1
00:00:03,800 --> 00:00:05,800
Welcome to Nobel conversations.

2
00:00:07,100 --> 00:00:08,100
In this episode,

3
00:00:08,200 --> 00:00:10,200
Josh Angrist and Guido Imbens

4
00:00:10,200 --> 00:00:14,600
sit down with Isaiah Andrews to discuss
how the field of econometrics is evolving.

5
00:00:16,100 --> 00:00:17,900
So he doing Josh, you're both

6
00:00:18,200 --> 00:00:21,500
pioneers and developing tools for
Empirical research in economics.

7
00:00:21,500 --> 00:00:25,300
And so I'd like to explore sort of where
you feel like the field is heading,

8
00:00:25,300 --> 00:00:29,400
sort of Economics econometrics.
The whole thing to start,

9
00:00:29,900 --> 00:00:31,400
I'd be interested to hear

10
00:00:32,200 --> 00:00:35,200
about whether you feel like
the sort of the way in which

11
00:00:35,200 --> 00:00:36,700
the local average treatment effects from.

12
00:00:36,800 --> 00:00:38,700
Work sort of took hold

13
00:00:38,800 --> 00:00:43,500
has any lessons for how new empirical
methods and economics develop and spread or

14
00:00:43,500 --> 00:00:45,700
how they should? That's a good question.

15
00:00:46,700 --> 00:00:47,500
You go first.

16
00:00:49,700 --> 00:00:55,100
Yeah, somebody I think the important
thing is to come up with with

17
00:00:55,300 --> 00:00:59,200
the good conferencing cases. Where

18
00:01:00,600 --> 00:01:02,100
The questions are clear

19
00:01:02,400 --> 00:01:06,200
and we're kind of the methods
apply in general. So maybe

20
00:01:06,400 --> 00:01:12,000
one thing I kind of looking back
at the subsequent literature.

21
00:01:12,200 --> 00:01:13,400
So I really like the

22
00:01:13,600 --> 00:01:17,300
the regression discontinuity
literature. Whenever

23
00:01:17,400 --> 00:01:21,000
clearly a bunch of really convincing
examples and that allowed people to kind of

24
00:01:22,300 --> 00:01:27,200
think more clearly look harder at
a methodological. The questions,

25
00:01:27,400 --> 00:01:30,200
can it do a clear applications that
allow you to kind of think about

26
00:01:30,500 --> 00:01:34,200
Wow, do this type of sumption
seem reasonable here. What kind of

27
00:01:34,300 --> 00:01:38,000
what kind of things do we not?
Like in this, the early papers?

28
00:01:38,500 --> 00:01:42,300
How can we improve things? So having
clear applications motivating,

29
00:01:43,400 --> 00:01:46,400
these legislators? I think
it's it's very helpful.

30
00:01:46,800 --> 00:01:49,300
I'm glad you mentioned the
regression discontinuity. He do.

31
00:01:49,300 --> 00:01:53,300
I think there's a lot of
complementarity between IV and RD.

32
00:01:54,700 --> 00:02:00,200
Instrumental variables and
regression discontinuity and the

33
00:02:00,600 --> 00:02:05,800
A lot of the Econo metric applications
of regression discontinuity are what used

34
00:02:05,800 --> 00:02:07,800
to be called? Fuzzy rdd where

35
00:02:08,500 --> 00:02:12,100
you know, it's not discrete or
deterministic at the cutoff, but

36
00:02:12,500 --> 00:02:14,900
just to change in rates or intensity.

37
00:02:14,900 --> 00:02:19,800
And and the late framework helps us
understand those applications and gives us

38
00:02:19,800 --> 00:02:25,000
a clear interpretation for say something
like, in my paper with Victor Lavie,

39
00:02:25,000 --> 00:02:28,100
where we use, maimonides
rule, the class size cut-offs.

40
00:02:28,600 --> 00:02:30,200
What are you getting there? So?

