:29:24,320
what's going on.

497
00:29:24,320 --> 00:29:26,740
What's going on-- there are
two possible answers.

498
00:29:26,740 --> 00:29:31,010
One is kind of tradition, and
the other is economics.

499
00:29:31,010 --> 00:29:37,800
So one is there's a set of
traditions which make bad

500
00:29:37,800 --> 00:29:38,570
things happen.

501
00:29:38,570 --> 00:29:40,760
But people are not willing
them to happen.

502
00:29:40,760 --> 00:29:45,840
The other is that people are
actually making choices to

503
00:29:45,840 --> 00:29:47,090
make these things happen.

504
00:29:50,250 --> 00:29:52,580
So what we're going to spend
some time doing is trying to

505
00:29:52,580 --> 00:29:54,810
distinguish between
these two views.

506
00:29:54,810 --> 00:29:55,580
Is it tradition?

507
00:29:55,580 --> 00:29:56,830
Is it economics?

508
00:29:59,760 --> 00:30:04,680
So you could imagine just
traditionally there's a theory

509
00:30:04,680 --> 00:30:07,300
which says that girls
need less food.

510
00:30:07,300 --> 00:30:08,190
That's false.

511
00:30:08,190 --> 00:30:09,600
But you believe it.

512
00:30:09,600 --> 00:30:11,750
And you don't actually
actively do anything

513
00:30:11,750 --> 00:30:14,170
to make them die.

514
00:30:14,170 --> 00:30:15,700
You have wrong beliefs.

515
00:30:15,700 --> 00:30:18,630
You think girls need less food,
because adult women are

516
00:30:18,630 --> 00:30:21,000
smaller than adult men.

517
00:30:21,000 --> 00:30:23,270
And so you feed your
girls less.

518
00:30:23,270 --> 00:30:27,140
This is not an attempt to kill
them, it's just a tradition.

519
00:30:27,140 --> 00:30:30,590
But then that could feed back
and have a consequence.

520
00:30:35,810 --> 00:30:37,440
Problem is how do
you test this?

521
00:30:42,740 --> 00:30:43,732
Yeah?

522
00:30:43,732 --> 00:30:51,260
AUDIENCE: If you actually look
at fertility rates, so if the

523
00:30:51,260 --> 00:30:57,210
number of people that are born
are essentially all 50-50, but

524
00:30:57,210 --> 00:30:59,308
then within five years you would
see a clear difference.

525
00:30:59,308 --> 00:31:00,740
So maybe at some point
that's happening.

526
00:31:00,740 --> 00:31:04,655
But even if in turns of birth
rate, if there's a difference,

527
00:31:04,655 --> 00:31:06,460
then it's something else.

528
00:31:06,460 --> 00:31:10,380
PROFESSOR: So one answer to this
question is if these guys

529
00:31:10,380 --> 00:31:14,470
are not getting born, then
obviously something

530
00:31:14,470 --> 00:31:15,890
else is going on.

531
00:31:15,890 --> 00:31:23,910
Now mostly still about five
years, not five, 15 years ago,

532
00:31:23,910 --> 00:31:27,960
you don't see a big trend
in birthrates.

533
00:31:27,960 --> 00:31:31,180
Now you actually see a trend
in birthrates as well,

534
00:31:31,180 --> 00:31:37,145
reported butts they're more
and more reported boys

535
00:31:37,145 --> 00:31:38,120
relative to girls.

536
00:31:38,120 --> 00:31:39,790
But over 15 years ago,
you don't see that.

537
00:31:39,790 --> 00:31:43,910
What you did see was more
different mortality rates for

538
00:31:43,910 --> 00:31:46,540
boys and girls.

539
00:31:46,540 --> 00:31:49,130
So imagine that you wanted
to test this view.

540
00:31:49,130 --> 00:31:52,260
Well the problem is how
do you test it?

541
00:31:52,260 --> 00:31:57,400
You don't actually observe who
eats what in a family.

542
00:31:57,400 --> 00:32:01,120
So it's hard to imagine you
going and figuring out oh,

543
00:32:01,120 --> 00:32:02,390
they're not feeding
their daughter.

544
00:32:02,390 --> 00:32:03,530
They're feeding their son.

545
00:32:03,530 --> 00:32:07,150
That's just too difficult to do,
because everybody eats out

546
00:32:07,150 --> 00:32:08,480
of the same cooked food.

547
00:32:08,480 --> 00:32:10,810
And so it's going to be
very hard to find.

548
00:32:10,810 --> 00:32:14,000
Somebody had a very good
idea of how to do it.

549
00:32:14,000 --> 00:32:17,895
We just looked at specialized
adult goods, like cigarettes.

550
00:32:21,066 --> 00:32:24,402
Your daughter, a five-year-old
daughter, doesn't smoke a

551
00:32:24,402 --> 00:32:28,130
cigarette, nor does your
five-year-old son.

552
00:32:28,130 --> 00:32:34,190
So when a son is born, if
there's an effect on how much

553
00:32:34,190 --> 00:32:43,720
smoking you do, it has nothing
to do with the children, the

554
00:32:43,720 --> 00:32:44,690
children smoking.

555
00:32:44,690 --> 00:32:48,110
So children are not going to be
a direct source of demand.

556
00:32:48,110 --> 00:32:51,270
So why is that good?

557
00:32:51,270 --> 00:32:55,870
Because then you can look at
what happens to smoking when

558
00:32:55,870 --> 00:33:00,550
the family gets a
boy or a girl.

559
00:33:00,550 --> 00:33:03,905
And if girls are fed less,
what would you expect?

560
00:33:08,016 --> 00:33:09,462
AUDIENCE: [INAUDIBLE].

561
00:33:09,462 --> 00:33:12,354
They would smoke more cigarettes
because they have a

562
00:33:12,354 --> 00:33:14,120
larger share of their budget
to spend on cigarettes,

563
00:33:14,120 --> 00:33:16,210
because they'll give
less to the girl.

564
00:33:16,210 --> 00:33:20,040
PROFESSOR: Right, smoking
goes down by less.

565
00:33:20,040 --> 00:33:23,310
So the basic observation you
should see is when a child is

566
00:33:23,310 --> 00:33:28,210
born, adult income goes down,
because adults have less money

567
00:33:28,210 --> 00:33:29,590
to spend on themselves,
because they

568
00:33:29,590 --> 00:33:31,120
will feed the child.

569
00:33:31,120 --> 00:33:36,330
But if they feed the girl less
than the boy, then you'd see

570
00:33:36,330 --> 00:33:43,960
that when a boy is born, they'll
really cut back on

571
00:33:43,960 --> 00:33:44,610
cigarettes.

572
00:33:44,610 --> 00:33:47,950
But when a girl is born,
they'll cut back less.

573
00:33:47,950 --> 00:33:51,970
Because they are feeding
the girl less.

574
00:33:51,970 --> 00:33:56,460
We have a pretty precise
measurement of demand for

575
00:33:56,460 --> 00:33:57,730
adult goods.

