How Reducing Employment Protection Affects Job Creation and Job Destruction

Key Question Being Explored in This Insight:

A reduction in employment protection is expected to encourage both job creation and job destruction. To estimate these two effects independently, Francesco Maria Esposito exploits a change in the Italian employment protection legislation, approved in 2012, which impacted firms differently depending on their size around the 15 employees cut-off.

View the page of: Francesco Maria Esposito

Institution: University of Birmingham

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 Birmingham . Reflect on how the institution's academic focus and research partnerships may shape the questions being explored.

Further Reading ▼

Read the Study

Image courtesy of interviewee. July 11, 2024

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

This video discusses the impact of employment protection legislation (EPL) on job creation and destruction. EPL consists of regulations that prevent unjustified dismissals by employers, aiming to protect workers. The author investigates how changes in EPL—specifically a reduction in employment protection—affect firms' behaviors regarding hiring (job creation) and layoffs (job destruction).

Previous literature suggests that reducing employment protection would lead to both increased job creation and job destruction because firms would be more willing to hire if they could easily let go of workers when necessary. However, the author finds through a study of Italian firms that after 2012, when EPL was relaxed, there was a notable increase in job destruction, but no corresponding increase in job creation for firms with around 15 employees—contrary to predictions.

This video study uses a difference-in-differences approach with treatment groups that include firms above the cutoff (15 employees) and those exactly at the cutoff, comparing them to a control group of smaller firms. Findings indicate that larger firms tended to fire more employees post-reform, while those at the cutoff did not increase hiring despite expectations.

The study highlights the importance of considering policy uncertainty when assessing the effects of labor laws, noting that firms might act conservatively due to fears of potential future policy reversals. The author emphasizes the need for stability in employment policies to ensure firms can make confident hiring decisions without the anxiety of forthcoming regulatory changes.

Overall, the text suggests that the real-world effects of labor regulations can diverge from theoretical expectations due to factors like policy uncertainty, which can complicate the objectives of employment protection reforms.

Language:

Export as PDF

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

00:00.000 --> 00:02.805
Employment protection legislation is the set

00:02.825 --> 00:05.605
of all regulations that are in place in order

00:05.665 --> 00:09.365
to prevent illegitimate dismissals of employers.

00:09.945 --> 00:11.325
So basically, when we want

00:11.325 --> 00:14.485
to protect workers from these illegitimate dismissals,

00:14.485 --> 00:17.445
governments put in place sort of regulations

00:17.555 --> 00:19.325
that include terms and

00:19.745 --> 00:23.415
and conditions for the worker to be, to be,

00:23.435 --> 00:24.575
to be basically fired.

00:24.795 --> 00:27.375
The, and this regulation must be respected somehow.

00:28.235 --> 00:30.095
Um, job creation

00:30.095 --> 00:33.175
and job destruction, instead referred to some measures

00:33.395 --> 00:36.295
of increasing or decreasing the firm side.

00:36.515 --> 00:38.415
So it's a firm level measure.

00:39.195 --> 00:40.215
So the,

00:41.005 --> 00:44.895
what I'm looking in this study is whether there is an effect

00:44.955 --> 00:49.215
and what is the effect of employ protection legislation on,

00:49.635 --> 00:52.815
behaviors, on firms behavior on creating

00:52.915 --> 00:53.935
and destroying jobs.

00:54.315 --> 00:58.535
So basically, my question specifically ask, what is,

00:59.875 --> 01:02.015
the effect of a reduction in employment,

01:02.035 --> 01:04.375
in employment protection on,

01:05.395 --> 01:06.975
job creation and job destruction?

01:07.435 --> 01:10.735
Now, the literature usually,

01:12.485 --> 01:17.275
about that says that, dynamic model of labor demand

01:18.515 --> 01:21.925
predict that when employment protection is reduced,

01:22.475 --> 01:25.565
this will increase both job creation and job destruction.

