Labor scarcity, technology adoption and innovation

Key Question Being Explored in This Insight:

How does labor scarcity, driven by cholera deaths linked to summer temperatures, affect short-run technology adoption and innovation in agriculture versus industry?

View the page of: Raphaël Franck

Institution: University of Stuttgart

Explore more insights in:

Economics Labor Economics Social Sciences

Media Literacy Toolkit

Faculti's Media Literacy Toolkit helps viewers critically engage with academic insights by analyzing the research context, identifying perspectives, and encouraging thoughtful evaluation.

Critical Questions to Consider ▼

What assumptions does the research make?
Are there alternative perspectives not explored?
What are the limitations of the research method?

Bias and Perspective Awareness ▼

This research comes from University of Stuttgart . Reflect on how the institution's academic focus and research partnerships may shape the questions being explored.

Further Reading ▼

Read the Study

How does labor scarcity, driven by cholera deaths linked to summer temperatures, affect short-run technology adoption and innovation in agriculture versus industry?

Image courtesy of interviewee. March 20, 2025

Log-in or Sign-up to Faculti
Currently viewing this subject insight as a guest. You have insight(s) remaining for this month. Login to view 8000+ figures on the platform.

Generating Workspace...

Export as PDF

Language:

Export as PDF

X-TIMESTAMP-MAP=LOCAL:00:00:00.000,MPEGTS:90000

1
00:01:13.975 --> 00:01:14.705
Okay, Rafael.

2
00:01:14.725 --> 00:01:17.305
So it's the, it's the 14th of October already.

3
00:01:17.415 --> 00:01:20.185
It's 1230, here in the, in the United Kingdom.

4
00:01:20.215 --> 00:01:21.305
Talk about labor scarcity,

5
00:01:21.305 --> 00:01:23.105
technology, adoption, and innovation.

6
00:01:23.105 --> 00:01:24.585
Thanks for taking the time to speak to us.

7
00:01:24.925 --> 00:01:26.905
Uh, well, what was the context of your work?

8
00:01:28.025 --> 00:01:31.485
So the study starts from a very straightforward question,

9
00:01:31.485 --> 00:01:33.285
which is what is the relationship

10
00:01:33.285 --> 00:01:35.445
between labor and technology?

11
00:01:36.305 --> 00:01:38.445
And for the sake of the argument, let's assume

12
00:01:38.445 --> 00:01:39.885
that labor is scarce.

13
00:01:40.585 --> 00:01:43.605
So your reaction will be, well, this is trivial.

14
00:01:44.025 --> 00:01:47.125
If labor is scarce, wages will increase.

15
00:01:47.275 --> 00:01:49.245
This will lead to more technology adoption.

16
00:01:50.505 --> 00:01:51.605
But that reasoning

17
00:01:51.605 --> 00:01:54.565
and plus the factors of production are substitutes.

18
00:01:55.105 --> 00:01:58.125
And then it immediately calls the other question, which is,

19
00:01:58.495 --> 00:02:00.005
given the level of technology,

20
00:02:00.055 --> 00:02:02.525
which we enjoy in modern societies nowadays,

21
00:02:03.065 --> 00:02:05.285
why don't we get 85% of unemployment?

22
00:02:06.295 --> 00:02:09.515
And the answer is, you need people to operate machines.

23
00:02:10.095 --> 00:02:11.155
In other words, the factors

24
00:02:11.215 --> 00:02:13.115
of production can be complimentary.

25
00:02:13.495 --> 00:02:15.475
And if indeed labor-based cares

26
00:02:15.535 --> 00:02:17.635
and wages are high under complimentary factors

27
00:02:17.735 --> 00:02:20.515
of production, this will actually lead to

28
00:02:21.145 --> 00:02:22.475
less technology adoption

29
00:02:22.475 --> 00:02:25.235
because it's not profitable to invest in new machines.

30
00:02:26.215 --> 00:02:28.965
So that's what the study is about.

31
00:02:28.985 --> 00:02:32.165
And the study starts from this natural experiment, which

32
00:02:32.765 --> 00:02:37.285
I will argue is quite convincing the labor pandemics in

33
00:02:37.285 --> 00:02:38.325
19th century France.

34
00:02:38.705 --> 00:02:40.525
So when labor is scarce, what do you think happens?

35
00:02:41.145 --> 00:02:43.485
Do you think the factors of production are complimentary

36
00:02:43.705 --> 00:02:45.525
or substitute in agriculture?

37
00:02:46.145 --> 00:02:50.215
Do you think that factors of production are complimentary,

38
00:02:50.305 --> 00:02:51.655
substitute in industry

39
00:02:52.855 --> 00:02:56.995
and beyond that, do you think that labor scarcity say,

40
00:02:57.455 --> 00:03:01.165
has a differential impact on other aspects

41
00:03:01.305 --> 00:03:03.465
of economic growth?

42
00:03:04.855 --> 00:03:07.965
Given that maybe the factors of production are complimentary

43
00:03:07.985 --> 00:03:10.285
or substitute in agriculture or industry?

44
00:03:10.785 --> 00:03:13.405
Do you think that there will be more human capital

45
00:03:13.555 --> 00:03:16.605
formation, more investments in literacy?

46
00:03:16.605 --> 00:03:18.725
Because we talk about the 19th century

47
00:03:20.335 --> 00:03:23.345
that there will be more investments in literacy in the

48
00:03:25.435 --> 00:03:26.735
sector where factors

49
00:03:26.755 --> 00:03:29.175
of production are complimentary or substitute.

50
00:03:30.895 --> 00:03:33.395
And so that's not that straightforward.

51
00:03:35.155 --> 00:03:36.165
It's not that straightforward.

52
00:03:36.585 --> 00:03:40.875
And so obviously the paper starts from a

53
00:03:41.505 --> 00:03:45.115
long literature on the impact of diseases,

54
00:03:46.135 --> 00:03:47.875
on economic growth and the environment.

55
00:03:47.875 --> 00:03:50.795
And of course, there's this literature arguing

56
00:03:50.905 --> 00:03:54.635
that the Black death was the reason why

57
00:03:55.605 --> 00:04:00.185
the western world grew, because there was less labor, and

58
00:04:00.185 --> 00:04:04.625
therefore there was more incentives to invest in machines.

59
00:04:04.625 --> 00:04:06.465
But also there was the end of Cdom.

60
00:04:07.085 --> 00:04:10.025
But here we are looking at different context, which is,

61
00:04:10.475 --> 00:04:12.345
we're looking at France in the 19th century,

62
00:04:12.485 --> 00:04:14.305
and in other words, we're looking at a country which is

63
00:04:14.715 --> 00:04:16.265
outta the tion trap.

64
00:04:17.375 --> 00:04:20.195
So what happens when you have the black death in the Middle

65
00:04:20.305 --> 00:04:25.195
Ages is that we live in a Malian world, world where people,

66
00:04:25.865 --> 00:04:28.715
when they get richer, they will have more children.

67
00:04:30.865 --> 00:04:32.685
But here it's different

68
00:04:32.685 --> 00:04:35.005
because you are out of the Malian trap

69
00:04:35.385 --> 00:04:38.805
and there's the disconnects between wealth

70
00:04:39.385 --> 00:04:40.485
and the number of children.

71
00:04:42.375 --> 00:04:46.995
And so the paper is also very relevant

72
00:04:47.215 --> 00:04:49.075
if there's any policy relevant,

73
00:04:49.345 --> 00:04:53.555
because we've been through a pandemic,

74
00:04:54.235 --> 00:04:58.595
recently, and that pandemic killed, fortunately killed,

75
00:04:59.295 --> 00:05:02.235
upon much fewer people.

76
00:05:03.605 --> 00:05:06.105
That's, if I can share my screen,

77
00:05:06.205 --> 00:05:07.505
I'm going to share my screen.

78
00:05:07.815 --> 00:05:09.385
Yeah. Brilliant, brilliant. So far.

79
00:05:09.405 --> 00:05:13.035
By the way, I might turn my camera off when you

80
00:05:13.035 --> 00:05:15.235
share, just so I'm not in it.

81
00:05:15.375 --> 00:05:16.375
I'm still here.

82
00:05:27.065 --> 00:05:28.885
So this is France, right?

83
00:05:29.145 --> 00:05:32.995
And so what's interesting about France is

84
00:05:33.065 --> 00:05:37.675
that France was divided in 1719 into administrative areas

85
00:05:38.655 --> 00:05:41.875
of, called apartments.

86
00:05:41.935 --> 00:05:45.515
And those administrative areas were actually,

87
00:05:47.165 --> 00:05:49.385
meant to be of small size, right?

88
00:05:49.525 --> 00:05:52.945
So on average, they are six seven square kilometers.

89
00:05:53.965 --> 00:05:57.485
And France is, you know, we're not going

90
00:05:57.485 --> 00:05:59.205
to say a large country, but still,

91
00:06:00.105 --> 00:06:01.555
basically the size of Texas.

92
00:06:02.895 --> 00:06:05.275
And what's interesting is that France,

93
00:06:06.625 --> 00:06:11.495
which had about 30 million people in the 1380s, so lost

94
00:06:12.445 --> 00:06:16.465
1% of its population over 22 years between the first

95
00:06:17.205 --> 00:06:20.425
and the last, epidemics of cholera, which,

96
00:06:20.605 --> 00:06:21.985
is analyzed in this study.

97
00:06:22.485 --> 00:06:25.265
And so obviously what is interesting is not only the timing,

98
00:06:25.405 --> 00:06:29.825
but also the differential impact in some areas of France.

99
00:06:29.825 --> 00:06:32.825
And some departments, no one were,

100
00:06:32.925 --> 00:06:35.425
no one died in other areas of France.

101
00:06:36.055 --> 00:06:37.065
Many people died.

