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

2
00:00:02,460 --> 00:00:03,870
Commons license.

3
00:00:03,870 --> 00:00:06,910
Your support will help MIT
OpenCourseWare continue to

4
00:00:06,910 --> 00:00:10,560
offer high quality educational
resources for free.

5
00:00:10,560 --> 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
00:00:22,550 --> 00:00:26,100
PROFESSOR: All right, so what
we're going to do for the next

9
00:00:26,100 --> 00:00:31,490
few lectures, really through the
end of the course, is turn

10
00:00:31,490 --> 00:00:34,280
away from what we focused on for
most of this course, which

11
00:00:34,280 --> 00:00:37,710
is efficiency, and start
talking about equity.

12
00:00:37,710 --> 00:00:42,620
So most of this course, we've
talked about efficiency and

13
00:00:42,620 --> 00:00:45,750
the principles that lead
to efficient outcomes.

14
00:00:45,750 --> 00:00:48,700
We haven't talked a whole lot
about fairness and equity.

15
00:00:48,700 --> 00:00:51,160
We've mentioned it in passing
for things like why you want a

16
00:00:51,160 --> 00:00:54,980
minimum wage, et cetera, but we
haven't really focused on

17
00:00:54,980 --> 00:00:58,710
how society should consider
efficiency and equity as two

18
00:00:58,710 --> 00:01:02,130
different goals, because
clearly it does.

19
00:01:02,130 --> 00:01:05,850
So for example, a classic
example of the fact that

20
00:01:05,850 --> 00:01:09,530
efficient outcomes may not be
equally equitable is the

21
00:01:09,530 --> 00:01:12,390
perfectly competitive market
versus the perfectly price

22
00:01:12,390 --> 00:01:16,850
discriminating monopolist.
Remember that both outcomes,

23
00:01:16,850 --> 00:01:20,200
both a perfectly competitive
market and a perfectly price

24
00:01:20,200 --> 00:01:23,080
discriminating monopolist,
both lead to

25
00:01:23,080 --> 00:01:25,610
maximum social welfare.

26
00:01:25,610 --> 00:01:29,790
So in the terms we've used in
this course, both provide

27
00:01:29,790 --> 00:01:32,230
equal social welfare.

28
00:01:32,230 --> 00:01:34,250
The difference, of course, being
that in the perfectly

29
00:01:34,250 --> 00:01:36,760
price discriminating monopolist
case, the producer

30
00:01:36,760 --> 00:01:39,920
gets it all, whereas in the
perfectly competitive case,

31
00:01:39,920 --> 00:01:42,710
it's shared between producers
and consumers.

32
00:01:42,710 --> 00:01:44,960
So we actually care.

33
00:01:44,960 --> 00:01:47,260
We actually in the end are going
to care as a society,

34
00:01:47,260 --> 00:01:50,160
not just about social welfare,
but who's getting the shares

35
00:01:50,160 --> 00:01:50,650
of the pie.

36
00:01:50,650 --> 00:01:52,610
In other words, we're not going
to care just about the

37
00:01:52,610 --> 00:01:54,985
size of the pie, but who's
getting what shares.

38
00:01:57,670 --> 00:02:01,410
So first of all, even for a
given size of the pie, we have

39
00:02:01,410 --> 00:02:03,330
to decide who gets
what shares.

40
00:02:03,330 --> 00:02:06,130
Moreover, it gets much more
complicated than that because

41
00:02:06,130 --> 00:02:08,550
the size of the pie
isn't fixed.

42
00:02:08,550 --> 00:02:12,570
And typically, in a society's
efforts to redistribute

43
00:02:12,570 --> 00:02:15,490
shares, they shrink the pie.

44
00:02:15,490 --> 00:02:19,290
That is, societies typically
face the classical equity

45
00:02:19,290 --> 00:02:20,540
efficiency trade off.

46
00:02:27,980 --> 00:02:32,430
Which is that in any effort
to redistribute resources,

47
00:02:32,430 --> 00:02:34,790
inevitably we'll shrink
the total amount

48
00:02:34,790 --> 00:02:37,670
of resources available.

49
00:02:37,670 --> 00:02:43,730
And that's roughly speaking,
because the efficient outcome

50
00:02:43,730 --> 00:02:45,280
is where you start from.

51
00:02:45,280 --> 00:02:46,620
You start from an outcome
where you've

52
00:02:46,620 --> 00:02:48,420
maximized social welfare.

53
00:02:48,420 --> 00:02:51,970
So any intervention that tries
to deviate from that by moving

54
00:02:51,970 --> 00:02:54,810
resources across groups, by
definition will move us away

55
00:02:54,810 --> 00:02:56,100
from that efficient outcome.

56
00:02:56,100 --> 00:02:58,260
We're starting with the
efficient outcome, let's say a

57
00:02:58,260 --> 00:03:00,870
permanently price discriminating
monopolist,

58
00:03:00,870 --> 00:03:01,760
that's efficient.

59
00:03:01,760 --> 00:03:04,060
Any efforts we have to take
some money away from that

60
00:03:04,060 --> 00:03:06,400
monopolist and give it to
other people will end up

61
00:03:06,400 --> 00:03:09,140
causing distortions and reducing
the total size of

62
00:03:09,140 --> 00:03:12,340
total social welfare.

63
00:03:12,340 --> 00:03:15,070
So basically, we're always
going to face this equity

64
00:03:15,070 --> 00:03:17,240
efficiency trade off, and really
the question we want to

65
00:03:17,240 --> 00:03:19,750
ask in this lecture is,
is it worth it?

66
00:03:19,750 --> 00:03:22,810
How do we think about whether
it's worth it to redistribute

67
00:03:22,810 --> 00:03:26,580
from rich to poor, even if along
the way, we shrink the

68
00:03:26,580 --> 00:03:29,200
total size of the pie, even if
we have to deviate from the

69
00:03:29,200 --> 00:03:31,870
welfare maximizing point.

70
00:03:31,870 --> 00:03:33,890
And the best way to think about
this I think is due to

71
00:03:33,890 --> 00:03:35,470
an analogy developed by
a famous economist

72
00:03:35,470 --> 00:03:38,460
named Arthur Okun.

73
00:03:38,460 --> 00:03:42,060
And this is Okun's leaky
bucket analogy.

74
00:03:42,060 --> 00:03:43,970
So Okun's leaky bucket analogy
is the following, he said,

75
00:03:43,970 --> 00:03:48,210
look, let's say as a society,
we agree that we should

76
00:03:48,210 --> 00:03:49,470
redistribute from
rich to poor.

77
00:03:49,470 --> 00:03:51,400
And I'll come later to why we
might agree that, but let's

78
00:03:51,400 --> 00:03:54,210
just say we agree as a society
that we should redistribute

79
00:03:54,210 --> 00:03:55,290
from the richest members to the

80
00:03:55,290 --> 00:03:57,810
poorest members of society.

81
00:03:57,810 --> 00:04:00,090
And let's think of that
redistribution as being

82
00:04:00,090 --> 00:04:02,840
literally the rich people put
money in a bucket, and you

83
00:04:02,840 --> 00:04:04,640
carry that bucket and then dump
it out in front of the

84
00:04:04,640 --> 00:04:05,650
poor people.

85
00:04:05,650 --> 00:04:08,240
Let's say that's the way
redistribution happened.

86
00:04:08,240 --> 00:04:11,540
Well, probably we'd have general
agreement that if Bill

87
00:04:11,540 --> 00:04:14,240
Gates could put $1 in a bucket
and it could be carried and

88
00:04:14,240 --> 00:04:17,890
given to a homeless person,
that's a good thing.

89
00:04:17,890 --> 00:04:19,839
We probably in general agree
that if you distribute from

90
00:04:19,839 --> 00:04:22,310
the richest people to the
poorest people, and $1 from a

91
00:04:22,310 --> 00:04:24,390
rich person becomes $1
for a poor person,

92
00:04:24,390 --> 00:04:26,560
that's a good thing.

93
00:04:26,560 --> 00:04:29,220
However, what if along the
way, some of that dollar

94
00:04:29,220 --> 00:04:30,890
leaked out?

95
00:04:30,890 --> 00:04:34,410
What if Bill Gates put $1 in the
bucket, but by the time it

96
00:04:34,410 --> 00:04:37,510
got to the poor person,
it was only $0.80.

97
00:04:37,510 --> 00:04:39,840
You might say, OK, that's still
fine. $0.80 to a poor

98
00:04:39,840 --> 00:04:40,740
person is a lot.

99
00:04:40,740 --> 00:04:42,440
Bill Gates isn't going
to miss $1, but $0.80

100
00:04:42,440 --> 00:04:43,750
to them might matter.

101
00:04:43,750 --> 00:04:45,920
What if it's only $0.50?

102
00:04:45,920 --> 00:04:49,090
What if Bill Gates put $1 in, by
the time you carry it over,

103
00:04:49,090 --> 00:04:51,690
it leaks out so much so the poor
person only gets $0.50?

104
00:04:51,690 --> 00:04:53,610
What if it's $0.10?

105
00:04:53,610 --> 00:04:56,380
What if literally every dollar
Bill Gates gives up, the poor

106
00:04:56,380 --> 00:04:58,670
person only gets $0.10, then you
might say, gee, it's not

107
00:04:58,670 --> 00:05:00,550
entirely clear it's
worth it anymore.

108
00:05:00,550 --> 00:05:02,990
I mean, yeah, $0.10 is worth
more to a poor person than a

109
00:05:02,990 --> 00:05:05,040
$1 is to Bill Gates, but that's
really wasteful to have

110
00:05:05,040 --> 00:05:07,800
$0.90 just leak out
along the way.

111
00:05:07,800 --> 00:05:12,700
So basically, what we have to
do is ask basically how does

112
00:05:12,700 --> 00:05:17,430
society think about transfers
when some of what happens when

113
00:05:17,430 --> 00:05:20,080
you transfer from rich to poor
is you lose some efficiency

114
00:05:20,080 --> 00:05:23,110
along the way, something leaks
out of the bucket?

115
00:05:23,110 --> 00:05:24,870
And that's basically what we're
going to talk about in

116
00:05:24,870 --> 00:05:27,320
the next two lectures.

117
00:05:27,320 --> 00:05:29,390
We're going to do that
in four steps.

118
00:05:29,390 --> 00:05:33,195
We're first going to ask, how
does society value transfers?

119
00:05:35,730 --> 00:05:40,240
How does society think about the
value of $1 to Bill Gates

120
00:05:40,240 --> 00:05:41,740
versus $1 to a poor person?

121
00:05:41,740 --> 00:05:43,800
How do we think about the value
of those dollars in

122
00:05:43,800 --> 00:05:45,360
different hands, something
we haven't

123
00:05:45,360 --> 00:05:48,240
talked about this semester?

124
00:05:48,240 --> 00:05:50,920
We then are going to talk about
the facts on income

125
00:05:50,920 --> 00:05:51,690
distribution.

126
00:05:51,690 --> 00:05:54,510
How has the distribution of
dollars changed over time in

127
00:05:54,510 --> 00:05:58,770
the US, and how do we compare
it internationally?

128
00:05:58,770 --> 00:06:01,185
Then we're going to talk about
the sources of leakage.

129
00:06:04,790 --> 00:06:08,310
That is, what causes this bucket
to be leaky, in fact,

130
00:06:08,310 --> 00:06:09,730
what causes the equity
efficiency

131
00:06:09,730 --> 00:06:12,460
trade off in practice.

132
00:06:12,460 --> 00:06:15,200
And then finally, we're going to
talk about what governments

133
00:06:15,200 --> 00:06:21,470
do, what government does to
redistribute resources.