41
00:02:30,400 --> 00:02:33,900
Of course, you can answer that question
with a linear constant effects model,

42
00:02:34,200 --> 00:02:37,600
but it turns out we're not limited
to that. And an RD is still

43
00:02:37,700 --> 00:02:39,900
very powerful and Illuminating,

44
00:02:40,900 --> 00:02:42,100
even when you know,

45
00:02:42,100 --> 00:02:46,000
the correlation between the cut
off and the variable of interest in

46
00:02:46,000 --> 00:02:51,000
this case class size is partial,
maybe even not that strong.

47
00:02:52,000 --> 00:02:56,400
So there was definitely a kind of a
parallel development. It's also interesting,

48
00:02:56,600 --> 00:03:00,200
you know, nobody talked about
regression discontinuity designs when we

49
00:03:00,300 --> 00:03:02,100
Were in graduate school. It was

50
00:03:02,400 --> 00:03:05,300
something that other social
scientists were interested in

51
00:03:05,800 --> 00:03:12,300
and that kind of grew up alongside the
late framework and we've both done work on

52
00:03:12,500 --> 00:03:15,000
both applications and methods there and

53
00:03:15,400 --> 00:03:19,800
it's been very exciting to see that
kind of develop and become so important.

54
00:03:20,000 --> 00:03:23,700
It's part of a general Evolution.
I think towards you know,

55
00:03:24,000 --> 00:03:30,200
credible identification, strategies, causal
effects less, you know, making a condom

56
00:03:30,400 --> 00:03:33,300
Tricks more about causal
questions that about models

57
00:03:33,700 --> 00:03:38,700
in terms of the future. I think one thing
that Slade has helped facilitate as a move

58
00:03:38,700 --> 00:03:42,700
towards more creative
randomized, trials, where,

59
00:03:42,800 --> 00:03:44,400
you know, there's something of interest,

60
00:03:45,500 --> 00:03:50,700
it's not possible or straightforward
to Simply turn it off or on

61
00:03:51,000 --> 00:03:52,900
but you can encourage it

62
00:03:53,200 --> 00:03:58,200
or discourage it. So, you subsidize
schooling with financial aid, for example,

63
00:03:59,000 --> 00:04:00,200
so now we have a whole

64
00:04:00,300 --> 00:04:02,300
Framework for interpreting that.

65
00:04:02,900 --> 00:04:03,500
And,

66
00:04:03,600 --> 00:04:06,900
and it kind of opens
the doors to randomized,

67
00:04:06,900 --> 00:04:09,400
Trials of things that that maybe would,

68
00:04:09,700 --> 00:04:10,200
you know,

69
00:04:10,300 --> 00:04:13,800
not have seen possible before we've,

70
00:04:14,500 --> 00:04:19,700
we've used that a lot in the work. We do
on schools in our in the blueprint Lab

71
00:04:20,000 --> 00:04:26,600
at MIT were exploiting random assignment
and in very creative ways, I think.

72
00:04:28,100 --> 00:04:32,300
Related to that. Do you see sort
of particular factors that make for

73
00:04:32,400 --> 00:04:34,300
useful research and econometrics.

74
00:04:34,400 --> 00:04:37,400
You've alluded to it?

75
00:04:37,400 --> 00:04:40,300
Having a clear connection to
problems that are actually coming up.

76
00:04:40,300 --> 00:04:45,000
And empirical practice is often a good
idea. I'll send it. Always a good idea.

77
00:04:45,700 --> 00:04:50,100
I often find myself sitting in an
economy metrics Theory, seminar.

78
00:04:50,700 --> 00:04:52,500
Say the Harvard MIT seminar

79
00:04:53,400 --> 00:04:57,000
and I'm thinking what problem is
this guy solving who has this?

80
00:04:57,100 --> 00:04:59,800
This problem and you know,

81
00:05:01,600 --> 00:05:04,700
sometimes there's an
embarrassing silence if I ask

82
00:05:04,900 --> 00:05:08,300
or there might be a
fairly contrived scenario.

83
00:05:08,800 --> 00:05:11,600
I want to see where the tool is useful.

84
00:05:12,500 --> 00:05:14,900
There are some purely foundational tools.