102
00:06:37.415 --> 00:06:39.815
Even, in,

103
00:06:40.725 --> 00:06:44.185
in the southwestern department of Esh, 4.2%

104
00:06:44.185 --> 00:06:45.345
of the population died.

105
00:06:45.645 --> 00:06:47.305
And when I say died, died in one summer.

106
00:06:48.515 --> 00:06:49.865
We'll get back to that in a moment.

107
00:06:50.205 --> 00:06:53.935
So just a few words on the cholera epidemics.

108
00:06:54.715 --> 00:06:59.625
Uh, it looks like already in India, the

109
00:07:00.885 --> 00:07:02.695
cholera bacterium was known,

110
00:07:03.435 --> 00:07:07.375
and it appeared in a western world in the 19th century.

111
00:07:08.445 --> 00:07:10.545
And it was only in, 1854.

112
00:07:10.565 --> 00:07:12.345
In 1854,

113
00:07:12.765 --> 00:07:15.945
Felipo Pacini was able net Italian biologist was able

114
00:07:15.965 --> 00:07:19.785
to isolate the bacterium, but it was only an 1855.

115
00:07:19.855 --> 00:07:23.825
Then John Snow, was able to prove

116
00:07:24.375 --> 00:07:26.065
that the cholera bacter,

117
00:07:26.165 --> 00:07:28.265
the cholera was spread through waters.

118
00:07:28.605 --> 00:07:31.425
And before, people thought

119
00:07:31.425 --> 00:07:33.825
that maybe the color was transmitted through air

120
00:07:34.055 --> 00:07:35.745
because they were miasma.

121
00:07:36.165 --> 00:07:39.765
And one of the reason for that is that the cholera

122
00:07:40.315 --> 00:07:43.325
bacterium can survive six to seven weeks on dry clothes,

123
00:07:43.335 --> 00:07:45.855
which were damped, and then at the,

124
00:07:45.855 --> 00:07:47.295
and which were damp and sweaty.

125
00:07:47.475 --> 00:07:50.895
And so very often emotive transmission of the cholera was

126
00:07:50.955 --> 00:07:54.775
during funerals because people would touch the dead

127
00:07:54.795 --> 00:07:56.335
and touch the body of the dead,

128
00:07:56.715 --> 00:07:58.135
and during those funeral wakes,

129
00:07:58.135 --> 00:08:02.345
and then it would also be infected with the chole bacterium,

130
00:08:02.965 --> 00:08:04.225
and people would say, it's

131
00:08:04.225 --> 00:08:06.465
because there's something in the air which is polluted.

132
00:08:06.805 --> 00:08:08.585
It was only in 1884 that,

133
00:08:09.285 --> 00:08:10.985
an effective treatment was provided.

134
00:08:12.045 --> 00:08:14.425
And clearly in France,

135
00:08:14.515 --> 00:08:18.555
where the weather is only,

136
00:08:19.335 --> 00:08:21.995
wet and humid during the summer, it's mostly

137
00:08:21.995 --> 00:08:24.675
during the summer, then the collateral is transmitted.

138
00:08:24.895 --> 00:08:29.795
And most people died from, the, the disease.

139
00:08:30.815 --> 00:08:34.955
And we're back to something which is that no one knew

140
00:08:35.255 --> 00:08:38.315
before 18 54, 18 55.

141
00:08:38.575 --> 00:08:41.075
And that's why I, the study only looks

142
00:08:41.855 --> 00:08:45.595
at the pandemics in 18 32, 18 49, and 1854.

143
00:08:46.655 --> 00:08:50.475
No one knew how the disease spread,

144
00:08:50.815 --> 00:08:52.715
and no one knew how to stop it.

145
00:08:53.735 --> 00:08:57.365
And that's why, and then that's why in French nowadays,

146
00:08:57.795 --> 00:09:01.885
there's an expression called bl blue, literally blue fear,

147
00:09:02.425 --> 00:09:04.365
which refers to an extreme fear.

148
00:09:04.425 --> 00:09:07.245
And that actually referred to first to the cholera,

149
00:09:07.355 --> 00:09:09.685
because when people were about to die,

150
00:09:09.685 --> 00:09:13.235
because they were dehydrated, they,

151
00:09:13.285 --> 00:09:15.355
their face would take a blue coloration.

152
00:09:16.685 --> 00:09:18.105
And so that's the consequence

153
00:09:18.105 --> 00:09:20.705
of the color pandemics in the French language to this day.

154
00:09:21.975 --> 00:09:22.435
You good?

155
00:09:27.615 --> 00:09:30.335
Uh, okay. So

156
00:09:30.685 --> 00:09:32.215
Yeah, sorry, Joe, I'll keep jumping in.

157
00:09:32.215 --> 00:09:33.535
But you are about to do your methods, aren't you?

158
00:09:34.775 --> 00:09:38.175
I'm sorry, I keep jumping in ffa. Sorry.

159
00:09:38.195 --> 00:09:39.415
But you are about to do your methods,

160
00:09:39.415 --> 00:09:40.695
aren't you? Your research design?

161
00:09:41.475 --> 00:09:44.675
Yes. So the, the methods,

162
00:09:47.265 --> 00:09:49.365
I was going to talk about the methods,

163
00:09:50.465 --> 00:09:51.805
but maybe I should leave the methods

164
00:09:51.865 --> 00:09:53.125
for the rest of the talk.

165
00:09:53.435 --> 00:09:55.925
Yeah, I was gonna say, because yeah, I mean,

166
00:09:55.985 --> 00:09:57.045
the way I see it is that you,

167
00:09:57.045 --> 00:09:59.325
your next slide is the methodology, isn't it?

168
00:09:59.325 --> 00:10:02.645
But I, I feel like, the design, the,

169
00:10:02.905 --> 00:10:04.685
and the data set is important as well.

170
00:10:04.745 --> 00:10:06.405
Or did you wanna do a bit more in literature?

171
00:10:07.965 --> 00:10:09.265
No, I think that's fine.

172
00:10:09.465 --> 00:10:13.655
I think that, we could do, I don't,

173
00:10:14.395 --> 00:10:17.135
I assume this will be edited. Yeah. Oh,

174
00:10:17.165 --> 00:10:18.165
This is edited.

175
00:10:18.305 --> 00:10:20.175
Don't worry. It's actually really been

176
00:10:20.175 --> 00:10:21.975
really very well delivered. But yeah, we can

177
00:10:21.975 --> 00:10:22.975
Chat. The question is, do

178
00:10:22.975 --> 00:10:25.255
you wanna, so that's, that's the moment

179
00:10:25.265 --> 00:10:29.785
where in a, an economics presentation, this is the moment

180
00:10:29.835 --> 00:10:31.865
where I would talk about the results.

181
00:10:32.525 --> 00:10:33.985
Oh, okay. The results.

182
00:10:34.445 --> 00:10:35.985
That's fine. I mean, that is the case,

183
00:10:36.205 --> 00:10:39.065
And I don't know whether that's something that

184
00:10:39.775 --> 00:10:41.305
Happens, Is really relevant, what happens.

185
00:10:41.375 --> 00:10:43.345
Yeah. What happens for us is that we've got this sort

186
00:10:43.345 --> 00:10:45.905
of structure where we go sort of chronologically context,

187
00:10:46.745 --> 00:10:49.665
literature, research, findings, conclusions.

188
00:10:49.665 --> 00:10:52.585
But every now and then, maybe one in 20 economists will say,

189
00:10:52.805 --> 00:10:53.945
can we change the order?

190
00:10:54.285 --> 00:10:57.625
And can we go context in empirical results lit,

191
00:10:57.725 --> 00:10:58.745
or not even literature set.

192
00:10:58.745 --> 00:11:00.665
Third, that's often at the end contributions.

193
00:11:00.665 --> 00:11:02.385
And it's interesting, the economists, I think, want

194
00:11:02.385 --> 00:11:04.105
to deliver it the way they do in a presentation,

195
00:11:04.405 --> 00:11:06.625
but we, we always accommodate. So what's best for

196
00:11:06.625 --> 00:11:07.625
You? Uh, so I

197
00:11:07.625 --> 00:11:10.265
think it's more like game theorists who do that,

198
00:11:10.725 --> 00:11:13.545
okay.

199
00:11:13.725 --> 00:11:18.305
So that, let's, let's reju in with, then the,

200
00:11:18.305 --> 00:11:20.305
the data, and then we can leave the,

201
00:11:20.365 --> 00:11:21.465
the results for lighter.

202
00:11:21.815 --> 00:11:24.945
Yeah, no problem. So I'm just gonna turn my video off,

203
00:11:24.965 --> 00:11:26.425
and then, yeah, so let's go data.

204
00:11:27.345 --> 00:11:30.025
I just, one thing, I,

205
00:11:30.145 --> 00:11:31.385
I think I missed your first name.

206
00:11:32.205 --> 00:11:33.705
Sam. Sam. Sam.

207
00:11:34.695 --> 00:11:37.495
I think actually I'd like you, I'd like to you to, I,

208
00:11:37.615 --> 00:11:39.575
I would like to see you if there's a way of, of

209
00:11:39.575 --> 00:11:40.575
Course. No, of course, of course. In

210
00:11:40.575 --> 00:11:41.255
all interviews,

211
00:11:41.535 --> 00:11:44.535
I, I, I find that In all interviews, I'm on camera.

212
00:11:44.805 --> 00:11:46.255
Zoom was mentally exhausting.

213
00:11:46.615 --> 00:11:49.135
Teaching on Zoom was mentally exhausting. Yeah.

214
00:11:49.635 --> 00:11:52.695
The only issue that fa is on the is, so I'm happy

215
00:11:52.695 --> 00:11:54.485
to be on camera throughout, I'm always on camera on every

216
00:11:54.485 --> 00:11:57.885
interview, but it's with these, share screen things,

217
00:11:58.395 --> 00:12:00.445
it's in those moments, I have to come off camera

218
00:12:00.445 --> 00:12:02.405
because they will, it will capture my face,

219
00:12:03.105 --> 00:12:04.845
and then the editors, they can't have me.