576
00:33:57,730 --> 00:33:58,800
So we can use that.

577
00:33:58,800 --> 00:34:00,540
We don't know who smokes.

578
00:34:00,540 --> 00:34:03,080
But we know that the kids are
not doing the smoking.

579
00:34:03,080 --> 00:34:06,060
So therefore, we can by
observing what happens to a

580
00:34:06,060 --> 00:34:08,420
family when the child is born.

581
00:34:08,420 --> 00:34:14,010
We can back out whether or not
there is a difference in the

582
00:34:14,010 --> 00:34:17,219
way you feed boys or girls.

583
00:34:17,219 --> 00:34:20,920
Is that clear why that's better
than trying to measure?

584
00:34:20,920 --> 00:34:23,199
It's very hard to measure
who eats what.

585
00:34:23,199 --> 00:34:28,340
But clearly the kids don't
smoke cigarettes.

586
00:34:28,340 --> 00:34:29,469
And they don't drink alcohol.

587
00:34:29,469 --> 00:34:32,040
So if you look at alcohol, and
cigarettes, and a few other

588
00:34:32,040 --> 00:34:35,900
things that are essentially
adult goods, we should be able

589
00:34:35,900 --> 00:34:38,780
to back out what's happening
when the child is born.

590
00:34:52,219 --> 00:34:53,469
Somebody did that.

591
00:35:01,890 --> 00:35:03,500
So take beverages.

592
00:35:03,500 --> 00:35:06,910
Beverages doesn't
include milk.

593
00:35:06,910 --> 00:35:12,330
So it's soda, and tea, and
all of those things.

594
00:35:16,500 --> 00:35:21,620
This table is horrible, but this
paper was published in a

595
00:35:21,620 --> 00:35:26,260
obscure journal many
years ago.

596
00:35:26,260 --> 00:35:27,660
This is just a scan of that.

597
00:35:27,660 --> 00:35:29,120
There's nothing else
I could find.

598
00:35:29,120 --> 00:35:30,450
There's no electronic
version, et cetera.

599
00:35:33,530 --> 00:35:38,100
What you're looking for is the
effect of a boy versus a girl,

600
00:35:38,100 --> 00:35:42,780
so number of males between
zero and four, versus the

601
00:35:42,780 --> 00:35:45,270
number of females between zero
and four, and the effect of

602
00:35:45,270 --> 00:35:49,350
that on the consumption
of beverages.

603
00:35:49,350 --> 00:35:51,730
And the answer is it has no
effect, that what's in the

604
00:35:51,730 --> 00:35:53,250
bracket is the tea statistic.

605
00:35:53,250 --> 00:35:54,500
The tea statistic is zero.

606
00:35:59,600 --> 00:36:01,160
Both the numbers are zero.

607
00:36:01,160 --> 00:36:02,550
And that's basically
what you find.

608
00:36:08,570 --> 00:36:14,370
You see what you'd expect, which
is that, for example,

609
00:36:14,370 --> 00:36:17,090
when a child is born,
you're just poorer.

610
00:36:17,090 --> 00:36:21,420
So you buy less rice,
less wheat.

611
00:36:21,420 --> 00:36:24,750
But there's no difference
between girls and boys.

612
00:36:24,750 --> 00:36:27,907
Boys and girls have the same
effect, if anything, on both.

613
00:36:32,730 --> 00:36:38,790
Basically there's no evidence
that when boy child is born,

614
00:36:38,790 --> 00:36:42,220
families cut back more than
when a girl child is born.

615
00:36:42,220 --> 00:36:43,970
We don't find any evidence.

616
00:36:43,970 --> 00:36:46,280
And this will be looked
in many countries.

617
00:36:46,280 --> 00:36:50,410
So another tradition story
is the following.

618
00:36:50,410 --> 00:36:55,950
Imagine that tradition is when
the child gets sick, do I take

619
00:36:55,950 --> 00:36:57,650
it to the hospital?

620
00:36:57,650 --> 00:37:00,010
And with the boy, the tradition
is I always take it

621
00:37:00,010 --> 00:37:00,830
to the hospital.

622
00:37:00,830 --> 00:37:09,480
With a girl, I just give her
some water and generally sing

623
00:37:09,480 --> 00:37:12,770
songs to her, but I don't take
her to the hospital,

624
00:37:12,770 --> 00:37:15,900
something like that.

625
00:37:15,900 --> 00:37:18,890
Do we see any evidence
of this?

626
00:37:18,890 --> 00:37:24,360
So this is hard to study,
because we don't actually

627
00:37:24,360 --> 00:37:29,000
observe that many episodes of
life-threatening diseases.

628
00:37:29,000 --> 00:37:34,210
But there's a nice fact that
Elaina Rose identified, which

629
00:37:34,210 --> 00:37:35,460
is the following.

630
00:37:40,330 --> 00:37:45,890
Over time, different parts
of India have droughts at

631
00:37:45,890 --> 00:37:47,370
different points of time.

632
00:37:47,370 --> 00:37:54,120
So some areas have
droughts in 2006.

633
00:37:54,120 --> 00:37:55,730
Some have droughts in 2004.

634
00:37:55,730 --> 00:37:58,100
Some have droughts in 1996.

635
00:37:58,100 --> 00:38:01,005
So a drought is a time
of economic stress.

636
00:38:07,280 --> 00:38:10,210
What's a drought?

637
00:38:10,210 --> 00:38:11,460
Why is it economic stress?

638
00:38:14,629 --> 00:38:17,491
AUDIENCE: Lower end qualities
to low crop yield.

639
00:38:17,491 --> 00:38:20,200
PROFESSOR: Yeah, basically
you can't grow anything.

640
00:38:20,200 --> 00:38:21,340
It's too dry.

641
00:38:21,340 --> 00:38:26,100
So when the rains fail,
crops fail.

642
00:38:26,100 --> 00:38:27,480
Families have no money.

643
00:38:31,890 --> 00:38:33,540
Well, how are they dealing with

644
00:38:33,540 --> 00:38:36,450
this particular emergency?

645
00:38:36,450 --> 00:38:40,150
And there she finds some
evidence that relative to

646
00:38:40,150 --> 00:38:45,640
non-drought years, in drought
years, girls die more.

647
00:38:45,640 --> 00:38:48,060
So that suggests that
emergencies might be

648
00:38:48,060 --> 00:38:49,440
a part of the story.

649
00:38:49,440 --> 00:38:53,140
In emergencies, girls are
not treated as well.

650
00:38:53,140 --> 00:38:56,510
So I gave one example,
which is not

651
00:38:56,510 --> 00:38:57,570
taking her to the hospital.

652
00:38:57,570 --> 00:39:00,350
The other one might be, when I
don't have money, I don't buy

653
00:39:00,350 --> 00:39:03,350
food for her.

654
00:39:03,350 --> 00:39:07,160
So some tradition of that
kind might explain it.

655
00:39:07,160 --> 00:39:09,850
Problem is that droughts are
not frequent enough.

656
00:39:09,850 --> 00:39:11,600
And remember the trends.