85
00:05:14,900 --> 00:05:17,600
I do take the point, you
know, there are people who are

86
00:05:18,200 --> 00:05:22,500
working on conceptual
foundations of you know,

87
00:05:22,600 --> 00:05:25,300
it's more becomes more like
mathematical statistics.

88
00:05:25,800 --> 00:05:27,000
I mean, I remember an early example

89
00:05:27,100 --> 00:05:28,100
I believe that that I,

90
00:05:28,200 --> 00:05:32,500
you know, I struggled to understand was
the idea of stochastic Equity continuity,

91
00:05:32,500 --> 00:05:36,900
which my one of my thesis advisors Whitney
knew he was using to great effect and

92
00:05:37,500 --> 00:05:39,900
I was trying to understand
that and there isn't really.

93
00:05:40,600 --> 00:05:45,200
It's really foundational. It's not
but an application that's driving that

94
00:05:46,100 --> 00:05:47,300
at least not immediately

95
00:05:48,600 --> 00:05:53,200
but but most things are not like that
and so there should be a problem.

96
00:05:53,800 --> 00:05:56,900
And the I think it's on the it's on

97
00:05:57,100 --> 00:05:59,600
On the, the seller of that sort of thing,

98
00:06:00,100 --> 00:06:04,400
you know, because there's opportunity
cost the time and attention and effort

99
00:06:04,400 --> 00:06:07,200
to understand things to, you
know, it's on the seller to say.

100
00:06:07,400 --> 00:06:08,900
Hey, I'm solving this problem

101
00:06:09,400 --> 00:06:12,900
and and here's a set of results
that show that it's useful.

102
00:06:12,900 --> 00:06:15,200
And here's some insight that I get.

103
00:06:16,200 --> 00:06:17,200
As you said, Josh, great,

104
00:06:17,200 --> 00:06:20,700
sort of there's been a move in the
direction of thinking more about causality

105
00:06:20,700 --> 00:06:22,800
in economics and empirical
work in economics,

106
00:06:22,900 --> 00:06:24,800
any consequences of sort of the Wilds,

107
00:06:24,800 --> 00:06:27,000
the spread of that view. That
surprised you or anything.

108
00:06:27,100 --> 00:06:31,400
UB was downsides of sort of the way that
he could empirical, economics has gone

109
00:06:31,500 --> 00:06:32,100
sometimes.

110
00:06:32,100 --> 00:06:38,500
I see somebody does Ivy and they get a
result which seems implausibly large.

111
00:06:38,800 --> 00:06:40,200
That's the usual case.

112
00:06:42,500 --> 00:06:43,900
So it might be, you know,

113
00:06:43,900 --> 00:06:48,600
an extraordinarily large causal effect
of some relatively minor Intervention,

114
00:06:49,100 --> 00:06:52,800
which was randomized or for
which you could make a case that

115
00:06:52,900 --> 00:06:56,900
that there's a good design.
And then when I see that,

116
00:06:57,100 --> 00:06:57,600
That and,

117
00:06:57,600 --> 00:06:58,500
you know, I think,

118
00:06:58,500 --> 00:06:58,900
you know,

119
00:06:58,900 --> 00:07:02,600
it's very hard for me to believe that this
relatively minor intervention has such

120
00:07:02,600 --> 00:07:04,000
a large defect,

121
00:07:04,100 --> 00:07:07,400
the author. Well, sometimes
resort to the local average,

122
00:07:07,400 --> 00:07:09,100
treatment effects theorem and say,

123
00:07:09,400 --> 00:07:12,700
wow, these compliers, you know,
they're special in some way.

124
00:07:13,300 --> 00:07:17,600
And, you know, they just benefit
extraordinarily from this intervention

125
00:07:18,100 --> 00:07:22,100
and I'm reluctant to take that
at face value. I think, you know,

126
00:07:22,100 --> 00:07:24,100
often when effects are too big,

127
00:07:24,300 --> 00:07:26,900
it's because the exclusion
restriction is failing. So

128
00:07:27,100 --> 00:07:31,700
Don't really have the right endogenous
variable to scale that result.

129
00:07:32,000 --> 00:07:35,700
And so I'm not too happy to see

130
00:07:35,800 --> 00:07:38,800
you know, just sort of
a generic heterogeneity

131
00:07:38,900 --> 00:07:43,800
argument being used to excuse something
that I think might be a deeper problem.