220
00:12:04.845 --> 00:12:07.205
Whereas I think when I, turn my camera off,

221
00:12:07.705 --> 00:12:10.645
it just captures your face and the screen.

222
00:12:10.945 --> 00:12:12.405
But in these normal interviews, of course,

223
00:12:12.545 --> 00:12:13.565
I'm, you know, more than

224
00:12:13.565 --> 00:12:14.565
Happy. Okay. Okay. Okay.

225
00:12:14.565 --> 00:12:17.445
No, no. Uh, okay. Okay. I get that. Okay.

226
00:12:17.465 --> 00:12:21.405
So I'll do without it. Okay, no worries. Okay.

227
00:12:21.625 --> 00:12:26.445
So, let me, okay, let's get back to it.

228
00:12:26.585 --> 00:12:30.245
Let me reshare and then we can restart

229
00:12:30.465 --> 00:12:31.645
as if nothing happened.

230
00:12:35.525 --> 00:12:37.665
So let's look at the, the maps of France

231
00:12:38.485 --> 00:12:40.945
or the share of color of death in each department.

232
00:12:40.965 --> 00:12:44.385
And as you can see, there was a differential timing.

233
00:12:45.405 --> 00:12:49.145
Uh, it's only in 1830 to 18 49, 18 54 that we get

234
00:12:50.035 --> 00:12:51.125
cholera pandemics.

235
00:12:52.185 --> 00:12:56.565
And the reason that, we see it usually,

236
00:12:57.545 --> 00:12:59.625
you see that very often,

237
00:13:01.715 --> 00:13:04.615
the most death altered in the north of France.

238
00:13:05.635 --> 00:13:08.535
And we'll get back to that, later

239
00:13:08.535 --> 00:13:11.895
because it'll be useful for the empirical, methodology.

240
00:13:12.675 --> 00:13:15.975
So there's a reason for that is that very often,

241
00:13:16.555 --> 00:13:20.285
or very often, what's usually thought is

242
00:13:20.285 --> 00:13:24.005
that the RA came from the UK by boat,

243
00:13:24.705 --> 00:13:27.125
and then from the north it spread to Paris.

244
00:13:27.225 --> 00:13:29.205
And from Paris, it spread to the rest of France.

245
00:13:30.065 --> 00:13:33.325
But aside from 1854, where

246
00:13:35.685 --> 00:13:37.935
Cero is mostly transmitted, we think

247
00:13:37.935 --> 00:13:39.975
that the coral was transmitted by the soldiers

248
00:13:40.805 --> 00:13:43.415
embarking from the harbors of Tulo in Marse

249
00:13:43.755 --> 00:13:45.255
to fight the war in Crimea.

250
00:13:45.965 --> 00:13:50.335
Most of the time, the pandemics remained in the north,

251
00:13:51.155 --> 00:13:52.775
the east, and the west of France.

252
00:13:53.625 --> 00:13:57.775
Right? And it has to do with the climate of France.

253
00:13:57.835 --> 00:14:00.335
And I will get back to a link between humidity

254
00:14:01.605 --> 00:14:06.305
and, the weather and, and, hot temperatures.

255
00:14:06.405 --> 00:14:10.345
So what we, what I do in the, in this study is just look at

256
00:14:10.695 --> 00:14:11.945
what happens in the summer.

257
00:14:12.165 --> 00:14:14.985
But it's interesting to look at the progression

258
00:14:15.085 --> 00:14:18.945
of calera in 1854 throughout the year, the only year

259
00:14:18.945 --> 00:14:20.985
where we have month by month data.

260
00:14:21.085 --> 00:14:25.505
And as you can see, very few people die in the winter.

261
00:14:26.845 --> 00:14:28.945
And we do see a little rise

262
00:14:30.055 --> 00:14:32.035
in June when the weather gets warmer,

263
00:14:33.615 --> 00:14:35.435
and then we see that more people die in

264
00:14:35.465 --> 00:14:36.635
July, August, September.

265
00:14:37.395 --> 00:14:42.225
And once again, it dies out in,

266
00:14:42.225 --> 00:14:43.625
October, November, December.

267
00:14:44.605 --> 00:14:47.305
So what's interesting is obviously

268
00:14:47.305 --> 00:14:51.365
that there's a huge differential in the share of people

269
00:14:51.465 --> 00:14:55.845
who died in every department across those three years.

270
00:14:56.465 --> 00:14:59.605
So on average, it's one 31%,

271
00:15:00.095 --> 00:15:03.405
which is 10 times more than the share

272
00:15:03.405 --> 00:15:05.125
of the French population that died

273
00:15:05.125 --> 00:15:07.125
during the Covid pandemic, right?

274
00:15:07.145 --> 00:15:10.765
And so, as I've said before, it's 20, over 22 years, 1%

275
00:15:10.765 --> 00:15:13.045
of the population died in France.

276
00:15:14.265 --> 00:15:16.765
So there is something about the methodology

277
00:15:18.025 --> 00:15:22.245
and, something which I'm going to first say by

278
00:15:23.055 --> 00:15:24.315
to find some causality.

279
00:15:25.375 --> 00:15:28.725
Uh, we use summer temperatures, so,

280
00:15:29.215 --> 00:15:30.445
we'll get back to that maybe.

281
00:15:30.505 --> 00:15:33.645
But I wanna show you something, which is summer temperatures

282
00:15:34.345 --> 00:15:38.605
are really what explains the spread of the calera.

283
00:15:39.865 --> 00:15:44.475
What is interesting is that, as you probably know, the north

284
00:15:44.475 --> 00:15:48.715
of France is not as warm as the south

285
00:15:48.715 --> 00:15:49.955
of France during the summer.

286
00:15:50.375 --> 00:15:52.435
And here there's something which is that

287
00:15:53.655 --> 00:15:57.465
it's not only the warm weather that matters, it's also

288
00:15:58.425 --> 00:16:00.025
humidity and evaporation.

289
00:16:00.845 --> 00:16:03.345
And it's not because the weather is warm,

290
00:16:03.725 --> 00:16:07.345
but there is more humidity, obviously, as humans, we feel,

291
00:16:08.125 --> 00:16:10.425
more, humidity, more accu,

292
00:16:10.455 --> 00:16:12.905
more accurately in the summer than in the winter.

293
00:16:13.485 --> 00:16:14.785
But it's always the case.

294
00:16:15.405 --> 00:16:19.145
But the south is always more humid than the North.

295
00:16:19.925 --> 00:16:22.665
And for this, this can, that's true in France,

296
00:16:23.005 --> 00:16:25.465
the north is more humid than the South,

297
00:16:26.365 --> 00:16:28.785
and, you just have, it's also true in China.

298
00:16:29.285 --> 00:16:33.705
And it's also true, simply speaking in Africa, where the

299
00:16:34.345 --> 00:16:37.585
northern coast of Africa on the Mediterranean Sea is

300
00:16:37.585 --> 00:16:41.065
obviously more humid than the Sara Desert, right?

301
00:16:41.765 --> 00:16:45.955
So based on that, I'm going to

302
00:16:46.865 --> 00:16:49.485
simply say, we're going to look at the Cora, see

303
00:16:49.485 --> 00:16:52.485
how the cholera affects economic variables,

304
00:16:52.905 --> 00:16:54.525
and to find some causality.

305
00:16:55.615 --> 00:16:57.285
We'll, I will use some temperatures.

306
00:16:57.585 --> 00:16:58.765
So we have, so it,

307
00:16:58.765 --> 00:17:01.765
there's a question about the technology adoption at

308
00:17:01.765 --> 00:17:02.845
the department level.

309
00:17:03.635 --> 00:17:06.495
So tools in agriculture per day laborers,

310
00:17:06.595 --> 00:17:08.655
and also engines in the mining industry.

311
00:17:09.115 --> 00:17:13.855
So the interesting thing is that for agricultural data

312
00:17:14.405 --> 00:17:17.455
only have two years, 1852 and 1862.

313
00:17:17.805 --> 00:17:22.615
However, for the mining statistics, I have yearly data,

314
00:17:23.725 --> 00:17:26.985
and it's also true for the production of agricultural crops

315
00:17:26.985 --> 00:17:29.625
and combustibles for agricultural crops only for two years.

316
00:17:29.965 --> 00:17:34.785
And for combustibles, it's, these are yearly data.

317
00:17:35.575 --> 00:17:39.425
Just one thing is that this is a small thing that says

318
00:17:40.055 --> 00:17:44.625
what is the value of machines as opposed to the, the price,

319
00:17:44.685 --> 00:17:47.225
the value of important machines as opposed to the price

320
00:17:47.325 --> 00:17:48.905
of other machines

321
00:17:48.905 --> 00:17:52.385
or sites are very,

322
00:17:52.705 --> 00:17:54.785
a very simple agricultural tools.

323
00:17:54.965 --> 00:17:58.865
And the answer is, there is no difference in the price

324
00:17:59.525 --> 00:18:01.545
of these different machines

325
00:18:01.545 --> 00:18:03.745
and tools suggesting that any

326
00:18:03.745 --> 00:18:07.065
of the results which we see in this paper are not about any

327
00:18:07.365 --> 00:18:10.105
budget constraint, and not due to any budget constraint,

328
00:18:10.105 --> 00:18:12.105
but really choices of

329
00:18:13.025 --> 00:18:15.195
individuals based on the relations

330
00:18:15.195 --> 00:18:17.155
between labor and technology.

331
00:18:18.485 --> 00:18:22.145
Uh, as a another source of data,

332
00:18:23.065 --> 00:18:25.865
I use a patents per depar per department.

333
00:18:25.925 --> 00:18:29.425
And so these are the 20 categories that the French state

334
00:18:30.415 --> 00:18:32.305
used in the 19th century.