657
00:39:11,600 --> 00:39:13,020
The trends are all bad.

658
00:39:13,020 --> 00:39:15,910
They're all getting worse.

659
00:39:15,910 --> 00:39:17,610
India and China are
getting richer.

660
00:39:17,610 --> 00:39:20,670
The dependence on agriculture
in general, these kinds of

661
00:39:20,670 --> 00:39:24,560
extreme events, is
just going down.

662
00:39:24,560 --> 00:39:27,160
Droughts don't matter as
much in India now.

663
00:39:29,890 --> 00:39:33,160
30 years ago, agriculture
was 50% of GDP.

664
00:39:33,160 --> 00:39:35,010
Now it's 25% of GDP.

665
00:39:35,010 --> 00:39:38,060
So basically, droughts matter
less to people.

666
00:39:38,060 --> 00:39:41,090
And since droughts matter less
to people, you would imagine

667
00:39:41,090 --> 00:39:43,940
the trend being favorable
rather than against.

668
00:39:43,940 --> 00:39:45,320
Think of China.

669
00:39:45,320 --> 00:39:48,850
Most people are now
in industry.

670
00:39:48,850 --> 00:39:50,290
They're also richer.

671
00:39:50,290 --> 00:39:52,780
And as you get richer, and you
would imagine that emergencies

672
00:39:52,780 --> 00:39:53,900
are less serious.

673
00:39:53,900 --> 00:39:55,850
Because you always
have some money.

674
00:39:55,850 --> 00:39:58,520
You're no longer at the
margin of starvation.

675
00:39:58,520 --> 00:40:02,290
So the problem with this story
is that it's just inconsistent

676
00:40:02,290 --> 00:40:05,090
with the fact the things are
getting worse over time.

677
00:40:05,090 --> 00:40:07,370
You would think that under this
story things would be

678
00:40:07,370 --> 00:40:09,940
getting better, because it's
an emergency story.

679
00:40:09,940 --> 00:40:13,806
And as countries get richer and
less exposed to risk, you

680
00:40:13,806 --> 00:40:15,500
would see less and less
of this, rather

681
00:40:15,500 --> 00:40:17,300
than more and more.

682
00:40:17,300 --> 00:40:19,465
So that's why this can't be
the whole explanation.

683
00:40:19,465 --> 00:40:20,940
It could be a piece of it.

684
00:40:24,950 --> 00:40:29,700
So going back to our distinction
between kind of

685
00:40:29,700 --> 00:40:37,380
the tradition-based stories and
the more economic stories,

686
00:40:37,380 --> 00:40:40,150
here is an economic fact,
which is striking.

687
00:40:47,660 --> 00:40:50,910
So in other words, what I mean
by an economic story is that

688
00:40:50,910 --> 00:40:54,050
tradition-based stories say that
there's some rule people

689
00:40:54,050 --> 00:40:56,210
use, fixed rule.

690
00:40:56,210 --> 00:40:57,670
They have not decided
on this rule.

691
00:40:57,670 --> 00:40:59,100
This is given to them.

692
00:40:59,100 --> 00:41:02,700
And that rule somehow, under
some circumstances, has bad

693
00:41:02,700 --> 00:41:04,590
outcome for girls.

694
00:41:04,590 --> 00:41:07,860
So when the girl is
sick, you don't

695
00:41:07,860 --> 00:41:09,400
take her to the hospital.

696
00:41:09,400 --> 00:41:10,650
That's a rule.

697
00:41:12,620 --> 00:41:16,450
We're now asking a question,
when the economic incentives

698
00:41:16,450 --> 00:41:20,270
to treat girls better changes,
do you treat them better?

699
00:41:20,270 --> 00:41:23,170
That suggests that you don't
have a fixed rule.

700
00:41:23,170 --> 00:41:27,810
You're actually behaving in a
rational way, quote unquote

701
00:41:27,810 --> 00:41:31,770
"rational way." You're actually
planning to have bad

702
00:41:31,770 --> 00:41:34,540
things happen to girls.

703
00:41:34,540 --> 00:41:36,990
So do you see what the
distinction is between

704
00:41:36,990 --> 00:41:39,810
tradition-based stories
and economic stories?

705
00:41:39,810 --> 00:41:43,320
Economic stories are ones which
essentially say that we

706
00:41:43,320 --> 00:41:46,890
are making active choices which
have consequences for

707
00:41:46,890 --> 00:41:52,990
girls, rather than this is a
rule we use and sometimes bad

708
00:41:52,990 --> 00:41:56,630
things happen, because we
use a particular rule.

709
00:41:56,630 --> 00:41:58,880
Here is a nice example.

710
00:41:58,880 --> 00:42:04,920
So in China, after 1979,
agriculture is liberalized,

711
00:42:04,920 --> 00:42:07,910
meaning people are allowed to
keep the money from going

712
00:42:07,910 --> 00:42:09,270
their crops.

713
00:42:09,270 --> 00:42:12,630
This is the start of China's
economic miracle.

714
00:42:12,630 --> 00:42:15,270
It's called the family
responsibility system, and

715
00:42:15,270 --> 00:42:17,850
it's sort of when everything
in China starts.

716
00:42:21,010 --> 00:42:26,130
Tea is one crop where women have
a comparative advantage

717
00:42:26,130 --> 00:42:27,430
and even absolute advantage.

718
00:42:27,430 --> 00:42:32,230
Tea is something that has to
be picked very patiently.

719
00:42:32,230 --> 00:42:36,970
The most valuable leaves in
tea are the smallest ones.

720
00:42:36,970 --> 00:42:40,360
So here to pick the small
leaves, you have to pick them

721
00:42:40,360 --> 00:42:41,600
without bruising them.

722
00:42:41,600 --> 00:42:44,310
If the leaf is bruised,
the price goes down.

723
00:42:44,310 --> 00:42:54,820
The best leaf teas, like Chinese
teas, sell for $600 a

724
00:42:54,820 --> 00:42:58,180
kilogram, so something of
the order of magnitude.

725
00:42:58,180 --> 00:42:59,660
It's pretty expensive.

726
00:42:59,660 --> 00:43:03,430
So if you actually have the best
teas, you bruise a leaf,

727
00:43:03,430 --> 00:43:04,570
that costs money.

728
00:43:04,570 --> 00:43:06,630
So you better have somebody
who picks with

729
00:43:06,630 --> 00:43:09,380
very delicate fingers.

730
00:43:09,380 --> 00:43:11,670
Whether it's true or not, I
don't know, but it's believed

731
00:43:11,670 --> 00:43:14,180
in China that women are
better for picking

732
00:43:14,180 --> 00:43:16,380
expensive tea leaves.

733
00:43:16,380 --> 00:43:17,410
Maybe they are.

734
00:43:17,410 --> 00:43:21,745
Maybe this is a story that men
made up to make them do it.

735
00:43:21,745 --> 00:43:26,130
But whatever the explanation,
this is the belief.

736
00:43:26,130 --> 00:43:28,280
So given that belief,
what matters is what

737
00:43:28,280 --> 00:43:29,530
people believe here.