01:25.625 --> 01:26.885
So the reason it's simple,

01:26.885 --> 01:29.565
because if I reduce employment protection,

01:29.825 --> 01:33.125
so I basically make it easier for firms to fire workers,

01:33.755 --> 01:36.605
then firms, of course, they will far more easily.

01:36.785 --> 01:37.885
So they will far more,

01:38.345 --> 01:41.485
but at the same time, they'll also be incentivized

01:41.625 --> 01:43.485
and encouraged to hire more

01:43.555 --> 01:46.205
because they know that they can freely hire

01:46.435 --> 01:48.965
because in case they will need to fire these workers,

01:49.115 --> 01:50.245
they can easily do that.

01:50.825 --> 01:54.365
So there is basically a double effect, a two margins effect

01:54.625 --> 01:55.845
of a reduction

01:55.945 --> 01:58.485
or in general policy on employment protection.

01:59.065 --> 02:02.005
And this is the first paper that try to analyze

02:02.275 --> 02:05.925
what is the effect on these two margins separately.

02:06.225 --> 02:09.085
So that's more or less the context of this paper.

02:09.505 --> 02:12.245
Uh, it's better probably not to mention,

02:12.715 --> 02:14.085
authors in particular,

02:14.265 --> 02:17.845
but in general, the literature is concentrated

02:18.025 --> 02:20.925
and focused on firm sites, on the fact, on

02:21.185 --> 02:23.285
of employment production, on firm sites,

02:23.465 --> 02:25.605
and not on these two different margins

02:25.605 --> 02:27.325
of job creation and job destruction.

02:27.585 --> 02:30.565
So basically what they see, they simply see

02:30.595 --> 02:35.125
what they simply study, what is the effect of, a, a,

02:35.365 --> 02:39.405
a change in employ protection on the sides of the companies,

02:39.425 --> 02:42.685
but they don't know whether they have more incentive to grow

02:42.985 --> 02:46.005
or more incentive to decrease and to destroy jobs.

02:46.425 --> 02:49.765
So again, like employment protection may have this double

02:49.825 --> 02:51.325
effect, and it is really important

02:51.505 --> 02:56.325
for policy evaluation reasons to try to understand what is,

02:56.545 --> 02:58.685
the effect separately on

02:58.685 --> 03:00.325
job creation and job destruction.

03:00.585 --> 03:02.205
So the first thing is that most

03:02.205 --> 03:04.405
of the retailer refer to firm sites.

03:05.745 --> 03:08.525
The second thing is about the sample sites.

03:09.345 --> 03:14.005
So as I will explain later, I use, the entire population

03:14.005 --> 03:15.405
of Italian firms here.

03:15.545 --> 03:17.845
So I have administered, I have access

03:17.845 --> 03:19.365
to administrative dataset.

03:19.665 --> 03:22.125
So I'm, I'm actually able to see what happens

03:22.265 --> 03:25.565
to every company in a country, not just to a

03:26.815 --> 03:29.885
small subset of the, of the firms.

03:30.585 --> 03:33.245
One thing is saying, what is the effect of an introduction

03:33.245 --> 03:34.525
of employment protection?

03:34.845 --> 03:37.205
A different thing is saying, what is the effect

03:37.205 --> 03:38.965
of a reduction in employment protection?

03:38.965 --> 03:41.885
So there are some papers that study what is

03:42.025 --> 03:44.965
and analyze what is the effect of an introduction

03:44.965 --> 03:46.045
of employment protection,

03:46.345 --> 03:50.165
but that's not exactly symmetric in terms

03:50.265 --> 03:51.565
of, of the effect.

03:51.625 --> 03:55.085
So we may expect for some reason

03:55.115 --> 03:58.485
that if we increment the protection, there is something

03:58.485 --> 04:01.885
that happens, and if we reduce protection,

04:01.885 --> 04:04.365
there may be something else that actually happens.

04:05.135 --> 04:07.605
Until 2012 in Italy,

04:07.615 --> 04:10.405
there was a specific employment protection legislation

04:10.405 --> 04:12.845
that was different according

04:12.945 --> 04:15.445
to the firm sides of companies.