335
00:18:33.285 --> 00:18:35.985
Um, there's a question which is, are,

336
00:18:36.095 --> 00:18:37.865
what are those categories exactly?

337
00:18:38.085 --> 00:18:40.225
And obviously, those data will self-reported.

338
00:18:41.125 --> 00:18:44.185
Now, I think that it's best to use those data

339
00:18:44.405 --> 00:18:47.625
and to take them at face value than to try to reconstruct,

340
00:18:47.645 --> 00:18:51.825
any of those data based on, you know,

341
00:18:51.825 --> 00:18:53.065
some modern classification.

342
00:18:54.695 --> 00:18:57.785
Also at the department level, I have data on the share

343
00:18:57.785 --> 00:19:00.305
of literate individuals among French Army conscript.

344
00:19:00.725 --> 00:19:03.425
So, throughout the 19th century,

345
00:19:04.635 --> 00:19:07.815
not everybody was subjected to a mandatory civil,

346
00:19:08.175 --> 00:19:09.935
military mandatory service.

347
00:19:10.645 --> 00:19:15.415
However, all French men aged 20 Hudson report

348
00:19:16.155 --> 00:19:18.295
for military service in the department

349
00:19:18.385 --> 00:19:19.655
where their father lived.

350
00:19:19.755 --> 00:19:22.935
So we have a good measure of regional

351
00:19:23.735 --> 00:19:25.935
literacy in France for the 19th century.

352
00:19:26.075 --> 00:19:27.695
We also data on the crude birth rates

353
00:19:28.475 --> 00:19:31.215
and measure of financial development deposits

354
00:19:31.215 --> 00:19:32.575
and saving banks per capita.

355
00:19:34.215 --> 00:19:37.435
In addition, I have individual data.

356
00:19:37.695 --> 00:19:42.035
So these are data from the survey of the 3000 families

357
00:19:42.095 --> 00:19:46.555
or in French, the, so what is, what is the,

358
00:19:46.865 --> 00:19:49.155
what is this survey about?

359
00:19:49.415 --> 00:19:53.475
It follows people whose first name, whose, I'm sorry,

360
00:19:53.485 --> 00:19:56.395
whose last name starts with the letter TRA.

361
00:19:57.015 --> 00:19:59.995
And so, the letters in

362
00:20:00.075 --> 00:20:01.675
and out of themselves are not rare,

363
00:20:01.815 --> 00:20:05.755
but the order of those letters are, is relatively rare, such

364
00:20:05.755 --> 00:20:09.315
that there are only three southern families whose last name

365
00:20:10.055 --> 00:20:12.595
starts with the letters TRA,

366
00:20:13.795 --> 00:20:18.255
and we can check, we can look at those people,

367
00:20:18.925 --> 00:20:21.255
over gen over several generations

368
00:20:21.915 --> 00:20:25.135
and know whether those people were able to

369
00:20:25.885 --> 00:20:27.735
sign their wedding license as opposed

370
00:20:27.735 --> 00:20:28.855
to market with a cross.

371
00:20:30.235 --> 00:20:33.815
And we also, they also offer other data such as, you know,

372
00:20:33.835 --> 00:20:35.895
age of marriage, the gender.

373
00:20:36.675 --> 00:20:37.775
And obviously we,

374
00:20:37.775 --> 00:20:40.735
we can also know whether they left an inheritance or not.

375
00:20:42.475 --> 00:20:46.815
Uh, few control variables, data on GDP per capita,

376
00:20:47.595 --> 00:20:49.005
some geographic and founders,

377
00:20:49.185 --> 00:20:51.965
and also a measure of seasonal,

378
00:20:52.005 --> 00:20:54.645
rainfall deviation from seasonal rainfall.

379
00:20:55.445 --> 00:20:59.645
So the empirical strategy is very simple.

380
00:21:01.225 --> 00:21:05.075
What I do is that I look at the share of cholera

381
00:21:06.325 --> 00:21:09.745
and see whether it explains technology adoption,

382
00:21:10.035 --> 00:21:13.065
technology innovation, human capital formation

383
00:21:14.085 --> 00:21:17.745
in a department over the years.

384
00:21:19.245 --> 00:21:21.465
Now, if you,

385
00:21:22.855 --> 00:21:26.355
and this is the moment where I say maybe I've convinced you

386
00:21:26.355 --> 00:21:29.235
that the spread of the cholera is random,

387
00:21:29.655 --> 00:21:32.395
at least in the 19th century, before 1855

388
00:21:32.855 --> 00:21:36.755
before John Snow, no one knows how the cholera is spread.

389
00:21:37.645 --> 00:21:39.865
And therefore, because no one knows

390
00:21:39.865 --> 00:21:41.585
how cholera is spread, it's completely random.

391
00:21:42.475 --> 00:21:45.375
Uh, some, someone has the cholera, it travels to

392
00:21:45.375 --> 00:21:48.015
that region rather than that region that explains

393
00:21:48.015 --> 00:21:49.495
that more people die here or there.

394
00:21:50.885 --> 00:21:53.295
Some of you may say this is fine,

395
00:21:53.635 --> 00:21:56.615
but there might still be some omitted variable,

396
00:21:57.545 --> 00:22:00.335
which is simultaneously correlated

397
00:22:00.365 --> 00:22:03.015
with the Sheriff Cora death in an area,

398
00:22:03.675 --> 00:22:06.735
and later on with technology adoption

399
00:22:06.915 --> 00:22:10.175
or income per capita, in which case you would want

400
00:22:10.175 --> 00:22:11.655
to see a first stage.

401
00:22:12.395 --> 00:22:16.795
And in that case, what I argue is that the temperature

402
00:22:17.835 --> 00:22:19.835
somewhat, it specifically summit temperature,

403
00:22:20.595 --> 00:22:22.635
explains the share of coral deaths.

404
00:22:23.335 --> 00:22:25.315
Now, as I've hinted

405
00:22:25.375 --> 00:22:28.265
before, it's actually

406
00:22:28.845 --> 00:22:32.425
the summer temperature is obviously warmer in the south,

407
00:22:33.085 --> 00:22:35.225
but it'll also for less humidity.

408
00:22:35.805 --> 00:22:39.145
And therefore, the prediction, which we find is indeed that

409
00:22:39.845 --> 00:22:44.315
summer temperature is negatively correlated

410
00:22:44.425 --> 00:22:46.035
with the share of coral deaths.

411
00:22:47.875 --> 00:22:50.905
So based on that,

412
00:22:51.765 --> 00:22:53.465
and the effect is quite substantial, right?

413
00:22:53.465 --> 00:22:55.625
So there's a quantitative effect, which is quite substantial

414
00:22:55.625 --> 00:22:59.505
because a 1% decrease in summer temperature would increase

415
00:22:59.565 --> 00:23:02.145
the share of coral death by 11.8%.

416
00:23:05.145 --> 00:23:08.775
There are plenty of robustness checks for those

417
00:23:08.775 --> 00:23:11.635
who are interested, but I won't show them to you.

418
00:23:12.215 --> 00:23:16.215
But in a nutshell, it's really summer temperature

419
00:23:16.915 --> 00:23:18.975
in the years of the Chole app pandemics.

420
00:23:19.005 --> 00:23:21.255
It's not summer temperatures in other years,

421
00:23:22.325 --> 00:23:24.975
it's not temperatures in other seasons,

422
00:23:25.085 --> 00:23:27.015
it's not rainfall in other seasons.

423
00:23:27.715 --> 00:23:31.615
Uh, it's not collected to some spread of

424
00:23:32.495 --> 00:23:33.775
pandemics in the 18th century

425
00:23:33.995 --> 00:23:36.535
or viral diseases in the modern era.

426
00:23:37.205 --> 00:23:40.375
It's also not linked to the characteristics

427
00:23:40.395 --> 00:23:43.935
of the population such as life expectancy, the size

428
00:23:43.935 --> 00:23:45.055
of a population, age

429
00:23:45.085 --> 00:23:48.045
and gender occupations of the population.

430
00:23:48.625 --> 00:23:53.465
And it's not even related to the, the diffusion,

431
00:23:53.605 --> 00:23:55.705
trade, internal trade in France

432
00:23:56.365 --> 00:24:01.305
or, even the wealth of individuals, their ability

433
00:24:01.305 --> 00:24:02.905
to sign, et cetera, et cetera.

434
00:24:03.885 --> 00:24:06.265
So coming to the main results of the paper,

435
00:24:09.595 --> 00:24:13.175
it appears that factors of production are

436
00:24:14.025 --> 00:24:16.435
complimentary in the industrial sector,

437
00:24:17.065 --> 00:24:20.395
that in the areas which were hit the hardest by the

438
00:24:21.305 --> 00:24:26.165
cholera pandemics, fewer machines were

439
00:24:26.675 --> 00:24:29.385
used in the year that follows.

440
00:24:29.925 --> 00:24:31.915
And so that's true

441
00:24:31.975 --> 00:24:35.035
for steam powered engine horsepower, steam powered engine.

442
00:24:35.775 --> 00:24:37.875
We, we do find an effect, which is not always,

443
00:24:38.005 --> 00:24:39.395
which is relatively robust,

444
00:24:39.395 --> 00:24:41.315
right across the different, tools.

445
00:24:42.495 --> 00:24:47.155
Uh, there is however no impact on the numbers of workers

446
00:24:47.495 --> 00:24:48.595
and the wage of workers.

447
00:24:49.475 --> 00:24:51.275
I don't wanna spend too much time on,

448
00:24:51.275 --> 00:24:54.835
arguing about the value of an insignificant reason result.

449
00:24:54.835 --> 00:24:57.355
But what is interesting is when we look at the production

450
00:24:57.355 --> 00:25:00.835
of combustibles, and what we find is

451
00:25:00.995 --> 00:25:04.875
that there is basically no effect on the value on the

452
00:25:04.875 --> 00:25:07.955
production of coal, but there is an effect

453
00:25:08.495 --> 00:25:09.875
on the value of Pete.

