TRADE

ADJUSTMENT
IN ASIA
Past Experiences and Lessons Learned

Edited by Marc Bacchetta and Matthias Helble

Trade Adjustment in Asia:
Past Experiences and
Lessons Learned

Edited by
Marc Bacchetta and Matthias Helble
© Asian Development Bank Institute and World Trade Organization 2020

Asian Development Bank Institute (ADBI)

ISBN 9784899742173 (Print)
ISBN 9784899742180 (PDF)

World Trade Organization (WTO)

ISBN 9789287050212 (Print)
ISBN 9789287050229 (PDF)

All rights reserved. No part of this publication may be reproduced, translated and/or
disseminated, in any form, whether for commercial/for profit or non-profit purposes,
without prior permission in writing from the respective rights holders. For this purpose,
the ADBI can be contacted through info@adbi.org, and the WTO can be contacted through
enquiries@wto.org.

The opinions expressed and arguments employed in this publication rest solely
with their authors and are not necessarily those of ADBI, the WTO, or any of their
respective members. They should also not be interpreted as any kind of endorsement or
recommendation on the part of ADBI or the WTO, their respective members, or partners.

Neither ADBI nor the WTO guarantees the accuracy of the data included in this
publication and accepts no liability for any consequence of their use.

By making any designation of or reference to a particular territory, city, or geographic

area, or by using the term “country” in this publication, neither ADBI nor the WTO intend
to make any judgments as to the legal or other status of any territory, city, or area. This
publication, as well as any data and any map included herein, is without prejudice to the
status of or sovereignty over any territory, to the delimitation of international frontiers and
boundaries, and to the name of any territory, city or area.

Reference to names of firms and commercial products and processes does not imply
their endorsement by ADBI or the WTO, and any failure to mention a particular firm,
commercial product, or process is not a sign of disapproval.

ADB recognizes “China” as the People’s Republic of China, “Korea” as the Republic of
Korea, and “Vietnam” as Viet Nam.

Note: In this publication, “$” refers to US dollars.

Cover artwork by Clivewa, “Abstract Watercolor Painting.”

Asian Development Bank Institute

Kasumigaseki Building 8F
3-2-5, Kasumigaseki, Chiyoda-ku
Tokyo 100-6008, Japan
www.adbi.org

World Trade Organization

Centre William Rappard
Rue de Lausanne 154
CH-1211 Geneva 21, Switzerland
www.wto.org
 Contents

Tables and Figures v

Forewordix
Contributorsxi

Introduction1
Marc Bacchetta and Matthias Helble

Part I: Overview: Trade Adjustment in Asia

1. Responses to Trade Opening:
Evidence and Lessons from Asia 17
Devashish Mitra

Part II: Labor Market Adjustments in Asia

2. Industry Wages and Tariffs of the Rest of the World 67
Marcelo Olarreaga, Roberta Piermartini, and Guido Porto
3. Trade Liberalization and the Hukou System of the
People’s Republic of China: How Migration Frictions
Can Amplify the Unequal Gains from Trade 81
Yuan Zi
4. The Impact of Tariff Liberalization on the Labor Share
in India’s Manufacturing Industry  115
Prachi Gupta and Matthias Helble
5. Export Boom, Employment Bust? The Paradox
of Indonesia’s Displaced Workers, 2000–2014 141
Rashesh Shrestha and Ian Coxhead

Part III: Firm-level Adjustments in Asia

6. Firm Adjustment to Trade Policy Changes in East Asia 179
Dionisius Narjoko and Shujiro Urata

iii
iv Contents

7. The Rise of the People’s Republic of China

and Its Competition Effects on Innovation in Japan 207
Nobuaki Yamashita and Isamu Yamauchi
8. Trade Reform, Managers, and Skill Intensity:
Evidence from India 229
Pavel Chakraborty
9. Multiproduct Firms, Tariff Liberalization, and Product
Churning in Vietnamese Manufacturing 276
Ha Thi Thanh Doan

Index299
 Tables and Figures

Tables
A1 Responses to Trade Opening:
Summary of Evidence from Asia 57
2.1 Average Wages and Average Tariffs in ROW:
OLS-FE Estimation 73
2.2 Average Wages and Average Tariffs in ROW:
Asian Economies 75
2.3 Average Wages and Average Tariffs in ROW:
Losers from Higher Tariffs 76
2.4 Average Wages and Average Tariffs in ROW:
Winners from Lower Tariffs 76
3.1 Imports and Tariffs 89
3.2 Effect of Input Tariff Cuts 97
3.3 Regional Adjustments to Trade Liberalization 102
3.4 Regional Effects of Hukou Abolishment 106
3.5 Regional Adjustments to Tariff Reductions,
without Hukou Frictions 108
3.6 Gains from Tariff Reductions Role of Internal Geography 111
4.1 Economic Growth and Openness in India, 1960–2010 (%) 116
4.2 Share in India’s Manufacturing Export and Import
by NIC-2 Digit Classification (%) 119
4.3 Average Labor Share of Income, 1999–2008 120
4.4 Average Employment and Wages/Sales Ratio  121
4.5 Summary Statistics 124
4.6 Average Tariff for Manufacturing Industry Groups
Analyzed in the Study, 1997–2008 126
4.7 Trade Liberalization and Labor Share 129
4.8 Impact of Tariff Reforms on Employment
(Plant Fixed Effects Model) 131
4.9 Trade Liberalization, Labor Share, and Factor Intensity
of Production (Plant Fixed Effects Model) 132
4.10 Trade Liberalization, Labor Share, and Technology
Intensity of Production (Plant Fixed Effects Model) 134
4.11 Trade Liberalization, Labor Share,
and Labor Market Flexibility 136
5.1 Employment Characteristics in Indonesian Family Life
Survey Data, 1993–2014 151
5.2 Share of Formal Jobs by Occupation and Sector 155

v
vi Tables and Figures

5.3 Characteristics of Formal Jobs—Earnings and Benefits 156

5.4 Distribution of Formal Workers by Occupation,
Sector, and Education Level in 2000 158
5.5 Likelihood of Staying Formally Employed
by Education Level and Sector in 2000 160
5.6 Distribution of Formal Workers by Occupation,
Sector, and Education Level in 2007 161
5.7 Likelihood of Staying Formally Employed
by Education Level and Sector in 2007 162
5.8 Impact of Displacement on Earnings in 2007 163
5.9 Impact of Displacement on Earnings in 2014 164
5.10 Characteristics of Workers by Displacement
Category in 2007 167
5.11 Characteristics of Workers by Displacement
Category in 2014 167
5.12 Regression Results from Self-Reported
Displacement, 2007 169
5.13 Earnings Growth and Displacement, 2014 170
A5.1 Distribution of Workers in 2000 across Sector
and Schooling by Their Presence in the 2007 Sample 175
A5.2 Distribution of Workers in 2007 across Sector
and Schooling by Their Status in the 2014 Sample 175
6.1 Summary of Key Findings 181
7.1 Change of Import Competition by Source Economies
in Japan’s Manufacturing Industry, 1994 and 2005 211
7.2 Patent Usage and Variable Definitions 216
7.3 Import Competition from the People’s Republic of China
and Patent Usage, 1994 and 2005 220
7.4 Import Competition from the People’s Republic of China
and Patent Usage (Implied Volatility Regressions),
1994 and 2005 221
7.5 Import Competition from the People’s Republic of China
and Patent Usage (OLS with Firm-Level Characteristic
Controls), 1994 and 2005 222
7.6 Import Competition from the People’s Republic of China
and Patent Usage (OLS with Other Import Competition
Variables), 1994 and 2005 223
A7.1 Descriptive Statistics 227
A7.2 Descriptive Statistics for Variables Used in Regressions 228
8.1 Descriptive Statistics 239
8.2 Correlation Matrix—Imports, Exports,
and Managerial Compensation 242
8.3 Imports, Exports, and Relative Demand for Managers 245
 Tables and Figures vii

8.4 Imports, Exports, Relative Demand for Managers,

and Skill Premium 246
8.5 Output Tariffs, Input Tariffs, and Relative Demand
for Managers 250
8.6 Input Tariffs, Relative Demand for Managers,
and Skill Premium—Benchmark Results 251
8.7 Input Tariffs, Relative Demand for Managers,
and Skill Premium—Additional Channels 255
8.8 Input Tariffs, Relative Demand for Managers,
and Skill Premium—Firm Characteristics 258
8.9 Import of Capital Goods—Top 10 Destinations 261
8.10 Employment Share, by Educational Status 264
8.11 Tertiary-Secondary Wage Premium, by Age Group 265
9.1 Frequency and Output Share of Firms 283
9.2 Superiority of Multiproduct Firms 285
9.3 Sales Distribution across Products 286
9.4 Frequency of Product Turnover 287
9.5 Product Turnover of Multiproduct Firms
by Ownership Type 288
9.6 Contribution of Product Turnover to Output Growth 289
9.7 Tariff Reduction and Product Churning 292

Figures
2.1 Correlations between Wages and Tariffs in Asia 70
2.2 Correlations between Wages and Tariffs in
Developed Countries 70
2.3 Correlations between Wages and Tariffs in Latin America 71
2.4 Correlations between Wages and Tariffs in Africa 72
3.1 Tariff Changes and Preliberalization Tariff Levels 88
3.2 Regional Employment and Import Changes 91
3.3 Individual and Regional Gains from Trade 104
3.4 Individual Gains from Trade with and without
Hukou Frictions 110
4.1 Average Tariff (Weighted Average) for Industrial Products,
1990–2008117
4.2 Tariff Rates by Industry, 2000 and 2008 118
5.1 Indonesia’s Export and Import Composition
by Sector, 1996 146
5.2 Trends in Indonesia’s Exports of Key Products, 1990–2016 146
5.3 Manufacturing Employment as Share of the Population
Aged 20–49 Years for Males and Females 148
5.4 Trends in Shares of Male and Female Manufacturing
Employment by Age Cohort 150

vii
viii Tables and Figures

5.5 Distribution of Log Hourly Earnings

by Displacement Status in 2007 168
5.6 Distribution of Log Hourly Earnings
by Displacement Status in 2014 168
6.1 Theoretical Prediction of the Relationship
between Productivity and Product Scope 191
6.2 Global Value Chain Participation Index of Electronics
Industry in Selected Economies, 2009 198
7.1 The Rise of the People’s Republic of China
in World Trade, 1990–2011 209
7.2 Structural Changes in the People’s Republic of China’s
Export Product Composition, 1990–2011 210
8.1 Trade and Managerial Compensation, 1990–2011 233
8.2 Managerial Compensation, Importers and Nonimporters,
1990–2011234
8.3 Managerial Compensation, across Industries, 1990–2011 241
8.4 Managerial Compensation, across Size Distribution,
1990–2011243
8.5 Tariff Reform in India, Manufacturing Industries,
1990–2011248
8.6 Wage Premium, Tertiary and Secondary Degree Workers,
1987–2004265
8.7 Wage Premium, Different Age Groups, Tertiary
and Secondary Degree Workers, 1987–2004 266

viii
 Foreword

Over the last decade, attitudes toward globalization have shifted in a

number of developed countries, contributing to rising trade tensions.
A growing public perception holds that the integration of goods, services,
labor, and capital markets only benefits a happy few while leaving many
people behind. This change in attitudes has been an important factor
in the transformation of the political landscape with the election in
several countries of politicians who question the effects of international
cooperation and who have adopted fewer cooperative approaches to
trade and migration.
Economists have been actively involved in the public debate, which
has shifted the attention of researchers to the labor market effects of
trade. New research has contributed to a better understanding of the
mechanisms through which firms and workers adjust to trade. Overall,
the findings show that although technological advances and trade yield
important benefits for economies overall, certain types of workers and/
or regions could be negatively affected. They suggest that policies to
help workers adjust to changes in the labor market and ensure that the
benefits are spread more widely can increase the positive impact of open
trade and technological progress. Making trade more inclusive may also
help defuse anti-trade sentiment.
Asia, the focus of this book, is a prime example of the positive effects
of more open trade. This is reflected in attitudes that are generally
more favorable to trade in emerging Asian countries than in developed
countries. A 2018 Pew Research Center report shows that adults in
emerging economies are more likely to credit trade with boosting wages
and creating employment than those in most developed countries. As
Asia opened up, the region experienced sustained economic growth that
improved the welfare of hundreds of millions of people. Yet, not much is
known about how Asian economies have adjusted to trade. Reaping the
benefits of trade opening is not automatic. It requires firms and workers
to adjust to a new, more competitive environment, including seeking
opportunities abroad. Adjustment may have been different and possibly
easier in Asian countries, but in the presence of mobility frictions, it is
unlikely to have been costless. Moreover, depending on the industry and
firm, some workers may have experienced painful declines in pay or
even layoffs, while others saw their salaries increase or found new jobs.
The objective of this book is to shed light on the adjustment of
labor markets to trade opening in Asian countries. Leading academics

ix
x Foreword

around the world have worked on this edited volume to present the
latest evidence on the topic. Their contributions illustrate that trade
indeed played a transformative role in Asia’s development. Trade
opening helped reallocate capital and labor to more productive use.
Frictions in the movement of capital and labor at times have limited
the potential gains. Some groups of workers were less well-prepared to
embrace new opportunities. Overall, the book concludes that trade is a
powerful source for a more efficient allocation of resources. However,
various frictions generate adjustment costs. A better understanding of
the adjustment process can lead to policies designed to reduce those
costs and ease the process. A re-skilling of workers or the introduction
of social safety nets can also help.
Closing the economy to international trade would not achieve the
outcomes its advocates promise. It would throw out the substantial gains
from trade, while disrupting processes of adjustment that are often fairly
well advanced. The question is how to make the adjustment process
smoother and relatively frictionless. We hope this book contributes to
better understanding of the impact of trade opening on sectors, firms,
and labor markets. This will help in designing better policies that ensure
that the benefits of trade are enjoyed by all.

Naoyuki Yoshino Xiaozhun Yi

Dean Deputy Director-General
Asian Development Bank Institute World Trade Organization
 List of Contributors

Marc Bacchetta is a counsellor in the Economic Research and Statistics

Division of the World Trade Organization, Geneva.

Pavel Chakraborty is a lecturer at the Department of Economics,

Lancaster University, United Kingdom.

Ian Coxhead is a professor in the Department of Agricultural and

Applied Economics, University of Wisconsin-Madison, United States.

Ha Thi Thanh Doan is an economist at the Economic Research Institute

for ASEAN and East Asia, Jakarta.

Prachi Gupta is an adjunct professor of political science and international

affairs at Temple University, Tokyo.

Matthias Helble is an economist in the Economic Research and Regional

Cooperation Department of the Asian Development Bank, Manila, and
an adjunct fellow at the Asian Development Bank Institute, Tokyo.

Devashish Mitra is a professor of economics and Gerald B. and Daphna

Cramer professor of global affairs at the Maxwell School, Syracuse
University, United States.

Dionisius Narjoko is a senior economist at the Economic Research

Institute for ASEAN and East Asia, Jakarta.

Marcelo Olarreaga is a professor of economics at the University

of Geneva, and a research fellow at the Centre for Economic Policy
Research, London.

Roberta Piermartini is a counsellor in the Economic Research and

Statistics Division of the World Trade Organization, Geneva.

Guido Porto is professor of economics at Universidad Nacional de La

Plata, Argentina.

Rashesh Shrestha is an economist at the Economic Research Institute

for ASEAN and East Asia, Jakarta.

xi
xii List of Contributors

Shujiro Urata is professor of economics at the Graduate School of Asia-

Pacific Studies, Waseda University, Tokyo, senior research advisor to the
president of the Economic Research Institute for ASEAN and East Asia,
Jakarta, and a visiting fellow at the Asian Development Bank Institute,
Tokyo.

Nobuaki Yamashita is a senior lecturer at the School of Economics,

Finance and Marketing of the Royal Melbourne Institute of Technology
University, Australia.

Isamu Yamauchi is an associate professor at Meiji University, Japan.

Yuan Zi is an assistant professor in the Department of Economics at the

University of Oslo, Norway.
 Introduction
Marc Bacchetta and Matthias Helble

Context
Asia has successfully integrated into the world economy over the past
forty years. After failed attempts to use import substitution strategies
to stimulate economic development, most economies in Asia started to
adopt export-oriented growth policies in the 1980s. As a consequence,
Asia’s trade has grown substantially over the past 40 years, reaching
well over half the region’s gross domestic product in recent years.
Similarly, foreign direct investment (FDI) increased steadily in the
region, especially after the 1985 Plaza Accord. The surge of FDI flows
was linked to the buildup of regional and global value chains (GVCs).
More open trade regimes, combined with modern communication
and transportation technologies, allowed the setting up of production
networks across borders. Subsequently, Asia’s trade transformed itself
from interindustry trade to intra-industry trade.
Hand in hand with trade opening and economic integration, Asia
experienced strong economic growth. The region’s gross domestic
product (in nominal terms) grew about 15-fold from 1980 to 2016.
This exceptional economic progress translated into a substantial
improvement of well-being throughout the populations. For example,
in 1990, 56% of those living in East Asia lived in extreme poverty, while
in South Asia 54% were extremely poor. Within 20 years, those numbers
were reduced significantly. By 2010, 12% remained extremely poor in
East Asia and 31% in South Asia. Given these impressive achievements
and positive experience with globalization, it comes as little surprise
that many in Asia strongly embrace globalization. According to the Pew
Research Center’s Spring 2014 Global Attitudes Survey, a large majority
of Asians (about 86%) believe that trade is good, and that it creates jobs
and boosts wages.
However, as we know from theory and experience, reaping the
benefits from trade opening requires some adjustments. As economies
liberalize their trade regimes and trade costs between trading partners
decrease, each of the trading partners starts exporting more of those
products for which it has a comparative advantage and importing
more of the products for which its trading partner has a comparative

1
2 Trade Adjustment in Asia: Past Experiences and Lessons Learned

advantage. Exporting firms tend to grow and hire more workers, while
import-competing firms need to adjust or shrink and lay off workers.
Some workers need to transition to new positions either within the
same company or in other firms. In any case, firms and workers will
need to adjust in response to changes in trade costs. This adjustment can
be seamless and painless, but it may also be costly and difficult. Overall,
the gains from trade are larger than the losses, but some workers bear
a disproportionate share of the costs of adjustment and some are left
with a lower wage. This also happened in Asia. The success of Asia
in integrating into the world economy was not automatic; it required
sectors, firms, and workers to cope with more international competition
and respond to new opportunities.
While several other developed countries’ political developments
in the last decade attracted the attention of researchers on adjustment
and the labor market effects of trade, there was less interest regarding
adjustment to trade in Asia. In developed countries in the mid-2000s,
the bright light shed by a number of political campaigns on those
left behind by globalization renewed economists’ interest in the link
between trade and jobs, inciting researchers to reexamine previous
findings suggesting that the effect of trade on labor was not significant.
Over the past years, researchers have reexamined more closely the
adjustment process and the effects of trade opening, using more micro-
level data and new methodologies. They found new and more significant
effects of trade on labor market outcomes, in particular of trade with
the People’s Republic of China (Autor, Dorn, and Hanson 2013). The
new evidence indicated that labor market adjustments to trade were
more difficult and costly than what earlier evidence suggested, and that
the adverse impacts of trade are highly concentrated among specific
worker groups and locations (World Trade Organization 2017). These
studies also suggested that trade had contributed more significantly
than earlier studies suggested to the loss of manufacturing jobs in the
US. Yet more recent research factoring in GVCs offers a more benign
picture of the effects of trade on aggregate employment and wages as
well as on manufacturing jobs, but it tends to confirm the findings of
previous studies regarding the concentration of adverse effects of trade
on certain workers and certain regions (Bacchetta and Stolzenburg
2019). This work focused mainly on the manufacturing sector in
developed countries and Latin America. Unfortunately, much less is
known about adjustment and the labor market effects of trade in other
sectors and regions of the world.
For Asia, relatively little empirical evidence has been emerging on
how labor markets and firms in the region have adjusted to trade opening.
Topalova and Khandelwal (2011), for example, provided empirical
evidence for a procompetitive impact of output tariff liberalization in the
 Introduction 3

case of India. Hasan et al. (2012) found that tariff cuts lead to reductions
in unemployment rates in India. Yu (2015) found a positive impact of both
input and output tariff reduction on total factor productivity in the People’s
Republic of China. More recently, McCaig and Pavcnik (2018) showed
that the share of manufacturing workers in Viet Nam in the formal sector
substantially increased in response to US tariff reductions. This emerging
literature suggests that Asia has adjusted well to trade opening.
Our book adds to this relatively new strand of literature by providing
additional insights on how Asia has coped with trade opening. As
mentioned, substantive research efforts have gone into analyzing the
effects of trade opening in developed countries. A particular focus has
been on the impact of the People’s Republic of China’s integration into
the world economy on manufacturing jobs in the US. We believe that
more research is needed to understand how the economies in developing
Asia have handled the adjustment to trade opening.
There are still many open questions regarding the labor market
effects of trade in developing countries. Labor markets in developing
countries are often dual, with a large share of informal jobs and weaker
social safety nets. There is also evidence suggesting that labor mobility
costs are higher in developing countries. At the same time, given the
speed at which many Asian economies have integrated in world trade
and the significant differences between trade developments in Asia
compared to the rest of the world, the effects of trade in this region may
be different—if not qualitatively then at least quantitatively—from what
they are in other regions.
The volume also aims to provide new insights on adjustment to trade
by examining both how firms and how labor markets adjust to trade.
A better understanding of how firms adjust can help us understand
how workers are affected by trade. Firms have different margins of
adjustment, and depending on which of those margins they use in
reaction to trade shocks, workers will be affected differently. The book
shows that there is much to learn from the Asian experience.
Our book comes at a timely moment. In recent years, political
developments in the developed countries have drawn increasing
attention to the distributional impact of globalization and trade. A
number of countries have elected politicians who argue that multilateral
and regional trade opening have been harmful to their countries and, in
particular, to certain workers in certain regions. Consequently, we have
witnessed an increase in trade tensions and in the use of trade restrictive
measures. The trend toward less liberal trade policies is not limited to
developed countries. In several developing countries, efforts to further
open trade have come to a halt or have been even reversed. Our book
provides new evidence on how trade opening affects labor markets and
firms. We hope it serves as useful guidance to respond to some of the
4 Trade Adjustment in Asia: Past Experiences and Lessons Learned

concerns that have been raised and to promote more inclusive trade
opening.

Labor Market Adjustment

Trade and Labor: A Review of the Recent Literature

Recent reviews of the literature on the labor market effects of trade

show that trade leads to employment and wage gains at the national
level, although in the case of employment these are typically small
(Bacchetta and Stolzenburg 2019; World Trade Organization 2017). In
the first half of the 2010s, a series of empirical studies focusing on import
competition from emerging markets and formerly planned Eastern
European economies in the 2000s found that trade liberalization
episodes had had a detrimental impact on labor market outcomes. These
studies suggested that trade was one of the main factors behind the
labor market adjustments that have taken place over the last decades.
A prominent role in this regard was given to the People’s Republic of
China’s accession to the World Trade Organization in 2001 (Autor, Dorn,
and Hanson 2013; Pierce and Schott 2016), the conclusion of the North
American Free Trade Agreement in 1994 (Hakobyan and McLaren 2016),
and the enlargement of the European Union in 2004 (Braakmann and
Vogel 2011). Since the results of these studies were difficult to reconcile
with the common view among economists that trade has only minor
employment impacts, further research was undertaken, which, among
other effects, sheds light on the impact of the expansion of value chains
on the relationship between trade and labor markets.
Factoring in GVCs when studying the impact of trade on labor
markets reveals that trade has not been a significant contributor to
declines in manufacturing jobs in developed economies, and that job
gains in services have offset job losses in manufacturing. While import
competition can hurt employment in import-competing firms and
their suppliers, access to cheaper imported inputs lowers costs in firms
that use them, allowing them to expand. Export expansion, which has
benefited several manufacturing industries, has been another offsetting
factor so that even when considering only manufacturing employment
the overall effect of trade is likely to be minor.
There is also empirical evidence showing that GVC integration has
supported jobs and earnings as well as other spillovers that operate
through labor markets in developing countries (Hollweg 2019). Job and
wage gains have been achieved not only within the exporting sector but
also indirectly through linkages of exporting firms to domestic input-
supplying firms.
 Introduction 5

Importantly, however, moving from the nationwide and sectoral level

to regional and individual outcomes reveals significant heterogeneity
both in developed and developing countries. For instance, when local
labor markets within countries are not sufficiently diversified, trade can
widen regional disparities. Regions specialized in import-competing
and upstream industries can fall behind, while areas with industries that
export or benefit from cost savings pull away. The result that trade can
widen regional disparities in terms of wages and employment seems to
be quite general. It has been found both for developed and developing
countries such as Brazil, India, and Viet Nam (Dix-Carneiro and Kovak
2017; Topalova 2010; McCaig 2011).
Similarly, while in developed countries trade typically leads to labor
market polarization by favoring high-skilled employment over medium-
skilled employment, it also seems to raise the demand for skilled workers
in developing countries, at least in those for which there is evidence
(Bacchetta and Stolzenburg 2019; Bacchetta et al. 2017; Hollweg 2019).
Detailed information on the skill structure within French manufacturing
firms shows that firms employ relatively more skilled workers in
marketing and development when they sell their products outside of
France (Maurin, Thoenig, and Thesmar 2002). Other studies show that
import competition leads to skill upgrading through its impact on product
and process innovation (Bloom, Draka, and Van Reenen 2016; Mion and
Zhu 2013). Firm- and worker-level evidence shows that offshoring and
import competition have a small positive impact on the demand for
nonroutine occupations and thus on job polarization (Becker et al. 2013;
Keller and Utar 2016; Hakkala and Huttunen 2016; Utar 2016; Hummels
et al. 2014). Two recent studies stand out for accounting explicitly for the
rise of GVCs. The first (Reijnders and de Vries 2018) finds that while both
automation and offshoring have contributed to polarization in developed
economies, the effect of automation is dominant. The results of the
second study (Beverelli et al. 2018) suggest that import competition
from the People’s Republic of China increased the share of low-skilled
employment in the US, while participation in GVCs increased the share
of high-skilled employment. Trade as a combination of the two has thus
contributed to polarization. The results for trade are, however, dwarfed
by the estimates for the role of technology.
In developing countries, employment creation and wage gains have
been biased toward more skilled workers, which contrasts with the
predictions of traditional trade theory. Some early evidence pointed
toward skill-biased technical change due to increased competitive
pressure brought about by trade liberalization as one of the mechanisms
at play (Attanasio, Golderg, and Pavcnik 2004). There is also evidence of
a complementary mechanism in which exporters have more incentive to
upgrade their technology when trade costs decrease (Bustos 2011a, 2011b).
6 Trade Adjustment in Asia: Past Experiences and Lessons Learned

When they get better access to the markets of rich countries, exporters
may also have an incentive to upgrade the quality of their products, which
in turn requires more skilled workers (Verhoogen 2008; Brambilla,
Lederman, and Porto 2012). Finally, imported technological change may
be an important driver of demand for skills in developing countries that
rely on imports for most of their capital equipment (Burstein, Cravino,
and Vogel 2013). GVCs reinforce this trend by supporting more complex
industrial organization and relying on complementary skill-intensive
services inputs (Hollweg 2019).

Trade and Labor in Asia

A review of the literature by Devashish Mitra (Chapter 1) confirms
that trade has benefited Asian economies through a number of
channels and in many different respects—through higher productivity,
lower markups through import competition, higher wages, higher
employment, lower unemployment, and, above all, lower poverty
rates. At the same time, however, studies of the labor market and
distributional effects of trade identify a number of challenges. First,
trade can increase informality. Second, high adjustment costs can
significantly slow down the adjustment process following a trade shock
and reduce the benefits from trade opening. Third, rigidities in the
labor market may limit the movement of workers and increase the costs
of firms to make the necessary adjustments. Fourth, there is evidence
that trade raised inequality in some Asian countries but reduced it in
others. Finally, evidence suggests that, in some respects, globalization
has put pressure on workers, decreasing their bargaining power relative
to their employers.
It is important to emphasize that none of these results can be
generalized to all countries in the region or beyond the region. The
empirical studies in this book focus on a small set of countries, including
the People’s Republic of China and India, which differ significantly
from other countries in the region.
The other studies in this volume examine various dimensions of
worker and firm adjustments to trade in Asia. The same caveat regarding
generalization of their results applies.
Marcelo Olarreaga, Roberta Piermartini, and Guido Porto
(Chapter 2) look systematically and for a large sample of developed
and developing countries at the correlation between tariffs faced in
the export destination markets and domestic wages. Tariffs are often
used as an instrument to protect domestic workers from foreign
competition. Arguably, as tariffs protect domestic workers, they also
hurt foreign workers. Indeed, using sector-level data across developing
 Introduction 7

and developed countries during 1976–2004, the authors find that an

increase of 10 percentage points in tariffs faced in international markets
leads on average to a 0.8% reduction in wages at home. They also
explore the winners and losers from the observed changes in tariffs of
the rest of the world (ROW) and show that, because countries export
different products to different countries, there are important differences
in terms of changes in market access. This implies that workers in some
countries have benefited from better market access, whereas others
experienced losses in wages due to a deterioration of their market access
abroad. Of the 14 Asian economies in the sample, eight experienced
declines in average wages as a result of increases in ROW tariffs, and
six experienced wage increases as a result of declines in ROW tariffs.
Japan, Malaysia, Pakistan, and Hong Kong, China experienced some of
the highest average wage growth due to improvements in market access.
Yuan Zi (Chapter 3) analyzes how the hukou system—the People’s
Republic of China’s household registration system—has restricted
the geographical mobility of workers, and the spatial adjustments of
labor to trade shocks and limited the (re)distribution of the gains from
trade. The remarkable expansion of the People’s Republic of China’s
participation in world trade has been accompanied by a high level of
internal migration within the country. Hundreds of millions of workers
in the People’s Republic of China have moved from inland areas to
coastal cities, contributing to the country’s manufacturing growth
and export surges. Zi shows, however, that by prohibiting migrant
workers from accessing various social benefits in their actual cities
of residence, the hukou system limited migration inflows into urban
areas experiencing positive trade shocks. At the worker level, a larger
proportion of the gains accrued to workers holding local hukou. Hence,
the uneven regional gains from trade translated into uneven gains across
people holding different types of hukou.
Prachi Gupta and Matthias Helble (Chapter 4) examine how trade
liberalization affected the labor share in India during 1998–1999 to
2007–2008. Overall, they find that a decline in output tariffs led to an
increase in the labor share of income, while a fall in input tariffs led
to a decrease in the labor share. Interestingly, controlling for factor
intensity, they also find that in sectors that are technology and human
capital intensive, both declines in input and output tariff rates led to a
decline in the labor share. A fall in tariffs only led to an increase in the
labor share for labor-intensive and low-technology plants. Finally, the
empirical results show that labor adjustment occurred more efficiently
in Indian states with flexible labor laws.
Rashesh Shrestha and Ian Coxhead (Chapter 5) analyze the impact
of Indonesia’s export boom and rapid economic growth on formality
8 Trade Adjustment in Asia: Past Experiences and Lessons Learned

in the job market. Indonesia had relatively lower competitiveness in

sectors with high formal employment rates. Trade opening triggered
some involuntary labor movement from formal to informal modes of
employment. Using an econometric model, the authors provide new
evidence that the earnings of workers displaced from formal jobs to
informal ones are significantly lower than those of workers who remain
in the formal market. This finding shows that welfare gains from trade
opening and economic growth are not necessarily shared by everyone.
Furthermore, they shed some light on the causes of Indonesia’s
unprecedented increase in inequality during the same growth period.

Firm-level Adjustment
A review by Dionisius Narjoko and Shujiro Urata (Chapter 6) of the
recent literature on firm-level adjustments in Asia shows strong
evidence of the positive impact of trade opening on productivity as
well as a rapid growth in the number of new firms entering the market.
More and more studies are finding evidence that exporting increases
their process innovation activities and promotes the creation of new
products. Finally, several studies show a positive impact of input tariff
reduction on quality upgrading.
Nobuaki Yamashita and Isamu Yamauchi (Chapter 7) examine
the question whether firms that have been more exposed to import
competition from developing countries undertake more innovative
activity or less. Using a firm-level panel dataset for Japan for 1994–2005,
the authors find that intensified import competition from the People’s
Republic of China has resulted in more innovative activity of Japanese
firms, consistent with similar findings of firms in Europe. Moreover,
such competition has also led to both an increase in nonused patents.
Pavel Chakraborty (Chapter 8) analyzes the effect of tariff
liberalization in India between 1990 and 2011 on wage inequality,
measured as the ratio of managerial to nonmanagerial compensation.
India underwent a significant structural transformation through trade
liberalization and other reforms (domestic) in the 1990s because of a
balance-of-payments crisis. Using this episode to identify the causal
effect of a drop in tariffs on wage inequality, he finds that a 10% reduction
in input tariffs (and not output) significantly increases the share of
managerial compensation by 0.5%–3.5%, raising within-firm wage
inequality. His results also suggest that a rise in skill intensity possibly
explains the increase in managerial compensation, but only for firms
below the median of the size distribution. On the other hand, there is
no evidence of a demand shift away from nonmanagers due to the drop
in tariffs.
 Introduction 9

Using firm-level data covering the 2010–2015 period, Ha Thi

Thanh Doan (Chapter 9) analyzes the characteristics of multiproduct
manufacturing firms in Viet Nam. She finds that multiproduct firms are
larger, more capital intensive, more productive, and more likely to export.
Looking into the link between tariff reduction and product shedding, her
results reveal that exporters have an important role in product adding,
which suggests they may contribute to aggregate growth through the
channel of product scope expansion. Finally, while she finds that state-
owned enterprises are more likely to spread economic activities across
products and industries, there is little difference in terms of product
churning among FDI, state-owned enterprises, and the domestic private
sector.