04:15.825 --> 04:20.005
So if you were up until 15 worker sites,

04:20.265 --> 04:23.445
so if you basically have small sites,

04:23.595 --> 04:26.085
then like a really light kind

04:26.085 --> 04:29.685
of employment protection was in place for firms instead

04:30.115 --> 04:34.685
that, are larger, were larger than 50 workers.

04:35.225 --> 04:36.725
Uh, before 2012,

04:36.775 --> 04:40.365
there was a much stricter employment protection,

04:40.675 --> 04:44.925
much stricter in the sense that all the,

04:45.065 --> 04:49.365
all the reasons allowed, for workers to,

04:49.385 --> 04:52.885
for employers to fire workers were really limited.

04:53.705 --> 04:57.125
And you needed to prove actually that there was a, a,

04:57.365 --> 05:01.965
a serious, issue with that worker in order to fire,

05:02.585 --> 05:03.845
to fire him or her.

05:04.385 --> 05:06.645
So, what happens is that

05:06.775 --> 05:10.965
after 2012, this regulation was changed in the sense

05:10.965 --> 05:15.645
that also it was, it, it was made much easier for firms

05:15.935 --> 05:18.725
above 15 workers as well to fire workers.

05:19.425 --> 05:20.565
Why this is the case.

05:20.875 --> 05:23.645
This is the case because a new reason

05:23.825 --> 05:26.005
for firing workers was introduced,

05:26.615 --> 05:28.605
which was the economic reason.

05:28.825 --> 05:33.125
So that, that was the, the biggest, introduction.

05:33.265 --> 05:37.125
So basically for firms above this cutoff, it was much easier

05:37.255 --> 05:39.925
after two, 2012 to fire workers

05:40.115 --> 05:42.685
because they could use the economic reason

05:42.825 --> 05:44.005
in order to firework.

05:44.145 --> 05:48.925
So if the firm was facing some cha some economic challenges,

05:48.925 --> 05:51.965
then Europe could use these,

05:52.385 --> 05:53.925
clouds in order to find work.

05:54.105 --> 05:56.805
So of course, the introduction of these

05:56.985 --> 05:58.765
For firms above 15 worker

05:59.435 --> 06:03.845
made substantial change in the behaviors of firms.

06:04.625 --> 06:05.845
So what happened is that

06:05.985 --> 06:08.085
before the, the 2012,

06:08.135 --> 06:12.445
there was a cutoff at 15 workers in firms size distribution.

06:12.745 --> 06:16.005
So for firms below the cutoff, it's easy to fire

06:16.185 --> 06:17.765
for firm above the cutoff.

06:18.075 --> 06:19.965
It's not that easy to fire

06:20.295 --> 06:24.965
after 2012, we start for firms both below and aft, below

06:25.025 --> 06:27.925
and, above the cutoff, it's easier to fire.

06:28.265 --> 06:31.685
So basically, firms above the cutoff have experienced

06:31.855 --> 06:33.205
after 2012

06:34.695 --> 06:38.515
and an easiness of, firing workers.

06:39.495 --> 06:43.305
So I wanna mention something which is really important

06:43.365 --> 06:46.385
for this study is that the cutoff is sharp in the sense

06:46.385 --> 06:50.825
that up until 15 workers included is much easier

06:50.965 --> 06:55.885
to fire workers before 2012, strictly

06:56.095 --> 06:57.605
above 15 workers in.

06:57.795 --> 06:59.965
Instead, it was much, much harder.

07:00.345 --> 07:02.085
Now why I'm saying like this,

07:02.085 --> 07:06.005
because we expect that firms strictly above 15 workers

07:06.495 --> 07:09.685
after the policy, since now it's easier to fire,

07:10.075 --> 07:14.665
they will start firing the case in which

07:14.775 --> 07:18.625
instead we will see whether there is also an effect on the

07:18.625 --> 07:19.705
hiring margins

07:19.705 --> 07:24.145
or on the job creation Margin is on firms at 15 workers

07:24.145 --> 07:27.945
because those firms were able to fire workers

07:28.655 --> 07:31.465
without issues before their form.