738
00:43:29,530 --> 00:43:34,320
Because would imagine that
another crop, which is the

739
00:43:34,320 --> 00:43:36,780
opposite, is fruits.

740
00:43:36,780 --> 00:43:39,840
Fruits need someone to climb
a tree and pick them.

741
00:43:39,840 --> 00:43:44,080
And again in China, the belief
is that men are better at

742
00:43:44,080 --> 00:43:46,770
climbing trees--

743
00:43:46,770 --> 00:43:49,910
maybe again false.

744
00:43:49,910 --> 00:43:53,530
So men are better at
picking fruits.

745
00:43:53,530 --> 00:43:56,460
Women are better
at picking tea.

746
00:43:56,460 --> 00:43:58,390
That's starting fact.

747
00:43:58,390 --> 00:43:59,730
At that's what people believe.

748
00:43:59,730 --> 00:44:01,220
Let's start from there.

749
00:44:01,220 --> 00:44:09,940
Now suppose parents respond
to economic incentives.

750
00:44:09,940 --> 00:44:11,580
And what would you expect
happens when the

751
00:44:11,580 --> 00:44:13,910
price of tea goes up?

752
00:44:13,910 --> 00:44:16,780
What do you think will happen
to their daughter?

753
00:44:16,780 --> 00:44:17,550
Yeah?

754
00:44:17,550 --> 00:44:19,020
AUDIENCE: They'll treat
their daughter better.

755
00:44:19,020 --> 00:44:20,290
PROFESSOR: They'll treat
their daughter better.

756
00:44:20,290 --> 00:44:25,170
They're going to feed her more,
or they're going to take

757
00:44:25,170 --> 00:44:28,640
her to the hospital more
willingly, et cetera.

758
00:44:28,640 --> 00:44:29,790
So you'd expect that.

759
00:44:29,790 --> 00:44:34,690
And likewise, when the price
of fruits goes up, you'd

760
00:44:34,690 --> 00:44:38,350
expect boys to be
treated better.

761
00:44:38,350 --> 00:44:43,810
Now in fact, both price of
fruit and trees go up.

762
00:44:43,810 --> 00:44:45,850
And that's a trend.

763
00:44:45,850 --> 00:44:48,080
So it's hard to use a
trend to study this.

764
00:44:48,080 --> 00:44:49,890
Because everything's going up.

765
00:44:49,890 --> 00:44:51,870
Both tea and fruits
are going up.

766
00:44:51,870 --> 00:44:54,220
So you can't distinguish
between these two.

767
00:44:54,220 --> 00:44:57,030
But one thing you can do is you
can look at areas which

768
00:44:57,030 --> 00:45:02,800
are suitable for tea, areas just
suitable for fruits, and

769
00:45:02,800 --> 00:45:06,670
areas that were suitable
for neither.

770
00:45:06,670 --> 00:45:10,780
Now you can say when tea prices
go up, do we see a

771
00:45:10,780 --> 00:45:16,510
bigger change in the gender
ratio in the tea-suitable

772
00:45:16,510 --> 00:45:19,600
areas relative to the others?

773
00:45:19,600 --> 00:45:25,940
Or you can ask, when the price
of fruits go up, do we see a

774
00:45:25,940 --> 00:45:30,760
bigger change in the gender
ratio in the fruit-favored

775
00:45:30,760 --> 00:45:35,860
areas than in the areas where
neither is favored?

776
00:45:35,860 --> 00:45:39,620
So we're always comparing areas
where, basically, wheat

777
00:45:39,620 --> 00:45:44,300
or rice grows with tea areas
and fruit areas.

778
00:45:44,300 --> 00:45:45,940
Is that clear what
we're doing?

779
00:45:45,940 --> 00:45:49,920
We're asking when the price of
fruits goes up, do we see a

780
00:45:49,920 --> 00:45:53,260
different trend in fruit-growing
areas, relative

781
00:45:53,260 --> 00:45:54,510
to wheat-growing areas?

782
00:45:56,830 --> 00:46:00,690
Is that clear why that's
a good way to

783
00:46:00,690 --> 00:46:01,940
look at this question?

784
00:46:04,230 --> 00:46:05,970
What are we worried about?

785
00:46:05,970 --> 00:46:09,120
If we just said, well, the
price of tea is going up,

786
00:46:09,120 --> 00:46:14,170
what's happening to girls?

787
00:46:14,170 --> 00:46:16,660
Well, the price of tea
is going up, but it's

788
00:46:16,660 --> 00:46:18,200
going up over time.

789
00:46:18,200 --> 00:46:21,880
If the fraction of girls is
going down, we'll get a

790
00:46:21,880 --> 00:46:23,530
negative association
out of that.

791
00:46:23,530 --> 00:46:25,290
Because there are two
things happening.

792
00:46:25,290 --> 00:46:28,000
One is that it's just going
down for other reasons.

793
00:46:28,000 --> 00:46:30,270
The other is prices of
tea is going up for

794
00:46:30,270 --> 00:46:32,170
whatever economic reasons.

795
00:46:32,170 --> 00:46:33,390
And I'm going to
correlate them.

796
00:46:33,390 --> 00:46:35,030
I'll find a negative
correlation.

797
00:46:35,030 --> 00:46:36,190
So that's not what
we want to do.

798
00:46:36,190 --> 00:46:39,140
We want to look at areas which
are suitable for tea.

799
00:46:39,140 --> 00:46:41,570
That's where the price of tea
should have an effect, should

800
00:46:41,570 --> 00:46:44,250
not have an effect in
areas which are not

801
00:46:44,250 --> 00:46:45,320
suitable for tea.

802
00:46:45,320 --> 00:46:48,810
So we're going to compare the
effect of an increase in price

803
00:46:48,810 --> 00:46:53,140
of tea in girls in tea-suitable
areas versus

804
00:46:53,140 --> 00:46:57,260
non-tea-suitable areas,
not compare trends.

805
00:46:57,260 --> 00:46:59,600
This is basically what is called
doing a difference in

806
00:46:59,600 --> 00:46:59,980
difference.

807
00:46:59,980 --> 00:47:05,500
And we'll take high tea price
periods, low tea price

808
00:47:05,500 --> 00:47:08,770
periods, and compare
tea-suitable areas and

809
00:47:08,770 --> 00:47:10,020
non-tea-suitable areas.

810
00:47:13,580 --> 00:47:16,430
You look more lost than
I would expect.

811
00:47:16,430 --> 00:47:19,110
It's a very easy idea.

812
00:47:19,110 --> 00:47:23,240
Somebody else want to say back
what I'm trying to say?

813
00:47:23,240 --> 00:47:27,646
Who's going to volunteer to
improve on that explanation?

814
00:47:27,646 --> 00:47:29,610
I'm going to [INAUDIBLE].

815
00:47:29,610 --> 00:47:30,860
Somebody [INAUDIBLE]?

816
00:47:35,993 --> 00:47:37,466
OK, go ahead.

