Popular in Social Sciences

Books

City Time: On Being Sentenced to Rikers Island

Independent Scholar David A. Campbell

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

Economics

Gender and Economics

University of California Santa Barbara Shelly Lundberg

American Politics

Keeping the Republic

Boston College Marc Landy

Banks and Policy Organisations

Greening the financial sector

European Investment Bank Matteo Ferrazzi

British Politics

Entrepreneurial thinking: the politics and practice of policy impact

University of Sheffield Matthew Flinders

Culture and Society

Carceral Entrapments: Views from the Prison/Street Interface in India

Indian Institute of Technology Delhi Mahuya Bandyopadhya

Shaping Institutions

View more insights in Social Sciences - Economics - Microeconomics

View more figures from Bocconi University

View the page of WilliamFuchs and Satoshi Fukuda

Read the Study

William Fuchs and Satoshi Fukuda discuss a simple model of the evolution of institutions, where leaders’ actions
have a persistent effect by shaping the norms of the institutions they lead

Image courtesy of interviewee. February 1, 2024

Generating Workspace

Export as PDF

This video video discusses a study motivated by understanding why countries or institutions with similar legal frameworks can produce vastly different outcomes. The examples of Argentina and the United States are used to illustrate this phenomenon, given that Argentina's constitution was modeled after the U.S. version, yet their political and economic outcomes diverged significantly.

The study examines the role of political leaders in shaping institutional norms and behaviors, emphasizing that leaders' actions can set precedents that influence future governance. It highlights the phenomenon of democratic backsliding, where leaders might gradually undermine democratic processes, and how varying degrees of political norms affect this relationship.

Using economic and game theory, the study develops a formal model to analyze these dynamics, allowing for comparative analysis of different political settings, including corporations. It posits that leaders' incentives to respect or abuse institutions depend significantly on the strength of institutional norms and leader types.

Key findings suggest that strong political norms can act as a check against leader misconduct, while weak norms may encourage it. The paper also explores the implications of term limits, indicating that longer terms can potentially incentivize early abuses of power by leaders, complicating governance further.

Through empirical support, including indices of electoral democracy, the study advocates for maintaining strong institutional norms while recognizing the nuanced interactions between leadership behavior and democratic health. It calls for attention on how free media and competitive political environments can shape leaders' accountability. Overall, the research contributes to understanding the long-term implications of leadership on institutional integrity, encouraging future research and policy considerations in political science and economics.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.225 --> 00:04.285
The study is, is, is motivated by, by trying to understand

00:04.705 --> 00:07.425
how, countries or,

00:07.425 --> 00:12.145
or more generally, institutions that have very similar,

00:12.805 --> 00:14.865
sort of legal or, or, or,

00:14.865 --> 00:19.675
or written rules, can have very different outcomes.

00:19.935 --> 00:21.395
Uh, to some extent.

00:21.615 --> 00:23.795
One of the motivating examples is,

00:23.855 --> 00:25.995
is Argentina versus the the us.

00:26.775 --> 00:28.715
Um, I happen to be Argentinian

00:28.775 --> 00:31.875
and, the Argentine constitution was actually modeled

00:32.125 --> 00:33.995
after the, the US Constitution.

00:34.805 --> 00:36.545
And, needless to say, our,

00:36.565 --> 00:38.705
our outcomes have been somewhat, dissimilar.

00:39.005 --> 00:41.145
Uh, and so trying to understand that

00:41.285 --> 00:44.465
and how, you know, is it that the institutions are good?

00:44.465 --> 00:46.025
Is it that the US was sort of lucky

00:46.125 --> 00:50.425
and had, very good, people sitting in the chair, so

00:50.425 --> 00:53.545
to speak, or, or, or is it something else?

00:53.565 --> 00:54.905
And, and, you know, that's sort

00:54.905 --> 00:55.945
of a little bit the motivation.

00:55.965 --> 00:57.265
And, and I guess these days,

00:57.865 --> 00:59.305
although that was not the original motivation,

00:59.325 --> 01:03.065
but with, with with the Trump, you know,

01:03.625 --> 01:06.555
pushing on the institutions, pushing on the norms, it sort

01:06.555 --> 01:08.195
of became a little bit more relevant again,

01:08.895 --> 01:11.995
How the behavior of the leader shapes institution

01:12.215 --> 01:13.555
is actually very recognized.

01:14.375 --> 01:17.555
So during the course of our study, we find, we found

01:17.555 --> 01:21.155
that indeed, George Washington even mentioned that

01:21.265 --> 01:24.475
what he does is gonna be, become a precedent in the future.

01:24.615 --> 01:27.475
So he was very careful about what he did indeed.

01:28.295 --> 01:33.015
So the idea that, again, you know, the, the behavior

01:33.015 --> 01:36.335
of the reader, current reader affects the behaviors

01:36.335 --> 01:39.455
of the future reader is recognizing political science,

01:39.675 --> 01:40.695
the role, and,

01:40.755 --> 01:42.575
and more general in the social science in,

01:42.675 --> 01:43.935
in, in, in general.