07:31.605 --> 07:34.385
But, but after the reform for them, now,

07:34.415 --> 07:36.905
they should have an incentive to harm new workers

07:37.135 --> 07:39.625
because they are at 15, probably

07:39.625 --> 07:42.985
before the reform, they were afraid of surpassing the cutoff

07:43.055 --> 07:46.345
because once you enter in the 15,

07:46.345 --> 07:50.225
above the 15 firm sites regulation, then it's hard to,

07:50.225 --> 07:51.345
for any of your workers.

07:51.805 --> 07:53.065
So before the, the

07:53.645 --> 07:57.585
before the reform, you don't wanna, absolutely,

07:58.615 --> 08:02.105
surpass the cutoff, but after the reform study you want.

08:02.485 --> 08:05.745
So that's the double margin effect that they evaluate.

08:05.745 --> 08:08.305
So basically I want to evaluate what is the fact

08:08.305 --> 08:10.105
of firms above the cutoff.

08:10.135 --> 08:13.745
They should be more incentivized to fire workers as opposed

08:13.845 --> 08:18.625
to, hiring new ones, while instead firms exactly 15,

08:18.895 --> 08:22.025
they should have, been encouraged by these reform

08:22.205 --> 08:26.505
to hire new workers, while instead they should not,

08:26.565 --> 08:28.745
see any effects in terms of firing

08:28.775 --> 08:30.105
because they were able

08:30.105 --> 08:32.345
to fire workers also before the reform.

08:32.725 --> 08:36.945
So basically this reform made us a big change in the firm

08:37.085 --> 08:39.345
site's distribution and in this cutoff.

08:39.725 --> 08:42.785
So I'm using a difference in difference design

08:42.895 --> 08:44.545
with two different treatment groups.

08:45.165 --> 08:47.865
One which is made of firms above the cutoff,

08:48.165 --> 08:51.385
one other which is made of firms exactly at the cutoff

08:51.725 --> 08:53.865
and one control group, which is made

08:53.865 --> 08:56.485
by firms which are below the cutoff.

08:56.555 --> 09:00.325
Because remember that this firm do not see any effect

09:00.545 --> 09:03.805
of the reform on them because they were able to hire

09:03.865 --> 09:07.525
and fire before the reform as well, without any issue.

09:07.785 --> 09:09.485
So that's the perfect control group.

09:09.825 --> 09:12.205
So what they do, I do a difference in difference

09:12.205 --> 09:14.205
between these two treatment groups

09:14.385 --> 09:17.285
and the control group respectively, to estimate

09:17.285 --> 09:20.605
what is the effect on job destruction and job creation.

09:20.615 --> 09:24.365
Again, job destruction using the treatment made of firms

09:24.655 --> 09:27.285
above, strictly above the cutoff, while instead,

09:27.485 --> 09:31.365
I evaluate the fact on job creation using firms which are

09:31.395 --> 09:33.925
exactly at the cutoff, surprisingly,

09:33.985 --> 09:35.725
and I'm saying surprisingly

09:35.725 --> 09:39.605
because again, dynamic law, dynamic model

09:39.625 --> 09:41.485
of labor demand predict

09:41.675 --> 09:44.525
that a reduction in employment protection would increase

09:44.555 --> 09:46.445
both job creation and job destruction.

09:46.505 --> 09:48.805
For the reasons that I've explained at the beginning

09:48.805 --> 09:52.885
of this interview, what I find is that as a response

09:52.945 --> 09:56.925
to this policy, there has been a change in, in, for firms

09:57.095 --> 09:58.805
above the cutoff because they have

09:58.805 --> 10:00.205
started to fire more workers.

10:00.785 --> 10:01.805
But at the same time,

10:01.935 --> 10:05.125
there was no effect on firms at the cutoff in the sense

10:05.125 --> 10:09.445
that firms with exactly 15 workers didn't start hiring

10:09.785 --> 10:10.885
and creating new jobs.

