Labor protection laws and the drain on productivity: Evidence from India

• Published date: 01 May 2020
• Date: 01 May 2020
• DOI: http://doi.org/10.1111/rode.12656
• Author: Daniel Schwab

Rev Dev Econ. 2020;24:383–401. wileyonlinelibrary.com/journal/rode

|

383

© 2020 John Wiley & Sons Ltd

1

|

INTRODUCTION

This paper explores unintended consequences of employment protection legislation (EPL), which

limits the ability of employers to fire workers. Nearly every country in the world has some form of

EPL, which can be a requirement of advance notice, a severance payment, a prohibition on firing, or

some combination of these (World Bank, 2015). One common argument in favor of these laws is that

they reduce income risk in the absence of perfectly functioning insurance markets (Pissarides, 2001).

Received: 29 June 2019

|

Revised: 24 January 2020

|

Accepted: 27 January 2020

DOI: 10.1111/rode.12656

REGULAR ARTICLE

Labor protection laws and the drain on

productivity: Evidence from India

DanielSchwab

College of the Holy Cross, 1 College St.,

Worcester, MA, 01610, USA

Correspondence

College of the Holy Cross, 1 College St.,

Worcester, MA, 01610, USA.

Email: dschwab@holycross.edu

Abstract

Employment protection legislation (EPL) is designed to

promote securityers by placing restrictions on firing, but it

generates unintended consequences. With India as a setting,

I argue that EPL shifts jobs from younger to older workers

in two ways: by discouraging the hiring of unproven young

workers and by preventing the firing of low-productivity

workers. The identification strategy is motivated by Rajan

and Zingales (1998): I assume that EPL is more binding in

those manufacturing sectors where the involuntary separa-

tion rate in other countries is high. The data show that older

workers are more likely to have formal jobs, and the effect

is strongest in high-firing sectors, which indicates that EPL

shifts jobs from young to old. Additionally, EPL reduces

plant-level total factor productivity, and this effect is seen

only in plants which are large enough to fall within the pur-

view of EPL, which provides a useful placebo test.

JEL CLASSIFICATION

O17; J63

384

|

SCHWAB

Although there may be benefits, EPL carries the risk of creating distortions by preventing firms from

firing low-productivity workers.

The setting is India's manufacturing sector, which has some of the strictest EPL in the world

(World Bank, 2008), so the distortions it creates are large and important for the Indian economy. In ad-

dition, there is considerable variation in state-level EPL, meaning that it is possible to restrict attention

to a single country, ensuring that much of the institutional environment is the same across the data.

India's main labor law is the Industrial Disputes Act (IDA), which requires large plants to obtain

government permission before the plants can fire a permanent worker. Contract workers, who are em-

ployed through contractors and not directly employed by the firm where they work, are not protected

and can be fired freely in all plants. Additionally, EPL does not apply to plants below a size threshold.

Finally, employees who are above the retirement age of 58 are also not protected, and are not included

in the sample. Permanent workers in large plants are protected by EPL, and I call them “formal,” while

informal workers are either contract workers or in a small plant.

EPL leads to an older formal workforce by distorting the hiring decision and by preventing firing.

A forward-looking firm is reluctant to hire a young worker if it will not be able to fire her if she turns

out to be unproductive. Older workers are especially “safe” in India, because is no risk of an unproduc-

tive worker staying on past retirement age. Now suppose that an employee's performance is poor. In

the absence of EPL, the firm can fire the worker and hire someone else. New hires tend to be young, so

this dismissal makes the labor force younger. On the other hand, if a firm is forbidden to fire because

of EPL, then the worker will remain with the firm, so the labor force will be older.

First I demonstrate a correlation: in states with stricter EPL, older workers are more likely to be

formal. The strength of a state's EPL is calculated by aggregating the ratings of three earlier papers.

To argue that stronger EPL causes older workers to be in the formal sector, as opposed to the results

being driven by an omitted variable or reverse causality, I consider heterogeneity across manufactur-

ing sectors. The broad strategy is to argue that some manufacturing sectors are more impacted by EPL

than others, which is motivated by Rajan and Zingales (1998). If EPL causes the formal sector to be

older, then older workers should be especially likely to be in the formal sector if they work in strict-

EPL states and in sectors where EPL often binds.

It is not possible to observe directly which sectors would have the most firing in the absence of

EPL, but US involuntary separation rates are instructive because the United States has very weak

EPL.1

There is a strong correlation between countries’ involuntary separation rates (Micco & Pagés,

2006), which provides some reassurance that separation rates are driven by technology or other char-

acteristics of the sector rather than by idiosyncratic country-specific factors.2

That is, EPL causes the

shift in jobs from young to old because older workers are especially likely to be formal in strict-EPL

states in manufacturing sectors where the involuntary separation rate in the USA is high.

A similar empirical strategy can be used to address whether EPL has a measurable effect on plant-

level total factor productivity (TFP) of large plants. I test whether TFP is lower when the state has a

high value for the interaction between strict state-level EPL and the high sector-level involuntary sepa-

ration rate. Among plants large enough to be affected by EPL, there is a negative relationship between

TFP and this interaction term. Smaller plants provide a useful placebo test, and there is no relationship

between TFP and this interaction for plants that are too small to be affected by EPL, which suggests

that EPL does in fact reduce TFP.

A plausible explanation is that EPL reduces TFP because it distorts the age profile of workers,

but I do not address the mechanism for this in the present paper. For example, it is also possible that

the effect on TFP of the distortion of the age profile is quite small, and the entire reduction in TFP

operates through the channel of distorting the size of firms (Amirapu & Gechter, 2019) or deterring

potential entrepreneurs.