454
00:25:09.975 --> 00:25:14.475
And so Pete is a combustible, which is less valuable

455
00:25:14.475 --> 00:25:17.435
because it produces less energy than coal.

456
00:25:17.935 --> 00:25:21.955
And so here we have a very rational decision by,

457
00:25:22.275 --> 00:25:26.515
suppliers, which is that, when you have,

458
00:25:26.585 --> 00:25:28.435
complimentary factors of production

459
00:25:28.575 --> 00:25:32.795
and you have fewer machines, then you will obviously produce

460
00:25:33.745 --> 00:25:36.155
less of the least of valuable good.

461
00:25:36.415 --> 00:25:38.075
That's exactly what we see here.

462
00:25:39.985 --> 00:25:44.785
Uh, so the effect exists, they are clearly there,

463
00:25:44.925 --> 00:25:46.225
but they're not large,

464
00:25:46.845 --> 00:25:49.545
and they're also very much in the short term, right?

465
00:25:49.565 --> 00:25:52.945
So it's 1, 2, 3 years maximum, no more than that.

466
00:25:55.355 --> 00:25:58.615
And by the way, there is no long term impact, as far

467
00:25:58.615 --> 00:26:02.535
as I can tell on the share of,

468
00:26:02.645 --> 00:26:04.615
workers or professionals.

469
00:26:05.035 --> 00:26:06.335
And there's no effect,

470
00:26:06.335 --> 00:26:08.815
even on GDP per capita one 50 years later.

471
00:26:10.795 --> 00:26:13.535
Now, when we move to the impacts

472
00:26:13.915 --> 00:26:17.215
of the endemics in agriculture,

473
00:26:17.795 --> 00:26:22.005
and what appears is that the factors

474
00:26:22.105 --> 00:26:24.925
of production in agriculture are substitute.

475
00:26:25.665 --> 00:26:29.805
In other words, that indeed, even though the robot,

476
00:26:29.865 --> 00:26:33.205
the reasons are not always robust are all specifications.

477
00:26:33.785 --> 00:26:36.085
But once again, it depends whether you care more about the

478
00:26:36.165 --> 00:26:37.685
ULS or the two SLS.

479
00:26:39.145 --> 00:26:42.365
But what we find is that the share of Colorado death in,

480
00:26:42.625 --> 00:26:46.495
in the population led to more adoption

481
00:26:46.795 --> 00:26:48.415
of agricultural tools.

482
00:26:50.435 --> 00:26:52.525
What is interesting is

483
00:26:52.525 --> 00:26:55.125
that we do see the impact here on the number

484
00:26:55.225 --> 00:26:56.285
of day laborers.

485
00:26:56.945 --> 00:26:59.845
Uh, there is somehow a positive effect

486
00:27:00.385 --> 00:27:04.805
on the average day wage of day laborers, at least in the,

487
00:27:05.465 --> 00:27:08.655
two SLS specifications, which is hard

488
00:27:08.675 --> 00:27:10.855
to provide an in a straightforward interpretation

489
00:27:10.855 --> 00:27:14.095
because it is a general equilibrium interpretation.

490
00:27:14.275 --> 00:27:17.255
On the one hand, there are, there are fewer day laborers,

491
00:27:18.115 --> 00:27:19.455

492
00:27:19.475 --> 00:27:21.775
but on the other hand, there are,

493
00:27:21.785 --> 00:27:24.095
those daily laborers are more productive

494
00:27:24.095 --> 00:27:26.215
because they use more tools, and

495
00:27:26.215 --> 00:27:28.335
therefore it's not really surprising

496
00:27:28.335 --> 00:27:31.175
that we get a positive effect, positive

497
00:27:31.175 --> 00:27:32.215
and significant effect.

498
00:27:33.435 --> 00:27:37.285
Uh, so the effect, once again, the effect is there,

499
00:27:37.505 --> 00:27:41.805
but it's not very, it's not in the long term.

500
00:27:42.185 --> 00:27:43.525
And what is interesting is that

501
00:27:44.305 --> 00:27:47.525
the co pandemics did not have any effect on the adoption of

502
00:27:48.555 --> 00:27:50.205
same powered threshing machines,

503
00:27:50.205 --> 00:27:52.325
which were the most developed tools

504
00:27:53.065 --> 00:27:55.525
in agriculture in 19th century France.

505
00:27:56.825 --> 00:28:01.165
Uh, also we see very little effect on the production

506
00:28:01.185 --> 00:28:03.285
and on the value of the land.

507
00:28:05.065 --> 00:28:09.205
So, what is interesting when we look at innovation is

508
00:28:09.205 --> 00:28:11.445
that there is somehow an effect on innovation.

509
00:28:12.385 --> 00:28:15.765
It's very limited, and this effect is mostly driven

510
00:28:15.865 --> 00:28:18.245
by agricultural hydraulic patents,

511
00:28:19.175 --> 00:28:21.365
which is not surprising

512
00:28:21.715 --> 00:28:26.635
because France, as is, well known,

513
00:28:27.295 --> 00:28:28.715
was short on call

514
00:28:28.895 --> 00:28:33.135
and rich on, water, rich on rivers,

515
00:28:33.265 --> 00:28:35.215
which explains the slow spread

516
00:28:35.215 --> 00:28:36.695
of the steam engine in France.

517
00:28:37.195 --> 00:28:39.935
And it's not. And given the importance of agriculture,

518
00:28:39.965 --> 00:28:43.415
it's not really surprising that most patents,

519
00:28:43.515 --> 00:28:46.455
if there's an effect on more technology adoption

520
00:28:46.795 --> 00:28:49.615
and factors of production are substitute in agriculture,

521
00:28:49.865 --> 00:28:52.535
there will be more innovation in that sector

522
00:28:52.595 --> 00:28:55.375
and specifically in agricultural hydraulic patents,

523
00:29:00.595 --> 00:29:00.815
Sam,

524
00:29:02.575 --> 00:29:03.675
To say, I was gonna say.

525
00:29:03.675 --> 00:29:05.315
So that's brilliant actually, those results.

526
00:29:05.495 --> 00:29:07.395
Um, yeah, I was gonna follow up with a mechanism.

527
00:29:07.395 --> 00:29:08.635
What do you think caused all of this

528
00:29:11.305 --> 00:29:15.315
S How do

529
00:29:15.315 --> 00:29:16.435
you feel about doing a bit of this?

530
00:29:16.495 --> 00:29:17.955
Do you wanna do this one on, on? Um,

531
00:29:18.825 --> 00:29:20.995
Yeah, I was wondering what's, what's the best thing?

532
00:29:20.995 --> 00:29:24.195
Because on the one hand, I think conclusions, I think

533
00:29:24.195 --> 00:29:26.355
that this, this is probably better

534
00:29:26.415 --> 00:29:27.955
to show it like this Yeah, yeah.

535
00:29:27.955 --> 00:29:32.275
For economi, but this is probably, and then I can come back,

536
00:29:32.335 --> 00:29:35.505
but I can do the, I can do this,

537
00:29:35.565 --> 00:29:38.545
and I, I think it makes sense that we start from this,

538
00:29:39.135 --> 00:29:42.285
that I cut it end with it cut to myself.

539
00:29:42.905 --> 00:29:44.805
Yes. And then we come back to the

540
00:29:47.195 --> 00:29:48.195
Slide.

541
00:29:48.205 --> 00:29:51.235
To the slide. So 1, 2, 3.

542
00:29:51.975 --> 00:29:54.715
So, so what I'm gonna do is, so the only thing is if we,

543
00:29:54.735 --> 00:29:56.195
if we're now going to video, we've got

544
00:29:56.195 --> 00:29:57.875
to now put the video back on as full.

545
00:29:57.875 --> 00:30:01.435
So I think you've got stop being slide at your end at to,

546
00:30:01.745 --> 00:30:02.745
Okay. So do

547
00:30:02.745 --> 00:30:04.995
you think people saw that before? I think so, right?

548
00:30:05.455 --> 00:30:06.675
No, absolutely. Absolutely.

549
00:30:06.675 --> 00:30:09.155
That is captured and it, and it was super well delivered,

550
00:30:09.475 --> 00:30:11.195
actually, really fluid.

551
00:30:11.405 --> 00:30:13.275
Thank you. Thank you. Thank you.

552
00:30:13.495 --> 00:30:18.295
I'm, let's, there's,

553
00:30:18.325 --> 00:30:20.055
it's the first time I'm doing this.

554
00:30:20.525 --> 00:30:22.775
Yeah, really good. Um, so first of all, just for me,

555
00:30:22.955 --> 00:30:24.855
if you are going to go mechanisms on,

556
00:30:25.355 --> 00:30:26.895
on whatever it's called, PowerPoint,

557
00:30:26.895 --> 00:30:27.935
is it still called PowerPoint?

558
00:30:28.275 --> 00:30:29.895
Um, then I'm gonna come off,

559
00:30:29.955 --> 00:30:32.375
but if you're going to go mechanisms on video,

560
00:30:32.635 --> 00:30:33.815
we have to stay on, don't we?

561
00:30:34.965 --> 00:30:36.815
Okay. So, let's

562
00:30:39.905 --> 00:30:41.965
let, basically I'm going to talk over,

563
00:30:42.145 --> 00:30:43.245
I'm going to talk like this.

564
00:30:43.635 --> 00:30:47.765
Yeah. I'm going to basically say this Yeah.

565
00:30:47.945 --> 00:30:51.765
But without reading it, and then I'm going to share that.

566
00:30:52.225 --> 00:30:54.365
So comment. So what would happen is that I share that.