817
00:47:37,466 --> 00:47:40,657
AUDIENCE: Basically you're
trying to see that if this

818
00:47:40,657 --> 00:47:44,340
does actually have an effect on
how parents treat children.

819
00:47:44,340 --> 00:47:49,800
You'll see girls treated better
in places where tea is

820
00:47:49,800 --> 00:47:53,720
the cash crop or something and
the price increases, versus

821
00:47:53,720 --> 00:47:55,680
areas where tea is not
the cash crop,

822
00:47:55,680 --> 00:47:57,150
but the price increases.

823
00:47:57,150 --> 00:47:59,600
There shouldn't be a change in
how the girls are treated.

824
00:47:59,600 --> 00:48:00,090
PROFESSOR: Correct.

825
00:48:00,090 --> 00:48:04,220
So we're going to compare years
when tea prices went up

826
00:48:04,220 --> 00:48:09,100
with years when tea prices
didn't go up and compare them

827
00:48:09,100 --> 00:48:13,010
in areas where tea is a
plausible crop and areas where

828
00:48:13,010 --> 00:48:15,770
tea cannot be grown.

829
00:48:15,770 --> 00:48:18,800
So the increase in tea prices
should have no effect where

830
00:48:18,800 --> 00:48:21,100
tea cannot be grown.

831
00:48:21,100 --> 00:48:23,760
It will have a spurious
effect, maybe,

832
00:48:23,760 --> 00:48:24,850
but no really effect.

833
00:48:24,850 --> 00:48:26,620
It should have an effect
where tea can be grown.

834
00:48:26,620 --> 00:48:32,010
So we're going to compare tea
possible areas with other

835
00:48:32,010 --> 00:48:35,542
areas, when tea prices go up
versus when it doesn't.

836
00:48:35,542 --> 00:48:36,150
Yeah?

837
00:48:36,150 --> 00:48:38,278
AUDIENCE: How about we think
about it just like you have

838
00:48:38,278 --> 00:48:42,613
you control population, where
the tea price going up

839
00:48:42,613 --> 00:48:43,245
shouldn't have an effect.

840
00:48:43,245 --> 00:48:45,661
And then you have your
non-control, and you just take

841
00:48:45,661 --> 00:48:47,866
you data and subtract your
control from it to see you

842
00:48:47,866 --> 00:48:48,310
[INAUDIBLE].

843
00:48:48,310 --> 00:48:49,630
PROFESSOR: That's
exactly right.

844
00:48:49,630 --> 00:48:51,950
So that's another way to
say the same thing.

845
00:48:51,950 --> 00:48:54,840
Anybody else want to add
anything to that?

846
00:48:54,840 --> 00:48:58,040
So let me show you
some trends.

847
00:48:58,040 --> 00:49:01,950
So that's the basic fact.

848
00:49:01,950 --> 00:49:06,430
So that's tea prices.

849
00:49:06,430 --> 00:49:10,400
Suddenly in '79, you see stock
tea prices just skyrocketed.

850
00:49:10,400 --> 00:49:12,820
That's the first fact.

851
00:49:12,820 --> 00:49:14,420
Tea prices are going up.

852
00:49:14,420 --> 00:49:16,360
Tea production is going up.

853
00:49:16,360 --> 00:49:20,150
Second thing is that if you
look at tea versus other

854
00:49:20,150 --> 00:49:23,510
crops, tea prices are going up
faster than other crops.

855
00:49:23,510 --> 00:49:25,400
So other crops are like grain.

856
00:49:25,400 --> 00:49:29,070
Grain is going up, but tea
is going up faster.

857
00:49:29,070 --> 00:49:31,830
Oil is going up, but tea
is going up faster.

858
00:49:31,830 --> 00:49:33,560
So tea is the crop
that is going up

859
00:49:33,560 --> 00:49:37,750
fastest among these crops.

860
00:49:37,750 --> 00:49:39,450
Look at this one.

861
00:49:39,450 --> 00:49:42,680
Here we're comparing
fruits with tea.

862
00:49:42,680 --> 00:49:45,000
And fruits are going
up much faster.

863
00:49:45,000 --> 00:49:50,180
Fruits with other crops, like
grain, oil, and cotton, and

864
00:49:50,180 --> 00:49:52,310
again, fruits are going
up much faster.

865
00:49:52,310 --> 00:49:57,360
So basically, what matters is
not the absolute price.

866
00:49:57,360 --> 00:50:02,410
The relative price matters,
because it's what makes you

867
00:50:02,410 --> 00:50:04,280
want to do more tea.

868
00:50:04,280 --> 00:50:07,330
So tea areas are benefiting,
relative to those other areas.

869
00:50:13,190 --> 00:50:15,010
So this is a complicated
graph, so

870
00:50:15,010 --> 00:50:16,560
let's do this slowly.

871
00:50:16,560 --> 00:50:19,010
Try to understand what's
in this graph.

872
00:50:19,010 --> 00:50:27,690
So what's in this graph is areas
which grow a lot of tea.

873
00:50:35,240 --> 00:50:38,055
This is the coefficient
in the regression.

874
00:50:41,610 --> 00:50:49,080
So this is the effect of having
more area devoted to

875
00:50:49,080 --> 00:50:50,970
tea in a particular year.

876
00:50:50,970 --> 00:51:00,490
So what this is saying is that
before 1979, you see tea

877
00:51:00,490 --> 00:51:01,650
prices are low.

878
00:51:01,650 --> 00:51:07,260
In those years, if you look at
the fraction of males, the

879
00:51:07,260 --> 00:51:09,510
fraction of males
bounces around.

880
00:51:09,510 --> 00:51:16,800
But the fact that tea is sown
has essentially very close to

881
00:51:16,800 --> 00:51:19,110
zero effect.

882
00:51:19,110 --> 00:51:23,040
After 1979, the fact that
tea is sown has

883
00:51:23,040 --> 00:51:23,980
a much bigger effect.

884
00:51:23,980 --> 00:51:28,950
So this is the effect of how
much land you put into tea in

885
00:51:28,950 --> 00:51:31,660
a particular year,
so year by year.

886
00:51:31,660 --> 00:51:39,470
So in that year, what's
happening, in 1983, how many

887
00:51:39,470 --> 00:51:46,800
boys versus girls were born
in high tea areas

888
00:51:46,800 --> 00:51:48,410
relative to low tea areas?

889
00:51:48,410 --> 00:51:50,150
That's always the exercise.

890
00:51:50,150 --> 00:51:56,480
So think of one more percent
of land devoted to tea.

891
00:51:56,480 --> 00:52:00,920
What is the effect of that
on the fraction of

892
00:52:00,920 --> 00:52:04,940
boys born in 1983?

893
00:52:04,940 --> 00:52:07,430
That's what that says.

894
00:52:07,430 --> 00:52:11,300
So the main point to take
away is you see a

895
00:52:11,300 --> 00:52:15,110
big drop after 1979.

896
00:52:15,110 --> 00:52:21,740
So tea areas, the fraction of
boys born goes down right when

897
00:52:21,740 --> 00:52:22,850
you think it should go down.

