1
00:00:00,090 --> 00:00:02,490
The following content is
provided under a Creative

2
00:00:02,490 --> 00:00:04,059
Commons license.

3
00:00:04,059 --> 00:00:06,330
Your support will help
MIT OpenCourseWare

4
00:00:06,330 --> 00:00:10,720
continue to offer high quality
educational resources for free.

5
00:00:10,720 --> 00:00:13,350
To make a donation or
view additional materials

6
00:00:13,350 --> 00:00:17,280
from hundreds of MIT courses,
visit MIT OpenCourseWare

7
00:00:17,280 --> 00:00:18,480
at ocw.mit.edu.

8
00:00:27,484 --> 00:00:28,770
ROBERT TOWNSEND: Today is--

9
00:00:28,770 --> 00:00:30,960
it's not exactly a
midpoint of the course,

10
00:00:30,960 --> 00:00:35,940
but it is a transition
lecture between

11
00:00:35,940 --> 00:00:40,080
these macro financial models
with micro underpinnings.

12
00:00:40,080 --> 00:00:42,990
Today we're going to go
through the measurement,

13
00:00:42,990 --> 00:00:49,560
and then in subsequent
lectures, focus on testing

14
00:00:49,560 --> 00:00:51,295
for the micro underpinnings.

15
00:00:51,295 --> 00:00:53,420
I mean, everything is
connected to everything else.

16
00:00:56,130 --> 00:00:56,940
There's a quote--

17
00:00:56,940 --> 00:00:59,550
I don't know, many of you
may be familiar with it.

18
00:00:59,550 --> 00:01:01,410
It's really quite good.

19
00:01:01,410 --> 00:01:03,950
Lord Kelvin.

20
00:01:03,950 --> 00:01:06,540
In physical science,
the first essential step

21
00:01:06,540 --> 00:01:10,500
in the direction of
learning any subject

22
00:01:10,500 --> 00:01:12,840
is define the principles
of numerical reckoning

23
00:01:12,840 --> 00:01:14,790
and practicable
methods for measuring

24
00:01:14,790 --> 00:01:18,090
some quality connected with it.

25
00:01:18,090 --> 00:01:20,730
I often say that
you can measure--

26
00:01:20,730 --> 00:01:23,460
when you can measure what you're
speaking about and express it

27
00:01:23,460 --> 00:01:27,130
in numbers, you know
something about it,

28
00:01:27,130 --> 00:01:29,490
but when you cannot measure
it, you cannot express it

29
00:01:29,490 --> 00:01:33,540
in numbers, your knowledge is
of a meager and unsatisfactory

30
00:01:33,540 --> 00:01:35,700
kind.

31
00:01:35,700 --> 00:01:37,440
It may be the
beginning of knowledge,

32
00:01:37,440 --> 00:01:40,260
but you have scarcely
in your thoughts

33
00:01:40,260 --> 00:01:43,380
advanced to the
stage of science,

34
00:01:43,380 --> 00:01:47,310
whatever that matter may be.

35
00:01:47,310 --> 00:01:50,340
So of course, today we're going
to talk about measurement,

36
00:01:50,340 --> 00:01:52,660
and this is not an
economics quote.

37
00:01:52,660 --> 00:02:00,510
This is a little more related
to both economics and impact

38
00:02:00,510 --> 00:02:01,440
evaluation.

39
00:02:01,440 --> 00:02:03,390
It's Bill Gates.

40
00:02:03,390 --> 00:02:05,205
He's talking about
the steam engine.

41
00:02:05,205 --> 00:02:10,229
In his book I think he must have
read on the most powerful idea

42
00:02:10,229 --> 00:02:14,790
in the world, which is
about measuring the energy

43
00:02:14,790 --> 00:02:19,130
output of engines with
this my micrometer.

44
00:02:22,290 --> 00:02:26,690
The point-- he's
quoting Mr. Rosen--

45
00:02:26,690 --> 00:02:28,920
is to see incremental
design changes

46
00:02:28,920 --> 00:02:30,480
that led to improvement--

47
00:02:30,480 --> 00:02:33,690
mainly what you want, higher
power and less consumption

48
00:02:33,690 --> 00:02:35,510
of the coal.

49
00:02:35,510 --> 00:02:39,710
And it was this interaction
between experimenting, trying

50
00:02:39,710 --> 00:02:43,400
things, measuring
precisely given the goal

51
00:02:43,400 --> 00:02:46,040
and then innovating
some more that

52
00:02:46,040 --> 00:02:52,250
led to this high-powered steam
engine which revolutionized

53
00:02:52,250 --> 00:02:55,370
the world, essentially.

54
00:02:55,370 --> 00:02:57,230
So he's saying he's
struck how important it

55
00:02:57,230 --> 00:03:01,370
is to measure-- to improve
the human condition

56
00:03:01,370 --> 00:03:05,060
and how incredible the progress
could be if you set a goal

57
00:03:05,060 --> 00:03:08,090
and measure it.

58
00:03:08,090 --> 00:03:11,990
But then he goes on
to say that he is also

59
00:03:11,990 --> 00:03:14,030
amazed how little
this is actually done

60
00:03:14,030 --> 00:03:17,020
and how hard it is to
do it to get it right.

61
00:03:17,020 --> 00:03:19,460
He gives an example
of foreign aid,

62
00:03:19,460 --> 00:03:23,510
just measuring the dollars,
say, that flow rather than

63
00:03:23,510 --> 00:03:26,270
the underlying impact
that it may or may not

64
00:03:26,270 --> 00:03:31,375
have on household business
or country welfare.

65
00:03:34,270 --> 00:03:36,870
So the type of measurements
that are needed or those that

66
00:03:36,870 --> 00:03:40,275
allow us to assess
performance and gauge impact.

67
00:03:45,150 --> 00:03:49,640
So this is the end of
what Bill Gates was saying

68
00:03:49,640 --> 00:03:51,590
and the rest is
something I want to say.

69
00:03:54,600 --> 00:03:56,520
Measurement is
tricky in the sense

70
00:03:56,520 --> 00:03:59,730
that he's really not saying
just go out and measure

71
00:03:59,730 --> 00:04:04,650
anything you want to
consistent with the goal.

72
00:04:04,650 --> 00:04:09,570
For a while the Gates Foundation
was promoting savings accounts,

73
00:04:09,570 --> 00:04:12,690
having poor people have
access to savings account

74
00:04:12,690 --> 00:04:17,610
as a kind of panacea,
and indeed, wanted

75
00:04:17,610 --> 00:04:19,800
to measure the
number of households

76
00:04:19,800 --> 00:04:24,150
out there now who have
them and track improvement.

77
00:04:24,150 --> 00:04:31,320
And there's a new world survey
done by the Gallup Foundation--

78
00:04:31,320 --> 00:04:33,390
Gallup people who do
the polls, actually,

79
00:04:33,390 --> 00:04:36,360
have survey
techniques worldwide.

80
00:04:39,440 --> 00:04:41,840
But of course, from
our perspective,

81
00:04:41,840 --> 00:04:44,690
just having a savings account
may not accomplish much.

82
00:04:44,690 --> 00:04:46,700
It's not a kind of
measurement that's really

83
00:04:46,700 --> 00:04:49,070
assessing the impact on people.

84
00:04:49,070 --> 00:04:52,940
You'd like to know what impact
it's having on welfare--

85
00:04:52,940 --> 00:04:54,920
that is to say on
actually using it

86
00:04:54,920 --> 00:04:58,520
or better ability to smooth
consumption or finance

87
00:04:58,520 --> 00:05:00,980
investments and so on.

88
00:05:03,850 --> 00:05:05,250
Well fortunately
in economics, we

89
00:05:05,250 --> 00:05:10,440
have one set of measurement
tools that are common,

90
00:05:10,440 --> 00:05:14,070
and they're also consistent
with the theme of the lectures--

91
00:05:14,070 --> 00:05:17,160
namely there is measurement
at the individual household

92
00:05:17,160 --> 00:05:21,300
level that's also
consistent with measurement

93
00:05:21,300 --> 00:05:24,780
of the flow of funds across
countries and everything

94
00:05:24,780 --> 00:05:26,290
in between.

95
00:05:26,290 --> 00:05:31,560
And what I'm going to
try to go through today

96
00:05:31,560 --> 00:05:37,380
but here's a quick summary of
it is what these tools are.

97
00:05:37,380 --> 00:05:43,070
Basically you start with
the financial accounts

98
00:05:43,070 --> 00:05:45,440
for a household or a firm.

99
00:05:45,440 --> 00:05:48,030
So that would be the balance
sheet, the income statement,

100
00:05:48,030 --> 00:05:51,290
and the statement of cash flow.

101
00:05:51,290 --> 00:05:55,430
You may associate that language
with corporate financial

102
00:05:55,430 --> 00:06:00,530
accounting and wonder
perhaps about its relevance,

103
00:06:00,530 --> 00:06:03,650
but they are the
benchmark standards

104
00:06:03,650 --> 00:06:07,160
and they can be applied to
households and especially

105
00:06:07,160 --> 00:06:09,050
to households running
businesses, which

106
00:06:09,050 --> 00:06:11,880
is the typical enterprise
in developing countries.

107
00:06:11,880 --> 00:06:20,130
So I'll show you in a moment
exactly how this can be done.

108
00:06:20,130 --> 00:06:23,550
What use can you
make of these things?

109
00:06:23,550 --> 00:06:26,670
Again, I'm going to try to
buttress my case that it's not

110
00:06:26,670 --> 00:06:28,560
just measurement for
the sake of measurement,

111
00:06:28,560 --> 00:06:33,130
but actually that you can
use these measurements.

112
00:06:33,130 --> 00:06:37,380
So in this case, we'll
draw a distinction between

113
00:06:37,380 --> 00:06:42,670
productivity and the return
on assets versus liquidity,

114
00:06:42,670 --> 00:06:45,870
some distinction you could
not make if you were not

115
00:06:45,870 --> 00:06:50,610
clear-headed as in applying
these financial accounts--

116
00:06:50,610 --> 00:06:53,790
or at least it would be
quite easy to make mistakes.

117
00:06:59,600 --> 00:07:05,410
And show how-- where
wealth is coming from

118
00:07:05,410 --> 00:07:08,020
and how wealth
changes over time.

119
00:07:08,020 --> 00:07:11,620
At the household level that
gets you into poverty dynamics,

120
00:07:11,620 --> 00:07:15,940
you can start thinking about the
mechanics that underlie changes

121
00:07:15,940 --> 00:07:16,960
in inequality.

122
00:07:22,010 --> 00:07:25,520
And again, in an
application, take a look

123
00:07:25,520 --> 00:07:31,840
at how households are making
a distinction, apparently,

124
00:07:31,840 --> 00:07:35,200
between savings rates
and productivity

125
00:07:35,200 --> 00:07:38,420
in terms of what they do
in their own investment.

126
00:07:38,420 --> 00:07:42,130
I think you've seen in
the recitation section

127
00:07:42,130 --> 00:07:44,590
sort of a precursor of
this in terms of the Thai

128
00:07:44,590 --> 00:07:47,660
data and some case studies.

129
00:07:47,660 --> 00:07:50,230
So in some sense, we're going
to go back to the basics

130
00:07:50,230 --> 00:07:53,360
and talk about the
financial accounts,

131
00:07:53,360 --> 00:07:59,830
but at least you have
a motivation already

132
00:07:59,830 --> 00:08:04,050
for what households are doing.

133
00:08:04,050 --> 00:08:07,560
Then we can aggregate
up from households--

134
00:08:07,560 --> 00:08:13,010
say all households in a
village and create something

135
00:08:13,010 --> 00:08:14,510
like the national
income accounts,

136
00:08:14,510 --> 00:08:17,090
except they're not at
the national level,

137
00:08:17,090 --> 00:08:19,580
they're village income accounts.

138
00:08:19,580 --> 00:08:26,450
But we have the standard savings
and investment, GDP numbers,

139
00:08:26,450 --> 00:08:31,850
and indeed, we get balance
of payments at the village

140
00:08:31,850 --> 00:08:35,929
level or regional
level just so you

141
00:08:35,929 --> 00:08:39,500
can see what's going on within
the country, not just of

142
00:08:39,500 --> 00:08:41,280
the country relative to other--

143
00:08:41,280 --> 00:08:44,330
and again, you've seen this
come up in the lectures--

144
00:08:44,330 --> 00:08:47,840
China, provincial-level
China, at the national level

145
00:08:47,840 --> 00:08:51,920
relative to the
rest of the world,

146
00:08:51,920 --> 00:08:55,340
flows across sectors that
differ in obstacles to trade.

147
00:08:59,980 --> 00:09:05,350
And finally, we'll get to the
flow of funds accounting, which

148
00:09:05,350 --> 00:09:07,690
I like to think of
as the nuts and bolts

149
00:09:07,690 --> 00:09:10,550
of the financial system
actually measured.

150
00:09:10,550 --> 00:09:13,330
So you don't just
have to imagine

151
00:09:13,330 --> 00:09:16,390
what's going on in
the country over time

152
00:09:16,390 --> 00:09:19,030
in response to shocks,
you can actually

153
00:09:19,030 --> 00:09:21,152
see it in the measurement.

154
00:09:26,820 --> 00:09:31,820
So first paper up.

155
00:09:31,820 --> 00:09:32,720
Not going to work?

156
00:09:36,560 --> 00:09:39,910
Let's not cut off too
much, maybe it's all right.

157
00:09:39,910 --> 00:09:50,000
Is to create-- well, I'm
just not fast enough.

158
00:09:55,260 --> 00:09:57,160
Create the balance
sheet, income statement,

159
00:09:57,160 --> 00:10:00,300
and statement of cash
flow for these households

160
00:10:00,300 --> 00:10:03,720
in order to make the
distinction between productivity

161
00:10:03,720 --> 00:10:05,530
and liquidity.

162
00:10:05,530 --> 00:10:12,330
So this is a quote from
Deaton that I ask often show.

163
00:10:12,330 --> 00:10:14,580
The only way to
obtain these measures

164
00:10:14,580 --> 00:10:16,680
is by imposing an
accounting framework

165
00:10:16,680 --> 00:10:19,950
on the data, painstakingly
construct estimates

166
00:10:19,950 --> 00:10:21,870
from myriad responses
to questions

167
00:10:21,870 --> 00:10:25,470
about the specific components
that contribute to the total.

168
00:10:25,470 --> 00:10:29,370
So actually, there's
a computer code

169
00:10:29,370 --> 00:10:34,410
that's you know 18 to 20
pages single-line code,

170
00:10:34,410 --> 00:10:36,880
and it operates
on the Thai data.

171
00:10:36,880 --> 00:10:40,830
It, of course, searches
over all the questions

172
00:10:40,830 --> 00:10:42,300
in the questionnaire
in order to be

173
00:10:42,300 --> 00:10:44,940
able to extract the
key variables that

174
00:10:44,940 --> 00:10:47,910
are needed line
by line to create

175
00:10:47,910 --> 00:10:49,500
these financial accounts.

176
00:10:49,500 --> 00:10:51,990
We didn't begin with
the financial accounts.

177
00:10:51,990 --> 00:10:55,530
I wish we had, because
then certain instances,

178
00:10:55,530 --> 00:10:58,290
it would have helped
us measure better.

179
00:10:58,290 --> 00:11:02,460
This was something
we came to afterwards

180
00:11:02,460 --> 00:11:04,470
when we were just struggling
with a simple idea

181
00:11:04,470 --> 00:11:07,920
of what do we really
mean by income?

182
00:11:07,920 --> 00:11:09,300
Something very basic.