567
00:30:56.625 --> 00:30:58.785
Brilliant. And then at the end, you think for conclusions,

568
00:30:58.785 --> 00:30:59.785
we go back to just video

569
00:31:01.215 --> 00:31:06.185
And for conclusions, we, yeah, we, so

570
00:31:07.885 --> 00:31:09.625
yes, we're going to just for video,

571
00:31:09.825 --> 00:31:11.105
I think I'll do all, all of this.

572
00:31:12.335 --> 00:31:17.005
And then for conclusion, yeah,

573
00:31:17.005 --> 00:31:20.115
we CI mean, yeah, we can do, yeah,

574
00:31:20.625 --> 00:31:22.155
yeah. Okay.

575
00:31:22.325 --> 00:31:26.645
Right. Sorry if I'm being so, you know.

576
00:31:26.865 --> 00:31:27.765
No, you're actually being

577
00:31:27.925 --> 00:31:29.085
brilliant. It's actually really good.

578
00:31:29.875 --> 00:31:31.325
Okay. Thank you. Thank you.

579
00:31:34.405 --> 00:31:35.505
1, 2, 3, 4. Okay.

580
00:31:35.925 --> 00:31:40.765
And so, so now the

581
00:31:40.965 --> 00:31:44.415
question is what is the mechanism behind all this?

582
00:31:44.595 --> 00:31:48.535
And for this, we have to bill on, recent research

583
00:31:49.315 --> 00:31:53.215
on the link between human capital formation

584
00:31:53.315 --> 00:31:55.735
and technology adoption in the 19th century.

585
00:31:57.115 --> 00:32:02.055
And what is this new research about, papers, by,

586
00:32:02.115 --> 00:32:06.215
attack and co-authors in 2019, also, a paper which I,

587
00:32:06.255 --> 00:32:08.415
wrote with ODed Glar in 2021.

588
00:32:09.235 --> 00:32:12.805
And, and on a one in 2022, the,

589
00:32:13.965 --> 00:32:17.595
there is absolutely no doubt that in the 20th century

590
00:32:18.995 --> 00:32:21.085
education, human capital formation

591
00:32:21.085 --> 00:32:24.165
and technology are complimentary factors, right?

592
00:32:24.185 --> 00:32:27.125
We see the horizon mass education.

593
00:32:28.105 --> 00:32:31.565
But in the 19th century, there was often this idea that

594
00:32:33.195 --> 00:32:37.305
education was not, condu was not

595
00:32:38.755 --> 00:32:42.905
actually driven by technology adoption.

596
00:32:43.705 --> 00:32:46.465
Actually during the first industrial revolution, that is

597
00:32:47.015 --> 00:32:50.505
from the 1750s to the 1870s,

598
00:32:51.255 --> 00:32:54.545
basically all the new new technology was destroying the,

599
00:32:56.625 --> 00:32:58.685
the craft of autisms

600
00:32:59.105 --> 00:33:03.045
and actually led people to invest less in

601
00:33:03.575 --> 00:33:04.925
human capital formation.

602
00:33:05.505 --> 00:33:08.765
And so this new research by attack, and its co-authors

603
00:33:08.865 --> 00:33:12.605
and myself and ODed gallo, what appears is that

604
00:33:13.485 --> 00:33:16.805
actually human capital formation and by this literacy

605
00:33:17.265 --> 00:33:20.765
and technology adoption were complimentary factors.

606
00:33:21.985 --> 00:33:25.805
And this, obviously is the question whether

607
00:33:26.345 --> 00:33:28.845
if this is the case, then we should see

608
00:33:28.845 --> 00:33:32.365
that in the places which are hit the harshest by the

609
00:33:33.665 --> 00:33:38.185
pandemics, by the Cora pandemics, we should see that

610
00:33:38.735 --> 00:33:40.425
more people are literate.

611
00:33:40.525 --> 00:33:42.425
We should see some effect in the long run.

612
00:33:43.365 --> 00:33:46.605
And for this, I will use the,

613
00:33:48.845 --> 00:33:50.905
data on literacy.

614
00:33:54.015 --> 00:33:54.305
That,

615
00:33:59.065 --> 00:33:59.415
sorry.

616
00:33:59.475 --> 00:34:00.895
That's okay. Just good To share. Sorry.

617
00:34:00.895 --> 00:34:03.275
No, no, it's me. Yeah. 1, 2, 3.

618
00:34:03.555 --> 00:34:08.105
I will use the data on the literacy of,

619
00:34:08.245 --> 00:34:12.505
of the, at the individual level, looking at the share of

620
00:34:13.075 --> 00:34:16.465
bride and grooms who were able to sign their wedding license

621
00:34:17.085 --> 00:34:21.535
as opposed to market with a cross, which is obviously

622
00:34:22.355 --> 00:34:26.815
the lowest bound possible on literacy, which is already a,

623
00:34:27.035 --> 00:34:30.175
a significant step in liman capital formation

624
00:34:30.555 --> 00:34:33.615
and an indication on literacy.

625
00:34:34.775 --> 00:34:35.855
F-I-O-F-I-O,

626
00:34:36.035 --> 00:34:38.495
you're not sharing the screen currently. Do, are you aware of that?

627
00:34:39.395 --> 00:34:41.375
Yes. That's why I was about to do,

628
00:34:41.795 --> 00:34:43.735
that's why I said, let's go back. Oh,

629
00:34:43.795 --> 00:34:46.875
Man, I said, let's go back to the moment

630
00:34:46.875 --> 00:34:47.995
where I'm saying this,

631
00:34:48.575 --> 00:34:50.555
and I was doing the transition quietly.

632
00:34:56.055 --> 00:34:58.315
An order is an order, and discipline is discipline.

633
00:35:06.975 --> 00:35:11.885
And so indeed, when we look at the share

634
00:35:11.945 --> 00:35:14.605
of brides and grooms who were able

635
00:35:14.665 --> 00:35:17.485
to sign their wedding license, those people

636
00:35:17.545 --> 00:35:19.525
who were born one to 20 years

637
00:35:19.615 --> 00:35:23.565
after each pandemic, we see that the larger share

638
00:35:24.715 --> 00:35:29.135
of those people is able to sign their wedding license

639
00:35:29.755 --> 00:35:31.135
as opposed to market with a cross.

640
00:35:32.705 --> 00:35:35.815
There is also an effect at the,

641
00:35:35.815 --> 00:35:38.485
departmental level, less strong.

642
00:35:38.515 --> 00:35:42.925
It's mostly in the two SLS regressions for the individuals

643
00:35:42.925 --> 00:35:45.445
who were born in the year of a pandemic.

644
00:35:45.505 --> 00:35:50.045
And then the state evaluates their ability to

645
00:35:50.595 --> 00:35:54.005
send the, to their, their, their ability to read

646
00:35:54.005 --> 00:35:55.205
and write 20 years later.

647
00:35:55.745 --> 00:35:58.885
And there's also a persistent effect for the individuals

648
00:35:58.885 --> 00:36:00.725
who were born 20 years after each pandemic.

649
00:36:01.185 --> 00:36:04.525
So the effect disappears entirely for the individuals

650
00:36:04.525 --> 00:36:07.525
who were born 35 years after each pandemic.

651
00:36:07.585 --> 00:36:09.005
But that's not really surprising

652
00:36:09.005 --> 00:36:12.965
because we are basically looking at people, who were,

653
00:36:13.825 --> 00:36:17.805
who benefited from the, laws on men

654
00:36:18.385 --> 00:36:20.725
and free public schooling in France.

655
00:36:21.145 --> 00:36:23.485
Uh, therefore that's not really surprising

656
00:36:23.485 --> 00:36:27.005
that the slight advantage that,

657
00:36:27.065 --> 00:36:29.205
the slight free journal advantage caused

658
00:36:29.225 --> 00:36:30.605
by the Colorado disappeared.

659
00:36:31.735 --> 00:36:33.675
Uh, so there is an effect.

660
00:36:34.185 --> 00:36:36.355
It's there, but once again, it's small,

661
00:36:37.215 --> 00:36:38.425
it's slightly persistent.

662
00:36:40.045 --> 00:36:43.785
Uh, what's interesting is that if we look at the courses

663
00:36:44.005 --> 00:36:46.065
for male and female adults

664
00:36:46.065 --> 00:36:47.585
and apprentices, which we want to think

665
00:36:47.585 --> 00:36:50.865
as the 19th century equivalence of, courses

666
00:36:51.125 --> 00:36:54.305
for retraining workers, there is an impact.

667
00:36:55.135 --> 00:36:56.395
Uh, there's a slight impact.

668
00:36:56.465 --> 00:36:58.875
It's mostly for men, not so much for women.

669
00:37:00.095 --> 00:37:02.315
And there is an impact,

670
00:37:02.415 --> 00:37:06.885
but it's not really consistent on spending a slight impact.

671
00:37:07.995 --> 00:37:10.815
And there's absolutely no impact on primary school

672
00:37:10.815 --> 00:37:12.535
attendance as far as you can tell.

673
00:37:13.555 --> 00:37:18.335
And there's really no systematic impact on spending

674
00:37:18.515 --> 00:37:22.555
by the various branches of the French states.

675
00:37:22.805 --> 00:37:27.235
There are three branches, states, departments in commune,

676
00:37:27.235 --> 00:37:30.995
which are the, and there's not this very systematic impact

677
00:37:31.485 --> 00:37:36.405
suggesting that it's not really driven by

678
00:37:37.115 --> 00:37:38.165
this effect that we see.

679
00:37:38.185 --> 00:37:41.325
The, the impact of human capital formation is not driven by

680
00:37:43.255 --> 00:37:45.125
state spending, but it's really driven

681
00:37:45.225 --> 00:37:49.165
by the private investments, the preferences of parents

682
00:37:49.305 --> 00:37:51.805
for their children and their own human capital,

683
00:37:52.425 --> 00:37:55.325
and also for their, and the, their own human capital

684
00:37:55.385 --> 00:37:56.805
of the parents and also of the children.