134
00:06:21,470 --> 00:06:24,460
How does the US government in
particular redistribute

135
00:06:24,460 --> 00:06:27,790
resources from higher to
lower income groups?

136
00:06:27,790 --> 00:06:28,610
So that's what we're
going to talk about

137
00:06:28,610 --> 00:06:31,370
the next two lectures.

138
00:06:31,370 --> 00:06:35,510
So we're going to start with
the question of how does

139
00:06:35,510 --> 00:06:39,870
society value transfers, or
more generally, how do we

140
00:06:39,870 --> 00:06:46,330
think about the socially
optimal

141
00:06:46,330 --> 00:06:47,580
allocation of resources?

142
00:06:54,820 --> 00:06:57,520
The key word here being
allocation.

143
00:06:57,520 --> 00:06:59,960
So we're no longer just going to
talk about the entire size

144
00:06:59,960 --> 00:07:02,540
of the social welfare triangle,
we're now going to

145
00:07:02,540 --> 00:07:06,010
talk about who gets what, and
how do we think about the

146
00:07:06,010 --> 00:07:11,760
socially optimal distribution of
those resources, allocation

147
00:07:11,760 --> 00:07:14,010
or distribution of
those resources.

148
00:07:14,010 --> 00:07:17,530
And to do so, we're going to
have to take an extra somewhat

149
00:07:17,530 --> 00:07:20,500
uncomfortable step we haven't
taken before.

150
00:07:20,500 --> 00:07:23,120
Before, we've just talked about
a purely mathematical

151
00:07:23,120 --> 00:07:26,120
concept of maximizing the
size of a triangle.

152
00:07:26,120 --> 00:07:29,460
Now we've got to actually put
some value judgment on who

153
00:07:29,460 --> 00:07:31,970
gets what within
that triangle.

154
00:07:31,970 --> 00:07:33,850
And the way we make that value
judgment is through the

155
00:07:33,850 --> 00:07:36,095
introduction of what we call
a social welfare function.

156
00:07:40,430 --> 00:07:43,730
A social welfare function,
which is literally the

157
00:07:43,730 --> 00:07:46,440
mathematical representation,
because we're all

158
00:07:46,440 --> 00:07:48,460
about the math here.

159
00:07:48,460 --> 00:07:51,930
It's a mathematical
representation of how society

160
00:07:51,930 --> 00:07:54,740
values different groups.

161
00:07:54,740 --> 00:07:58,700
So social welfare function is
some function of the utility

162
00:07:58,700 --> 00:08:02,370
of individual one, the utility
of individual two, all the way

163
00:08:02,370 --> 00:08:04,980
to the utility of individual
350 million, however many

164
00:08:04,980 --> 00:08:07,530
people we have now in the US.

165
00:08:07,530 --> 00:08:10,510
It's literally a mathematical
representation of how we take

166
00:08:10,510 --> 00:08:14,400
every individual's utility and
come up with a measure of

167
00:08:14,400 --> 00:08:15,650
social welfare.

168
00:08:17,810 --> 00:08:20,170
That's what the social welfare
function's going to be.

169
00:08:20,170 --> 00:08:20,850
So think about that.

170
00:08:20,850 --> 00:08:24,120
Let's look at figure 23-1,
which shows what we call

171
00:08:24,120 --> 00:08:26,900
isowelfare curves.

172
00:08:26,900 --> 00:08:31,320
Think about a society with two
people, Ned and Homer, and

173
00:08:31,320 --> 00:08:33,850
think about the government's
decision,

174
00:08:33,850 --> 00:08:35,870
or think about society--

175
00:08:35,870 --> 00:08:37,070
excuse the word government,
think about

176
00:08:37,070 --> 00:08:38,270
society right now--

177
00:08:38,270 --> 00:08:42,610
society's valuation of the
allocation of resources across

178
00:08:42,610 --> 00:08:45,370
these two individuals.

179
00:08:45,370 --> 00:08:49,030
Each of these curves is meant
to represent a social

180
00:08:49,030 --> 00:08:49,990
indifference curve.

181
00:08:49,990 --> 00:08:51,510
In other words, think
of society as

182
00:08:51,510 --> 00:08:52,550
having a utility function.

183
00:08:52,550 --> 00:08:56,140
Think of this being a social
utility function, and the

184
00:08:56,140 --> 00:08:58,930
utility function has
indifference curves.

185
00:08:58,930 --> 00:09:00,870
So society has some indifference
curves.

186
00:09:00,870 --> 00:09:06,380
So what this depiction says is
society's indifferent between

187
00:09:06,380 --> 00:09:10,270
Homer having u1h and Ned having
u1n, that is, with Ned

188
00:09:10,270 --> 00:09:12,520
having a lot and Homer
having a little.

189
00:09:12,520 --> 00:09:16,640
Or Homer having u2h and Ned
having u2n, that is Homer

190
00:09:16,640 --> 00:09:18,980
having a lot and Ned
having a little.

191
00:09:18,980 --> 00:09:20,770
So I've just made this up.

192
00:09:20,770 --> 00:09:23,075
I've drawn this curve such that
society is indifferent

193
00:09:23,075 --> 00:09:24,325
between those two allocations.

194
00:09:26,920 --> 00:09:30,090
And basically, these isowelfare
curves will look

195
00:09:30,090 --> 00:09:32,000
just like our indifference
curves.

196
00:09:32,000 --> 00:09:35,620
So we've talked in this course
about choosing between goods.

197
00:09:35,620 --> 00:09:38,300
We've talked about choosing
between states of the world,

198
00:09:38,300 --> 00:09:39,800
injured not injured.

199
00:09:39,800 --> 00:09:42,660
We've talked about choosing
between periods of time, today

200
00:09:42,660 --> 00:09:44,090
versus tomorrow.

201
00:09:44,090 --> 00:09:46,220
Now I'm talking about the
hardest one of all, choosing

202
00:09:46,220 --> 00:09:51,520
between people, choosing between
the utility of people.

203
00:09:51,520 --> 00:09:53,960
But it's the same damn
principle, it's the same stuff

204
00:09:53,960 --> 00:09:57,780
we've always done, which is
there's some utility function,

205
00:09:57,780 --> 00:09:59,300
now we call it a social
welfare function.

206
00:09:59,300 --> 00:10:02,330
You maximize it across
its elements--

207
00:10:02,330 --> 00:10:05,020
here it's the utility of Ned
and Homer, and you develop

208
00:10:05,020 --> 00:10:06,190
some indifference curves.

209
00:10:06,190 --> 00:10:07,780
And they have all the properties
indifference curves

210
00:10:07,780 --> 00:10:10,410
always had, further out is
better, that a society is

211
00:10:10,410 --> 00:10:13,850
happier if both Ned and Homer
have more, and along the

212
00:10:13,850 --> 00:10:16,760
curve, you're indifferent.

213
00:10:16,760 --> 00:10:20,910
So these are representations
of how society feels about

214
00:10:20,910 --> 00:10:23,090
transfers across people.

215
00:10:23,090 --> 00:10:26,190
So the big question then
becomes, what does this social

216
00:10:26,190 --> 00:10:28,270
welfare function look like?

217
00:10:28,270 --> 00:10:30,580
Mathematically, how do you
represent this incredibly hard

218
00:10:30,580 --> 00:10:34,790
question of the trade
off across people?

219
00:10:34,790 --> 00:10:38,560
And this is an incredibly deep
question, but the standard

220
00:10:38,560 --> 00:10:42,080
social welfare function that's
used is something due to

221
00:10:42,080 --> 00:10:45,190
Jeremy Bentham, a famous
English philosopher.

222
00:10:45,190 --> 00:10:47,650
You can actually still see his
head if you go to University

223
00:10:47,650 --> 00:10:49,390
College London.

224
00:10:49,390 --> 00:10:50,660
Actually, it's not a
representation of his head,

225
00:10:50,660 --> 00:10:53,510
they actually had him stuffed
and his head on display, but

226
00:10:53,510 --> 00:10:55,050
then apparently, students would
take it at night and use

227
00:10:55,050 --> 00:10:56,060
it for soccer.

228
00:10:56,060 --> 00:10:57,880
So they now just have a
representation of his head

229
00:10:57,880 --> 00:10:59,560
there at UCL.

230
00:10:59,560 --> 00:11:05,190
But Jeremy Bentham developed
what we call utilitarianism,

231
00:11:05,190 --> 00:11:08,160
the utilitarian social
welfare function.

232
00:11:08,160 --> 00:11:13,160
And utilitarianism simply says
that the social welfare of

233
00:11:13,160 --> 00:11:21,100
society is simply the sum of
each individual's utility.

234
00:11:21,100 --> 00:11:27,890
So u1 plus u2 plus dot dot
dot plus u350 million.

235
00:11:27,890 --> 00:11:30,720
That's social welfare.

236
00:11:30,720 --> 00:11:34,660
That is the social welfare of
US society, is the sum of

237
00:11:34,660 --> 00:11:37,070
everyone's utility.

238
00:11:37,070 --> 00:11:39,000
Very straightforward.

239
00:11:39,000 --> 00:11:42,650
So it's a linear social
welfare function.

240
00:11:42,650 --> 00:11:47,530
So what this says is this says
you would maximize social

241
00:11:47,530 --> 00:11:53,330
welfare by transferring from
any one individual to any

242
00:11:53,330 --> 00:11:56,910
other individual, if the first
individual has a higher

243
00:11:56,910 --> 00:12:00,090
utility that the second
individual.

244
00:12:00,090 --> 00:12:04,750
So basically you maximize
social welfare, is that

245
00:12:04,750 --> 00:12:07,300
basically if you--

246
00:12:07,300 --> 00:12:08,620
I'm sorry, let me say
it another way.

247
00:12:08,620 --> 00:12:13,350
If transferring $1 from me to
you makes you better off than

248
00:12:13,350 --> 00:12:15,660
it makes me worse off,
then we should do it.

249
00:12:18,590 --> 00:12:22,370
Now, your first instinct might
be, well, this isn't very

250
00:12:22,370 --> 00:12:24,830
liberal in a sense, this is
just saying everybody's

251
00:12:24,830 --> 00:12:25,680
utility is the same.

252
00:12:25,680 --> 00:12:28,590
That is, we consider Bill Gates
is one of these people,

253
00:12:28,590 --> 00:12:30,110
and I'm one of these people,
and you're one of these

254
00:12:30,110 --> 00:12:31,750
people, and the homeless guy
in Harvard Square is one of

255
00:12:31,750 --> 00:12:32,540
these people.

256
00:12:32,540 --> 00:12:36,190
And they're all just added up.

257
00:12:36,190 --> 00:12:39,110
How is that fair?

258
00:12:39,110 --> 00:12:40,840
In fact, let me ask the
question differently.

259
00:12:40,840 --> 00:12:45,020
Why despite that is utilitarian
wealth, social

260
00:12:45,020 --> 00:12:47,370
welfare function can it be
perceived as fairly liberal,

261
00:12:47,370 --> 00:12:49,160
despite the fact that everyone
just gets added up?

262
00:12:49,160 --> 00:12:52,390
Why in fact is social welfare
function that's utilitarian

263
00:12:52,390 --> 00:12:54,880
call for transfers from Bill
Gates to homeless person?

264
00:12:54,880 --> 00:12:55,360
Yeah?

265
00:12:55,360 --> 00:12:57,290
AUDIENCE: [UNINTELLIGIBLE]

266
00:12:57,290 --> 00:12:58,400
PROFESSOR: Exactly.