132
00:07:45,300 --> 00:07:47,400
I think it played somewhat
of an unfortunate roll pin.

133
00:07:47,400 --> 00:07:52,300
The discussions kind of between reduced
form and structural approaches where

134
00:07:52,600 --> 00:07:54,200
I feel that wasn't quite

135
00:07:55,000 --> 00:07:59,300
right. The instrumental
variables assumptions are

136
00:08:00,400 --> 00:08:05,200
at the core structural assumptions about
Behavior. They were coming from economic

137
00:08:07,100 --> 00:08:09,900
thinking about the economic
behavior of agents,

138
00:08:10,300 --> 00:08:15,000
and it's somehow it got
pushed in a Direction.

139
00:08:15,100 --> 00:08:17,600
That I think wasn't
really very helpful. If

140
00:08:18,800 --> 00:08:21,700
the way I think, initially the

141
00:08:22,800 --> 00:08:27,300
we wrote things up. It was it was describing
what was happening, there was set of

142
00:08:27,500 --> 00:08:32,200
methods. People were using be
clarified what those methods were doing

143
00:08:32,900 --> 00:08:38,500
and in a way that I think
contain a fair amount of insight,

144
00:08:39,100 --> 00:08:45,000
but it somehow it got pushed into a corner
that I think was not necessarily very

145
00:08:45,100 --> 00:08:48,600
or even just the language of
reduced form versus structural.

146
00:08:48,600 --> 00:08:51,100
I find kind of funny in
the sense that the right

147
00:08:51,100 --> 00:08:53,100
the local average treatment
effect model, right?

148
00:08:53,100 --> 00:08:55,300
The potential outcomes
model is a nonparametric.

149
00:08:55,300 --> 00:08:56,200
Structural model,

150
00:08:56,200 --> 00:08:58,600
if you want to think about it, as
you sort of suggested, he does.

151
00:08:58,600 --> 00:08:59,400
So, there's something,

152
00:09:00,100 --> 00:09:03,700
there's something a little funny about
putting these two things in a position when

153
00:09:03,800 --> 00:09:06,600
yes, well, that language, of
course, comes from the area, the

154
00:09:06,900 --> 00:09:09,800
70s equations framework that we inherited.

155
00:09:10,400 --> 00:09:12,400
It has the advantage that people seem

156
00:09:12,400 --> 00:09:15,000
to know what you mean
when you use it, but might

157
00:09:15,100 --> 00:09:18,200
That people are hearing different. Different
people are hearing different things.

158
00:09:18,300 --> 00:09:20,900
Yeah. I think I think veggies
Farmers had become use

159
00:09:20,900 --> 00:09:23,200
a little bit of the
pejoratives. Okay? Yeah.

160
00:09:23,300 --> 00:09:28,300
The word, which is not really quite
what it was originally intended for.

161
00:09:30,100 --> 00:09:34,100
I guess something else that strikes
me in thinking about the effects of

162
00:09:34,100 --> 00:09:38,200
the local average treatment effect
framework is that often folks will appeal to

163
00:09:38,200 --> 00:09:41,800
a local average, treatment effects
intuition for settings. Well, beyond

164
00:09:42,000 --> 00:09:44,900
ones, where any sort of formal
results has actually been

165
00:09:45,000 --> 00:09:49,700
Shhhhht. And I'm curious given
all the work that you guys did to,

166
00:09:49,900 --> 00:09:53,200
you know, establish late results in
different in different settings. I'm curious

167
00:09:53,300 --> 00:09:57,800
any thoughts on that. I think there's
going to be a lot of cases where

168
00:09:57,900 --> 00:10:02,200
the intuition does get.
You get you some distance,

169
00:10:02,800 --> 00:10:07,600
but it's going to be somewhat limited
and establishing formal results. There

170
00:10:08,400 --> 00:10:12,700
may be a little tricky and there may
be only work in special circumstances,

171
00:10:13,100 --> 00:10:13,500
you need.