Main Findings and Policy Conclusions

Mobility frictions prevent necessary adjustments. In the case of the
People’s Republic of China, for example, the hukou system appears to
have limited the workers in moving where the demand was. In India,
firms in states with rigid labor laws were unwilling to hire more labor
and instead replaced labor with capital. Several studies in this volume
indicate that mobility frictions can cause important adjustment costs.
Identifying these frictions and addressing them can improve overall
efficiency but it can also help make trade more inclusive.
Trade opening appears to have the potential to increase informality,
but it doesn’t necessarily have this effect. In the case of Indonesia,
trade opening triggered a movement from formal to informal modes of
employment. However, there is also evidence that trade can raise the
share of formal employment (Bacchetta et al. 2009). The benefits of
trade opening can be more easily shared if a large share of workers are
in the formal sector. Being in the formal sector usually means to enjoy
benefits that smoothen the adjustment process, such as unemployment
insurance. More research is needed to understand how governments can
ensure that more trade will not result in more informal employment.
Evidence in this volume also suggests that engaging in exports can
stimulate research and development activities, enlarge the product
scope, and increase the product quality, which tend to increase the
demand for skilled labor. Not having access to skilled labor undermines
the competitiveness of firms and entire sectors. Training and education
policies play an important role in helping respond to the changes in
labor demand induced by trade (Bacchetta et al. 2017).
Trade opening increases competition among firms. At the
same time, it allows for improved access to inputs at lower prices.
Typically, it therefore results in an increase in productivity of firms.
10 Trade Adjustment in Asia: Past Experiences and Lessons Learned

However, some firms might have to exit the market or are absorbed
by other firms. These dynamics suggest that the quick reallocation of
capital from less to more productive uses is crucial to achieve high
welfare benefits. Well-developed capital markets can help achieve this
objective. In Asia, the financing of firms is typically bank dominated, and
capital markets are shallow. As Asian economies grow, it is important
to deepen their financial markets to ensure an easier redeployment of
capital across sectors.
Trade opening is a blessing for some sectors, while others might
suffer. It can therefore have a significant impact on income distribution.
Despite adjustment costs, trade opening usually yields net overall
welfare benefits. While there is an efficiency argument for governments
to address mobility frictions and reduce adjustment costs, governments
may have different preferences regarding the need to redistribute
income between those who lose and those who gain from trade. In
developing countries, tax collection can be a challenge for governments
that wish to redistribute. For these countries, improving tax collection
should therefore become a policy priority. The experience of Asia
clearly shows that trade delivers; governments can help ensure that
trade delivers for all.
 Introduction 11

References
Attanasio, O., P.K. Goldberg, and N. Pavcnik. 2004. Trade Reforms and
Wage Inequality in Colombia. Journal of Development Economics.
74 (2). pp. 331–66.
Autor, D.H., D. Dorn, and G.H. Hanson. 2013. The China Syndrome:
Local Labor Market Effects of Import Competition in the United
States. American Economic Review. 103 (6). pp. 2121–68.
Bacchetta, M., and V. Stolzenburg. 2019. Trade, Value Chains and
Labour Markets in Advanced Economies. In World Trade
Organization, Institute of Developing Economies – Japan External
Trade Organization, Organisation for Economic Co-operation and
Development, University of International Business and Economics,
and World Bank Group, eds. Global Value Chain Development Report
2019: Technological Innovation, Supply Chain Trade, and Workers in
a Globalized World. Geneva: World Trade Organization.
Bacchetta, M., C. Gregg, S. Rubínová, and B. Tumurchudur Klok. 2017.
Investing in Skills for Inclusive Trade. Geneva: International Labour
Organization and World Trade Organization.
Bacchetta, M., E. Ernst, and J.P. Bustamante. 2009. Globalization and
Informal Jobs in Developing Countries. Geneva: International
Labour Organization and World Trade Organization.
Becker, S., K. Ekholm, and M.A. Muendler. 2013. Offshoring and the
Onshore Composition of Tasks and Skills. Journal of International
Economics. 90 (1). pp. 91–106.
Beverelli, C., S. Rubinova, V. Stolzenburg, and N. Woessner. 2018.
Did Global Value Chains Contribute to Rising Labour Market
Polarization? World Trade Organization, unpublished manuscript.
Bloom, N., M. Draka, and J. Van Reenen. 2016. Trade Induced Technical
Change? The Impact of Chinese Imports on Innovation, IT and
Productivity. The Review of Economic Studies. 83 (1). pp. 87–117.
Braakmann, N., and A. Vogel. 2011. How Does Economic Integration
Influence Employment and Wages in Border Regions? The Case of
the EU Enlargement 2004 and Germany’s Eastern Border. Review of
World Economics. 147 (2). pp. 303–23.
Brambilla, I., D. Lederman, and G. Porto. 2012. Exports, Export
Destinations, and Skills. The American Economic Review. 102 (7).
pp. 3406–38.
Burstein, A., J. Cravino, and J. Vogel. 2013. Importing Skill-Biased
Technology. American Economic Journal: Macroeconomics. 5 (2).
pp. 32–71.
Bustos, P. 2011a. The Impact of Trade Liberalization on Skill Upgrading:
Evidence From Argentina. Universitat Pompeu Fabra Economics
12 Trade Adjustment in Asia: Past Experiences and Lessons Learned

and Business Working Paper. No. 1189. Barcelona: Universitat

Pompeu Fabra.
____. 2011b. Trade Liberalization, Exports, and Technology Upgrading:
Evidence on the Impact of MERCOSUR on Argentinian Firms.
American Economic Review. 101 (1). pp. 304–40.
Dix-Carneiro, R., and B.K. Kovak. 2017. Trade Liberalization and Regional
Dynamics. The American Economic Review. 107 (10). pp. 2908–46.
Hakkala, K.N., and K. Huttunen. 2016. Worker-Level Consequences of
Import Shocks. VATT Institute for Economic Research Working
Paper. No. 74. Helsinki: VATT Institute for Economic Research.
Hakobyan, S., and J. McLaren. 2016. Looking for Local Labor Market
Effects of NAFTA. Review of Economics and Statistics. 98 (4).
pp. 728–41.
Hasan, R., D. Mitra, P. Ranjan, and R.N. Ahsan. 2012. Trade Liberalization
and Unemployment: Theory and Evidence from India. Journal of
Development Economics. 97 (2). pp. 269–80.
Hollweg, C.H. 2019. Global Value Chains and Employment in Developing
Countries. In World Trade Organization, Institute of Developing
Economies – Japan External Trade Organization, Organisation
for Economic Co-operation and Development, University of
International Business and Economics and World Bank Group,
eds. Global Value Chain Development Report 2019: Technological
Innovation, Supply Chain Trade, and Workers in a Globalized World.
Geneva: World Trade Organization.
Hummels, D.L., R. Jørgensen, J.R. Munch, and C. Xiang. 2014. The Wage
Effects of Offshoring: Evidence from Danish Matched Worker-Firm
Data. American Economic Review. 104 (6). pp. 1597–629.
Keller, W., and H. Utar. 2016. International Trade and Job Polarization:
Evidence at the Worker Level. CESifo Working Paper. No. 5978.
Munich: Center for Economic. Studies and Ifo Institute.
Maurin, E., M. Thoenig, and D. Thesmar. 2002. Globalization and the
Demand for Skill: An Export Based Channel. CEPR Discussion
Paper. No. 3406. Washington, DC: Center for Economic and Policy
Research.
McCaig, B. 2011. Exporting Out of Poverty: Provincial Poverty in Vietnam
and US Market Access. Journal of International Economics. 85 (1).
pp. 102–13.
McCaig, B., and N. Pavcnik. 2018. Export Markets and Labor Allocation
in a Low-Income Country. American Economic Review. 108 (7).
pp. 1899–941.
Mion, G., and L. Zhu. 2013. Import Competition from and Offshoring
to China: A Curse or Blessing for Firms? Journal of International
Economics. 89 (1). pp. 202–15.
 Introduction 13

Pierce, J.R., and P.K. Schott. 2016. The Surprisingly Swift Decline of
US Manufacturing Employment. The American Economic Review.
106 (7). pp. 1632–62.
Reijnders, L.S.M., and G.J. de Vries. 2018. Technology, Offshoring and
the Rise of Non-routine Jobs. Journal of Development Economics.
135. pp. 412–32.
Topalova, P. 2010. Factor Immobility and Regional Impacts of Trade
Liberalization: Evidence on Poverty from India. American Economic
Journal: Applied Economics. 2 (4). pp. 1–41.
Topalova, P., and A. Khandelwal. 2011. Trade Liberalization and Firm
Productivity: The Case of India. Review of Economics and Statistics.
93 (3). pp. 995–1009.
Utar, H. 2016. Workers beneath the Floodgates: Impact of Low-Wage
Import Competition and Workers’ Adjustment. CESifo Working
Paper Series. No. 6224. Munich: Center for Economic. Studies and
Ifo Institute.
Verhoogen, E.A. 2008. Trade, Quality Upgrading, and Wage Inequality
in the Mexican Manufacturing Sector. The Quarterly Journal of
Economics. 123 (2). pp. 489–530.
World Trade Organization. 2017. World Trade Report 2017: Trade,
Technology and Jobs. Geneva.
Yu, M. 2015. Processing Trade, Tariff Reductions and Firm Productivity:
Evidence from Chinese Firms. The Economic Journal. 125 (585).
pp. 943–88.
 PART I
Overview:
Trade Adjustment
in Asia
 1
Responses to Trade Opening:
Evidence and Lessons from Asia
Devashish Mitra

1.1 Introduction
Over the last 3 decades, several developing countries have liberalized
their trade regimes. This may have happened either due partially to
conditionalities imposed by international organizations, such as the
International Monetary Fund in response to emergency requests for loans,
in the context of a country’s accession to the World Trade Organization, or
as a result of the signature of a preferential trade agreement. In many cases,
the reforms may have stemmed from a country’s own disappointment
with its growth performance during its import substitution phase. While
movement toward free trade is expected to expand the size of the overall
economic pie, such changes always produce both losers and winners.
In fact, it is this creation of winners and losers, along with “individual-
specific uncertainty” (Fernandez and Rodrik 1991) about who benefits
and who loses from reforms, that has led to the delays in trade reforms,
appropriately called “status quo bias.”
Guided by theoretical work, a large number of empirical papers
focus on identifying the losing and winning sections of society from these
reforms and even quantifying the impact of these reforms on the various
economic classes, such as the poor relative to the rich and unskilled
relative to skilled workers. While there certainly will be winners and
losers, who exactly they will be also matters. If the rich benefit and the
poor lose, then, despite economic growth, there will be a new situation
with higher poverty and inequality. This does not necessarily mean that
countries should not open up to trade. It only means that they will have
to have social protection schemes in place at the time of trade reforms.
Of course, if the incomes of the poor grow along with the overall
economic growth brought about by trade opening, then trade reforms
will be highly desirable. Even in these situations, social protection and

17
18 Trade Adjustment in Asia: Past Experiences and Lessons Learned

redistributive policies might be necessary to maximize the progress in

poverty reduction and minimize any possible increase in inequality.
As will be apparent from the evidence reviewed in this chapter,
researchers associate trade liberalization with poverty reduction in Asia.
This has probably happened through economic growth. At the same
time, they also sometimes and in some countries see a rise in inequality.
In the case of the People’s Republic of China, there has been a very rapid
rise in income inequality but at the same time a particularly impressive
reduction in poverty. On the one hand, there is no question about the
desirability of poverty reduction. The desirability of lower inequality, on
the other hand, is sometimes questionable under certain conditions. For
instance, some degree of inequality is optimal to preserve incentives in
any economy. However, beyond a point, inequality could lead to social,
political, and economic instability that could hurt a country’s growth
performance. It could also lead to inequality in educational attainment.
Especially in the presence of credit market imperfections, inequality
reduces a country’s aggregate human capital and handicaps local
businesses in their expansion (or even in their entry decisions), thereby
adversely affecting growth.
In this chapter, I first examine the evidence on the impact of trade
opening on productivity, since productivity levels normally determine
incomes. A rise in productivity, holding other things equal, improves
workers’ wages and lifts them out of poverty. In addition, it is always
necessary to determine the distribution of any increase in per capita
income. The first distribution of importance is that between producers
and consumers. While every individual in society is both a producer and
a consumer, owners of firms or capitalists actually receive the producer
surplus. Thus, looking at the impact of trade on producer and consumer
surplus indicates something about the distribution of welfare changes.
A reduction in price–cost markups indicates a shift of the surplus from
firm owners to consumers, most of whom are ordinary citizens whose
primary income source is their raw labor power. Therefore, I also study
the impact of trade reforms on markups.
As mentioned, for ordinary citizens, the main source of income is
their labor. It is therefore necessary to know whether trade provides
a higher or lower reward for their labor and whether it provides
greater and better opportunities to use their labor for their livelihoods.
Therefore, I next review the evidence on the impact of trade reforms on
wages, employment, and unemployment. Productivity and markups are
just inputs into these outcomes that matter directly to citizens.
However, it is important to consider not only how many people have
jobs but also the quality of their jobs. In developing countries, a large
number of people work in the informal sector or informally for formal
 Responses to Trade Opening: Evidence and Lessons from Asia 19

firms. These workers lack job security, health insurance, a pension plan,
and so on. In addition, they receive a wage that is a fraction of the formal
worker wage. Therefore, it is important to investigate how trade affects
informal employment as a proportion of the overall employment and
whether there are any complementary domestic policies and institutions
that affect this relationship. Investigating these relationships will allow
me to make recommendations on domestic policies such as labor market
policies.
In addition, it is necessary to examine the immediate impact of trade
reforms or shocks on workers. While workers might move to a better
steady state due to trade liberalization, they may incur mobility costs to
move to the new steady state. As a result of incurring trade adjustment
costs, workers might be hurt in the transition, and they may lose a
significant part of the gross welfare gains as a result. I discuss policies
aimed at reducing labor mobility costs and minimizing the pain from
trade adjustment.
I continue by directly examining the evidence on the impact of
trade reforms on poverty, as well as the channels through which this
impact takes place. In this regard, I consider rural and urban poverty
separately. I also investigate the impact on the people just below the
poverty line relative to those far below it. The impact of trade on poverty
is important, especially when evaluating social welfare using a Rawlsian
welfare function, which measures how well a society is performing by
how well the least well-off are performing.
Next, this chapter reviews the impact of trade liberalization on
inequality in its various forms. While there are summary measures
of inequality, such as the Gini coefficient, every such measure has its
weaknesses and no measure is able to capture all aspects of inequality.
Therefore, starting with overall inequality, I move to more specific
forms of inequality, such as the inequality among workers, which
I capture through the ratio of the wages of the skilled to those of the
unskilled, often using the ratio of wages of nonproduction workers to
those of production workers as a proxy. I also study the interindustry
wage differentials and the way in which trade reforms affect them. It
is important to analyze this heterogeneity, because different industries
receive different amounts of tariff cuts. For example, in most countries,
the tariff cuts were deepest in labor-intensive industries, simply because
they were the most protected initially.
Another way in which trade affects the distribution of the overall pie
is through its impact on the bargaining power of workers relative to their
employers. In addition, as mentioned, trade affects the monopoly power
of firms, which I measure using the price–marginal cost markups. Apart
from affecting the way in which consumers and producers share the
20 Trade Adjustment in Asia: Past Experiences and Lessons Learned

surplus, this impact affects the wedge between the value of the marginal
product of labor and the wage that labor receives. As a result, this may
affect labor’s share in the sales or output (and, at the macro level, the
national income). It is also important to analyze the impact of trade on
these labor shares (along with the bargaining power of workers), since,
during the last 2–3 decades, these shares have been falling all over the
world, along with countries opening their trade regimes, in the presence
of skill-biased technical change. It is important to investigate which of
the two (trade or technological change) is the culprit here.
I find that, for Asian countries, on the whole has been able to
stimulate productivity and, to a certain extent, discipline firms to
reduce their markups (if I focus on the impact of reductions in output
tariff cuts). As a result, wages have risen on average. In addition, large
reductions in poverty have occurred. However, inequality has increased,
for which the blame partially falls on trade. Thus, there is a need for
redistributive policies and social protection policies, especially in the
form of public works programs.

1.2 Trade Reforms, Productivity, and Markups

The standard gains from trade are those that countries obtain through
specialization and exchange. Moving away from free trade to a state of
protection leads to production as well as consumption distortion costs.
Under protection, for an import-competing good, consumers pay and
domestic producers receive a higher price than that under free trade.
This distorts both the consumption and the production of the good,
with too little of the former and too much of the latter. It is possible
to demonstrate these standard costs of protectionism or the standard
gains from trade under the very basic conditions of perfect competition.
Once countries move to less competitive market structures, allowing
firms to have some degree of monopoly power, trade results in another
gain. Trade destroys the monopoly power of domestic firms along with
the deadweight losses that accompany it. Domestic firms, while still not
facing any competition from other domestic firms, now face competition
from foreign firms within the same industry. Domestic consumers are
no longer at the mercy of domestic firms, and, as a result, firms cannot
charge unusually high markups over their costs. In other words, trade
has a way of disciplining domestic firms, the extent of which it is
possible to estimate by examining the impact of trade on price–marginal
cost markups.
In addition to a decline in markups, which benefits consumers for
given production costs, import competition can also have an impact
on these costs themselves through the induced changes in technology
 Responses to Trade Opening: Evidence and Lessons from Asia 21

and efficiency. Procompetitive effects of trade lead to an increase in

the incentives for import-competing firms to invest in research and
development and function more efficiently. The way in which these
effects work is that reductions in the production costs of domestic firms
relative to those of foreign firms now lead to a gain in the market share of
the former at the expense of the latter. This opportunity to grab market
share from a foreign competitor or the danger of losing some market
share to it due to lagging productivity leads to this procompetitive
increase in the incentive for firms to invest in productivity increases.
There is also a market size effect in the opposite direction, arising from
the fact that the benefits of any reduction in the production costs of
a domestic import-competing firm now applies to a smaller domestic
market (for any domestic firm) under freer trade, thereby reducing the
returns to cost reduction. In addition, the trade and endogenous growth
literature highlights several other channels through which trade can
affect productivity growth. These effects often move in opposing
directions. For example, depending on whether skilled labor becomes
more or less expensive through trade due to Stolper–Samuelson-type
effects, research and development output may increase or decrease.
Trade will also reduce the duplication of research efforts as well as lead
to greater knowledge flows, resulting in higher productivity. Grossman
and Helpman (1991) describe and rigorously model a number of other
channels. Due to these numerous mutually opposing effects, how trade
affects productivity and productivity growth becomes an empirical
question.
While all the theory on markups points in only one direction, namely
that trade reforms, in the form of reductions in import protection for the
output of an industry (leading to greater import competition), should
unambiguously lower price–cost markups, the various theoretical
models on the impact of trade reforms on productivity together predict
little. However, it is important to study both empirically for a few
reasons. While price depends on cost and markup, cost depends on
productivity. In other words, consumer and overall welfare ultimately
depends on productivity and markup, and it is thus important to study
them, especially to gain some idea of the sizes of the actual changes.
Furthermore, inputs may become cheaper due to trade liberalization,
and that might have an opposite impact on markups.
In many studies, in the process of estimating productivity and the
impact of trade on it, researchers end up measuring markups and the
impact of trade on them. It is important to note that, while the latter
is often a by-product of the former, markup estimations are sometimes
necessary to achieve accurate estimates of productivity and changes in
productivity.
22 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Pioneered by Nobel laureate Robert Solow in 1957, the oldest

approach to measuring total factor productivity (TFP) growth was
called growth accounting. The approach assumed perfect competition
and constant returns to scale, with the implication that the share of a
factor in the total output equals the elasticity of the output with respect
to the factor. This is another way of saying that the reward to the
factor equals the value of its marginal product. However, the elasticity
of output with respect to this factor, measured as its share in output,
would be underestimated in the presence of imperfect competition and
increasing returns to scale. This would overestimate the unexplained
growth in output or TFP growth. Robert Hall’s (1988) pathbreaking
work in macroeconomics incorporated imperfect competition and
markups as well as nonconstant returns into the analysis of TFP growth.
In international trade, Harrison (1994) was the first to incorporate these
two features into TFP growth estimation using firm-level data from Côte
d’Ivoire. The same regression estimates both markups and productivity
growth as well as changes in them in response to trade liberalization.
Krishna’s and Mitra’s (1998) paper was the first to estimate these
parameters using the same methodology (though slightly modified to
incorporate changes in returns to scale) for an Asian country.

1.2.1 Empirics
I start with the case of India. The oldest study on the impact of trade
reforms on markups and productivity in India is the one by Krishna
and Mitra (1998). They use firm-level data for a few industries in the
period 1986–1993 on output, capital, labor raw materials, energy use,
and input shares in output (using the data on input expenditures) to
produce estimates of both markups and productivity along with changes
in them between the pre- and postreform periods. The idea is that, if
they regress the growth rate of output on a weighted sum of the growth
rates in the various inputs (the weights being the shares of the various
inputs in the value of the output), the coefficient of this variable is the
price-to-marginal cost markup and the intercept term is the estimated
TFP growth. Additionally, using an interaction dummy variable and the
dummy variable itself (where the dummy variable takes the value 1 for
the postreform years and 0 otherwise), we can estimate the change in
the markup and productivity growth.
Such a study is meaningful only if the trade reform is not endogenous
to changes in the relevant economic variables. The Indian trade reforms
initiated in 1991 provide such an opportunity, since the reforms, as
Krishna and Mitra argue, were unexpected in that the Government
of India approached the International Monetary Fund to rescue it
 Responses to Trade Opening: Evidence and Lessons from Asia 23

from a bad macroeconomic situation. The loans came with the strict
conditionality of economic reforms, which included trade liberalization
as an important component. There were other reasons as well to believe
that the reforms were exogenous.
For three out of the four industries studied—non-electrical
machinery, electronics, and transport equipment—Krishna and Mitra
find statistically significant reductions in markups. In the pre-reform
phase, the markups are in the range of 1–2 percentage points, and they
all fall below 1 percentage point after the reforms. This is consistent
with the idea that a firm might lose money while adapting to a new and
changing environment. In another industry—electrical machinery—
the markup is below 1 percentage point initially, and no statistically
significant change in the markup is observable.
Moving to productivity growth, in three out of four industries—
electrical machinery, non-electrical machinery, and electronics—the
point estimates of growth increases are positive, ranging between 3 and 6
percentage points, but these estimates are not as precise as the markups
and markup changes. In the case of transport equipment, there is a
decline in productivity growth, but the estimate is highly insignificant
statistically.
One important point to note is that this method involves choosing
inputs and output simultaneously, as a result of which both are correlated
with technology shocks. Consequently, the right-hand side input
variables are correlated with the error term. Researchers argue that
this will lead to biased estimates of productivity growth and markups.
Krishna and Mitra assert that the estimates of the change in markup and
change in productivity growth will be biased only if the abovementioned
correlation changes after the reform. They do not expect this reform to
have a systematic impact on this correlation. Krishna and Mitra support
their arguments with Monte Carlo simulations.
Since Krishna’s and Mitra’s study, the methodologies for markup
and productivity estimation have improved, and they address the above
concerns directly. The recent studies on productivity and markup are
separate. Topalova and Khandelwal (2011) use the Levinsohn–Petrin
approach to address the simultaneity problem that I described above
(as well as the measurement error problem). The approach recognizes
that the choice of materials responds to technology shocks and changes
in the capital stock. Under such conditions, inverting this function of
technology shocks and capital stock gives the technology shocks as
a function of material inputs and capital. Further assuming a Markov
process for technology, the authors are able to control for simultaneity
problems. The authors also have a longer sample period for their firm-
level analysis, spanning a 15-year period, 1987–2001. Additionally, they
24 Trade Adjustment in Asia: Past Experiences and Lessons Learned

investigate the impact of both tariffs on final goods as well as inputs.

Topalova and Khandelwal find a procompetitive impact of output
tariffs in that lower tariffs lead to higher productivity.1 However, they
conclude that the positive impact of an equal input tariff reduction is
much greater in size. A 10% reduction in the output tariff increases
productivity by 0.3%, while a 10% reduction in the input tariff leads to a
4.8% productivity increase. Between 1989 and 1996, the output and input
tariff declines led to about 1.7% and 10.6% increases in productivity,
respectively. The authors view this result as indicative of a much
stronger impact of trade liberalization through the greater availability
of a broader range of inputs of higher quality as well as “exposure to
new technologies” rather than through greater competition from
final products coming from abroad. Goldberg et al. (2010) confirm
this channel, finding that trade liberalization led to a 21% decline in
the prices of intermediate inputs and an 8% increase in the variety of
intermediate inputs.
De Loecker et al. (2016) rigorously study the impact of trade
liberalization on the markups of Indian firms. Using an improved version
of the Levinsohn–Petrin approach to deal with the fact that physical
quantities of output and inputs are rarely observable (leading to biases
in production function estimation), the authors arrive at estimates of the
elasticities of output with respect to the various inputs. Under constant
returns to scale, the price–marginal cost markup equals the elasticity
of output with respect to an input divided by the input’s share in the
sales of output. They calculate these markups and investigate their
relationship to trade liberalization for the period 1989–1997. Reductions
in input tariffs increase markups. Average costs are lowered through
reductions in input tariffs but partly offset by the fact that only some of
these benefits are passed on to consumers as a result of an increase in
markups. A 10-percentage-point reduction in input tariffs can lead to
an 8% increase in markups; a 10-percentage-point reduction in output
tariffs has a procompetitive effect of reducing markups by 1.2%–1.5%.
While output tariffs declined on average by 62 percentage points during
the 1989–1996 period, input tariffs declined by 24 percentage points.
The net impact was an increase in markups during this period of about
13%. Given that costs declined by 31% due to lower input prices as well
as a greater variety of inputs (consistent with Goldberg et al. [2010] as
well as Topalova and Khandelwal [2011]), this meant that there was
an average decline in prices (relative to the overall two-digit sectoral
price level) of about 18%. The authors also find that the procompetitive

1
Note that the new literature, unlike the old literature (Krishna and Mitra 1998;
Harrison 1994), focuses on productivity levels rather than productivity growth rates.
 Responses to Trade Opening: Evidence and Lessons from Asia 25

effect of output tariff reduction was concentrated in the initially

high-markup firms. While a 10-percentage-point reduction in output
tariffs led to a 4.4% reduction in markups in firms that were among the
top 10% of the initial distribution of markups, the remaining firms, on
average, experienced a 1.3% markup reduction.
Nataraj (2011) also examines formal- and informal-sector firms in
India separately for the period 1989–2001 and finds that, while mainly
output tariff reductions affect informal-firm productivity, input tariff
reductions affect only formal-sector firm productivity. A 10-percentage-
point reduction in the output tariff increases formal-firm productivity
by up to 0.76% but increases informal firm productivity by 4.8%. A
10-percentage-point reduction in the input tariff increases formal-firm
productivity by 4.6%, with no statistically significant effect on informal-
firm productivity. This makes intuitive sense, since informal sector
firms rarely buy imported inputs but might feel the competition from
imported final products.
Amiti and Konings (2007) study the impact of the 1990s trade reforms
on firm-level productivity in Indonesia. Indonesia’s trade reforms are
linked to its accession to the World Trade Organization in 1995, and the
sample period for this study is 1991–2001. This is, in fact, the first study
to investigate the impact of input and output tariffs simultaneously on
firm productivity, which it calculates using the Olley–Pakes approach
to correct for both simultaneity in input and output choice and sample
selection bias. While a 1%–6% increase in productivity is attributable to
a 10-percentage-point reduction in the output tariff, they find that firms
that import inputs can experience up to a 13% increase in productivity
from a 10-percentage-point input tariff reduction. They also find that
the Asian financial crisis somewhat muted the latter effect from 1997
onward. This is understandable, since domestic currency devaluations
would have led to an increase in the domestic price of inputs. The
authors also find the beneficial effects of output tariff reforms to be
concentrated in the more competitive industries (as compared with
the high-markup ones), while the positive effects of input tariffs on
productivity do not vary by the degree of within-industry competition.
The results here qualitatively survive correction for endogeneity of
trade reforms through an instrumental variable approach in which the
initial (1991) tariffs instrument the change in tariffs.
Brandt et al. (2017) study the impact of trade liberalization on firm-
level markups and TFP in the People’s Republic of China around the
time of its accession to the World Trade Organization. As in the case of
India, input tariff reductions increase markups, implying an incomplete
pass-through of input cost reductions to consumers. On the other
hand, output tariff reductions reduce the markups only of the relatively
26 Trade Adjustment in Asia: Past Experiences and Lessons Learned

large firms, mainly incumbents, and have no impact on the markups of

other firms, especially new entrants. On average, a 10-percentage-point
reduction in the output tariff leads to up to a 1% markup reduction,
while a 10-percentage-point reduction in the input tariff leads to a
7% markup increase.2 The authors argue that the endogeneity of the
trade liberalization is not a major concern by showing that neither past
industry productivity nor past productivity change determine tariff
reductions. The procompetitive effects in this study and in that of De
Loecker et al. (2016) are not comparable, since the latter observe firm-
level prices to which they can apply their estimated markups to compute
firm-level marginal costs, which, in turn, they can control for in their
estimation of the procompetitive effects of output tariff reductions.
To the extent that there are procompetitive effects on productivity
and costs that are incompletely passed on to consumers, this study of
the People’s Republic of China will underestimate the output tariff
effects on markups. The procompetitive effect of a 10-percentage-point
decline in the output tariff is a 1.7% increase in TFP, while, for the same
percentage decline in the input tariff, the TFP gain is 16%–18%. While
the bulk of the gain in industry-level productivity comes from increases
in within-firm productivity, some of the gains arise through the entry
of new productive firms that are flexible enough to incorporate newer,
more efficient technology. The exit of relatively low-productivity firms
is another channel but is not as strong. The authors also find that
the industries that experience deeper input tariff cuts are those that
experience a relatively smaller reallocation of output from less to more
productive firms.
Yu (2015) also considers the impact of trade liberalization on firm-
level TFP in the People’s Republic of China for the period 2000–2006.
While he also finds a positive impact of both input and output tariff
reduction on TFP, in contrast to other studies, he identifies a much
bigger impact of output tariff reductions than input tariff reductions.
With a 10-percentage-point output tariff reduction, TFP increases
by 9%. A 10-percentage-point input tariff reduction increases TFP by
only 5%. Many reasons could explain the difference in results. The
sample period is slightly different (in Brandt et al. [2017]it is 1998–
2007), but the tariff measures are also firm specific in Yu’s study,
with firm-specific weights based on the multiple product lines that
each firm sells in the case of output tariffs and firm-level imports of

2
Fan et al. (2017) obtain qualitatively similar results on markups. They run regressions
separately for firms engaged in processing trade and other firms. As they expect from
the theory, these effects empirically do not appear for the former but are apparent in
the latter type of firms.
 Responses to Trade Opening: Evidence and Lessons from Asia 27

various imported inputs in the case of input tariffs. Another result that
Yu obtains is that the impact of these tariff reductions on TFP decreases
with an increase in the share of processing imports in total imports. This
is not surprising, since these processing imports do not lead to greater
import competition nor were they ever subject to tariffs during the
author’s sample period.
Bas and Causa (2013) examine the impact of input tariff reductions,
among many policy changes, on the labor productivity of firms in the
People’s Republic of China. In this study, they take the productivity
heterogeneity among firms quite seriously. Using a sample of firms
for 2001–2008, they find that input tariff reductions increase labor
productivity and that the effect is stronger for firms at the domestic
technological frontier than for other firms. Firms that are on or close to the
frontier experience a 0.74% increase in productivity from a 1-percentage-
point reduction in the input tariff. Firms whose productivity is half of
the domestic technological frontier will experience roughly a 0.5% rise
in productivity from the same 1-point-reduction in the input tariff. The
procompetitive impact of output tariff reductions seems to be stronger
for firms that are relatively distant from the technological frontier.
There is also an industry-level study by Kim (2000) for the Republic
of Korea for the period 1966–1988. He finds that a 10-percentage-point
reduction in the quota–coverage ratio leads to an increase in TFP growth
of about 0.26 percentage points and a reduction in the markup of 1.33
percentage points. A 10-percentage-point reduction in the nominal rate
of protection, on the other hand, leads to an increase in the TFP growth
rate of only 0.12 percentage points and a reduction in the markup of
0.4 percentage points. Trade liberalization, primarily quota–coverage
reduction from 10% to 30%, during the entire sample period raises the
annual TFP growth rate permanently by over 2 percentage points.3
Thus, we see that trade reforms generally lead to productivity
increases and that the reduction in input tariffs has a much bigger
productivity-enhancing impact than an output tariff reduction. Pavcnik
(2002) for Chile and Fernandes (2007) for Colombia also find a positive
impact of trade liberalization on productivity, but, for countries outside
Asia, no studies decompose the impacts of output and input tariffs.
As regards productivity growth, Harrison (1994) finds support for an
increase in firm-level productivity growth as a result of trade reforms in
Côte d’Ivoire. At the same time, she concludes that tariff cuts lead to a
reduction in markups. Levinsohn (1993) shows a reduction in markups
as a result of trade liberalization for Turkey. Overall, through these

3
The nominal rate of protection actually rose from 36% to 39%.
28 Trade Adjustment in Asia: Past Experiences and Lessons Learned

channels, trade should increase the average income and at the same time
improve the distribution of real incomes.
It is important to note that empirical studies on the impact of trade on
productivity and/or markups only exist for a handful of Asian countries:
India, Indonesia, the People’s Republic of China, the Republic of Korea,
and Turkey. However, the fact that the results are no different for all the
other non-Asian countries should offer a certain degree of confidence
that these relationships may be generalizable to other Asian countries.
The next logical point to study is the impact of trade on the two
most fundamental labor market outcomes—wages and employment—
since both depend on productivity and markups, as will be explained.

1.3 Trade, Wages, and Employment

Trade benefits the abundant factor and hurts the scarce factor.
Given that most Asian economies are abundant in labor, in particular
low-skilled labor, I expect trade to increase wages, in particular
low-skilled wages, in these economies. However, this result (the
Stolper–Samuelson theorem) is based on the assumption that factors
move freely between sectors, one of the key assumptions of the
Heckscher–Ohlin model. This assumption is unlikely to be valid
in developing Asia, especially in the short to medium run, thereby
reducing the relevance of the Heckscher–Ohlin model to this region.
Even with low levels of education, workers need to have sector- or
even firm-specific skills, and those take time to acquire. Workers
acquire some skills on the job. As a result, workers who become
displaced, through import competition, cannot easily find new jobs in
the expanding sectors. Those who remain employed in the shrinking
sectors may experience wage reductions. Furthermore, if the Stolper–
Samuelson effect holds, even weakly, the wage increase can result in
many firms reducing their employment in response.
However, the procompetitive productivity effect, for which
evidence exists, can result in an increase in firm-level and industry-level
employment and wages. In addition, the destruction of monopoly power,
brought about by import competition, shrinks the wedge between the
price and the marginal cost and thus between the value of the marginal
product and the wage, thereby possibly leading to an increase in the
wage. However, if there is rent sharing, the decline in the rents can
lower the wages that employees receive and/or the employment itself.
In addition, the incentive to remain unionized decreases. With de-
unionization, workers’ bargaining power lessens, representing another
channel through which wages can shrink. In addition, in a monopoly
situation, the seller produces too little and as a result the employment is
 Responses to Trade Opening: Evidence and Lessons from Asia 29

low. Trade, by reducing the monopoly power of domestic firms, might be

able to increase employment through this channel.