267
00:12:58,400 --> 00:13:01,570
If you maximize social welfare,
what it's going to

268
00:13:01,570 --> 00:13:04,870
say is set marginal
utilities equal.

269
00:13:04,870 --> 00:13:06,910
If you differentiate this with
respect to everybody's

270
00:13:06,910 --> 00:13:10,600
utility, you're going to get
that the rule that you get

271
00:13:10,600 --> 00:13:13,610
from maximizing, the first order
condition mathematically

272
00:13:13,610 --> 00:13:16,750
is going to be that everyone's
margin utility should be equal

273
00:13:16,750 --> 00:13:18,930
with utilitarian social welfare
function, which is

274
00:13:18,930 --> 00:13:20,820
very radical.

275
00:13:20,820 --> 00:13:23,550
That says that we should
redistribute until Bill Gates

276
00:13:23,550 --> 00:13:27,850
has the same marginal utility
as the homeless person.

277
00:13:27,850 --> 00:13:29,800
Thought of that way, that's
incredibly radical

278
00:13:29,800 --> 00:13:30,570
redistribution.

279
00:13:30,570 --> 00:13:32,420
It's not weighting Bill Gates
equally to the homeless

280
00:13:32,420 --> 00:13:34,860
person, it's weighting his
utility equal, but it's saying

281
00:13:34,860 --> 00:13:38,180
we should have massive
transfers.

282
00:13:38,180 --> 00:13:41,950
In fact, basically we should
equalize income in society,

283
00:13:41,950 --> 00:13:43,880
effectively.

284
00:13:43,880 --> 00:13:47,470
That basically, what
we should do is--

285
00:13:47,470 --> 00:13:49,260
in fact, let me be
more precise.

286
00:13:49,260 --> 00:13:52,360
If individuals are identical,
which they're not, but if

287
00:13:52,360 --> 00:13:55,820
individuals had identical
tastes, the utilitarian social

288
00:13:55,820 --> 00:13:58,860
welfare function would say
that social welfare is

289
00:13:58,860 --> 00:14:02,580
maximized with an equal
distribution of income.

290
00:14:02,580 --> 00:14:04,810
Pretty radical.

291
00:14:04,810 --> 00:14:07,570
If individuals are identical,
social welfare is maximized

292
00:14:07,570 --> 00:14:09,920
with the equal distribution of
income, despite the fact we

293
00:14:09,920 --> 00:14:12,910
started with this pretty
un-obnoxious looking function,

294
00:14:12,910 --> 00:14:14,850
we're simply going to weight
everybody equally.

295
00:14:14,850 --> 00:14:16,700
We're not even saying we care
more about poor people than

296
00:14:16,700 --> 00:14:18,730
rich people, we're saying weight
everybody equally.

297
00:14:18,730 --> 00:14:22,030
That fairly innocuous looking
function delivers the radical

298
00:14:22,030 --> 00:14:24,620
implication that if everyone is
identical, we should have

299
00:14:24,620 --> 00:14:27,970
an equal distribution of
income in society.

300
00:14:27,970 --> 00:14:30,770
And that's a typical social
welfare function

301
00:14:30,770 --> 00:14:31,590
that we look at.

302
00:14:31,590 --> 00:14:35,485
But that actually is not
considered really a

303
00:14:35,485 --> 00:14:36,840
particularly lefty social
welfare function.

304
00:14:36,840 --> 00:14:38,830
In fact, if you talk to the left
side of the philosophical

305
00:14:38,830 --> 00:14:41,380
distribution, they'll
say this is not an

306
00:14:41,380 --> 00:14:43,610
appropriate measure to use.

307
00:14:43,610 --> 00:14:49,350
They much prefer something like
the Rawlsian criteria,

308
00:14:49,350 --> 00:14:52,506
named for the philosopher
John Rawls.

309
00:14:52,506 --> 00:14:57,370
The Rawlsian social welfare
function is that the role that

310
00:14:57,370 --> 00:15:01,430
society's goal should be to
maximize the well-being of its

311
00:15:01,430 --> 00:15:03,780
worst off member.

312
00:15:03,780 --> 00:15:07,720
That a fair and just society
is one which maximizes the

313
00:15:07,720 --> 00:15:11,000
well-being of the worst
off person in society.

314
00:15:11,000 --> 00:15:18,610
So that says social welfare is
equal to the min of u1, u2,

315
00:15:18,610 --> 00:15:19,860
dot dot dot dot.

316
00:15:23,100 --> 00:15:25,330
So it's a max-a-min criteria.

317
00:15:25,330 --> 00:15:30,120
You want to maximize the min,
maximize the minimum utility

318
00:15:30,120 --> 00:15:33,130
in society is the Rawlsian
social welfare function.

319
00:15:35,780 --> 00:15:44,980
Now, in some situations,
this can deliver--

320
00:15:44,980 --> 00:15:48,820
in fact, if everyone's
identical, this would deliver

321
00:15:48,820 --> 00:15:51,820
in many cases the same outcome
as utilitarianism.

322
00:15:51,820 --> 00:15:54,750
But in some cases, it's much
more radically lefty.

323
00:15:54,750 --> 00:15:56,340
Think about it this way.

324
00:15:56,340 --> 00:16:01,500
Suppose we had a situation where
everybody but me and

325
00:16:01,500 --> 00:16:05,940
you, but me and this guy
have equal incomes.

326
00:16:05,940 --> 00:16:10,070
Everybody in the world's equal
income is at $40,000.

327
00:16:10,070 --> 00:16:14,490
But my income is $1 million,
and his income is $39,999,

328
00:16:14,490 --> 00:16:19,300
he's got $1 less income
than the rest of you.

329
00:16:19,300 --> 00:16:22,080
And let's say that a new
government is running for

330
00:16:22,080 --> 00:16:25,200
office that proposes to take
away all of my money, so I'm

331
00:16:25,200 --> 00:16:27,480
down to $40,000, and
give him $1 and

332
00:16:27,480 --> 00:16:30,242
bring him up to $40,000.

333
00:16:30,242 --> 00:16:33,060
Now in utilitarianism, that
would not be an appropriate

334
00:16:33,060 --> 00:16:36,120
thing to do, because clearly
my utility is going to fall

335
00:16:36,120 --> 00:16:38,580
more from going from 1 million
to 40,000 than his is going to

336
00:16:38,580 --> 00:16:42,240
go from going 39,999
to 40,000.

337
00:16:42,240 --> 00:16:45,020
So under utilitarianism this
would be a bad thing to do.

338
00:16:45,020 --> 00:16:47,280
But in Rawlsianism, it'd be a
good thing to do, because he's

339
00:16:47,280 --> 00:16:48,780
the worst off member.

340
00:16:48,780 --> 00:16:51,320
So it doesn't matter what
happens to me, it only matters

341
00:16:51,320 --> 00:16:53,200
what happens to him.

342
00:16:53,200 --> 00:16:55,640
So Rawlsiansim is an incredibly
radical theory,

343
00:16:55,640 --> 00:16:58,510
which posits that we only care
what happens to the poorest

344
00:16:58,510 --> 00:17:01,190
guy, even if we have to like
take all the money away from

345
00:17:01,190 --> 00:17:02,980
the richest guys,
we don't care.

346
00:17:02,980 --> 00:17:06,579
We only care about what happens
to the poorest guy.

347
00:17:06,579 --> 00:17:11,490
So that's a very radical sort of
lefty view of how we should

348
00:17:11,490 --> 00:17:12,740
think about distribution.

349
00:17:14,980 --> 00:17:20,020
On the other hand, we have
a very radical--

350
00:17:20,020 --> 00:17:22,829
well, some would say a more
radical view on the right,

351
00:17:22,829 --> 00:17:29,430
which is a Nozickian social
welfare function, named for

352
00:17:29,430 --> 00:17:33,830
the philosopher Robert Nozick,
also at Harvard.

353
00:17:33,830 --> 00:17:35,960
Rawls and Nozick were
both at Harvard.

354
00:17:35,960 --> 00:17:37,880
Nozick said this is just
way too radical.

355
00:17:37,880 --> 00:17:40,490
He said utilitarianism is
way too radical, way

356
00:17:40,490 --> 00:17:41,660
too far to the left.

357
00:17:41,660 --> 00:17:43,450
Forget Rawlsianiasm,
that's just nutty.

358
00:17:43,450 --> 00:17:45,420
Even utilitarianism is way too
far to the left, because he

359
00:17:45,420 --> 00:17:50,040
says, basically what should
matter in society is an equal

360
00:17:50,040 --> 00:17:53,160
distribution of resources--

361
00:17:53,160 --> 00:17:56,110
I'm sorry, an equal distribution
of opportunities.

362
00:17:56,110 --> 00:17:58,850
And conditional on equal
distribution of opportunities,

363
00:17:58,850 --> 00:18:01,190
if individuals take those
opportunities and choose to do

364
00:18:01,190 --> 00:18:03,400
outcomes which give them
different incomes, that's

365
00:18:03,400 --> 00:18:05,190
their problem.

366
00:18:05,190 --> 00:18:08,460
So for example, let's say that
we started with everyone in

367
00:18:08,460 --> 00:18:12,580
the world started with an equal
distribution of income.

368
00:18:12,580 --> 00:18:16,150
But let's say that MIT students
loved to hear me

369
00:18:16,150 --> 00:18:20,320
lecture, and were willing to pay
$100 to hear me lecture.

370
00:18:20,320 --> 00:18:23,750
And so at the end of the year,
every MIT student ended up

371
00:18:23,750 --> 00:18:27,430
$100 poorer and I ended
up a ton richer.

372
00:18:27,430 --> 00:18:29,670
Well, under both utilitarianism,
and certainly

373
00:18:29,670 --> 00:18:33,180
under Rawlsiansim, that is an
outcome which is not social

374
00:18:33,180 --> 00:18:36,550
welfare maximizing, because I
don't care about that last

375
00:18:36,550 --> 00:18:38,180
$100 whereas you guys care
a little bit about

376
00:18:38,180 --> 00:18:40,310
giving up the $100.

377
00:18:40,310 --> 00:18:42,390
So clearly social welfare would
be maximized if I gave

378
00:18:42,390 --> 00:18:44,660
some of that money
back to you.

379
00:18:44,660 --> 00:18:46,880
Under Rawlsianism, clearly I
should give it all back to

380
00:18:46,880 --> 00:18:51,610
you, because all we care about
is you guys being $100 poorer.

381
00:18:51,610 --> 00:18:52,850
Nozick says this is crazy.

382
00:18:52,850 --> 00:18:55,690
This is a voluntary
transaction.

383
00:18:55,690 --> 00:18:59,890
You guys voluntarily paid me
$100, making me richer.

384
00:18:59,890 --> 00:19:03,060
Why should there be any
redistribution?

385
00:19:03,060 --> 00:19:05,100
Why should social welfare
be any lower from that

386
00:19:05,100 --> 00:19:06,710
transaction?

387
00:19:06,710 --> 00:19:08,380
How could a voluntary
transaction

388
00:19:08,380 --> 00:19:10,280
lower social welfare?

389
00:19:10,280 --> 00:19:16,100
And this is sort of a right
wing approach to thinking

390
00:19:16,100 --> 00:19:19,210
about social welfare, which is
look, everybody should have

391
00:19:19,210 --> 00:19:21,730
equal opportunities.

392
00:19:21,730 --> 00:19:23,520
It's not fair to have some
person not have access to

393
00:19:23,520 --> 00:19:24,420
education and things.

394
00:19:24,420 --> 00:19:27,120
But once everybody's got equal
opportunities, there's no

395
00:19:27,120 --> 00:19:29,090
reason why if you make voluntary
choices that lead to

396
00:19:29,090 --> 00:19:32,680
unequal distribution of
income we should care.