10:11.345 --> 10:13.645
So there was an effect on job destruction,

10:13.705 --> 10:15.565
but no effect on job creation.

10:15.945 --> 10:17.245
And this is surprising

10:17.245 --> 10:19.925
and striking also in light of the motivations.

10:20.505 --> 10:24.205
Uh, if we look at the motivation that this policy had,

10:24.205 --> 10:28.405
at the beginning because this, when this policy was made,

10:28.545 --> 10:31.565
was made because there was a big discussion in Italy

10:32.095 --> 10:36.125
about firms being reluctant in hiring new workers

10:36.195 --> 10:37.605
because of this policy.

10:38.105 --> 10:42.005
So politicians were saying, look, if we remove this policy,

10:42.305 --> 10:46.925
you will see that, firms will hire new workers

10:47.355 --> 10:50.205
because they will know that they're free

10:50.225 --> 10:52.125
to fire them if they wanna to do so.

10:52.545 --> 10:54.805
But actually what they find in this study is

10:54.805 --> 10:56.405
that this is not what is happening.

10:56.865 --> 10:59.645
So always be careful when there are some policy

10:59.705 --> 11:02.045
and some expectation from theoretical model,

11:02.435 --> 11:04.325
because there may be something,

11:04.415 --> 11:07.045
there may be some assumptions of,

11:07.045 --> 11:10.645
theoretical models which are not verified in reality,

11:10.945 --> 11:14.565
and that's why we may observe different behaviors

11:14.585 --> 11:16.685
of firms in response of this policy.

11:17.145 --> 11:20.925
So in this particular case, what is the insight

11:21.035 --> 11:22.085
that we have?

11:22.395 --> 11:23.645
What is theoretical?

11:24.275 --> 11:27.725
What is the theoretical, assumptions that,

11:27.915 --> 11:31.325
that models are missing from the general picture?

11:31.905 --> 11:35.365
In my opinion, what they are missing is the uncertainties,

11:35.435 --> 11:37.365
what they call policy uncertainty.

11:37.545 --> 11:42.445
So the reason why we may observe a job destruction in

11:42.445 --> 11:44.925
this setting and not of creation is

11:44.925 --> 11:48.405
because there was policy uncertainty about

11:48.915 --> 11:52.605
this same policy to come back in place in the close future

11:53.305 --> 11:57.045
is indeed there was like a, a, a hard discussion

11:57.875 --> 12:00.045
between, policymaker

12:00.145 --> 12:01.365
but also politicians

12:01.745 --> 12:05.005
to reintroduce this policy in the next years.

12:05.385 --> 12:06.565
And if you're a firm

12:06.985 --> 12:10.125
and you know that now it's easier for you to file,

12:11.065 --> 12:13.685
but you may know that the year

12:13.735 --> 12:16.565
after the same policy come back in place again,

12:16.915 --> 12:20.165
then you know that if you are someone, then you'll,

12:20.865 --> 12:25.005
you may be stalked at 16, 17 workers say,

12:25.385 --> 12:26.885
not be able to fire workers

12:26.885 --> 12:29.125
because they have reintroduced this law.

12:29.465 --> 12:31.685
So always take into account

12:31.755 --> 12:36.445
that uncertainty may play a role, in the actual,

12:36.715 --> 12:38.925
effectiveness of this kind of policy.

12:39.225 --> 12:42.565
And they provide some suggestive evidence in the paper

12:42.635 --> 12:45.325
that this is the case because they show that also

12:45.335 --> 12:50.045
after the policy, the reform was implemented, firms

12:50.375 --> 12:54.525
above the cutoff are more likely to fire the exact number

12:54.545 --> 12:58.565
of workers in order to reach exactly the cutoff

12:58.985 --> 13:03.925
as opposed to fire, different numbers of workers.

13:04.025 --> 13:07.405
So basically, if you are 16 workers before the reform,

13:07.855 --> 13:10.445
after this reform is much more likely

13:10.755 --> 13:13.525
that you file just one worker as opposed

13:13.545 --> 13:15.605
to two workers or zero workers.