01:44.915 --> 01:48.495
And more recently, as we mentioned, you know,

01:48.845 --> 01:51.775
there's a phenomenon called the Democratic back riding.

01:52.435 --> 01:55.295
So if you take a look at like Chave in Venezuela

01:55.315 --> 01:57.535
or OB in Hung, you know, OB in Hungary

01:57.955 --> 02:01.815
and this intelligence, so forth, so forth,

02:01.815 --> 02:03.615
political reader basically tries

02:03.675 --> 02:06.855
to undermine some democratic practices gradually piece

02:06.855 --> 02:10.615
by piece, you know, for instance, by changing the,

02:11.035 --> 02:12.895
the competition of Supreme Court judges

02:13.475 --> 02:15.175
or changing the oversight of

02:15.695 --> 02:17.215
electrical rules and so forth, so forth.

02:17.835 --> 02:22.335
So our model actually indeed, tries to shed right on

02:22.635 --> 02:26.785
how the behavior of of the political leader, you know,

02:26.855 --> 02:29.545
gradually undermines institutional norms

02:30.205 --> 02:34.485
and our focus on like unwritten news, you know,

02:34.485 --> 02:37.045
because if we take a look at the current phenomena,

02:37.215 --> 02:41.085
oftentimes the behavior of the readers are more like,

02:42.405 --> 02:44.725
informal in the sense that the readers are elected

02:44.815 --> 02:46.965
under seemingly fair elections.

02:47.345 --> 02:50.445
You know, seemingly they, respect the constitutions,

02:50.505 --> 02:51.645
but they kind of

02:52.355 --> 02:55.045
changed the political norms, constitutional norms.

02:55.145 --> 02:58.935
In fact, you know, we try to, study the attack

02:58.965 --> 03:01.365
of the program, using the,

03:01.435 --> 03:02.965
like, economic and game theory.

03:03.385 --> 03:06.725
And one of the advantage of advantages of this is, you know,

03:06.965 --> 03:10.245
actually our paper speaks also to corporations.

03:11.065 --> 03:14.085
So, you know, we can also talk about how, you know,

03:14.265 --> 03:18.565
leader in corporation like tries to capture

03:19.185 --> 03:20.525
at the corporate board,

03:21.065 --> 03:23.445
You know, a lot of the literature and, and, and,

03:24.585 --> 03:27.605
and sort of more on the legal scholars and, and, and,

03:27.665 --> 03:29.045
and political science,

03:29.805 --> 03:31.885
although they've described some of this phenomenon,

03:31.915 --> 03:35.765
it's more, you know, to some extent verbal.

03:35.955 --> 03:38.365
It's more descriptive than, than than analytic,

03:38.425 --> 03:39.605
sort of to speak.

03:40.225 --> 03:43.205
And so part of our contribution is, is to write a,

03:43.285 --> 03:46.325
a formal model that that tries to,

03:46.625 --> 03:48.085
to capture this phenomena.

03:48.745 --> 03:51.045
And, and by having a formal model and,

03:51.045 --> 03:53.525
and, you know, putting in different elements to the model,

03:53.525 --> 03:57.205
like, you know, probability of reelection, you know,

03:57.305 --> 03:59.525
the benefits of being in power and, and,

03:59.545 --> 04:03.685
and different parts to it, it allows us also to, to do sort

04:03.685 --> 04:05.325
of comparative statics analysis

04:05.325 --> 04:07.125
or think about policy, think about

04:07.395 --> 04:09.005
what would happen if we change this?

04:09.275 --> 04:12.805
What, how would the leaders optimal response change

04:12.905 --> 04:14.045
and, and, and so on.

04:14.185 --> 04:16.965
So having this, you know, it's not

04:16.965 --> 04:19.325
that the idea is necessarily new to some extent,

04:19.325 --> 04:21.725
this is the, it is more the, the contribution

04:21.725 --> 04:25.525
and the modeling of, of this, using, again, more sort

04:25.525 --> 04:26.525
of economic techniques.

04:26.905 --> 04:29.525
And that allows us then to, to speak about changes

04:29.625 --> 04:32.805
to the environment or, or what we might expect

04:33.105 --> 04:37.965
In terms of the, the law of like readers readership,

04:38.145 --> 04:40.805
on like democratic practices, for instance.

04:41.025 --> 04:44.005
You know, we use, varieties of democracy v them

04:44.505 --> 04:46.725
to provide some, you know, empiric support of our theory.

04:47.185 --> 04:50.285
So like, for instance, like we measure, you know, how,

04:50.295 --> 04:53.125
again, how political readers undermine institutions by,

04:53.125 --> 04:55.605
you know, measuring, the index

04:55.745 --> 04:57.725
of like electric electoral democracy.

04:58.505 --> 05:01.565
So we can clearly see that how, you know, like,

05:01.595 --> 05:04.285
political readers, in fact, like for instance,

05:04.285 --> 05:05.645
Chave in Venezuela,

05:05.655 --> 05:08.525
under undermined the democracy, so forth, of course.