172
00:10:14,600 --> 00:10:19,500
And you end up with a lot of formality
that may not quite capture the intuition

173
00:10:19,900 --> 00:10:23,200
sometimes I'm somewhat uneasy with them
and they are not necessarily the papers.

174
00:10:23,200 --> 00:10:25,000
I would want to ride that the

175
00:10:25,100 --> 00:10:30,000
but I do think something do intuition
orphaned US capture part of the

176
00:10:30,200 --> 00:10:31,100
of the problem.

177
00:10:33,100 --> 00:10:36,300
I think, in some sense we were
kind of very fortunate there

178
00:10:36,900 --> 00:10:40,500
in the way. The late paper go handle. It.
Don't know if that, actually the editor,

179
00:10:40,600 --> 00:10:41,700
made it much shorter

180
00:10:42,100 --> 00:10:46,300
and that then allowed us to kind of
focus on very clear, crisp results

181
00:10:47,100 --> 00:10:49,800
where if, you know, this,

182
00:10:50,000 --> 00:10:54,200
this is somewhat unfortunate tendency in
the commercialization of having the papers.

183
00:10:54,600 --> 00:10:58,800
Well, you should be able to fix that, man.
I'm trying to take some time to fix that.

184
00:10:59,400 --> 00:11:02,700
I think this is an example where it's sort
of very clear that having it. Be sure.

185
00:11:02,900 --> 00:11:08,000
It's actually impose that no paper can
be longer than the late paper that wow.

186
00:11:08,800 --> 00:11:14,300
Great. At least no Theory. No Theory Pig.
Yeah, and I think, I think they're well,

187
00:11:14,500 --> 00:11:16,800
I'm trying very hard to get
the papers to be shorter.

188
00:11:16,800 --> 00:11:18,700
And I think there's a lot of value

189
00:11:19,200 --> 00:11:22,600
today because it's often the second
part of the paper that doesn't actually

190
00:11:23,700 --> 00:11:26,500
Get you much further
and understanding things

191
00:11:27,000 --> 00:11:31,700
but and it does make things much
harder to read and, you know,

192
00:11:32,400 --> 00:11:33,700
it sort of goes back to

193
00:11:34,200 --> 00:11:38,500
how I think he kind of a trick should
be done to you should focus on the see.

194
00:11:38,700 --> 00:11:41,300
It should be reasonably
close to empirical problems.

195
00:11:41,500 --> 00:11:43,900
They should be very clear problems.

196
00:11:44,800 --> 00:11:48,900
But then often the the theory
doesn't need to be quite so long.

197
00:11:49,000 --> 00:11:49,300
Yeah,

198
00:11:51,100 --> 00:11:53,400
I think they had things have

199
00:11:53,600 --> 00:11:54,700
On a little off track.

200
00:11:56,400 --> 00:11:58,400
The relatively recent change has been a

201
00:11:58,500 --> 00:12:02,200
seeming big increase in demand for
people with sort of econometrics.

202
00:12:02,200 --> 00:12:04,800
Causal effect, estimation
skills in the tech sector.

203
00:12:05,000 --> 00:12:09,000
I'm interested either of you have
thoughts on sort of how that's gonna

204
00:12:09,200 --> 00:12:11,600
how that's going to interact with
the development of empirical methods,

205
00:12:11,600 --> 00:12:14,300
or Empirical research, and
economics. Going forward, sort of

206
00:12:14,600 --> 00:12:21,000
whether sort of a meta point, which
is there's this new kind of employer

207
00:12:21,800 --> 00:12:26,000
the Amazons and the Uber and, you know,

208
00:12:26,200 --> 00:12:27,600
Riser world

209
00:12:28,000 --> 00:12:29,300
and I think that's great.

210
00:12:29,300 --> 00:12:33,200
And I'd like to tell my students about
that, you know, especially at MIT.

211
00:12:33,200 --> 00:12:37,000
We have a lot of computer science
Majors. That's our biggest major

212
00:12:37,400 --> 00:12:42,800
and I try to seduce some of those folks
into economics by saying, you know,

213
00:12:43,200 --> 00:12:45,700
you can go work for these,

214
00:12:45,800 --> 00:12:48,400
you know companies that
people are very keen to

215
00:12:48,700 --> 00:12:50,800
work for because the work seems exciting,

216
00:12:52,000 --> 00:12:56,000
you know that the skills that you get in
econometrics are are as good or better.