13:16.105 --> 13:20.765
And if you are 17, worker company is much more likely

13:20.765 --> 13:24.365
that you file two as opposed to fir three or one.

13:24.825 --> 13:25.885
Why is this the case?

13:25.885 --> 13:28.725
Because you know that the same regulation can

13:29.555 --> 13:30.965
come back in place in the future.

13:31.465 --> 13:34.765
So you want to make sure that you are at the cutoff so

13:34.765 --> 13:36.765
that if the regulation come back in place,

13:37.065 --> 13:40.125
you are not gonna constrain by this regulation.

13:40.345 --> 13:43.925
So again, this is suggestive of the fact that,

13:44.055 --> 13:47.365
there may be some, aspect

13:47.595 --> 13:50.285
that may be relevant about uncertainty

13:50.285 --> 13:51.445
and policy uncertainty

13:51.445 --> 13:54.205
that we should consider in future research.

13:54.835 --> 13:57.645
Depending on the setting, it may be more relevant

13:57.645 --> 14:01.125
to protect workers or not protect workers.

14:01.425 --> 14:04.525
So of course, it depends on the institutional settings

14:04.945 --> 14:08.805
and also on the behaviors of the usual behaviors of firms

14:09.145 --> 14:11.765
and, and workers in specific countries.

14:12.225 --> 14:14.605
So I wouldn't,

14:17.495 --> 14:21.485
bring out so much external validity in this terms.

14:22.145 --> 14:25.005
But what is really important is that, is

14:25.275 --> 14:28.405
that once the policy is implemented,

14:28.955 --> 14:32.165
implemented once the, politicians

14:32.345 --> 14:36.445
and policymaker have chosen one direction, not

14:36.445 --> 14:40.245
to change their mind in the close future,

14:40.515 --> 14:42.245
because if they do that,

14:42.835 --> 14:46.325
they may actually have opposite effect compared to

14:46.325 --> 14:48.085
what they wanna achieve.

14:48.825 --> 14:50.765
So nothing to

14:50.765 --> 14:55.325
Say about general, feasibility

14:56.025 --> 14:58.925
of employment protection depends on the setting.

14:59.345 --> 15:03.605
But be really careful in when you implement this, these

15:04.205 --> 15:08.685
policies to make them last for a few years at least, so

15:08.925 --> 15:11.445
that firms and consumers

15:11.705 --> 15:15.645
or whoever, is gonna be, in place in

15:15.645 --> 15:19.565
that moment feel okay doing whatever they wanna do.

15:19.595 --> 15:22.245
They don't need to feel constrained with pressure,

15:22.555 --> 15:26.325
otherwise opposite effect may be achieved.

Quick Answers

The Impact of Employment Protection Legislation on Job Creation and Destruction**