397
00:19:32,680 --> 00:19:36,030
Now this is actually
very compelling.

398
00:19:36,030 --> 00:19:40,530
It has a lot of merit, and
really should make us think

399
00:19:40,530 --> 00:19:42,140
about our emphasis should be on

400
00:19:42,140 --> 00:19:44,430
opportunities rather than outcomes.

401
00:19:44,430 --> 00:19:48,480
But it has one flaw, which is
that in fact, most of the

402
00:19:48,480 --> 00:19:52,375
differences in society aren't
due to choices, but rather due

403
00:19:52,375 --> 00:19:56,890
to outcomes outside people's
control, often just luck.

404
00:19:56,890 --> 00:20:00,210
So basically, if it was true
that we all started equal and

405
00:20:00,210 --> 00:20:03,460
that any differences in income
came through our choices, then

406
00:20:03,460 --> 00:20:06,000
this would be a very powerful
way of looking at the world.

407
00:20:06,000 --> 00:20:08,580
But in fact, if you look at
most differences in income

408
00:20:08,580 --> 00:20:10,890
across people, they're not due
to their choices, or to

409
00:20:10,890 --> 00:20:12,720
effort, or to other things you
can actually figure out how to

410
00:20:12,720 --> 00:20:14,460
measure, they're due to
unmeasured things.

411
00:20:14,460 --> 00:20:17,160
Often it looks like luck.

412
00:20:17,160 --> 00:20:22,952
And in that case, you would care
about the fact that I end

413
00:20:22,952 --> 00:20:24,150
up richer than you.

414
00:20:24,150 --> 00:20:25,420
So if I end up richer than you
because you're willing to pay

415
00:20:25,420 --> 00:20:26,090
me, that's one thing.

416
00:20:26,090 --> 00:20:27,740
But if I'm richer than you just
because I won the lottery

417
00:20:27,740 --> 00:20:32,060
and you didn't, that's a
different thing altogether.

418
00:20:32,060 --> 00:20:35,690
And so basically, whether how we
feel about the equality of

419
00:20:35,690 --> 00:20:39,230
opportunities view versus the
equality of outcomes view

420
00:20:39,230 --> 00:20:41,450
depends very much on what we
think is the source of

421
00:20:41,450 --> 00:20:42,860
difference in incomes.

422
00:20:42,860 --> 00:20:47,050
If we think it has to do with
effort and choices, then

423
00:20:47,050 --> 00:20:50,410
people can often favor this
view, that if people by their

424
00:20:50,410 --> 00:20:52,880
effort and choices made more
money, let them keep it.

425
00:20:52,880 --> 00:20:56,040
But if people made more money
through luck and things beyond

426
00:20:56,040 --> 00:20:58,810
their control, then people tend
towards more this view,

427
00:20:58,810 --> 00:21:01,010
saying look, if you got your
money through luck and things,

428
00:21:01,010 --> 00:21:03,010
there's no reason why we should
have such an unequal

429
00:21:03,010 --> 00:21:04,630
distribution of income.

430
00:21:04,630 --> 00:21:07,270
And that's sort of another
tension in thinking about

431
00:21:07,270 --> 00:21:08,600
these social welfare
functions.

432
00:21:08,600 --> 00:21:10,630
There are no right
answers here.

433
00:21:10,630 --> 00:21:12,395
I don't mean through my tone
or word choice to imply

434
00:21:12,395 --> 00:21:13,130
there's any right answer.

435
00:21:13,130 --> 00:21:15,400
I'm just trying to lay out the
arguments for these different

436
00:21:15,400 --> 00:21:18,750
ways of thinking about it.

437
00:21:18,750 --> 00:21:23,480
Then finally, the last way we
could think about it, and to

438
00:21:23,480 --> 00:21:26,430
many people, really the most
compelling, is through what we

439
00:21:26,430 --> 00:21:27,940
call commodity egalitarianism.

440
00:21:36,030 --> 00:21:39,140
Commodity egalitarianism, which
is a fancy name for

441
00:21:39,140 --> 00:21:43,360
saying look, it doesn't
matter who has what.

442
00:21:43,360 --> 00:21:46,960
It just matters that everybody
can survive.

443
00:21:46,960 --> 00:21:49,313
This is kind of like a
reinterpretation of Rawls.

444
00:21:52,080 --> 00:21:54,600
It's almost a mix of Rawls
and Nozick in a way.

445
00:21:54,600 --> 00:21:57,670
It's saying look, what matters
is that people aren't

446
00:21:57,670 --> 00:22:00,790
starving, that they have decent
shelter, decent health

447
00:22:00,790 --> 00:22:03,060
care, et cetera, you can
define your minimum.

448
00:22:03,060 --> 00:22:07,570
But once you've defined a
minimum, we shouldn't care

449
00:22:07,570 --> 00:22:09,520
about how rich people get
beyond that minimum.

450
00:22:12,310 --> 00:22:16,180
So basically, the key thing is
that we should not care--

451
00:22:16,180 --> 00:22:18,690
what commodity egalitarianism
focuses on, the insight it

452
00:22:18,690 --> 00:22:21,390
brings, which is an important
one, is it's not clear why

453
00:22:21,390 --> 00:22:22,780
relative should matter.

454
00:22:22,780 --> 00:22:24,740
What should matter
is absolute.

455
00:22:24,740 --> 00:22:26,040
It's not clear why I should
care how much

456
00:22:26,040 --> 00:22:27,770
richer I am than you.

457
00:22:27,770 --> 00:22:30,010
We should just care that you
have enough to live a decent,

458
00:22:30,010 --> 00:22:31,420
socially acceptable life.

459
00:22:31,420 --> 00:22:35,290
And then if I want to get
rich, God bless me.

460
00:22:35,290 --> 00:22:38,490
So it's kind of like saying we
care about the minimum, but

461
00:22:38,490 --> 00:22:40,270
it's not all we care about.

462
00:22:40,270 --> 00:22:42,420
We care about making sure
there's a minimum decent

463
00:22:42,420 --> 00:22:45,060
standard of living, defined
however you want to do it.

464
00:22:45,060 --> 00:22:46,100
Different people choose choose
different things.

465
00:22:46,100 --> 00:22:47,965
For some people, it's just
food and shelter.

466
00:22:47,965 --> 00:22:50,170
Other people could include
health care, or clothing, or

467
00:22:50,170 --> 00:22:51,330
other things.

468
00:22:51,330 --> 00:22:53,230
I'm not saying what the
right minimum is.

469
00:22:53,230 --> 00:22:56,500
Let's define some minimum, and
then say beyond that if guys

470
00:22:56,500 --> 00:22:57,940
get rich, God bless them.

471
00:22:57,940 --> 00:23:01,680
That's their prerogative, as
long as the poor people have

472
00:23:01,680 --> 00:23:03,870
enough to survive.

473
00:23:03,870 --> 00:23:06,490
And this is an incredibly
interesting, compelling view,

474
00:23:06,490 --> 00:23:08,680
because it's a radically
different view, which says

475
00:23:08,680 --> 00:23:11,480
relatives don't matter,
absolutes are all that matter.

476
00:23:11,480 --> 00:23:13,655
All they're making sure is that
the absolutely people at

477
00:23:13,655 --> 00:23:17,490
the bottom have a decent
standard of living.

478
00:23:17,490 --> 00:23:20,980
And that's a very different
view as well.

479
00:23:20,980 --> 00:23:25,750
So basically, social welfare,
when we talked about utility

480
00:23:25,750 --> 00:23:27,595
functions, I didn't really talk
much, I just said, you

481
00:23:27,595 --> 00:23:28,970
write down a utility function,
we often use

482
00:23:28,970 --> 00:23:30,790
square root of c, whatever.

483
00:23:30,790 --> 00:23:32,530
I didn't really talk much about
the form, I just said it

484
00:23:32,530 --> 00:23:34,250
should have diminishing
marginal utility, and

485
00:23:34,250 --> 00:23:36,470
otherwise I said sort
of all bets are off.

486
00:23:36,470 --> 00:23:39,450
With social welfare functions,
it's a lot more open-ended and

487
00:23:39,450 --> 00:23:41,520
a lot harder, and we don't
even really know what the

488
00:23:41,520 --> 00:23:42,790
right standards are.

489
00:23:42,790 --> 00:23:48,090
But it's just saying that we
can't get away from asking

490
00:23:48,090 --> 00:23:49,540
this awkward question.

491
00:23:49,540 --> 00:23:51,960
Because ultimately society does
end up with an unequal

492
00:23:51,960 --> 00:23:55,040
distribution of resources, and
given that, we can't get away

493
00:23:55,040 --> 00:23:58,180
from asking the question of
how do we feel about that?

494
00:23:58,180 --> 00:24:01,960
Economics is about asking
uncomfortable questions.

495
00:24:01,960 --> 00:24:04,330
And this is an uncomfortable
question we have to answer,

496
00:24:04,330 --> 00:24:06,150
which is how do we feel
about different

497
00:24:06,150 --> 00:24:08,400
distribution of income.

498
00:24:08,400 --> 00:24:09,690
And the reason this
is an issue--

499
00:24:09,690 --> 00:24:10,030
Yeah?

500
00:24:10,030 --> 00:24:11,280
AUDIENCE: [UNINTELLIGIBLE]

501
00:24:15,590 --> 00:24:17,050
PROFESSOR: That's a
great question.

502
00:24:17,050 --> 00:24:18,270
I don't know.

503
00:24:18,270 --> 00:24:19,730
I wouldn't ask you
to write it down.

504
00:24:19,730 --> 00:24:21,930
I mean, it's a much more
complicated-- there's not a

505
00:24:21,930 --> 00:24:23,550
simple mathematical function
with your Nozick

506
00:24:23,550 --> 00:24:24,850
or commodity egal.

507
00:24:24,850 --> 00:24:26,300
Commodity egalitarianism,
I guess the

508
00:24:26,300 --> 00:24:28,330
function would be that--

509
00:24:28,330 --> 00:24:31,300
the social welfare function
would be that there's some

510
00:24:31,300 --> 00:24:33,520
minimum u bar.

511
00:24:33,520 --> 00:24:36,960
Below u bar, we have an infinite
weight on you in our

512
00:24:36,960 --> 00:24:37,900
social welfare function.

513
00:24:37,900 --> 00:24:40,810
But once you get to u bar, we
only care about you the same

514
00:24:40,810 --> 00:24:42,060
as everybody else.

515
00:24:42,060 --> 00:24:43,810
That would be sort of commodity
egalitarianism.

516
00:24:43,810 --> 00:24:47,550
Nozick, I don't know how you
write it down mathematically.

517
00:24:47,550 --> 00:24:48,860
Why does this matter?

518
00:24:48,860 --> 00:24:51,670
This matters because it turns
out in society we do have a

519
00:24:51,670 --> 00:24:53,870
very unequal distribution
of resources.

520
00:24:53,870 --> 00:24:58,320
So to see that, let's go to the
next page, figure 23-2,

521
00:24:58,320 --> 00:25:01,860
this is from my textbook
that I use for my

522
00:25:01,860 --> 00:25:04,420
course on public policy.

523
00:25:04,420 --> 00:25:06,920
And this shows the income
received by

524
00:25:06,920 --> 00:25:08,010
quintile in the US.

525
00:25:08,010 --> 00:25:09,650
Let me explain what
this table means.

526
00:25:09,650 --> 00:25:13,910
Each row is a fifth
of the population.

527
00:25:13,910 --> 00:25:16,900
If we had an equal distribution
of income, then

528
00:25:16,900 --> 00:25:19,530
each of these numbers
would be 20%.