898
00:52:22,850 --> 00:52:24,637
It is when the price
starts going up.

899
00:52:27,380 --> 00:52:30,730
So that's what this is saying.

900
00:52:30,730 --> 00:52:35,390
Before that, the average
is kind of at 0.15.

901
00:52:35,390 --> 00:52:39,720
After that, the average
is about 0.35.

902
00:52:39,720 --> 00:52:44,650
So the fraction of boys goes
down after the change.

903
00:52:44,650 --> 00:52:49,130
This is what happens in
the fruit areas--

904
00:52:49,130 --> 00:52:51,510
the opposite.

905
00:52:51,510 --> 00:52:55,330
After '79, fruit prices
go up also a lot.

906
00:52:55,330 --> 00:52:56,580
And fruit favors boys.

907
00:52:59,600 --> 00:53:02,170
The ratio used to be lower.

908
00:53:02,170 --> 00:53:04,992
It goes up afterwards.

909
00:53:04,992 --> 00:53:05,483
Yeah?

910
00:53:05,483 --> 00:53:06,956
AUDIENCE: Are these significant
results?

911
00:53:06,956 --> 00:53:10,910
PROFESSOR: Yes, so the average
is significantly different.

912
00:53:10,910 --> 00:53:11,850
You're right.

913
00:53:11,850 --> 00:53:14,230
Year by year, they're
not significant.

914
00:53:14,230 --> 00:53:16,930
But that's because it's a
small sample of people.

915
00:53:16,930 --> 00:53:21,580
But if you take the average of
these periods, the averages

916
00:53:21,580 --> 00:53:22,900
are significantly different.

917
00:53:22,900 --> 00:53:25,470
AUDIENCE: I'm just having
trouble understanding how so

918
00:53:25,470 --> 00:53:36,809
if women are better at picking
tea leaves, the economics make

919
00:53:36,809 --> 00:53:40,753
sense, but I can't see families
consciously choosing

920
00:53:40,753 --> 00:53:44,600
to have more girls.

921
00:53:44,600 --> 00:53:47,720
In 12 years, they'll
begin to--

922
00:53:47,720 --> 00:53:49,290
PROFESSOR: Well, yes and no.

923
00:53:49,290 --> 00:53:50,450
You're right with that.

924
00:53:50,450 --> 00:53:52,910
But on the other hand, it also
sends a signal which reminds

925
00:53:52,910 --> 00:53:55,750
them, oh well, you know.

926
00:53:55,750 --> 00:53:56,420
Two things--

927
00:53:56,420 --> 00:53:59,280
one, you might imagine that
the women might be more

928
00:53:59,280 --> 00:54:03,780
pro-girls, plausibly.

929
00:54:03,780 --> 00:54:07,240
And if you think that
women income share

930
00:54:07,240 --> 00:54:08,440
goes up in the family--

931
00:54:08,440 --> 00:54:09,800
AUDIENCE: They have
more stakes.

932
00:54:09,800 --> 00:54:10,610
PROFESSOR: They have more say.

933
00:54:10,610 --> 00:54:12,530
So that's one possible
mechanism.

934
00:54:12,530 --> 00:54:19,780
The second mechanism is just
that maybe when girls' incomes

935
00:54:19,780 --> 00:54:27,330
go up, they just look at it and
say, well, after all, all

936
00:54:27,330 --> 00:54:28,770
the women are working
these days.

937
00:54:28,770 --> 00:54:30,090
Why not have another girl?

938
00:54:30,090 --> 00:54:31,620
They'll be fine.

939
00:54:31,620 --> 00:54:34,540
They'll find a job as well.

940
00:54:34,540 --> 00:54:37,880
I can't tell what's going
on at that level.

941
00:54:37,880 --> 00:54:40,040
But it's clear that this
result is significant.

942
00:54:40,040 --> 00:54:41,290
It's actually substantial.

943
00:54:45,550 --> 00:54:48,070
Final set of thoughts--

944
00:54:52,150 --> 00:54:53,400
suppose this goes on.

945
00:54:56,085 --> 00:54:57,990
Let me back up.

946
00:54:57,990 --> 00:54:59,415
What's going on?

947
00:54:59,415 --> 00:55:01,245
How are they losing
all these girls?

948
00:55:01,245 --> 00:55:06,330
If it's an economic mechanism,
that says that there is

949
00:55:06,330 --> 00:55:08,360
incentive to lose the girls.

950
00:55:08,360 --> 00:55:11,340
How are they actually
doing it?

951
00:55:11,340 --> 00:55:12,590
What's going on?

952
00:55:19,754 --> 00:55:24,160
Unfortunately, it's all not very
nice, what's going on.

953
00:55:24,160 --> 00:55:26,250
So two things are happening.

954
00:55:26,250 --> 00:55:30,750
One is basically the technology
for pre-birth

955
00:55:30,750 --> 00:55:34,530
detection of gender of the
child has improved a lot.

956
00:55:37,160 --> 00:55:41,690
If you do amniocentesis, which
is an expensive procedure, you

957
00:55:41,690 --> 00:55:44,550
can basically very high
probability predict the gender

958
00:55:44,550 --> 00:55:46,070
of the child.

959
00:55:46,070 --> 00:55:50,380
The reason why rich areas are
doing the worst is precisely

960
00:55:50,380 --> 00:55:51,120
because of that.

961
00:55:51,120 --> 00:55:53,345
Rich areas are using
this technology.

962
00:55:56,000 --> 00:55:59,320
It's an expensive technology.

963
00:55:59,320 --> 00:56:02,230
You spend money to use it.

964
00:56:02,230 --> 00:56:06,830
You see in like the richer parts
of India, signs up would

965
00:56:06,830 --> 00:56:15,030
say, would you rather
spend basically $100

966
00:56:15,030 --> 00:56:22,380
now or $10,000 later?

967
00:56:22,380 --> 00:56:25,660
The idea is that you're to give
dowry for your daughter

968
00:56:25,660 --> 00:56:26,720
to get her married.

969
00:56:26,720 --> 00:56:28,000
And that's your cost.

970
00:56:28,000 --> 00:56:33,190
And so you should really get
ready to spend, whatever, $100

971
00:56:33,190 --> 00:56:37,460
now, identify her,
and then have an

972
00:56:37,460 --> 00:56:40,380
abortion, rather than spend.

973
00:56:40,380 --> 00:56:41,960
You see these signs up.

974
00:56:41,960 --> 00:56:44,000
They're all the illegal.

975
00:56:44,000 --> 00:56:46,450
Amniocentesis about eight
years ago was

976
00:56:46,450 --> 00:56:48,490
made illegal in India.

977
00:56:48,490 --> 00:56:51,010
It's still completely
prevalent.

978
00:56:51,010 --> 00:56:55,420
And since people are willing
to try to have many babies,

979
00:56:55,420 --> 00:57:00,240
they're even willing to do
it by using a much weaker

980
00:57:00,240 --> 00:57:01,750
technology, which
is a sonography.