05:08.985 --> 05:11.765
And, our actually model also speaks to

05:12.865 --> 05:16.325
how difficult it is to restore democratic practices,

05:16.865 --> 05:21.085
you know, once the like, the institutional norms,

05:21.225 --> 05:22.685
became lower.

05:24.025 --> 05:28.565
So in fact, as you know, as measured by,

05:28.905 --> 05:31.805
or like as kind of supported by the poor record,

05:31.945 --> 05:33.045
of Arab Spring.

05:33.505 --> 05:37.605
So we, we, we measured, like demo, like the index

05:37.605 --> 05:41.205
of electrical democracy in those like Arab spring countries

05:41.205 --> 05:44.685
that like EMN, like Egypt and so forth, so forth.

05:44.985 --> 05:47.405
So we clearly see that, you know, even if

05:48.135 --> 05:52.125
after like revolutions, the in index

05:52.125 --> 05:54.685
of democratic in index, increased, but

05:54.935 --> 05:59.325
after some years in the index became kind

05:59.325 --> 06:00.765
of lower again.

06:00.865 --> 06:04.045
So we also contrast, as Willie said,

06:04.065 --> 06:05.885
the US versus Argentina.

06:05.905 --> 06:07.325
So we, you know, the, the, the,

06:07.465 --> 06:09.725
the data data set actually allows us

06:09.745 --> 06:11.605
to take a look at the wrong run outcomes.

06:11.985 --> 06:16.525
So we actually in fact compare, some like, for instance,

06:16.765 --> 06:18.285
judicial constraints on executives

06:18.475 --> 06:20.485
between the US and Argentina.

06:20.945 --> 06:23.005
The main results have have sort of two parts.

06:23.225 --> 06:27.725
Uh, the first part is, you know, giving the, the level

06:27.725 --> 06:28.845
of the institutions

06:29.145 --> 06:33.945
and, the, the type of the leader, which,

06:33.955 --> 06:35.505
which plays a role in our model.

06:35.505 --> 06:38.065
So the leader can be more honest

06:38.205 --> 06:40.145
or more dishonest or more patriotic.

06:40.145 --> 06:41.945
There's different ways to interpret that.

06:42.445 --> 06:45.145
Uh, but so the first part is we see like, given

06:45.145 --> 06:47.345
what the institutions look like and,

06:47.345 --> 06:49.505
and the type of the leader, what's the optimal

06:49.605 --> 06:50.825
policy for the leader?

06:50.825 --> 06:53.065
Where basically what we mean by

06:53.065 --> 06:55.185
that is does the leader have incentives

06:55.325 --> 06:58.065
to respect institutions or to abuse these institutions?

06:58.775 --> 07:02.705
Okay? And so the first, result is that we show that,

07:03.055 --> 07:05.545
when we have a stationary mall, by that I mean

07:05.545 --> 07:08.505
that there's no sort of explicit term limit.

07:08.565 --> 07:10.985
So it's more like a constitutional system, if you will.

07:11.525 --> 07:13.625
Uh, then the optimal action

07:13.625 --> 07:16.105
for the leader is gonna be constant over time.

07:16.605 --> 07:18.865
And so that simplifies a lot of the analysis and,

07:18.865 --> 07:23.585
and allows us to, to characterize, a threshold level of,

07:23.645 --> 07:24.825
of on the type space.

07:24.845 --> 07:25.945
So given the norm level.

07:26.045 --> 07:28.905
So if the norms, for example, are very strong,

07:29.815 --> 07:33.345
only relatively high, you know, sorry,

07:33.455 --> 07:36.865
only relatively very dishonest types would be

07:36.865 --> 07:38.025
still tempted to cheat.

07:38.175 --> 07:41.385
Whereas if the norm levels are, are weaker, you know,

07:41.775 --> 07:44.745
more types would be tempted to take abuse of those norms.

07:44.855 --> 07:46.305
Okay? That relates a little bit

07:46.305 --> 07:49.425
to the points Atosh was making earlier, how difficult it is

07:49.425 --> 07:52.465
to restore democracy once you come from a very bad system.

07:53.445 --> 07:54.705
And so that's the first part.

07:54.705 --> 07:59.145
And then the second part is looking at using that as to try

07:59.145 --> 08:01.465
to understand what's the long run,

08:01.465 --> 08:02.665
properties of the system.

08:02.725 --> 08:05.625
So we're basically, you have a leader in place,

08:05.845 --> 08:07.625
the leaders taking their optimal action

08:08.075 --> 08:10.305
after some time, that leader will get replaced.

08:10.305 --> 08:13.865
So replacement is endogenous, it depends both on the,

08:14.005 --> 08:17.465
on the level of the norms and on the leader type.

08:17.815 --> 08:22.385
Okay? So the idea here is that if you have,

08:24.025 --> 08:25.905
stronger norms and,

08:26.045 --> 08:30.185
and the leader, behaves well, okay,

08:30.535 --> 08:32.625
he's less likely to some extent to be replaced

08:33.585 --> 08:35.985
vis-a-vis the situation where the norms are very strong

08:35.985 --> 08:38.345
and the leader behaves badly, okay?