529
00:25:19,530 --> 00:25:22,100
Each fifth of the population
would have a

530
00:25:22,100 --> 00:25:24,310
fifth of the income.

531
00:25:24,310 --> 00:25:25,630
But that's not true.

532
00:25:25,630 --> 00:25:30,850
So if you look at 1967, the
poorest 20% of people only

533
00:25:30,850 --> 00:25:33,440
earned 4% of the income.

534
00:25:33,440 --> 00:25:36,710
And the richest 20% of people
earned 44% of the income, or

535
00:25:36,710 --> 00:25:41,200
11 times as much, 43.8, about
11 times as much.

536
00:25:41,200 --> 00:25:44,780
That's an unequal distribution
of income.

537
00:25:44,780 --> 00:25:46,600
What's interesting is
you can see over

538
00:25:46,600 --> 00:25:47,530
time how it's changed.

539
00:25:47,530 --> 00:25:50,090
From 1967 to 1980, things
actually got

540
00:25:50,090 --> 00:25:51,740
a little more equal.

541
00:25:51,740 --> 00:25:54,380
The poorest share actually grew,
and the richest share

542
00:25:54,380 --> 00:25:56,290
fell a bit.

543
00:25:56,290 --> 00:25:59,120
Then this since 1980, things
have gotten a lot more

544
00:25:59,120 --> 00:26:04,180
unequal, to the point where in
2007, the poorest 20% of

545
00:26:04,180 --> 00:26:06,470
people only controlled 3.4%
of our resources.

546
00:26:06,470 --> 00:26:11,050
That is, 1/33 of all resources
in society go to the bottom

547
00:26:11,050 --> 00:26:14,810
20% of the income distribution,
whereas 1/2 go

548
00:26:14,810 --> 00:26:16,450
to the top 20% of the
income distribution.

549
00:26:16,450 --> 00:26:20,420
So the richest 20% of people
get half the income that's

550
00:26:20,420 --> 00:26:21,110
earned in the US.

551
00:26:21,110 --> 00:26:24,660
The poorest 20% of people get
1/33 of the income that's

552
00:26:24,660 --> 00:26:26,010
earned in the US.

553
00:26:26,010 --> 00:26:28,350
So this is why this stuff
matters, because there is

554
00:26:28,350 --> 00:26:30,840
inequality in practice.

555
00:26:30,840 --> 00:26:33,050
In fact, it matters a lot in the
US in particular because

556
00:26:33,050 --> 00:26:34,970
the next table shows how
we do relative to

557
00:26:34,970 --> 00:26:36,900
the rest of the world.

558
00:26:36,900 --> 00:26:39,280
So this compares us to what's
called the OECD countries.

559
00:26:39,280 --> 00:26:41,540
OECD is the organization for
economically something

560
00:26:41,540 --> 00:26:43,820
something, it's a
French acronym.

561
00:26:43,820 --> 00:26:48,510
And basically, it's sort of
close to developed countries.

562
00:26:48,510 --> 00:26:50,400
It's a first and second
world if you will.

563
00:26:53,040 --> 00:26:54,710
It's different years because
the data is collected at

564
00:26:54,710 --> 00:26:55,420
different frequencies.

565
00:26:55,420 --> 00:26:58,010
But the rough facts are
constant over time.

566
00:26:58,010 --> 00:27:01,510
This shows the share of income
in each of the quintiles

567
00:27:01,510 --> 00:27:04,200
across all this list
of countries.

568
00:27:04,200 --> 00:27:07,450
And the bottom next to last
row shows the unweighted

569
00:27:07,450 --> 00:27:11,740
average across these developed
countries compared to the US.

570
00:27:11,740 --> 00:27:15,240
What you see is we are the
second most unequal nation on

571
00:27:15,240 --> 00:27:18,030
this list other than Mexico.

572
00:27:18,030 --> 00:27:20,500
We are more unequal than any
other nation on this list. On

573
00:27:20,500 --> 00:27:25,380
average, the poorest 20% of
society has more than twice as

574
00:27:25,380 --> 00:27:29,740
much in other countries, whereas
the richest 20% has

575
00:27:29,740 --> 00:27:33,130
about 20% less, 40% of the
resources instead of 50% of

576
00:27:33,130 --> 00:27:35,580
the resources.

577
00:27:35,580 --> 00:27:38,100
And so what you see is we
have a much more unequal

578
00:27:38,100 --> 00:27:42,650
distribution of income than any
other major economy except

579
00:27:42,650 --> 00:27:43,900
for Mexico.

580
00:27:45,840 --> 00:27:50,230
So basically, this says we have
an incredibly unequal

581
00:27:50,230 --> 00:27:54,320
distribution in American society
by some absolute

582
00:27:54,320 --> 00:27:56,930
standard, just if you look at
the numbers, it seems unequal,

583
00:27:56,930 --> 00:27:59,820
and relative to the
rest of the world.

584
00:27:59,820 --> 00:28:02,890
No country in history has ever
had all these numbers be 20%.

585
00:28:02,890 --> 00:28:04,780
Societies are always
unequal, the

586
00:28:04,780 --> 00:28:06,200
question is just how unequal.

587
00:28:06,200 --> 00:28:08,310
And compared to the rest of the
world at least, we seem

588
00:28:08,310 --> 00:28:11,000
pretty unequal.

589
00:28:11,000 --> 00:28:14,740
So that says that under a
utilitarian function, or

590
00:28:14,740 --> 00:28:17,730
especially in a Rawlsian
function, we should worry that

591
00:28:17,730 --> 00:28:21,750
we're not maximizing
social welfare.

592
00:28:21,750 --> 00:28:25,340
Under a utilitarian social
welfare function, at least,

593
00:28:25,340 --> 00:28:27,200
it's going to be hard to think
whether it maximizes social

594
00:28:27,200 --> 00:28:28,470
welfare when there's
such an unequal

595
00:28:28,470 --> 00:28:31,140
distribution of resources.

596
00:28:31,140 --> 00:28:33,030
Certainly in a Rawlsian
function we're not.

597
00:28:33,030 --> 00:28:35,510
Nozick, who the hell knows.

598
00:28:35,510 --> 00:28:38,780
That depends on where we start
with opportunities, how much

599
00:28:38,780 --> 00:28:41,620
is due to choice and how
much is due to luck.

600
00:28:41,620 --> 00:28:44,570
But of course, this doesn't
speak at all to the last view,

601
00:28:44,570 --> 00:28:47,910
which is a commodity
egalitarianism view, which

602
00:28:47,910 --> 00:28:49,840
says none of this matters.

603
00:28:49,840 --> 00:28:51,860
All that matters is the
measure of absolute

604
00:28:51,860 --> 00:28:53,270
deprivation.

605
00:28:53,270 --> 00:28:55,950
And we have such a measure in
the US, it's called the

606
00:28:55,950 --> 00:28:57,200
poverty line.

607
00:29:02,000 --> 00:29:04,510
The poverty line is a concept
that was actually thought up

608
00:29:04,510 --> 00:29:08,040
by the mid-level government
bureaucrat in the 1960s.

609
00:29:08,040 --> 00:29:11,500
What this woman, Molly
Orshansky, did is she said

610
00:29:11,500 --> 00:29:14,890
look, the typical family in the
US spends about a third of

611
00:29:14,890 --> 00:29:17,470
their income on food.

612
00:29:17,470 --> 00:29:20,230
Let's figure out what a
minimally nutritionally

613
00:29:20,230 --> 00:29:21,600
adequate diet costs.

614
00:29:21,600 --> 00:29:24,010
That is, how much do you have
to spend on food to have at

615
00:29:24,010 --> 00:29:26,250
least a minimally
adequate diet.

616
00:29:26,250 --> 00:29:30,320
No Starbucks, no Outback
Steakhouse, just enough to

617
00:29:30,320 --> 00:29:33,490
actually have a nutritionally
adequate diet, and then let's

618
00:29:33,490 --> 00:29:35,790
multiply it by three, and we'll
call that what a family

619
00:29:35,790 --> 00:29:37,650
needs to live in the US.

620
00:29:37,650 --> 00:29:40,550
Food's typically a third of
consumption, let's figure out

621
00:29:40,550 --> 00:29:42,850
a minimally adequate diet,
multiply it by three, call it

622
00:29:42,850 --> 00:29:45,280
the poverty line.

623
00:29:45,280 --> 00:29:48,300
And basically, we've taken
this exercise and

624
00:29:48,300 --> 00:29:49,210
used it ever since.

625
00:29:49,210 --> 00:29:53,150
We simply update it by inflation
to see how the

626
00:29:53,150 --> 00:29:56,450
poverty line changes
over time.

627
00:29:56,450 --> 00:29:58,600
So basically, we've just taken
this concept and updated it by

628
00:29:58,600 --> 00:30:00,390
inflation over time.

629
00:30:00,390 --> 00:30:04,190
The result of what you get
is in figure 23-4.

630
00:30:04,190 --> 00:30:07,740
This shows the US poverty
line in 2006.

631
00:30:07,740 --> 00:30:10,840
Actually, this is mislabeled,
it's 2009 actually.

632
00:30:10,840 --> 00:30:13,980
It's a typo in my book.

633
00:30:13,980 --> 00:30:18,610
What this says is that for a
family of one, the minimum

634
00:30:18,610 --> 00:30:24,240
standard that you need
to live is $10,830.

635
00:30:24,240 --> 00:30:26,450
And it goes up with family size
because more people need

636
00:30:26,450 --> 00:30:28,090
more to eat.

637
00:30:28,090 --> 00:30:32,960
Now, if you are from Mississippi
or maybe North

638
00:30:32,960 --> 00:30:36,280
Dakota, this might look like
kind of a reasonably large

639
00:30:36,280 --> 00:30:37,270
number to you.

640
00:30:37,270 --> 00:30:39,725
If you're from Boston, this
looks like insane, like how

641
00:30:39,725 --> 00:30:43,335
could someone live on $10,830 a
year in Boston, New York, or

642
00:30:43,335 --> 00:30:45,960
DC, or San Francisco,
or any major city?

643
00:30:45,960 --> 00:30:47,740
And the answer is they can't.

644
00:30:47,740 --> 00:30:51,060
The answer is that the poverty
line has not been kept up

645
00:30:51,060 --> 00:30:54,120
appropriately, because we've
simply taken this number and

646
00:30:54,120 --> 00:30:56,820
inflated it by inflation over
time, but it hasn't accounted

647
00:30:56,820 --> 00:30:58,110
for the fact that people's
consumption

648
00:30:58,110 --> 00:30:59,290
bundles have changed.

649
00:30:59,290 --> 00:31:02,190
In particular, food is no longer
a third of the typical

650
00:31:02,190 --> 00:31:05,230
person's consumption bundle,
it's now 1/6.

651
00:31:05,230 --> 00:31:07,170
So in fact, the poverty line
should be much, much higher

652
00:31:07,170 --> 00:31:10,020
than it is, but it hasn't been
updated over time in a

653
00:31:10,020 --> 00:31:11,340
meaningful way.

654
00:31:11,340 --> 00:31:14,620
So you can think this is a very
low minimum standard for

655
00:31:14,620 --> 00:31:16,790
what people need to live, and
certainly below what anyone

656
00:31:16,790 --> 00:31:18,590
needs to live in any major
metropolitan area.

657
00:31:21,870 --> 00:31:24,220
That said, we can look at what's
happened to the share

658
00:31:24,220 --> 00:31:26,870
of population in poverty, and
that's in figure 23-5.