1.3.1 Empirical Evidence from Asia

Amiti and Davis (2012) investigate how the average wage of a firm
changes with trade liberalization. They examine the differential effects
of output and input tariff cuts on firms that are purely domestic (do not
export or import) and those that export and/or import. The theoretical
model that guides their estimation is one of fair wages in which workers
receive a fraction of the profits. In other words, more profitable firms pay
higher wages. As a result, greater openness in trade leads to exporting
firms earning higher profits and, therefore, paying higher wages, while
those selling only in the domestic market but facing import competition
suffer from profit reductions and pay lower wages. Additionally, firms
that use imported inputs become more profitable and pay higher
wages, with the effect being smaller and statistically not significant for
nonimporting firms. A 10-percentage-point output tariff cut leads to
a 2.4% reduction in the wage paid by a nonexporting firm but a 2.4%
increase in the wage paid by an exporting firm. A 10-percentage-point
input tariff reduction results in a 2.3% wage increase in firms that do
not import their inputs, while it leads to a 7.5% wage increase in firms
that import at least some of their inputs. The reason for firms that do
not import any of their inputs possibly ending up paying higher wages
due to input tariff cuts might be the competition from other firms in the
labor market or a procompetitive effect on upstream industries.
Here again, the results are robust to controlling for endogeneity using
an instrumental variable approach, which runs the regression in long
differences (5-year differences) and instruments the long-differenced
output tariffs using the industry’s initial share of production workers
in total employment and its interaction with an initial export status
indicator and a nontariff barrier dummy (and a few other variables). For
the first-differenced input tariff variable, the instrument is the initial
input tariff interacted with the initial import status.
Dutt (2003), using industry-level data for India, also finds that real
wages rise after liberalization. He further discovers that real wages
are positively related to import penetration. While trade protection
has no significant effect on the wage level, he finds that wage growth
is negatively related to protection: wage growth is higher for relatively
less protected industries and during years with lower tariffs.
Relatively little research focuses on the impact of trade on
employment at the micro level. Kambhampati, Krishna, and Mitra
(1997) find that, controlling for wages and markups, after trade
30 Trade Adjustment in Asia: Past Experiences and Lessons Learned

liberalization, the firm-level labor demand increased in India by

4%–9%, depending on the industry. Not controlling for wages and
markups produces a statistically insignificant impact of trade reforms
on firm-level employment, due to the mutually opposing channels that
I have described.
Dutt (2003) finds results for employment and employment growth
that are similar to his results for wage and wage growth with respect to
import penetration and protection.
In a study on the Republic of Korea, Mitra and Shin (2012) find that
a 10-percentage-point reduction in industry-level tariff reduces labor
demand at the firm level in the Republic of Korea by about 0.6% and
that a 10-percentage-point increase in the ratio of exports to output
increases this labor demand by 0.7%.
Hasan et al. (2012) investigate the impact of trade liberalization on
industry- and state-level unemployment rates in India. A 10-percentage-
point decrease in the state-level employment-weighted average tariff
rate leads to a 7.5% decline in the state-level unemployment rate. In
addition, a 10-percentage-point reduction in a two-digit industry-level
tariff leads to a 0.08-percentage-point reduction in the probability of
being unemployed within an industry. An increase in the value of the
marginal product of labor brought about by trade liberalization seems to
drive all these results.
For all the Asian economies studied so far, it seems that trade does
not reduce firm wages or employment. In most cases, trade reforms have
led to an increase in wages and employment. In the case of India, it is
also apparent that tariff cuts lead to reductions in unemployment rates.
Empirical work investigating the impact of trade reforms on wages and
employment in non-Asian countries, on the other hand, does not provide
such a positive view. For example, Ravenga (1997) finds that, in addition
to the reduction in the demand for output of domestic import-competing
firms in Mexico, trade reforms led to a destruction of monopoly rents,
which firms shared with workers, in turn becoming another wage- and
employment-reducing effect. Ravenga finds the overall impact of trade
reforms on firm-level wages and employment in import-competing firms
to be negative. She is able to break down the overall negative effect into the
channel through the destruction of quasi-rents and the remaining ones.
Currie and Harrison (1997) find no statistically significant effect of trade
reforms on firm-level employment in Morocco. They conclude that, while
there are positive effects through markup reductions and productivity
increases, there is a negative effect through a switch in demand toward
imported substitutes for domestic products.
Clearly, based on my earlier theoretical discussion, there are many
channels that flow in different directions. For all the Asian economies, for
 Responses to Trade Opening: Evidence and Lessons from Asia 31

which rigorous research investigates the wage and employment effects of

trade reforms, researchers do not find any adverse impact. For non-Asian
countries, however, some adverse effects are apparent, which prompt
caution about generalizing the positive effects to the remaining Asian
countries. As a result, the use of redistribution and adjustment policies
becomes quite important. To add to this, the evidence within Asia that
exists on trade adjustment and the impact of trade on informality, which
I will discuss next, strengthens the case for such policies. As I will argue,
such policies can also create and maintain support for globalization.

1.4 Trade and Informality

Having a job is important for an individual’s well-being. However,
conditional on having a job, the quality of that job also matters. What
determines the quality of a job? One obvious determinant is the wage,
but there are other important determinants as well. These include job
security, whether the job has a pension plan, whether it provides health
insurance, what its working conditions are, and so on.
Some, albeit not all, labor regulations aim to provide some of the
elements of job quality. Firms in the formal sector have to follow their
country’s labor laws and other regulations, such as those related to
corporate taxes. However, adherence to these regulations results in
additional costs for these firms. Consequently, firms in developing
countries often want to remain small and in the informal sector. In
addition, firms in the formal sector employ casual or informal short-term
workers to gain the flexibility to hire and fire them, which is not possible
in the case of permanent workers. Casual or short-term jobs, including
those in informal-sector enterprises, lack adequate job security and
social insurance. Thus, the fraction of employment that is informal is an
important inverse indicator of job quality.
What are the trade-offs here for firms? Remaining small (and in
the informal sector) prevents firms from exploiting economies of scale
and from using modern technology, which is usually cost-effective only
when the scale of production is sufficiently large. In addition, temporary
workers have very little incentive to learn on the job and be efficient, since
a large proportion of the human capital that they acquire through on-
the-job training and experience is firm specific and will not be of much
use elsewhere. However, a contraction in demand makes an industry
less profitable, and, therefore, it is less cost-effective for any firm in that
industry to hire more costly (but more productive) regular (permanent)
workers, while an expansion in demand produces the opposite result.
As mentioned earlier, Nataraj (2011), in her study of India’s formal
and informal manufacturing enterprises, found that a given output tariff
32 Trade Adjustment in Asia: Past Experiences and Lessons Learned

reduction increases informal-firm productivity proportionally much

more than formal-firm productivity, while the comparison is reversed
in the case of an input tariff reduction. This finding is important, as
one expects it to result in the expansion of informal relative to formal
employment due to the former but a reduction due to the latter. Thus,
one expects informality to respond to trade liberalization.
Mitra and Ural (2008), in their study of the Indian manufacturing
sector, find that industry productivity, output, value added, and employment
increase with tariff reductions, with the impact being relatively greater in
states where labor regulations generate relatively flexible labor markets.
However, Sundaram, Ahsan, and Mitra (2013) discover the opposite
effect of trade openness on informal-sector firms with five or fewer
workers. These firms experience a greater increase in output, value
added, and employment due to tariff reductions in the relatively rigid
labor regulation states compared with others. These results indicate that
trade liberalization might reduce informality (the share of employment
or output in the informal sector) in states with relatively flexible labor
regulations and increase it in other states. This might be driven by the need
for formal-sector firms, due to restrictive labor regulations, to outsource
some of their work to informal-sector firms.
The results for India are quite consistent with those that Goldberg
and Pavcnik (2003) find in their study of how the informal sector in
Brazil and Colombia responded to trade liberalization in the 1980s and
1990s. While, in the case of Brazil, Goldberg and Pavcnik do not find any
evidence of a relationship between informality and trade liberalization,
for Colombia they find evidence of an increase in informality as a result
of trade liberalization for only the earlier part of their sample period.
This relationship disappears after the implementation of the labor
regulation reforms that made the Colombian labor market more flexible,
which is the latter part of their sample period.
In another paper, Ahsan and Mitra (2017) find that informality in
India was rising in low-productivity sectors relative to high-productivity
sectors, which were also the sectors that were expanding in relative
output and employment. However, this differential trend disappears
with trade liberalization, possibly due to the need for greater flexibility
in input choice, which the employment of casual workers can provide,
as explained earlier.
McCaig and Pavcnik (2018) study the impact of Viet Nam’s
exports on informality. They find that, as United States (US) tariffs on
exports from Viet Nam to the US fell from 23.4% to 2.4% through the
US–Viet Nam Bilateral Trade Agreement, these exports expanded from
$1.1 billion in 2001 to $5 billion in 2004, and individuals moved from
employment in small, informal enterprises to employment in large,
formal firms. Over the first 2 years after the start of the bilateral trade
 Responses to Trade Opening: Evidence and Lessons from Asia 33

agreement, the proportion of informal workers in the manufacturing

sector decreased from 66% to 60%. The authors also find that industries
with bigger US tariff cuts experience larger reductions in informality.
This movement contributes to aggregate productivity growth of about
1.5%–2.8% annually and economic development.
While it is difficult to find any study that examines the impact of
trade on informality in the People’s Republic of China, a study by Liang,
Appleton, and Song (2016) shows that the proportion of casual
employment in urban areas of the People’s Republic of China increased
from 24% in 2007 to 42% in 2013. The authors put much of the blame for
this on the 2008 New Labor Contract Law, which requires all employers
to write up contracts for each of their employees and to provide social
insurance for workers who have contracts longer than 2 years. However,
the implementation and compliance are far from perfect, and formal firms
often hire short-term workers without contracts despite the law that
prohibits such hiring. In other words, the study shows that a more stringent
labor law has resulted in greater evasion and noncompliance with the law,
so making the law more stringent has been counterproductive. However,
this study does not discuss identification issues. As a result, researchers
cannot rule out one of the causes of the increase in employment informality
being the opening of the People’s Republic of China’s economy.
I next look at a cross-economy study by Fiess and Fugazza (2012),
whose panel dataset includes, among others, a number of Asian
economies such as Bangladesh; Hong Kong, China; India; Indonesia;
Japan; Malaysia; Nepal; Pakistan; the Philippines; the People’s Republic
of China; Singapore; and Sri Lanka. While they find that output
informality rises with trade openness, employment informality falls. This
is possible if there is a large increase in the relative labor productivity of
the informal sector. However, employment informality falling with trade
openness is good news from the point of view of job quality. I also believe
that it is necessary to separate developed and developing economies or
allow an interaction of trade openness or restriction with economy per
capita income, as the relationship between trade and informality for
developed and developing economies can be quite different. In addition,
the interaction of the trade variable with the nature of labor regulations
or the flexibility of labor markets will provide valuable insights.
Thus, it seems fairly clear that, in the presence of labor regulations
that produce rigid labor markets, import competition increases
informality. However, export expansion seems to reduce informality.
Labor reforms can contain, or even reverse, the informality-increasing
effects of trade liberalization. The fact that researchers find the
complementarity between trade and labor market flexibility to hold for
all the Latin American countries studied with respect to these issues
makes it also plausible for the Asian economies not studied so far.
34 Trade Adjustment in Asia: Past Experiences and Lessons Learned

1.5 L
 abor Mobility, Trade Shocks,
and Adjustment Costs
I next consider the adjustment costs incurred by workers moving from
one sector to another in response to trade shocks. Artuc, Lederman, and
Porto (2015) carry out pathbreaking work on this issue. In their first step,
the authors estimate the average labor mobility costs stemming from labor
market frictions using industry-level data (within the manufacturing
sector) on employment and wages. They perform this exercise for several
countries (31 developing countries and 25 developed countries), using a
dynamic model of sectoral employment choices. The labor mobility cost or
the cost of moving incurred by a worker in moving to another sector from
his or her current sector of employment turns out to be a few multiples
of the annual average wage. It is 3.88 times the average wage in South
Asia, 3.95 times in Central Asia, and 3.46 times in East Asia and the Pacific.
Regarding individual countries, it is 2.71 for the People’s Republic of
China, 2.87 for India, 3.34 for Iran, 3.46 for Indonesia, 3.77 for the Republic
of Korea, 4 for Lithuania, 4.47 for Azerbaijan, and 4.89 for Bangladesh. For
most developed countries, the numbers are much lower, for example 1.43
for Finland, 1.70 for Germany, 1.82 for the Netherlands, 2.21 for the US,
and so on. The authors also examine the correlations between mobility
costs in the various countries and country-specific characteristics. Richer
countries have lower mobility costs. The mobility costs are also higher
in countries with a larger proportion of their labor force in “vulnerable
employment” or low-quality jobs, a higher number of procedures to
enforce a contract, and a higher number of days required to export.
Based on these mobility costs, the second stage involves the
estimation of the welfare effects on workers in the food and beverage
sector, following a 30% decline in the price of food and beverages due to
a trade shock. A 9.55% potential welfare increase is reduced to an 8.53%
actual welfare increase due to the presence of mobility costs in India.
For the People’s Republic of China, these numbers are 8.25% and 7.05%,
respectively. The difference is greater in Indonesia, with the numbers
being 11.28% and 9.02%, respectively. Clearly, the difference is increasing
in the mobility costs estimated. If I consider countries with even higher
mobility costs, the difference is much bigger between potential and
actual welfare gains, respectively 12.87% and 8.16% for Bangladesh and
10.38% and 5.23% for the Philippines. Evidently, mobility costs wipe out
a sizable proportion of the welfare gains from trade.
While the initial impact of a food and beverage price decline due to
a trade shock is a decline in real wages in the food and beverage sector,
the real wage starts to rise after some time and reaches a higher steady
state within a few years. The higher the labor mobility cost in a country,
 Responses to Trade Opening: Evidence and Lessons from Asia 35

the longer it takes to converge to the new steady state. While a country
such as the People’s Republic of China with a low mobility cost will reach
95% of the new steady-state real wage in about 3 years, it could take
10–11 years in Bangladesh or the Philippines.
Matusz (2003) also provides a calibration of a dynamic multisectoral
search model of unemployment. He uses the data from the National
Sample Survey Office (NSSO) in India to calculate the duration of
unemployment as 4.4 months and the rate of job separation from
a firm as 2% per year. He uses these to calibrate his model and finds
that the adjustment costs can be up to 60% of the gross benefits from
trade liberalization. This is considerably higher than the figures in the
previous study, but, in general, the broader point is that mobility or
adjustment costs can account for a significant proportion of aggregate
welfare changes. After a trade shock, according to the results of Matusz’s
exercise, it takes the economy over 10 years to reach the new steady state.
The above results are consistent with the cross-country results of
Dutt, Mitra, and Ranjan (2009), who study the impact of trade policy on
unemployment. The dataset includes a handful of Asian economies in
addition to countries from other parts of the world. The authors find a very
interesting response of unemployment to trade liberalization. Initially,
there is a rise in unemployment in the year of liberalization. However,
over a longer period of time, there is a reduction in unemployment
relative to the initial level. Trade liberalization leads to immediate
dislocations of workers, resulting in a short-term spike in unemployment
rates of about 0.6% on average. Over a longer time horizon of 2–3 years,
employment recovers and the rise in unemployment reverses, leading to
a 2.5% decline in the unemployment rate in the long run. The results are
similar at the industry level for India in the study by Hasan et al. (2012).
Thus, informality and labor adjustment costs are real problems in
developing Asia. Furthermore, trade reforms can magnify these problems.
Therefore, next I discuss policy options to address these problems.

1.6 P
 olicies to Tackle Trade-induced
Adjustment and Informality
I have shown that trade adjustment costs, as a result of worker mobility
costs, can destroy a significant part of the welfare gains from trade. In
addition, these costs can lead to an initial decline in real wages in the
sector that is hit with a negative trade shock, followed by a rise to the
higher, new steady state. The transition to this new steady state can be
slow, the speed being inversely related to the magnitude of labor mobility
costs. Hollweg et al. (2014) provide several policy options to tackle the
problems associated with adjustment costs. These policies aim to reduce
36 Trade Adjustment in Asia: Past Experiences and Lessons Learned

the mobility costs that workers incur, such as subsidizing destination-

specific relocation costs, training programs to provide skills specific to
destination sectors, unemployment benefits or insurance, job search
assistance, subsidized employment through public works programs,
announcing trade reforms in advance, or gradual trade liberalization
that would allow for advance planning, skill acquisition, searches, and
so on. They warn about the dangers of more stringent employment
protection laws through more restrictions on the firing of workers, as
such policies would slow down the job creation in expanding sectors
and the transition to the new, better steady state.
Mitra and Ranjan (2011) discuss social protection policies for
workers exposed to external or globalization shocks. They also note
the benefits of expanding unemployment benefits and insurance to
facilitate job searches to enable efficient matching of workers and jobs
but with strict monitoring to ensure that such searches are truly taking
place. Like Hollweg et al. (2014), they do not recommend excessive use
of employment protection policies in the form of firing restrictions,
which often end up being hiring restrictions. The reason for this is that
employers are reluctant to hire workers when they know they cannot
fire them in the event of a negative shock or in the case of incompetence.
Mitra and Ranjan advocate East Asian-style public works programs
that serve the twin purpose of providing interim employment for
people who have lost their jobs in sectors exposed to adverse external
shocks and improving the public infrastructure that is badly needed in
many developing countries. They also argue that these various forms
of unemployment support will build and sustain support for greater
openness of the economy.
As regards informality, there is strong evidence that trade
liberalization will increase it in the presence of restrictive labor
regulations. This is quite clear from Goldberg and Pavcnik’s empirical
work on Latin America as well as my own work on India with coauthors.
In addition, the People’s Republic of China has experienced an increase
in informality due to the recent introduction of more rigid labor
regulations in the presence of substantial openness. Therefore, reforms
of labor regulations are probably the solution.

1.7 Trade, Poverty, and Inequality

Trade can affect poverty and inequality through many channels.
According to Bhagwati (2004), trade raises growth, which, in turn,
reduces poverty. Researchers often consider trade to be “an engine
of growth,” especially in the light of the experiences of several Asian
 Responses to Trade Opening: Evidence and Lessons from Asia 37

economies, including the People’s Republic of China and India, which

have liberalized their trade regimes during the last few decades. The
theoretical literature on trade and growth, however, provides several
different channels heading in opposite directions, as the section on
trade and productivity explained. However, David Ricardo’s work from
centuries ago clearly shows that there are “gains from trade” when trade
is driven by comparative advantage. For policy purposes, this means
that international trade is likely to lead to an increase in real per capita
income. In other words, trade expands the size of the pie. This rise in
real per capita income is economic growth.
Not only economists but also world leaders have been aware for
many decades that growth is a necessary condition for poverty reduction.
Especially in the 1950s and 1960s, when many developing countries had
abysmally low per capita incomes, it was clear, as a matter of simple
arithmetic, that redistribution would not lead to poverty reduction. For
example, perfect equality with a very low per capita income can put
everyone below the poverty line.
While there is empirical evidence, after controlling for reverse
causation through an instrumental variable approach, that trade
increases per capita income (Frankel and Romer 1999; Irwin and Tervio
2002), there is also evidence that the poor usually share a country’s
growth. In fact, the growth in incomes of the poor is no less than the rate
of growth of per capita income (Dollar and Kraay 2002). Additionally,
there are redistributive effects of trade in that trade redistributes in
favor of the abundant factor and away from the scarce factor. In poor
countries, the abundant factor is unskilled labor and those below the
poverty line are all unskilled workers. Thus, trade, by increasing the
incomes of the poor, is expected to raise the poor above the poverty line.
However, the above logic depends very much on intersectoral factor
mobility. In the absence of such mobility, workers in declining sectors
are trapped there, losing incomes and/or jobs. Trade also makes capital
goods cheaper, though these capital goods are often complementary to
skilled labor and not unskilled labor, thereby raising wage inequality and
poverty under certain conditions (Davis and Mishra 2007). Furthermore,
for firms to export successfully, they need to be able to offer goods of
higher quality than under autarky. However, higher-quality goods are
more intensive in skilled labor; thus, trade can increase the demand
for skilled labor and reduce the demand for unskilled labor, thereby
raising poverty and inequality. In other words, in such scenarios, simple
Heckscher–Ohlin predictions will not hold.
I next look empirically at the impact of trade on poverty and
inequality.
38 Trade Adjustment in Asia: Past Experiences and Lessons Learned

1.7.1 India–People’s Republic of China Comparisons

I start here by discussing some basic evidence that I present in Mitra

(2016), in which I discuss the experiences of two large Asian countries:
the People’s Republic of China and India. India’s trade-to-gross domestic
product (GDP) ratio increased from roughly 13% in 1988 to 48% in 2010,
while its average tariff rate fell from 80% to 10% during the same period.
The $1.25-a-day poverty rate decreased during the period from 53% to
32%. The most rapid decline in poverty, from 41.6% to 32.7%, occurred
during 2005–2010, when growth was also the most rapid, in the range of
8%–10% (except for 2008). Based on my cross-country regression, I find
in Mitra (2016) that the increase in trade as a fraction of GDP accounts
for a fourth of the reduction in poverty in India. The income inequality
in India during this period was fairly stable. The Gini coefficient
increased only a little from about 32 to 34, the entire blame for this (or
even more) falling on trade liberalization, based on Mitra’s (2016) cross-
country regression analysis.4 The ratio of the incomes of the top 10% to
the bottom 10% rose by only slightly above 10% from 6.9 to 7.7.
In the People’s Republic of China, the trade-to-GDP ratio rose from
17% in 1984 to 70% in 2005 but then fell to 62% in 2008 and 49% in
2009 due to the Great Recession. The country’s average tariff decreased
from 32% to 4% between 1984 and 2010, while its $1.25-a-day poverty
rate fell from 69% to 12% during this period. Based on my cross-country
regression in Mitra (2016), the increase in trade as a fraction of GDP
accounts for one seventh of the reduction in poverty in the People’s
Republic of China. The residual is possibly attributable to growth that
other factors drive, such as infrastructure and skill development, as
well as policies to promote labor market flexibility. However, this topic
requires rigorous investigation. In addition, the country’s Gini coefficient
increased from 28 to 42 during 1984–2009, one fifth of which is due to
trade liberalization, based on Mitra’s (2016) cross-country regression
analysis. The ratio of the incomes of the top 10% to the bottom 10% rose
from 6 to 18 during the same period.
The comparison makes clear that the People’s Republic of
China, where inequality increased much more, actually performed
much better in poverty reduction than India. While both India and
the People’s Republic of China have grown rapidly during the last
decades, the People’s Republic of China’s growth performance has
been significantly better than India’s and has remained steady at that

4
While the actual increase in inequality, according to the Gini coefficient, was 2 points,
the increase predicted by trade liberalization (the tariff reduction that actually took
place) was 3.5 points.
 Responses to Trade Opening: Evidence and Lessons from Asia 39

rate for a longer period of time. This relatively rapid growth has led
to faster poverty reduction. It is possible that the People’s Republic of
China was spending more of its tax revenues on infrastructure, while
India was using them mainly for redistributive purposes and public
works programs that were not so productive. Thus, equality was
unable to rise substantially in India, but its poor performance with
regard to infrastructure probably hurt its growth performance as well
as its progress in poverty reduction.

1.7.2 T
 rade and Poverty: Astructural, Reduced-form
Intra-country Studies from Asia

I next look at some intra-country studies on trade, poverty, and

inequality. I start with studies based on a direct, astructural, reduced-
form approach. Topalova (2007) examines a panel of districts in India
by creating measures of rural, urban, and overall poverty rates (the
proportion of people below the poverty line) and a measure of district-
level protection, which is a weighted average of industry-level tariff
rates, the weights being the initial share of the labor force in individual
districts (also calculated at the rural, urban, and overall levels). Topalova
finds that districts in which rural workers were more exposed to a rise in
import competition (in the form of a greater reduction in the weighted
average tariff ) experienced a relatively slower reduction in rural poverty.
Compared with a district that did not experience any change in tariffs, a
district that experienced the mean level of tariff reduction witnessed a
2-percentage-point rise in poverty. In the case of urban poverty, Topalova
does not find a statistically significant relationship with district-level
protection, but the coefficient sign is the same as in the case of rural
poverty. In fact, she describes one of her findings as trade liberalization
leading to a “significant setback” in rural poverty reduction (equaling
about 15% of the poverty reduction that took place in India in the 1990s).
Hasan, Mitra, and Ural (2007), later updated by Cain, Hasan, and
Mitra (2012), perform a cross-state panel analysis for India (as opposed
to Topalova’s cross-district analysis). Apart from considering states,
they examine NSSO regions (roughly three times the number of states).
Their main finding is that states where workers were more exposed to
foreign competition, through a greater employment concentration in
industries that were more open to trade, had lower rural, urban, and
overall poverty rates. States with greater trade liberalization (a greater
reduction in employment-weighted tariffs) also experienced greater
poverty reduction. These effects were more pronounced in states
with labor laws that allowed more flexible labor markets, greater road
density, and greater financial development. The authors find that trade
40 Trade Adjustment in Asia: Past Experiences and Lessons Learned

liberalization between 1987 and 2004 led to a 38% reduction in poverty.

A 1-percentage-point greater reduction in the employment-weighted
average tariff led to a 0.57% additional reduction in the poverty rate.
There could be a number of reasons for the differences in results
between Topalova and Hasan and Mitra and their coauthors. First,
results can differ because compositional changes can drive poverty
reduction (relatively poor districts within a state shrinking and rich
districts expanding). This is a plausible story, since researchers find
that, while labor is quite immobile between states, there is no such
evidence for the lack of mobility of workers between districts within a
state. There are other differences, such as differences in the treatment
of nontradable sectors in the calculation of the weighted protection,
the noninclusion of the 1993 NSSO round in Topalova’s study, and the
greater variety of protection measures used in the Cain–Hasan–Mitra–
Ural studies.
Another study (Mukim and Panagariya 2012) shows that, contrary
to the previous claims, socially disadvantaged classes have also
experienced declines in poverty rates during the period since the trade
reforms, with some evidence existing that trade reforms led to a decline
in their poverty.
Kis-Katos and Sparrow (2015) conduct a similar analysis of the
impact of the Indonesian trade reforms on poverty for 259 Indonesian
districts in the period 1993–2002. One difference between the studies
on India and this one on Indonesia is that these authors examine both
the employment-weighted output and the input tariffs, with the unit
of observation being a district every 3 years from 1993 to 2002. The
measures that the authors use are the poverty rate, the poverty gap, and
the squared poverty gap. Kis-Katos and Sparrow find that poverty (as
defined through any of the three measures) declines with a reduction
in the input tariffs, but that output tariff reductions increase poverty.
A 1-standard-deviation (or 2-percentage-point) larger reduction in the
employment-weighted input tariff leads to a reduction in the poverty rate
that is half a standard deviation (6.7 percentage points) greater. Running
these regressions for separate education levels (no education, primary
education, junior secondary education, and senior secondary education),
the two extreme education levels seem to drive the results. Furthermore,
the impact of input tariff reductions seems to work through an increase in
the share of working adults in the population, and there is some evidence
of this happening through an increase in the wage rate, possibly due to
an increase in firm productivity, as Amiti and Konings (2007) show for
the Indonesian case. In the case of wage increases, the channel is more
pronounced for medium-skilled workers, while the worker participation
channel works mainly for low-skilled workers.
 Responses to Trade Opening: Evidence and Lessons from Asia 41

McCaig (2011) studies the impact of tariff reductions in the US as

part of the bilateral trade agreement between the US and Viet Nam on
the poverty rates in Viet Nam's provinces. He constructs a province-
specific US import tariff using employment levels in various industries
within a province as weights. McCaig finds that there is negligible
interprovince migration, making this analysis meaningful. A 1-standard-
deviation reduction in the weighted average US tariff that a province
faces leads to a 33%–40% reduction in poverty within 2 years.
Thus, there seems to be considerable evidence that unilateral trade
reforms have reduced poverty at the state level in India (although
the district-level evidence available so far is different). In the case of
Viet Nam, state-level poverty declined as a result of the US reciprocal
tariff cuts. In addition, input tariff reductions have led to poverty
reductions in Indonesian provinces, while output tariff reductions
have led to slight increases in the incidence of poverty. Research
finds that trade liberalization had a poverty-reducing effect in Poland
through greater wage increases in labor-intensive industries that also
experienced deeper tariff cuts (Goh and Javorcik 2007). Furthermore,
in the 1990s, Mexican provinces that were more open to foreign direct
investment and imported and exported more relative to the value of
their output experienced greater poverty reduction relative to the more
closed provinces based on these measures (Hanson 2007). Goldberg and
Pavcnik (2007b), on the other hand, find that urban poverty is unrelated
to tariffs in the Colombian case but negatively related to the volume of
competing imports.
Overall, as long as the right kinds of complementary domestic
policies and institutions are in place, trade liberalization has a favorable
impact on poverty reduction in Asia, and this also applies to many other
parts of the world.
Next, I discuss studies that are more model driven.

1.7.3 T
 rade and Welfare: Empirical General Equilibrium
Studies from Asian Economies
I now move on to studies that use the approach called empirical general
equilibrium analysis, which Porto (2006) pioneered. These studies
focus on the changes in the cost of consumption as measured using the
compensating variation of the various price changes, both in tradable
and in nontradable industries, as a consequence of trade liberalization
(mainly tariff changes) as well as changes in wages resulting from tariff
declines. In the formula for compensating variation, price changes
interact with budget shares, which researchers can allow to vary
across income classes. Ural Marchand (2012), in an empirical general
42 Trade Adjustment in Asia: Past Experiences and Lessons Learned

equilibrium study for India, allows wage responses to vary by skill level
and age. For that purpose, she creates a quasi-panel based on skill level,
age, and industry over time, since the available data are repeated across
sections and not a true panel. She also assumes no interindustry labor
mobility, so wages for different skills and ages in an industry respond
only to changes in that industry’s price. Unlike Porto, she assumes no
impact of tariff changes on nontradable prices. However, like most of
the literature, Ural Marchand also ignores land and capital income.
The benefit of this analysis is that it is possible to compute the welfare
changes at each point in the income distribution. The price transmission
of tariffs in this analysis can differ between rural and urban areas as well
as the remoteness of the location from the nearest port. Overall, Ural
Marchand finds a pro-poor effect of trade reforms in India in that the
poor benefited proportionally more than others. Her analysis shows that
households at all levels of per capita expenditure benefited from the
reforms. Both the consumption and the wage effects separately benefit
the poor, especially those significantly below the poverty line relative
to those just below it. The benefits are greater for households in urban
areas than for those in rural areas and for those in relatively less remote
areas due to muted transmission of the price effects of tariffs. Over the
entire 1988–2000 period studied, the overall estimated welfare gain
was 27% for those at the lowest per capita expenditure levels in rural
areas and 13% at the highest levels in rural India. For urban areas, these
numbers are 40% and 18%, respectively.
Han et al. (2016) evaluate the pass-through of tariff reductions
into price reductions and through that the impact on urban poverty
in the People’s Republic of China. Having calculated the pass-through
into tradable prices (allowing for this pass-through to change with
the share of the private sector in a city’s economy) and the general
equilibrium impact of tradable price changes on nontradable prices, it is
possible to calculate the percentage welfare change (through a change
in the cost of consumption) for each urban household in the People’s
Republic of China using a compensating variation formula. While the
overall percentage increase in welfare is about 7%, poor households
experience about a 14% increase in welfare. The percentage gain in
welfare keeps decreasing with the overall household expenditure. This
is not surprising, since tradables, such as food, clothing, and household
appliances, form a bigger share of the household budget in the case of
relatively poorer households. Richer households buy relatively more
expensive nontradables, such as high-quality education, health care,
housing, and entertainment. Both the tariff pass-through and the welfare
gain from a given tariff reduction increase with the share of the private
sector in the economy.
 Responses to Trade Opening: Evidence and Lessons from Asia 43

Seshan (2014) performs a similar empirical general equilibrium

analysis for Viet Nam for 1993–1998. The author modifies the approach to
incorporate household production in agriculture, in which consumption
and supply decisions (including input use decisions) are not separable.
This is an important feature of agriculture in Viet Nam. Given that relaxing
export restrictions on agricultural products, along with reducing import
restrictions on goods such as chemicals and fertilizers, was an important
part of the overall trade reforms in Viet Nam, making this change to the
model was essential. Seshan finds both overall poverty and inequality
reduction arising from trade liberalization. While rural poverty fell by
9 percentage points, urban poverty rose slightly. Moreover, Seshan finds
that trade liberalization was responsible for a third of the decline in
overall poverty and half of the decline in rural poverty during this period.
Thus, based on the empirical general equilibrium welfare studies,
we find that, while the welfare impact of trade reforms in Asia has been
positive, it is special in that it has been pro-poor. In fact, Porto (2006)
also finds such an impact of Mercosur on Argentina. While Mercosur
is a customs union that people expect to lead to both trade creation
and diversion, the focus of Porto’s study is only trade creation. In that
respect, it is not especially different from the studies on Asia. However,
certain assumptions of his model differ slightly, especially compared
with the Indian study, in that he allows for some intersectoral labor
mobility as well as the possibility of tariff cuts to affect nontradable
prices. In addition, unlike Ural Marchand, he does not allow the price
transmission of tariff cuts to vary by distance from the nearest port or by
rural and urban areas, nor the wage responses to vary by age, skill level,
and so on. Furthermore, unlike Seshan, he does not allow for household
production in agriculture, in which consumption and production
decisions occur jointly.