08:38.345 --> 08:39.745
So there's like, when the norms are strong,

08:39.745 --> 08:41.345
it's like there's more scrutiny in some sense.

08:41.885 --> 08:44.505
So, so that the leader is more likely to be re

08:45.105 --> 08:47.145
replaced if he behaves or she behaves badly.

08:47.875 --> 08:50.865
Conversely, when, when the norms are weak, right?

08:51.125 --> 08:53.905
Uh, now misbehaving

08:53.965 --> 08:56.765
or abusing the norms has less punishment from a,

08:56.765 --> 08:58.245
from an electoral point of view,

08:58.245 --> 08:59.525
you're less likely to be replaced.

08:59.785 --> 09:01.045
And actually, it's even possible.

09:01.045 --> 09:02.805
We allow in them all the possibility that

09:03.525 --> 09:06.845
actually abusing the norms helps you remain power, okay?

09:06.845 --> 09:09.405
So that helps capture some of these democratic backsliding

09:09.405 --> 09:10.845
and autocratic regimes, right?

09:10.845 --> 09:14.485
So, so once the norms are very weak, by abusing them, it,

09:14.485 --> 09:17.565
it makes it very unlikely you get replaced, okay?

09:18.025 --> 09:20.085
So with that, which is an important building blog

09:20.145 --> 09:22.005
of them all, we, we see

09:22.005 --> 09:24.365
what would happen in the long run once, you know,

09:24.365 --> 09:26.565
leaders are endogenously being replaced.

09:27.305 --> 09:31.965
And, and basically what we show is that,

09:33.315 --> 09:36.405
once, once the norms are endogenous, so this is the,

09:36.465 --> 09:38.885
the importance of the endogeneity of the norms.

09:39.905 --> 09:43.165
You, you can have the possibility of this situation

09:43.165 --> 09:47.295
where starting from the same level of written rules, okay?

09:48.195 --> 09:52.015
The, if you happen to have bad early leadership,

09:52.925 --> 09:54.295
okay, what's gonna happen is

09:54.295 --> 09:56.815
that bad early leadership is gonna abuse the norms,

09:57.115 --> 09:58.535
so it's gonna weaken the norms.

09:59.155 --> 10:02.255
And then even if that leadership gets replaced later on,

10:02.795 --> 10:06.855
the norms are gonna be at such weak situation that more

10:06.855 --> 10:10.175
and more types are gonna be tempted to, to cheat

10:10.175 --> 10:11.215
and, and abuse the norms.

10:11.215 --> 10:12.735
And potentially every, you know,

10:12.735 --> 10:15.495
we have a finite type space, every type would wanna abuse.

10:16.455 --> 10:19.675
Uh, contrary in contrast to this, if you happen

10:19.675 --> 10:21.115
to have good leaders early on,

10:21.455 --> 10:23.635
and they set up strong good norms,

10:24.265 --> 10:25.755
then the norm gets stronger,

10:25.755 --> 10:28.555
and the moral, the norms are evolving endogenously, and

10:28.795 --> 10:30.795
therefore, you, you know,

10:30.795 --> 10:34.435
even if you get a relatively bad leader type, the

10:35.135 --> 10:36.955
the leader will still wanna respect the

10:36.955 --> 10:38.195
norms because they're very strong.

10:38.545 --> 10:41.715
Okay? So that's, that's what can lead to these two,

10:42.275 --> 10:43.515
outcomes that I was mentioning earlier,

10:43.515 --> 10:46.835
comparing Argentina versus the US where, you know,

10:46.835 --> 10:49.395
in the long run, in one situation, you got a sequence

10:49.395 --> 10:51.875
of good draws, and then you have strong norms,

10:52.015 --> 10:54.875
versus in the other one, a sequence of bad draws leads

10:54.875 --> 10:57.515
to eventually a bad equilibrium where, where there's abuse,

10:57.655 --> 10:58.955
of the norms all the time.

10:59.645 --> 11:01.855
Okay? There's also possibilities

11:01.875 --> 11:03.015
for other long run outcomes.

11:03.015 --> 11:05.295
We characterize that as well in the paper,

11:05.475 --> 11:07.855
but it's a combination of both the distribution of,

11:07.855 --> 11:10.455
of the leader types that you have, and the,

11:10.455 --> 11:13.095
and the evolution of how norms evolve over time,

11:13.145 --> 11:14.535
which is endogenous as well.

11:15.315 --> 11:17.995
Um, and then one,

11:18.095 --> 11:21.715
one additional thing we allow in the paper, is

11:21.905 --> 11:24.945
that the set of types themselves could evolve over time.

11:25.655 --> 11:30.145
Okay? So, what, what we mean by this is, for example,

11:30.405 --> 11:33.585
if you have a very corrupt political system, okay?

11:34.405 --> 11:38.465
Now, it actually could become very hard to make it,

11:38.565 --> 11:40.105
to become a viable candidate

11:40.725 --> 11:42.545
unless you're willing to, you know,

11:42.545 --> 11:44.345
grease the hands of others, right?