659
00:31:26,870 --> 00:31:28,890
So here's the sort of commodity
egalitarianism thing

660
00:31:28,890 --> 00:31:32,410
to look at, which is what's
happened to poverty.

661
00:31:32,410 --> 00:31:34,880
And what you see is there was
a massive decline in poverty

662
00:31:34,880 --> 00:31:38,130
for everyone in the 1960s
to early '70s.

663
00:31:38,130 --> 00:31:41,560
We call this the War On Poverty
under the Kennedy and

664
00:31:41,560 --> 00:31:46,100
Johnson administrations, a huge
reduction in poverty.

665
00:31:46,100 --> 00:31:50,780
What you see is that since about
1970, that reduction has

666
00:31:50,780 --> 00:31:53,270
continued for one group, for
the elderly, their poverty

667
00:31:53,270 --> 00:31:54,430
rates continued to fall.

668
00:31:54,430 --> 00:31:58,960
So in 1959, 35% of elderly
people lived in poverty.

669
00:31:58,960 --> 00:32:01,480
Then through massive expanses of
a program we'll learn about

670
00:32:01,480 --> 00:32:05,640
in a couple of lectures called
Social Security, that fell, so

671
00:32:05,640 --> 00:32:08,480
that today, only about
10% of elderly

672
00:32:08,480 --> 00:32:10,020
people live in poverty.

673
00:32:10,020 --> 00:32:12,010
On the other hand, for everyone
else, and especially

674
00:32:12,010 --> 00:32:15,000
for kids, it's basically
flattened out or increased

675
00:32:15,000 --> 00:32:17,280
since 1970.

676
00:32:17,280 --> 00:32:20,305
So basically, overall poverty
rate hit about 13% in 1970,

677
00:32:20,305 --> 00:32:22,210
and it's been pretty
flat since.

678
00:32:22,210 --> 00:32:25,440
For kids, it fell to about
15%, and it's been

679
00:32:25,440 --> 00:32:26,220
bounced up and down.

680
00:32:26,220 --> 00:32:28,770
It's now back up to about 20%.

681
00:32:28,770 --> 00:32:32,850
So from a commodity
egalitarianism view, even if

682
00:32:32,850 --> 00:32:36,060
we take the poverty line as a
minimum acceptable level,

683
00:32:36,060 --> 00:32:38,740
there are currently about 40
million families in the US

684
00:32:38,740 --> 00:32:40,420
living below poverty.

685
00:32:40,420 --> 00:32:43,710
40 million people, I'm sorry,
living below the poverty line.

686
00:32:43,710 --> 00:32:47,460
So even at this minimum
standard, we've got a failure

687
00:32:47,460 --> 00:32:48,620
of redistribution.

688
00:32:48,620 --> 00:32:51,460
We're not maximizing social
welfare given that we've got

689
00:32:51,460 --> 00:32:55,410
40 million people living
below this standard.

690
00:32:55,410 --> 00:32:59,640
So based on whatever standard
you want to use, except the

691
00:32:59,640 --> 00:33:03,070
hard to evaluate Nozickian
standard, we're clearly not

692
00:33:03,070 --> 00:33:04,810
maximizing social welfare
at the current

693
00:33:04,810 --> 00:33:06,240
distribution of income.

694
00:33:06,240 --> 00:33:08,170
We've got an incredibly unequal
distribution of

695
00:33:08,170 --> 00:33:10,460
income, and even if all we
care about is making sure

696
00:33:10,460 --> 00:33:12,990
people have a decent standard of
living, we're also failing

697
00:33:12,990 --> 00:33:15,680
on that standard for upwards
of 40 million people.

698
00:33:15,680 --> 00:33:17,940
For a more realistic poverty
line, it might be more like 60

699
00:33:17,940 --> 00:33:20,410
or 70 million people.

700
00:33:20,410 --> 00:33:25,010
So clearly there is an argument
here for some income

701
00:33:25,010 --> 00:33:28,310
redistribution, whatever your
social welfare function is.

702
00:33:28,310 --> 00:33:29,190
So we can write down--

703
00:33:29,190 --> 00:33:30,640
I'm not going to ask you to
write down mathematically

704
00:33:30,640 --> 00:33:32,750
social welfare function and
maximize it and solve for

705
00:33:32,750 --> 00:33:35,340
outcomes and stuff like that,
but what I want to leave you

706
00:33:35,340 --> 00:33:37,890
with here is simply
two points.

707
00:33:37,890 --> 00:33:42,560
First of all, there are lots of
reasons why we might want

708
00:33:42,560 --> 00:33:45,720
income redistribution in society
under any of these

709
00:33:45,720 --> 00:33:47,360
alternatives.

710
00:33:47,360 --> 00:33:48,655
That's point one.

711
00:33:48,655 --> 00:33:51,870
Point 2 is our current
distribution of resources

712
00:33:51,870 --> 00:33:55,740
seems sufficiently unequal that
we likely do want some

713
00:33:55,740 --> 00:33:59,110
redistribution of resources
in society.

714
00:33:59,110 --> 00:34:02,030
For most social welfare
functions you can write down,

715
00:34:02,030 --> 00:34:05,010
given the facts I've shown
you, we would not be

716
00:34:05,010 --> 00:34:08,139
maximizing social welfare at
the current distribution of

717
00:34:08,139 --> 00:34:09,389
income in society.

718
00:34:11,370 --> 00:34:13,150
I'm trying to couch this in
a way that doesn't make me

719
00:34:13,150 --> 00:34:14,760
particularly lefty or righty.

720
00:34:14,760 --> 00:34:17,520
I'm just trying to couch this
in a way which just draws on

721
00:34:17,520 --> 00:34:20,300
the uncomfortable fact that we
have to talk about this.

722
00:34:20,300 --> 00:34:22,480
We have to talk about income
redistribution.

723
00:34:22,480 --> 00:34:24,929
The only way to do so is
to have some framework.

724
00:34:24,929 --> 00:34:27,510
I've given you four frameworks
and said under all of them, or

725
00:34:27,510 --> 00:34:30,980
at least most of them, we would
be uncomfortable with

726
00:34:30,980 --> 00:34:34,159
the existing distribution
of income in society.

727
00:34:34,159 --> 00:34:35,409
Questions about that?

728
00:34:37,989 --> 00:34:40,040
So what do we do about that?

729
00:34:40,040 --> 00:34:44,820
What do we do about the fact
that we have unequal

730
00:34:44,820 --> 00:34:46,150
distribution of resources
in society?

731
00:34:46,150 --> 00:34:47,989
Well, we redistribute.

732
00:34:47,989 --> 00:34:50,610
We say look, let's get out
Okun's bucket and start

733
00:34:50,610 --> 00:34:52,710
carrying some money around.

734
00:34:52,710 --> 00:34:54,630
We don't like the fact that
there's people living in

735
00:34:54,630 --> 00:34:55,500
poverty who can't eat.

736
00:34:55,500 --> 00:34:56,989
We don't like the fact that
the poor have so much less

737
00:34:56,989 --> 00:34:58,090
than the rich.

738
00:34:58,090 --> 00:35:00,200
Let's start redistributing.

739
00:35:00,200 --> 00:35:04,550
Once you do that is you
have to recognize

740
00:35:04,550 --> 00:35:06,720
the leak in the bucket.

741
00:35:06,720 --> 00:35:09,320
And recognize there's a trade
off, that it's not so simple

742
00:35:09,320 --> 00:35:11,000
as saying, fine, let's take from
the rich and give to the

743
00:35:11,000 --> 00:35:15,130
poor, because there's a
leak in the bucket.

744
00:35:15,130 --> 00:35:18,040
And so that as we take from the
rich and give to the poor,

745
00:35:18,040 --> 00:35:21,560
we shrink the total size
of social welfare.

746
00:35:21,560 --> 00:35:24,760
And that gives a trade off.

747
00:35:24,760 --> 00:35:30,270
And in particular, there's two
sources of leakage that we

748
00:35:30,270 --> 00:35:32,560
have to think about.

749
00:35:32,560 --> 00:35:35,870
The first source of leakage is
that when you tax a rich

750
00:35:35,870 --> 00:35:41,260
person to take their money away,
they work less hard.

751
00:35:41,260 --> 00:35:43,690
And when they work less hard,
there's less goods produced in

752
00:35:43,690 --> 00:35:47,360
society, and that
shrinks the pie.

753
00:35:47,360 --> 00:35:49,140
The other source of leakage
you have to think about is

754
00:35:49,140 --> 00:35:52,790
that when you give money to
poor people, they may

755
00:35:52,790 --> 00:35:57,320
themselves quit their jobs to
qualify as poor so they can

756
00:35:57,320 --> 00:36:01,170
get the money, and that further
shrinks the pie.

757
00:36:01,170 --> 00:36:05,040
So both the taxing the rich and
the giving to the poor are

758
00:36:05,040 --> 00:36:07,670
sources of leakage
in the bucket.

759
00:36:07,670 --> 00:36:09,860
And that's a trade off with
dealing with these enormous

760
00:36:09,860 --> 00:36:13,850
distributional problems
that I laid out.

761
00:36:13,850 --> 00:36:16,580
So let's just go through one
example, general example to

762
00:36:16,580 --> 00:36:18,650
make this point.

763
00:36:18,650 --> 00:36:23,820
Let's say we have a society
where everyone is equal, and

764
00:36:23,820 --> 00:36:28,000
everyone earns a wage
of $20 an hour.

765
00:36:28,000 --> 00:36:30,160
Equal society, everyone's
equal in terms of their

766
00:36:30,160 --> 00:36:35,470
underlying preferences, and
everyone earns $20 an hour.

767
00:36:35,470 --> 00:36:39,750
But people work different
amounts of hours.

768
00:36:39,750 --> 00:36:41,620
Now they could work these
different amount of hours

769
00:36:41,620 --> 00:36:44,860
because they're lazy,
Nozick might say.

770
00:36:44,860 --> 00:36:46,200
They could work these different
amount of hours

771
00:36:46,200 --> 00:36:50,450
because they just can't find
a job, or because they're

772
00:36:50,450 --> 00:36:53,610
disabled, or because they're not
skilled enough to work a

773
00:36:53,610 --> 00:36:54,520
full-time job.

774
00:36:54,520 --> 00:36:56,740
Whatever the reason is, we're
going to stay away from what

775
00:36:56,740 --> 00:36:57,410
that reason is.

776
00:36:57,410 --> 00:36:59,200
Although obviously, what the
reason is matters for these

777
00:36:59,200 --> 00:37:00,580
social welfare functions.

778
00:37:00,580 --> 00:37:02,740
Let's just say we have a society
where everyone earns

779
00:37:02,740 --> 00:37:05,800
$20 an hour, but we have a
distribution on how many hours

780
00:37:05,800 --> 00:37:08,460
people work.

781
00:37:08,460 --> 00:37:13,560
And let's say that we want
to start with a commodity

782
00:37:13,560 --> 00:37:16,320
egalitarianism view, and say
we want to make sure that

783
00:37:16,320 --> 00:37:17,410
everyone has enough to eat.

784
00:37:17,410 --> 00:37:22,790
We want to make sure that
everyone has at least $10,000.

785
00:37:22,790 --> 00:37:25,440
So we come in and we say look,
we're worried about this

786
00:37:25,440 --> 00:37:27,420
distribution of resources
in society.

787
00:37:27,420 --> 00:37:32,430
We want to make sure everyone
has at least $10,000.

788
00:37:32,430 --> 00:37:33,696
So what we're going to
do is we're going to

789
00:37:33,696 --> 00:37:36,850
give everyone a transfer.