1.7.4 T
 rade and Inequality: Empirical Evidence from Asia
Overall Income Inequality
While the above studies focus mainly on wage inequality, the question
as to whether trade affects overall income inequality is also important.
In Mitra (2016), I show, using cross-country regressions for 46 countries
over the period 1981–2013, that a 10-percentage-point tariff reduction
raises inequality, as measured by the Gini coefficient (on a 0–100 scale),
by half a point. Based on these regressions, all the blame (and more)
for the slight increase in inequality in India can be placed on trade
reforms, while for the People’s Republic of China, trade liberalization is
responsible for one fifth of the inequality increase. A country-by-country
examination of what happened to inequality following trade reforms,
44 Trade Adjustment in Asia: Past Experiences and Lessons Learned

however, does not produce any clear patterns (Goldberg and Pavcnik
2007a).
Krishna and Sethupathy (2012) construct the Theil index of
inequality for the various states and for rural and urban areas for all the
years of the “thick” NSSO rounds in India in 1988–2005. The advantage
of this index is that it is additively separable into “within-group” and
“between-group” inequality (within and between states and within and
between rural and urban areas). The authors find that inequality, 70%
of which is within group, immediately after the reforms first decreased
during 1988–1994, then increased during 1994–2000, and decreased
thereafter. These results are robust to the use of other measures of
inequality. In addition, protection does not seem to be significantly
related to inequality.

Wage Inequality
Kumar and Mishra (2008) focus on the impact of trade liberalization
on the industry wage premium and overall wage inequality in India for
1983–2000. They estimate the value of the three-digit industry fixed
effects in Mincerian wage regressions run year by year with individual-
level household survey data from the NSSO for each survey round on
individual age, employment, and educational and other demographic
characteristics. The study controls for state and occupational indicators.
The final wage premiums take the form of percentage deviations from
the average industry for each year. Pooling all these wage premiums for
the various industries over all the years to create a panel, the authors
regress the industry wage premium on the nominal rate of protection,
the nontariff barrier coverage ratio, and the import penetration ratio.
The two preferred specifications are the ones in levels with year and
industry fixed effects and those in first differences with year effects.
A 1-percentage-point reduction in the industry’s import tariff leads
to a 0.17% increase in the industry wage premium, which the authors
explain using the available evidence on the procompetitive effects of
tariff reductions on productivity. To the extent that industries with a
larger share of unskilled workers in total employment experienced a
greater tariff reduction, the wage inequality must have decreased. The
authors confirm this by running the above regression separately for
skilled and unskilled workers. The results in Kumar’s and Mishra’s
paper are qualitatively robust to controlling for endogeneity through the
use of instrumental variables, namely 1980 nominal rates of protection
interacted with foreign exchange reserves as well as the initial share
of unskilled workers in employment interacted with foreign exchange
reserves. The results are also robust to controlling for gross fixed capital
formation.
 Responses to Trade Opening: Evidence and Lessons from Asia 45

Amiti and Cameron (2012) study the impact of trade liberalization

in Indonesia on wage inequality (measured by the ratio of the wage
rate of nonproduction to that of production labor) at the firm level for
1991–2000. Using five-digit output tariff rates and the share of various
inputs at the firm level as weights to arrive at input tariffs, the authors
find that a 10-percentage-point input tariff reduction lowers wage
inequality by 2.6% on average for all firms but by 4.5% for importing
firms. For firms with imports as a share of the value of all inputs that
are in the top 10%, this effect is 8.5%. This shows that, while imported
inputs might be, on average, substitutes for skilled workers, they are
probably complements for unskilled workers. The regression of firm-
level skill intensity (the ratio of nonproduction to production labor)
on the interaction of importing status and input tariff confirms this,
showing that importing firms reduce their skill intensity with input tariff
liberalization. Output tariffs seem to have no impact on wage inequality.
Chen, Yu, and Yu (2013) investigate the impact of input tariff cuts on
wage inequality between skilled and unskilled workers within firms in
the People’s Republic of China. The authors argue that skilled workers
share the profits of the firm primarily so that they are incentivized to
perform well rather than due to the presence of firm–worker bargaining.
Since input tariff reductions increase profits, skilled workers’ wages rise
relative to those of unskilled workers, whose wages firms determine in
a perfectly competitive labor market. The authors find evidence that
input trade liberalization, in the presence of profit sharing between
skilled workers and firms, leads to an increase in skilled–unskilled wage
inequality. Running equations in first differences, they instrument the
first-differenced tariff with the lagged tariff. Interestingly, their result
is the opposite of Amiti and Cameron’s (2012) finding for Indonesia.
The coefficient is smaller when the sample includes processing firms,
demonstrating that this channel is not valid for processing firms.
While there is evidence that trade liberalization reduced wage
inequality in India and Indonesia, the impact was the reverse in the
People’s Republic of China. The results from Latin America are similar
to those from the People’s Republic of China. Feenstra and Hanson (1996,
1997) find that wage inequality rose in response to trade reforms during
1975–1988 in Mexico, primarily from input tariff cuts that resulted in
relatively skill-intensive activities in processing these inputs moving to
Mexico from the US (where these activities were the least skill intensive
in the US but still more skill intensive than the existing production
activities that Mexican workers were already undertaking). Attanasio,
Goldberg, and Pavcnik (2004) find that trade liberalization increases
wage inequality in Colombia. However, Pavcnik et al. (2004) do not find
any impact of tariff cuts on wage inequality in Brazil. The difference
46 Trade Adjustment in Asia: Past Experiences and Lessons Learned

between these results is attributable to differences in labor market

flexibility as well as informal-to-formal sector labor mobility between
Colombia and Brazil. Overall, for Asia as well as the rest of the world,
researchers find mixed results regarding this question. More recent
work by Helpman et al. (2017) and by Krishna, Poole, and Senses (2014)
shows that wage inequality can have a nonmonotonic or increasing
relationship with respect to trade openness. Openness in trade leads
to higher returns to exporting firms from investment in screening to
find the best workers. This screening process leads to the recognition
of worker characteristics that are usually unobservable (especially to
the econometrician). The match quality also improves in exporting
firms with greater openness. As a result, exporting firms are able to pay
higher wages. As trade costs fall, initially only a few firms export; but
with further reductions in trade costs, more and more firms export.
This can contribute to a nonmonotonic response of wage inequality to
greater trade openness. In fact, it is possible in this case, as Helpman et
al. (2017) show, for inequality to increase first and then start to decrease.
They show the latter part in their model but not in the data, as trade
costs have not yet fallen to that level in Brazil.
I next move on to labor share, which researchers view as being
negatively correlated with income inequality.

Labor Share
Recently, there has been an interest in the factor shares or the so-called
functional distribution of income. While the share of labor had remained
constant for many decades all over the world, the past 2–3 decades have
witnessed a decline in this share in many parts of the world, especially
in developing countries (International Labour Organization 2011).
At the same time, globalization has taken place at a rapid pace all over the
world, especially trade reforms. Thus, there is a tendency to blame the
declining labor share on trade. The issue of declining labor shareneeds
further investigation for a few reasons. First, the rich derive their income
mainly from capital and land, the distribution of which is highly unequal
throughout the world. Second, the poor derive their income mainly from
their raw labor, and labor incomes are relatively more equally distributed.
Third, overall inequality and the share of labor are strongly negatively
correlated (Atkinson 2009). Last, while globalization has taken place at a
rapid pace, skill-biased technological change has occurred equally rapidly.
Therefore, the exact cause of the declining labor share and whether
globalization has sped up this decline or whether the decline would have
been even greater in the absence of globalization are unknown.
For India, Ahsan and Mitra (2014) use firm-level data to investigate
the impact of tariff cuts on the share of the wage bill in firm sales.
 Responses to Trade Opening: Evidence and Lessons from Asia 47

This study indicates something quite nuanced. For relatively small

firms (those that lie in the bottom third of the distribution), which
also turn out to be relatively labor intensive, based on the within-firm
variation and controlling for macro effects, the authors find that tariff
reductions increase labor share, while for large firms (in the top third
of the distribution), which are relatively less labor intensive, labor share
decreases with tariff reductions. While the elasticity of the labor share
with respect to the industry-level tariff is −0.5 in the former set of firms,
it is 0.8 for the latter. The study also finds that there is a decline in the
bargaining power of workers in the sharing of profits across the board
(as Rodrik [1997] argues and predicts), arising from these tariff cuts, but
there is an offsetting force coming from the destruction of the monopoly
power of domestic firms, which shrinks the wedge between the value of
the marginal product and the wage. The second force is the dominant
one in the case of small firms.
Mitra and Shin (2014) find qualitatively similar but quantitatively
slightly different results for the Republic of Korea using firm-level
survey data both for the labor share and for the bargaining power of
workers. Kamal, Lovely, and Mitra (2014), on the other hand, find that
48 Trade Adjustment in Asia: Past Experiences and Lessons Learned

in the case of firms in the People’s Republic of China between 1998

and 2007, tariff cuts lead to increases in firm-level labor share across the
board. This effect is stronger for coastal firms than for interior firms and
varies by ownership type (domestic private, foreign, and state owned).
Both input and output tariff cuts seem to have qualitatively similar
effects. The effective rates of protection have qualitatively similar effects
on the share of wages in the firm value added.
Regarding the case of India, while a reduction in the bargaining
power of workers (through either direct competition from imported
inputs or through indirect competition from goods produced by foreign
inputs and labor) is an important force arising from trade liberalization,
it is also possible that the decline in rents leads to declining unionization.
This can happen due to the reduced incentives to meet the costs
of forming new unions and maintaining old ones. Using household
survey data for 1993–2004 from India, from which the authors create
measures of union presence (the proportion of workers in an industry in
unionized activities) and union membership (the proportion of workers
in an industry who are union members) in various industries by state,
Ahsan, Ghosh, and Mitra (2017) find evidence that, in net-importer
industries, a 10-percentage-point reduction in the import tariff led to a
0.8-percentage-point reduction in the proportion of workers working in
unionized activities as well as in the proportion of workers who are union
members. This investigation also finds evidence from firm-level data that
industry quasi-rents per plant declined with tariff cuts (where quasi-rents
are total sales minus material and fuel costs minus the wage bill evaluated
at the prevailing nonunion wage in the household survey data).

Trade and Labor-demand Elasticities

Rodrik (1997) argues that trade liberalization makes the labor demand
more elastic. Through trade liberalization, cheaper and a greater variety
of imported substitutes for domestically produced goods are available.
This makes the demand for domestically produced goods more elastic.
The demand for labor, being derived from the demand for goods and
services, also becomes more elastic. In addition, imported inputs can
directly substitute domestic labor.
The study of the impact of trade on labor-demand elasticities is
important for a few reasons. An increase in the magnitude of labor-
demand elasticity results in lower bargaining power for workers relative
to employers, greater volatility in employment and wages for the given
volatility in productivity, and a greater negative impact of rises in input
and fuel costs on workers.
Hasan, Mitra, and Ramaswamy (2007) show that, with tariff cuts,
the absolute value of the elasticity of labor demand at the industry level
in Indian manufacturing rose from 0.076 to 0.186 in states with labor
 Responses to Trade Opening: Evidence and Lessons from Asia 49

regulations that ensure a relatively rigid labor market, while in the other
states it increased from 0.206 to 0.316. These calculations are based on
a change in the average manufacturing tariff from 150% in 1988 to 40%
in 1997. In contrast, Slaughter (2001) finds no systematic impact of trade
liberalization on labor-demand elasticities for the US, while this impact
is statistically insignificant for Turkey (Krishna, Mitra, and Chinoy 2001).
For the Republic of Korea, Mitra and Shin (2012) find weak evidence
of trade liberalization increasing labor-demand elasticities. However,
there is some evidence that the Republic of Korea's exports have
increased their firm-level labor-demand elasticities. A 10-percentage-
point increase in the share of exports in firm-level output leads to an
increase in absolute labor-demand elasticity of up to 0.04.

1.8 Discussion and Concluding Remarks

As is apparent from the above discussion, trade has been beneficial
to Asian economies through a number of channels and in many
different respects: through higher productivity, lower markups
through import competition, higher wages, higher employment, lower
unemployment, and, above all, lower poverty rates.5 At this juncture,
however, two caveats are in order. First, the beneficial impacts might
not be generalizable to Asian countries that so far remain unstudied.
Second, there are a few adverse consequences of trade even in the Asian
countries that are studied rigorously, and public policy needs to address
these. First, trade can increase informality, especially in the presence
of labor market rigidities. Second, there is an adverse effect stemming
from trade adjustment as a result of worker mobility costs. Then, there
is rising income inequality. However, the diversion of government funds
from social expenditures to infrastructure building during this period of
globalization is quite possibly the reason for some countries not being
able to contain the rise in inequality. However, at the same time, this
could be the reason for rapid growth and, in turn, through its “pull-up”
effect (Bhagwati 2004), the reduction in poverty.
My review of the evidence also shows that, in some respects,
globalization has put pressure on workers. For example, their bargaining
power relative to their employers has decreased, most likely due to the
greater options available to employers in terms of obtaining inputs from
abroad or even the wider variety of imported final goods and services
available to consumers, thereby making the services of domestic
workers more replaceable. This reflects in the rise of labor-demand

5
For a summary of the research discussed here, please see the literature table in the
appendix at the end of this chapter.
50 Trade Adjustment in Asia: Past Experiences and Lessons Learned

elasticities in the two Asian countries (India and the Republic of Korea)
for which evidence is available and in some other countries outside Asia.
As explained earlier, this, apart from reducing the bargaining power
of labor, makes workers’ incomes more volatile and their jobs more
uncertain. In addition, firms have to bear a higher burden of rises in
input and fuel costs.
Bhagwati (2004) argues that “appropriate policies” are necessary
to reap and harness the gains from trade. For example, he reasons that
countries can specialize away from goods for which the world prices
are falling steeply but still specialize according to their comparative
advantage. He is also in favor of other complementary “appropriate
policies,” especially with respect to agriculture, financial development,
property rights, infrastructure building, and so on. Cain, Hasan, and
Mitra (2012) find that Indian states that were financially more developed,
had higher road density, were closer to ports, and had labor regulations
that enabled more flexible labor markets, were able to achieve a greater
reduction in urban poverty as a result of trade reforms. The work of
Krishna, Mitra, and Sundaram (2010) also supports this result, finding
that “lagging” regions or states (those that are distant from their
respective nearest ports) within South Asia have been relatively less
successful in reducing poverty through trade reforms.
Thus, the above evidence stresses the need for better infrastructure,
especially a denser and better network of roads and greater investment
in the building of new ports. Also needed would be a larger number of
bank branches and labor regulations that can provide workers with the
right kind of protection without sacrificing flexibility for employers
to respond nimbly to the demand and supply shocks that they face in
a more globalized environment. In addition, social protection and
“appropriate” redistributive policies would help make sure that the
losers from globalization are appropriately compensated to minimize
the chances of a reversal of reforms (which policy makers should not
underestimate). Countries could use public works programs, which
provide the unemployed and underemployed with productive job
opportunities, as much as possible as a means of social protection and, at
the same time, infrastructure building, which is essential for maximizing
the gains from trade in its many forms. Public works programs can be
especially important in Asian developing countries where the informal
sector accounts for a substantial share of employment.
 Responses to Trade Opening: Evidence and Lessons from Asia 51

References
Ahsan, R. N., A. Ghosh, and D. Mitra. 2017. International Trade and
Unionization: Evidence from India. Canadian Journal of Economics.
50 (2). pp. 398–425.
Ahsan, R. N., and D. Mitra. 2014. Trade Liberalization and Labor’s Slice
of the Pie: Evidence from Indian Firms. Journal of Development
Economics. 108. pp. 1–16.
____. 2017. Can the Whole Actually Be Greater Than the Sum of Its
Parts? Lessons from India’s Growing Economy and Its Evolving
Structure (with Reshad N. Ahsan). In M. McMillan, D. Rodrik, and
C. Sepulveda, eds. Structural Change, Fundamentals and Growth:
A Framework and Country Studies. Washington, DC: International
Food Policy Research Institute. pp. 39–79.
Amiti, M., and L. Cameron. 2012. Trade Liberalization and the Wage
Skill Premium: Evidence from Indonesia. Journal of International
Economics. 87 (2). pp. 277–287.
Amiti, M., and D. R. Davis. 2012. Trade, Firms, and Wages: Theory and
Evidence. Review of Economic Studies. 79 (1). pp. 1–36.
Amiti, M., and J. Konings. 2007. Trade Liberalization, Intermediate
Inputs, and Productivity: Evidence from Indonesia. American
Economic Review. 97 (5). pp. 1611–38.
Artuc, E., D. Lederman, and G. Porto. 2015. A Mapping of Labor Mobility
Costs in the Developing World. Journal of International Economics
95 (1). pp. 28–41.
Atkinson, A. B. 2009. Factor Shares: The Principal Problem of Political
Economy? Oxford Review of Economic Policy. 25 (1). pp. 3–16.
Attanasio, O., P. K. Goldberg, and N. Pavcnik. 2004. Trade Reforms and
Income Inequality in Colombia. Journal of Development Economics.
74. pp. 331–66.
Bas, M., and O. Causa. 2013. Trade and Product Market Policies in
Upstream Sectors and Productivity in Downstream Sectors: Firm-
Level Evidence from China. Journal of Comparative Economics.
41 (3). pp. 843–62.
Bhagwati, J. 2004. In Defense of Globalization. New York: Oxford
University Press.
Brandt, L., J. Van Biesebroeck, L. Wang, and Y. Zhang. 2017. WTO
Accession and Performance of Chinese Manufacturing Firms.
American Economic Review. 107 (9). pp. 2784–820.
Cain, J., R. Hasan, and D. Mitra. 2012. Trade Liberalization and Poverty
Reduction: New Evidence from Indian States. In J. Bhagwati and
A. Panagariya, eds. India’s Reforms: How They Produced Inclusive
Growth. New York: Oxford University Press. pp. 91–185.
52 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Chen, B., M. Yu, and Z. Yu. 2013. Wage Inequality and Input Trade
Liberalization: Firm-Level Evidence from China. Mimeo. SSRN:
https://ssrn.com/abstract=2329292 or http://dx.doi.org/10.2139
/ssrn.2329292.
Currie, J., and A. E. Harrison. 1997. Sharing Costs: The Impact of
Trade Reform on Capital and Labor in Morocco. Journal of Labor
Economics. 15. pp. 44–71.
Davis, D. R., and P. Mishra. 2007. Stolper–Samuelson is Dead: And Other
Crimes of Both Theory and Data. In A. Harrison, ed. Globalization
and Poverty. Chicago, IL: University of Chicago Press. pp. 87–108.
De Loecker, J., P. K. Goldberg, A. K. Khandelwal, and N. Pavcnik.
2016. Prices, Markups, and Trade Reform. Econometrica. 84 (2).
pp. 445–510.
Dollar, D., and A. Kraay. 2002. Growth Is Good for the Poor. Journal of
Economic Growth. 7 (3). pp. 195–225.
Dutt, P. 2003. Labor-Market Outcomes and Trade Reforms: The Case
of India. In R. Hasan and D. Mitra, eds. The Impact of Trade on
Labor: Issues, Perspectives and Experiences from Developing Asia.
Amsterdam, Netherlands: North Holland. pp. 245–74.
Dutt, P., D. Mitra, and P. Ranjan. 2009. International Trade and
Unemployment: Theory and Cross-National Evidence. Journal of
International Economics. 78 (1). pp. 32–44.
Fan, H., X. Gao, Y. A. Li, and T. A. Luong. 2017. Trade Liberalization
and Markups: Micro Evidence from China. Journal of Comparative
Economics. 46 (1). pp. 103–130.
Feenstra, R. C., and G. H. Hanson. 1996. Foreign Investment, Outsourcing
and Relative Wages. In R. C. Feenstra et al., eds. Political Economy of
Trade Policy: Essays in Honor of Jagdish Bhagwati. Cambridge, MA:
MIT Press. pp. 89–127.
____. 1997. Foreign Direct Investment and Relative Wages: Evidence
from Mexico’s Maquiladoras. Journal of International Economics.
42. pp. 371–93.
Fernandes, A. M. 2007. Trade Policy, Trade Volumes and Plant-Level
Productivity in Colombian Manufacturing Industries. Journal of
International Economics. 71 (1). pp. 52–71.
Fernandez, R., and D. Rodrik. 1991. Resistance to Reforms: Status-Quo
Bias in the Presence of Individual-Specific Uncertainty. American
Economic Review. 81 (5). pp. 1146–55.
Fiess, N., and M. Fugazza. 2012. Informality and Openness to Trade:
Insights from Cross-Sectional and Panel Analyses. Margin – The
Journal of Applied Economic Research. 6 (2). pp. 235–75.
Frankel, J., and D. Romer. 1999. Does Trade Cause Growth? American
Economic Review. 89 (3). pp. 379–99.
 Responses to Trade Opening: Evidence and Lessons from Asia 53

Goh, C.-C., and B. S. Javorcik. 2007. Trade Protection and Industry Wage
Structure in Poland. In A. E. Harrison, ed. Globalization and Poverty.
Chicago, IL: University of Chicago Press. pp. 337–72.
Goldberg, P. K., A. K. Khandelwal, N. Pavcnik, and P. Topalova. 2010.
Imported Intermediate Inputs and Domestic Product Growth:
Evidence from India. Quarterly Journal of Economics. 125 (4).
pp. 1727–67.
Goldberg, P. K., and N. Pavcnik. 2003. The Response of the Informal
Sector to Trade Liberalization. Journal of Development Economics.
72 (2). pp. 463–96.
____. 2007a. Distributional Effects of Globalization in Developing
Countries. Journal of Economic Literature. 45 (1). pp. 39–82.
____. 2007b. The Effects of the Colombian Trade Liberalization on Urban
Poverty. In A. E. Harrison, ed. Globalization and Poverty. Chicago,
IL: University of Chicago Press. pp. 241–89.
Grossman, G. M., and E. Helpman. 1991. Innovation and Growth in the
Global Economy. Cambridge, MA: MIT Press.
Hall, R. E. 1988. The Relation between Price and Marginal Cost in US
Industry. Journal of Political Economy. 96 (5). pp. 921–47.
Han, J., R. Liu, B. U. Marchand, and J. Zhang. 2016. Market Competition,
Imperfect Pass-Through and Household Welfare in Urban China.
Journal of International Economics. 100. pp. 220–32.
Hanson, G. H. 2007. Globalization, Labor Income, and Poverty in
Mexico. In A.E. Harrison, ed. Globalization and Poverty. Chicago,
IL: University of Chicago Press. pp. 417–56.
Harrison, A. E. 1994. Productivity, Imperfect Competition and Trade
Reform: Theory and Evidence. Journal of International Economics.
36 (1–2). pp. 53–73.
Hasan, R., D. Mitra, and K. V. Ramaswamy. 2007. Reforms, Labor
Regulations and Labor Demand Elasticities: Empirical Evidence
from India. Review of Economics and Statistics. 89 (3). pp. 466–81.
Hasan, R., D. Mitra, P. Ranjan, and R. N. Ahsan. 2012. Trade Liberalization
and Unemployment: Evidence from India. Journal of Development
Economics. 97 (2). pp. 269–80.
Hasan, R., D. Mitra, and B. P. Ural. 2007. Trade Liberalization, Labor
Market Institutions, and Poverty Reduction: Evidence from Indian
States. India Policy Forum, 2006–07.
Helpman, E., O. Itskhoki, M. A. Muendler, and S. Redding. 2017. Trade
and Inequality: From Theory to Estimation. Review of Economic
Studies. 84 (1). pp. 357–405.
Hollweg, C. H., D. Lederman, D. Rojas, and E. R. Bulmer. 2014. Sticky
Feet: How Labor Market Frictions Shape the Impact of International
Trade on Jobs and Wages. Washington, DC: World Bank.
54 Trade Adjustment in Asia: Past Experiences and Lessons Learned

International Labour Organization. 2011. World of Work Report 2011:

Making Markets Work for Jobs. Geneva: International Labour Office.
Irwin, D., and J. M. Tervio. 2002. Does Trade Raise Income? Evidence
from the Twentieth Century. Journal of International Economics.
58. pp. 1–18.
Kamal, F., M. E. Lovely, and D. Mitra. 2014. Trade Liberalization and
Labor Shares in China. Mimeo, Syracuse University.
Kambhampati, U., P. Krishna, and D. Mitra. 1997. The Effect of Trade
Policy Reforms on Labor Markets: Evidence from India. Journal of
International Trade & Economic Development. 6 (2). pp. 287–97.
Kim, E. 2000. Trade Liberalization and Productivity Growth in Korean
Manufacturing Industries: Price Protection, Market Power, and
Scale Efficiency. Journal of Development Economics. 62. pp. 55–83.
Kis-Katos, K., and R. Sparrow. 2015. Poverty, Labor Markets and Trade
Liberalization in Indonesia. Journal of Development Economics. 117
(C). pp. 94–106.
Krishna, P., and D. Mitra. 1998. Trade Liberalization, Market Discipline
and Productivity Growth: New Evidence from India. Journal of
Development Economics. 56 (2). pp. 447–62.
Krishna, P., D. Mitra, and S. Chinoy. 2001. Trade Liberalization and Labor
Demand Elasticities: Evidence from Turkey. Journal of International
Economics. 55 (2). pp. 391–409.
Krishna, P., J. P. Poole, and M. Senses. 2014. Wage Effects of Trade
Reform with Labor Market Frictions and Endogenous Worker
Mobility. Journal of International Economics. 93 (2). pp. 239–52.
Krishna, P., and G. Sethupathy. 2012. Trade and Income Inequality in
India. In Jagdish Bhagwati and Arvind Panagariya, eds. India’s
Reforms: How They Produced Inclusive Growth. New York: Oxford
University Press. pp. 247–78.
Krishna, P., D. Mitra, and A. Sundaram. 2010. Do Lagging Regions Benefit
from Trade? In Ejaz Ghani, ed. The Poor Half Billion in South Asia:
What Is Holding Back Lagging Regions? New Delhi and New York:
Oxford University Press. pp. 137–77.
Kumar, U., and P. Mishra. 2008. Trade Liberalization and Wage
Inequality: Evidence from India. Review of Development Economics.
12 (2). pp. 291–311.
Levinsohn, J. 1993. Testing the Imports-as-Market-Discipline
Hypothesis. Journal of International Economics. 35 (1–2). pp. 1–22.
Liang, Z., S. Appleton, and L. Song. 2016. Informal Employment in
China: Trends, Patterns and Determinants of Entry. IZA Discussion
Paper. No. 10139.
Matusz, S. J. 2003. Trade Policy Reform and Labor Market Dynamics:
Issues and an Agenda for Future Research. In Rana Hasan and
Devashish Mitra, eds. The Impact of Trade on Labor: Issues,
 Responses to Trade Opening: Evidence and Lessons from Asia 55

Perspectives, and Experiences from Developing Asia. Amsterdam:

North-Holland Publishers. pp. 99–125.
McCaig, B. 2011. Exporting out of Poverty: Provincial Poverty in Vietnam
and US Market Access. Journal of International Economics. 85 (1).
pp. 102–13.
McCaig, B., and N. Pavcnik. 2018. Export Markets and Labor Allocation
in a Low-Income Country. American Economic Review. 108 (7).
pp. 1899–1941.
Mitra, D. 2016. Trade, Poverty, and Inequality. In J. Bhagwati, P. Krishna,
and A. Panagariya, eds. The World Trade System. Cambridge, MA:
MIT Press. pp. 55–90.
Mitra, D., and P. Ranjan. 2011. Social Protection in Labor Markets
Exposed to External Shocks. In M. Bacchetta and M. Jansen, eds.
Making Globalization Socially Sustainable. Geneva: International
Labour Organization and World Trade Organization. pp. 199–231.
Mitra, D., and J. Shin. 2012. Import Protection, Exports and Labor-
Demand Elasticities: Evidence from Korea. International Review of
Economics & Finance. 23 (C). pp. 91–109.
____. 2014. Globalization and Workers: Firm-Level Evidence from
Korea. In N. Khor and D. Mitra, eds. Trade and Employment in Asia.
August 2013. London: Routledge. pp. 204–45.
Mitra, D., and B. P. Ural. 2008. Indian Manufacturing: A Slow Sector
in a Rapidly Growing Economy. Journal of International Trade and
Economic Development. 17 (4). pp. 525–60.
Mukim, M., and A. Panagariya. 2012. Growth, Openness and the Socially
Disadvantaged. In J. Bhagwati and A. Panagariya, eds. India’s
Reforms: How They Produced Inclusive Growth. New York: Oxford
University Press. pp. 186–246.
Nataraj, S. 2011. The Impact of Trade Liberalization on Productivity:
Evidence from India’s Formal and Informal Manufacturing Sectors.
Journal of International Economics. 85 (2). pp. 292–301.
Pavcnik, N. 2002. Trade Liberalization, Exit, and Productivity
Improvements: Evidence from Chilean Plants. Review of Economic
Studies. 69. pp. 245–76.
Pavcnik, N., A. Blom, P.K. Goldberg, and N. Schady. 2004. Trade
Liberalization and Industry Wage Structure: Evidence from
Brazil. World Bank Economic Review. 18 (3). pp. 319–343.
Porto, G.G. 2006. Using Survey Data to Assess the Distributional Effects of
Trade Policy. Journal of International Economics. 70 (1). pp. 140–60.
Ravenga, A. 1997. Employment and Wage Effects of Trade Liberalization:
The Case of Mexican Manufacturing. Journal of Labor Economics.
15. pp. 20–43.
Rodrik, D. 1997. Has Globalization Gone Too Far? Washington, DC:
Peterson Institute for International Economics.
56 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Seshan, G. 2014. The Impact of Trade Liberalization on Household

Welfare in a Developing Country with Imperfect Labor Markets.
Journal of Development Studies. 50 (2). pp. 226–43.
Slaughter, M.J. 2001. International Trade and Labor-Demand
Elasticities. Journal of International Economics 54(1). pp. 27–56.
Solow, R. 1957. Technical Change and the Aggregate Production
Function. Review of Economics and Statistics. 39 (3). pp. 312–20.
Sundaram, A., R. Ahsan, and D. Mitra. 2013. Complementarity between
Formal and Informal Manufacturing in India. In Reforms and
Economic Transformation in India, edited by Jagdish Bhagwati and
Arvind Panagariya, pp. 49–85. New York: Oxford University Press.
Topalova, P. 2007. Trade Liberalization, Poverty and Inequality:
Evidence from Indian Districts. In A. E. Harrison, ed. Globalization
and Poverty, pp. 291–335. Chicago, IL: University of Chicago Press.
Topalova, P., and A. Khandelwal. 2011. Trade Liberalization and Firm
Productivity: The Case of India. Review of Economics and Statistics.
93 (3). pp. 995–1009.
Ural Marchand, B. 2012. Tariff Pass-Through and the Distributional
Effects of Trade Liberalization. Journal of Development Economics.
99 (2). pp. 265–81.
Yu, M. 2015. Processing Trade, Tariff Reductions and Firm Productivity:
Evidence from Chinese Firms. Economic Journal. 125 (585).
pp. 943–88.
 Responses to Trade Opening: Evidence and Lessons from Asia 57

Appendix A1
Table A1 Responses to Trade Opening: Summary of Evidence from Asia

Economies
Topic Authors Covered Main Finding
Trade reforms, Krishna and India Statistically significant reductions in markups
productivity Mitra (1998) and increases in productivity growth in firms in
growth, and majority of industries studied.
markups at the
firm level
Trade and Topalova and India Procompetitive impact of output tariff
productivity Khandelwal reduction: lower tariffs lead to higher
(2011) productivity. Much greater increase in
productivity from an equal input tariff
reduction.
Trade, Goldberg et India Trade liberalization led to a considerable
intermediate al. (2010) decline in the prices of intermediate inputs
goods prices, and an increase in their variety.
and variety
Trade De Loecker India Output tariff reductions led to a reduction in
liberalization et al. (2016) firm-level price–marginal cost markups, while
and firm-level input tariff reductions led to an increase in
markups these markups.
Trade Nataraj India Output tariff reductions increased informal
liberalization (2011) firm productivity proportionally much more
and than formal-firm productivity. The reverse was
formal- and the case with input tariff reductions.
informal-firm
productivity
Trade Amiti and Indonesia While a 1%–6% increase in productivity
liberalization Konings can be attributed to a 10-percentage-point
and (2007) reduction in output tariff, firms importing
productivity inputs experience up to a 13% increase in
productivity from a 10-percentage-point input
tariff reduction.
Trade Brandt et al. People’s A 10-percentage-point reduction in output
liberalization, (2017) Republic of tariff leads to 1% markup reduction, while a
markups, and China 10-percentage-point reduction in the input
productivity tariff leads to a 7% markup increase. The
procompetitive effect of a 10-percentage-
point decline in output tariff is 1.7% increase
in total factor productivity (TFP), while for a
10-percentage-point decline in input tariff the
TFP gain is 16%–18%.
continued on next page
 58 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table A1 continued
Economies
Topic Authors Covered Main Finding
Trade Fan et al. People’s Looking separately at firms engaged in
liberalization (2017) Republic of processing trade and other firms, effects in
and markups China Brandt et al. (2017) empirically do not show up
for the former but are seen in the latter.
Trade Yu (2013) People’s With a 10-percentage-point output tariff
liberalization Republic of reduction, firm TFP increases by 9%. A
and firm China 10-percentage-point input tariff reduction
productivity increases firm TFP by only 5%. These impacts
go down with an increase in the share of
processing imports in total imports.
Trade Bas and People’s Firms on or close to the frontier experience
liberalization, Causa Republic of a 0.74% increase in productivity from a
firm (2013) China 1-percentage-point reduction in input
productivity, tariffs. Firms whose productivity is half of
and the domestic technological frontier will see
technological roughly a 0.5% rise in productivity from the
frontier same tariff reduction. The procompetitive
impact of output tariff reductions is
stronger for firms relatively distant from the
technological frontier.
Trade Kim (2000) Republic of A 10-percentage-point reduction in the
liberalization, Korea quota-coverage ratio led to a TFP growth
industry increase of 0.26 percentage points and a
productivity, markup reduction of 1.33 percentage points. A
and markups 10-percentage-point reduction in the nominal
rate of protection led to a TFP growth rate
increase of only 0.12 percentage points and a
markup reduction of 0.4 percentage points.
Trade liberalization overall during the entire
sample period raised annual TFP growth rate
permanently by over 2 percentage points.
Trade Amiti and Indonesia A 10-percentage-point output tariff
liberalization Davis (2012) cut leads to a 2.4% reduction in the
and firm-level average nonexporting firm wage but a
wages 2.4% increase in the average export firm
wage. A 10-percentage-point input tariff
reduction results in a 2.3% wage increase in
nonimporting firms, but a 7.5% wage increase
in firms that import at least some of their
inputs.
Trade Kambhampati, India Controlling for wages and markups, after
liberalization, Krishna, and trade liberalization, firm-level labor demand
firm-level Mitra (1997) increased in India by 4%–9%, depending on
wages, and the industry.
markups
continued on next page
 Responses to Trade Opening: Evidence and Lessons from Asia 59

Table A1 continued
Economies
Topic Authors Covered Main Finding
Trade and Dutt (2003) India Industry-level employment goes up after
industry-level liberalization and is positively related to import
employment penetration. But there is no significant effect
and wages of trade protection on the employment level.
However, employment growth is negatively
related to protection.
Trade and Mitra and Republic of A 10-percentage-point reduction in industry-
firm-level Shin (2012) Korea level tariff reduces firm-level labor demand by
employment 0.6%. A 10-percentage-point increase in the
ratio of exports to output increases firm-level
labor demand by 0.7%.
Trade Hasan et al. India A 10-percentage-point decrease in the state-
liberalization (2012) level employment-weighted average tariff
and rate leads to a 7.5% decline in the state-level
unemployment unemployment rate. Also, a 10-percentage-
point reduction in a two-digit industry-level
tariff leads to a 0.08-percentage-point
reduction in the probability of being
unemployed within an industry.
Trade Sundaram, India Informal sector firms with five or fewer
liberalization, Ahsan, and workers experience a greater increase in
labor market Mitra (2013) output, value added, and employment due
flexibility, and to tariff reductions in the relatively rigid labor
informality regulation states as compared to others. Trade
liberalization might be reducing informality
(the share of employment or output in the
informal sector) in states with relatively
flexible labor regulations and increasing it in
other states.
Informality, Ahsan and India Informality was rising in low-productivity
structural Mitra (2017) sectors relative to high-productivity sectors,
change, which were also the sectors which were
and trade expanding in relative output and employment.
liberalization This differential trend goes away with trade
liberalization.
Exports and McCaig Viet Nam As United States (US) tariffs on exports by
informality and Pavcnik Viet Nam to the US were lowered from 23.4%
(2018) to 2.4% through the US–Viet Nam Bilateral
Trade Agreement, individuals moved from
employment in small, informal enterprises to
large, formal firms. Within the first 2 years,
the proportion of informal workers in the
manufacturing sector went down from 66% to
60%. Industries with bigger US tariff cuts also
experienced larger reductions in informality. This
contributed to aggregate productivity growth of
about 1.5%–2.8% annually.
continued on next page
 60 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table A1 continued
Economies
Topic Authors Covered Main Finding
Labor Liang, People’s The proportion of casual employment in urban
regulations Appleton, Republic of People’s Republic of China increased from
and informality and Song China 24% in 2007 to 42% in 2013, probably due
in an open (2016) to the 2008 New Labor Contract Law (in an
economy open economy).
Trade Fiess and People’s While output informality rises with trade
openness and Fugazza Republic of openness, employment informality falls.
informality (2012) China; India;
Indonesia;
Japan;
Bangladesh;
Pakistan;
Sri Lanka;
Nepal;
Malaysia; Hong
Kong, China;
Singapore; the
Philippines; and
many non-
Asian countries
Adjustment Artuc, People’s The labor mobility cost is 3.88 times the
costs of trade Lederman, Republic of average wage in South Asia, 3.95 in Central
and worker and Porto China, India, Asia, and 3.46 in East Asia and the Pacific. A
mobility costs (2015) Iran, Indonesia, sizable proportion of the welfare gains from
Republic trade are being wiped out by adjustment costs.
of Korea,
Lithuania,
Azerbaijan,
Bangladesh,
the Philippines,
and many
non-Asian
developed and
developing
countries
Policies to Hollweg et al. Cross- Policies aimed at reducing labor mobility
minimize trade (2014) country study, costs include subsidizing destination-
adjustment includes above specific relocation costs, training programs to
costs countries provide skills specific to destination sectors,
unemployment benefits or insurance, job
search assistance, subsidized employment
through public works programs, announcing
trade reforms in advance or gradual trade
liberalization, etc.
continued on next page
 Responses to Trade Opening: Evidence and Lessons from Asia 61

Table A1 continued
Economies
Topic Authors Covered Main Finding
Social Mitra and No specific The study notes the benefits of expanding
protection Ranjan country unemployment benefits and insurance but
policies for (2011) with strict monitoring. The study recommends
workers against excessive use of employment
exposed protection policies and advocates for East
to external Asian-style public works programs. The
authors also argue that various forms of
unemployment support will build and sustain
support for greater openness of economy.
Trade and per Frankel Cross-country Trade increases per capita income.
capita incomes and Romer study
(1999)

Irwin and Cross-country Trade increases per capita income.