11:44.345 --> 11:47.185
In the political apparatus to support you, right?

11:47.245 --> 11:49.985
So, so, and that, that can lead to, you know,

11:49.985 --> 11:53.445
in some sense, a, a faster convergence to, to bad outcomes

11:53.625 --> 11:55.245
or, or potentially to good outcomes, right?

11:55.245 --> 11:57.725
The, the converse is this would be, you know,

11:57.915 --> 11:59.005
when there's a lot of scrutiny,

11:59.115 --> 12:01.525
when the norms are very high, you know, as soon

12:01.525 --> 12:04.485
as you do something badly, even if, when you're lower levels

12:04.585 --> 12:07.565
of the political party that say you would get, you know,

12:07.565 --> 12:09.045
you would, you would get, get kicked out

12:09.045 --> 12:10.205
of the party or something like that.

12:10.755 --> 12:13.645
Okay? So, so we also allow for this sort of endogeneity of,

12:13.645 --> 12:14.925
of the, of the type space,

12:15.175 --> 12:17.085
Given that our model is very general.

12:17.665 --> 12:21.405
So actually our model also speaks to, corporations,

12:21.625 --> 12:23.205
in, in addition to politic.

12:24.065 --> 12:25.965
So if we take a look at corporations, you know,

12:25.965 --> 12:27.805
young corporations actually would like

12:27.805 --> 12:30.565
to invest in leadership more, you know,

12:30.565 --> 12:33.325
because CEOs, you know, recognize the importance

12:33.905 --> 12:36.125
of the early leadership, to,

12:36.125 --> 12:39.165
to shape the wrong run outcomes of the corporation.

12:39.465 --> 12:42.245
The paper on one hand is, I would say,

12:42.465 --> 12:44.285
abstract in general.

12:44.465 --> 12:47.285
But, one of the things we, we try to do in, in one

12:47.285 --> 12:51.925
of the last actions is to try to tie it down to, to several,

12:52.305 --> 12:55.165
you know, more concrete things, that,

12:55.195 --> 12:57.725
that indeed have, policy implications.

12:58.665 --> 13:01.085
Um, so here to some extent, I would say we, we,

13:01.105 --> 13:03.125
we are not reinventing the wheel,

13:03.385 --> 13:07.445
but we show how, how to think about

13:08.185 --> 13:11.205
the role of, for example, of a free media, okay?

13:11.225 --> 13:14.005
So in this, in this, parameterization we have

13:14.185 --> 13:16.685
of the probability of getting replaced, right?

13:16.785 --> 13:20.325
So that can be a stand in for, for the role of, for example,

13:20.345 --> 13:22.165
of, of free media or,

13:22.305 --> 13:26.645
or a, a, you know, a level competitive, political field

13:26.645 --> 13:29.645
where, where there's a lot of political competition,

13:29.905 --> 13:31.405
and, and things of that sort.

13:31.745 --> 13:33.405
Um, one area,

13:33.455 --> 13:36.045
which I would say is a little bit more surprising in terms

13:36.045 --> 13:39.165
of our results is in terms of term limits.

13:39.465 --> 13:41.845
So although I said in the, in the, in the baseline model,

13:41.845 --> 13:43.725
we, we have a,

13:44.565 --> 13:46.245
a stationary mall with, with no term limits.

13:46.345 --> 13:47.525
We, we show an extension

13:47.525 --> 13:49.085
what would happen if you have term limits.

13:50.025 --> 13:52.165
So term limits introduce something interesting,

13:52.165 --> 13:56.045
because on one hand you have this, sort of incentive in,

13:56.045 --> 13:58.245
in the last period you're in office, you say, well,

13:58.465 --> 13:59.965
you know, and sometimes I don't care if the,

14:00.085 --> 14:02.605
I will be replaced tomorrow if I do something improper,

14:02.605 --> 14:04.205
because I'm gonna be outta power anyway.

14:04.705 --> 14:06.165
So you sort of have this, this,

14:06.165 --> 14:08.445
this final period in which you, you're tempted to,

14:08.445 --> 14:09.525
to cheat in some sense.

14:10.265 --> 14:12.725
Now, one would think that that would lead

14:13.025 --> 14:16.005
to better policies be to having longer term limits, so

14:16.005 --> 14:18.605
that this effect is, is, is sort of smaller or,

14:18.625 --> 14:21.005
or, you know, affects less periods in some sense.

14:21.545 --> 14:24.765
But we show in the paper that actually, that intuition is,

14:24.825 --> 14:25.965
is not so straightforward.

14:26.745 --> 14:29.885
Uh, and for example, if we go from a two period setting

14:29.885 --> 14:32.765
where, you know, think of the, the two term limits, so,

14:32.785 --> 14:34.005
and you don't cheat in the first,

14:34.065 --> 14:35.205
and you might cheat in the second,

14:35.465 --> 14:37.365
and we now have like three periods.

14:37.365 --> 14:41.365
So, it, it's possible depending on, on,

14:41.385 --> 14:44.445
on different pa parts of the mall parameterizations,

14:44.915 --> 14:46.605
that you, you might even wanna

14:46.605 --> 14:47.765
cheat now in the first period.