217
00:12:56,100 --> 00:13:01,100
Better than than any competing discipline
has to offer. So you should at least

218
00:13:01,400 --> 00:13:04,200
take some econ, take some
econometrics. And some econ.

219
00:13:04,800 --> 00:13:07,000
I did a fun project with a uber

220
00:13:07,600 --> 00:13:12,900
on labor supply of Uber drivers and was
very, very exciting to be part of that.

221
00:13:13,100 --> 00:13:15,400
Plus. I got to drive for Uber for a while

222
00:13:15,900 --> 00:13:20,700
and I thought that was fun tonight. I did
not make enough that I was attempted to

223
00:13:21,100 --> 00:13:25,100
give up by a mighty job, but
I enjoyed the experience.

224
00:13:25,300 --> 00:13:26,000
I see a

225
00:13:26,200 --> 00:13:30,900
Cho challenge to our model
of graduate education here,

226
00:13:31,700 --> 00:13:37,400
which is if we're trading people
to go work at Amazon, you know,

227
00:13:37,900 --> 00:13:42,900
it's not clear. Why? You know, we should
be paying graduate stipends for that.

228
00:13:43,200 --> 00:13:45,400
Why should the taxpayer effectively

229
00:13:46,100 --> 00:13:51,400
be subsidizing? That our graduate education
in the u.s. Is generously subsidized?

230
00:13:51,400 --> 00:13:56,000
Even in private universities. It's
ultimately there's a lot of public money.

231
00:13:56,100 --> 00:13:59,300
Me there. And I think the
traditional rationale for that is,

232
00:13:59,500 --> 00:14:03,900
you know, we were training, Educators and
Scholars, and there's a great externality

233
00:14:04,300 --> 00:14:05,700
from the work that we do.

234
00:14:05,700 --> 00:14:09,600
It's either the research externality,
or a teaching externality.

235
00:14:10,100 --> 00:14:14,600
But, you know, if many of our students
are going to work in the private sector,

236
00:14:16,300 --> 00:14:21,700
that's fine, but that maybe their
employers should pay for that.

237
00:14:22,300 --> 00:14:25,100
He says, so different from
people working for a Consulting.

238
00:14:26,300 --> 00:14:26,900
Trust me.

239
00:14:27,200 --> 00:14:33,000
It's not clear to me that the number
of jobs in academics has changed.

240
00:14:33,100 --> 00:14:37,600
It's just, I feel like this is a
growing sector whereas Consulting,

241
00:14:37,700 --> 00:14:42,100
your right to raise that, it might
be the same for for Consulting.

242
00:14:43,300 --> 00:14:44,400
But this,

243
00:14:44,500 --> 00:14:47,500
you know, I'm placing more and
more students in these businesses.

244
00:14:47,500 --> 00:14:50,400
So, it's on my mind in
a way that I've sort of,

245
00:14:50,800 --> 00:14:55,500
you know, not been attentive to consulting
jobs, you know, Consulting was always,

246
00:14:55,600 --> 00:15:00,400
It's important and I think they'll so
there's some movement from Consulting back

247
00:15:00,400 --> 00:15:02,600
into research. It's a little more fluid.

248
00:15:02,900 --> 00:15:03,500
The,

249
00:15:03,900 --> 00:15:05,400
a lot of the work in the

250
00:15:06,400 --> 00:15:09,800
in both domains. I have to say,
it's not really different but

251
00:15:10,100 --> 00:15:13,700
you know, people who are working
in the tech sector are doing things

252
00:15:13,700 --> 00:15:16,800
that are potentially of scientific
interest, but mostly it's hidden.

253
00:15:17,100 --> 00:15:20,900
Then you really I have to say, you know,
why, why is the government paying for this?

254
00:15:21,800 --> 00:15:25,500
Yeah, although yeah, I mean taquitos point,
I guess it. There's a, there's a data.