00:00:00 - Employment protection legislation is the set of all regulations that are in place in order to prevent illegitimate dismissals of employers. So basically, when we want to protect workers from these illegitimate dismissals, governments put in place sort of regulations that include terms and and conditions for the worker to be, to be, to be basically fired. The, and this regulation must be respected somehow. job creation and job destruction, instead referred to some measures of increasing or decreasing the firm side. So it's a firm level measure. So the, what I'm looking in this study is whether there is an effect
00:00:49 - behaviors, on firms behavior on creating and destroying jobs. So basically, my question specifically ask, what is, the effect of a reduction in employment, in employment protection on, job creation and job destruction? Now, the literature usually, about that says that, dynamic model of labor demand predict that when employment protection is reduced, this will increase both job creation and job destruction.
00:01:26 - because if I reduce employment protection, so I basically make it easier for firms to fire workers, then firms, of course, they will far more easily. So they will far more, but at the same time, they'll also be incentivized and encouraged to hire more because they know that they can freely hire because in case they will need to fire these workers, they can easily do that. So there is basically a double effect, a two margins effect of a reduction or in general policy on employment protection. And this is the first paper that try to analyze what is the effect on these two margins separately.
00:02:09 - it's better probably not to mention, authors in particular, but in general, the literature is concentrated and focused on firm sites, on the fact, on of employment production, on firm sites, and not on these two different margins of job creation and job destruction. So basically what they see, they simply see what they simply study, what is the effect of, a, a, a change in employ protection on the sides of the companies, but they don't know whether they have more incentive to grow or more incentive to decrease and to destroy jobs.
00:02:49 - effect, and it is really important for policy evaluation reasons to try to understand what is, the effect separately on job creation and job destruction. So the first thing is that most of the retailer refer to firm sites. The second thing is about the sample sites. So as I will explain later, I use, the entire population of Italian firms here. So I have administered, I have access to administrative dataset. So I'm, I'm actually able to see what happens to every company in a country, not just to a
00:03:30 - One thing is saying, what is the effect of an introduction of employment protection? A different thing is saying, what is the effect of a reduction in employment protection? So there are some papers that study what is and analyze what is the effect of an introduction of employment protection, but that's not exactly symmetric in terms of, of the effect. So we may expect for some reason that if we increment the protection, there is something that happens, and if we reduce protection, there may be something else that actually happens. Until 2012 in Italy,

"Changes in Employment Protection Legislation: Impacts on Firm Behavior and Job Security"

00:04:10 - that was different according to the firm sides of companies. So if you were up until 15 worker sites, so if you basically have small sites, then like a really light kind of employment protection was in place for firms instead that, are larger, were larger than 50 workers. before 2012, there was a much stricter employment protection, much stricter in the sense that all the, all the reasons allowed, for workers to,
00:04:53 - And you needed to prove actually that there was a, a, a serious, issue with that worker in order to fire, to fire him or her. So, what happens is that after 2012, this regulation was changed in the sense that also it was, it, it was made much easier for firms above 15 workers as well to fire workers. Why this is the case. This is the case because a new reason for firing workers was introduced, which was the economic reason. So that, that was the, the biggest, introduction.
00:05:37 - after two, 2012 to fire workers because they could use the economic reason in order to firework. So if the firm was facing some cha some economic challenges, then Europe could use these, clouds in order to find work. So of course, the introduction of these For firms above 15 worker made substantial change in the behaviors of firms. So what happened is that before the, the 2012, there was a cutoff at 15 workers in firms size distribution. So for firms below the cutoff, it's easy to fire
00:06:18 - It's not that easy to fire after 2012, we start for firms both below and aft, below and, above the cutoff, it's easier to fire. So basically, firms above the cutoff have experienced after 2012 and an easiness of, firing workers. So I wanna mention something which is really important for this study is that the cutoff is sharp in the sense that up until 15 workers included is much easier to fire workers before 2012, strictly above 15 workers in.
00:07:00 - Now why I'm saying like this, because we expect that firms strictly above 15 workers after the policy, since now it's easier to fire, they will start firing the case in which instead we will see whether there is also an effect on the hiring margins or on the job creation Margin is on firms at 15 workers because those firms were able to fire workers without issues before their form. But, but after the reform for them, now, they should have an incentive to harm new workers because they are at 15, probably
00:07:43 - because once you enter in the 15, above the 15 firm sites regulation, then it's hard to, for any of your workers. So before the, the before the reform, you don't wanna, absolutely, surpass the cutoff, but after the reform study you want. So that's the double margin effect that they evaluate. So basically I want to evaluate what is the fact of firms above the cutoff. They should be more incentivized to fire workers as opposed to, hiring new ones, while instead firms exactly 15, they should have, been encouraged by these reform

"Analyzing the Impact of Employment Protection Reforms on Job Creation and Destruction: A Policy Uncertainty Perspective"**