14:47.765 --> 14:50.485
So you might wanna start cheating from time zero,

14:50.705 --> 14:51.925
all through your, your tenure.

14:52.705 --> 14:55.685
And, and the intuition for this has to do a little bit

14:55.875 --> 14:59.125
with the idea of, democratic backsliding and,

14:59.425 --> 15:02.045
and the fact that if you erode the institutions right,

15:02.045 --> 15:04.445
to some extent, from, from the perspective of,

15:04.465 --> 15:05.965
of a bad leader, right?

15:06.365 --> 15:08.765
Eroding institutions is a way to be able

15:08.765 --> 15:11.325
to profit more from office in the future.

15:12.025 --> 15:14.245
And so, if, if the leader has a longer horizon

15:14.425 --> 15:16.325
and is not a good leader type,

15:16.955 --> 15:20.485
then having a longer horizon makes it more worthy for

15:20.485 --> 15:22.405
that leader to start cheating early

15:22.405 --> 15:24.845
because it's gonna be able to reap those, those benefits

15:24.845 --> 15:26.685
of eroding institutions early on.

15:27.665 --> 15:31.935
Um, so, so we show that that intuition is, is, is subtle

15:32.235 --> 15:34.735
and, and so longer term limits are,

15:34.755 --> 15:35.895
are, could be problematic.

15:36.315 --> 15:40.035
And, so it's important to have term limits

15:40.455 --> 15:41.915
To compliment, the role

15:42.135 --> 15:44.115
or independence of media as we reset.

15:44.455 --> 15:47.635
So in our model, the replacement probability

15:47.735 --> 15:49.555
of a leader plays a very important role.

15:50.375 --> 15:53.235
So, you know, whether the reader stays in power

15:53.375 --> 15:56.235
or not, depends on what he or she did currently,

15:56.455 --> 16:00.675
and the level of, norms, institutional norms, okay?

16:01.615 --> 16:04.635
So, given that our model is actually very simple, you know,

16:04.635 --> 16:09.545
we can actually interpret how the replacement probability

16:09.605 --> 16:12.705
of reader it's shaped, you know, can be explained,

16:12.705 --> 16:14.505
for instance, as the independence media.

16:14.605 --> 16:17.145
So the reader wants to undermine a sensor,

16:17.325 --> 16:18.345
the independence media,

16:18.845 --> 16:20.305
or whether the reader wants

16:20.305 --> 16:23.745
to suppress the political competition, you know, by,

16:24.165 --> 16:27.305
you know, also, changing the liberal playing field

16:27.445 --> 16:31.385
by changing the supervision of election, so forth, so forth.

16:31.725 --> 16:33.745
We don't have, implications in terms

16:33.745 --> 16:36.305
of policy implications, like in,

16:36.305 --> 16:37.385
in some sense we're known.

16:37.445 --> 16:41.545
But, we hope that, our models like,

16:42.105 --> 16:45.185
attracts attention or attracts, an allows

16:45.325 --> 16:48.945
for extensions, you know, in political science and economics

16:48.945 --> 16:50.025
and social science in general.

00:00 - The study is, is, is motivated by, by trying to understand how, countries or, or more generally, institutions that have very similar, sort of legal or, or, or, or written rules, can have very different outcomes. to some extent.

00:21 - One of the motivating examples is, is Argentina versus the the us. I happen to be Argentinian and, the Argentine constitution was actually modeled after the, the US Constitution. And, needless to say, our, our outcomes have been somewhat, dissimilar. and so trying to understand that and how, you know, is it that the institutions are good? Is it that the US was sort of lucky

00:46 - and had, very good, people sitting in the chair, so to speak, or, or, or is it something else? And, and, you know, that's sort of a little bit the motivation. And, and I guess these days, although that was not the original motivation, but with, with with the Trump, you know, pushing on the institutions, pushing on the norms, it sort of became a little bit more relevant again, How the behavior of the leader shapes institution

01:12 - is actually very recognized. So during the course of our study, we find, we found that indeed, George Washington even mentioned that what he does is gonna be, become a precedent in the future. So he was very careful about what he did indeed. So the idea that, again, you know, the, the behavior of the reader, current reader affects the behaviors of the future reader is recognizing political science, the role, and, and more general in the social science in,

01:42 - in, in, in general. And more recently, as we mentioned, you know, there's a phenomenon called the Democratic back riding. So if you take a look at like Chave in Venezuela or OB in Hung, you know, OB in Hungary and this intelligence, so forth, so forth, political reader basically tries to undermine some democratic practices gradually piece by piece, you know, for instance, by changing the, the competition of Supreme Court judges

02:13 - or changing the oversight of electrical rules and so forth, so forth. So our model actually indeed, tries to shed right on how the behavior of of the political leader, you know, gradually undermines institutional norms and our focus on like unwritten news, you know, because if we take a look at the current phenomena, oftentimes the behavior of the readers are more like, informal in the sense that the readers are elected under seemingly fair elections.