183
00:11:09,300 --> 00:11:12,490
AUDIENCE: [INAUDIBLE] more
like 100 page, you think.

184
00:11:12,490 --> 00:11:13,740
ROBERT TOWNSEND: That's a lot.

185
00:11:13,740 --> 00:11:15,750
AUDIENCE: And the way that
you look at it is initial.

186
00:11:15,750 --> 00:11:16,667
ROBERT TOWNSEND: Yeah.

187
00:11:16,667 --> 00:11:19,470
Version 1, version 2.

188
00:11:19,470 --> 00:11:21,600
OK.

189
00:11:21,600 --> 00:11:22,980
So what are the
problems when you

190
00:11:22,980 --> 00:11:25,740
have these high frequency
monthly data or to some degree

191
00:11:25,740 --> 00:11:27,540
even annual data?

192
00:11:27,540 --> 00:11:29,760
There is a difference
in timing between, say,

193
00:11:29,760 --> 00:11:33,540
when inputs are purchased and
when their outputs are sold.

194
00:11:33,540 --> 00:11:37,380
If you think about farmers,
they only sell at harvest time

195
00:11:37,380 --> 00:11:39,930
and they're using
inputs along the way.

196
00:11:42,810 --> 00:11:47,790
It's true for businesses, too,
who don't sell instantaneously.

197
00:11:47,790 --> 00:11:51,150
They have to acquire inputs
to produce something or buy

198
00:11:51,150 --> 00:11:55,455
goods for inventory to stock,
and then subsequently sell.

199
00:11:59,930 --> 00:12:02,850
We are worried about
liquidity, on the other hand.

200
00:12:02,850 --> 00:12:05,090
So since smoothing
of consumption,

201
00:12:05,090 --> 00:12:07,010
protection of
investment from cash

202
00:12:07,010 --> 00:12:10,440
flow, financing a
budget deficits.

203
00:12:10,440 --> 00:12:14,600
So we want to make
distinction between cash flow

204
00:12:14,600 --> 00:12:20,090
in terms of liquidity and kind
of the overall performance.

205
00:12:20,090 --> 00:12:24,710
The key idea with performance
is an accrual notion

206
00:12:24,710 --> 00:12:28,250
of accounting, which
is simple to say--

207
00:12:28,250 --> 00:12:31,220
mainly do not book
or subtract expenses

208
00:12:31,220 --> 00:12:33,080
until the output is sold.

209
00:12:33,080 --> 00:12:35,300
It's the idea of
corporate finance--

210
00:12:35,300 --> 00:12:38,330
you have projects,
not occupations.

211
00:12:38,330 --> 00:12:39,680
You buy assets.

212
00:12:39,680 --> 00:12:44,420
You have an idea, you need to
fund it, you buy equipment,

213
00:12:44,420 --> 00:12:48,830
and you don't expect
a return right away.

214
00:12:48,830 --> 00:12:53,780
So for the short-run, you
have lots of negative profits,

215
00:12:53,780 --> 00:12:56,600
but it's just a
cash outflow, so you

216
00:12:56,600 --> 00:12:59,360
wait till you can get
the big picture, which

217
00:12:59,360 --> 00:13:04,260
is when you get the revenue then
you subtract off the inputs.

218
00:13:04,260 --> 00:13:06,270
But financing the
inputs along the way,

219
00:13:06,270 --> 00:13:08,700
that's a liquidity issue.

220
00:13:08,700 --> 00:13:10,890
And indeed, if
you're running out,

221
00:13:10,890 --> 00:13:15,840
you may as not invest even
though your underlying project

222
00:13:15,840 --> 00:13:17,970
or idea is quite successful.

223
00:13:22,820 --> 00:13:25,730
So we go to the
financial accounts.

224
00:13:25,730 --> 00:13:33,090
Now viewing households as firms,
we have to draw some parallels.

225
00:13:33,090 --> 00:13:36,660
Assets are kind of easy to think
about both on the household

226
00:13:36,660 --> 00:13:42,930
side and on the firm side.

227
00:13:42,930 --> 00:13:47,740
Households have debt just
like firms have liabilities,

228
00:13:47,740 --> 00:13:54,120
so those are straightforward
parallel concepts.

229
00:13:54,120 --> 00:13:56,950
When you think
about wealth, wealth

230
00:13:56,950 --> 00:14:00,415
is really net worth
on the household side.

231
00:14:03,010 --> 00:14:06,750
The language for firms
is something like equity.

232
00:14:06,750 --> 00:14:13,220
It's the amount of the owners'
investment in the enterprise

233
00:14:13,220 --> 00:14:18,600
that's built up over
time that is not

234
00:14:18,600 --> 00:14:21,980
as different from the fact
that some of the equipment

235
00:14:21,980 --> 00:14:24,210
gets owned by
outsiders who financed

236
00:14:24,210 --> 00:14:25,940
the project by lending.

237
00:14:29,670 --> 00:14:31,860
What is the sources
of this wealth?

238
00:14:31,860 --> 00:14:36,180
Well basically savings,
like retained earnings.

239
00:14:36,180 --> 00:14:40,050
Savings for households, retained
earnings for businesses.

240
00:14:40,050 --> 00:14:44,280
That part of the profits that
you put back into the business.

241
00:14:44,280 --> 00:14:48,210
But also potentially
contributed capital

242
00:14:48,210 --> 00:14:50,160
which I'll come to in a minute.

243
00:14:50,160 --> 00:14:54,840
This is this may strike you
as odd at first, I'm not sure.

244
00:14:54,840 --> 00:14:56,640
Consumption on
the household side

245
00:14:56,640 --> 00:15:00,200
is the same as dividends
on the firm side.

246
00:15:00,200 --> 00:15:03,620
So the consumption
we think we know,

247
00:15:03,620 --> 00:15:09,800
but dividends is
like a firm paying

248
00:15:09,800 --> 00:15:13,460
either the inside owners
or outside investors part

249
00:15:13,460 --> 00:15:14,420
of the return stream.

250
00:15:14,420 --> 00:15:17,870
So that stuff isn't going
into the firm anymore,

251
00:15:17,870 --> 00:15:20,060
it's not retained earnings.

252
00:15:20,060 --> 00:15:26,220
It's kind of the fruit
that you get to enjoy.

253
00:15:26,220 --> 00:15:31,350
This is also a bit of a
stretch and can be problematic.

254
00:15:31,350 --> 00:15:34,800
Incoming gifts on
the household side,

255
00:15:34,800 --> 00:15:38,550
what do we do with them
in the financial accounts?

256
00:15:38,550 --> 00:15:43,350
Gifts aren't really
income in some sense--

257
00:15:43,350 --> 00:15:45,120
in the corporate
financial sense.

258
00:15:45,120 --> 00:15:48,990
Because income is a
return stream of projects,

259
00:15:48,990 --> 00:15:54,570
but gifts, that's just
something somebody gave you.

260
00:15:54,570 --> 00:15:57,870
Well, so we ended up thinking
about gifts as financing.

261
00:15:57,870 --> 00:16:00,830
So the idea is, someone
gives you a gift,

262
00:16:00,830 --> 00:16:05,160
you have some implicit
idea of reciprocity.

263
00:16:05,160 --> 00:16:08,640
So they're in effect
helping your enterprise,

264
00:16:08,640 --> 00:16:12,950
they're investing
in your enterprise

265
00:16:12,950 --> 00:16:15,800
even though the contract
may not be nearly as formal

266
00:16:15,800 --> 00:16:18,765
as, say, an equity arrangement.

267
00:16:21,440 --> 00:16:23,720
And finally, the household
budget constraint

268
00:16:23,720 --> 00:16:28,680
is like the firm
cash flow constraint.

269
00:16:28,680 --> 00:16:31,740
So again, we can
therefore distinguish

270
00:16:31,740 --> 00:16:34,200
assets versus wealth.

271
00:16:34,200 --> 00:16:36,750
Note how easy it would be
to make a mistake there.

272
00:16:36,750 --> 00:16:37,860
Isn't that the same thing?

273
00:16:37,860 --> 00:16:38,400
No.

274
00:16:38,400 --> 00:16:44,270
Remember, the liabilities
get subtracted from assets

275
00:16:44,270 --> 00:16:46,460
to create, quote,
wealth or net worth.

276
00:16:46,460 --> 00:16:49,670
Accrual versus cash flow
I've said often enough.

277
00:16:49,670 --> 00:16:52,760
Savings as a budget
surplus as in cash flow--

278
00:16:52,760 --> 00:16:57,050
so you had more coming in in
revenue than you had outflows

279
00:16:57,050 --> 00:17:02,780
in terms of expenses versus a
different notion of savings,

280
00:17:02,780 --> 00:17:05,839
and that's savings as in
wealth accumulation, that's

281
00:17:05,839 --> 00:17:13,670
where do you put the increasing
wealth that you're acquiring?

282
00:17:13,670 --> 00:17:15,619
That's more like a
portfolio problem.

283
00:17:17,958 --> 00:17:19,500
And therefore, we
need to distinguish

284
00:17:19,500 --> 00:17:22,401
liquidity management
from asset management.

285
00:17:25,290 --> 00:17:27,900
So quick review.

286
00:17:27,900 --> 00:17:30,900
By the way, I'm really
going to struggle today

287
00:17:30,900 --> 00:17:35,460
to make sure you don't just
sort of pass out on me.

288
00:17:35,460 --> 00:17:37,650
Because accounting
has a reputation

289
00:17:37,650 --> 00:17:40,530
as being really
boring, and it is true

290
00:17:40,530 --> 00:17:43,710
that it gets more
and more tedious.

291
00:17:43,710 --> 00:17:45,690
And so you kind
of like tune out,

292
00:17:45,690 --> 00:17:51,270
like OK, I'll tolerate today.

293
00:17:51,270 --> 00:17:54,100
But no, you'll
discover that when

294
00:17:54,100 --> 00:17:57,310
you start doing your research,
whether it's looking at data

295
00:17:57,310 --> 00:18:01,780
from surveys or gathering
data or trying to do--

296
00:18:01,780 --> 00:18:05,313
and then you'll come back and
you'll be much more interested

297
00:18:05,313 --> 00:18:05,980
in these things.

298
00:18:05,980 --> 00:18:10,120
So the spirit with which I give
the lecture today is not the

299
00:18:10,120 --> 00:18:13,660
bore you with the details,
even though at times it

300
00:18:13,660 --> 00:18:17,560
will be tempting, because once
you've invested in this stuff,

301
00:18:17,560 --> 00:18:19,880
you want to share
that knowledge.

302
00:18:19,880 --> 00:18:24,160
But rather, to give you a
reader's guide or viewer's

303
00:18:24,160 --> 00:18:26,170
guide of what the
key concepts are

304
00:18:26,170 --> 00:18:27,980
and how things fit together.

305
00:18:27,980 --> 00:18:31,910
So that latter thing is what
I want you to take away.

306
00:18:31,910 --> 00:18:36,100
So here is a typical
balance sheet.

307
00:18:36,100 --> 00:18:40,120
You've got assets and
you've got liabilities.

308
00:18:40,120 --> 00:18:43,670
The bottom line
has to be the same.

309
00:18:43,670 --> 00:18:45,920
Now how does that happen?

310
00:18:45,920 --> 00:18:47,980
Well, you really
subtract liabilities

311
00:18:47,980 --> 00:18:52,820
off from the assets, and
hopefully assets are larger,

312
00:18:52,820 --> 00:18:57,390
so the residual is basically
equity or net worth, OK?

313
00:18:57,390 --> 00:18:59,100
And then this is
just an enumeration

314
00:18:59,100 --> 00:19:01,530
of different kinds
of assets, which

315
00:19:01,530 --> 00:19:05,700
include financial and
trade credit and so on as

316
00:19:05,700 --> 00:19:10,530
well as if-- lending as well
as your real physical fixed

317
00:19:10,530 --> 00:19:17,950
assets, and liabilities are
kind of obvious, and so on.

318
00:19:17,950 --> 00:19:25,710
So now let's get
into some cases here.

319
00:19:25,710 --> 00:19:28,320
The Living Standards
Measurement Survey

320
00:19:28,320 --> 00:19:34,810
makes a mistake if you want
to think of it that way.

321
00:19:34,810 --> 00:19:38,200
They ask about inputs used over
a specified cropping season

322
00:19:38,200 --> 00:19:39,115
and the amount spent.

323
00:19:42,470 --> 00:19:45,150
And implicitly
they're equating them,

324
00:19:45,150 --> 00:19:48,170
but they don't have to
be equal to each other.

325
00:19:48,170 --> 00:19:52,070
For one thing, you could be
using inputs in your farm

326
00:19:52,070 --> 00:19:55,280
that you acquired previously--
especially with monthly data,

327
00:19:55,280 --> 00:20:00,800
it's likely you have some
kind of input inventory.

328
00:20:00,800 --> 00:20:04,430
And for that matter,
you could spend money

329
00:20:04,430 --> 00:20:06,890
and buy inputs but not
actually put them on the field.

330
00:20:12,140 --> 00:20:19,370
And the same thing happens on
the output side, essentially.

331
00:20:19,370 --> 00:20:20,090
What do we do?

332
00:20:20,090 --> 00:20:23,660
Fortunately in the Thai
survey, we actually

333
00:20:23,660 --> 00:20:26,480
keep track of both
inputs required

334
00:20:26,480 --> 00:20:28,550
since the previous
interview as well as

335
00:20:28,550 --> 00:20:32,080
the actual use, both value
and quantity of inputs.

336
00:20:32,080 --> 00:20:35,555
So we have this input
inventory account.

337
00:20:39,260 --> 00:20:43,928
And revenues I say here,
but don't dwell on it.

338
00:20:43,928 --> 00:20:45,720
So what do you do when
you have the output?

339
00:20:45,720 --> 00:20:48,320
Well, you could store
it or you could sell it

340
00:20:48,320 --> 00:20:49,820
or you could eat it.

341
00:20:49,820 --> 00:20:52,460
So you kind of have to
create an output account.

342
00:20:52,460 --> 00:20:57,710
We act as if output is sold
at the time of harvest just

343
00:20:57,710 --> 00:21:01,190
to put an accurate value on
it, but when, in fact, they

344
00:21:01,190 --> 00:21:05,960
don't sell all of it or they
eat some and store some,

345
00:21:05,960 --> 00:21:10,290
then we create another sort
of output inventory account.

346
00:21:10,290 --> 00:21:14,240
So in high frequency data,
this sort of inventory

347
00:21:14,240 --> 00:21:16,753
that your mind might
glaze over when

348
00:21:16,753 --> 00:21:18,170
you look at a
financial statement,

349
00:21:18,170 --> 00:21:24,250
inventory may be a really
key smoothing device.

350
00:21:24,250 --> 00:21:27,730
And at the risk of
getting carried away,

351
00:21:27,730 --> 00:21:31,270
there's something called
work-in-progress inventory,

352
00:21:31,270 --> 00:21:34,060
and that happens when
you're using this accrual

353
00:21:34,060 --> 00:21:35,950
notion of accounting
where you don't

354
00:21:35,950 --> 00:21:39,010
want to subtract something
off as an expense yet.

355
00:21:39,010 --> 00:21:41,080
If you don't subtract
it off as an expense,

356
00:21:41,080 --> 00:21:44,120
then it's not on the
income statement.

357
00:21:44,120 --> 00:21:46,240
Well where the heck is it?

358
00:21:46,240 --> 00:21:49,120
Well basically,
you act like you've

359
00:21:49,120 --> 00:21:52,750
created an inventory
of something,

360
00:21:52,750 --> 00:21:54,880
like buying an asset.