981
00:57:01,750 --> 00:57:06,280
Sonography, which is a
technology that is legal, it

982
00:57:06,280 --> 00:57:09,760
doesn't identify the gender of
the child all the time, but

983
00:57:09,760 --> 00:57:12,060
over 25% of the time, it does.

984
00:57:12,060 --> 00:57:16,040
And when it does, then you can
immediately abort the child.

985
00:57:16,040 --> 00:57:18,900
So they're using technologies
to identify.

986
00:57:18,900 --> 00:57:21,650
This is why this is happening
more in richer areas, because

987
00:57:21,650 --> 00:57:23,390
technologies cost money.

988
00:57:23,390 --> 00:57:27,030
And the poorest areas are ones
where the technology is least

989
00:57:27,030 --> 00:57:29,580
used, because it's
just expensive.

990
00:57:29,580 --> 00:57:35,190
So you see basically a very,
very clear correlation between

991
00:57:35,190 --> 00:57:37,980
where the technology arrives
and the fraction

992
00:57:37,980 --> 00:57:40,390
of boys being born.

993
00:57:40,390 --> 00:57:42,210
So that's one thing
that's going on.

994
00:57:42,210 --> 00:57:46,370
Second thing is going on almost
surely is a lot of

995
00:57:46,370 --> 00:57:51,730
post-birth, I think, murder is
the only word you could use.

996
00:57:51,730 --> 00:57:54,250
I think unfortunately there's
a lot of, I think, children

997
00:57:54,250 --> 00:57:56,820
being disposed of.

998
00:57:56,820 --> 00:58:00,900
There are villages which report
to have, probably

999
00:58:00,900 --> 00:58:04,110
report, there's a news report I
think from like the Chinese

1000
00:58:04,110 --> 00:58:12,220
news agency which says that we
haven't had a girl born in our

1001
00:58:12,220 --> 00:58:13,900
village in the last
five years.

1002
00:58:13,900 --> 00:58:17,370
There's a news report of the
village chief proudly

1003
00:58:17,370 --> 00:58:21,400
announcing that there was no
girl born in his village in

1004
00:58:21,400 --> 00:58:22,140
the last five years.

1005
00:58:22,140 --> 00:58:25,940
So I think there is a lot of
just direct, deliberate action

1006
00:58:25,940 --> 00:58:27,860
going on there.

1007
00:58:27,860 --> 00:58:32,540
Now, there are many things
to think about here.

1008
00:58:32,540 --> 00:58:33,350
How do you stop it?

1009
00:58:33,350 --> 00:58:34,430
The government has tried.

1010
00:58:34,430 --> 00:58:37,210
It tried to have technologies.

1011
00:58:37,210 --> 00:58:39,905
There's a bunch of programs
around now, for example.

1012
00:58:42,460 --> 00:58:45,860
India has a number of states
where they have programs that

1013
00:58:45,860 --> 00:58:51,580
basically promise parents a
certain amount of money if

1014
00:58:51,580 --> 00:58:54,380
your girl graduates
from high school--

1015
00:58:54,380 --> 00:58:55,350
quite a lot of money.

1016
00:58:55,350 --> 00:58:56,640
So that's one way
to [INAUDIBLE]

1017
00:58:56,640 --> 00:58:59,560
incentives based on
girls' arrival.

1018
00:59:06,870 --> 00:59:10,000
But it's not clear how
well that works.

1019
00:59:10,000 --> 00:59:15,200
Because in some ways, it's
something that's based on 20

1020
00:59:15,200 --> 00:59:15,940
years later.

1021
00:59:15,940 --> 00:59:19,540
And you don't necessarily think
that people are that

1022
00:59:19,540 --> 00:59:21,870
responsive to it.

1023
00:59:21,870 --> 00:59:24,530
The other thing you can think
about is is there any economic

1024
00:59:24,530 --> 00:59:32,090
force that this, by itself, sets
off, which will then make

1025
00:59:32,090 --> 00:59:33,340
re-equilibrium.

1026
00:59:33,340 --> 00:59:36,910
So one issue that arises is
what's happening to the

1027
00:59:36,910 --> 00:59:38,150
marriage market?

1028
00:59:38,150 --> 00:59:39,150
So what do you think
will happen to the

1029
00:59:39,150 --> 00:59:44,484
marriage market in China?

1030
00:59:44,484 --> 00:59:47,340
AUDIENCE: Well, if there are
fewer women, then each one is

1031
00:59:47,340 --> 00:59:49,250
essentially worth more.

1032
00:59:49,250 --> 00:59:50,500
PROFESSOR: Yeah.

1033
00:59:53,290 --> 00:59:56,570
Dowry should go down, and bride
price should go up.

1034
01:00:05,600 --> 01:00:08,540
Why this hasn't happened already
in India, especially,

1035
01:00:08,540 --> 01:00:11,170
is that men tend to marry
women who are

1036
01:00:11,170 --> 01:00:13,914
younger than them.

1037
01:00:13,914 --> 01:00:18,930
If the population is growing,
and people marry women who are

1038
01:00:18,930 --> 01:00:28,790
younger than them, then even if
there are fewer girls born,

1039
01:00:28,790 --> 01:00:33,640
the girls are from a different
cohort, from a younger cohort.

1040
01:00:33,640 --> 01:00:36,830
So the population is growing, so
there's more girls around.

1041
01:00:41,220 --> 01:00:42,630
So that's one problem.

1042
01:00:42,630 --> 01:00:44,750
This force might operate,
but only if

1043
01:00:44,750 --> 01:00:46,110
population growth stops.

1044
01:00:46,110 --> 01:00:48,640
So it's going to operate more in
China where the population

1045
01:00:48,640 --> 01:00:51,170
growth has stopped than in India
where the population

1046
01:00:51,170 --> 01:00:52,800
growth is still going on.

1047
01:00:52,800 --> 01:00:54,810
As long as there's population
growth, you're going to see

1048
01:00:54,810 --> 01:00:57,975
this possibility of just going
down in the cohort chain.

1049
01:01:01,200 --> 01:01:04,430
So you might imagine that the
price of girls will go up, but

1050
01:01:04,430 --> 01:01:08,850
it won't go up if people marry
much younger women.

1051
01:01:08,850 --> 01:01:10,340
That's one issue.

1052
01:01:10,340 --> 01:01:17,830
Second issue, so in China, one
of things that people are

1053
01:01:17,830 --> 01:01:20,750
often puzzled about
is what's going on

1054
01:01:20,750 --> 01:01:22,000
with the savings rate?

1055
01:01:24,750 --> 01:01:27,910
The savings rate in China,
personal savings, is

1056
01:01:27,910 --> 01:01:29,830
historically unprecedented.

1057
01:01:29,830 --> 01:01:34,602
People save 30% of their
incomes, as it is compared to

1058
01:01:34,602 --> 01:01:38,760
the US, which used to save zero
till recently, and after

1059
01:01:38,760 --> 01:01:42,090
the crisis has gone up to 7%.