02:47 - You know, seemingly they, respect the constitutions, but they kind of changed the political norms, constitutional norms. In fact, you know, we try to, study the attack of the program, using the, like, economic and game theory. And one of the advantage of advantages of this is, you know, actually our paper speaks also to corporations. So, you know, we can also talk about how, you know, leader in corporation like tries to capture

03:19 - at the corporate board, You know, a lot of the literature and, and, and, and sort of more on the legal scholars and, and, and, and political science, although they've described some of this phenomenon, it's more, you know, to some extent verbal. It's more descriptive than, than than analytic, sort of to speak. And so part of our contribution is, is to write a, a formal model that that tries to,

03:46 - to capture this phenomena. And, and by having a formal model and, and, you know, putting in different elements to the model, like, you know, probability of reelection, you know, the benefits of being in power and, and, and different parts to it, it allows us also to, to do sort of comparative statics analysis or think about policy, think about what would happen if we change this? What, how would the leaders optimal response change

04:12 - and, and, and so on. So having this, you know, it's not that the idea is necessarily new to some extent, this is the, it is more the, the contribution and the modeling of, of this, using, again, more sort of economic techniques. And that allows us then to, to speak about changes to the environment or, or what we might expect In terms of the, the law of like readers readership, on like democratic practices, for instance.

04:41 - You know, we use, varieties of democracy v them to provide some, you know, empiric support of our theory. So like, for instance, like we measure, you know, how, again, how political readers undermine institutions by, you know, measuring, the index of like electric electoral democracy. So we can clearly see that how, you know, like, political readers, in fact, like for instance, Chave in Venezuela, under undermined the democracy, so forth, of course.

05:08 - And, our actually model also speaks to how difficult it is to restore democratic practices, you know, once the like, the institutional norms, became lower. So in fact, as you know, as measured by, or like as kind of supported by the poor record, of Arab Spring. So we, we, we measured, like demo, like the index of electrical democracy in those like Arab spring countries that like EMN, like Egypt and so forth, so forth.

05:44 - So we clearly see that, you know, even if after like revolutions, the in index of democratic in index, increased, but after some years in the index became kind of lower again. So we also contrast, as Willie said, the US versus Argentina. So we, you know, the, the, the, the data data set actually allows us to take a look at the wrong run outcomes.

06:11 - So we actually in fact compare, some like, for instance, judicial constraints on executives between the US and Argentina. The main results have have sort of two parts. the first part is, you know, giving the, the level of the institutions and, the, the type of the leader, which, which plays a role in our model. So the leader can be more honest or more dishonest or more patriotic.

06:40 - There's different ways to interpret that. but so the first part is we see like, given what the institutions look like and, and the type of the leader, what's the optimal policy for the leader? Where basically what we mean by that is does the leader have incentives to respect institutions or to abuse these institutions? Okay? And so the first, result is that we show that, when we have a stationary mall, by that I mean

07:05 - that there's no sort of explicit term limit. So it's more like a constitutional system, if you will. then the optimal action for the leader is gonna be constant over time. And so that simplifies a lot of the analysis and, and allows us to, to characterize, a threshold level of, of on the type space. So given the norm level. So if the norms, for example, are very strong, only relatively high, you know, sorry,

07:33 - only relatively very dishonest types would be still tempted to cheat. Whereas if the norm levels are, are weaker, you know, more types would be tempted to take abuse of those norms. Okay? That relates a little bit to the points Atosh was making earlier, how difficult it is to restore democracy once you come from a very bad system. And so that's the first part. And then the second part is looking at using that as to try to understand what's the long run,

08:01 - properties of the system. So we're basically, you have a leader in place, the leaders taking their optimal action after some time, that leader will get replaced. So replacement is endogenous, it depends both on the, on the level of the norms and on the leader type. Okay? So the idea here is that if you have, stronger norms and, and the leader, behaves well, okay, he's less likely to some extent to be replaced

08:33 - vis-a-vis the situation where the norms are very strong and the leader behaves badly, okay? So there's like, when the norms are strong, it's like there's more scrutiny in some sense. So, so that the leader is more likely to be re replaced if he behaves or she behaves badly. Conversely, when, when the norms are weak, right? now misbehaving or abusing the norms has less punishment from a, from an electoral point of view,

08:58 - you're less likely to be replaced. And actually, it's even possible. We allow in them all the possibility that actually abusing the norms helps you remain power, okay? So that helps capture some of these democratic backsliding and autocratic regimes, right? So, so once the norms are very weak, by abusing them, it, it makes it very unlikely you get replaced, okay? So with that, which is an important building blog of them all, we, we see

09:22 - what would happen in the long run once, you know, leaders are endogenously being replaced. And, and basically what we show is that, once, once the norms are endogenous, so this is the, the importance of the endogeneity of the norms. You, you can have the possibility of this situation where starting from the same level of written rules, okay? The, if you happen to have bad early leadership, okay, what's gonna happen is that bad early leadership is gonna abuse the norms,