685
00:37:57.155 --> 00:38:01.215
Because even during the first industrial revolution,

686
00:38:01.835 --> 00:38:04.515
it looks like education

687
00:38:04.735 --> 00:38:07.235
and technology where complimentary factors,

688
00:38:10.755 --> 00:38:13.865
there are of course many alternative explanations,

689
00:38:14.765 --> 00:38:18.065
and you might think that there are other factors than

690
00:38:18.635 --> 00:38:22.145
human capital formation that can explain the results.

691
00:38:22.805 --> 00:38:27.515
And I've listed some on those slides, migration,

692
00:38:27.515 --> 00:38:29.235
urbanization, fertility, edge

693
00:38:29.235 --> 00:38:31.035
of marriage, ity, or local finance.

694
00:38:31.375 --> 00:38:34.435
And the answer is the robustness checks

695
00:38:34.985 --> 00:38:36.005
do not show anything,

696
00:38:36.495 --> 00:38:39.845
which is probably not surprising the case of France, where

697
00:38:39.945 --> 00:38:43.245
as we know, the fertility patterns,

698
00:38:43.305 --> 00:38:47.685
the demographic transition started in the 18th century where

699
00:38:48.265 --> 00:38:50.045
ity had already declined in the 18th century

700
00:38:51.025 --> 00:38:54.285
and where there was a lot of internal migration and,

701
00:38:54.865 --> 00:38:57.125
but it's not really related to the Cora pandemics.

702
00:38:58.915 --> 00:39:00.175
So coming to a conclusion,

703
00:39:03.795 --> 00:39:07.345
what is interesting is to

704
00:39:08.085 --> 00:39:12.035
ask ourselves to what happens, what is the impact

705
00:39:12.035 --> 00:39:15.475
of the Cora outbreaks in 19th century France

706
00:39:16.535 --> 00:39:21.045
and more generally for us today.

707
00:39:22.105 --> 00:39:24.785
And so what appears is that?

708
00:39:25.165 --> 00:39:28.115
Yes, well, it appears that factors

709
00:39:28.175 --> 00:39:32.115
of production are complimentary in the industrial sector,

710
00:39:33.215 --> 00:39:36.935
substitute in the agricultural sector, the effect driven

711
00:39:36.955 --> 00:39:38.455
by human capital formation.

712
00:39:40.555 --> 00:39:43.055
And that's a result which exists

713
00:39:43.325 --> 00:39:47.295
because of the Malian world,

714
00:39:47.515 --> 00:39:49.235
or rather the fact

715
00:39:49.235 --> 00:39:51.715
that France is not in the maltus trap

716
00:39:51.715 --> 00:39:53.075
anymore in the 19th century.

717
00:39:54.095 --> 00:39:56.275
And it raises this question about the way

718
00:39:57.515 --> 00:40:00.075
countries should deal with pandemics nowadays.

719
00:40:01.215 --> 00:40:04.315
And, I don't wanna sound too callous about this.

720
00:40:04.325 --> 00:40:06.115
Economists have a reputation

721
00:40:06.115 --> 00:40:07.595
for being a bit callous sometimes.

722
00:40:08.815 --> 00:40:13.555
And, it's obvious that, you know, we should try

723
00:40:13.655 --> 00:40:17.535
as much as possible to avoid the death

724
00:40:17.595 --> 00:40:19.055
of people from pandemics,

725
00:40:21.295 --> 00:40:26.045
but it doesn't look like pandemics in a world

726
00:40:26.045 --> 00:40:28.125
where there's a disconnect between wealth

727
00:40:29.065 --> 00:40:31.405
and the number of people is very important.

728
00:40:35.975 --> 00:40:37.965
Great way to end. Thank you very much

729
00:40:38.065 --> 00:40:39.205
for talking to us, Raphael.

730
00:40:39.205 --> 00:40:40.125
Lemme tell you about your next steps.

Quick Answers

00:00:00 - What is the relationship between labor and technology? And for the sake of the argument, let's assume that labor is scarce. So your reaction will be, well, this is trivial. If labor is scarce, wages will increase. This will lead to more technology adoption. But that reasoning and plus the factors of production are substitutes. And then it immediately calls the other question, which is, given the level of technology, which we enjoy in modern societies nowadays, why don't we get 85% of unemployment? And the answer is, you need people to operate machines.

00:00:39 - of production can be complimentary. And if indeed labor-based cares and wages are high under complimentary factors of production, this will actually lead to less technology adoption because it's not profitable to invest in new machines. So that's what the study is about. And the study starts from this natural experiment, which I will argue is quite convincing the labor pandemics in 19th century France. So when labor is scarce, what do you think happens?

00:01:11 - or substitute in agriculture? Do you think that factors of production are complimentary, substitute in industry and beyond that, do you think that labor scarcity say, has a differential impact on other aspects of economic growth? Given that maybe the factors of production are complementary or substitute in agriculture or industry? Do you think that there will be more human capital

00:01:44 - Because we talk about the 19th century that there will be more investments in literacy in the sector where factors of production are complementary or substitute. And so it's not that straightforward. Obviously, the paper starts from a l long literature on the impact of diseases, on economic growth and the environment. And of course, there's a, this literature arguing

00:02:19 - the western world grew, because there was less labor and therefore there was more incentives to invest in machines. But also there was the end of Cdom. But here we are looking at different context, which is we're looking at France in the 19th century, and in other words, we're looking at a country which is out of the Milian trap. So what happens when you have the black death in the Middle Ages is that we live in a milian world, world where people,

00:02:54 - But here it's different because you are out of the Malian trap, And there's the disconnect between wealth and the number of children. And so the paper is also very relevant if there's any policy relevant, because we've been through a pandemic, recently. And that pandemic killed, fortunately killed upon

00:03:26 - What's interesting about France is that France was divided in 1719 into administrative areas of, called apartments. And those administrative areas were actually, meant to be of small size, right? So on average, there are 6,000 square kilometers, and France is, you know, going to say a large country, but still, basically the size of Texas.

00:04:02 - which had about 30 million people in the 1380s, so lost 1% of its population over 22 years between the first and the last, epidemics of cholera, which, is analyzed in this study. And so obviously what is interesting is not only the timing, but also the differential impact in some areas of France and some departments, no one were,

00:04:32 - Many people died. Even, in, in the southwestern department of Esh, 4.2% of the population died. And when I say died, died in one summer. We'll get back to that in a moment. So just a few words on the cholera epidemics. it's looks like already in India, the cholera bacterium was known,

00:05:04 - And it was only in, 1854. In 1854, Felipo Pacini was able, an Italian biologist was able to isolate the bacterium, but it was only in 1855. Then John Snow, was able to prove that the cholera bacteria, the cholera was spread through waters. And before, people thought that maybe the color was transmitted through air

00:05:32 - And one of the reason for that is that the cholera bacterium can survive six to seven weeks on the dry clothes, which were damped, and then at the, and which would damp and sweaty. And so very often a mode of transmission of the cholera was during funerals because people would touch the dead and touch the body of the dead and during those funeral wakes, and then it would also be infected with the color bacterium, and people would say, it's

00:06:03 - It was only in 1884 that, an effective treatment was provided. And clearly in France, where the weather is only, wet and humid during the summer, it's mostly during the summer, then the collateral is transmitted. And most people died from, the, the disease. And we're back to something which is that no one knew

00:06:34 - And that's why I, the study only looks at the pandemics in 18 32, 18 49, and 1854. No one knew how the disease spread, and no one knew how to stop it. And that's why, and then that's why in French nowadays, there's an expression called bl blue, literally blue fear, which refers to an extreme fear. And that actually referred to first to the cholera,

00:07:05 - because they were dehydrated, they would, their face would take a blue coloration. And so that's the consequence of the color pandemics in the French language to this day. So let's look at, the maps of France or the share of color death in each department. And as you can see, there was a differential timing. it's only in 1830 to 18 49, 18 54 that we get cholera pandemics.

00:07:40 - you see that very often, the most death ed in the north of France. And we'll get back to that, later because it'll be useful for the empirical, methodology. What's usually thought is that the cholera came from the UK by bolt, and then from the north it spread to Paris. And from Paris it spread to the rest of France.

00:08:11 - chole is mostly transmitted, we think that the call was transmitted by the soldiers embarking from the harbors of Tulo in Marse to fight the war in Crimea. Most of the time the pandemics remained in the north, the east, and the west of France. Right. And it has to do with the climate of France. And I will get back a link between humidity

00:08:43 - So what we, what I do in the, in this study is just look at what happens in the summers. But it's interesting to look at the progression of calera in 1854 throughout the year, the only year where we have month by month data. And as you can see, very few people die in the winter. And we do see a little rise in June when the weather gets warmer.

00:09:11 - July, August, September. And once again, it dies out in, October, November, December. So what's interesting is obviously that there's a huge differential in the share of people who died in every department across those three years. So on average, it's one 31%, which is 10 times more than the share of the French population that died

00:09:43 - And so, as I've said before, it's 20, over 22 years, 1% of the population died in France. So there is something about the methodology and, something which I'm going to first say by to find some causality. we use summer temperatures, so, we'll get back to that maybe. But I wanna show you something which is summer temperatures are really what explains the spread of the calera.

00:10:21 - of France is not as warm as the south of France during the summer. And here there's something which is that it's not only the warm weather that matters, it's also humidity and evaporation. And it's not because the weather is warm, but there is more humidity, obviously, as humans, we feel, more, humidity, more accu,

00:10:50 - But it's always the case that the south is always more humid than the north. And for this just can, that's true in France. The north is more humid than the South, and you just have, it's also true in China. And it's also true simply speaking in Africa, where the northern coast of Africa on the Mediterranean Sea is obviously more humid than the Sarah Desert, right?