1060
01:01:42,090 --> 01:01:46,300
So personal savings rates
in China are massive.

1061
01:01:46,300 --> 01:01:49,700
So one puzzle people have often
been thinking about is

1062
01:01:49,700 --> 01:01:51,980
what's going on with that?

1063
01:01:51,980 --> 01:01:54,170
There's a very nice paper which
does the following.

1064
01:01:56,680 --> 01:02:03,110
If I look at areas which have
high gender ratios, where

1065
01:02:03,110 --> 01:02:08,890
women are really scarce,
are people saving

1066
01:02:08,890 --> 01:02:11,690
more in those areas?

1067
01:02:11,690 --> 01:02:12,940
Why would they save more?

1068
01:02:17,310 --> 01:02:22,270
Why would you save more if
you're a parent of boy?

1069
01:02:22,270 --> 01:02:25,520
So he particularly asked, if
you're a parent of a boy, in

1070
01:02:25,520 --> 01:02:27,800
China it's easy, because
people have one child.

1071
01:02:27,800 --> 01:02:29,760
So they're either a parent of
a boy or parent of a girl.

1072
01:02:29,760 --> 01:02:34,040
If you're parent of a boy in an
area where the gender ratio

1073
01:02:34,040 --> 01:02:39,190
is particularly bad, so very
few girls, it turns

1074
01:02:39,190 --> 01:02:40,700
out you save more.

1075
01:02:40,700 --> 01:02:41,950
Why would that be the case?

1076
01:02:46,062 --> 01:02:47,470
AUDIENCE: The price
of a bride.

1077
01:02:47,470 --> 01:02:50,350
PROFESSOR: Yeah, you have
to buy your son a bride.

1078
01:02:50,350 --> 01:02:52,160
So people are saving more.

1079
01:02:52,160 --> 01:02:55,720
So basically they suggest that
people are beginning to save

1080
01:02:55,720 --> 01:03:00,760
more in order to buy their
child a bride.

1081
01:03:00,760 --> 01:03:02,840
And the way you buy it
is you don't go and

1082
01:03:02,840 --> 01:03:03,960
buy her in the market.

1083
01:03:03,960 --> 01:03:07,020
But basically women have become
very picky about not

1084
01:03:07,020 --> 01:03:11,040
marrying into families, which
can't afford to give their son

1085
01:03:11,040 --> 01:03:18,380
a separate apartment to move
into when they get married.

1086
01:03:18,380 --> 01:03:20,450
China has very poor
mortgage markets.

1087
01:03:20,450 --> 01:03:24,230
So you can't really take out
a mortgage to buy a house.

1088
01:03:24,230 --> 01:03:26,670
So the only you buy a house is
you borrow from your parents.

1089
01:03:26,670 --> 01:03:28,450
Your parents give you a gift.

1090
01:03:28,450 --> 01:03:32,520
So parents now, it's become a
status issue for parents to

1091
01:03:32,520 --> 01:03:33,920
marry this sons.

1092
01:03:33,920 --> 01:03:37,190
And the way you marry your son
is by buying him an apartment

1093
01:03:37,190 --> 01:03:38,640
when he gets married.

1094
01:03:38,640 --> 01:03:40,550
So if you don't have
enough wealth, you

1095
01:03:40,550 --> 01:03:41,720
can't buy an apartment.

1096
01:03:41,720 --> 01:03:43,610
And then your son can't
get married.

1097
01:03:43,610 --> 01:03:49,410
So this seems to be sort of one
of the facts that might

1098
01:03:49,410 --> 01:03:52,520
actually re-equilibriate China,
because it seems like

1099
01:03:52,520 --> 01:03:56,390
one thing that's going on in
China right now is this

1100
01:03:56,390 --> 01:03:58,035
massive increase in savings.

1101
01:04:00,900 --> 01:04:04,560
The problem with this method in
increasing savings is it's

1102
01:04:04,560 --> 01:04:06,490
a zero sum game.

1103
01:04:06,490 --> 01:04:08,370
There are only so many girls.

1104
01:04:08,370 --> 01:04:13,890
So the fact that I'm saving a
lot for my son just means my

1105
01:04:13,890 --> 01:04:16,680
son can have a bride, but
not anybody else.

1106
01:04:16,680 --> 01:04:18,220
It's not now that the number
of women go up

1107
01:04:18,220 --> 01:04:19,880
because they save.

1108
01:04:19,880 --> 01:04:23,150
So this is not really
solving the problem.

1109
01:04:23,150 --> 01:04:25,920
It's squeezing parents.

1110
01:04:25,920 --> 01:04:30,870
So if I were to save $1 more in
order to buy my son a house

1111
01:04:30,870 --> 01:04:35,010
so he can get married, then
somebody else is going to want

1112
01:04:35,010 --> 01:04:35,890
to do the same thing.

1113
01:04:35,890 --> 01:04:38,650
He's going to try to save even
more than me and buy a house

1114
01:04:38,650 --> 01:04:40,160
for his son.

1115
01:04:40,160 --> 01:04:44,050
So that competition is a
very kind of a negative

1116
01:04:44,050 --> 01:04:44,880
competition.

1117
01:04:44,880 --> 01:04:47,810
Everybody's getting hurt
by that competition.

1118
01:04:47,810 --> 01:04:53,700
So that should make parents
really rethink this strategy.

1119
01:04:53,700 --> 01:04:57,230
The one sign of hope I see is
this one, which is the market

1120
01:04:57,230 --> 01:05:00,390
forces seem to be kind
of operating to make

1121
01:05:00,390 --> 01:05:01,430
this problem better.

1122
01:05:01,430 --> 01:05:05,610
That's true in China, much less
true in India, because

1123
01:05:05,610 --> 01:05:07,230
India's population is growing.

1124
01:05:07,230 --> 01:05:10,550
So there's still this minimizing
of adjustment.

1125
01:05:10,550 --> 01:05:14,810
OK, so questions, comments,
before we stop?

1126
01:05:14,810 --> 01:05:15,804
Yeah?

1127
01:05:15,804 --> 01:05:19,780
AUDIENCE: In China, what you
see about the increase in

1128
01:05:19,780 --> 01:05:22,140
girls again, because parents
see that they don't have to

1129
01:05:22,140 --> 01:05:23,260
save money to have a girl?

1130
01:05:23,260 --> 01:05:24,640
PROFESSOR: So you
might see that.

1131
01:05:24,640 --> 01:05:26,310
You aren't seeing it yet.

1132
01:05:26,310 --> 01:05:30,460
But maybe once it gets clear
that your daughter gets to

1133
01:05:30,460 --> 01:05:34,630
live in a wonderful apartment,
et cetera, and having a son is

1134
01:05:34,630 --> 01:05:38,030
a huge pain because you have
to now find ways to pay for

1135
01:05:38,030 --> 01:05:41,800
his apartment, you might start
preferring daughters.

1136
01:05:41,800 --> 01:05:43,220
So maybe we'll see that.

1137
01:05:49,180 --> 01:05:50,860
OK, thank you.