361
00:21:54,880 --> 00:21:57,400
So all the fertilizer that
you've put on the land

362
00:21:57,400 --> 00:21:59,500
and all of that stuff
in the accrual notion

363
00:21:59,500 --> 00:22:03,010
is work-in-progress inventory,
it's producing the crop.

364
00:22:06,970 --> 00:22:07,770
Yep?

365
00:22:07,770 --> 00:22:10,200
AUDIENCE: So how do you--

366
00:22:10,200 --> 00:22:13,150
literally-- so how
do you kind of--

367
00:22:13,150 --> 00:22:16,380
so I'm assuming
that in your survey

368
00:22:16,380 --> 00:22:18,460
you ask kind of a
simpler question,

369
00:22:18,460 --> 00:22:20,593
like just count
assets and name--

370
00:22:20,593 --> 00:22:21,510
ROBERT TOWNSEND: Yeah.

371
00:22:21,510 --> 00:22:22,908
AUDIENCE: --expenses and what--

372
00:22:22,908 --> 00:22:23,950
ROBERT TOWNSEND: Exactly.

373
00:22:23,950 --> 00:22:26,340
AUDIENCE: --like that.

374
00:22:26,340 --> 00:22:30,900
How can you be sure that
your classification kind of

375
00:22:30,900 --> 00:22:34,560
is what it is and it's in
the mind of the households?

376
00:22:34,560 --> 00:22:40,410
So like I understand sort of
the asset versus wealth thing

377
00:22:40,410 --> 00:22:44,070
and so on, but when you go into
this like inventories and stuff

378
00:22:44,070 --> 00:22:44,867
like that, how--

379
00:22:44,867 --> 00:22:46,200
ROBERT TOWNSEND: We create that.

380
00:22:46,200 --> 00:22:47,930
We don't ask the
household what they're

381
00:22:47,930 --> 00:22:50,430
a work-in-progress inventory is.

382
00:22:50,430 --> 00:22:52,710
We have that as a
conceptual category

383
00:22:52,710 --> 00:22:54,990
as measured from
other variables,

384
00:22:54,990 --> 00:22:58,242
like what did they buy,
not yet put on the land.

385
00:22:58,242 --> 00:23:00,810
AUDIENCE: Would you think
it's interesting to go

386
00:23:00,810 --> 00:23:04,393
see whether that maps
in into some sort of--

387
00:23:04,393 --> 00:23:05,310
ROBERT TOWNSEND: Yeah.

388
00:23:05,310 --> 00:23:05,880
AUDIENCE: --way--

389
00:23:05,880 --> 00:23:07,755
ROBERT TOWNSEND: So
Chris Woodruff had done--

390
00:23:07,755 --> 00:23:10,710
the inventory thing is the
hardest thing to get right.

391
00:23:13,240 --> 00:23:15,130
And in fact, when
you think about it,

392
00:23:15,130 --> 00:23:18,250
you buy maybe
inputs at one price

393
00:23:18,250 --> 00:23:19,580
and you sell them in another.

394
00:23:19,580 --> 00:23:21,100
So you can have--

395
00:23:21,100 --> 00:23:24,580
or-- so you can have
capital gains and losses

396
00:23:24,580 --> 00:23:27,370
on the inventory account.

397
00:23:27,370 --> 00:23:28,180
AUDIENCE: Yeah--

398
00:23:28,180 --> 00:23:30,570
ROBERT TOWNSEND:
And-- and so yeah.

399
00:23:30,570 --> 00:23:33,210
So what Chris has
done is actually--

400
00:23:33,210 --> 00:23:36,040
he wrote a paper called--

401
00:23:36,040 --> 00:23:38,320
and we take the
opposite point of view,

402
00:23:38,320 --> 00:23:42,140
but his is, how is
the sausage made?

403
00:23:42,140 --> 00:23:44,090
Or don't ask what's
in the sausage.

404
00:23:44,090 --> 00:23:48,540
So he just asks households
basically what their income is,

405
00:23:48,540 --> 00:23:52,690
and the view there is
that somehow they're

406
00:23:52,690 --> 00:23:57,070
going to do better than asking
all these detailed questions.

407
00:23:57,070 --> 00:24:00,010
We think it's better to
ask the detailed questions

408
00:24:00,010 --> 00:24:03,700
and then construct the income,
but you can certainly--

409
00:24:03,700 --> 00:24:05,800
and we didn't do enough of this.

410
00:24:05,800 --> 00:24:06,940
So yes.

411
00:24:06,940 --> 00:24:12,050
You should compare the answers
of the households to so--

412
00:24:12,050 --> 00:24:15,130
and do the households have
in mind some liquidity notion

413
00:24:15,130 --> 00:24:18,010
of income or do they have
in mind some sort of rate

414
00:24:18,010 --> 00:24:21,010
of return calculation when you
ask them what their income is

415
00:24:21,010 --> 00:24:23,480
in a given year?

416
00:24:23,480 --> 00:24:25,460
AUDIENCE: Maybe even
like at an earlier stage.

417
00:24:25,460 --> 00:24:28,280
So I think that in the
whole boredom of accounting,

418
00:24:28,280 --> 00:24:29,600
there is some--

419
00:24:29,600 --> 00:24:31,960
what I remember as
more interesting stuff.

420
00:24:31,960 --> 00:24:35,840
It's like, everything that goes
on those statements you can--

421
00:24:35,840 --> 00:24:38,650
it's either market
value, cost value.

422
00:24:38,650 --> 00:24:41,440
So there's all sorts of--

423
00:24:41,440 --> 00:24:43,520
and you can get all
sorts of numbers

424
00:24:43,520 --> 00:24:47,190
depending on how
you value things.

425
00:24:47,190 --> 00:24:50,810
So my question would be
in those in these settings

426
00:24:50,810 --> 00:24:53,150
where I would
assume there's even

427
00:24:53,150 --> 00:24:56,720
measurement error in
prices and everything else,

428
00:24:56,720 --> 00:24:58,370
like how does that
all come together?

429
00:24:58,370 --> 00:25:00,578
And how do you think noise--

430
00:25:00,578 --> 00:25:02,370
ROBERT TOWNSEND: So
I'm going to take you--

431
00:25:02,370 --> 00:25:03,110
yeah.

432
00:25:03,110 --> 00:25:05,360
So I'll take you through
a paper we wrote.

433
00:25:05,360 --> 00:25:08,780
I'm going to formalize the
notion of rate of return

434
00:25:08,780 --> 00:25:10,880
on assets, and
then I'll show you

435
00:25:10,880 --> 00:25:14,720
one slide where we
did a lot of what--

436
00:25:14,720 --> 00:25:17,550
the great virtue of
having all the data

437
00:25:17,550 --> 00:25:19,370
is we could do these
experiments, say, what

438
00:25:19,370 --> 00:25:22,460
if we didn't count something
we actually measured,

439
00:25:22,460 --> 00:25:24,497
or what if we know
change categories

440
00:25:24,497 --> 00:25:26,080
and put it in one
place and the other?

441
00:25:26,080 --> 00:25:29,270
So actually rank-ordered
the three things

442
00:25:29,270 --> 00:25:32,490
that matter most to
measuring productivity

443
00:25:32,490 --> 00:25:36,170
so that we and especially
other researchers

444
00:25:36,170 --> 00:25:38,420
might be able to do better.

445
00:25:38,420 --> 00:25:40,840
Or not have to worry about
all the details on everything,

446
00:25:40,840 --> 00:25:41,990
that's more to the point.

447
00:25:41,990 --> 00:25:42,490
Yes?

448
00:25:42,490 --> 00:25:44,490
AUDIENCE: So what do you
do with households that

449
00:25:44,490 --> 00:25:45,750
also are running a business?

450
00:25:45,750 --> 00:25:48,866
Do you ask them only about
household inventories

451
00:25:48,866 --> 00:25:49,698
or the equivalent?

452
00:25:49,698 --> 00:25:51,740
ROBERT TOWNSEND: No, we
don't separate it at all.

453
00:25:51,740 --> 00:25:52,820
AUDIENCE: OK.

454
00:25:52,820 --> 00:25:56,210
ROBERT TOWNSEND: So we treat
it as an integrated unit.

455
00:25:56,210 --> 00:26:01,520
And again, my view of it is,
households don't necessarily

456
00:26:01,520 --> 00:26:02,860
make that separation.

457
00:26:02,860 --> 00:26:05,750
So we asked them
what is the profit

458
00:26:05,750 --> 00:26:07,250
from running the business?

459
00:26:07,250 --> 00:26:11,510
You're asking them
mentally to to separate out

460
00:26:11,510 --> 00:26:14,480
all these things, to have
some money from wage earning

461
00:26:14,480 --> 00:26:15,720
and so on.

462
00:26:15,720 --> 00:26:20,150
So we prefer to
measure everything--

463
00:26:20,150 --> 00:26:21,020
AUDIENCE: Sorry.

464
00:26:21,020 --> 00:26:21,895
ROBERT TOWNSEND: Yes?

465
00:26:21,895 --> 00:26:26,030
AUDIENCE: --quick follow-up
on [INAUDIBLE] question.

466
00:26:26,030 --> 00:26:29,060
In this sense in
which households

467
00:26:29,060 --> 00:26:32,960
that are also a
business or just treated

468
00:26:32,960 --> 00:26:35,720
as a business and
consumption decisions

469
00:26:35,720 --> 00:26:37,850
are treated as a
unitary things--

470
00:26:37,850 --> 00:26:42,290
really this notion of some sort
of corporate thing going on,

471
00:26:42,290 --> 00:26:45,800
how do you then go into--

472
00:26:45,800 --> 00:26:47,010
have you thought of--

473
00:26:47,010 --> 00:26:49,670
do you think it would be
interesting to like integrate

474
00:26:49,670 --> 00:26:54,710
that with household
decision-making modeling that

475
00:26:54,710 --> 00:26:57,570
also treats the household
as a corporate firm?

476
00:26:57,570 --> 00:27:01,670
So the idea of transaction
costs within the households,

477
00:27:01,670 --> 00:27:05,390
how like-- or the household then
is still a unitary thing which

478
00:27:05,390 --> 00:27:07,790
maybe doesn't make much sense
because there should be--

479
00:27:07,790 --> 00:27:10,953
there could be like the
woman is a manager of the--

480
00:27:10,953 --> 00:27:11,870
ROBERT TOWNSEND: Yeah.

481
00:27:11,870 --> 00:27:16,200
So there are two
things going on here.

482
00:27:16,200 --> 00:27:19,880
One is household non-business
activities and the other

483
00:27:19,880 --> 00:27:23,180
is business activities,
and we prefer

484
00:27:23,180 --> 00:27:26,150
to try to measure all of that
and have an integrated account.

485
00:27:26,150 --> 00:27:28,600
Now in terms of--

486
00:27:28,600 --> 00:27:31,760
a household is collection
of individuals,

487
00:27:31,760 --> 00:27:34,925
could we do the accounting
at the individual level?

488
00:27:37,460 --> 00:27:40,040
In some instances,
we're painfully aware

489
00:27:40,040 --> 00:27:44,150
that this notion we have
that this group of people

490
00:27:44,150 --> 00:27:47,960
is a household and is a
unit is stretching things,

491
00:27:47,960 --> 00:27:50,720
and yet that was our
starting point, to have

492
00:27:50,720 --> 00:27:52,910
a household-level survey.

493
00:27:52,910 --> 00:27:57,560
For one thing, the demands
even in the measurement

494
00:27:57,560 --> 00:28:02,060
of interviewing everybody
individually is horrendous.

495
00:28:02,060 --> 00:28:04,820
Also, you know some
assets are held jointly

496
00:28:04,820 --> 00:28:08,180
or people may
disagree about who's

497
00:28:08,180 --> 00:28:12,210
got the real claims on them.

498
00:28:12,210 --> 00:28:14,180
I mean, that said, there
are instances clearly

499
00:28:14,180 --> 00:28:21,020
where the husband may be
hiding some of his income

500
00:28:21,020 --> 00:28:24,950
or the young teenager
still living at home

501
00:28:24,950 --> 00:28:28,670
actually have some earnings
in his or her bank account

502
00:28:28,670 --> 00:28:30,920
and that's kind of not
commingled with the rest

503
00:28:30,920 --> 00:28:36,290
of the family budget, and those
are real, real problems for us.

504
00:28:36,290 --> 00:28:36,830
Yes?

505
00:28:36,830 --> 00:28:38,997
AUDIENCE: I mean, I think
to be fair to some extent,

506
00:28:38,997 --> 00:28:41,690
you can make that criticism of
corporate accounting as well,

507
00:28:41,690 --> 00:28:43,290
that maybe they're like--

508
00:28:43,290 --> 00:28:47,600
they're not acting as a unit or
there are people doing things

509
00:28:47,600 --> 00:28:49,245
that you can't see
in the accounts

510
00:28:49,245 --> 00:28:52,260
and that just looking
at a firm statement

511
00:28:52,260 --> 00:28:54,260
would be different if you
would like to sit down

512
00:28:54,260 --> 00:28:55,850
with every member
of a corporate firm

513
00:28:55,850 --> 00:28:58,735
and ask, what did you
spend money on this month?

514
00:28:58,735 --> 00:29:00,360
Where did you-- all
this sort of stuff.

515
00:29:00,360 --> 00:29:03,260
So I feel like to some
extent, the criticisms here

516
00:29:03,260 --> 00:29:06,260
are criticisms that apply to the
general framework of accounting

517
00:29:06,260 --> 00:29:08,860
and not to the household
problem specifically.

518
00:29:08,860 --> 00:29:11,193
ROBERT TOWNSEND: Yeah, there's
is some stuff I took out,

519
00:29:11,193 --> 00:29:13,730
otherwise the lecture
would be even longer, which

520
00:29:13,730 --> 00:29:18,380
is, do these problems-- you have
a conglomerate which consists

521
00:29:18,380 --> 00:29:22,360
of different firms, and
there are issues there

522
00:29:22,360 --> 00:29:26,260
about the flows from
one to the other.

523
00:29:26,260 --> 00:29:28,480
Here, you could treat a
marriage, for example,

524
00:29:28,480 --> 00:29:32,680
as a merger of two firms, and
there is development literature

525
00:29:32,680 --> 00:29:33,310
like that.

526
00:29:33,310 --> 00:29:35,950
Mark Rosenzweig
has written I think

527
00:29:35,950 --> 00:29:39,400
quite convincingly on
the marriage of daughters

528
00:29:39,400 --> 00:29:42,820
as basically insurance
and business transactions

529
00:29:42,820 --> 00:29:45,880
and so on.

530
00:29:45,880 --> 00:29:53,120
So there are-- a divorce is
like a breakup and so on.

531
00:30:01,700 --> 00:30:05,240
We try to measure
liquidity, and so

532
00:30:05,240 --> 00:30:08,090
it's tempting to just
have everything in cash,

533
00:30:08,090 --> 00:30:12,410
but actually here and in
India, grain, rice here,

534
00:30:12,410 --> 00:30:17,180
and sorghum in India is
people borrow and lend in it,

535
00:30:17,180 --> 00:30:18,890
they make commitments
to pay in it.

536
00:30:18,890 --> 00:30:23,630
So we actually have in
kind barter transactions.

537
00:30:23,630 --> 00:30:25,490
Now here again,
you could imagine--

538
00:30:25,490 --> 00:30:28,460
and when I teach
monetary economics,

539
00:30:28,460 --> 00:30:31,130
we get into this for sure.

540
00:30:31,130 --> 00:30:34,910
The notion of liquidity is how
quickly one could sell an asset

541
00:30:34,910 --> 00:30:37,520
or is there a deep
market for it.

