Trade Liberalization, Rival Exporters and Reallocation of Production: An Analysis of Indian Manufacturing

• Date: 01 August 2017
• Author: Lawrence Edwards,Asha Sundaram
• Published date: 01 August 2017
• DOI: http://doi.org/10.1111/roie.12279

Trade Liberalization, Rival Exporters and

Reallocation of Production: An Analysis of Indian

Manufacturing

Lawrence Edwards and Asha Sundaram*

Abstract

Employing a difference-in-difference estimation technique on firm-level data on Indian exporters, we show

that the removalof US textile and apparel quotas was associated with a relativeincrease in sales of products

where India was previously quota restricted, but a relative decrease in sales and the unit value of products

where China was previously quota restricted. Our study hence highlights the importance of accounting for

falling trade barriers for rival exporters in analyzing trade liberalization effects. Additionally, we find evi-

dence indicating thatquota rights were not allocated efficiently, suggesting potential gains from reallocation

with the dismantling of the Indian quota licensing regime.

1. Introduction

The final removal of quota restrictions on the import of clothing and textile products

under the Agreement on Textile and Clothing (ATC), the successor of the Multifibre

Arrangement (MFA), on 1 January 2005 led to a surge in imports of these products

into previously protected markets of advanced economies such as the USA, EU and

Canada.

1

The increase in imports from China was most dramatic and has been studied

widely in the context of US quota restrictions (Barrows and Harrigan, 2009; Brambilla

et al., 2010; Bernhofen et al., 2011). Yet, quotas were also binding on other large

labor-abundant economies with a comparative advantage in clothing and textile prod-

ucts. India, for example, exhibited a US quota fill rate of 87%, next to only China and

Bangladesh, who both exhibited a quota fill rate of 88% (Brambilla et al., 2010).

2

The effect of the end of the MFA on other countries is likely to be different to that

on China. Quotas were applied at a product level on a country-specific basis, implying

potentially differential effects of their removal across countries. The effects of the

quota removal in each country are also potentially interdependent. The response by

exporters in each country to the removal of restrictions on their exports will have been

influenced by the simultaneous dismantling of quotas on rival exporters in other coun-

tries. Improved market access in response to the removal of quotas on one country’s

* Sundaram: School of Economics, University of Cape Town, Rondebosch, Cape Town 7701, South

Africa. Also affiliated to Department of Economics, University of Auckland, Auckland, 1010, New

Zealand. Edwards: School of Economics, University of Cape Town, Rondebosch, Cape Town, South

Africa. Tel: 127-21-650-2730; E-mail: assundar.wk@gmail.com. The authors would like to thank the two

anonymous referees, conference participants at the Annual Conference on Economic Growth and Devel-

opment, ISI Delhi, Midwest International Economics Group Meetings, and seminar participants at the

Michigan International and Macro Lunch and the Departmental Seminar at the University of Cape

Town for extremely useful comments. They are grateful to Alastair Codrington for excellent research

assistance and also thank Economic Research Southern Africa (ERSA) for financial support. All errors

and omissions remain their own.

V

C2016 John Wiley & Sons Ltd

Review of International Economics, 25(3), 677–693, 2017

DOI:10.1111/roie.12279

exports may be overshadowed by competition effects arising from the removal of quo-

tas on rival country exports.

In this paper, we investigate the effect of the removal of the MFA/ATC (hereinafter

MFA) quotas on sales of Indian exporting firms. We focus on quotas applied by the

USA on imports from India and on imports from India’s largest competitor, China.

Further, we ask if quota licenses were allocated efficiently by the Indian quota licens-

ing regime by analyzing the productivity and product-specific unit values of entering,

incumbent and exiting exporters and by exploring heterogenous effects of the Indian

quota removal across these firms (Khandelwal et al., 2013). Our results are robust to

controlling for the EU quota phase-down, the domestic Indian import tariff on textile

and apparel imports, an alternate definition of quota restrictiveness and a falsification

test.

We argue that our study is important for various reasons. First, we are interested in

studying the impact of a reduction in trade barriers on firm behavior and the realloca-

tion of production resulting from the absence of this market distortion, an exercise

that, we believe, has important implications for ascertaining welfare gains. The ending

of the MFA is useful in this regard as it was an externally imposed reduction in trade

barriers and hence presents a suitable opportunity to control for some of the endoge-

neity in trade policy.

Second, we wish to look at the impact of the removal of quotas conditional on the

removal of quotas on imports from rival countries. We believe that this is crucial for

identifying nuances in the effects of the textiles and apparel quota regime on firms in

developing countries.

3

In a broader context, we posit that trade liberalization effects

can interact with the nature of trade barriers in competing, or rival markets.

4

Again,

the removal of the MFA quotas provides us with a tool to delve into reallocation

resulting from a simultaneous fall in trade barriers for two large, exporting countries.

Quotas on China were far more restrictive than those for India. Hence, for a few prod-

ucts, while China was quota restricted under the MFA regime, India was not, provid-

ing us with sufficient variation to disentangle the two effects.

We exploit a firm-level panel dataset on Indian textile firms that also has informa-

tion on firm sales at the product level. This product-level information allows us to

explore the differential impact of the MFA on quota restricted and unrestricted prod-

ucts within the firm. Unlike other studies that focus on the export response by firms,

we focus on the sales response.

5

An advantage of focusing on total firm sales is that

we are able to capture the net effect of the quota removal on production by firms,

after any substitution of exports for domestic sales, or exports to alternative

destinations.

6

Focusing on firms that exported in any year during 2002–2006, we adopt a

difference-in-difference estimation technique that looks at differential sales of prod-

ucts that were subject to quotas (the treatment group) and those that were not (the

control group), before and after 2005, to identify the effect of the removal of quotas

on firm behavior.

7

This empirical strategy enables us to control for unobservable

shocks such as changing consumer preferences, changes in technology or the policy

environment in the textiles and clothing sector, which might affect firm behavior.

Additionally, the detailed nature of the data and our empirical strategy allow us to

account for some of the firm-specific unobserved heterogeneity driving firm outcomes

and selection into products jointly.

We find that the quota removal was associated with an increase in sales of 39% in

previously quota-restricted products and a decrease in sales of 24% in products where

China was previously quota restricted among Indian textile and apparel exporters.

678 Lawrence Edwards and Asha Sundaram

V

C2016 John Wiley & Sons Ltd