790
00:37:36,850 --> 00:37:43,240
And that transfer function is
going to be equal to the max

791
00:37:43,240 --> 00:37:51,600
of 0 and $10,000 minus
your income.

792
00:37:51,600 --> 00:37:53,430
In other words, if your
income is 0, we're

793
00:37:53,430 --> 00:37:55,860
going to give you $10,000.

794
00:37:55,860 --> 00:37:59,960
If your income's $5,000, we're
going to give you $5,000.

795
00:37:59,960 --> 00:38:03,140
If you income is $9,999, we're
going to give you $1.

796
00:38:03,140 --> 00:38:06,450
Once your income is $10,000,
we give you nothing.

797
00:38:06,450 --> 00:38:09,090
So this is very much like a
commodity egalitarianism view.

798
00:38:09,090 --> 00:38:11,203
We're going to bring
you up to 10,000.

799
00:38:11,203 --> 00:38:13,550
We're going to bring you up to
10,000, but once you're above

800
00:38:13,550 --> 00:38:16,410
10,000, we don't
care about you.

801
00:38:16,410 --> 00:38:19,780
So this is much less radical
than the utilitarian solution.

802
00:38:19,780 --> 00:38:20,900
This is just saying,
we're just going to

803
00:38:20,900 --> 00:38:22,390
bring the poor up.

804
00:38:22,390 --> 00:38:23,640
We're just going to
deal with that.

805
00:38:26,920 --> 00:38:28,980
However, if we're going to have
this program, we've got

806
00:38:28,980 --> 00:38:29,560
to pay for it.

807
00:38:29,560 --> 00:38:30,850
We've got to give money
to poor people,

808
00:38:30,850 --> 00:38:32,010
where's that come from?

809
00:38:32,010 --> 00:38:34,650
Let's say we finance the program
by taxing higher

810
00:38:34,650 --> 00:38:36,080
income people.

811
00:38:36,080 --> 00:38:37,590
What we're going to say in
particular is we're going to

812
00:38:37,590 --> 00:38:39,840
have a tax rate tau--

813
00:38:39,840 --> 00:38:42,910
so regular t is transfers,
tau is taxes--

814
00:38:42,910 --> 00:38:52,290
which is 0 if your income is
less than $20,000, and 20% if

815
00:38:52,290 --> 00:38:55,640
your income is greater than
$20,000, greater than or equal

816
00:38:55,640 --> 00:38:56,970
to $20,000.

817
00:38:56,970 --> 00:38:58,380
So we're going to have
a tax schedule.

818
00:38:58,380 --> 00:39:00,840
We're not going to tax you if
you're less than $20,000.

819
00:39:00,840 --> 00:39:02,820
But once you earn more than
$20,000, we're going to tax

820
00:39:02,820 --> 00:39:06,220
you at 20%.

821
00:39:06,220 --> 00:39:08,880
So this is our redistributive
scheme in society.

822
00:39:08,880 --> 00:39:10,760
This is our Okun's bucket.

823
00:39:10,760 --> 00:39:12,420
The rich put the money in.

824
00:39:12,420 --> 00:39:14,530
We tax them to get the money
to put in the bucket.

825
00:39:14,530 --> 00:39:16,470
We bring it over and give it
to the poor in the form of

826
00:39:16,470 --> 00:39:18,490
bringing everybody
up to $10,000.

827
00:39:18,490 --> 00:39:20,550
That's our redistributive
scheme, the simplest possible

828
00:39:20,550 --> 00:39:23,750
redistributive scheme.

829
00:39:23,750 --> 00:39:25,590
What does this do?

830
00:39:25,590 --> 00:39:29,990
Let's now go to figure 23-6.

831
00:39:29,990 --> 00:39:31,380
Pretty complicated figure,
so let's walk

832
00:39:31,380 --> 00:39:33,840
through this slowly.

833
00:39:33,840 --> 00:39:39,630
We initially have a budget
constraint that runs from the

834
00:39:39,630 --> 00:39:44,470
intercept at 40,000 on the
y-axis to the intercept of

835
00:39:44,470 --> 00:39:45,590
2,000 on the x-axis.

836
00:39:45,590 --> 00:39:48,400
So the initial budget constraint
is the outer budget

837
00:39:48,400 --> 00:39:52,270
constraint here running from
40,000 on the y-axis all the

838
00:39:52,270 --> 00:39:54,170
way down to 2,000
on the x-axis.

839
00:39:54,170 --> 00:39:57,250
That's because we have a
$20 hour an hour wage--

840
00:39:57,250 --> 00:39:58,080
I'm sorry, let me go back.

841
00:39:58,080 --> 00:40:01,000
This is a consumption leisure
trade off, like we do whatever

842
00:40:01,000 --> 00:40:02,460
we do in labor supply.

843
00:40:02,460 --> 00:40:05,420
When we do labor supply, we do
consumption leisure trade off,

844
00:40:05,420 --> 00:40:07,530
so you're deciding
how hard to work.

845
00:40:07,530 --> 00:40:09,340
How do you decide how
hard to work?

846
00:40:09,340 --> 00:40:13,340
You can do it by trading off
consumption and leisure.

847
00:40:13,340 --> 00:40:14,790
You can either take
2,000 hours of

848
00:40:14,790 --> 00:40:17,750
leisure and consume nothing.

849
00:40:17,750 --> 00:40:21,720
Or you could take zero leisure,
work 2,000 hours to

850
00:40:21,720 --> 00:40:23,600
consume $40,000.

851
00:40:23,600 --> 00:40:26,110
That's your trade-off, or
combinations in between.

852
00:40:26,110 --> 00:40:29,660
And we have three individuals,
A, B, and C. A is someone who

853
00:40:29,660 --> 00:40:30,870
works very little.

854
00:40:30,870 --> 00:40:35,170
They take a lot of leisure and
have very low consumption.

855
00:40:35,170 --> 00:40:37,340
C is someone who works a lot.

856
00:40:37,340 --> 00:40:39,570
They have low leisure and
high consumption.

857
00:40:39,570 --> 00:40:42,230
And B's in the middle.

858
00:40:42,230 --> 00:40:47,040
Basic labor leisure trade off,
questions about that?

859
00:40:47,040 --> 00:40:51,650
Now, how does this new
government policy affect the

860
00:40:51,650 --> 00:40:52,850
budget constraint?

861
00:40:52,850 --> 00:40:54,410
Well, it affects
it in two ways.

862
00:40:54,410 --> 00:40:58,640
First of all, for those
above $20,000, we

863
00:40:58,640 --> 00:41:01,140
now lower their wage.

864
00:41:01,140 --> 00:41:04,560
Instead of taking home $20 hours
an hour, they're only

865
00:41:04,560 --> 00:41:07,960
going to take home $16 an hour,
because we're taxing

866
00:41:07,960 --> 00:41:09,950
them at 20%.

867
00:41:09,950 --> 00:41:12,710
So for everyone above $20,000,
we shifted the budget

868
00:41:12,710 --> 00:41:13,590
constraint in.

869
00:41:13,590 --> 00:41:17,240
We have lowered the
price of leisure.

870
00:41:17,240 --> 00:41:21,630
We've lowered the price of
leisure by taxing them.

871
00:41:21,630 --> 00:41:25,090
So we've shifted that budget
constraint in.

872
00:41:25,090 --> 00:41:26,940
It's the same budget constraint
up to 20,000, but

873
00:41:26,940 --> 00:41:29,180
now it pivots, and at
20,000, you're now

874
00:41:29,180 --> 00:41:30,570
on the lower segment.

875
00:41:30,570 --> 00:41:32,860
With a slope, you can write
there, the slope is $20 at the

876
00:41:32,860 --> 00:41:37,000
higher segment, it's only $16
for that lower segment.

877
00:41:37,000 --> 00:41:40,610
That's the first change you've
made to the budget constraint.

878
00:41:40,610 --> 00:41:43,410
That's the tax part.

879
00:41:43,410 --> 00:41:47,450
The second change is that for
anyone with income below

880
00:41:47,450 --> 00:41:52,400
$10,000, we've now said no
matter how hard you work,

881
00:41:52,400 --> 00:41:54,980
we're going to give
you $10,000.

882
00:41:54,980 --> 00:42:00,210
So now we've said, once your
leisure is 1,500 hours, or

883
00:42:00,210 --> 00:42:04,450
your work is 500 hours, anywhere
to the right of that,

884
00:42:04,450 --> 00:42:09,060
your income is always $10,000
no matter how hard you work.

885
00:42:09,060 --> 00:42:11,890
If you earn $5,000, your
income is $10,000.

886
00:42:11,890 --> 00:42:14,030
If you earn $1, your
income is $10,000.

887
00:42:14,030 --> 00:42:18,870
If you earn $9,999, your
income is $10,000.

888
00:42:18,870 --> 00:42:23,610
So the new budget constraint is
a flat segment starting at

889
00:42:23,610 --> 00:42:27,740
the intersection of 10,000 and
1,500 and moving all the way

890
00:42:27,740 --> 00:42:29,690
to the right.

891
00:42:29,690 --> 00:42:34,290
That is, no matter how much
leisure you take, above 1,500

892
00:42:34,290 --> 00:42:37,190
hours of leisure, your
consumption is always $10,000,

893
00:42:37,190 --> 00:42:39,050
so it's flat.

894
00:42:39,050 --> 00:42:43,410
So the new budget constraint
is the inner segment above,

895
00:42:43,410 --> 00:42:45,840
then the old budget constraint
from $20,000 to $10,000, and

896
00:42:45,840 --> 00:42:50,570
then a flat from point B to
point D. I'm sorry, not from B

897
00:42:50,570 --> 00:42:53,950
to D. Kill that last comment.

898
00:42:53,950 --> 00:42:56,830
From the intersection of 10,000
and 1,500 all the way

899
00:42:56,830 --> 00:43:00,710
to the right, that's the
new budget constraint.

900
00:43:00,710 --> 00:43:01,910
Questions about that?

901
00:43:01,910 --> 00:43:03,220
This is very important.

902
00:43:03,220 --> 00:43:05,270
This is just an application to
understand how these things

903
00:43:05,270 --> 00:43:06,580
affect budget constraints.

904
00:43:06,580 --> 00:43:09,450
What does this do to
people's choices?

905
00:43:09,450 --> 00:43:13,470
Well, for a person like
C, they are now

906
00:43:13,470 --> 00:43:14,670
taxed on their labor.

907
00:43:14,670 --> 00:43:17,170
Person C doesn't care about the
welfare program, they're

908
00:43:17,170 --> 00:43:19,660
rich, they don't care about
this $10,000 thing.

909
00:43:19,660 --> 00:43:22,840
But they do care about the
fact they're now taxed.

910
00:43:22,840 --> 00:43:26,780
Assuming substitution effects
dominate, the price of leisure

911
00:43:26,780 --> 00:43:31,050
has fallen, so they choose more
leisure and less labor.

912
00:43:31,050 --> 00:43:36,850
So person C, who was at C moves
to point E, working less

913
00:43:36,850 --> 00:43:40,060
assuming substitution effects
dominate, more

914
00:43:40,060 --> 00:43:42,940
leisure, less labor.

915
00:43:42,940 --> 00:43:48,420
Person A, for them, this
is a no-brainer.

916
00:43:48,420 --> 00:43:51,880
They can have both more leisure
and more consumption

917
00:43:51,880 --> 00:43:55,370
by moving from point A to point
D. Under this new budget

918
00:43:55,370 --> 00:44:00,490
constraint, they used to have
about-- it's not labeled

919
00:44:00,490 --> 00:44:03,035
there-- but have about 1,700
hours of leisure and consume

920
00:44:03,035 --> 00:44:05,120
maybe $5,000.