542
00:30:37,520 --> 00:30:41,420
So this is, again, both
a judgment call here

543
00:30:41,420 --> 00:30:44,840
because we aggregate it
up, but also opens up

544
00:30:44,840 --> 00:30:49,460
a really interesting sort of
line of research where you

545
00:30:49,460 --> 00:30:53,360
can study liquidity and
transactions in these village

546
00:30:53,360 --> 00:30:55,550
economies.

547
00:30:55,550 --> 00:30:56,870
What to do with livestock?

548
00:30:59,630 --> 00:31:01,300
What does a cow--

549
00:31:01,300 --> 00:31:07,460
a cash cow-- well, so we
just separate the asset

550
00:31:07,460 --> 00:31:09,560
from the income
and expense part.

551
00:31:09,560 --> 00:31:11,720
So the cow is the asset.

552
00:31:11,720 --> 00:31:16,170
The cow, if it's young, actually
appreciates in value over time.

553
00:31:16,170 --> 00:31:18,410
You get a capital
gain out of it.

554
00:31:18,410 --> 00:31:22,400
Cattle can die prematurely,
you'll get a capital loss.

555
00:31:22,400 --> 00:31:25,490
We actually try to
book all those things.

556
00:31:25,490 --> 00:31:28,940
Meanwhile, the cow-- the
dairy cattle up in Lopburi

557
00:31:28,940 --> 00:31:30,710
are delivering milk.

558
00:31:30,710 --> 00:31:33,590
So that's the revenue stream
off the asset, actually.

559
00:31:33,590 --> 00:31:36,990
It's a great analogy,
like fruit off the tree.

560
00:31:36,990 --> 00:31:40,030
They're not Lucas trees,
they're, I don't know,

561
00:31:40,030 --> 00:31:41,120
Lopburi animals.

562
00:31:45,220 --> 00:31:48,910
And gifts, as I've said--

563
00:31:48,910 --> 00:31:52,200
so I don't think
I'll belabor that.

564
00:31:52,200 --> 00:31:55,020
I spared you a lot.

565
00:31:55,020 --> 00:31:57,450
Just want to
introduce the concept.

566
00:31:57,450 --> 00:32:00,240
In the underlying
data, you'll see

567
00:32:00,240 --> 00:32:05,140
a transaction, like received
wage income in cash.

568
00:32:05,140 --> 00:32:09,080
Now what to do with
the answer to that?

569
00:32:09,080 --> 00:32:10,530
50 baht.

570
00:32:10,530 --> 00:32:18,480
Well, it enters in all three
incomes, all three statements.

571
00:32:18,480 --> 00:32:23,070
Basically the cash enters an
increase in the balance sheet

572
00:32:23,070 --> 00:32:25,020
as an increase in
a kind of asset,

573
00:32:25,020 --> 00:32:27,300
it's revenue on the
income statement,

574
00:32:27,300 --> 00:32:32,440
and it's incoming cash flow
on the statement of cash flow.

575
00:32:32,440 --> 00:32:35,160
Not every transaction
in the data

576
00:32:35,160 --> 00:32:37,110
enter on all three accounts.

577
00:32:37,110 --> 00:32:39,810
The struggle was
deciding indeed where

578
00:32:39,810 --> 00:32:43,620
to put all of these things,
although the accounting

579
00:32:43,620 --> 00:32:47,040
standards double-entry
bookkeeping does provide

580
00:32:47,040 --> 00:32:51,300
a discipline-- and yes,
we made mistakes initially

581
00:32:51,300 --> 00:32:53,880
and things didn't add up.

582
00:32:53,880 --> 00:32:57,810
So then we realized we hadn't
entered things appropriately.

583
00:32:57,810 --> 00:33:00,210
That's what
double-entry bookkeeping

584
00:33:00,210 --> 00:33:02,580
is, some clerk having
all the transactions

585
00:33:02,580 --> 00:33:05,220
and then doing the books,
quote-unquote, consists

586
00:33:05,220 --> 00:33:09,300
of entering all those daily
transactions into ledgers

587
00:33:09,300 --> 00:33:13,670
and creating statements.

588
00:33:13,670 --> 00:33:16,460
So these happened to all be
cash transactions, that's

589
00:33:16,460 --> 00:33:19,070
kind of an accident,
it's just the first page

590
00:33:19,070 --> 00:33:21,930
of about eight or 10
of them in the book,

591
00:33:21,930 --> 00:33:25,080
and I'll skip the rest.

592
00:33:25,080 --> 00:33:28,560
Now let's talk about
liquidity management.

593
00:33:28,560 --> 00:33:32,720
Here's a statement of
cash flow, D for deficit.

594
00:33:32,720 --> 00:33:34,100
So you run a deficit.

595
00:33:34,100 --> 00:33:37,310
The deficit, say, is the
difference between your income

596
00:33:37,310 --> 00:33:39,290
in a month--

597
00:33:39,290 --> 00:33:42,260
cash revenue, I should say--

598
00:33:42,260 --> 00:33:47,630
and, say, consumption--
minus consumption,

599
00:33:47,630 --> 00:33:49,700
minus investment, OK?

600
00:33:49,700 --> 00:33:54,410
So it's a deficit, you
had more demands for cash

601
00:33:54,410 --> 00:33:55,855
than you received in cash.

602
00:33:58,670 --> 00:34:04,870
Well, again, by the
construction of the accounts,

603
00:34:04,870 --> 00:34:06,460
everything has to balance.

604
00:34:06,460 --> 00:34:11,480
So we know exactly how
a deficit is financed.

605
00:34:11,480 --> 00:34:13,960
We even try to
measure the currency.

606
00:34:13,960 --> 00:34:18,070
So we have F1 through Fn
are these financing devices,

607
00:34:18,070 --> 00:34:20,650
and then you can
do a little math,

608
00:34:20,650 --> 00:34:24,969
basically, multiply
both sides by D,

609
00:34:24,969 --> 00:34:28,520
but D is the sum of the F,
so basically you get this,

610
00:34:28,520 --> 00:34:30,370
and then you square
it, and then you

611
00:34:30,370 --> 00:34:33,820
divide by the different squared,
and all of a sudden you're

612
00:34:33,820 --> 00:34:36,040
looking at a variance--

613
00:34:36,040 --> 00:34:41,810
covariance decomposition,
which is exactly true.

614
00:34:41,810 --> 00:34:46,360
So if you want to know the role
that a particular financing

615
00:34:46,360 --> 00:34:50,409
device like F1 is playing
in terms of smoothing

616
00:34:50,409 --> 00:34:54,790
the deficits, it's
basically like running

617
00:34:54,790 --> 00:35:02,440
a regression of the deficit
onto the financing device

618
00:35:02,440 --> 00:35:04,780
and getting a
regression coefficient.

619
00:35:04,780 --> 00:35:06,070
It's really cute.

620
00:35:06,070 --> 00:35:09,070
That you could do it
one financing device

621
00:35:09,070 --> 00:35:11,260
at a time and the sum
of those coefficients

622
00:35:11,260 --> 00:35:13,330
has to add up to 1.

623
00:35:13,330 --> 00:35:16,390
So this isn't the only
way to talk about--

624
00:35:19,010 --> 00:35:22,040
this is variance/covariance,
not order of magnitude.

625
00:35:22,040 --> 00:35:25,320
If you have two devices that
track each other perfectly

626
00:35:25,320 --> 00:35:28,220
but there's always a
gap, this thing will say,

627
00:35:28,220 --> 00:35:31,790
it's the perfect smoother,
that's the thing that's used,

628
00:35:31,790 --> 00:35:33,380
but in fact, the gap remains.

629
00:35:33,380 --> 00:35:35,480
So you have to be a
little cautious about

630
00:35:35,480 --> 00:35:37,100
the interpretation sometimes.

631
00:35:43,480 --> 00:35:44,810
So how do you finance?

632
00:35:44,810 --> 00:35:47,380
Well you could
decrease your deposits

633
00:35:47,380 --> 00:35:52,990
if you have them in a financial
institution, you could--

634
00:35:52,990 --> 00:35:55,960
this is an odd way to put it--
decrease in your net ROSCA

635
00:35:55,960 --> 00:35:58,390
position, which means basically
you don't want to pay in,

636
00:35:58,390 --> 00:36:00,640
you want to get paid out.

637
00:36:00,640 --> 00:36:03,420
You want to be a recipient.

638
00:36:03,420 --> 00:36:07,240
You could call in your
loans, you could borrow,

639
00:36:07,240 --> 00:36:09,595
you could get gifts or
decrease your cash holdings.

640
00:36:12,130 --> 00:36:17,410
And for two case
study households,

641
00:36:17,410 --> 00:36:18,355
we actually measure--

642
00:36:21,450 --> 00:36:26,960
and you probably unfortunately
can't see it very well,

643
00:36:26,960 --> 00:36:33,430
but this cash is close to
100 or 1, basically, 100%.

644
00:36:36,100 --> 00:36:39,430
These numbers, though,
are the quartiles

645
00:36:39,430 --> 00:36:42,650
of the province in which this
case study household A lives.

646
00:36:42,650 --> 00:36:47,590
So the median is 76%, this
guy's actually on the high side,

647
00:36:47,590 --> 00:36:51,590
but still, in this
variance/covariance sense,

648
00:36:51,590 --> 00:36:58,750
3/4 of the deficits is
households in Lopburi

649
00:36:58,750 --> 00:37:03,640
are financing their
deficits with cash.

650
00:37:03,640 --> 00:37:04,140
Cash.

651
00:37:04,140 --> 00:37:06,150
They're holding
great gobs of it.

652
00:37:06,150 --> 00:37:09,000
Unbelievable amounts
of it in the house.

653
00:37:09,000 --> 00:37:12,450
And when they get
these adverse shocks

654
00:37:12,450 --> 00:37:16,683
or they have more expenses than
inputs, it outcomes to cash.

655
00:37:16,683 --> 00:37:18,850
You want to say, well why
isn't-- is it in a savings

656
00:37:18,850 --> 00:37:19,350
account?

657
00:37:19,350 --> 00:37:20,470
Not much.

658
00:37:20,470 --> 00:37:21,190
Negligible.

659
00:37:24,740 --> 00:37:28,800
Now I was a little
critical of this sort

660
00:37:28,800 --> 00:37:30,690
of measuring savings accounts.

661
00:37:30,690 --> 00:37:33,180
This is a measure
of savings account.

662
00:37:33,180 --> 00:37:35,340
I guess at the end
of the day, the issue

663
00:37:35,340 --> 00:37:38,430
is, are they managing
their cash well?

664
00:37:38,430 --> 00:37:39,792
This is the starting point.

665
00:37:39,792 --> 00:37:41,250
We have the
measurement, maybe they

666
00:37:41,250 --> 00:37:43,980
should be putting
more of it in the bank

667
00:37:43,980 --> 00:37:45,480
and taking it out
when they need it.

668
00:37:45,480 --> 00:37:51,450
They don't do it much, so
that is a bit of a puzzle.

669
00:37:51,450 --> 00:37:54,360
Gifts and loans are--

670
00:37:54,360 --> 00:37:55,620
I think this is not--

671
00:37:55,620 --> 00:37:58,922
this is out of the book
and not the latest version.

672
00:37:58,922 --> 00:38:00,130
AUDIENCE: This is the latest.

673
00:38:00,130 --> 00:38:02,340
ROBERT TOWNSEND: Is it?

674
00:38:02,340 --> 00:38:06,190
So OK.

675
00:38:06,190 --> 00:38:12,070
So you can see these upper
quartiles are getting up

676
00:38:12,070 --> 00:38:14,350
to 15%, 20%, 25%.

677
00:38:14,350 --> 00:38:21,220
So gifts are used
by some households

678
00:38:21,220 --> 00:38:23,650
substantially as well.

679
00:38:23,650 --> 00:38:26,647
It's not just a
pure cash economy.

680
00:38:29,815 --> 00:38:31,440
And like-- well that's
cash management.

681
00:38:31,440 --> 00:38:33,970
We can talk about
asset management.

682
00:38:33,970 --> 00:38:37,050
It's really the same
concept, except instead

683
00:38:37,050 --> 00:38:39,300
of operating on
the cash flow, you

684
00:38:39,300 --> 00:38:41,310
do this accrual notion
of accounting and look

685
00:38:41,310 --> 00:38:44,730
at sort of net savings.

686
00:38:44,730 --> 00:38:49,680
And I'll come back to
that later, essentially.

687
00:38:49,680 --> 00:38:51,570
And the portfolio management.

688
00:38:51,570 --> 00:38:55,010
So here, in answer
to the question that

689
00:38:55,010 --> 00:38:58,100
was raised earlier, is
looking at something

690
00:38:58,100 --> 00:39:02,690
like profits over assets,
which is a return on assets,

691
00:39:02,690 --> 00:39:09,830
and then creating 10
different measurements of it.

692
00:39:09,830 --> 00:39:13,340
Do you want to exclude household
consumption of their own output

693
00:39:13,340 --> 00:39:16,040
and not count it into income?

694
00:39:16,040 --> 00:39:19,550
We counted total utility
expenses as a business expense,

695
00:39:19,550 --> 00:39:21,560
maybe we shouldn't
have done that.

696
00:39:21,560 --> 00:39:24,680
Maybe it was a household
expense for, quote,

697
00:39:24,680 --> 00:39:25,745
consumption purposes.

698
00:39:28,420 --> 00:39:30,730
So service flow
from fixed assets,

699
00:39:30,730 --> 00:39:33,290
which would be very natural
to do for durable goods,

700
00:39:33,290 --> 00:39:34,130
we tried that.

701
00:39:37,160 --> 00:39:40,047
We could take income but
then add gifts and transfers

702
00:39:40,047 --> 00:39:41,630
even though it's
really not associated

703
00:39:41,630 --> 00:39:44,960
with a project, and
especially this sort of cash.

704
00:39:47,630 --> 00:39:50,390
So it turned out
that this distinction

705
00:39:50,390 --> 00:39:54,890
of cash versus accrual
was very important.

706
00:39:54,890 --> 00:39:57,470
The treatment of gifts
was very important.

707
00:39:57,470 --> 00:40:00,260
And up here, which where I
didn't start, but I could,

708
00:40:00,260 --> 00:40:05,120
is in trying to get at the
cost of household labor--

709
00:40:05,120 --> 00:40:08,010
own household labor
used in production,

710
00:40:08,010 --> 00:40:13,460
that is a monster difficulty,
because you don't see for many

711
00:40:13,460 --> 00:40:17,510
of them what they would have
earned as wages in someone

712
00:40:17,510 --> 00:40:21,020
else's farm and so on, yet it's
clear that's very important

713
00:40:21,020 --> 00:40:22,891
and it makes a very
big difference.

714
00:40:26,120 --> 00:40:28,970
So hopefully knowing
that, one can bend over

715
00:40:28,970 --> 00:40:32,450
backwards to try to get a
better sense of the opportunity

716
00:40:32,450 --> 00:40:33,710
cost of labor.

717
00:40:33,710 --> 00:40:39,320
So as I said, let's look
at some uses of these.

718
00:40:39,320 --> 00:40:49,540
This is a paper with Anan
on the wealth distribution,

719
00:40:49,540 --> 00:40:53,340
and let's look at that 1%.

720
00:40:53,340 --> 00:40:54,330
Let's go get 'em.

721
00:40:54,330 --> 00:40:56,300
Let's get the top 1%.

722
00:40:56,300 --> 00:41:00,530
How much wealth does
the top 1% have?

723
00:41:00,530 --> 00:41:02,810
It's huge.

724
00:41:02,810 --> 00:41:08,540
They start out in 1999 with
about 35% of the total wealth.

725
00:41:11,530 --> 00:41:15,430
Now it's actually diminished.