Tervio study
(2002)
Growth in Dollar Cross-country The growth in incomes of the poor is no less
incomes of and Kraay study than the rate of growth of per capita income.
the poor (2002)
Trade, growth, Mitra (2016) India and The People’s Republic of China, where
poverty and (first relevant the People’s inequality went up much more, actually did
inequality part) Republic of much better at poverty reduction than India.
China While both India and the People’s Republic
of China have grown quite rapidly during the
last decades, the People’s Republic of China’s
growth performance has been quite a bit better,
and that has led to a faster reduction in poverty.
Trade and Topalova India Districts whose rural workers were more
poverty (2007) exposed to an import competition rise
reduction saw a relatively slower reduction in rural
poverty. Compared to a district that did not
experience any change in this exposure, a
district experiencing the mean change saw
a 2-percentage-point poverty rise. Urban
poverty showed no statistically significant
relationship but had the same coefficient sign
as rural poverty. Trade liberalization led to a
“significant setback” in rural poverty reduction.
Trade and Hasan, Mitra, India States with workers more exposed to foreign
poverty and Ural competition had lower rural, urban, and overall
reduction (2007), later poverty rates. States with greater reduction in
updated by employment-weighted tariffs also experienced
Cain, Hasan, greater poverty reduction. These effects
and Mitra were more pronounced in states with labor
(2012) laws making for more flexible labor markets,
greater road density, and greater financial
development.
continued on next page
 62 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table A1 continued
Economies
Topic Authors Covered Main Finding
Trade and Mukim and India Socially disadvantaged classes also
poverty Panagariya experienced declines in poverty rates during
reduction (2012) the period since trade reforms, with some
of socially evidence that trade reforms led to a decline in
disavantaged their poverty.
classes
Trade and Kis-Katos Indonesia Poverty declined with a reduction in the
poverty and Sparrow input tariffs but increased with output tariff
reduction (2015) reductions. A 1-standard-deviation larger
reduction in the employment-weighted input
tariff leads to half a standard deviation greater
reduction in the district-level poverty rate.
The results seem to be driven by people at the
extremes in education levels.

McCaig Viet Nam A 1-standard-deviation reduction in the

(2011) employment-weighted average US tariff faced
by a province leads to a 33%–40% reduction in
poverty within 2 years.
Trade and Ural India There was a pro-poor effect of trade reforms in
poverty Marchand that the poor benefited proportionally more than
(2012) others. The benefits were greater for households
in urban areas than those in rural areas and for
those in relatively less remote areas.

Han et al. People’s The poor saw a great percentage increase in

(2016) Republic of their real incomes as compared to the rich due
China to trade liberalization.
Trade, poverty Seshan Viet Nam Overall poverty as well as inequality fell due
and inequality (2014) to trade liberalization. While rural poverty
fell drastically, urban poverty went up slightly.
Trade liberalization has been responsible for
a third of the decline in overall poverty and
half of the decline in rural poverty during the
postreform period.
Trade and Mitra (2016) India and All of the slight increase in inequality in India
inequality (second the People’s over the last decades can be attributed to trade
relevant Republic reforms, while in the People’s Republic of China
part) of China trade liberalization is responsible for a fifth of
(cross-country the inequality increase.
regressions
using data from
42 countries
were run)

continued on next page

 Responses to Trade Opening: Evidence and Lessons from Asia 63

Table A1 continued
Economies
Topic Authors Covered Main Finding
Trade and Krishna and India Inequality, right after the reforms, first went
inequality Sethupathy down during 1988–1994, then went up during
(2012) 1994-2000, and went down thereafter.
Protection does not seem to be significantly
related to inequality.
Trade Kumar and India A 1-percentage-point reduction in the
and wage Mishra industry’s import tariff leads to a 0.17%
inequality (2008) increase in the industry wage premium. Bigger
tariff reductions and, therefore, bigger wage
premium increases in unskilled labor-intensive
industries resulted in a reduction in wage
inequality.

Amiti and Indonesia A 10-percentage-point input tariff reduction

Cameron lowers wage inequality by 2.6% on average for
(2012) all firms but by 4.5% for importing firms. For
firms whose imports as a share of the value of
all input are in the top 10%, this effect is 8.5%.

Chen, Yu, People’s Input trade liberalization, in the presence of

and Yu Republic of profit sharing between skilled workers and
(2013) China firms, leads to an increase in skilled–unskilled
wage inequality. This effect is not valid for
processing firms.
Trade and Ahsan and India For the relatively small and labor-intensive
labor share Mitra (2014) firms, tariff reductions raised labor share, while
for the large and relative low labor intensity
firms, labor share fell with tariff reductions.
While the elasticity of the labor share with
respect to the industry-level tariff is -0.5 in the
former set of firms, it is 0.8 for the latter.

Tariff cuts lead to increases in firm-level labor

Kamal, People’s share across the board. This effect is stronger
Lovely, and Republic of for coastal firms than for the interior firms
Mitra (2015) China and vary by ownership type: domestic private,
foreign, and state owned. Both input and
output tariff cuts have qualitatively similar
effects.
Mitra and Republic of Results are qualitatively similar but
Shin (2014) Korea quantitatively different from Ahsan and Mitra
(2014).
continued on next page
 64 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table A1 continued
Economies
Topic Authors Covered Main Finding
Trade and Ahsan, India In net-importer industries, a 10-percentage-
unionization Ghosh, and point reduction in the import tariff led to
Mitra (2017) a 0.8-percentage-point reduction in the
proportion of workers working in unionized
activities as well as in the proportion of workers
that are union members. Industry quasi-rents
per plant were declining with tariff cuts.
Trade and Krishna, Turkey The impact of trade liberalization on firm-level
labor-demand Mitra, and labor-demand elasticities was statistically
elasticities Chinoy insignificant.
(2001)

Hasan, India The absolute elasticity of labor demand at the

Mitra, and industry level in Indian manufacturing went
Ramaswamy up, due to trade liberalization, from 0.076 to
(2007) 0.186 in states with rigid labor regulation and
from 0.206 to 0.316 in states with flexible
labor regulation.

Mitra and Republic of There was weak evidence of trade liberalization

Shin (2014) Korea increasing labor-demand elasticities. However,
exports from the Republic of Korea have
increased their firm-level labor-demand
elasticities. A 10-percentage-point increase in
the share of exports in firm-level output led to
an increase in absolute labor-demand elasticity
by up to 0.04.
 PART II
Labor Market
Adjustments in Asia
 2
Industry Wages and Tariffs of
the Rest of the World1
Marcelo Olarreaga, Roberta Piermartini, and Guido Porto

2.1 Introduction
There is widespread evidence that countries use trade policy to protect
their workers. Sector-level tariffs typically correlate positively with
sector wages and employment. This chapter examines the mirror
question of how tariffs of other countries of the world affect industry
wages at home. To answer this question, we rely on an industry-level
analysis of wages in a sample of developing and developed countries
spanning from 1976 to 2004. The effect of trade policy is identified
through differential exposure of trade policy changes abroad for workers
in different industries.
As expected, we find that there is a robust negative correlation
between tariffs faced in export markets and sector-level wages. Our
estimates suggest that a 10-percentage-point higher tariff in the rest of
the world (ROW) implies a 0.8% lower wage at home. Because during
the period under study most countries benefited from improvements
in market access, sector-level wages tended to increase through this
channel. But we uncovered a large degree of heterogeneity across
countries. Asian economies such as Japan, Malaysia, Pakistan, and Hong
Kong, China are among those in our sample that experienced the highest
average wage growth during the period due to improvements in market
access. At the other end of the spectrum, Latin American and African
countries such as Argentina, Uruguay, Senegal, and Nigeria experienced
the largest declines in average wages due to higher ROW tariffs faced by

1
The authors are grateful to Marc Bacchetta for useful suggestions. Olarreaga and
Porto also gratefully acknowledge support from the r4d program on employment
funded by the Swiss National Science Foundation and the Swiss Agency for
Development and Cooperation.

67
68 Trade Adjustment in Asia: Past Experiences and Lessons Learned

their exporters. We also find that there is strong heterogeneity within

regions. In Asia, countries such as Nepal and Sri Lanka saw their wages
decline due to a deterioration of their market access. In Latin America
and Africa, some countries experience increases in wages due to
declines in the tariffs they face abroad. These results are important for at
least three reasons. First, they provide a rationale for trade negotiations
based on wages that is probably much easier to grasp than the terms-of-
trade rationale offered by the standard trade model (even though they
are driven by the same mechanism). Second, it enables us to disentangle
the significant heterogeneity across countries in terms of wage growth
benefits associated with trade reforms over the period. Workers in
some countries have been left behind by the general move toward trade
liberalization, and there is a need to address these imbalances if we
want those countries and workers back at the negotiating table. Last,
but not least, if individuals care about worker welfare not only at home,
but also abroad, as the corporate social responsibility literature seems
to suggest, then the case for using tariffs to protect workers is seriously
undermined by our results.
The literature on trade policy and wages is vast. Earlier studies
typically find significant negative effects of removing trade protection
on wages, especially for developing countries. For example, in a study
on Colombia, Attanasio, Goldberg, and Pavcnik (2004) document a
reduction of the skill premium in sectors that face the strongest tariff
reduction compared to industries that faced a lower reduction. In her
study of the effects of India’s liberalization in 1991, Topalova (2010)
finds that regions with a higher concentration of industries that lost
protection as a result of import tariff reductions experienced a slower
decline in poverty. Looking at Brazil’s early 1990s trade liberalization,
Kovak (2013) estimates a relative fall in wages in regions facing larger
liberalization. In general, the literature finds larger employment
effects than effects on wages of import tariff reductions in developed
countries (Grossman 1986; Trefler 2004; Pierce and Schott 2016, who
look at reduction of trade policy uncertainty rather than tariff cuts).
McLaren and Hakobyan (2016) find that wages grew significantly less
for workers in the United States located in areas more affected by tariff
cuts following the implementation of the North American Free Trade
Agreement.
We focus on changes in tariffs faced in the export destination
markets. To our knowledge, our study is the first to look systematically
at how the correlation between tariffs and wages depend on the nature
of the tariff change and for a large sample of developed and developing
countries. In his study on the United States–Viet Nam free trade
agreement, McCaig (2011) finds that provinces in Viet Nam that were
 Industry Wages and Tariffs of the Rest of the World 69

more exposed to the United States’ tariff cuts experience greater declines
in poverty rates. Similarly, Porto (2010) predicts that the elimination of
trade and barriers on exports of agro-manufactures to industrialized
countries would cause poverty to decline in Argentina. These studies,
however, do not look at wages.

2.2 Data
Our main source of data is the Trade, Production and Protection database
put together by Nicita and Olarreaga (2007). The cross-country data
include information on export values and export quantities, production,
value added, employment, wages, and number of establishments for
28 manufacturing industries corresponding to the three-digit level of
the International Standard Industrial Classification (ISIC), Revision 2.
The database is available at the World Bank trade website (www.
worldbank.org/trade). We combine the Nicita and Olarreaga data
with supplementary data on country characteristics from the World
Development Indicators. These characteristics include per capita gross
domestic product (GDP), GDP, population, and bilateral exchange rates.
The basic premise of our analysis is the correlation between the
average tariff across export destinations and the level of wages. Using
the Nicita and Olarreaga (2007) data, we calculate the average industry
wage for each source country as the ratio of total industry wage bill to
total employment. Using wic as the average wage in industry i in country
c, we construct a measure of exposure to foreign tariffs by computing the
average tariff across export markets. We define the average tariff of an
industry’s exports as:

𝜏𝜏𝑖𝑖𝑖𝑖𝑖𝑖 = ∑ 𝑠𝑠𝑖𝑖𝑖𝑖𝑑𝑑𝑡𝑡 ∗ 𝜏𝜏𝑖𝑖𝑑𝑑𝑡𝑡  (1)

𝑑𝑑

where τidt is the tariff faced𝑙𝑙𝑙𝑙𝑙𝑙

by industry i in destination
𝑤𝑤𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛽𝛽𝛽𝛽 𝑖𝑖 country d at time t
𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑥𝑥𝑖𝑖𝑖𝑖𝑖𝑖 𝛾𝛾 + 𝜑𝜑𝑡𝑡 + 𝜑𝜑𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑢𝑢𝑖𝑖𝑖𝑖𝑖𝑖 ǡ
and sicdt the share of destination d in exports of industry i of source country
c. Average tariffs abroad vary by industry, source country, and year
through differences in tariffs themselves and the export-share weights.
Figure 2.1 showcases our basic hypothesis for four Asian
countries in our sample: the Republic of Korea, India, Indonesia, and
Singapore. In these cases, there is clear negative correlation between
the log of the average wage paid in industry i in country c and the
average tariff across export destinations. Higher tariffs in the rest
of the world de-protect Asian workers and make them worse-off in
terms of wages. Figure 2.2 displays this negative correlation for four
developed countries: the United States, Denmark, Finland, and Norway.
70 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 2.1 Correlations between Wages and Tariffs in Asia

(a) Republic of Korea (b) India

3.5 1.5

1.0
3.0

0.5

2.5 0.0

–0.5
2.0
0.0 0.1 0.2 0.3 0.00 0.05 0.10 0.15

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

(c) Indonesia (d) Singapore

1.5 4.0

1.0
3.5
0.5

0.0 3.0

–0.5
2.5
0.0 0.2 0.4 0.6 0.0 0.1 0.2 0.3

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

Source: Authors’ estimation.

Figure 2.2 Correlations between Wages

and Tariffs in Developed Countries
(a) United States (b) Denmark

4.0 3.8

3.6
3.5
3.4

3.0 3.2

3.0
2.5
0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

(c) Finland (d) Norway

4.0 continued on next page

3.4

3.8
3.2

3.6
3.0

2.8 3.4

2.6 3.2

2.4 3.0
0.00 0.05 0.10 0.15 0.20 0.0 0.5 1.0 1.5

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t
 4.0 3.8

3.6
3.5
3.4

3.0 Industry Wages

3.2 and Tariffs of the Rest of the World 71
3.0
2.5

Figure 2.2 continued

0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

(c) Finland (d) Norway

4.0
3.4

3.8
3.2

3.6
3.0

2.8 3.4

2.6 3.2

2.4 3.0
0.00 0.05 0.10 0.15 0.20 0.0 0.5 1.0 1.5

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

Source: Authors’ estimation.

Figure 2.3 shows examples for Latin America (Colombia, Chile, Peru,
and Uruguay) and Figure 2.4 for Africa (Ethiopia, Cameroon, Malawi,
and Nigeria). In these countries, the data reveal that industries that
faced higher tariffs in the rest of the world paid lower wages, on average.

Figure 2.3 Correlations between Wages

and Tariffs in Latin America
(a) Colombia (b) Chile

2.5
3.5

2.0
3.0

1.5
2.5

1.0 2.0

0.5 1.5
0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

(c) Peru (d) Uruguay

3.0 3.5

2.5 3.0

2.0 2.5

1.5 2.0

1.0 1.5

0.5 1.0
0.00 0.05 0.10 0.15 0.20 0.0 0.1 0.2 0.3 0.4

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

Source: Authors’ estimation.

72 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 2.4 Correlations between Wages and Tariffs in Africa

(a) Ethiopia (b) Cameroon

3
0.0

–0.2
2
–0.4

–0.6 1

–0.8

–1.0 0

0.0 0.1 0.2 0.3 0.4 0.0 0.1 0.2 0.3 0.4

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

(c) Malawi (d) Nigeria

2 1.5

1 1.0

0 0.5

-1 0.0

-2 -0.5
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.1 0.2 0.3 0.4 0.5

Fitted values Log Wage 2000 US dollars, year t Fitted values Log Wage 2000 US dollars, year t

Source: Authors’ estimation.

2.3 Econometric Model and Results

To study these
𝜏𝜏𝑖𝑖𝑖𝑖𝑖𝑖questions formally,
= ∑ 𝑠𝑠𝑖𝑖𝑖𝑖𝑑𝑑𝑡𝑡 ∗ 𝜏𝜏𝑖𝑖𝑑𝑑𝑡𝑡  we set up the following regression
specification for wages: 𝑑𝑑

𝑖𝑖
𝑙𝑙𝑙𝑙𝑙𝑙 𝑤𝑤𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛽𝛽𝛽𝛽𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑥𝑥𝑖𝑖𝑖𝑖𝑖𝑖 𝛾𝛾 + 𝜑𝜑𝑡𝑡 + 𝜑𝜑𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑢𝑢𝑖𝑖𝑖𝑖𝑖𝑖 ǡ (2)

where log wict is the log of the average wage paid in industry i in country
of origin c at time t, τict is the export-share weighted average foreign
tariff across destination markets as defined in equation (1), xict is a
vector of controls that varies across several specifications, φt are year
fixed-effects, φic are country of origin-industry fixed effects, and uict is
the error term. Given the set of fixed effects used in equation (2), the
coefficient of interest β, which captures the impact of tariffs abroad on
sector wages is identified using the average variation across time and
within country and industries. Standard errors are clustered by industry-
 Industry Wages and Tariffs of the Rest of the World 73

country (source). Because we are looking at the impact of changes in

trade protection abroad on home wages, we do not expect endogeneity
to be an important problem and estimate equation (2) using ordinary
least squares.
Table 2.1 presents our results. Conditioning only on the fixed
effects φt and φic, the coefficient is negative and statistically significant
(column 1). An industry facing an average tariff across destinations
which is 10 percentage points higher pays on average almost 1% lower
wages. Since this model includes year effects, any aggregate shock
is accounted for, while the origin-industry effects control for time-
invariant characteristics of an industry in a given country (such as
certain technological characteristics or policies that remain constant).
This negative correlation between wages and foreign tariffs is robust
to the inclusion of various important controls. Following Brambilla
and Porto (2016), we add the log of the per capita GDP of the origin

Table 2.1 Average Wages and Average Tariffs in ROW:

OLS-FE Estimation

(1) (2) (3) (4) (5) (6)

Average –0.0984** –0.0976** –0.0972* -0.0835* –0.0806** –0.0808**
ROW Tariff [0.0496] [0.0483] [0.0497] [0.0495] [0.0400] [0.0400]
Log Origin 0.459*** 0.338*** 0.410*** 0.412***
p/c GDP [0.101] [0.103] [0.0884] [0.0883]
Log Industry –0.000104 –1.4e-04** –0.000121* –0.000122*
Exports [7.19e-05] [6.96e-05] [6.78e-05] [6.80e-05]
Log Industry 0.145*** 0.0300 0.0300
Output [0.0333] [0.0229] [0.0229]
Productivity 0.189*** 0.190***
[0.0250] [0.0250]
Average 0.00852
p/c GDP [0.0119]
Observations 7,083 7,083 7,083 6,795 6,166 6,166
R-squared 0.016 0.024 0.016 0.072 0.154 0.154
Origin- 1,539 1,539 1,539 1,539 1,539 1,539
Industry
Groups
FE = fixed effects, GDP = gross domestic product, OLS = ordinary least squares, p/c = per capita, ROW = rest
of the world.
Significance at 1%, 5%, and 10% levels indicated by ***, **, and *.
Note: Proportional changes in average wages caused by the observed change in ROW tariffs, 2000–1995.
Source: Authors’ estimation.
74 Trade Adjustment in Asia: Past Experiences and Lessons Learned

country in column 2. Higher income implies a higher domestic demand

and thus higher wages. The level of per capita GDP also accounts
for differential country effects across time, such as periods of booms or
crises. In column 3, we exclude per capita GDP but include the log of
industry exports. In column 4, we further control for per capita GDP in
the country of origin, the log of industry exports and the log of industry
output (column 4). The results are robust: in column 4, for instance,
β̂ = –0.0835, which is smaller but similar to the β̂ estimated when
conditioning only on the fixed effects (column 1). Finally, we control for
the average productivity of an industry (output per worker) in column 5
and for the average GDP of the export destination (as in Brambilla and
Porto 2016) in column 6. These controls do not affect the results either,
with an estimate β̂ = –0.0808 so that a 10-percentage-point higher tariff
in the rest of the world implies 0.8% lower industry wages.
If the impact of foreign tariffs on domestic wages is on average
relatively small, the impact of foreign trade reforms on home wages
will depend on the change in tariffs in foreign countries faced by each
country’s exporters. As discussed in the next section, there are some
important differences.

2.4 Wages and Rest of the World Tariff Changes

In this section, we explore the winners and losers from the observed
changes in ROW tariffs. To this end, we compute for each country in
our sample the observed change in the average tariff faced in the rest
of the world from 1995 to 2000. Then, we use the estimated coefficient
in column 5 of Table 2.1 and make predictions for the change in average
log wages.
The average impacts on wages in Asia are shown in Table 2.2. There
are 14 Asian economies in our sample, with eight losers (those with
increases in ROW tariffs and declines in average wages) and six winners
(those with ROW tariff declines and wage increases). The largest loss is
estimated for Nepal, at 0.61% decline in wages. In Sri Lanka, Bangladesh,
and Thailand, the losses are −0.19%, −0.17%, and −0.10%, respectively.
In Jordan, Singapore, the Philippines, and India, the losses become
smaller and are negligible. The largest increase in wages is observed
in Hong Kong, China, at 1.23%. In Malaysia with an increase of 0.52%,
Pakistan with 0.36%, Japan with 0.29%, and Indonesia with 0.18%, the
impacts are sizable, but the gains in the Republic of Korea are small.
These differences in the impact of foreign tariffs on home wages in
Asia are simply explained by differences in changes in ROW tariffs faced
by each of these countries. The largest reductions in ROW tariffs are
experienced by exporters in Hong Kong, China with declines in average
 Industry Wages and Tariffs of the Rest of the World 75

Table 2.2 Average Wages and Average Tariffs in ROW:

Asian Economies

ROW Tariff Changes Change in Log Wages

Nepal 0.0759 –0.0061
Sri Lanka 0.0240 –0.0019
Bangladesh 0.0214 –0.0017
Thailand 0.0125 –0.0010
Jordan 0.0063 –0.0005
Singapore 0.0012 –0.0001
Philippines 0.0007 –0.0001
India 0.0006 –0.0000
Republic of Korea –0.0026 0.0002
Indonesia –0.0226 0.0018
Japan –0.0358 0.0029
Pakistan –0.0451 0.0036
Malaysia –0.0639 0.0052
Hong Kong, China –0.1519 0.0123
ROW = rest of the world.
Notes: Dependent variable is average wage. Controls in all columns: origin-industry effects, year effects.
Standard errors clustered at origin-industry level.
Sources: Data from A. Nicita and M. Olarreaga. 2007. Trade, Production and Protection Database,
1976–2004. World Bank Economic Review 21 (1). pp. 165–171; and World Bank. World Development
Indicators. http://datatopics.worldbank.org/world-development-indicators/ (accessed June 2017).

tariffs of 15 percentage points, followed by Malaysia (6 percentage

points) and Pakistan (5 percentage points). Sri Lanka, Bangladesh, and
Thailand have the largest increases in ROW tariffs with increases in
average tariffs of 2, 2, and 1 percentage point(s), respectively. Note that
these changes in foreign trade protection only include tariffs. If nontariff
measures were to be included, it is likely that we would observe larger
changes in foreign trade protection.
The estimated impacts on wages for the remaining 16 losing countries
are reported in Table 2.3. ROW tariffs increase in these countries, and
this creates losses in real wages ranging from −0.43% in Argentina to
very negligible losses in Ireland. This again is simply explained by the
degree of ROW tariff increases experienced in each of these countries
varying from an increase of 5 percentage points in Argentina to
0.05 percentage points in Ireland. In the remaining 52 countries in our
sample (Table 2.4), real wages increase following reduction in foreign
76 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 2.3 Average Wages and Average Tariffs in ROW:

Losers from Higher Tariffs

ROW Tariff Changes Change in Log Wages

Argentina 0.0532 –0.0043
Uruguay 0.0508 –0.0041
Senegal 0.0496 –0.0040
Nigeria 0.0482 –0.0039
Oman 0.0363 –0.0029
Iran 0.0354 –0.0029
Ethiopia 0.0351 –0.0028
Chile 0.0144 –0.0012
Malawi 0.0100 –0.0008
Egypt 0.0092 –0.0007
Côte d’Ivoire 0.0071 –0.0006
Guatemala 0.0068 –0.0005
Brazil 0.0022 –0.0002
Kyrgyz Republic 0.0017 –0.0001
Venezuela 0.0013 –0.0001
Ireland 0.0005 –0.0000
ROW = rest of the world.
Note: Proportional changes in average wages caused by the observed change in ROW tariffs, 2000-1995.
Source: Author’s calculation using the data in A. Nicita and M. Olarreaga. 2007. Trade, Production and
Protection Database, 1976–2004. World Bank Economic Review 21 (1). pp. 165–171.

Table 2.4 Average Wages and Average Tariffs in ROW:

Winners from Lower Tariffs

ROW Tariff Changes Change in Log Wages

El Salvador –0.0004 0.0000
United Kingdom –0.0035 0.0003
Latvia –0.0037 0.0003
Peru –0.0040 0.0003
Ecuador –0.0047 0.0004
Gabon –0.0050 0.0004
Greece –0.0052 0.0004
Finland –0.0054 0.0004
continued on next page
 Industry Wages and Tariffs of the Rest of the World 77

Table 2.4 continued

ROW Tariff Changes Change in Log Wages
Colombia –0.0055 0.0004
United States –0.0057 0.0005
Kenya –0.0062 0.0005
Netherlands –0.0063 0.0005
Sweden –0.0067 0.0005
Bolivia –0.0068 0.0005
Honduras –0.0072 0.0006
Portugal –0.0072 0.0006
United Republic of Tanzania –0.0077 0.0006
Cameroon –0.0079 0.0006
Canada –0.0082 0.0007
Denmark –0.0092 0.0007
Italy –0.0106 0.0009
Mexico –0.0135 0.0011
Slovenia –0.0149 0.0012
Spain –0.0154 0.0012
Australia –0.0176 0.0014
Germany –0.0191 0.0015
Austria –0.0194 0.0016
France –0.0200 0.0016
Slovakia –0.0209 0.0017
South Africa –0.0224 0.0018
Lithuania –0.0227 0.0018
Botswana –0.0239 0.0019
New Zealand –0.0240 0.0019
Costa Rica –0.0277 0.0022
Ukraine –0.0293 0.0024
Algeria –0.0306 0.0025
Russian Federation –0.0313 0.0025
Bulgaria –0.0352 0.0028
Azerbaijan –0.0368 0.0030
Armenia –0.0373 0.0030
Hungary –0.0423 0.0034
Trinidad and Tobago –0.0425 0.0034
continued on next page
78 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 2.4 continued

ROW Tariff Changes Change in Log Wages
Czech Republic –0.0442 0.0036
Panama –0.0454 0.0037
Poland –0.0555 0.0045
Mauritius –0.0585 0.0047
Israel –0.0593 0.0048
Norway –0.0662 0.0054
Morocco –0.0722 0.0058
Moldova –0.0724 0.0058
Tunisia –0.1084 0.0088
Turkey –0.1110 0.0090
ROW = rest of the world.
Notes: Dependent variable is average wage. Controls in all columns: origin-industry effects, year effects.
Standard errors clustered at origin-industry level.
Sources: Data from A. Nicita and M. Olarreaga. 2007. Trade, Production and Protection Database,
1976–2004. World Bank Economic Review 21 (1). pp. 165–171; and World Bank. World Development
Indicators. http://datatopics.worldbank.org/world-development-indicators/ (accessed June 2017).

tariffs. The gains range from very small increases in real wages due to
very low cuts in ROW tariffs in El Salvador to more sizable real wage
increases of 0.9% in Turkey.