00:11:23 - simply say we're going to look at the cholera, see how the cholera affects economic variables, and to find some causality. We'll, I will use summer temperatures. So we have, so there's a question about the technology adoption at the department level. So tools in agriculture per day, laborers and also engines and the mining industry. So the interesting thing is that for agricultural data

00:11:54 - However, for the mining statistics, I have yearly data, and it's also true for the production of agricultural crops and combustibles for agricultural crops only for two years. And for combustibles, it's, these are yearly data. Just one thing is that this is a small thing that says what is the value of machines as opposed to the, the price,

00:12:23 - of other machines or sites. A very, a very simple agricultural tools. And the answer is, there is no difference in the price of these different machines and tools suggesting that any of the results which we see in this paper are not about any budget constraints, and not due to any budget constraint, but really choices of individuals based on the relations

00:12:55 - as a another source of data, I use a patents per depa per department. And so these are the 20 categories that the French state used in the 19th century. there's a question which is, are what are those categories exactly? And obviously those data was self-reported. Now, I think that it's best to use those data and to take them at face value than to try to reconstruct,

00:13:28 - some modern classification. Also at the department level, I have data on the share of literate individuals among French Army conscript. So, throughout the 19th century, not everybody was subjected to, mandatory civil, military, military service. However, all French men aged 20 Hudson report for military service in the department

00:13:56 - So we have a good measure of regional literacy in France for the 19th century. We also data on the crude, both rates and measure of financial development deposits and saving banks per capita. In addition, I have individual data. So these are data from the survey of the 3000 families or in French, the De Me. So what is, what is the, what is this survey about?

00:14:30 - whose last name starts with the letter TRA. And so, the letters in and out of themselves are not rare, but the order of those letters are, is relatively rare, such that there are only three southern families whose last name starts with the letters TRA, and we can check, we can look at those people at over gen over several generations

00:15:02 - sign their wedding license as opposed to market with a cross. And we also, they also offer other datas as, you know, age of marriage, the gender. And obviously we, we can also know whether they left an inheritance or not. few control variables, data on GDP per capita, some geographic and founders, and also a measure of seasonal, rainfall

00:15:32 - So the empirical strategy is very simple. What I do is that I look at the share of cholera and see whether it explains technology adoption, technology innovation, human capital formation in a department over the years. Now, if you,

00:16:02 - that the spread of the cholera is random, at least in the 19th century, before 1855 before John Snow, no one knows how the cholera is spread. And therefore, because no one knows how cholera is spread, it's completely random. Some, someone has the cholera, it travels to that region rather than that region that explains that more people die here or there. Some of you may say this is fine,

00:16:34 - which is simultaneously correlated with the share of coral death in an area. And later on with technology adoption or incomp capita, in which case you would want to see a first stage. And in that case, what I argue is that the temperature summit, specifically summit temperature, explains the share of color of deaths. Now, as I've hinted

00:17:05 - the summer temperature is obviously warmer in the south, but it'll also for less humidity. And therefore the prediction which we find is indeed that summer temperature is negatively correlated with the share of Corona death. So based on that, and the effect is quite substantial, right? So there's a quantitative effect, which is quite substantial

00:17:36 - the share of car death by 11.8%. There are plenty of robustness checks for those who are interested, but I won't show them to you. But in a nutshell, it's really summer temperature in the years of the color epidemics. It's not summer temperatures in other years, it's not temperatures in other seasons, it's not rainfall in other seasons.

00:18:09 - pandemics in the 18th century or viral diseases in the modern era. It's also not linked to the characteristics of the population such as life expectancy, the size of a population, age and gender occupations of the population. And it's not even related to the, the diffusion, trade, internal trade in France or, even the wealth of individuals,

00:18:40 - So coming to the main results of the paper, it appears that factors of production are complimentary in the industrial sector, that in the areas which were hit the hardest by the cholera pandemics, fewer machines were used in the year that follows. And so that's true

00:19:12 - We, we do find an effect, which is not always, which is relatively robust, right across the different, tools. there is however no impact on the numbers of workers and the wage of workers. I don't wanna spend too much time on, arguing about the value of an insignificant result. But what is interesting is when we look at the production of combustibles, and what we find is that there is basically no effect on the value on the

00:19:43 - but there is an effect on the value of peat. And so peat is a combustible, which is less valuable because it produces less energy than coal. And so here we have a very rational decision by, suppliers, which is that, when you have, complimentary factors of production and you have fewer machines, then you will obviously produce less of the least valuable good.

00:20:16 - so the effect exists, they are clearly there, but they're not large, and they're also very much in the short term, right? So it's 1, 2, 3 years maximum, no more than that. And by the way, there is no long term impact, as far as I can tell on the share of, workers or professionals, and there's no effect, even on GDP per capita one run 50 years later.

00:20:53 - in agriculture, and what appears is that the factors of production in agriculture are substitute. In other words, that indeed, even though the robot, the reasons are not always robust occur all specifications. But once again, it depends whether you care more about the OLS or the two SLS. But what we find is that the share

00:21:22 - adoption of agricultural tools. What is interesting is that we do see the impact here on the number of day laborers. there is somehow a positive effect on the average day wage of day laborers, at least in the, two SLS specifications, which is hard to provide an in a straightforward interpretation because it is a general equilibrium interpretation.

00:21:55 - but on the other hand, there are those day laborers are more productive because they use more tools and therefore it's not really surprising that we get a positive effect, positive and significant effect. so the effect, once again, the effect is there, but it's not very, it's not in the long term. And what is interesting is that the co pandemics did not have any effect on the adoption of

00:22:26 - which were the most developed tools in agriculture in 19th century France. also we see very little effect on the production and on the value of the land. So, what is interesting when we look at innovation is that there is somehow an effect on innovation. It's very limited, and this effect is mostly driven by agricultural hydraulic patents,

00:22:58 - because France, as is, well known, was short on call and rich on, water rich on rivers, which explains the slow spread of the steam engine in France. And it's not. And given the importance of agriculture, it's not really surprising that most patents, if there's an effect on more technology adoption and factors of production are substitute in agriculture,

00:23:29 - and specifically in agricultural hydraulic patents. Now, the question is what is the mechanism behind all this? And for this, we have to build on, recent research on the link between human capital formation and technology adoption in the 19th century. And what is this new research about, papers, by, at attack and co-authors in 2019, also, a paper which I,

00:24:03 - and on a one in 2022, the, there is absolutely no doubt that in the 20th century education, human capital formation and technology are complimentary factors, right? We see the rise in mass education. But in the 19th century, there was often this idea that education was not condu was not

00:24:36 - That actually during the first industrial revolution, that is from the 1750s to the 1870s, that basically all the new new technology was destroying the, the craft of autisms and actually led people to invest less in human capital formation. And so this new research by attack, and its co-authors and myself

00:25:07 - actually human capital formation and by this literacy and technology adoption were complimentary factors. And this, obviously is the question whether if this is the case, then we should see that in the places which are hit the harshest by the pandemics, by the Colorado pandemics, we should see that more people are literate.

00:25:36 - And so indeed, when we look at the share of brides and grooms who were able to sign their wedding license, those people who were born one to 20 years after each pandemic, we see that the larger share of those people is able to sign their wedding license as opposed to market with a cross. There is also an effect at the,

00:26:07 - It's mostly in the two SLS regressions for the individuals who were born in the year of a pandemic. And then the state evaluates their ability to send the, to their, their, their ability to read and write 20 years later. And there's also a persistent effect for the individuals who were born 20 years after each pandemic. So the effect disappear entirely for the individuals who were born 35 years after each pandemic.

00:26:38 - because we are basically looking at people, who were, who benefited from the, laws on mandatory and free public schooling in France. And therefore that's not really surprising that the slight advantage that, the slight free journal advantage caused by the Colorado disappeared. so there is an effect, it's there, but once again, it's small, it's slightly persistent.

00:27:13 - for male and female adults and apprentices, which we want to think as the 19th century equivalence of, courses for retraining workers, there is an impact. there's a slight impact. It's mostly for men, not so much for women. And there is an impact, but it's not really consistent on spending a slight impact. And there's absolutely no impact on primary school

00:27:43 - And there's really no systematic impact on spending by the various branches of the French states. There are three branches, states, departments in commune, which are the cities, and there's not this very systematic impact suggesting that it's not really driven by this effect that we see. The, the impact of human capital formation is not driven by

00:28:14 - by the private investments, the preferences of parents for their children and their own human capital, and also for their, and the, their own human capital of the parents and also of the children. Because even during the first industrial revolution, it looks like education and technology where complementary factors, there are of course many alternative explanations

00:28:48 - human capital formation that can explain the results. And I've listed some on those slides, migration, urbanization, fertility, edge of marriage, ity, or local finance. And the answer is the robustness checks do not show anything, which is probably not surprising the case of France, where as we know, the fertility patterns, the demographic transition started in the 18th century

00:29:20 - and where there was a lot of internal migration and, but it's not really related to the Corona pandemics. So coming to conclusion, what is interesting is to ask ourselves to what happens, What is the impact of the Cora and outbreaks in 19th century France and more generally for us today?

00:29:54 - Yes, well, it appears that factors of production are complimentary in the industrial sector, substitute in the agricultural sector, that the effect is driven by human capital formation, and that's a result which exists because of the maltus world, or rather a fact that France is not in the maltus trap anymore in the 19th century.

00:30:27 - countries should deal with pandemics nowadays. And, I don't wanna sound too callous about this. Economists have a reputation for being a bit callous sometimes, and, it's obvious that, you know, we should try as much as possible to avoid the death of people from pandemics. But it doesn't look like pandemics in a world

00:30:58 - and the number of people is very important.

Faculti AI can make mistakes

Research

Policy

Analysis

Data

Originals

Explainers

Podcasts

Watch the latest episode of The Exchange – the show where influential policymakers and leading researchers engage in insightful discussions shaping the future.

Episode: Brazil's Economic Resilience