726
00:41:15,430 --> 00:41:18,910
This is the-- frankly,
a wonderful thing

727
00:41:18,910 --> 00:41:21,190
to have all these many,
many years of data.

728
00:41:23,710 --> 00:41:26,110
It's diminished to 26%.

729
00:41:26,110 --> 00:41:29,400
What's going on is this
stuff at the bottom end

730
00:41:29,400 --> 00:41:35,770
where the bottom 50%
have 5% of the wealth.

731
00:41:35,770 --> 00:41:41,900
By the way, this is typical of
the US and European history,

732
00:41:41,900 --> 00:41:45,230
and it's very skewed
wealth distributions.

733
00:41:45,230 --> 00:41:49,790
But this thing actually
doubles over the space

734
00:41:49,790 --> 00:41:53,110
of about 12 years here.

735
00:41:53,110 --> 00:41:55,043
AUDIENCE: Is this
including cash?

736
00:41:55,043 --> 00:41:55,960
ROBERT TOWNSEND: Yeah.

737
00:41:55,960 --> 00:41:57,390
AUDIENCE: So it
could be like cash.

738
00:41:57,390 --> 00:41:58,598
ROBERT TOWNSEND: It could be.

739
00:42:01,140 --> 00:42:02,830
There was a question.

740
00:42:02,830 --> 00:42:04,320
AUDIENCE: Oh yes.

741
00:42:04,320 --> 00:42:08,730
Not related to this, but what
household characteristics does

742
00:42:08,730 --> 00:42:15,463
[INAUDIBLE] economic ones like
[INAUDIBLE] household members--

743
00:42:15,463 --> 00:42:16,880
ROBERT TOWNSEND:
Yeah, we have all

744
00:42:16,880 --> 00:42:19,400
of that, all the demography.

745
00:42:19,400 --> 00:42:23,015
Age, education.

746
00:42:23,015 --> 00:42:25,000
Yeah.

747
00:42:25,000 --> 00:42:30,550
So in answer to your
question, we'll see what the--

748
00:42:30,550 --> 00:42:32,520
that among some of
these poor people,

749
00:42:32,520 --> 00:42:35,450
they have a very high return on
assets, real physical assets,

750
00:42:35,450 --> 00:42:37,450
and we'll see where they're
putting their money.

751
00:42:41,970 --> 00:42:45,470
So inequalities going
down in some sense, right?

752
00:42:45,470 --> 00:42:49,160
So anyway, this is not the
story of a poverty trap.

753
00:42:51,710 --> 00:42:54,020
Now I'm not claiming
this is universal truth,

754
00:42:54,020 --> 00:42:56,960
I'm not claiming there
aren't poverty traps, really,

755
00:42:56,960 --> 00:43:01,100
and models and in
the real world.

756
00:43:01,100 --> 00:43:02,600
Many countries still have--

757
00:43:02,600 --> 00:43:08,750
like India-- vast
populations of poor people,

758
00:43:08,750 --> 00:43:12,430
but maybe be a little leery
that at least in the case

759
00:43:12,430 --> 00:43:16,190
of Thailand, you think about
starting at a low level--

760
00:43:16,190 --> 00:43:18,230
and some of these guys
remain in poverty,

761
00:43:18,230 --> 00:43:22,090
but there is a huge
growth dynamic going on.

762
00:43:22,090 --> 00:43:22,700
Yes?

763
00:43:22,700 --> 00:43:26,270
AUDIENCE: And how
much is the very--

764
00:43:26,270 --> 00:43:28,230
in terms of the actual
distribution of income,

765
00:43:28,230 --> 00:43:30,890
how much of the very
rich got surveyed?

766
00:43:30,890 --> 00:43:32,877
How much-- of this--

767
00:43:32,877 --> 00:43:34,460
the people who got
surveyed, were they

768
00:43:34,460 --> 00:43:36,080
like a top truncated
distribution,

769
00:43:36,080 --> 00:43:38,168
like their very interested
and you know, the ones

770
00:43:38,168 --> 00:43:39,210
who answered this survey.

771
00:43:39,210 --> 00:43:40,400
It's usual--

772
00:43:40,400 --> 00:43:43,200
ROBERT TOWNSEND: Yeah, it's--

773
00:43:43,200 --> 00:43:45,420
well, it's a stratified
random sample.

774
00:43:45,420 --> 00:43:48,000
In some cases we actually
ended up with everybody

775
00:43:48,000 --> 00:43:52,620
in the village, but maybe
only two out of the 16 or so

776
00:43:52,620 --> 00:43:54,495
villages did that happen.

777
00:43:57,020 --> 00:43:59,660
I think it's more
typical that we're

778
00:43:59,660 --> 00:44:02,210
worried the rich guys are
hiding some of their assets

779
00:44:02,210 --> 00:44:07,620
rather than them not
responding at all.

780
00:44:07,620 --> 00:44:09,630
I could not run these
surveys on my own,

781
00:44:09,630 --> 00:44:13,050
I have built up close
friendships and collaborators

782
00:44:13,050 --> 00:44:18,000
in Thailand who are amazingly
good at getting households

783
00:44:18,000 --> 00:44:21,300
to participate, and there
are occasional scares where

784
00:44:21,300 --> 00:44:24,130
they think the government's
going to start taxing them,

785
00:44:24,130 --> 00:44:28,530
the chicken farmers were making
great gobs of money in profits,

786
00:44:28,530 --> 00:44:34,650
and rumors circulated that they
better conceal their wealth.

787
00:44:34,650 --> 00:44:36,780
They dropped out, but we
got almost all of them

788
00:44:36,780 --> 00:44:39,684
back in after a while.

789
00:44:39,684 --> 00:44:42,930
AUDIENCE: So I told them that
attrition is very low, but what

790
00:44:42,930 --> 00:44:44,690
about the initial response rate?

791
00:44:44,690 --> 00:44:45,620
What is that?

792
00:44:45,620 --> 00:44:46,505
Close to 100%?

793
00:44:46,505 --> 00:44:48,005
Like when you
actually went in and--

794
00:44:48,005 --> 00:44:48,922
ROBERT TOWNSEND: Yeah.

795
00:44:48,922 --> 00:44:49,680
It was very high.

796
00:44:53,488 --> 00:44:55,868
AUDIENCE: What do they get
for answering the survey?

797
00:44:55,868 --> 00:44:56,344
ROBERT TOWNSEND: Hmm?

798
00:44:56,344 --> 00:44:57,719
AUDIENCE: I think
Harry mentioned

799
00:44:57,719 --> 00:44:59,740
that people get gifts for--

800
00:44:59,740 --> 00:45:02,050
AUDIENCE: No, it's
not substantial.

801
00:45:02,050 --> 00:45:03,770
It's just like you
get a new year's gift.

802
00:45:03,770 --> 00:45:05,470
So it's not worth the time.

803
00:45:05,470 --> 00:45:07,787
So they don't like look
at it from a financial--

804
00:45:07,787 --> 00:45:09,370
ROBERT TOWNSEND: We
give them a bottle

805
00:45:09,370 --> 00:45:11,770
of fish sauce or a
blanket, or something,

806
00:45:11,770 --> 00:45:15,070
and that's a Thai
sort of cultural thing

807
00:45:15,070 --> 00:45:19,540
to establish goodwill and
that you care about a person.

808
00:45:19,540 --> 00:45:21,700
We just did not
and we deliberately

809
00:45:21,700 --> 00:45:26,642
avoided getting into the
business of paying people.

810
00:45:26,642 --> 00:45:29,100
Because for one thing, then
other households in the village

811
00:45:29,100 --> 00:45:35,910
feel like they're getting
cheated and it's just-- it's--

812
00:45:35,910 --> 00:45:39,180
So this is all
voluntary participation.

813
00:45:39,180 --> 00:45:41,760
13 years of households
answering questions

814
00:45:41,760 --> 00:45:46,515
monthly with only, whatever
it is, 2% per year dropping.

815
00:45:50,670 --> 00:45:58,030
So here is-- there's a
movie called Falling Down--

816
00:45:58,030 --> 00:46:00,390
I don't even remember
that guy went berserko

817
00:46:00,390 --> 00:46:01,380
with a machine gun.

818
00:46:05,220 --> 00:46:06,240
I always think of that.

819
00:46:10,980 --> 00:46:13,720
So not everybody's
wealth is going up,

820
00:46:13,720 --> 00:46:15,750
and in fact, some
people actually

821
00:46:15,750 --> 00:46:18,780
find their wealth is falling.

822
00:46:18,780 --> 00:46:21,540
So we can track
their net position

823
00:46:21,540 --> 00:46:24,930
in the distribution of
wealth, and here, you

824
00:46:24,930 --> 00:46:28,980
can see these
people climbing up,

825
00:46:28,980 --> 00:46:31,980
and other people,
quote, falling down.

826
00:46:31,980 --> 00:46:35,400
This is the histogram of
sort of in percentiles.

827
00:46:35,400 --> 00:46:40,460
By the way, the average
has to be 0 because it's--

828
00:46:40,460 --> 00:46:44,290
the net change is 0
over all the households,

829
00:46:44,290 --> 00:46:49,220
but you can see
some substantial--

830
00:46:49,220 --> 00:46:52,886
1 standard deviation is
like 14 percentage points.

831
00:46:52,886 --> 00:46:57,473
AUDIENCE: So do you have any
[INAUDIBLE] falling down?

832
00:46:57,473 --> 00:46:58,390
ROBERT TOWNSEND: Yeah.

833
00:46:58,390 --> 00:47:02,990
So I guess one thing that sort
of got lost in the shuffle

834
00:47:02,990 --> 00:47:08,060
because we didn't decide
clearly whether it would show

835
00:47:08,060 --> 00:47:11,150
within the TA session or
I would show it here is

836
00:47:11,150 --> 00:47:13,400
kind of a life cycle picture.

837
00:47:13,400 --> 00:47:19,090
So we've created
by cohorts, by age.

838
00:47:19,090 --> 00:47:23,250
And you can see, especially
in the areas near Bangkok,

839
00:47:23,250 --> 00:47:25,980
a classic life cycle.

840
00:47:25,980 --> 00:47:29,410
So as these
households are young,

841
00:47:29,410 --> 00:47:33,280
they start to accumulate
wealth, but after a while,

842
00:47:33,280 --> 00:47:34,990
especially as their
income falls off,

843
00:47:34,990 --> 00:47:36,680
they start running
down their assets.

844
00:47:36,680 --> 00:47:39,440
So this does not
distinguish by age.

845
00:47:39,440 --> 00:47:42,820
So it's a bit misleading,
it's not like there's some--

846
00:47:42,820 --> 00:47:46,240
necessarily some young household
that's just consistently eating

847
00:47:46,240 --> 00:47:49,090
more than their income.

848
00:47:49,090 --> 00:47:53,100
There are households like that,
and that raises this issue

849
00:47:53,100 --> 00:47:55,350
before of what's going on.

850
00:47:55,350 --> 00:47:59,910
I mean, if the husband
isn't there, for example,

851
00:47:59,910 --> 00:48:03,030
and the wife is running
some hair salon,

852
00:48:03,030 --> 00:48:07,110
she's consistently eating more
than then the household income,

853
00:48:07,110 --> 00:48:11,200
but you kind of wonder if
there isn't something going on,

854
00:48:11,200 --> 00:48:13,080
and we just don't
have the husband there

855
00:48:13,080 --> 00:48:17,430
who chooses either for
personal or occupation reasons

856
00:48:17,430 --> 00:48:20,790
to be in Bangkok the whole time.

857
00:48:20,790 --> 00:48:25,300
So then we get a
misleading picture

858
00:48:25,300 --> 00:48:34,020
that she may be in trouble or
she may know more the future

859
00:48:34,020 --> 00:48:34,620
than we do.

860
00:48:37,150 --> 00:48:41,750
So let's do a simple
decomposition.

861
00:48:41,750 --> 00:48:43,660
Here's savings.

862
00:48:43,660 --> 00:48:48,220
Savings on both sides
in absolute amount.

863
00:48:48,220 --> 00:48:50,860
And then divide and
multiply by profits

864
00:48:50,860 --> 00:48:53,470
from the household and
businesses, all of them

865
00:48:53,470 --> 00:48:57,040
summed up, and divide
and multiply by assets.

866
00:48:57,040 --> 00:48:59,200
So that has to be true.

867
00:48:59,200 --> 00:49:00,520
It's an identity.

868
00:49:00,520 --> 00:49:05,830
But savings over
basically profits

869
00:49:05,830 --> 00:49:10,810
is a savings rate out of
income, and profits over assets

870
00:49:10,810 --> 00:49:14,800
is the return on
assets post-multiplied

871
00:49:14,800 --> 00:49:17,510
by the level of assets.

872
00:49:17,510 --> 00:49:22,600
So basically you can look at--

873
00:49:36,060 --> 00:49:41,010
you can look at the
change in wealth

874
00:49:41,010 --> 00:49:45,440
or the percentage change in
wealth, and wealth comes--

875
00:49:45,440 --> 00:49:48,100
where is a change in
wealth coming from?

876
00:49:48,100 --> 00:49:52,170
It's coming from savings
plus income and gifts.

877
00:49:52,170 --> 00:49:55,450
The savings thing is
decomposed up here.

878
00:49:55,450 --> 00:49:59,370
So part of the mechanical
accounting drivers

879
00:49:59,370 --> 00:50:03,990
of increasing wealth has
to do with savings rates

880
00:50:03,990 --> 00:50:09,550
and rates of return on assets.

881
00:50:09,550 --> 00:50:12,080
And maybe gifts if
they're important.

882
00:50:12,080 --> 00:50:16,630
So here is the
correlation of the growth

883
00:50:16,630 --> 00:50:21,903
of net worth with savings
rates, and depending

884
00:50:21,903 --> 00:50:23,320
on whether the
unit of observation

885
00:50:23,320 --> 00:50:26,770
is household month or household
year or households averaged up

886
00:50:26,770 --> 00:50:30,010
over all of the months
and years of the data,

887
00:50:30,010 --> 00:50:36,940
you do see some positive
significant correlations.

888
00:50:36,940 --> 00:50:40,390
So in fact, not
too surprisingly,

889
00:50:40,390 --> 00:50:43,450
some of the increase
in net worth

890
00:50:43,450 --> 00:50:48,580
is coming from households
who have high savings rates.

891
00:50:48,580 --> 00:50:54,740
But the bigger part of the
story is the correlation

892
00:50:54,740 --> 00:50:57,620
with the return on assets.

893
00:50:57,620 --> 00:51:02,420
And there, it's consistent
pretty much at all levels.

894
00:51:06,190 --> 00:51:10,030
So household by household
with our measurement,

895
00:51:10,030 --> 00:51:13,930
we look at their productivity
in terms of rate of return,

896
00:51:13,930 --> 00:51:17,720
and then see whether their net
wealth is going up or down,

897
00:51:17,720 --> 00:51:20,650
and it's the households who
have high rates of return who

898
00:51:20,650 --> 00:51:22,420
have the high--

899
00:51:22,420 --> 00:51:24,810
highest rates of increase
in their net worth.

900
00:51:30,040 --> 00:51:31,620
And it's quite persistent.

901
00:51:31,620 --> 00:51:35,820
Here, we look at return on
assets in the first half

902
00:51:35,820 --> 00:51:40,830
and second half of the sample
at the time that we did it.

903
00:51:40,830 --> 00:51:46,350
So if they were identical, they
would lie on a 45-degree line,

904
00:51:46,350 --> 00:51:50,670
and these scatter points are
close with one exception, which

905
00:51:50,670 --> 00:51:53,802
is Buriram.