2.5 Conclusions
We show that industries that face higher tariffs in their export markets
pay lower wages. We show that this result is robust and holds for both
developed and developing countries. However, because countries export
different products to different countries, there are important differences
in terms of changes in market access, which implies that workers in
some countries have benefited from better market access, whereas
others experienced losses in wages due to a deterioration of their
market access abroad. Even though there are important intraregional
differences, workers in parts of Asia seem to be among those that have
benefited the most from wage increases associated with trade reforms
during the period under study, whereas workers in Latin America and
Africa are among those that have actually experienced wage losses due
to trade reforms during the same period.
 Industry Wages and Tariffs of the Rest of the World 79

Differences in sector specialization and trading partners explain our

results, but differences in the impact that changes in market access have
on wages at home are also likely explained by labor market frictions,
different types of worker endowments, product market power, and
taxation and redistribution that we leave for further research. There
is definitely a need for within-country studies to further disentangle
differences across countries in terms of benefits accruing to workers
from trade reforms abroad.
Our findings further suggest that there is a positive international
spillover from unilateral liberalization that countries do not internalize
when setting their trade policies to protect workers’ wages. As a
result, the level of protection that countries individually set to protect
workers employed in the industry with low wages is inefficiently high.
Existing evidence supports this prediction. In their study of tariffs
faced in international markets by Indian workers, Mendoza, Nayyar,
and Piermartini (2018) show that trade barriers tend to be increasingly
higher for workers earning lower wages. More research is needed to
assess the general equilibrium effects of a global reduction of tariffs
for the goods produced by low-wage workers. However, the analysis
of existing tariff profiles and our results appear to suggest the need to
further international cooperation.
80 Trade Adjustment in Asia: Past Experiences and Lessons Learned

References
Attanasio, O., P. K. Goldberg, and N. Pavcnik. 2004. Trade Reforms and
Wage Inequality in Colombia. Journal of Development Economics.
74 (2). pp. 331–66.
Brambilla, I., and G. Porto. 2016. High-Income Export Destinations,
Quality and Wages. Journal of International Economics. 98 (C).
pp. 21–35.
Grossman, G. M. 1986. Imports as a Cause of Injury: The Case of the US
Steel Industry. Journal of International Economics. 20 (3). pp.201–23.
Kovak, B. 2013. Regional Effects of Trade Reform: What Is the Correct
Measure Of Liberalization? American Economic Review. 103 (5).
pp. 1960–76.
McCaig, B. 2011. Exporting Out of Poverty: Provincial Poverty in Vietnam
and US Market Access. Journal of International Economics. 85 (1).
pp. 101–13.
McLaren, J., and S. Hakobyan. 2016. Looking for Local Labor Market
Effects of the NAFTA. The Review of Economics and Statistics.
98 (4). pp. 728–41.
Mendoza, A., G. Nayyar, and R. Piermartini. 2018. Are the “Poor” Getting
Globalised? In World Trade Organization and World Bank Trade
and Poverty Reduction: New Evidence of Impacts on Developing
Countries. Mimeo, WTO.
Nicita, A., and M. Olarreaga. 2007. Trade, Production and Protection
Database, 1976–2004. World Bank Economic Review. 21 (1).
pp. 165–71.
Pierce, J., and P. Schott. 2016. The Surprisingly Swift Decline of US
Manufacturing Employment. American Economic Review. 106 (7).
pp. 1632–62.
Porto, G. 2010. International Market Access and Poverty in Argentina.
Review of International Economics. 18 (2). pp. 396–407.
Topalova, P. 2010. Factor Immobility and Regional Impacts of Trade
Liberalization: Evidence on Poverty from India. American Economic
Journal: Applied Economics. 2 (4). pp. 1–41.
Trefler, D. 2004. The Long and Short of the Canada–US Free Trade
Agreement. American Economic Review. 94 (4). pp. 870–89.
World Bank. World Development Indicators. http://datatopics.worldbank.
org/world-development-indicators/ (accessed June 2017).
 3
Trade Liberalization
and the Hukou System of the
People’s Republic of China:
How Migration Frictions
Can Amplify the Unequal
Gains from Trade
Yuan Zi

3.1 Introduction
The emergence of the People’s Republic of China as a great economic
power has stimulated an epochal shift in patterns of world trade, in
contradiction to the conventional wisdom regarding the impact of trade
on labor markets in developed countries (Autor, Dorn, and Hanson
2016). The global effects of the People’s Republic of China’s trade and
economic growth has been widely documented (Autor, Dorn, and
Hanson 2013; Bugamelli, Fabiani, and Sette 2015; Balsvik, Jensen, and
Salvanes 2015; Giovanni, Levchenko, and Zhang 2014; Hsieh and Ossa
2011), reshaping our understanding of the consequences of trade for
wages, unemployment, and other labor market outcomes.
On the other hand, equally significant transformations can be
identified within the People’s Republic of China itself, including the
remarkable degree of internal migration occurring within the country.
Hundreds of millions of the workers have moved from inland areas to
coastal cities, contributing to manufacturing growth and export surges.
However, the extent to which they have benefited from the country’s
trade liberalization remains less clear: migrant workers are usually
treated as second-class citizens and are prevented from accessing

81
82 Trade Adjustment in Asia: Past Experiences and Lessons Learned

various social benefits provided at the local level due to the country’s
unique household registration system (hukou). Does the hukou system
contribute to the unequal distribution of gains from trade? Does this
labor market distortion prevent the People’s Republic of China from
fully reaping the gains from trade reforms? These are relevant policy
questions that require empirical underpinning.
From a theoretical point of view, trade liberalization is often considered
an important driver of economic development, as it can raise a country’s
income through increasing specialization in sectors with a comparative
advantage, providing access to cheap foreign inputs, and facilitating
the adoption of new technologies. However, prominent trade theories
typically focus on long-run equilibria, assuming that the reallocation of
resources across economic activities is frictionless. However, in reality,
factor adjustments tend to be slow, costly, and heterogeneous across
firms, sectors, and space. As long as some production factors are spatially
immobile and trade is not frictionless, the extent to which a country can
gain from trade is ultimately contingent on labor mobility. This point has
long been recognized but has become increasingly emphasized by trade
and labor economists, as we find more and more evidence regarding the
adverse impact of trade shocks on labor market outcomes.
Nevertheless, demonstrating this empirically is not easy. First, it
is very difficult to find a clear measure of migration friction, as most
factors or policy shocks affect both goods and people at the same time.
Second, according to most studies, internal migration reacts negligibly
to trade shocks. Some indirect evidence exists that labor immobility can
explain a large proportion of the negative impact of trade,1 but owing
to the aforementioned difficulties, a direct test remains missing. I
exploit the People’s Republic of China’s liberalization episode following
its accession to the World Trade Organization (WTO) to test how its
internal migration reacts to trade shocks. Using a hukou friction measure
constructed in Zi (2018), I shed light on the interaction between trade
and migration frictions.
Drawing on a rich dataset that I assembled on the People’s Republic
of China’s regional economy, I find that prefectures that experience
more positive trade shocks have seen a relative increase in employment,
and the effect is strongest in provinces with a lower amount of hukou
friction. A prefecture at the 75th percentile of effective tariff exposure
experiences an employment increase 3.4 percentage points greater (or
smaller decrease) than a prefecture at the 25th percentile. In a prefecture

1
See, for instance, Autor, Dorn, and Hanson (2013); Topalova (2010); Dix-Carneiro and
Kovak (2017) for the cases of the United States (US), India, and Brazil, respectively.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 83

with the lowest amount of hukou friction, the effect is three times larger
than the average effect. On average, over 30% of the regional variation
in employment changes can be attributed to trade liberalization.
Moreover, the total population and the working-age population of
prefectures react to trade shocks and their interaction with the hukou
measure in a quantitatively similar way to employment, suggesting
that the observed regional employment changes are primarily driven
by interregional labor adjustments. Direct focus on migration flows
yields similar results. Most importantly, I only find that trade shocks
result in increases in the population holding local hukou in prefectures
where hukou frictions are low. This result suggests that in spite of labor
mobility between prefectures, migrant workers can only obtain a local
hukou in prefectures with less stringent hukou systems. This supports
the validity of my hukou measure and confirms the existence of hukou
frictions.
Although the focus of this chapter is on the People’s Republic of
China, the message and policy implications are not limited to this country.
According to World Population Policies 2013 (United Nations 2013),
60% of governments in the world desired a major change in their spatial
labor distribution and 80% of these countries had policies in place to
influence internal migration. In general, gaining a greater understanding
of when and where trade is costly and how various domestic frictions
shape the impact of trade on workers, individuals, and/or households of
different groups is central to the research agenda of trade economists.
Implementing effective policies to eliminate or mitigate these frictions
and effectively targeting the most adversely affected individuals should
represent a salient issue to policy makers and applied economists.
The content of this chapter is based on the analysis of Zi (2018). I
begin by discussing the People’s Republic of China’s trade reforms since
the late 1970s and their acceleration following the country’s accession to
the WTO. I then provide a detailed description of the country’s various
hukou reforms. In the following section, I identify the trade shocks
and migration frictions embedded in the hukou system at the local
level, present evidence regarding the ways in which trade shocks have
stimulated a substantial degree of spatial labor reallocation in the People’s
Republic of China, as well as how regional hukou frictions influence this
effect. I subsequently offer a simple conceptual framework that guides
our inquiry on measuring and interpreting the welfare impacts of the
observed labor reallocation. Finally, I present welfare calculations from
Zi (2018), demonstrating how the hukou system has prevented optimal
spatial adjustments of labor to trade shocks in the country, and how this
in turn amplifies the negative distributional consequences of trade.
84 Trade Adjustment in Asia: Past Experiences and Lessons Learned

3.2 T
 he Trade and Hukou Reforms
of the People’s Republic of China
3.2.1 Trade Liberalization in the People’s Republic
of China: Before and after the WTO

Prior to its economic reform in the early 1980s, the average tariff
level in the People’s Republic of China was 56%.2 This tariff schedule
was implemented in 1950, with almost no change since, partly due to
the relative unimportance of trade policy under the centrally planned
economy. Under the planned economy, import and export quantities
represented government decisions rather than reflections of market
supply and demand (Ianchovichina and Martin 2001). During this period,
the People’s Republic of China’s trade was run by 10 to 16 foreign-trade
corporations that were de facto monopolies in their specified product
ranges (Lardy 1991).
In 1982, the People’s Republic of China commenced its first tariff
modification, and gradually reduced its average tariff by 13% in the
following 5 years. From 1992 onward, in order to pave the way for the
country’s accession to the WTO, the government engaged in a series of
voluntary tariff cuts on over 5,000 products, driving the simple average
tariff down from 43% in 1992 to 24% in 1996 (Li 2013).
However, these episodes of tariff reductions were accompanied
by pervasive and complex import and export controls. Import quotas,
licenses, designated trading practices, and other nontariff barriers
were widely used (Blancher and Rumbaugh 2004). There was also a
substantial level of tariff redundancy resulting from various preferential
arrangements. To name a few, imports for processing purposes, for
military uses, by special economic zones (SEZs), and in certain areas
near the People’s Republic of China’s border were subject to waivers
or reductions in import duties. According to Ianchovichina and Martin
(2001), only 40% of imports were subject to official tariffs. In addition,
the renminbi depreciated by more than 60% in the 1980s, and a further
44% in 1994 to help firms export (Li 2013). As a result, changes in tariff
duties do not fully reflect the changes in actual protection faced by firms
or the accessibility of imported inputs during these periods.
In 1996, the government implemented substantial reforms that
removed most restrictive nontariff barriers to fulfill the preconditions
of WTO accession. Trade licenses, special import arrangements,

2
This is the 1982 unweighted average tariff documented by Blancher and Rumbaugh
(2004).
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 85

and discriminatory policies against foreign goods were reduced or

eliminated to render tariffs the primary instruments of protection.
The share of all imports subject to licensing requirements fell from a
peak of 46% in the late 1980s to less than 4% of all commodities by
the time the People’s Republic of China acceded to the WTO. The state
abolished import substitution lists and authorized tens of thousands
of companies to engage in foreign trade transactions, undermining the
monopoly powers of state trading companies for all but a handful of
commodities. The transformation was similarly far-reaching on the
export side (Lardy 2005). The duty-free policy on imports for personal
use by SEZs was gradually abolished in the 1990s, and preferential
duty in some border provinces were abolished in 2001. Moreover, the
People’s Republic of China also abolished, modified, or added over
1,000 national regulations and policies. At the regional level, more than
3,000 administrative regulations and about 188,000 policy measures
implemented by provincial and municipal governments were ceased.
From 2001, phased tariff reductions were implemented following
the People’s Republic of China’s WTO accession, with the goal of
reducing both the average tariff levels and the dispersion of tariffs across
industries. In 2000, the People’s Republic of China’s simple average
applied tariff was 17%, with a standard deviation across the Harmonized
System six-digit level (HS6) products of 12%. By the end of 2005, the
average tariff level was reduced to 6%, and the standard deviation had
almost halved. The average tariff level stabilized after 2005.3

3.2.2 The Hukou System

A hukou is a household registration record that identifies a person
as a resident of a particular area in the People’s Republic of China. It
officially identifies a person as a resident of an area in the country and
determines where he or she is officially allowed to live. The hukou
system was introduced in the early 1950s to harmonize the old household
registration systems across regions. However, under the centrally
planned economy, economic resources were mostly devoted to urban
areas, as the government hoped to extract the country’s agricultural
economic surplus to fuel urban industrialization. This uneven allocation
of resources led to a massive influx of migrants into the main cities,
which in turn resulted in substantial unemployment in urban areas while
threatening agricultural production in rural areas (Kinnan, Wang, and

3
All numbers are calculated using the simple average of most-favored nation applied
tariffs at the Harmonized System six-digit level (HS6) from the United Nations
Conference on Trade and Development Trade Analysis Information System database.
86 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Wang 2015). As a result, the hukou system was soon repurposed to restrict
both interregional and rural-to-urban migration. In 1958, the Standing
Committee of the National People’s Congress adopted the Household
Registration Regulations. According to the regulations, citizens could
only apply to move after the registration authority had granted them the
local hukou. From then on, the People’s Republic of China entered an era
with strict internal migration controls, with the hukou being at the center
of the migration control system.
By the end of the 1950s, free migration became extremely rare.
Migrant workers would require six passes to work in provinces other
than their own. Moreover, rural-to-urban migrants would have to adhere
to the above restrictions and also first acquire an urban hukou, the annual
quota of which was 0.15% to 0.2% of the nonagricultural population of
each city (Cheng 2007). Under the central planning system, coupons for
consumption goods, employment, housing, education, health care, and
other social benefits were entirely allocated based on the local hukou,
and urban dwellers without a local hukou would be fined, arrested, and
deported. Thus, it was impossible for people to work and live outside
their authorized domain (Cheng and Selden 1994).
In the early 1980s, the People’s Republic of China latched onto a
labor-intensive, export-oriented development strategy that created
increasingly large labor demand in cities. Accordingly, migration policy
began to relax over time. In 1984, the State Council allowed rural
populations to reside in villages with self-sustained staples. In the
following year, the Ministry of Public Security allowed people to migrate
freely conditional on applying for a temporary residential permit upon
arrival. In 1993, the People's Republic of China officially ended the
food rationing system, and internal migration was no longer limited by
hukou-based consumption coupons. Gradually, the distinction between
the rural and urban hukou also became less important (Bosker et al.
2012). The rural-to-urban migration quotas were officially abolished in
1997; for many cities and towns, the rural/urban distinction of the hukou
type was also eliminated (Chan 2009).
Nevertheless, the hukou system continues to serve as the primary
instrument for regulating interregional migration. Certain cities
have limited capacity for large quantities of labor due to historical or
environmental issues, so they continue to seek to keep migration under
tight control. Some regions that are close to national borders or that
contain large proportions of ethnic minority groups are also sensitive to
migration inflows, largely due to stability concerns. In addition, without
fiscal transfers from the central government, prefectures generally
have very little incentive to provide public services to migrant workers.
Consequently, discrimination against migrant workers on the basis of the
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 87

hukou status remains widespread. Individuals who do not have a local

hukou in the place where they live are not able to access certain jobs,
schooling, subsidized housing, health care, and other benefits enjoyed by
those who do. As a result, the ease of obtaining a local hukou still influences
one’s migration decisions to a considerable extent.
Importantly, as part of a contemporaneous reform devolving
fiscal and administrative powers to lower-level governments, local
governments have largely acquired the authority to determine the
number of hukou to issue in their jurisdictions. Since 1992, some
provinces have introduced temporary resident permits for anyone with
a legitimate job or business in one of their major cities, and some grant
a hukou to high-skilled professionals or businesspeople who make
large investments in their region (Kinnan, Wang, and Wang 2015). The
most significant change is the introduction of two types of residential
registration, the so-called temporary residential permit and the blue-
stamp hukou. Unlike the regular hukou, these are not administered
by the central government; instead, their design and implementation
are determined by local governments. While the temporary resident
permit can be issued to anyone who has a legitimate job or business in
the city, citizens who want a blue-stamp hukou are usually required to
pay a one-time entry fee called the urban infrastructural construction
fee, which varies between a few thousand yuan in small cities and
CNY50,000 in more “attractive” cities. However, the stringency of
these policies and general hukou-issuing rules differ significantly
across regions. For instance, it is notoriously difficult to obtain a hukou
in Beijing or Shanghai, while Dongguan, a coastal city in Guangdong
province, offers relatively generous granting rules to attract low-
skilled migrants for its booming manufacturing sectors. It is this
heterogeneity in hukou-granting practices that provides variation in
the hukou friction measure.
The aforementioned practices resulted in formal hukou reform,
launched by the central government in 1997. The major aspects of the
reform included officially abolishing the rural-to-urban migration
quotas and approving the selective migration policies in cities. Following
an experimental period, a national implementation of the reform began
in 2001. However, this reform, which is largely an affirmation of local
policies that were already in place, has been largely put on hold since
mid-2002 due to stability concerns (Wang 2004). According to Chan and
Buckingham (2008), it only had a small impact in facilitating internal
migration. In spite of the general increase in the number of migrants
in the country during the last quarter century, the annual number of
hukou migrants recorded by the Ministry of Public Security remained
stable between 1992 to 2008 (Chan 2013). In 2011, “a hukou reform” was
88 Trade Adjustment in Asia: Past Experiences and Lessons Learned

mentioned again in the country’s Five-Year Plan, but the exact plan only
began to take shape in 2014.

3.2.3 Exogeneity of Trade Liberalization

The validity of the empirical analysis relies on the variation in tariff

changes across industries. To draw any causal implications of input
trade liberalization, tariff changes must be unrelated to counterfactual
industry employment growth. As discussed in Kovak (2013), such a
correlation may arise if trade policy makers impose smaller tariff cuts to
protect weaker industries, or if larger industries lobby for smaller tariff
cuts (Grossman and Helpman 1994).
There are several reasons to believe that these concerns are
less important for the People’s Republic of China. Viewing WTO
membership as a means of locking the country on a path of deepening
economic reform and openness, the government has demonstrated a
greater desire to open rather than protect its domestic industries (Woo
2001). Additional supporting evidence comes from examining the
relationship between tariff cuts and preliberalization employment. If
policy makers had permitted “stronger” industries to bear larger tariff

Figure 3.1 Tariff Changes and Preliberalization Tariff Levels

Ferrous Metals Gas

0.00 Nonferrous Metals Coal
Other Minerals
Petroleum Processing and Coking
Nonferrous Pressing RubberMetal Products
Electric Power Forestry Animal Husbandry
Ferrous Pressing Chemicals
SpecialNonmetal Products
2000–2005 change in log(1+t)

Petroleum Extraction Machinery

–0.05 Pharmaceuticals
Machinery Electrics
Other Manufacturing
Office Machinery
Timbers Plastics Sports Goods
Telecommunications Foods
Farming Apparel
Paper ProductsTransports
–0.10 Chemical FiberTextiles

Tobacco
–0.15
Furniture

–0.20

–0.25
Beverages

0.0 0.1 0.2 0.3 0.4 0.5

2000 prelibralization log(1+t)

Notes: This figure plots log tariff changes over 2000–2005 against the log 2000 tariff levels. The
sectoral tariff is calculated based on the simple average of most-favored nation applied tariff rates
at the Harmonized System HS6 product level from the Trade Analysis Information System database.
Correlation: –0.84; regression coefficient: –0.43; standard error: 0.044; t: –9.60.
Source: Author’s calculation.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 89

cuts, industries with higher employment growth between 1990 and 2000
would have experienced greater tariff reductions; if large industries
lobbied more or were more likely to be protected due to employment
concerns, industries with larger employment (in levels) in 2000 would
have experienced lower tariff cuts. However, I find only marginal and
statistically insignificant correlation between tariff changes and pre-
WTO industry employment in both changes and levels, the simple
correlations being 0.13 and 0.16, respectively.
Following the approach of Goldberg and Pavcnik (2005), Figure 3.1
demonstrates that industries with high tariffs in 2000 experienced the
largest tariff cuts, with the correlation between the 2000 tariff levels
and the change in tariffs being –0.84. The fact that the pre-WTO tariff
levels largely determined the tariff changes following the People’s
Republic of China’s WTO accession implies that the primary goal of
policy makers was to reduce tariff rates and to smoothen cross-industry
variations. This rules out the industry protection and political economy
concerns.
Most importantly, even after rounds of voluntary tariff reductions,
the country’s tariff structure in 2000 remained similar to that of 1992,4

Table 3.1 Imports and Tariffs

Import Values Varieties Unit Values

All Products Intermediates All Products Intermediates All Products Intermediates

(1) (2) (3) (4) (5) (6)
Output tariff –0.22*** –0.16*** –1.06*** –1.01*** 0.02 0.05**
(0.05) (0.06) (0.12) (0.13) (0.02) (0.02)
Year fixed Yes Yes Yes Yes Yes Yes
effect
HS6 fixed Yes Yes Yes Yes Yes Yes
effect
Observations 35,457 26,380 35,457 26,380 33,695 25,193
R-squared 0.08 0.09 0.27 0.29 0.10 0.09
Number of 5,222 3,904 5,222 3,904 5,124 3,830
hs2002
HS = Harmonized System.
Notes: Coefficient on tariffs from HS6 product level regression of log import value, HS8 variety numbers, unit
value on lagged output tariffs, HS6 product fixed effects, and year fixed effects. An observation is HS6-year. The
data cover nonprocessing trade in 2000–2006. Robust standard errors in parentheses. *** p<0.01, ** p<0.05,
* p<0.1.
Source: Author’s calculation.

4
1992 is the earliest year that tariff data for the People's Republic of China at the HS6
level are available.
90 Trade Adjustment in Asia: Past Experiences and Lessons Learned

with a correlation of 0.93. On the other hand, the bound duties after
joining the WTO were largely imposed externally, benchmarking the
tariff levels of other WTO members. Unlike in many other developing
countries, there is almost no gap between the People’s Republic of
China’s bound and applied duties, and the binding coverage is 100%.
This implies that the preliberalization tariffs in the People’s Republic
of China were based on a protection structure that was set a decade
earlier, while postliberalization tariffs were externally set. Therefore,
it is highly unlikely that tariff reductions between 2000 and 2005 are
correlated with counterfactual industry employment changes.

3.2.4 Tariff Reductions, Trade Surge,

and Employment Changes

Before analyzing the relationship between input tariff reduction and labor
reallocation, I first examine whether the tariff reduction induced by the
People’s Republic of China’s WTO accession was systematically related
to its trade expansion. To summarize the findings, I find that (i) lower
tariffs led to an overall increase in trade values, (ii) lower tariffs led to an
increase in imports in the number of varieties within HS6 categories, and
(iii) lower tariffs resulted in lower unit values of existing product lines,
with particularly pronounced effects on intermediate products.
These results are summarized in Table 3.1. I begin by examining the
responsiveness of import values to tariffs by regressing the log import
value of an HS6 product on HS6 log tariff levels, HS6 fixed effects, and
year fixed effects. I restrict my analysis to the period 2000–2006, in
which I have access to customs data to calculate HS6 product variety
numbers. Extending the analysis to the period 2000–2010 yields
similar results. For all regressions, I exclude processing trade flows
as they are not affected by tariff reductions. Column (1) of Table 3.1
reports the coefficient estimates of tariffs for all sectors, and column
(2) for intermediate sectors based on the Broad Economic Categories
classification. In both cases, we can note that declines in tariffs are
associated with higher import values.
Recent theory also emphasizes the benefits gained by increasing
the imported varieties. As we can see from columns (3) and (4), this
channel also plays a role in our context. I define varieties as the number
of distinct HS8 products within a given HS6 product category. I then
regress the number of varieties on the tariff, HS6 fixed effects, and year
fixed effects. For all sectors and only intermediate sectors, a decline
in tariffs is associated with an increased number of varieties. Lastly,
I examine the impact of tariff reduction on the unit price of imports. The
estimation results are presented in columns (5) and (6). Tariff declines
are associated with decreases in import unit values, but this relationship
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 91

Figure 3.2 Regional Employment and Import Changes

0.6

Regional Employment Changes, 2000–2010

0.4

0.2

0.0

–0.2

–0.4

5 10 15 20 25
Nonprocessing Import Changes, 2000–2006

Note: See text for details.

Source: Author’s calculation.

is only statistically significant for intermediate sectors. That is, the

benefits from trade that occur through increasing imports at a lower
price are particularly true of intermediate goods, consistent with the
beneficial effects of input tariff reduction on regional employment,
which I probe formally in later sections.
Next, I examine the relationship between increased imports and
employment changes across cities in the People’s Republic of China. In
Figure 3.2, I plot the difference in log employment between 2000 and
2010 against the change in log nonprocessing imports between 2000 and
2006 at the prefecture city level. As we can see from Figure 3.2, cities
with larger import increases are also associated with larger increases in
regional employment. Combining the examination of the relationships
between tariff reductions and trade surges, we can be confident that the
impact of input tariff reductions on spatial labor reallocation is indeed
channeled through changes in trade flows.

3.3 Measurements and Specifications

3.3.1 Local Labor Markets

Throughout the empirical analysis, local labor markets are defined as

prefectures. A prefecture is an administrative division of the People’s
Republic of China that ranks below a province and above a county. Given
92 Trade Adjustment in Asia: Past Experiences and Lessons Learned

that most regional policies, including the overall planning of public

transportation, are conducted at the prefecture level (Xue and Zhang
2001), I expect counties within the same prefecture to have strong
commuting ties and to be economically integrated. In order to account
for prefecture boundary changes, I use information concerning the
administrative division changes published by the Ministry of Civil Affairs
of the People’s Republic of China to create time-consistent county groups
based on prefecture boundaries in the year 2000. This results in 337
geographic units, which I describe as prefectures or regions, including
four direct-controlled municipalities and 333 prefecture-level divisions
that cover the entire country. Relative to commuting zones in the United
States, the prefectures in the People’s Republic of China are about twice
as large on average and 1.5 times the size when the 10 largest (but sparsely
populated) prefectures in autonomous regions are excluded.
The empirical analysis in this chapter studies 10-year changes in
prefecture employment, total and working-age populations, the most
recent 5-year migrant inflows from other provinces, and the population
holding local hukou in each prefecture. I collect these variables at the
county level from the Tabulation on Population Census of the People’s
Republic of China by county for 2000 and 2010, and then aggregate them
to prefectures based on the time-consistent county groups. Notably,
the employment measure includes informal workers, the lion’s share
of whom are migrants.5 Between 2000 and 2010, the People’s Republic
of China underwent a significant change in its spatial distribution of
employment, with some prefectures seeing over a 50% increase in local
employment, while others experienced more than a 30% decrease.

3.3.2 Regional Trade Shock Exposures

To construct the exposure of local labor markets to input tariff reductions,

I combine data on regional industry employment with data on tariffs
and industry cost shares. Data on regional employment by industry
in 2000 were collected from the Tabulation on the 2000 Population
Census published by each province. The original data are by county

5
According to Park, Wu, and Du (2012), informal employment in the People’s Republic
of China is defined either based on (i) whether or not the employer fails to provide
all of the three most important types of social insurance that employers are expected
to provide in the People’s Republic of China (i.e., pensions, health insurance, and
unemployment insurance), or (ii) whether workers have a labor contract. Migrant
workers contribute to 49.0% of the informal employment in the People’s Republic of
China under the first definition and 65.7% under the second. The employment data
from population census include all informal workers, as long as they engaged in at
least 1 hour of paid work the week before the survey date, or were on leave.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 93

and by 92 two-digit 1994 Chinese Standard Industrial Classification

(CSIC1994), which I aggregate to prefecture level.6 I use the simple
average of most-favored nation applied tariffs at the HS6 product level
from the United Nations Conference on Trade and Development Trade
Analysis Information System database to calculate tariff changes. The
cost share of each industry is constructed as its share of value in the
output industry using the 2002 Chinese national input–output (IO)
table.7 To utilize these various datasets, I also construct a common
industry classification, which consists of 71 industries, including five
agricultural and 28 nontraded industries.8
As is standard in the literature, I measure input tariff cuts (ΔIT) as
the input-cost weighted average of tariff reductions:

Δ𝐼𝐼𝑇𝑇𝑠𝑠 = ∑ 𝛼𝛼𝑠𝑠 (𝑘𝑘)𝑑𝑑ln(1 + t 𝑘𝑘 ), (1)

𝑘𝑘ϵK
where αs(k) represents the cost share of industry s due to purchases
from industry k, tk is theΔ𝐼𝐼𝑇𝑇 tariff rate of industry k, and d represents the
Δ𝐼𝐼𝑇𝑇𝑠𝑠 = ∑ 𝛼𝛼𝑠𝑠 (𝑘𝑘)𝑑𝑑ln(1 + t 𝑘𝑘 ), 𝑠𝑠 = ∑ 𝛼𝛼𝑠𝑠 (𝑘𝑘)𝑑𝑑ln(1 + t 𝑘𝑘 ),
long-difference (Autor, Dorn, and Hanson framework) between 2000
and 2005. FollowingΔRIT 𝑖𝑖 =(2013)∑ 𝑘𝑘ϵK
𝛿𝛿𝑖𝑖𝑖𝑖and
ΔITDix-Carneiro
s , 𝑠𝑠 = ∑ 𝛼𝛼𝑠𝑠 (𝑘𝑘)𝑑𝑑ln(1 and
𝑘𝑘ϵK Δ𝐼𝐼𝑇𝑇 + t 𝑘𝑘 ),
Kovak 𝑘𝑘ϵK
Kovak (2017),
I calculate the regional input tariff 𝑠𝑠ϵK cuts (∆RIT) as follows:
ΔRIT𝑖𝑖 = ∑ 𝛿𝛿𝑖𝑖𝑖𝑖 ΔITs ,
ΔRIT𝑖𝑖 = ∑ 𝛿𝛿𝑖𝑖𝑖𝑖 ΔITs ,
𝑠𝑠ϵK
ΔRIT𝑖𝑖 = ∑ 𝛿𝛿𝑖𝑖𝑖𝑖 ΔITs𝑠𝑠ϵK, (2)
1
1 𝐿𝐿 𝑠𝑠ϵK
𝑖𝑖𝑖𝑖 𝜙𝜙 1
𝐿𝐿𝑖𝑖𝑖𝑖 𝜙𝜙 𝑠𝑠 𝐿𝐿𝑖𝑖𝑖𝑖 𝜙𝜙
𝑠𝑠 𝛿𝛿𝑖𝑖𝑖𝑖 = 1 amount
𝑠𝑠
where 𝑖𝑖𝑖𝑖 ∑ 𝐿𝐿 ( 1 ) , Lis is the initial
𝛿𝛿 = 𝛿𝛿𝑖𝑖𝑖𝑖 = 1of labor allocated to
𝑠𝑠∈𝛫𝛫 𝑖𝑖𝑖𝑖 𝜙𝜙 ∑ 𝐿𝐿
𝑠𝑠∈𝛫𝛫 s 𝑖𝑖𝑖𝑖 ( 𝜙𝜙𝑠𝑠 ) ∑𝑠𝑠∈𝛫𝛫 𝐿𝐿𝑖𝑖𝑖𝑖 (𝜙𝜙 )
industry in region i, and ϕs is 1 – the wage bill
𝑠𝑠
𝑠𝑠

share of the industry value added.𝐿𝐿 In a specific-factor model with a 1

𝑖𝑖𝑖𝑖 𝜙𝜙1
constant1 returns production 𝛿𝛿𝑖𝑖𝑖𝑖 =function, 𝜙𝜙𝑠𝑠 represents
𝑠𝑠 the labor demand
elasticity
𝜙𝜙𝑠𝑠 (Kovak 2013). The weight δis captures 1 the intuition behind the
1 ∑𝑠𝑠∈𝛫𝛫 𝐿𝐿𝑖𝑖𝑖𝑖 (𝜙𝜙 )
construction of ∆RIT: a prefecture willΔ𝑌𝑌experience 𝑠𝑠 a larger
′ increase in
employment if its workers 𝜙𝜙𝑠𝑠 ′ are specialized𝑖𝑖 =inβ1industries ΔRITi + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 𝟏𝟏 𝛾𝛾 + 𝜖𝜖𝑖𝑖
with large input
Δ𝑌𝑌𝑖𝑖 = β1 ΔRITi + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 𝟏𝟏 𝛾𝛾 + 𝜖𝜖𝑖𝑖
Δ𝑌𝑌𝑖𝑖 = β2 ΔRITi + β3 ΔRITi ∗ 𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝑢𝑢𝑝𝑝 + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′𝟐𝟐 𝛾𝛾 + 𝜖𝜖𝑖𝑖 ,
6
The Δ𝑌𝑌
2010 employment 1
by iindustry
∗ 𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝑢𝑢𝑝𝑝has
+ 𝐷𝐷many
𝑝𝑝 + 𝐗𝐗 missing values, so I perform all analyses
′
𝑖𝑖 = β 2 ΔRITi + β3 ΔRIT 𝟐𝟐 𝛾𝛾 + 𝜖𝜖𝑖𝑖 ,
′
at the regional rather Δ𝑌𝑌
than
𝑖𝑖 =the
𝜙𝜙 β
𝑠𝑠 1 ΔRIT
region-industry
i + 𝐷𝐷 𝑝𝑝 +
level. 𝐗𝐗 𝟏𝟏 𝛾𝛾 + 𝜖𝜖𝑖𝑖
7
Given that trade liberalization began in 2001, I use the IO table of the closest year.
I do so under the assumption that industries’ cost structures adjust slowly to trade
reforms. I do not use the 1997 IO table for two reasons: first, the 1997 IO table uses an
industry classification that is less consistent with employment data; second, it might
Δ𝑌𝑌𝑖𝑖 =
understate the importance
Δ𝑌𝑌 β2= ΔRIT
β1 ΔRIT
of𝑖𝑖tradable
+ βi3due
iinputs +ΔRIT
𝐷𝐷to𝑝𝑝the
i+ 𝐗𝐗 ′𝟏𝟏 𝛾𝛾financial
∗Asian
𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝑢𝑢 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′𝟐𝟐 𝛾𝛾
+ 𝑝𝑝𝜖𝜖𝑖𝑖+ crisis. + 𝜖𝜖𝑖𝑖 ,
8
The common industry classification is created to achieve the maximum disaggregation
between different classifications; the 2002 IO table consists of 122 industries and
is coded similarly to the 1994/2002 Chinese Standard Industrial Classification
(CSIC1994/CSIC2002). Δ𝑌𝑌𝑖𝑖 = β2 ΔRITi + β3 ΔRITi ∗ 𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝑢𝑢𝑝𝑝 + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′𝟐𝟐 𝛾𝛾 + 𝜖𝜖𝑖𝑖 ,
94 Trade Adjustment in Asia: Past Experiences and Lessons Learned

tariff declines, and more so if these industries are elastic in labor demand.
Nevertheless, my empirical results are robust to using a weight that is
based on employment only.
Disparities in industry weights across regions generate substantial
variations in their exposure to input trade liberalization. The three
hubs of the People’s Republic of China’s trade and economic growth—
the Bohai Economic Rim, the Yangtze River Delta, and the Pearl River
Delta—are among the top beneficiaries of input trade liberalization.
Western prefectures that are specialized in animal husbandry or basic
food processing and manufacturing benefited greatly from tariff cuts
in farming industries, and hence also experienced large reductions in
regional input tariffs.
Similarly to calculating the regional input tariff cuts, I compute
regional output tariff reductions as a δis -weighted average of industry-
specific tariff reductions over 2000–2005. In order to calculate external
tariff reductions, I first use customs data for the People’s Republic of
China for 2000 to compute prefecture exports and calculate the export
share by destination country for each industry and prefecture. I then
take the export-share weighted average of the tariff changes across
destination countries to obtain prefecture industry-specific tariff
reductions. In the final step, I compute the weighted average tariff
changes across industries using δis for each prefecture.