09:57 - so it's gonna weaken the norms. And then even if that leadership gets replaced later on, the norms are gonna be at such weak situation that more and more types are gonna be tempted to, to cheat and, and abuse the norms. And potentially every, you know, we have a finite type space, every type would wanna abuse. contrary in contrast to this, if you happen to have good leaders early on, and they set up strong good norms,

10:24 - then the norm gets stronger, and the moral, the norms are evolving endogenously, and therefore, you, you know, even if you get a relatively bad leader type, the the leader will still wanna respect the norms because they're very strong. Okay? So that's, that's what can lead to these two, outcomes that I was mentioning earlier, comparing Argentina versus the US where, you know, in the long run, in one situation, you got a sequence

10:49 - of good draws, and then you have strong norms, versus in the other one, a sequence of bad draws leads to eventually a bad equilibrium where, where there's abuse, of the norms all the time. Okay? There's also possibilities for other long run outcomes. We characterize that as well in the paper, but it's a combination of both the distribution of, of the leader types that you have, and the, and the evolution of how norms evolve over time,

11:13 - which is endogenous as well. and then one, one additional thing we allow in the paper, is that the set of types themselves could evolve over time. Okay? So, what, what we mean by this is, for example, if you have a very corrupt political system, okay? Now, it actually could become very hard to make it, to become a viable candidate unless you're willing to, you know, grease the hands of others, right?

11:44 - In the political apparatus to support you, right? So, so, and that, that can lead to, you know, in some sense, a, a faster convergence to, to bad outcomes or, or potentially to good outcomes, right? The, the converse is this would be, you know, when there's a lot of scrutiny, when the norms are very high, you know, as soon as you do something badly, even if, when you're lower levels of the political party that say you would get, you know, you would, you would get, get kicked out

12:09 - of the party or something like that. Okay? So, so we also allow for this sort of endogeneity of, of the, of the type space, Given that our model is very general. So actually our model also speaks to, corporations, in, in addition to politic. So if we take a look at corporations, you know, young corporations actually would like to invest in leadership more, you know, because CEOs, you know, recognize the importance

12:33 - of the early leadership, to, to shape the wrong run outcomes of the corporation. The paper on one hand is, I would say, abstract in general. But, one of the things we, we try to do in, in one of the last actions is to try to tie it down to, to several, you know, more concrete things, that, that indeed have, policy implications. so here to some extent, I would say we, we, we are not reinventing the wheel,

13:03 - but we show how, how to think about the role of, for example, of a free media, okay? So in this, in this, parameterization we have of the probability of getting replaced, right? So that can be a stand in for, for the role of, for example, of, of free media or, or a, a, you know, a level competitive, political field where, where there's a lot of political competition, and, and things of that sort. one area,

13:33 - which I would say is a little bit more surprising in terms of our results is in terms of term limits. So although I said in the, in the, in the baseline model, we, we have a, a stationary mall with, with no term limits. We, we show an extension what would happen if you have term limits. So term limits introduce something interesting, because on one hand you have this, sort of incentive in, in the last period you're in office, you say, well,

13:58 - you know, and sometimes I don't care if the, I will be replaced tomorrow if I do something improper, because I'm gonna be outta power anyway. So you sort of have this, this, this final period in which you, you're tempted to, to cheat in some sense. Now, one would think that that would lead to better policies be to having longer term limits, so that this effect is, is, is sort of smaller or, or, you know, affects less periods in some sense.

14:21 - But we show in the paper that actually, that intuition is, is not so straightforward. and for example, if we go from a two period setting where, you know, think of the, the two term limits, so, and you don't cheat in the first, and you might cheat in the second, and we now have like three periods. So, it, it's possible depending on, on, on different pa parts of the mall parameterizations, that you, you might even wanna

14:46 - cheat now in the first period. So you might wanna start cheating from time zero, all through your, your tenure. And, and the intuition for this has to do a little bit with the idea of, democratic backsliding and, and the fact that if you erode the institutions right, to some extent, from, from the perspective of, of a bad leader, right? Eroding institutions is a way to be able to profit more from office in the future.

15:12 - And so, if, if the leader has a longer horizon and is not a good leader type, then having a longer horizon makes it more worthy for that leader to start cheating early because it's gonna be able to reap those, those benefits of eroding institutions early on. so, so we show that that intuition is, is, is subtle and, and so longer term limits are, are, could be problematic. And, so it's important to have term limits

15:40 - To compliment, the role or independence of media as we reset. So in our model, the replacement probability of a leader plays a very important role. So, you know, whether the reader stays in power or not, depends on what he or she did currently, and the level of, norms, institutional norms, okay? So, given that our model is actually very simple, you know, we can actually interpret how the replacement probability of reader it's shaped, you know, can be explained,

16:12 - for instance, as the independence media. So the reader wants to undermine a sensor, the independence media, or whether the reader wants to suppress the political competition, you know, by, you know, also, changing the liberal playing field by changing the supervision of election, so forth, so forth. We don't have, implications in terms of policy implications, like in, in some sense we're known.

16:37 - But, we hope that, our models like, attracts attention or attracts, an allows for extensions, you know, in political science and economics and social science in general.

Faculti AI can make mistakes