906
00:51:53,802 --> 00:51:55,930
Oh, you should be thinking
models at this point.

907
00:51:55,930 --> 00:52:01,110
Remember the productivity,
the TFP thing?

908
00:52:01,110 --> 00:52:03,975
That's the thing sitting outside
multiplicatively in front

909
00:52:03,975 --> 00:52:05,100
of the production function?

910
00:52:05,100 --> 00:52:08,730
These are household-specific
TFP numbers sort of.

911
00:52:08,730 --> 00:52:11,280
It's not a very sophisticated--
production function

912
00:52:11,280 --> 00:52:12,930
is just linear.

913
00:52:12,930 --> 00:52:15,890
So it's not really just--

914
00:52:15,890 --> 00:52:19,080
you'd rather have like the
marginal product of capital

915
00:52:19,080 --> 00:52:23,160
and pre-multiply, this is a
linear production function.

916
00:52:23,160 --> 00:52:27,770
The marginal product
is just the average.

917
00:52:27,770 --> 00:52:28,840
But it's very persistent.

918
00:52:28,840 --> 00:52:34,300
So in these models, the issue
is, are those TFP shocks IID

919
00:52:34,300 --> 00:52:36,070
over households,
are they the same

920
00:52:36,070 --> 00:52:38,190
unless you get blasted
with some shock?

921
00:52:38,190 --> 00:52:42,010
And this suggests that
the household level

922
00:52:42,010 --> 00:52:43,950
it's really quite persistent.

923
00:52:43,950 --> 00:52:50,690
Buriram is an
interesting exception,

924
00:52:50,690 --> 00:52:52,760
and I'll just say
it in words, there

925
00:52:52,760 --> 00:52:57,020
are many more occupations
shifts in Buriram

926
00:52:57,020 --> 00:52:59,390
than in the other
three provinces.

927
00:52:59,390 --> 00:53:03,870
It turns out these villages
aren't really villages anymore,

928
00:53:03,870 --> 00:53:06,350
they've paved over our
soil moisture readers

929
00:53:06,350 --> 00:53:09,350
and they become
part of the town.

930
00:53:09,350 --> 00:53:11,720
There's a big construction
boom going on.

931
00:53:11,720 --> 00:53:14,540
Part of that is
pararubber, part of it

932
00:53:14,540 --> 00:53:19,580
is Scandinavian tourists
coming in marrying Thai women

933
00:53:19,580 --> 00:53:23,282
and staying.

934
00:53:23,282 --> 00:53:26,990
The place is just like
a construction boom.

935
00:53:26,990 --> 00:53:30,320
So people shifted
from rice-growing

936
00:53:30,320 --> 00:53:32,780
or money-lending or
other things to laying

937
00:53:32,780 --> 00:53:35,420
tile and all these
things, and they

938
00:53:35,420 --> 00:53:37,310
had a higher return
on their assets

939
00:53:37,310 --> 00:53:40,700
as a result on average
of those switches.

940
00:53:40,700 --> 00:53:42,120
So that's kind of reassuring.

941
00:53:42,120 --> 00:53:44,360
It's another sense
in which households

942
00:53:44,360 --> 00:53:47,960
are aware of their success.

943
00:53:47,960 --> 00:53:51,560
And here, coming back
to Witt's question,

944
00:53:51,560 --> 00:53:54,600
if you partition
into low, medium,

945
00:53:54,600 --> 00:53:56,510
and high return on
assets and looked

946
00:53:56,510 --> 00:54:01,850
at what they did with
their increasing wealth,

947
00:54:01,850 --> 00:54:08,050
these high ROA guys are putting
money back in the business.

948
00:54:08,050 --> 00:54:10,460
They're buying equipment,
they're investing and putting

949
00:54:10,460 --> 00:54:12,940
it into their business.

950
00:54:12,940 --> 00:54:16,100
And that's not true of
the medium or low guys.

951
00:54:16,100 --> 00:54:18,120
These guys are doing
the financial assets.

952
00:54:18,120 --> 00:54:21,680
So more of their
increasing financial wealth

953
00:54:21,680 --> 00:54:27,440
is potentially in cash, but to
the extent that we measured ROA

954
00:54:27,440 --> 00:54:29,390
accurately, which
we have tried very

955
00:54:29,390 --> 00:54:32,120
hard to do in all kinds of
different ways, we're getting

956
00:54:32,120 --> 00:54:35,240
the reinforcing message
here that we've measured it

957
00:54:35,240 --> 00:54:38,775
accurately because they seem
to do what you would think,

958
00:54:38,775 --> 00:54:41,150
which is, you're not going to
put your money in a savings

959
00:54:41,150 --> 00:54:45,150
account when you've got
a really high return.

960
00:54:45,150 --> 00:54:47,910
And later in another lecture
will try to adjust for risk.

961
00:54:51,340 --> 00:54:57,380
So now we'll aggregate
it up to villages

962
00:54:57,380 --> 00:54:59,090
and use those same
financial accounts.

963
00:55:03,120 --> 00:55:06,260
We do have to struggle
again with this dual role

964
00:55:06,260 --> 00:55:11,210
that households play as
consumers and producers.

965
00:55:11,210 --> 00:55:14,140
So much of economics
is driven by here

966
00:55:14,140 --> 00:55:17,690
is firms maximizing profits,
here is households maximizing

967
00:55:17,690 --> 00:55:19,820
utility-- these are
households running businesses,

968
00:55:19,820 --> 00:55:21,660
so what are we going to do?

969
00:55:21,660 --> 00:55:25,590
And in particular, what are
we going to do with the labor?

970
00:55:25,590 --> 00:55:28,910
Well, we made a decision,
although here, we do

971
00:55:28,910 --> 00:55:31,180
check and do robustness checks.

972
00:55:31,180 --> 00:55:33,320
And what I'm going
to show you today

973
00:55:33,320 --> 00:55:36,500
is treating even the
labor income as something

974
00:55:36,500 --> 00:55:42,620
like consulting income, like
they're providing a service--

975
00:55:42,620 --> 00:55:45,110
their business is
to provide a labor

976
00:55:45,110 --> 00:55:48,440
service to other households,
and they get profits from that.

977
00:55:54,820 --> 00:55:58,420
Now if you think about a
production sector and aggregate

978
00:55:58,420 --> 00:56:03,850
up, to the extent that inputs
from one business are--

979
00:56:03,850 --> 00:56:06,310
the outputs from one business
are an input to the other,

980
00:56:06,310 --> 00:56:08,620
the net is zero and
it doesn't count.

981
00:56:08,620 --> 00:56:12,220
But in these villages,
the output of one business

982
00:56:12,220 --> 00:56:15,760
could be someone
else's consumption,

983
00:56:15,760 --> 00:56:18,200
and it doesn't net out.

984
00:56:18,200 --> 00:56:20,030
Both get measured.

985
00:56:20,030 --> 00:56:22,580
As it would if you're
distinguishing households

986
00:56:22,580 --> 00:56:25,190
from firms.

987
00:56:25,190 --> 00:56:28,940
We can actually measure and
ask about the transactions.

988
00:56:28,940 --> 00:56:32,750
So we know the transaction
partners almost.

989
00:56:32,750 --> 00:56:36,050
We did screw up with
consumption because we

990
00:56:36,050 --> 00:56:38,150
forgot to ask
consistently where they

991
00:56:38,150 --> 00:56:42,230
buy stuff, which has to do with
the price thing also mentioned

992
00:56:42,230 --> 00:56:43,730
earlier.

993
00:56:43,730 --> 00:56:45,680
But fortunately,
and I'm sure you'll

994
00:56:45,680 --> 00:56:48,260
be more than happy to
take my word for it,

995
00:56:48,260 --> 00:56:52,550
there is an accounting identity,
so you can kind of back up

996
00:56:52,550 --> 00:56:57,770
as a residual how much of
that output is basically

997
00:56:57,770 --> 00:57:01,850
sold and consumed in the village
because we measured everything

998
00:57:01,850 --> 00:57:03,300
else.

999
00:57:03,300 --> 00:57:05,197
In other words, on
the buying side,

1000
00:57:05,197 --> 00:57:07,780
we don't see it, but-- because
we don't know where they bought

1001
00:57:07,780 --> 00:57:10,210
it, but we actually figured
out on the selling side

1002
00:57:10,210 --> 00:57:12,965
that it came from somebody who
produced it in the village.

1003
00:57:20,060 --> 00:57:21,740
Measurement error,
you asked-- somebody

1004
00:57:21,740 --> 00:57:25,070
asked about the rich
guys not participating.

1005
00:57:25,070 --> 00:57:28,730
If a moneylender were
like the central node

1006
00:57:28,730 --> 00:57:32,690
and the money lender refused
to answer our questions,

1007
00:57:32,690 --> 00:57:34,190
we're going to get
really screwed up

1008
00:57:34,190 --> 00:57:38,127
in terms of the flow of funds
and the balance of payments

1009
00:57:38,127 --> 00:57:38,960
and everything else.

1010
00:57:38,960 --> 00:57:41,600
Because we are going to act
like the average overall--

1011
00:57:41,600 --> 00:57:45,260
the households in the
village is the representation

1012
00:57:45,260 --> 00:57:49,397
of what's going on in the
village itself as a whole.

1013
00:57:49,397 --> 00:57:51,230
To the extent that
there's measurement error

1014
00:57:51,230 --> 00:57:53,935
but there is IID
over households,

1015
00:57:53,935 --> 00:57:56,060
and to the extent that we
have a lot of households,

1016
00:57:56,060 --> 00:57:59,120
then hopefully that
helps make the case

1017
00:57:59,120 --> 00:58:02,750
that we've got a
reasonably good measure.

1018
00:58:10,690 --> 00:58:14,225
So how do you create
these accounts?

1019
00:58:17,550 --> 00:58:20,035
What is GNP anyway?

1020
00:58:20,035 --> 00:58:20,535
GDP?

1021
00:58:23,130 --> 00:58:26,674
Everyone seems to be
obsessed about it.

1022
00:58:26,674 --> 00:58:31,180
OK, it just comes from
the income statement.

1023
00:58:31,180 --> 00:58:33,660
So basically, here's
a statement of income,

1024
00:58:33,660 --> 00:58:35,340
it's what you would expect.

1025
00:58:35,340 --> 00:58:39,000
You've got revenues come from
production, or for that matter,

1026
00:58:39,000 --> 00:58:42,390
if you're a lender or
saver from interest,

1027
00:58:42,390 --> 00:58:43,620
you've got capital gains.

1028
00:58:43,620 --> 00:58:46,230
You subtract off
capital losses, maybe

1029
00:58:46,230 --> 00:58:51,870
you have sort of a paid payment
from an insurance indemnity.

1030
00:58:51,870 --> 00:58:54,900
So those are revenues,
here is basically expenses,

1031
00:58:54,900 --> 00:58:56,940
production expenses,
interest expenses

1032
00:58:56,940 --> 00:59:01,910
on loans, depreciation,
insurance premia, taxes, et

1033
00:59:01,910 --> 00:59:02,440
cetera.

1034
00:59:02,440 --> 00:59:10,960
So net income before tax--

1035
00:59:10,960 --> 00:59:13,920
before income tax but
after property tax

1036
00:59:13,920 --> 00:59:16,350
is what makes these things
balance because this

1037
00:59:16,350 --> 00:59:17,610
is basically the profit--

1038
00:59:17,610 --> 00:59:21,920
it's basically
revenues less expenses.

1039
00:59:21,920 --> 00:59:26,060
So that should be
comfortable, sound sort of

1040
00:59:26,060 --> 00:59:28,310
reassuringly familiar.

1041
00:59:28,310 --> 00:59:33,740
All that we do is
basically take this stuff

1042
00:59:33,740 --> 00:59:36,440
and subtract it off
from both sides.

1043
00:59:36,440 --> 00:59:38,420
That cancels it out
on the right-hand side

1044
00:59:38,420 --> 00:59:41,160
and puts it on the
left-hand side.

1045
00:59:41,160 --> 00:59:46,970
So basically you have production
revenue less production

1046
00:59:46,970 --> 00:59:59,070
expenses, and that's
basically the rest

1047
00:59:59,070 --> 01:00:02,470
is equal to the payments
to factors of production.

1048
01:00:02,470 --> 01:00:05,220
So you're paying
stuff out in interest,

1049
01:00:05,220 --> 01:00:11,230
you're paying out in taxes,
depreciation is over here,

1050
01:00:11,230 --> 01:00:14,940
and you're paying out to
the owners of the enterprise

1051
01:00:14,940 --> 01:00:17,250
namely profits.

1052
01:00:17,250 --> 01:00:20,430
So you may remember
vaguely that there's

1053
01:00:20,430 --> 01:00:24,810
two ways to measure
income on the output side

1054
01:00:24,810 --> 01:00:28,140
and on the factor payment side,
and they add up to each other.

1055
01:00:28,140 --> 01:00:33,240
So this is
value-added, basically.

1056
01:00:33,240 --> 01:00:37,170
The only catch here is
the way we treated labor,

1057
01:00:37,170 --> 01:00:40,080
you don't see a labor
expense over here,

1058
01:00:40,080 --> 01:00:42,330
because normally
value-added, you're

1059
01:00:42,330 --> 01:00:45,810
only subtracting off
intermediate inputs

1060
01:00:45,810 --> 01:00:48,720
and other purchased inputs,
but here, we actually subtract

1061
01:00:48,720 --> 01:00:53,820
off the labor because of
the way we conceptualize

1062
01:00:53,820 --> 01:00:59,200
that problem or that issue.

1063
01:00:59,200 --> 01:01:03,150
So you want to see what's
happening over time in the four

1064
01:01:03,150 --> 01:01:09,400
provinces, four villages
per province, Chachoengsao

1065
01:01:09,400 --> 01:01:17,080
GDP or really village product is
going down pretty consistently.

1066
01:01:17,080 --> 01:01:20,080
Now what happened is the
shrimp ponds, basically,

1067
01:01:20,080 --> 01:01:21,970
people had
converted-- yeah, that

1068
01:01:21,970 --> 01:01:24,160
stuff you're buying
in Costco or Walmart,

1069
01:01:24,160 --> 01:01:25,660
those are typically Thai shrimp.

1070
01:01:28,840 --> 01:01:33,390
But the ponds-- so
there's some disease.

1071
01:01:33,390 --> 01:01:38,010
And so-- and household,
some of them switched.

1072
01:01:38,010 --> 01:01:41,180
And the fish and others
now are actually,

1073
01:01:41,180 --> 01:01:43,670
with the world grain prices
being relatively high,

1074
01:01:43,670 --> 01:01:49,027
they're refilling in the
ponds and planting rice again.

1075
01:01:49,027 --> 01:01:49,610
AUDIENCE: Rob?

1076
01:01:49,610 --> 01:01:50,050
ROBERT TOWNSEND: Yes?

1077
01:01:50,050 --> 01:01:52,370
AUDIENCE: What are the different
colored lines [INAUDIBLE]

1078
01:01:52,370 --> 01:01:53,580
ROBERT TOWNSEND: Each
of four villages.

1079
01:01:53,580 --> 01:01:53,930
So--

1080
01:01:53,930 --> 01:01:54,763
AUDIENCE: Oh, I see.

1081
01:01:54,763 --> 01:01:56,740
[INAUDIBLE]

1082
01:01:56,740 --> 01:01:59,560
ROBERT TOWNSEND: So you
can see, is every village

1083
01:01:59,560 --> 01:02:00,640
like every other village?

1084
01:02:00,640 --> 01:02:03,550
Well, to some degree, yes,
but the levels are different,

1085
01:02:03,550 --> 01:02:09,460
and sometimes even the movements
over time are a bit different.