00:08:26 - see any effects in terms of firing because they were able to fire workers also before the reform. So basically this reform made us a big change in the firm site's distribution and in this cutoff. So I'm using a difference in difference design with two different treatment groups. One which is made of firms above the cutoff, one other which is made of firms exactly at the cutoff and one control group, which is made by firms which are below the cutoff. Because remember that this firm do not see any effect of the reform on them because they were able to hire
00:09:07 - So that's the perfect control group. So what they do, I do a difference in difference between these two treatment groups and the control group respectively, to estimate what is the effect on job destruction and job creation. Again, job destruction using the treatment made of firms above, strictly above the cutoff, while instead, I evaluate the fact on job creation using firms which are exactly at the cutoff, surprisingly, and I'm saying surprisingly because again, dynamic law, dynamic model of labor demand predict that a reduction in employment protection would increase
00:09:46 - For the reasons that I've explained at the beginning of this interview, what I find is that as a response to this policy, there has been a change in, in, for firms above the cutoff because they have started to fire more workers. But at the same time, there was no effect on firms at the cutoff in the sense that firms with exactly 15 workers didn't start hiring and creating new jobs. So there was an effect on job destruction, but no effect on job creation. And this is surprising and striking also in light of the motivations. if we look at the motivation that this policy had, at the beginning because this, when this policy was made,
00:10:32 - about firms being reluctant in hiring new workers because of this policy. So politicians were saying, look, if we remove this policy, you will see that, firms will hire new workers because they will know that they're free to fire them if they wanna to do so. But actually what they find in this study is that this is not what is happening. So always be careful when there are some policy and some expectation from theoretical model, because there may be something, there may be some assumptions of, theoretical models which are not verified in reality,
00:11:14 - of firms in response of this policy. So in this particular case, what is the insight that we have? What is theoretical? What is the theoretical, assumptions that, that models are missing from the general picture? In my opinion, what they are missing is the uncertainties, what they call policy uncertainty. So the reason why we may observe a job destruction in this setting and not of creation is because there was policy uncertainty about this same policy to come back in place in the close future
00:11:57 - between, policymaker but also politicians to reintroduce this policy in the next years. And if you're a firm and you know that now it's easier for you to file, but you may know that the year after the same policy come back in place again, then you know that if you are someone, then you'll, you may be stalked at 16, 17 workers say, not be able to fire workers because they have reintroduced this law. So always take into account that uncertainty may play a role, in the actual,

"Navigating Policy Uncertainty: The Impact of Employment Protection Reforms on Firm Behavior"

00:12:39 - And they provide some suggestive evidence in the paper that this is the case because they show that also after the policy, the reform was implemented, firms above the cutoff are more likely to fire the exact number of workers in order to reach exactly the cutoff as opposed to fire, different numbers of workers. So basically, if you are 16 workers before the reform, after this reform is much more likely that you file just one worker as opposed to two workers or zero workers. And if you are 17, worker company is much more likely
00:13:24 - Why is this the case? Because you know that the same regulation can come back in place in the future. So you want to make sure that you are at the cutoff so that if the regulation come back in place, you are not gonna constrain by this regulation. So again, this is suggestive of the fact that, there may be some, aspect that may be relevant about uncertainty and policy uncertainty that we should consider in future research. Depending on the setting, it may be more relevant to protect workers or not protect workers.
00:14:04 - and also on the behaviors of the usual behaviors of firms and, and workers in specific countries. So I wouldn't, bring out so much external validity in this terms. But what is really important is that, is that once the policy is implemented, implemented once the, politicians and policymaker have chosen one direction, not to change their mind in the close future, because if they do that,
00:15:22 - what they wanna achieve. So nothing to Say about general, feasibility of employment protection depends on the setting. But be really careful in when you implement this, these policies to make them last for a few years at least, so that firms and consumers or whoever, is gonna be, in place in that moment feel okay doing whatever they wanna do. They don't need to feel constrained with pressure, otherwise opposite effect may be achieved.

Faculti AI can make mistakes

Research

Policy

Analysis

Data

Originals

Explainers

Podcasts

Faculti

New on Faculti

Topic Support

How Reducing Employment Protection Affects Job Creation and Job Destruction

Media Literacy Toolkit

Critical Questions to Consider ▼

Bias and Perspective Awareness ▼

Further Reading ▼