3.3.3 The Hukou Measure

The primary dataset that I use to construct the hukou measure is the
0.095% random sampled data of the Population Census in 2000. The
complete dataset covers the entire population of the for the People’s
Republic of China, and the sample was randomly drawn at the household
level, with a unique identifier linking individuals in the same household.
The dataset contains rich individual-level information including one’s
hukou registration status and migration history in the last 5 years, from
which I can infer the stringency of a prefecture’s hukou system based on
the likelihood of an individual obtaining a local hukou after settling in that
prefecture. In reality, the likelihood of an individual acquiring or being
granted a local hukou also depends on various individual characteristics.
In order to draw out these effects, I calculate the hukou measure as
follows. I focus on individuals who moved between 1995 and 2000 to
a prefecture that is not their birthplace.9 I regress a dummy equal to 1

9
In the early 1990s, most internal migration was state planned, guaranteeing local
hukou to migrants. I therefore focus on the most recent 5 years. The raw dataset
contains 1,180,111 observations; given that most people never migrate, the number of
observations in my regressions is 62,289.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 95

if the individual had already obtained a local hukou before November

2000 (when the census was conducted) on age, age squared, gender,
ethnicity (Han versus other), marriage status (ever married), difference
in log gross domestic product (GDP) per capita between the migrate-out
and migrate-in provinces,10 migrate-from-rural-areas dummy, migrate-
within-province dummy, categorical variables for education and for
the years of residence Δ𝐼𝐼𝑇𝑇𝑠𝑠 =in∑ 𝑠𝑠 (𝑘𝑘)𝑑𝑑ln(1
the𝛼𝛼current t 𝑘𝑘 )prefecture
city,+and , fixed effects.
Δ𝐼𝐼𝑇𝑇
I then 𝑠𝑠 =a∑
take simple𝛼𝛼𝑠𝑠 (𝑘𝑘)𝑑𝑑ln(1
average + the
of t 𝑘𝑘 ),estimated prefecture fixed effects by
𝑘𝑘ϵK
province and𝑘𝑘ϵK normalize it from 0 to 1 to obtain the final measure.
The hukou measure is an inverse indicator of migration frictions
associated with the hukou system: it equals 0 if a province has the most
stringent hukou-granting ΔRIT𝑖𝑖 =practice.
∑ 𝛿𝛿𝑖𝑖𝑖𝑖 ΔIT Consistent
s, with common knowledge,
my hukou = ∑ 𝛿𝛿𝑖𝑖𝑖𝑖suggest
ΔRIT𝑖𝑖 measure ΔITs , that Beijing, Shanghai, and Guangdong are
𝑠𝑠ϵK
among the most 𝑠𝑠ϵK difficult provinces to obtain a local hukou. In addition,
there is no correlation between the GDP per capita of a province and
its hukou policy. For instance, Qinghai and Xinjiang have very stringent
1
hukou policies, which 1 are more 𝐿𝐿𝑖𝑖𝑖𝑖 𝜙𝜙
likely driven by limited farming land and
𝐿𝐿
political stability𝑖𝑖𝑖𝑖𝛿𝛿concerns.
𝜙𝜙
𝑠𝑠
Hukou stringency is not determined by the
𝑖𝑖𝑖𝑖𝑠𝑠 = 1
𝛿𝛿𝑖𝑖𝑖𝑖population
initial = density of a 𝐿𝐿
region
1 ∑𝑠𝑠∈𝛫𝛫 𝑖𝑖𝑖𝑖 (𝜙𝜙 either,
) with some densely populated
provinces ∑ 𝑠𝑠∈𝛫𝛫 𝐿𝐿
such (𝜙𝜙 ) having a𝑠𝑠rather liberal hukou system, while
as𝑖𝑖𝑖𝑖 Henan𝑠𝑠
other densely populated regions such as Beijing have a stringent system.

3.3.4 Empirical1 Specification

1
𝜙𝜙𝑠𝑠
𝜙𝜙𝑠𝑠 the regional input tariff cuts and the hukou measure at hand,
Given
I estimate the following equations in the next subsection:

Δ𝑌𝑌𝑖𝑖 = β1 ΔRITi + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′𝟏𝟏 𝛾𝛾 + 𝜖𝜖𝑖𝑖 (3)

Δ𝑌𝑌𝑖𝑖 = β1 ΔRITi + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′𝟏𝟏 𝛾𝛾 + 𝜖𝜖𝑖𝑖
and

Δ𝑌𝑌𝑖𝑖 = β2 ΔRITi + β3 ΔRITi ∗ 𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝑢𝑢𝑝𝑝 + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′ 𝛾𝛾 + 𝜖𝜖𝑖𝑖 ,

Δ𝑌𝑌𝑖𝑖 = β2 ΔRITi + β3 ΔRITi ∗ 𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝐻𝑢𝑢𝑝𝑝 + 𝐷𝐷𝑝𝑝 + 𝐗𝐗 ′𝟐𝟐 𝛾𝛾 + 𝜖𝜖𝑖𝑖 , (4)𝟐𝟐

where the second specification explores the heterogeneous regional

effect of input tariff reductions depending on the hukou frictions. Here,
ΔYi is the decadal change of the log value of a regional outcome variable
such as employment or total population; β1 captures the regional effect
of input trade liberalization on the variable of interest during the
2000–2010 period, while β2 and β3 represent the heterogeneous impact

10
I obtain GDP per capita data from the 2000 provincial statistical yearbooks. Note that
it is important to control for GDP differences, as a migrant from a more developed
area might not be willing to switch and acquire a local hukou.
96 Trade Adjustment in Asia: Past Experiences and Lessons Learned

of input tariff reductions depending on hukou frictions; Dp are province

fixed effects; and X represents a set of additional controls. In the main
specification, X includes regional output tariff reductions, external
tariff reductions, and the preliberalization level of the outcome variable
to control for increased import competition, improved market access,11
and possible mean convergence. Hukoup is the hukou friction measure;
in equation (4), I also control for its interaction with external and output
tariff reductions.

3.3.5 Empirical Results

The impacts of input tariff cuts on the People’s Republic of China’s

regional labor adjustments are summarized in Table 3.2. In columns (1)
and (2), I first present the results of regressing employment changes
on regional input tariff cuts. The standard errors are clustered at the
provincial level, accounting for the possible covariance between the
error terms across prefectures within the same province. Regressions
are weighted by the log of the employment in the initial period. As we
can see from column (1), the coefficient of ∆RIT is significant at the 1%
level and has the expected positive sign. The estimate of 4.92 implies
that a prefecture facing a 1-percentage-point regional input tariff cut
experiences an almost 5-percentage-point employment increase. The
difference between regional input tariff cuts in regions at the 25th and
75th percentiles is 0.7 percentage points. Evaluated using the estimate in
column (1), a region at the 75th percentile experiences a 3.4-percentage-
point larger employment increase than a region at the 25th percentile.
Consistent with the existing literature, I find that regional output
tariff reductions have a negative impact on employment, although at
a smaller magnitude compared to the impact of input tariff cuts. The
effect of external tariff reductions has the expected positive sign but is
statistically insignificant.
In column (2), I add the interaction term between input trade
liberalization and the hukou measure, probing whether input
liberalization-induced employment adjustments are more pronounced
in provinces with relatively free hukou systems. Given that I normalized
my hukou measure to the unit interval, coefficients of ∆RIT directly
reflect the impact of input tariff cuts in prefectures with the highest
hukou frictions. The coefficient for the interaction term is positive

11
External tariff reductions capture the positive impact of tariff reductions by the
People’s Republic of China’s trading partners following its WTO accession. However,
this is less of a concern as most countries had already granted the People’s Republic
of China most-favored nation status prior to 2001.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 97

and statistically significant. Input tariff reductions have no impact on

regional employment in the provinces with the most stringent hukou
systems. In contrast, in regions with the most relaxed hukou systems, a
1-percentage-point increase in input tariff cuts leads to a 16-percentage-
point relative increase in employment, which is much larger than the
5-percentage-point average found in column (1). I also find a weak
relationship between the effect of output tariff reductions and hukou
stringency, although the result is only significant when fixed effects
are included. This is consistent with the fact that the hukou system is
primarily designed to control for migration inflows. On the other hand,
the interaction terms between external tariff change and hukou frictions
have the opposite sign. Calculated based on the specification in column
(2), the partial R-squared of regional input tariff cuts, regional output
tariff cuts and their interactions with the hukou measure is 0.35. This
suggests that when considering both input and output channels, over
30% of the regional variation in employment changes could be accounted
for by trade liberalization.

Table 3.2 Effect of Input Tariff Cuts

Working-Age
Employment Population
(1) (2) (3) (4)
Regional input tariff cuts (∆RIT) 4.92*** –0.06 4.33*** –1.00
(1.44) (1.53) (1.46) (1.50)
Regional input tariff cuts x Hukou 15.70*** 16.43***
(4.45) (4.31)
Regional output tariff change –2.73*** –3.81** –2.20*** –2.83***
(0.67) (0.92) (0.59) (0.83)
Regional output tariff change x Hukou 4.52** 3.53*
Regional input tariff cuts (∆RIT) (2.04) (1.88)
External tariff change 0.10 0.73 0.16 1.10*
(0.19) (0.54) (0.22) (0.54)
Controls Yes Yes Yes Yes
Province fixed effects (31) Yes Yes Yes Yes
Observations 337 337 337 337
R-squared 0.62 0.65 0.58 0.63
continued on next page
98 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 3.2 continued

Migrant Inflows Hukou Population
(5) (6) (7) (8)
Regional input tariff cuts (∆RIT) 13.16** –5.55** 1.25 –2.51
(5.56) (2.05) (0.77) (2.59)
Regional input tariff cuts x Hukou 61.99*** 10.23**
(15.41) (4.65)
Regional output tariff change –3.73 –2.92 –2.84*** –2.06
(2.49) (2.61) (0.70) (1.35)
Regional output tariff change x Hukou 3.14 –1.26
Regional input tariff cuts (∆RIT) (5.39) (2.66)
External tariff change 0.72 –0.06 0.17 0.68*
(1.23) (3.76) (0.12) (0.36)
Controls Yes Yes Yes Yes
Province fixed effects (31) Yes Yes Yes Yes
Observations 337 337 337 337
R-squared 0.41 0.44 0.70 0.72
Notes: The dependent variable is the difference in log employment, log working-age population (15 to
64 years old), log population that migrated from other provinces between 2005–2010 and 1995–1990,
and the 10-year change in log prefecture population holding local hukou permit for columns (1) and (2),
(3) and (4), (5) and (6), and (7) and (8), respectively. The sample contains 333 prefectures and four direct-
controlled municipalities. All regressions include the full vector of control variables; models with interaction
terms further include the interaction between the hukou measure and other tariff changes as in column (6)
of Table 3.1. Prefecture birth and death rates are also controlled in columns (7) and (8). Robust standard
errors in parentheses are adjusted for 31 province clusters. Models are weighted by the log of beginning-
period prefecture population. *** p<0.01, ** p<0.05, * p<0.1.

Source: Author’s calculation.

In combining the results from columns (1) and (2), we know

that prefectures facing larger input tariff cuts experience a relative
increase in employment, and the effect is stronger in provinces
with a lower amount of hukou friction. This observed change in
regional employment can be caused by both intra- and interregional
adjustments. A positively affected region may experience a decline
in unemployment and an increase in labor force participation,
both of which may result in an increase in local employment. To
ensure that it is the spatial reallocation of labor that drives the
employment adjustment, I next look at how working-age (15 to 64
years) populations respond to input tariff reductions. If the observed
employment changes are mainly due to intraregional adjustments,
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 99

trade shocks should have no impact on the local population, whereas

if the change is primarily due to interregional adjustments, the local
population should react to trade shocks in a quantitatively similar
way to that of employment.
Columns (3) and (4) report the results of regressing the regional
change of log working-age populations on regional input tariff cuts, with
and without interactions. I include the full set of controls and cluster
standard errors at the provincial level. The results strongly support
interregional labor reallocation: column (3) shows that prefecture-
level working-age populations react positively and significantly to input
tariff cuts, and the coefficients are quantitatively similar to those of
employment. On average, a 1-percentage-point increase in regional input
tariff cuts leads to a 4.33-percentage-point increase in the working-age
population of a prefecture. In a prefecture with the least hukou frictions,
a 1-percentage-point increase in regional input tariff cuts leads to a
16.43-percentage-point increase in the working-age population, which
reinforces the notion that the observed regional employment adjustment
is due to interregional labor reallocation.
Compared to indirectly inferring spatial adjustments in labor
from regional population changes, it would be preferable to examine
migration directly. However, the ideal measure, i.e., the decadal
change in net migration inflows, is not available. Therefore, I instead
consider the most similar variable available in the census: the number
of migrants from other provinces in the past 5 years. It is important to
note that compared to the ideal measure, this variable is likely to provide
an insignificant estimate. First, interregional migration occurs much
more frequently within provinces than across them. Second, since this
variable counts migrant inflows in 5-year periods, I compare the number
of migrants between 1995 and 2000 with those between 2005 and 2010.
As tariff reductions began in 2001, if their impact levels off quickly, I will
not be able to find a significant result.
With the above concerns in mind, I regress the change in the log
5-year inflow of population from other provinces on regional input tariff
reductions, with and without interactions. The results are presented in
columns (5) and (6) of Table 3.2, respectively. Column (5) reports that
a 1-percentage-point increase in regional input tariff reduction leads
to a 13.16-percentage-point increase in migrant inflows from other
provinces. Column (6) confirms that input tariff cuts result in larger
migrant inflows when the hukou system is less stringent. Both estimates
are significant at the 5% level. Since migration is a flow rather than a
stock variable, the magnitude of the estimates is much larger. These
results provide additional support to the finding that regional input
tariff cuts increase local employment through attracting labor from
100 Trade Adjustment in Asia: Past Experiences and Lessons Learned

other locations, and this effect crucially depends on frictions caused by

the hukou system.
Interestingly, I find that neither the regional output tariff reductions
nor the external tariff changes have a significant effect on migrant inflows,
nor does their impact heterogeneously depend on hukou frictions. These
results further suggest that among various trade shocks associated with
the accession to the WTO, input tariff liberalization seems to have played
the dominant role in shaping labor reallocation in the country.
Finally, columns (7) and (8) of Table 3.2 show how the number of
individuals holding local hukou (hukou population) in a prefecture
responds to input tariff reductions. In these regressions, I further control
for prefecture birth and death rates to address two additional concerns.
One is that input tariff cuts may generate different life expectancies
across regions, affecting hukou population changes. The second concern
is that it may generate different family planning behavior across regions
(i.e., in positively affected areas, families may be willing to have more
children). The latter is going to impact the hukou population via birth
rate changes, as the children of local hukou holders are automatically
granted a local hukou.
If a local hukou can be obtained without cost, the hukou population
should be highly correlated with total population in a given region, and
hence react positively to input tariff reductions. The empirical results,
however, point to the contrary: column (7) indicates that on average,
reductions in regional tariffs do not cause significant changes in the
hukou population. However, in prefectures with less stringent hukou
systems, the hukou population does increase in positively affected
regions. Column (8) indicates that in a prefecture with the freest hukou
system, a 1-percentage-point increase in regional input tariff cuts leads
to a 10.23-percentage-point increase in the hukou population. The
magnitude, however, is only two-thirds of the input liberalization-
induced increase in the working-age population (column (4), Table 3.2).
This implies that hukou frictions are substantial even in regions with the
least stringent system. This represents evidence that even though many
migrant workers moved to cities that were positively affected by trade
shocks, few were able to obtain a local hukou.

3.4 Welfare Quantifications

The empirical results presented in the previous section can be viewed
through the lens of a multisector spatial equilibrium model, which
allows for IO linkages and migration frictions. Falling trade costs allow
firms to access cheaper intermediate inputs and hence produce less
expensive final goods. As a result, demand for local production increases.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 101

Regions whose labor is concentrated in industries facing larger input

tariff reductions are more positively affected, thus pushing up wages and
attracting workers from other places. Immigration increases the price
of nontradables and depresses wages until the economy reaches a new
equilibrium. When migration frictions are high, employment changes
are small due to limited migration, and real wages instead react strongly
to trade shocks. In contrast, when migration friction is low, employment
reacts more strongly to trade shocks and real wages react less.
In standard economic geography models with no migration
frictions, workers will move to arbitrage away the utility difference of
living in different cities. When we allow for amenity or productivity
difference across locations and heterogeneous workers, real wages can
differ across cities. In this case, a city with a high wage in equilibrium
either reflects its relative technology superiority, or the fact that it is
less livable than the other cities, and so people demand higher wages
to compensate for the relatively poor local amenities. In equilibrium,
we may observe that some cities offer more employment than others,
but from the workers’ perspective, they are indifferent about where to
live given the equilibrium wage and amenity in each location. Therefore,
when the economy is hit by a trade shock, the expected utility change
will be the same among all workers.
However, with the presence of the hukou system, a larger fraction
of the gains accrues to workers with a local hukou. This is intuitive: as
migrant inflows are limited in positively affected regions when hukou
friction is high, real wages increase. Workers with a local hukou face no
constraints on living in the particular region, and hence benefit more
significantly from the local wage increase. In other words, the presence
of a hukou affects the welfare distribution across otherwise identical
worker groups.
In Table 3.3, I present the quantification of the welfare change of
workers holding a hukou of different provinces in the People’s Republic
of China, and the simulated regional adjustment in other margins
following trade liberalization. The five provinces with the largest
increases in employment are Beijing, Shanghai, Guangdong, Tianjin, and
Fujian, with Beijing experiencing an increase in employment of 0.55%
and Shanghai of 0.40%. The five provinces with the lowest employment
increase are Hubei, Hunan, Sichuan, Anhui, and Jiangxi, consistent
with our observations in the real data. These simulations are based on
static models, hence the magnitudes of adjustments are smaller than our
observations from real data.12

12
The simulated numbers may change slightly depending on model assumptions; see
Zi (2018) for details.
102 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 3.3 Regional Adjustments to Trade Liberalization

(1) (2) (3) (4) (5) (6) (7)

Province or
Autonomous Real
Region Employment Wage GDP Price Exports Imports Welfare
Guangdong 0.19 0.95 1.12 –1.77 5.27 7.05 0.93
Beijing 0.55 1.80 2.27 –2.06 9.62 3.34 1.69
Shanghai 0.40 1.61 1.93 –1.85 6.30 4.02 1.50
Tianjin 0.17 1.43 1.57 –1.96 7.62 4.65 1.38
Fujian 0.08 0.97 1.04 –1.69 6.67 7.14 0.96
Zhejiang 0.06 0.89 0.93 –1.60 5.18 5.69 0.87
Jiangsu 0.05 1.14 1.17 –1.79 9.40 6.85 1.12
Xinjiang 0.03 0.67 0.67 –1.53 6.42 5.25 0.64
Uygur
Liaoning 0.02 0.96 0.97 –1.76 8.12 6.50 0.95
Hainan 0.01 0.74 0.76 –1.66 2.76 6.44 0.75
Jilin 0.00 0.73 0.73 –1.74 6.71 5.08 0.73
Yunnan –0.01 0.49 0.48 –1.49 7.48 5.15 0.49
Shandong –0.01 0.70 0.68 –1.71 5.92 8.72 0.69
Inner –0.02 0.52 0.50 –1.54 4.76 6.75 0.52
Mongolia
Ningxia Hui –0.02 0.40 0.38 –1.51 9.17 3.99 0.41
Qinghai –0.02 0.40 0.38 –1.51 6.72 5.12 0.40
Shanxi –0.02 0.37 0.35 –1.48 7.23 4.27 0.37
Shaanxi –0.02 0.46 0.43 –1.60 5.15 5.58 0.46
Heilongjiang –0.03 0.59 0.56 –1.66 4.06 6.97 0.59
Chongqing –0.03 0.48 0.45 –1.54 2.67 4.32 0.48
Gansu –0.03 0.31 0.28 –1.52 5.72 3.82 0.31
Hebei –0.05 0.41 0.36 –1.62 5.20 7.88 0.42
Henan –0.05 0.38 0.33 –1.56 5.15 5.23 0.40
Hubei –0.05 0.48 0.43 –1.59 5.14 5.13 0.50
Guangxi –0.05 0.55 0.50 –1.50 6.59 4.81 0.57
Zhuang
Guizhou –0.05 0.40 0.35 –1.53 4.76 4.50 0.42
Hunan –0.07 0.46 0.39 –1.53 7.48 5.12 0.49
Sichuan –0.08 0.47 0.39 –1.58 4.27 4.81 0.49
continued on next page
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 103

Table 3.3 continued

(1) (2) (3) (4) (5) (6) (7)
Province or
Autonomous Real
Region Employment Wage GDP Price Exports Imports Welfare
Anhui –0.11 0.52 0.41 –1.53 4.34 5.18 0.55
Jiangxi –0.12 0.39 0.28 –1.47 3.85 5.36 0.44
Weighted 0.63
average
Standard 0.27
deviation
GDP = gross domestic product.
Notes: Counterfactual percentage changes in regional employment, real wage, real GDP (total value added
divided by local consumption price index), consumption price index, exports, and imports when the tariff
structure of the People’s Republic of China changed from its 2000 to 2005 level, holding hukou frictions
constant. The nominal wage of the constructed rest of the world is the numeraire.
Source: Author’s calculation.

Column (2) shows that real wages increase in all provinces and
that they are positively correlated with changes in employment. When
comparing changes in real wages and employment, two patterns stand
out. First, regional employment reacts less to trade shocks than do wages
(regressing employment changes on wage changes yields coefficient of
0.47), indicating substantial internal migration frictions in the People’s
Republic of China. Second, a region with a larger real wage increase is
not necessarily a region with a greater increase in employment. To see
this dynamic, compare Fujian with Guangdong. The latter has a smaller
rise in real wage in equilibrium, but its labor inflows rise twice as much.
This suggests that migration frictions differ significantly across regions
in the People’s Republic of China.
Column (3) presents changes in provincial real GDP, adjusted
for the local price index. Every region gains from tariff reductions,
but the level differs significantly across regions. The most positively
affected provinces are those that were the most developed before the
introduction of tariff reductions, implying that trade liberalization
has exacerbated regional inequality in the People’s Republic of China.
Column (4) presents changes in the local consumption price index.
Beijing, Tianjin, and Shanghai experienced the largest price decrease,
suggesting that they are the top beneficiaries of cheaper foreign goods.
Columns (5) and (6) present the total changes in exports and imports,
and show that both increased in all provinces, with some provinces
experiencing a larger increase in total exports than imports. There are
104 Trade Adjustment in Asia: Past Experiences and Lessons Learned

two main economic forces behind these changes in trade flows. The first
is related to industry composition. When sectors with limited regional
importance experience substantial tariff cuts, limited import competition
is introduced but a broad range of sectors may benefit. This boosts local
exports more than imports. The other subtler force works through trade
diversion. Cheaper intermediates directly lower production costs in
all regions in the People’s Republic of China. For a Chinese province,
it therefore becomes optimal to source more intermediates locally and
from other provinces in the country. This also suppresses growth in
imports from the rest of the world.
As suggested by the last column of Table 3.3, all regions (in terms of
people’s hukou status) gain from tariff reductions, but the distribution
of the gains is uneven. Individuals with a Beijing and Shanghai hukou
experience welfare improvements of 1.69% and 1.50%, respectively,
while individuals holding a hukou from Shanxi and Gansu provinces only
gain 0.37% and 0.31%, approximately 80% less. The hukou population-
weighted average welfare increase is 0.63%, with the standard deviation
being 0.27%.
Note that when labor is perfectly mobile, workers may choose to
move to different places due to idiosyncratic amenity draws, but the
welfare changes should be equal across individuals due to migration.
When labor is perfectly immobile, the labor reallocation term equals 0,

Figure 3.3 Individual and Regional Gains from Trade

0.020

Beijing
Individual gains from trade, with hukou

0.015 Shanghai
Tianjin

Jiangsu
0.010 Fujian

0.005 Ningxia
Henan Qinghai
Shanxi
Gansu

0.000
0.000 0.005 0.010 0.015 0.020
Regional gains from trade

Notes: This figure plots individual welfare changes in terms of hukou provinces (individual gains
from trade) against the changes in provincial real income per capita, i.e., regional gains from trade).
The lighter line is the linear fit and the darker is the 45-degree line. Correlation: 1.00; regression
coefficient: 0.97; t: 178.04; R-squared: 0.999.
Source: Author’s calculation.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 105

and hence individual gains from trade equal the real income increase of
the individual’s hukou province. To explore the extent to which internal
migration has alleviated the uneven welfare gains, Figure 3.3 plots
individual welfare changes in terms of their hukou (individual gains
from trade) against the changes in provincial real income per capita
(regional gains from trade). The relationship is strikingly linear with
the data points lying around the 45-degree reference line. This suggests
that the redistribution of wealth via migration is limited: although we
can see large changes in real income, most of the gains in booming areas
accrue to local hukou holders due to the high costs of migration.

3.5 Counterfactual Hukou Abolishment

The empirical exercise in section 3.3 suggests that the People’s Republic
of China’s internal labor reallocation over 2000–2010 can be explained
in large part by its integration into the global economy. Moreover, the
labor market distortion caused by the hukou system has a significant
impact in shaping the effects of trade on migration. A question that
naturally follows is, what would happen if the People’s Republic of China
abolished the hukou system? In particular, what is its direct impact on
wages, employment, trade, and GDP growth in different regions? What
are its distributional consequences? Would the country have reaped
more gains from trade reforms if the hukou system had been abolished
before its accession to the WTO? These are intriguing questions not
only of academic interest but also relevant for today’s policy debate.
For example, reforming this system appears high on the agenda of the
government today, but the exact road map is much less clear. Exploring
the counterfactual result of hukou abolishment could provide valuable
guidance regarding the ongoing hukou reforms.

3.5.1 Effects of Hukou Abolishment

Table 3.4 presents the regional adjustments that would have taken
place following the abolishment of the hukou system. I report the five
provinces that would experience the most significant expansions or
contractions. Beijing, Shanghai, and Guangdong are the top migrant-
receiving provinces, with an employment increase of more than 10%.
Jiangxi, Sichuan, Anhui, Hunan, and Guangxi Zhuang Autonomous
Region have the largest migrant outflows. The large migrant outflows in
Guangxi and Jiangxi are (among other factors) due to their geographic
proximity to Guangdong, while for Anhui these outflows are due to
its proximity to Shanghai. In the case of Hunan and Sichuan, locals
may face fewer migration frictions for other reasons, such as their
106 Trade Adjustment in Asia: Past Experiences and Lessons Learned

strong historical ties with Guangdong. This is also reflected in the fact
that their regional employment reacts strongly to tariff reductions
(Table 3.3). In expanding provinces, increased labor supply lowers real
wages and boosts local GDP; given the increased economic size, more
intermediates can now be sourced locally with a cost advantage, hence
the local consumption price decreases.
There are two forces that govern changes in trade flows. A province
experiencing expansion requires more intermediate inputs, which
implies an increase in both exports and imports; at the same time,
increased economic size also means the region gains a cost advantage
in producing a wider range of intermediates, suggesting an increase in
exports and a decrease in imports. These two forces work in the opposite
direction in contracting provinces. Therefore, exports should always rise
while the changes in imports are ambiguous in provinces with worker
inflows, and the opposite is true in provinces with worker outflows. The
calibration exercise shows that imports in all top expanding provinces
decrease, suggesting that the latter force prevails. On the other hand,
imports increase in some contracting provinces but decrease in others.
In the last column of Table 3.4, I present the individual welfare
changes. Although increased regional employment hurts local hukou
holders by bidding up structure rents and lowering wages, relaxations in
the system makes it easier for individuals to move to provinces where they
have higher amenity draws, which always improves welfare. Therefore,
while individuals holding a hukou from provinces with worker outflows
benefit from hukou reforms unambiguously, those with a hukou from
migrant-receiving provinces may not necessarily lose. As shown in the
last column of Table 3.4, the top expanding provinces’ hukou holders do

Table 3.4 Regional Effects of Hukou Abolishment

(1) (2) (3) (4) (5) (6) (7)

Province or
Autonomous Real
Region Employment Wage GDP Price Exports Imports Welfare
Beijing 18.99 –5.68 11.75 –1.43 17.47 –2.57 –5.93
Shanghai 18.33 –5.45 11.47 –1.31 13.80 –2.89 –5.52
Guangdong 13.20 –0.37 8.90 –1.23 10.46 –2.55 –3.62
Tianjin 5.26 –1.48 3.62 –0.46 3.54 –1.17 –1.24
Xinjiang 4.69 –0.73 3.91 –0.57 4.61 –1.25 –0.46
Uygur
Hainan 4.36 –0.77 3.44 –0.33 3.22 –0.97 –0.27
continued on next page
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 107

Table 3.4 continued

(1) (2) (3) (4) (5) (6) (7)
Province or
Autonomous Real
Region Employment Wage GDP Price Exports Imports Welfare
Zhejiang 3.42 –0.91 2.46 –0.39 1.97 –1.03 0.27
Fujian 2.85 –0.64 2.18 –0.48 1.16 –0.96 0.23
Qinghai 2.09 –0.34 1.77 –0.44 3.00 –1.27 0.21
Yunnan 1.26 0.25 1.47 –0.21 1.10 –0.82 0.46
Inner 0.60 0.39 0.94 –0.07 0.26 –0.62 0.98
Mongolia
Liaoning 0.53 0.01 0.54 –0.19 0.52 –0.47 0.31
Shanxi 0.48 0.40 0.83 –0.24 1.04 –1.08 0.66
Jiangsu 0.08 0.04 0.12 –0.20 –0.15 –0.56 0.94
Ningxia Hui 0.02 0.39 0.36 –0.14 –0.03 –0.70 0.88
Shandong –0.46 0.39 –0.09 –0.11 –1.35 –0.47 0.78
Jilin –0.59 0.71 0.07 –0.07 –0.49 –0.54 1.32
Hebei –0.81 0.78 –0.09 –0.03 –2.51 –0.82 1.39
Shaanxi –1.10 0.67 –0.48 0.02 –0.99 –0.24 1.32
Heilongjiang –1.26 0.79 –0.48 –0.05 –0.87 –0.38 1.69
Gansu –1.39 0.84 –0.60 –0.03 –1.36 –0.48 1.49
Chongqing –1.83 1.37 –0.59 0.06 –2.91 –0.90 2.26
Henan –2.44 1.03 –1.48 0.08 –3.31 –0.25 2.04
Guizhou –2.67 1.41 –1.36 0.08 –3.21 –0.51 2.73
Hubei –3.12 0.99 –2.20 0.26 –4.20 0.05 2.44
Guangxi –3.46 1.66 –1.91 0.10 –4.80 –0.88 3.34
Zhuang
Hunan –4.73 1.82 –3.10 0.60 –8.54 0.32 3.76
Anhui –5.06 2.38 –2.90 0.46 –6.17 –0.21 4.40
Sichuan –5.37 2.02 –3.60 0.82 –8.10 0.72 4.21
Jiangxi –6.17 2.08 –4.30 0.63 –8.65 0.05 4.66
Weighted 1.51
average
Standard 2.22
deviation
GDP = gross domestic product.
Notes: Counterfactual percentage changes in regional employment, real wage, real GDP (total value added
divided by local consumption price index), consumption price index, exports, imports, and hukou population’s
welfare when hukou frictions are reduced to 0 in all provinces, holding tariffs constant. The nominal wage of the
constructed rest of the world is the numeraire.
Source: Author’s calculation.
108 Trade Adjustment in Asia: Past Experiences and Lessons Learned

experience significant welfare losses. However, of the 17 provinces that

experience employment increases, their hukou holders’ welfare only
decreases in six. The average gains across provinces is 1.56%, which is
twice as high as the gains from trade reforms.

3.5.2 Effects of Tariff Reductions Given the Elimination

of Hukou Frictions

I next explore whether the impact of trade liberalization would be different

if the hukou system had been abolished before the People’s Republic
of China’s accession to the WTO. Starting from the postabolishment
equilibrium, I repeat the first quantitative exercise by shocking the
system with tariff changes. Table 3.5 presents the regional effects for the
five provinces with the biggest and smallest increases in employment.
Comparing these results with those in Table 3.3, we observe that regional
employment reacts more strongly to trade shocks with the elimination
of hukou frictions, while real wages react less. For instance, the change
in Beijing employment increases by more than 50%, while the change
in its real wage declines by 6%. The absolute changes, however, remain
small. One plausible explanation regards data aggregation: calibrating
the initial labor distribution at the province level overestimates the initial
migration frictions, therefore abolishing the hukou system seems to have
only a marginal effect in shaping the impact of trade as the model suggests
very high migration frictions in levels even after abolishing the hukou.