1086
01:02:09,460 --> 01:02:11,620
Another thing to watch
out for is the scale here,

1087
01:02:11,620 --> 01:02:16,450
because in order to clearly
display each village--

1088
01:02:16,450 --> 01:02:18,910
each of the four villages in
a province, where changing

1089
01:02:18,910 --> 01:02:20,570
the scale of a province.

1090
01:02:20,570 --> 01:02:23,920
Sisaket is a lot poorer than--

1091
01:02:23,920 --> 01:02:27,760
even in their worst days,
Chachoengsao villages

1092
01:02:27,760 --> 01:02:38,060
are better off
substantially than Sisaket.

1093
01:02:38,060 --> 01:02:42,500
So the income levels
are maybe on average

1094
01:02:42,500 --> 01:02:44,400
three times higher
in Chachoengsao.

1095
01:02:48,130 --> 01:02:52,200
They're almost ordered by
their distance to Bangkok.

1096
01:02:52,200 --> 01:02:54,440
I won't bore you with this.

1097
01:02:54,440 --> 01:02:56,130
This is just to
what do households

1098
01:02:56,130 --> 01:02:59,375
do with their profits?

1099
01:02:59,375 --> 01:03:04,410
And then the story is mostly
in the northeast. they eat it.

1100
01:03:04,410 --> 01:03:08,010
And in central region, they're
more likely to put it back

1101
01:03:08,010 --> 01:03:12,750
in the business or their
farm as retained earnings.

1102
01:03:12,750 --> 01:03:19,250
Savings investment
account, that's

1103
01:03:19,250 --> 01:03:21,540
probably worth looking at.

1104
01:03:21,540 --> 01:03:23,010
Here's the balance sheet.

1105
01:03:23,010 --> 01:03:26,940
So remember, we had assets
and liabilities and net worth.

1106
01:03:26,940 --> 01:03:31,190
This is the change over
time in the balance sheet.

1107
01:03:31,190 --> 01:03:34,710
So it's like a
first-time difference.

1108
01:03:34,710 --> 01:03:40,960
So changes in liabilities
and changes in net worth

1109
01:03:40,960 --> 01:03:46,060
must equal to the change
in all the assets,

1110
01:03:46,060 --> 01:03:48,880
especially if you include
change in net worth over there.

1111
01:03:51,630 --> 01:03:55,270
These could be
financial or real.

1112
01:03:55,270 --> 01:03:58,770
Well hopefully this is
comfortable and familiar.

1113
01:03:58,770 --> 01:04:01,830
Just the notion of having assets
and liabilities and net worth

1114
01:04:01,830 --> 01:04:04,170
and taking a time difference.

1115
01:04:04,170 --> 01:04:07,310
What is the savings
investment account?

1116
01:04:07,310 --> 01:04:10,970
Well, you just basically take
the change in liabilities

1117
01:04:10,970 --> 01:04:14,420
and put it on the
left-hand side.

1118
01:04:14,420 --> 01:04:18,870
That leaves the change in net
worth on the right-hand side,

1119
01:04:18,870 --> 01:04:21,800
and then what you do with
it on the left-hand side.

1120
01:04:21,800 --> 01:04:22,580
Yes?

1121
01:04:22,580 --> 01:04:26,660
AUDIENCE: So before we were
talking about household gifts

1122
01:04:26,660 --> 01:04:28,395
equity.

1123
01:04:28,395 --> 01:04:29,270
ROBERT TOWNSEND: Yep.

1124
01:04:29,270 --> 01:04:30,978
AUDIENCE: So does that
mean that they get

1125
01:04:30,978 --> 01:04:32,380
counted as financial assets?

1126
01:04:36,270 --> 01:04:38,613
Where would gifts
in a village enter?

1127
01:04:38,613 --> 01:04:40,155
ROBERT TOWNSEND:
Contributed capital.

1128
01:04:40,155 --> 01:04:41,270
AUDIENCE: Contributed capital.

1129
01:04:41,270 --> 01:04:41,720
ROBERT TOWNSEND: Yeah.

1130
01:04:41,720 --> 01:04:43,127
That's what we called it.

1131
01:04:43,127 --> 01:04:43,544
AUDIENCE: All right.

1132
01:04:43,544 --> 01:04:43,961
Cool.

1133
01:04:43,961 --> 01:04:44,380
Thanks.

1134
01:04:44,380 --> 01:04:45,140
I'm sorry, it just occurred to--

1135
01:04:45,140 --> 01:04:45,980
ROBERT TOWNSEND:
No, no, it's fine.

1136
01:04:45,980 --> 01:04:47,070
AUDIENCE: --no idea
where they're going.

1137
01:04:47,070 --> 01:04:48,237
ROBERT TOWNSEND: Yeah, yeah.

1138
01:04:48,237 --> 01:04:51,018
It's there, because if
they're getting gifts and not

1139
01:04:51,018 --> 01:04:53,060
eating too much, it should
increase their wealth,

1140
01:04:53,060 --> 01:04:55,640
and therefore, it should show
up somewhere as either reduced

1141
01:04:55,640 --> 01:04:58,640
debt or increased assets.

1142
01:04:58,640 --> 01:05:04,850
And that phenomenon depends
a lot on where you are.

1143
01:05:04,850 --> 01:05:07,210
AUDIENCE: But entered
both sides, right?

1144
01:05:07,210 --> 01:05:10,762
It entered as increased cash,
and then also increased--

1145
01:05:10,762 --> 01:05:11,720
ROBERT TOWNSEND: Well--

1146
01:05:11,720 --> 01:05:12,560
AUDIENCE: --contributed capital.

1147
01:05:12,560 --> 01:05:14,102
ROBERT TOWNSEND:
The concept is here.

1148
01:05:14,102 --> 01:05:16,610
So basically say
this is positive,

1149
01:05:16,610 --> 01:05:20,550
and your change in
net worth is positive

1150
01:05:20,550 --> 01:05:23,760
because either retained
earnings or contributed capital.

1151
01:05:23,760 --> 01:05:25,920
And then the issue is
portfolio management,

1152
01:05:25,920 --> 01:05:27,900
it's what did you do with it?

1153
01:05:27,900 --> 01:05:32,260
And you could put it in cash,
that's in financial assets,

1154
01:05:32,260 --> 01:05:38,580
but you could also have it in
inventory as I was alluding to,

1155
01:05:38,580 --> 01:05:43,990
or have livestock assets
or other fixed assets.

1156
01:05:43,990 --> 01:05:46,180
So it's worthy of
study to figure out

1157
01:05:46,180 --> 01:05:48,268
what they're doing with it.

1158
01:05:48,268 --> 01:05:49,726
AUDIENCE: And
sometimes the idea is

1159
01:05:49,726 --> 01:05:51,730
there's a reciprocity,
and the gift that I

1160
01:05:51,730 --> 01:05:55,690
give to my buddy in
the next village is

1161
01:05:55,690 --> 01:05:56,840
in some sense a residual--

1162
01:05:56,840 --> 01:05:57,757
ROBERT TOWNSEND: Yeah.

1163
01:05:57,757 --> 01:05:58,690
AUDIENCE: So yeah.

1164
01:05:58,690 --> 01:05:58,930
ROBERT TOWNSEND: Yeah.

1165
01:05:58,930 --> 01:06:00,290
AUDIENCE: So yea, I
guess that's what--

1166
01:06:00,290 --> 01:06:01,690
ROBERT TOWNSEND: I mean,
the extreme versions of that

1167
01:06:01,690 --> 01:06:04,030
would be the kids
that go off to Bangkok

1168
01:06:04,030 --> 01:06:06,150
and they send money back.

1169
01:06:06,150 --> 01:06:08,110
Now can you distinguish
they're just

1170
01:06:08,110 --> 01:06:09,640
caring about their parents?

1171
01:06:09,640 --> 01:06:14,500
And there's a very strong
Thai tradition there.

1172
01:06:14,500 --> 01:06:18,010
But then eventually they
come back and run the farm.

1173
01:06:24,590 --> 01:06:27,080
But it is-- it's a
real incoming flow,

1174
01:06:27,080 --> 01:06:29,060
so we have to do
something with it.

1175
01:06:29,060 --> 01:06:32,297
We could just not call
it equity or something,

1176
01:06:32,297 --> 01:06:34,130
and in effect, that's
what we're doing here,

1177
01:06:34,130 --> 01:06:39,510
is just keeping track of it
separately as a line item.

1178
01:06:39,510 --> 01:06:40,010
OK.

1179
01:06:40,010 --> 01:06:47,300
So I guess I said this before.

1180
01:06:47,300 --> 01:06:50,480
Oh, by the way, on the
lifecycle diagrams,

1181
01:06:50,480 --> 01:06:56,350
it's really dramatic
that in the northeast,

1182
01:06:56,350 --> 01:06:59,890
wealth isn't dropping
as these guys get older,

1183
01:06:59,890 --> 01:07:01,780
and likewise, they're
not accumulating

1184
01:07:01,780 --> 01:07:05,800
a lot of financial or real
assets as they're young

1185
01:07:05,800 --> 01:07:07,540
and getting older either.

1186
01:07:07,540 --> 01:07:09,970
There doesn't seem to
be this life cycle,

1187
01:07:09,970 --> 01:07:12,470
and the gifts just
get bigger and bigger.

1188
01:07:12,470 --> 01:07:15,910
You don't have to get
older to get those gifts.

1189
01:07:15,910 --> 01:07:16,840
It's the huge--

1190
01:07:16,840 --> 01:07:21,940
I think it's overlapping
generations, essentially.

1191
01:07:21,940 --> 01:07:25,440
It's really quite different
between the northeast

1192
01:07:25,440 --> 01:07:28,000
and the central area,
and that's related,

1193
01:07:28,000 --> 01:07:30,585
because it's kind
of showing up here.

1194
01:07:30,585 --> 01:07:32,335
So then we get to the
balance of payments.

1195
01:07:39,830 --> 01:07:44,420
So basically, villages trade.

1196
01:07:44,420 --> 01:07:47,480
They produce stuff and
sell it and they buy stuff

1197
01:07:47,480 --> 01:07:50,040
as inputs and by consumption.

1198
01:07:50,040 --> 01:07:54,260
So we got sort of a trade
balance that we measure.

1199
01:07:54,260 --> 01:07:59,180
You could add in interest
earnings from, quote, abroad,

1200
01:07:59,180 --> 01:08:01,520
but it's really mostly
from the same country.

1201
01:08:01,520 --> 01:08:03,440
They're getting interest
on savings accounts

1202
01:08:03,440 --> 01:08:07,410
that they held outside the
village or in those gifts

1203
01:08:07,410 --> 01:08:07,910
again.

1204
01:08:11,470 --> 01:08:14,740
So you may or may
not have seen this,

1205
01:08:14,740 --> 01:08:17,080
this gets really
kind of confusing

1206
01:08:17,080 --> 01:08:20,140
the way the standard
accounting is done.

1207
01:08:20,140 --> 01:08:26,410
Trade balance plus this other
earnings on gifts and interest

1208
01:08:26,410 --> 01:08:28,870
is called the current
account, and then you

1209
01:08:28,870 --> 01:08:31,990
have the current
account balanced

1210
01:08:31,990 --> 01:08:35,109
with the financial count
and the capital account.

1211
01:08:35,109 --> 01:08:36,970
But they didn't
write current account

1212
01:08:36,970 --> 01:08:42,490
equal financial account,
so this stuff has

1213
01:08:42,490 --> 01:08:48,609
to be-- if you're trading
a lot, like China,

1214
01:08:48,609 --> 01:08:51,640
and you have a big sort
of current account,

1215
01:08:51,640 --> 01:08:54,569
this stuff has to be negative.

1216
01:08:54,569 --> 01:08:56,660
A negative in the
financial account

1217
01:08:56,660 --> 01:08:58,710
is actually, quote, a
good thing in the sense

1218
01:08:58,710 --> 01:09:03,050
that you're accumulating assets.

1219
01:09:03,050 --> 01:09:06,270
But believe me, it'll
throw you every time when

1220
01:09:06,270 --> 01:09:09,330
you start thinking about it.

1221
01:09:09,330 --> 01:09:12,210
And then I won't
drive this for you,

1222
01:09:12,210 --> 01:09:14,700
but it turns out that
if you take this sort

1223
01:09:14,700 --> 01:09:19,109
of Keynesian-looking
thing, which is really just

1224
01:09:19,109 --> 01:09:23,340
an accounting statement,
income less consumption less

1225
01:09:23,340 --> 01:09:27,029
investment plus transfers--
that's not taxes--

1226
01:09:27,029 --> 01:09:32,340
has to be identically
equal to the trade balance

1227
01:09:32,340 --> 01:09:35,220
plus sort of capital
account balance, which

1228
01:09:35,220 --> 01:09:39,069
we keep track of separately,
plus the change in inventories.

1229
01:09:39,069 --> 01:09:44,430
So largely, if they
run an export surplus,

1230
01:09:44,430 --> 01:09:47,064
they're going to save
it, but not quite.

1231
01:09:47,064 --> 01:09:48,689
Because they have
other ways of saving,

1232
01:09:48,689 --> 01:09:51,569
you have to adjust for
changes in the capital,

1233
01:09:51,569 --> 01:09:54,990
buying and selling with
foreigners, so to speak,

1234
01:09:54,990 --> 01:09:56,490
and the changes in inventory.

1235
01:09:56,490 --> 01:09:59,424
AUDIENCE: So the left-hand
side is a financial account

1236
01:09:59,424 --> 01:10:01,325
in this case?

1237
01:10:01,325 --> 01:10:02,700
ROBERT TOWNSEND:
It's not obvious

1238
01:10:02,700 --> 01:10:05,750
how you get from this to this.

1239
01:10:05,750 --> 01:10:10,250
And I decided not to
try to derive it today.

1240
01:10:10,250 --> 01:10:12,050
But it is an identity.

1241
01:10:12,050 --> 01:10:13,370
One follows from the other.

1242
01:10:17,520 --> 01:10:18,970
And here, the
balance of payments

1243
01:10:18,970 --> 01:10:24,220
accounts for these
villages, mostly

1244
01:10:24,220 --> 01:10:27,440
this story is they're
running surpluses.

1245
01:10:27,440 --> 01:10:31,410
The trade or trade and
current account is this blue

1246
01:10:31,410 --> 01:10:34,860
or blue-red stuff, and--

1247
01:10:34,860 --> 01:10:36,720
although here you
see a difference.

1248
01:10:36,720 --> 01:10:39,680
Chachoengsao's
getting in trouble,

1249
01:10:39,680 --> 01:10:43,530
its balance of payments
is going down in the sense

1250
01:10:43,530 --> 01:10:47,640
that it's running less
and less of a surplus.

1251
01:10:47,640 --> 01:10:50,140
Whereas Lopburi is
the other story.

1252
01:10:50,140 --> 01:10:51,480
It's always increasing.

1253
01:10:51,480 --> 01:10:54,330
The difference
between red and blue

1254
01:10:54,330 --> 01:10:59,700
is basically the gifts that play
a big role in the northeast.

1255
01:10:59,700 --> 01:11:03,180
And here's an instance
of Buriram coming off

1256
01:11:03,180 --> 01:11:08,040
the financial crisis,
which was 1997-98,

1257
01:11:08,040 --> 01:11:11,640
but anyway, they actually ran
a balance of payments deficit.

1258
01:11:16,280 --> 01:11:20,860
So this opens up the whole
vista of international economics

1259
01:11:20,860 --> 01:11:23,200
and what's going on
using those kinds of ways

1260
01:11:23,200 --> 01:11:25,660
of conceptualizing the problem.