921
00:44:05,120 --> 00:44:07,050
Now they can have more leisure,
they can have 2,000

922
00:44:07,050 --> 00:44:08,930
hours of leisure and
consume $10,000.

923
00:44:08,930 --> 00:44:14,920
So they move to point D. They
work less hard as well.

924
00:44:14,920 --> 00:44:19,030
Person B, the way I've drawn
this, this is critical.

925
00:44:19,030 --> 00:44:23,160
The way I've drawn this, their
indifference curve cuts

926
00:44:23,160 --> 00:44:26,580
through the horizontal segment
running from the old budget

927
00:44:26,580 --> 00:44:29,470
constraint to point D. Why
is that important?

928
00:44:29,470 --> 00:44:32,950
Because that means that
there'll be a higher

929
00:44:32,950 --> 00:44:36,180
difference curve out at point D,
because indifference curves

930
00:44:36,180 --> 00:44:37,400
can't cross.

931
00:44:37,400 --> 00:44:39,970
So if their old indifference
curve cuts through that

932
00:44:39,970 --> 00:44:43,210
horizontal segment, it must
mean that they would be

933
00:44:43,210 --> 00:44:47,020
happier out at point D.
They're giving up some

934
00:44:47,020 --> 00:44:48,170
consumption.

935
00:44:48,170 --> 00:44:51,680
This isn't a no-brainer like
for A, but they get so much

936
00:44:51,680 --> 00:44:54,340
leisure it's worth it.

937
00:44:54,340 --> 00:44:57,210
So you could have drawn B, if
I'd drawn B higher up, their

938
00:44:57,210 --> 00:44:58,910
indifference curve wouldn't have
crossed this horizontal

939
00:44:58,910 --> 00:45:00,720
segment, and then they wouldn't
have wanted to move

940
00:45:00,720 --> 00:45:02,370
out to point D.

941
00:45:02,370 --> 00:45:04,060
So for A, it's a no-brainer.

942
00:45:04,060 --> 00:45:05,890
they move to D. For B
it depends on their

943
00:45:05,890 --> 00:45:06,290
indifference curve.

944
00:45:06,290 --> 00:45:06,670
Yeah?

945
00:45:06,670 --> 00:45:09,611
AUDIENCE: For C, doesn't he
change to working harder and

946
00:45:09,611 --> 00:45:10,553
consuming less?

947
00:45:10,553 --> 00:45:13,850
Because he moves [INAUDIBLE],
so his leisure [INAUDIBLE].

948
00:45:16,680 --> 00:45:17,870
PROFESSOR: Yeah, you're right.

949
00:45:17,870 --> 00:45:18,650
This is drawn wrong.

950
00:45:18,650 --> 00:45:19,950
You're right, we draw it so
that income effect is

951
00:45:19,950 --> 00:45:20,440
dominating.

952
00:45:20,440 --> 00:45:21,180
Good point.

953
00:45:21,180 --> 00:45:23,090
I'm sorry, that should
have been--

954
00:45:23,090 --> 00:45:24,240
good catch.

955
00:45:24,240 --> 00:45:27,970
Point E should be to the
right of point C,

956
00:45:27,970 --> 00:45:29,280
that's a mistake here.

957
00:45:29,280 --> 00:45:30,750
We should have drawn it with
substitution effects

958
00:45:30,750 --> 00:45:31,560
dominating.

959
00:45:31,560 --> 00:45:33,660
That would have been more
leisure and less consumption.

960
00:45:33,660 --> 00:45:36,790
He get less consumption,
but he works more

961
00:45:36,790 --> 00:45:37,460
because of the tax.

962
00:45:37,460 --> 00:45:38,810
But you're right, that should
have been to the right of

963
00:45:38,810 --> 00:45:41,390
point C. Please correct that.

964
00:45:41,390 --> 00:45:43,350
They should have been to the
right of point C. Substitution

965
00:45:43,350 --> 00:45:46,950
effect should lead
him to earn less.

966
00:45:46,950 --> 00:45:50,330
So he earns less, although once
again, it's ambiguous.

967
00:45:50,330 --> 00:45:53,090
A clearly works less,
that's unambiguous.

968
00:45:53,090 --> 00:45:54,900
And B, it depends on their
difference curves.

969
00:45:54,900 --> 00:45:56,950
But the way we've drawn it here,
B works less-- in fact,

970
00:45:56,950 --> 00:45:58,860
B works a ton less.

971
00:45:58,860 --> 00:46:04,510
B goes from taking maybe 1,400
hours of leisure to taking

972
00:46:04,510 --> 00:46:06,040
2,000 hours of leisure.

973
00:46:06,040 --> 00:46:07,620
This is a huge change for B.

974
00:46:07,620 --> 00:46:11,170
What this means is this tax
and transfer system has

975
00:46:11,170 --> 00:46:16,310
massively reduced the amount of
labor supplied in society.

976
00:46:16,310 --> 00:46:17,115
Why do we care?

977
00:46:17,115 --> 00:46:22,160
Go to figure 23-7.

978
00:46:22,160 --> 00:46:24,680
We care because we
initially were in

979
00:46:24,680 --> 00:46:28,150
equilibrium at point e1.

980
00:46:28,150 --> 00:46:32,560
But when people work less hard,
that's a shift inward in

981
00:46:32,560 --> 00:46:37,410
the labor supply curve that
reduces the total amount of

982
00:46:37,410 --> 00:46:41,110
labor supplied and causes
a dead weight loss.

983
00:46:41,110 --> 00:46:44,910
We've distorted the economy by
people working less hard.

984
00:46:44,910 --> 00:46:46,950
Trades that would've
made both parties

985
00:46:46,950 --> 00:46:50,240
better off are not happening.

986
00:46:50,240 --> 00:46:51,810
Trades that would have made both
parties better off are

987
00:46:51,810 --> 00:46:53,520
not happening.

988
00:46:53,520 --> 00:46:56,430
We've distorted the economy and
caused a dead weight loss.

989
00:46:56,430 --> 00:46:58,580
And that is the equity
efficiency

990
00:46:58,580 --> 00:47:02,360
trade off in a nutshell.

991
00:47:02,360 --> 00:47:04,990
The equity efficiency trade off
is we knew we wanted this

992
00:47:04,990 --> 00:47:05,990
tax and transfer system.

993
00:47:05,990 --> 00:47:07,940
We were very upset about
the number of

994
00:47:07,940 --> 00:47:09,550
people living in poverty.

995
00:47:09,550 --> 00:47:13,120
But by putting this in, we've
caused a dead weight loss,

996
00:47:13,120 --> 00:47:15,750
because it caused the rich
people to work less hard and

997
00:47:15,750 --> 00:47:17,290
the poor people to
work less hard.

998
00:47:17,290 --> 00:47:20,680
There's leaks in the bucket
on both sides.

999
00:47:20,680 --> 00:47:23,310
This dead weight loss triangle
is the leak in the bucket, the

1000
00:47:23,310 --> 00:47:29,070
social waste that comes from
this redistribution system.

1001
00:47:29,070 --> 00:47:32,020
Questions about that?

1002
00:47:32,020 --> 00:47:37,170
So that leads us to mask,
was it worth it?

1003
00:47:37,170 --> 00:47:40,640
Is a transfer system
like this worth it?

1004
00:47:40,640 --> 00:47:42,510
Well, there's a simple answer.

1005
00:47:42,510 --> 00:47:44,330
You solve the social
welfare function.

1006
00:47:44,330 --> 00:47:45,490
Once again putting
Nozick aside, we

1007
00:47:45,490 --> 00:47:46,960
can't write that down.

1008
00:47:46,960 --> 00:47:51,643
But if you're utilitarian, you
just solve it, because, in

1009
00:47:51,643 --> 00:47:53,130
other words, what you say
is look, two things

1010
00:47:53,130 --> 00:47:54,770
are going on here.

1011
00:47:54,770 --> 00:47:57,410
Everyone's utility is
falling because

1012
00:47:57,410 --> 00:47:58,490
there's a social waste.

1013
00:47:58,490 --> 00:48:03,420
So each of those u's goes down,
but the rich u's go down

1014
00:48:03,420 --> 00:48:06,470
a little bit, and the poor
u's go up a lot.

1015
00:48:06,470 --> 00:48:07,700
And you have to basically
ask which of

1016
00:48:07,700 --> 00:48:09,500
those effects are bigger.

1017
00:48:09,500 --> 00:48:11,220
In other words, if the dead
weight loss triangle is very

1018
00:48:11,220 --> 00:48:13,290
small, then clearly--

1019
00:48:13,290 --> 00:48:15,170
or if there's a small
leak in the bucket--

1020
00:48:15,170 --> 00:48:17,350
then clearly, total utilitarian
social welfare

1021
00:48:17,350 --> 00:48:19,780
will go up with a scheme like
this, because the poor people

1022
00:48:19,780 --> 00:48:21,600
will be made so much happier,
and the other

1023
00:48:21,600 --> 00:48:23,350
people won't care much.

1024
00:48:23,350 --> 00:48:26,010
But if that dead weight loss
triangle is huge, let's

1025
00:48:26,010 --> 00:48:29,050
imagine half of society's output
disappears for this

1026
00:48:29,050 --> 00:48:31,370
transfer system, well, then
probably social welfare will

1027
00:48:31,370 --> 00:48:34,500
fall, because so many people are
made sadder that the fact

1028
00:48:34,500 --> 00:48:36,900
a few poor people are made
happier isn't worth it.

1029
00:48:36,900 --> 00:48:38,950
Remember, 40 million people are
in poverty, that means 300

1030
00:48:38,950 --> 00:48:41,070
million people aren't.

1031
00:48:41,070 --> 00:48:43,430
So if 300 million people, if
they see their incomes cut in

1032
00:48:43,430 --> 00:48:46,020
half so that 40 million people
can see their incomes raised

1033
00:48:46,020 --> 00:48:49,970
some, for most utilitarian
social welfare functions, that

1034
00:48:49,970 --> 00:48:51,470
won't be a good deal.

1035
00:48:51,470 --> 00:48:54,110
So basically, we can literally
mathematically represent

1036
00:48:54,110 --> 00:48:59,130
whether this transfer was a
good thing or not by using

1037
00:48:59,130 --> 00:49:01,230
this social welfare function
to evaluate it.

1038
00:49:01,230 --> 00:49:04,020
Under a Rawlsian function,
we know the answer.

1039
00:49:04,020 --> 00:49:06,096
We know this is a good thing,
because Rawls doesn't care if

1040
00:49:06,096 --> 00:49:08,210
we destroy all of society,
as long as that bottom

1041
00:49:08,210 --> 00:49:10,240
guy gets pulled up.

1042
00:49:10,240 --> 00:49:14,110
So basically, Rawls we know that
this is a good answer.

1043
00:49:14,110 --> 00:49:17,520
Utilitarian, we don't know.

1044
00:49:17,520 --> 00:49:20,840
That's going to depend on the
loss to the rest of us from a

1045
00:49:20,840 --> 00:49:23,120
smaller pie versus the
gain to the poor from

1046
00:49:23,120 --> 00:49:25,370
getting more resources.

1047
00:49:25,370 --> 00:49:27,240
Questions about that?

1048
00:49:27,240 --> 00:49:29,870
All right, what we'll do is
we'll come back next time, and

1049
00:49:29,870 --> 00:49:33,440
we'll talk about what is the
government actually do to

1050
00:49:33,440 --> 00:49:35,260
actually affect redistribution
in society,

1051
00:49:35,260 --> 00:49:36,510
and how does it look?