255
00:15:25,600 --> 00:15:30,000
Question here of it has the sort
of total nanak. It sort of say

256
00:15:30,500 --> 00:15:30,900
private

257
00:15:31,300 --> 00:15:34,100
for-profit sector employment of econ Ph.D.

258
00:15:34,100 --> 00:15:37,700
Program graduates increased or has
it just been a substitution from

259
00:15:37,900 --> 00:15:40,200
finance and Consulting towards tack.

260
00:15:40,300 --> 00:15:44,300
I may be a reaction to something
that's not really happening

261
00:15:44,400 --> 00:15:48,200
so bad. I've actually done some work
with some of these tech companies.

262
00:15:49,100 --> 00:15:52,300
So I don't disagree with Justice
point that we need to think

263
00:15:52,300 --> 00:15:55,100
a little bit about the funding model
whose it was in the end paying for the

264
00:15:55,600 --> 00:15:59,400
It education. But from a
scientific perspective.

265
00:16:00,100 --> 00:16:03,500
The only do these places have
have great data and nowadays.

266
00:16:03,500 --> 00:16:07,100
They tend to be very careful
with that for privacy reasons,

267
00:16:07,500 --> 00:16:08,800
but also have great questions.

268
00:16:10,200 --> 00:16:11,200
I find it very

269
00:16:11,600 --> 00:16:13,300
inspiring kind of to listen to

270
00:16:13,300 --> 00:16:15,800
the people there and kind of see
what kind of questions they have

271
00:16:15,900 --> 00:16:17,300
and often their questions.

272
00:16:18,200 --> 00:16:20,600
That also come up outside of these.

273
00:16:20,700 --> 00:16:25,400
These companies have a couple of
papers with the rights in the chat.

274
00:16:25,800 --> 00:16:26,300
And then

275
00:16:26,500 --> 00:16:31,600
as soon as an atheist kind of where we
look at ways of combining experimental data

276
00:16:31,700 --> 00:16:34,100
and observational data, and can it there.

277
00:16:35,500 --> 00:16:38,600
Rights Chetty was interested
in what is the effect

278
00:16:38,700 --> 00:16:44,600
of Early Childhood programs on outcomes
later in life? Not just kind of test scores,

279
00:16:44,600 --> 00:16:48,300
but on earnings and stuff, and
we cannot be developed methods

280
00:16:48,600 --> 00:16:51,500
that would help you shed
light on that, on the some,

281
00:16:52,700 --> 00:16:55,000
in some settings and the same problems.

282
00:16:56,300 --> 00:17:00,500
Came up kind of in this
tech company settings.

283
00:17:00,800 --> 00:17:03,700
And so for my perspective, it's

284
00:17:04,400 --> 00:17:07,500
the same kind of a stocking two
people doing a protocol work.

285
00:17:07,600 --> 00:17:11,800
I tried to kind of look at these
specific problems and then try to come up

286
00:17:11,900 --> 00:17:18,300
with more General problems that we
formulating the problems at a higher level.

287
00:17:18,500 --> 00:17:22,900
So that I can think about solutions
that work in a range of settings.

288
00:17:23,400 --> 00:17:25,500
And so from that perspective, the

289
00:17:25,700 --> 00:17:30,300
His with the the tech companies I
just very valuable and very useful.

290
00:17:30,900 --> 00:17:31,400
It's know.

291
00:17:31,700 --> 00:17:33,700
We do have students. Now spent

292
00:17:33,800 --> 00:17:37,200
doing internships there and
then coming back and writing

293
00:17:37,400 --> 00:17:43,400
more interesting thesis, as a
result of their experiences there.

294
00:17:44,600 --> 00:17:47,800
If you'd like to watch more
Nobel conversations, click here,

295
00:17:48,200 --> 00:17:50,500
or if you'd like to learn
more about econometrics,

296
00:17:50,600 --> 00:17:53,200
check out Josh's mastering
econometrics series.

297
00:17:53,700 --> 00:17:55,500
If you'd like to learn more about he do.

298
00:17:55,600 --> 00:17:58,300
Josh and Isaiah check out
the links in the description.