1261
01:11:29,030 --> 01:11:33,980
And just to remind you, because
I promised an application

1262
01:11:33,980 --> 01:11:37,520
for each section, we've
had tons of models

1263
01:11:37,520 --> 01:11:43,160
of the role of financing
constraints or other obstacles

1264
01:11:43,160 --> 01:11:46,980
in growth and TFP and
so on and so forth.

1265
01:11:46,980 --> 01:11:56,310
So here's just some model with
arch where we do the following.

1266
01:11:56,310 --> 01:12:00,070
We take these villages to
be small open economies,

1267
01:12:00,070 --> 01:12:03,010
we see a drop in
the interest rate--

1268
01:12:03,010 --> 01:12:06,470
it's actually true worldwide--

1269
01:12:06,470 --> 01:12:10,940
and we see the
number of new loans

1270
01:12:10,940 --> 01:12:15,590
increasing this has to do in
part with this village fund

1271
01:12:15,590 --> 01:12:19,430
stuff that we'll talk about
later the Million Baht

1272
01:12:19,430 --> 01:12:20,240
Fund in Thailand.

1273
01:12:23,050 --> 01:12:26,750
This is the collateral
constraint--

1274
01:12:26,750 --> 01:12:29,500
this is a model thing.

1275
01:12:29,500 --> 01:12:34,110
So we're going to act as if that
innovation associated with more

1276
01:12:34,110 --> 01:12:37,590
loans was like a weakening
of the borrowing constraint

1277
01:12:37,590 --> 01:12:46,270
and feed it into
a model, and it's

1278
01:12:46,270 --> 01:12:51,680
a trade model with
financial underpinnings,

1279
01:12:51,680 --> 01:12:55,940
and it predicts things like the
relative price of agriculture

1280
01:12:55,940 --> 01:12:59,340
relative to manufactured goods.

1281
01:12:59,340 --> 01:13:03,240
In the model and in the
data, we're kind of close.

1282
01:13:03,240 --> 01:13:09,120
And a real goal was to try
to get the wages right,

1283
01:13:09,120 --> 01:13:13,785
and here is the actual and
model-driven real wages.

1284
01:13:16,540 --> 01:13:17,870
Why is this a struggle?

1285
01:13:17,870 --> 01:13:21,030
If you think about
Heckscher-Ohlin,

1286
01:13:21,030 --> 01:13:24,600
that has to do with
the area around Bangkok

1287
01:13:24,600 --> 01:13:29,070
being very capital-intensive
and not having so much labor,

1288
01:13:29,070 --> 01:13:33,120
so interest rates are
lower and wages are higher;

1289
01:13:33,120 --> 01:13:36,290
in the northeast, it's the
other way around because capital

1290
01:13:36,290 --> 01:13:38,500
and wealth is relatively low.

1291
01:13:38,500 --> 01:13:40,110
So you think that
over time you're

1292
01:13:40,110 --> 01:13:42,750
going to get
convergence, which means

1293
01:13:42,750 --> 01:13:46,710
wages ought to be dropping
in the central area,

1294
01:13:46,710 --> 01:13:49,120
and they're not.

1295
01:13:49,120 --> 01:13:53,070
So then you introduce
this financial constraint,

1296
01:13:53,070 --> 01:13:55,350
and you get the
dynamics that have

1297
01:13:55,350 --> 01:13:57,550
to do with choosing
occupations and so on.

1298
01:13:57,550 --> 01:14:01,170
So that hopefully you remember
enough from the other papers

1299
01:14:01,170 --> 01:14:06,030
that we covered to kind of
imagine how one would fill this

1300
01:14:06,030 --> 01:14:09,570
out, and we're actually at
these calibrated values--

1301
01:14:09,570 --> 01:14:12,720
we're actually
matching quite well

1302
01:14:12,720 --> 01:14:20,050
all the income shares for wages,
manufacturing, and agriculture.

1303
01:14:20,050 --> 01:14:22,230
Yep?

1304
01:14:22,230 --> 01:14:25,700
AUDIENCE: I might
remember wrong, but just--

1305
01:14:25,700 --> 01:14:29,520
yeah, it's very likely,
but in your paper

1306
01:14:29,520 --> 01:14:32,157
with Kaboski on the Million--

1307
01:14:32,157 --> 01:14:33,490
ROBERT TOWNSEND: Baht Fund, yes.

1308
01:14:36,050 --> 01:14:38,180
AUDIENCE: Is it
the interest rate

1309
01:14:38,180 --> 01:14:44,510
wasn't going to be with
the intervention, was it?

1310
01:14:44,510 --> 01:14:47,160
Now you just said that
the interest rate drop has

1311
01:14:47,160 --> 01:14:49,410
to do with the Million Baht--

1312
01:14:49,410 --> 01:14:51,460
ROBERT TOWNSEND:
Remember, what we're doing

1313
01:14:51,460 --> 01:14:53,760
is kind of a variation
in the cross-section

1314
01:14:53,760 --> 01:14:57,610
as a function of the percentage
of households in the village.

1315
01:14:57,610 --> 01:15:02,510
So the per-capita
treatment is we're

1316
01:15:02,510 --> 01:15:05,310
seeing whether that has any
impact on the interest rate.

1317
01:15:05,310 --> 01:15:08,230
What I showed you was the time
variation in the interest rate,

1318
01:15:08,230 --> 01:15:09,660
so it's a very different object.

1319
01:15:09,660 --> 01:15:10,310
AUDIENCE: OK.

1320
01:15:10,310 --> 01:15:11,227
ROBERT TOWNSEND: Yeah.

1321
01:15:11,227 --> 01:15:13,830
So that's kind of like
differenced out, basically.

1322
01:15:13,830 --> 01:15:18,095
It's the relative
difference from the trend

1323
01:15:18,095 --> 01:15:19,470
and whether the
Million Baht Fund

1324
01:15:19,470 --> 01:15:20,637
had something to do with it.

1325
01:15:23,260 --> 01:15:23,760
OK.

1326
01:15:23,760 --> 01:15:27,240
The last thing is flow of funds.

1327
01:15:27,240 --> 01:15:30,270
These are the flow of funds
accounts for Thailand,

1328
01:15:30,270 --> 01:15:31,080
basically.

1329
01:15:34,490 --> 01:15:37,180
A quick introduction
to flow of funds,

1330
01:15:37,180 --> 01:15:39,240
you've got non-financial
corporations,

1331
01:15:39,240 --> 01:15:43,650
financial corporations,
government, households,

1332
01:15:43,650 --> 01:15:47,620
and not-for-profit corporations,
and the rest of the world.

1333
01:15:47,620 --> 01:15:49,320
Those are the sectors.

1334
01:15:49,320 --> 01:15:53,640
And then you have
acquisition of assets,

1335
01:15:53,640 --> 01:15:57,870
and they're actually in
detailed financial accounts.

1336
01:15:57,870 --> 01:16:00,370
When the flow of funds
accounts-- when they exist,

1337
01:16:00,370 --> 01:16:03,330
you've got money and
gold and currency,

1338
01:16:03,330 --> 01:16:06,130
securities, loans, and so on.

1339
01:16:06,130 --> 01:16:10,500
So acquiring a loan means
basically you're a lender.

1340
01:16:10,500 --> 01:16:13,800
On the other hand, you've
got sort of liabilities

1341
01:16:13,800 --> 01:16:16,890
down here, and so that you can
have a loan which represents

1342
01:16:16,890 --> 01:16:19,140
an increase in borrowing.

1343
01:16:19,140 --> 01:16:22,560
It takes a lot of concentration
to constantly remember

1344
01:16:22,560 --> 01:16:25,500
uses means putting money--

1345
01:16:25,500 --> 01:16:27,000
so the idea is you run--

1346
01:16:29,530 --> 01:16:34,630
you have positive savings,
where do you put your money?

1347
01:16:34,630 --> 01:16:36,640
And you could put it
here or you could reduce

1348
01:16:36,640 --> 01:16:39,700
your deficits, or if you--

1349
01:16:39,700 --> 01:16:40,720
reduce your liabilities.

1350
01:16:40,720 --> 01:16:42,362
If you run it the
other way and you

1351
01:16:42,362 --> 01:16:43,820
have negative
savings or something,

1352
01:16:43,820 --> 01:16:45,820
and then you've got to
increase your liabilities

1353
01:16:45,820 --> 01:16:49,880
or run down and sell off
some assets that you own.

1354
01:16:49,880 --> 01:16:52,980
These are the
traditional labor labels

1355
01:16:52,980 --> 01:16:55,075
in the flow of funds by sector.

1356
01:16:58,150 --> 01:16:59,980
Is there a relationship
between flow

1357
01:16:59,980 --> 01:17:02,080
of funds and the
national income accounts?

1358
01:17:02,080 --> 01:17:03,250
Yes.

1359
01:17:03,250 --> 01:17:06,460
Actually, in principle,
they should all

1360
01:17:06,460 --> 01:17:09,280
be consistent with one another.

1361
01:17:09,280 --> 01:17:11,740
But they're often not.

1362
01:17:11,740 --> 01:17:15,010
The thing that is most
consistent is flow of funds

1363
01:17:15,010 --> 01:17:18,010
into and out of the
financial institutions

1364
01:17:18,010 --> 01:17:21,880
because commercial banks and
other financial corporations

1365
01:17:21,880 --> 01:17:24,370
are required to
report to regulators,

1366
01:17:24,370 --> 01:17:26,990
and the central bank
keeps track of things.

1367
01:17:26,990 --> 01:17:31,480
So we see most accurately
deposits coming in,

1368
01:17:31,480 --> 01:17:34,720
loans going out for
financial corporations.

1369
01:17:34,720 --> 01:17:38,230
We often see little or
anything of the transactions

1370
01:17:38,230 --> 01:17:41,170
between the other sectors.

1371
01:17:41,170 --> 01:17:42,970
There are national
income and product

1372
01:17:42,970 --> 01:17:48,880
account for those sectors,
and we see, therefore, savings

1373
01:17:48,880 --> 01:17:52,660
for those, but those
savings numbers

1374
01:17:52,660 --> 01:17:56,470
are often radically
different from the savings

1375
01:17:56,470 --> 01:18:00,640
we would impute from
just seeing transactions

1376
01:18:00,640 --> 01:18:03,190
with the measured
formal financial sector.

1377
01:18:03,190 --> 01:18:05,620
Another way to say
this, if you read Deaton

1378
01:18:05,620 --> 01:18:12,100
and so on, the savings you see
coming from household surveys

1379
01:18:12,100 --> 01:18:14,860
gives you very, very different
numbers than the savings

1380
01:18:14,860 --> 01:18:18,310
you'll see at the level of
the national income accounts,

1381
01:18:18,310 --> 01:18:21,220
and that's because the
measurement is missing,

1382
01:18:21,220 --> 01:18:26,030
so households get treated as the
residual in the national income

1383
01:18:26,030 --> 01:18:26,530
accounts.

1384
01:18:30,050 --> 01:18:35,990
But flow of funds data do
exist, and to some extent,

1385
01:18:35,990 --> 01:18:40,820
they can be created, and
Thailand is, for example--

1386
01:18:40,820 --> 01:18:46,310
and Mexico-- creating improved
financial flow of funds

1387
01:18:46,310 --> 01:18:48,080
accounts.

1388
01:18:48,080 --> 01:18:50,210
I'd like to think we had
something to do with that.

1389
01:18:54,170 --> 01:18:55,880
We can also do it at
the village level.

1390
01:18:55,880 --> 01:19:01,160
I've shown you a lot, but
let me just jump here.

1391
01:19:04,700 --> 01:19:06,860
At the national
aggregate level, these

1392
01:19:06,860 --> 01:19:10,220
are the traditional sectors,
and here's a diamond,

1393
01:19:10,220 --> 01:19:15,890
and basically you can see
from the standpoint of banks,

1394
01:19:15,890 --> 01:19:19,990
households are
interior-- basically

1395
01:19:19,990 --> 01:19:22,010
households are
investing in banks,

1396
01:19:22,010 --> 01:19:24,400
but that's a liability
for the banking sector.

1397
01:19:24,400 --> 01:19:29,300
Savings accounts
are to be paid back.

1398
01:19:29,300 --> 01:19:32,620
On the other hand, in this
period between '08 and--

1399
01:19:32,620 --> 01:19:37,090
'04 and '08, you know the
banking sector was investing

1400
01:19:37,090 --> 01:19:39,610
in the rest of the world and
that had to do with Thailand

1401
01:19:39,610 --> 01:19:41,965
running a balance
of payments surplus.

1402
01:19:44,660 --> 01:19:47,160
But this is at
the village level.

1403
01:19:47,160 --> 01:19:51,200
And here, you can distinguish
financial corporations,

1404
01:19:51,200 --> 01:19:54,680
transactions with businesses
outside the village,

1405
01:19:54,680 --> 01:19:57,230
transactions of the village
funds we just talked about,

1406
01:19:57,230 --> 01:20:01,310
and you can see them using
currency, running surpluses

1407
01:20:01,310 --> 01:20:04,460
or deficits with these
other entities in terms

1408
01:20:04,460 --> 01:20:06,020
of the objects.

1409
01:20:06,020 --> 01:20:10,010
So the point here is that
it's a common way of measuring

1410
01:20:10,010 --> 01:20:14,120
and it doesn't just have to be
applied as it is conventionally

1411
01:20:14,120 --> 01:20:15,650
at the national level.

1412
01:20:15,650 --> 01:20:19,430
Flow of funds can zoom
in and zoom out by region

1413
01:20:19,430 --> 01:20:23,040
and tell you which object's
moving around over time.

1414
01:20:23,040 --> 01:20:32,240
And in fact, you can read
it in a little more leisure

1415
01:20:32,240 --> 01:20:37,460
style on your own, there's
papers of Guerrieri and Guido

1416
01:20:37,460 --> 01:20:42,470
Lorenzoni and Monika Piazzesi
and Martin Schneider,

1417
01:20:42,470 --> 01:20:46,160
and these are,
granted, about the US,

1418
01:20:46,160 --> 01:20:49,910
but it's basically looking
at the household sector

1419
01:20:49,910 --> 01:20:52,100
and going to the flow
of funds data in the US

1420
01:20:52,100 --> 01:20:58,550
to measure changes in
savings and borrowing.

1421
01:20:58,550 --> 01:21:01,400
And this same
language that they use

1422
01:21:01,400 --> 01:21:03,890
to try to understand
what happens

1423
01:21:03,890 --> 01:21:07,310
in a financial crisis
when all of a sudden you

1424
01:21:07,310 --> 01:21:11,690
can't borrow as much
because the banks are tight,

1425
01:21:11,690 --> 01:21:15,300
or in Monika's case, what
happens when there are many,

1426
01:21:15,300 --> 01:21:19,080
many more young people
entering and they have

1427
01:21:19,080 --> 01:21:21,860
certain expectations
about inflation,

1428
01:21:21,860 --> 01:21:25,080
and each one of these
pairs of authors

1429
01:21:25,080 --> 01:21:29,460
is doing this kind of experiment
in the context of a model.

1430
01:21:29,460 --> 01:21:32,430
So these very same
models that rely

1431
01:21:32,430 --> 01:21:35,370
on a sort of macro-aggregate
looking, though,

1432
01:21:35,370 --> 01:21:40,500
at sectors in the US also
apply to developing countries.

1433
01:21:40,500 --> 01:21:44,250
In fact Guido's model is eerily
similar, although none of us

1434
01:21:44,250 --> 01:21:47,610
recognized it at the time, to
the model that Joe and I used

1435
01:21:47,610 --> 01:21:52,730
to analyze the village
fund in in Thailand.