Table 3.5 Regional Adjustments to Tariff Reductions,

without Hukou Frictions

(1) (2) (3) (4) (5) (6) (7)

Province or
Autonomous Real
Region Employment Wage GDP Price Exports Imports Welfare
Beijing 0.84 1.68 2.44 –2.07 9.97 3.36 1.57
Shanghai 0.58 1.51 2.01 –1.85 6.39 4.00 1.40
Tianjin 0.27 1.39 1.62 –1.96 7.65 4.70 1.33
Guangdong 0.25 0.92 1.15 –1.76 5.12 7.01 0.89
Fujian 0.12 0.95 1.07 –1.68 6.50 7.15 0.94
Zhejiang 0.09 0.87 0.94 –1.59 5.09 5.71 0.86
Jiangsu 0.07 1.13 1.19 –1.79 9.36 6.90 1.11
Liaoning 0.03 0.96 0.98 –1.75 8.11 6.54 0.94

continued on next page

 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 109

Table 3.5 continued

(1) (2) (3) (4) (5) (6) (7)
Province or
Autonomous Real
Region Employment Wage GDP Price Exports Imports Welfare
Xinjiang Uygur 0.03 0.66 0.67 –1.52 6.39 5.32 0.64
Hainan 0.02 0.73 0.77 –1.65 2.76 6.48 0.75
Jilin –0.01 0.73 0.72 –1.73 6.55 5.11 0.73
Shandong –0.02 0.70 0.67 –1.70 5.82 8.79 0.70
Yunnan –0.02 0.49 0.48 –1.48 7.40 5.20 0.50
Ningxia Hui –0.03 0.41 0.37 –1.50 9.09 4.04 0.41
Inner –0.03 0.53 0.49 –1.53 4.60 6.83 0.53
Mongolia
Shaanxi –0.04 0.46 0.42 –1.59 4.99 5.62 0.47
Shanxi –0.04 0.37 0.33 –1.48 6.99 4.37 0.38
Qinghai –0.04 0.41 0.36 –1.50 6.69 5.20 0.41
Heilongjiang –0.04 0.59 0.54 –1.65 3.94 7.04 0.60
Gansu –0.05 0.31 0.26 –1.52 5.55 3.86 0.32
Chongqing –0.05 0.49 0.44 –1.53 2.59 4.37 0.50
Guangxi –0.08 0.56 0.49 –1.49 6.44 4.90 0.59
Zhuang
Henan –0.08 0.40 0.31 –1.55 5.04 5.30 0.42
Hubei –0.08 0.49 0.41 –1.58 5.05 5.17 0.52
Hebei –0.08 0.42 0.33 –1.61 4.98 7.98 0.44
Guizhou –0.09 0.42 0.32 –1.52 4.63 4.56 0.44
Hunan –0.11 0.48 0.37 –1.52 7.27 5.16 0.52
Sichuan –0.12 0.49 0.36 –1.57 4.16 4.87 0.53
Anhui –0.17 0.54 0.36 –1.52 4.14 5.23 0.60
Jiangxi –0.19 0.42 0.23 –1.45 3.68 5.41 0.49
Weighted 0.64
average
Standard 0.25
deviation
GDP = gross domestic product.
Notes: This table presents the counterfactual percentage changes in regional employment, real wage, real GDP
(total value added divided by local consumption price index), consumption price index, exports, and imports
when the tariff structure of the People’s Republic of China changed from its 2000 to 2005 level after eliminating
hukou frictions, holding tariffs constant. The nominal wage of the constructed rest of the world is the numeraire.
Source: Author’s calculation.
110 Trade Adjustment in Asia: Past Experiences and Lessons Learned

The final column of Table 3.5 presents the changes in welfare of

hukou holders of a given province. Comparing these results to those in
Table 3.3, we can see that the top five beneficiaries remain hukou holders
from Beijing, Shanghai, Tianjin, Jiangsu, and Fujian. However, they gain
less due to the larger increase in migrant inflows. On the other hand,
provinces with net migration outflows also experience an increase in
regional employment response to trade shocks, and this is associated
with larger welfare improvements.
The last two rows of column (7) of Table 3.5 report the weighted
average and the standard deviation of welfare increases. Average gains
from trade increase by about 2%, from 0.71% in the case with hukou
frictions to 0.72%. Compared with Monte, Redding, and Rossi-Hansberg
(2015), who demonstrate that by permitting commuting across counties
in the United States, gains improve from a 20% reduction in domestic
trade costs by 0.8%, the additional gains from trade due to hukou friction
elimination are considerable. In addition, calibrating the model at the
prefecture city level would likely produce larger estimates. Therefore,
we can interpret the 2% gains from trade due to hukou abolishment as a
lower bound estimate.
The standard deviation of welfare gains across worker types falls
from 0.24% to 0.22%. Freer migration leads to greater employment

Figure 3.4 Individual Gains from Trade with

and without Hukou Frictions
0.020
Individual gains from trade, with hukou

Beijing
0.015
Shanghai
Tianjin
Jiangsu
0.010 Fujian

0.005 Ningxia
Henan Qinghai
Shanxi
Gansu

0.000
0.000 0.005 0.010 0.015 0.020
Regional gains from trade

Notes: This figure plots individuals’ welfare changes from tariff reductions in terms of hukou provinces
(individual gains from trade) with hukou abolishment against the changes without. The lighter line is
the linear fit and the darker line is the 45-degree line. Correlation: 0.999 Regression coefficient: 0.9;
t: 101.55; R-squared: 0.997.
Source: Author’s calculation.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 111

increases in more positively affected regions, and the opposite in less

positively affected regions. This narrows the spatial wage gap, meaning
that individuals who stay in contracting regions are less negatively
affected. In addition, freer migration makes individuals migrate to
booming areas to improve their welfare. Both effects lead to more
evenly distributed gains. Figure 3.4 plots individual gains from tariff
reductions without hukou frictions to those with; the plot is flatter than
the 45-degree line, suggesting that the elimination of hukou frictions
alleviates the distributional effects of trade.

3.5.3 The Role of Internal Geography:

General Assessment

I now investigate the importance of accounting for internal geography

in computing gains from trade. I compare the results of my model
(benchmark) to a multiregion model with no internal migration, as
well as to a two-country model treating the People’s Republic of China
as a unit of analysis. I calibrate each of these models to the year 2000
and compute the welfare response and its decomposition to the tariff
reductions from the country’s WTO accession. Table 3.6 presents the
simulated welfare effects implied by different models. The first row
presents the benchmark results, the second row the welfare results for
the no-migration model, and the third row the results for the model
treating the People’s Republic of China as a whole.
Table 3.6 suggests that the welfare effects are smaller for the model
without migration compared to the benchmark model. The average gain
decreases from 0.63% to 0.61%. It is also distributed more unevenly, with
the standard error rising from 0.27% to 0.31%. Allowing interregional
migration diminishes both effects.
When treating the People’s Republic of China as a unit of analysis,
the welfare increases substantially to 0.80%, almost 30% more than the

Table 3.6 Gains from Tariff Reductions: Role of Internal Geography

Average Standard Deviation

Model (%) (%)
Benchmark 0.63 0.27
No migration 0.61 0.31
People’s Republic of China as a whole 0.80 0.00
Notes: Average of counterfactual percentage changes in welfare in terms of individual hukou provinces and
its standard deviation. In models with many regions in the country, both the average and standard deviation
of welfare changes are hukou population weighted.
Source: Author’s calculation.
112 Trade Adjustment in Asia: Past Experiences and Lessons Learned

benchmark model. The results reflect the importance of accounting for

domestic geography. The intuition for this result is related to that of the
no-migration model. By treating the People’s Republic of China as a unit
of analysis, I implicitly assume that both goods and factors are perfectly
mobile within the country, and thus I get much larger gains from trade.
In short, the results from this subsection illustrate the importance of
accounting for economic geography within a country to evaluate the
effect of trade policies, especially where the country has significant
spatial heterogeneities.

3.6 Conclusion
Trade liberalization can lead to significant spatial labor adjustment
within a country, and internal migration frictions are important in
shaping the impact of trade. In the context of the People’s Republic of
China, input-liberalization has stimulated significant labor reallocation
across prefectures, as well as the presence of migration frictions caused
by the hukou system. Quantitatively, tariff reductions improve the People’s
Republic of China’s aggregate welfare by 0.71%, but magnify regional
disparities. Abolishing the hukou system leads to a sizable improvement
in aggregate welfare, but it also has a strong distributional impact. In
addition, it increases the gains from trade and alleviates its negative
distributional consequences. These results shed light on the benefits
of eliminating migration frictions and the importance of taking these
frictions into account when evaluating both aggregate and distributional
consequences of trade reforms.
While the focus of this chapter has been on the People’s Republic
of China, the existing literature suggests that migration frictions are
pervasive in many other countries. According to the World Population
Policies 2013 (United Nations 2013), 60% of governments around the
world desired a major change in their spatial labor distribution, and
80% of these had policies to influence internal migration. Therefore,
this chapter’s exercises may inform migration policy and stimulate
further studies in other national contexts.
 Trade Liberalization and the Hukou System of the People’s Republic of China:
How Migration Frictions Can Amplify the Unequal Gains from Trade 113

References
Autor, D. H., D. Dorn, and G. H. Hanson. 2013. The China Syndrome:
Local Labor Market Effects of Import Competition in the United
States. The American Economic Review. 103 (6). pp. 2121–68.
____. 2016. The China Shock: Learning from Labor-Market Adjustment to
Large Changes in Trade. Annual Review of Economics. 8 (1). pp. 205–40.
Balsvik, R., S. Jensen, and K. G. Salvanes. 2015. Made in China, Sold in
Norway: Local Labor Market Effects of an Import Shock. Journal of
Public Economics. 127. pp. 137–44.
Blancher, N.R., and R. Rumbaugh. 2004. China: International Trade and
WTO Accession. IMF Working Paper. No. WP/04/36. Washington,
DC: International Monetary Fund.
Bosker, M., S. Brakman, H. Garretsen, and M. Schramm. 2012. Relaxing
Hukou: Increased Labor Mobility and China’s Economic Geography.
Journal of Urban Economics. 72 (2). pp. 252–66.
Bugamelli, M., S. Fabiani, and E. Sette. 2015. The Age of the Dragon:
The Effect of Imports from China on Firm-Level Prices. Journal of
Money, Credit and Banking. 47 (6). pp. 1091–118.
Chan, K. W. 2009. The Chinese Hukou System at 50. Eurasian Geography
and Economics. 50 (2). pp. 197–221.
____. 2013. China: Internal Migration. In I. Ness, ed. The Encyclopedia
of Global Human Migration. Malden, MA: Blackwell Publishing.
doi:10.1002/9781444351071.wbeghm124
Chan, K. W., and W. Buckingham. 2008. Is China Abolishing the Hukou
System? The China Quarterly. 195. pp. 582–606.
Cheng, M. 2007. How Do We Lose the Freedom of Movement: Formation
of China’s Household Registration System in the 1950s and Its
Evolution. Modern China Studies. 4. pp. 6.
Cheng, T., and M. Selden. 1994. The Origins and Social Consequences of
China’s Hukou System. The China Quarterly. 139. pp. 644–68.
Dix-Carneiro, R., and B. K. Kovak. 2017. Trade Liberalization and Regional
Dynamics. American Economic Review. 107 (10). pp. 2908–46.
Giovanni, J. D., A. A. Levchenko, and J. Zhang. 2014. The Global Welfare
Impact of China: Trade Integration and Technological Change.
American Economic Journal: Macroeconomics. 6 (3). pp. 153–83.
Goldberg, P. K., and N. Pavcnik. 2005. Trade, Wages, and the Political
Economy of Trade Protection: Evidence from the Colombian Trade
Reforms. Journal of International Economics. 66 (1). pp. 75–105.
Grossman, G. M., and E. Helpman. 1994. Protection for Sale. The
American Economic Review. 84 (4). pp. 833–50.
Hsieh, C.-T., and R. Ossa. 2011. A Global View of Productivity Growth in
China. NBER Working Paper. No. 16778. Cambridge, MA: National
Bureau of Economic Research. doi:10.3386/w16778
114 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Ianchovichina, E., and W. Martin. 2001. Trade Liberalization in China’s

Accession to WTO. Journal of Economic Integration. 16 (4). pp. 421–45.
http://www.jstor.org/stable/23000767
Kinnan, C., S.-Y. Wang, and Y. Wang. 2015. Relaxing Migration Constraints
for Rural Households. NBER Working Paper. No. 21314. Cambridge,
MA: National Bureau of Economic Research. doi:10.3386/w21314
Kovak, B. K. 2013. Regional Effects of Trade Reform: What Is the Correct
Measure of Liberalization? The American Economic Review. 103 (5).
pp. 1960–76.
Lardy, N.R. 1991. Foreign Trade and Economic Reform in China, 1978–1990.
Cambridge, UK, and New York, NY: Cambridge University Press.
____. 2005. Trade Liberalization and Its Role in Chinese Economic
Growth. In W. Tseng and D. Cowen, eds. India’s and China’s Recent
Experience with Reform and Growth. Procyclicality of Financial
Systems in Asia. London: Palgrave Macmillan. pp. 158–69.
Li, X. 2013. China as a Trading Superpower. Mimeo, London School of
Economics and Political Science, 1–7.
Monte, F., S. J. Redding, and E. Rossi-Hansberg. 2015. Commuting,
Migration and Local Employment Elasticities. NBER Working
Paper. No. 21706. Cambridge, MA: National Bureau of Economic
Research. doi:10.3386/w21706
Park, A., Y. Wu, and Y. Du. 2012. Informal Employment in Urban China.
Technical Report. Washington, DC: World Bank. http://documents
.worldbank.org/curated/en/383331468220789675/pdf/777370WP0
Infor00Box377299B00PUBLIC0.pdf
Topalova, P. 2010. Factor Immobility and Regional Impacts of Trade
Liberalization: Evidence on Poverty from India. American Economic
Journal: Applied Economics. 2 (4). pp. 1–41.
United Nations. 2013. World Population Policies 2013. Technical
Report. New York, NY: United Nations. http://www.un.org/en
/development/desa/population/publications/pdf/policy
/WPP2013/wpp2013.pdf
Wang, F. 2004. Reformed Migration Control and New Targeted
People: China’s Hukou System in the 2000s. The China Quarterly.
177. pp. 115–32.
Woo, W. T. 2001. Recent Claims of China’s Economic Exceptionalism:
Reflections Inspired by WTO Accession. China Economic Review.
12. pp. 107–36.
Xue, L., and H. Zhang. 2001. Shaping Up Sustainable Urban Transport
Authorities in China: Status Quo Assessment and Reforming
Solutions. Working Paper, World Resources Institute.
Zi, Y. 2018. Trade Liberalization and the Great Labor Reallocation.
Mimeo, University of Oslo.
 4
The Impact
of Tariff Liberalization
on the Labor Share in India’s
Manufacturing Industry
Prachi Gupta and Matthias Helble

4.1 Introduction
India opened its domestic market to international trade in the early
1990s. The decision had profound implications on the development
trajectory of the country. Trade opening allowed the country’s firms
to source inputs more cheaply and increase their competitiveness. At
the same time, it meant that firms in India were exposed to increased
competition from abroad, while consumers enjoyed lower prices for
final goods. While the trade opening helped boost economic growth,
it required an adjustment of capital and labor within and across
sectors.
In this chapter, we aim to uncover how the labor share in the
manufacturing sector changed after the trade opening. We use highly
disaggregated plant-level data covering the period 1998–2008 to study
the overall effect, as well as the effect across different manufacturing
sectors. We find that a decline in output tariffs increased the labor share
in manufacturing, whereas a fall in input tariffs led to a decrease in the
labor share. This general result becomes more nuanced once we take
into account the factor intensity of sectors. In technology-intensive and
human capital resource-intensive sectors, both a decline in input and
output tariff rates led to a decline in labor share. Separating the sample
into the Indian states with flexible and inflexible labor laws, our results
suggests that Indian plants subject to inflexible labor laws adjusted to
trade opening by systematically replacing labor with capital. Our study
significantly adds to the existing literature by studying the effect of both

115
116 Trade Adjustment in Asia: Past Experiences and Lessons Learned

output and input tariff liberalization. Furthermore, ours is the first study
to control for different levels of technology across sectors.

4.2 Background
Until the early 1990s, the Government of India pursued a policy of
import substitution. The underlying idea was that a protectionist trade
regime would insulate the domestic industry from the rigors of market
competition. As imports would be very expensive, domestic production
and industrialization would be promoted, and Indian firms would
sooner or later become competitive, including internationally. The
public sector was given a central role in running the economy. However,
in contrast to the economic objective of rapid industrialization with
equitable growth, this policy regime created an economy of complacent
rentiers rather than competitive entrepreneurs, and it stifled the growth
of the industry sector. In addition, government expenditures spiraled
and made the Indian economy vulnerable to external shocks.
When external economic shocks did hit the country’s economy in
1991, the Indian government approached the International Monetary
Fund and the World Bank for support. The two international
organizations made their assistance conditional on the implementation
of profound structural reforms. This created a milieu for the political
will to undertake large-scale economic reforms, which had already
been discussed in economic circles in India in the 1980s. Starting
in 1991, the government started a comprehensive liberalization
strategy that gradually dismantled the protective shield for domestic
producers. One important pillar of the new strategy was the opening
up of India’s highly protected trade regime. The government first

Table 4.1 Economic Growth and Openness in India, 1960–2010 (%)

GDP per Trade % Manufacturing

Time Period Capita Growth of GDP Output Growth
1960/61 to 1969/70 1.7 9.6 4.7
1970/71 to 1979/80 1.0 11.3 5.1
1980/81 to 1989/90 2.9 14.0 5.9
1990/91 to 1999/2000 3.6 21.2 7.0
2000/01 to 2009/10 5.0 39.3 8.5
GDP = gross domestic product.
Sources: World Bank Development Indicators (2019) and authors’ calculations using data from National
Income Accounts, India.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 117

started by opening its market for raw materials, intermediate goods,

and capital goods. The market for consumer goods remained highly
protected throughout the 1990s.
Table 4.1 compares the economic growth rate of India and the share
of trade in gross domestic product (GDP) from 1960 to 2010. While the
3 decades from 1960 to 1990 were marked by low growth and low trade-
to-GDP ratios, the latter 2 decades saw a sharp increase in both. Similar
to overall growth, the growth rate in manufacturing was boosted after
the trade opening.
Figure 4.1 illustrates how the average and weighted tariff rates for
industrial products were drastically reduced, respectively, from 81.9%
and 49.5% in 1990 to 57.4% and 27.8% in 1992. The process of tariff
rationalization continued over the next 2 decades. By 1999, the average
and weighted tariff rates stood at 33.0% and 28.6%, respectively, and the
peak tariff was reduced to 35.1%. The effective rate of protection was
reduced from 125.9% in 1986–1990 to 80.2% in 1990–1995 and further to
40.4% in 1996–2000 (Das 2003).
While in the 1990s, during the first decade of reforms, tariff rates
declined substantially, some experts argued that the level was still
sufficiently high to substantially discourage imports (Goldar 2012).
Furthermore, imports of most products were still subject to a range of
nontariff measures. Das (2003) estimated the nominal and effective rates
of protection and found that they remained high until 2000, reaching
almost 100% for intermediate and final goods.

Figure 4.1 Average Tariff (Weighted Average)

for Industrial Products, 1990–2008 (%)
100 200
90 180
80 160
Export and Import (₹ million)

70 140
60 120
50 100
40 80
30 60
20 40
10 20
0 0
90
91
92
93
94
95
96
97
98
99

20 0
20 1
20 2
03

20 4
20 5
20 6
20 7
08
0

0
0

0
0
0
19

19

19

19
19

19
19

19
19
19

20
20

Simple Average Tariff Weighted Average Tariff

Tariff Peak Rate (right vertical axis)

Source: Authors’ calculation based on R. Banga and A. Das, eds. 2012. Twenty Years of India’s
Liberalization: Experiences and Lessons. Geneva: United Nations Conference on Trade and
Development. http://unctad.org/en/PublicationsLibrary/osg2012d1_en.pdf. p. 8.
118 Trade Adjustment in Asia: Past Experiences and Lessons Learned

It was only in 2000, after India lost a World Trade Organization

dispute against the United States on quantitative restrictions, when
the Indian government decided to remove a large number of nontariff
restrictions (Banga and Das 2012). In the 2000s, the long-protected
consumer sector was opened to trade, which led to a sharp rise in
India’s imports of final goods, leading in turn to a significant increase
in competition for domestic producers. At the same time, the fall in the
price of intermediate inputs allowed Indian manufacturers to boost
their productivity. Figure 4.2 shows the fall in tariffs for the 10 most
important industries in India. The decline was most dramatic in the
apparel sector, which employed roughly a quarter of the workers in the
manufacturing sector.
The trade opening also contributed to a change of the import
and export basket of India. Table 4.2 lists the share of each industry
(following the National Industrial Classification of India, or NIC) in
total manufactured imports and exports. In textiles, wearing apparel,
and leather (NIC17–19) the export shares fell from close to 40% in
1990 to 13.7% in 2010. All three industries are labor intensive, and they
collectively employed about one quarter of total workers in 2010. At
the same time, we notice a sharp rise in the export of coke and refined
petroleum products, a very capital-intensive industry. In terms of
imports, India saw a sharp increase in furniture imports. Depending
on the technological sophistication, the furniture industry can be labor
intensive.

Figure 4.2 Tariff Rates by Industry, 2000 and 2008 (%)

45
40
35
30
25
20
15
10
5
0
uf cts

nd d p cts

pr ts

fix s

ro s
s

pr ts

ts
re

re

ct

t
c

uc

uc

uc

uc
u

Fu woo odu

du
l a le m ctu

tu
od

od

od

g a all rod

od

od
a
pr

pr

pr
p

p
bl r an and
d

ill

em an ed

an ied

d
an
re

te

te
m

li

i
e

l
nd

be ela

la
al

al
r
Pa nitu
o

re
d

d
ki

i
cc

r
xt
nd

nd
d

Te
ba

ra

r
an

tro Ch ing

re als
To

p e
od

re

ic

in
sh
m
pa

fin
Fo

i
Lu
Ap

pu

um
g,
tin

le
in
Pr

Pe

2000 2008

Source: Author’s calculation based on R. Banga and A. Das, eds. 2012. Twenty Years of India’s
Liberalization: Experiences and Lessons. Geneva: United Nations Conference on Trade and
Development. http://unctad.org/en/PublicationsLibrary/osg2012d1_en.pdf. p. 10.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 119

Table 4.2 Share in India’s Manufacturing Export

and Import by NIC-2 Digit Classification (%)
Imports Exports
NIC Industry 1990 2000 2010 1990 2000 2010
NIC 15: Food Products 2.8 6.1 4.1 10.1 9.3 6.0
and Beverages
NIC 16: Tobacco 0.0 0.0 0.0 0.3 0.1 0.1
NIC 17: Textiles 1.7 2.1 1.3 15.7 16.0 7.3
NIC 18: Wearing 0.0 0.1 0.1 16 13.7 4.7
Apparel
NIC 19: Leather 0.6 0.6 0.4 7.8 3.8 1.7
NIC 20: Wood 0.1 0.1 0.2 0.1 0.1 0.1
NIC 21: Paper 3.0 2.1 1.2 0.1 0.4 0.4
NIC 22: Publishing 0.5 1.9 0.3 0.3 1.0 0.2
NIC 23: Coke, Refined 19.0 8.1 4.7 3.5 3.6 18.9
Petroleum Products
NIC 24: Chemicals 23.8 20.7 17.3 9.1 11.6 12.5
NIC 25: Rubber 0.9 1.3 1.3 1.4 1.6 1.7
NIC 26: Non-metallic 0.8 0.7 0.8 0.6 1.5 1.0
Mineral Products
NIC 27: Basic Metals 12.6 21.4 25.3 3.0 4.5 9.8
NIC 28: Fabricated 1.2 1.5 1.6 2.7 3.2 2.2
Metal Products
NIC 29: Machinery 13.5 9.8 10.3 3.2 2.5 3.7
and Equipment
NIC 30: Office, 1.7 5.1 2.1 0.8 0.6 0.3
Accounting and
Computing Machinery
NIC 31: Electrical 3.1 3.2 3.3 1.2 1.5 2.2
Machinery
NIC 32: Radio, 3.5 4.7 7.6 0.6 0.6 2.0
Television and
Communication
Equipment
NIC 33: Medical, 4.1 3.5 2.6 0.4 0.7 0.8
Precision and Optical
Instruments
NIC 34: Motor 2.2 1.9 2.3 1.9 1.7 4.3
Vehicles
NIC 35: Transport 4.6 2.7 3.7 1.1 1.0 3.6
Equipment
NIC 36: Furniture 0.3 2.3 9.6 20.0 21.0 16.7
NIC = National Industrial Classification (2004).
Source: Authors’ calculation based on data from World Integrated Trade Solution.
120 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 4.3 Average Labor Share of Income, 1999–2008

Mean Standard Deviation

1999 0.126 0.130
2000 0.132 0.135
2001 0.124 0.131
2002 0.124 0.130
2003 0.124 0.129
2004 0.119 0.123
2005 0.108 0.113
2006 0.106 0.111
2007 0.100 0.106
2008 0.100 0.104
Source: Authors’ estimates.

In our study, we define the labor share as the total wage bill
(product of wages and number of employees) divided by total sales.
Table 4.3 shows a tendency of a falling labor shares during the period of
trade liberalization. This is a rather unexpected outcome with respect
to theory given the comparative advantage of India in labor-intensive
sectors.
The labor share can fall in two cases, holding all other variables
constant. First, when the total sales increase faster than the wage bill.
Second, in the case of a contraction—that is, the total sales fall more slowly
than the wage bill. To better understand the three variables involved in
the labor share, Table 4.4 gives an overview of how employment, the
wage rate (inflation adjusted), and sales revenue (inflation adjusted)
evolved.
In our data of large manufacturing plants, we find that the average
employment size fell by 1.1% annually during the period of our analysis.
Overall, the census sector witnessed a growth in employment of 4.9% per
year during the period. Given that firing workers in census plants is not
very easy due to stringent labor laws in India, the numbers in columns
(1) and (2) imply that the new entrant plants had a smaller labor force.
This is not surprising since there is substantial evidence in the literature
with respect to “jobless” growth and capital intensification in the case
of Indian manufacturing (Kannan and Raveendran 2009; Thomas 2013).
We also see that wages grew by 4.1%, while plant revenue grew at a much
higher rate of 7.0%. Studies that have analyzed wages and productivity in
the case of Indian manufacturing have found that capital intensification
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 121

Table 4.4 Average Employment and Wages/Sales Ratio

Average
Number of Total Number
Workers per of Workers Wage Rate Sales Revenue
Plant Employed (₹ per day) (₹ million)
Year (1) (2) (3) (4)
1999a 391.3 3,179,513 NA 4.28
2000a 357.9 3,100,965 131.7 4.43
2001 241.1 3,722,010 130.1 3.05
2002 233.7 3,601,330 137.7 3.06
2003 237.1 3,691,729 143.2 3.48
2004 202.3 3,882,585 145.3 3.20
2005 258.1 4,026,021 154.4 4.59
2006 233.2 4,348,107 157.7 4.37
2007 220.6 4,688,096 165.0 4.52
2008 214.0 5,124,493 176.5 4.51
Growthb –1.1% 4.9% 4.1% 7.0%
NA = not available.
Note: Wage rate and sales revenue numbers have been deflated using the wholesale price index with base
year 2004/05.
a The definition of the census sector was changed in 2001 to include plants with more than 100 workers,
which earlier was capped at 200 workers. Hence, the figures for 1999 and 2000 are not comparable with
the later years.
b We have excluded 1999 and 2000 for the purpose of growth estimation.
Source: Authors’ estimates.

in the Indian manufacturing sector has increased, though a large share of

the resultant labor productivity growth has been retained by employers
with little growth in the wages of workers (Kannan and Raveendran
2009). Overall, the numbers suggest that the main driver of the fall in
labor share was the smaller increase in the wage bill compared to sales.
The question that follows is what the impact of tariff liberalization has
been on labor share. In the analysis that follows, we try to answer this.

4.3 Literature Review

The case of India’s experience with trade opening has attracted scholarly
attention for many years. Development economists have attempted
to evaluate the impact of trade opening on various socioeconomic
outcomes in India.
122 Trade Adjustment in Asia: Past Experiences and Lessons Learned

With respect to the impact on poverty, the evidence remains mixed.

Using four rounds of household survey data (1983–1984, 1987–1988,
1993–1994, and 1999–2000), Topalova (2007) finds that districts that
were more exposed to tariff reductions experienced a lower reduction
in rural poverty. Contrary to her study, Cain, Hasan, and Mitra (2012),
also using household survey data, undertook a state-level analysis
with no evidence for worsened poverty due to the reductions in trade
protection. Instead, they find that states whose workers were on average
more exposed to foreign competition had lower rural, urban, and overall
poverty rates.
Another dimension covered was the impact on inequality. Using
Indian household expenditure survey data from 1988 to 2005, Krishna
and Sethupathy (2011) find that income inequality fell between 1988 and
1994, rose between 1994 and 2000, and fell again after 2000. They find
that the changes to the state-level measures for trade protection had no
significant impact on changes in inequality across households within
states. This is in line with the findings of Topalova (2007) that there is
no discernible effect of trade reforms on rural and urban inequality in
India.
More closely related to our study are the empirical papers that
evaluate the impact on firms. One strand of literature looks at the link
between trade and firm productivity. Sivadasan (2009) analyzes the
impact of tariff reduction and foreign direct investment reforms on cross-
sectional data of manufacturing plants during 1986–1987 to 1994–1995.
The study finds a positive impact of foreign direct investment and tariff
reforms on plant-level productivity toward the end of the observation
period. Topalova and Khandelwal (2011) use firm-level data for 1989–
2001 and analyze the impact of the reduction in output and input tariffs
on firm-level productivity growth. The study finds the reduction in both
tariffs to have had a significant impact on firm-level productivity growth,
and gains are found to be larger with respect to the fall in input tariffs.
Gupta (2016) uses plant-level data of Indian manufacturing for 1997–
1998 to 2007–2008 and analyzes the channels through which trade can
drive productivity growth. Her study finds that technology spillovers, in
addition to input and output variety growth, significantly contributed
to higher plant-level productivity in the case of India. In contrast to the
above, the studies by Balakrishnan, Pushpangadan, and Babu (2000) and
Bollard, Klenow, and Sharma (2013) find no significant role for trade
reforms in generating productivity gains in Indian manufacturing.
Our study builds on previous attempts to gauge the impact of trade
opening on employment and labor shares. Hasan et al. (2012) find urban
unemployment to have declined after trade opening in states with more
flexible labor markets and larger employment shares in net export
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 123

industries. Workers in industries experiencing greater reductions in

trade protection were less likely to become unemployed, especially in
net export industries. Using aggregate industry-level data, Hasan et
al. (2012) find evidence that trade liberalization had a positive impact
on labor demand elasticity. They also show that trade reforms led to
a reduction in labor share due to a decline in the bargaining power of
workers. However, using micro-level data, Ahsan and Mitra (2014)
uncover variation in terms of the impact of trade liberalization on labor
shares. For small and labor-intensive firms, the trade reforms seemingly
led to an increase in labor share, while the impact was found to be the
opposite for the larger and less labor-intensive firms. Mishra and Kumar
(2005) use household survey data and find that a negative relationship
forms over time between changes in trade policy and changes in industry
wage premiums. Since tariff reductions were proportionately larger in
sectors that employ a larger share of unskilled workers, their findings
imply that unskilled workers experienced an increase in their relative
incomes due to the trade reforms.
Traditional trade theory (the Heckscher–Ohlin model) predicts
that gains from trade favor abundant factors. In the case of a developing
country with abundant low-skilled labor like India, this would imply
that unskilled labor would benefit the most from globalization. However,
new theories such as Marjit, Beladi, and Chakrabarti (2004) suggest
that, even in a labor-abundant country, trade liberalization can reduce
the wages of unskilled labor, thereby widening the gap between the
rich and the poor. Moreover, such adjustments may be costly, with the
burden falling disproportionately on the poor (Banerjee and Newman
2004).

4.4 Data and Variables

4.4.1 Plant-level Data

The Indian government has relatively recently released plant-level data

for manufacturing firms. The data are collected through the Indian
Annual Survey of Industries (ASI). The ASI is the principal source of
industrial statistics in India and extends to almost the entire country.1

1
Many studies in the Indian context have earlier used industry-level and plant-
level cross-sections from the ASI (Hasan, Mitra, and Ramaswamy 2007; Hsieh and
Klenow 2009; Sivadasan 2009; Harrison, Martin, and Nataraj 2012; Bollard, Klenow,
and Sharma 2013), but the earlier data of the ASI did not disclose plant identifiers.
This study uses the recently released data with plant identifiers launched by the ASI
which enable us to create a panel of manufacturing plants in India.
124 Trade Adjustment in Asia: Past Experiences and Lessons Learned

It covers all factories that are registered under sections 2(m) (i) and 2(m)
(ii) of the Factories Act of 1948. This implies that all factories employing
10 or more workers using power and those employing 20 or more
workers without using power are surveyed under the ASI. The primary
unit of enumeration is a plant or factory in the case of the manufacturing
industry. The plant-level panel data of the ASI included in our sample
cover the period between 1998–1999 and 2007–2008.
The ASI data are collected annually by the Field Operations
Division of the National Sample Survey Office in consultation with the
Chief Inspector of Factories in the states. Under the ASI framework,
factories are classified into two sectors: Census and Sample. In the
“Census” sector, the data from all the factories employing 100 (or 200 up
to the accounting year 2002–2003) workers are collected on a complete
enumeration basis. The remaining factories fall under the “Sample”
sector, for which data are collected by drawing a representative sample
using sampling techniques. Since continuous data are only available for
this set, our study covers only those plants that fall under the “Census”
sector of the ASI and can be successfully analyzed in a panel form.
The data are an unbalanced panel and contain detailed information
on production-related factors such as output, fixed assets, inventories,
working capital, inputs, employment, labor costs, raw materials,
electricity, power and fuel consumption, location, ownership, year of
incorporation, and so on.
As per the National Industrial Classification (NIC), factories
are classified into industry categories up to the four-digit level of
disaggregation in the ASI.2 In this study, we only focus on those plants
that operate in the manufacturing sector (i.e., belong to the NIC15 to
NIC36 two-digit industry groups). Table 4.5 provides the summary
statistics of the key variables included in the regressions.

Table 4.5 Summary Statistics

Mean Standard Deviation

Age (years) 22.02 22.06
No. of workers 313.93 851.17
Capital stock (₹ million) 544.14 4,172.80
Output (₹ million) 1,146.53 9,619.45
Source: Authors’ estimates.

2
NIC classification of Indian industries closely corresponds to the International
Standard Industrial Classification of All Economic Activities classification.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 125

4.4.2 Tariff Data

In order to gauge the impact of trade opening on the labor share, we

use the applied tariff as a proxy. Nontariff measures are often similarly
important, but we only take into account tariffs due to data constraints.
We also do not consider any preferential tariffs, as by the early 2000s,
India had not yet signed preferential trade agreements with significant
trading partners.
Trade liberalization affects firms mainly in two ways. First, as tariffs
are lowered, the firms typically face more foreign competition for their
final goods. Second, lower tariffs allow them to source their inputs at a
lower price. As these two channels might affect labor shares differently,
we introduce them as separate variables.
Output tariffs. To measure the trade openness of a given industry,
we use the most disaggregate output tariff data possible. In our study,
this corresponds to the four-digit level data as per the International
Standard Industrial Classification of All Economic Activities, or ISIC
Rev 3. Since the ASI categorizes factories in this panel data per the NIC
1998 classification, we construct a concordance table between the four-
digit ISIC Rev 3 and four-digit NIC 1998 classifications in order to match
the tariff data with the ASI panel. The primary data series of tariffs
was obtained from the Trade Analysis Information System (TRAINS)
database of the United Nations Conference on Trade and Development.
However, tariff data were not available in TRAINS for some of the years,
for which instances we used data obtained from the Integrated Data
Base of the World Trade Organization.
Input tariffs. In addition to output tariffs, which measure
competition faced by domestic producers from the import of final
goods, trade liberalization also facilitates access to cheaper and
advanced intermediate inputs. To capture the impact of improved
access to advanced intermediate input, we construct a measure of
input tariffs following Topalova and Khandelwal (2011). It is defined
as follows:

𝐼𝐼𝑛𝑛_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑖𝑖𝑡𝑡= Σj⍺𝑖𝑖𝑗𝑗 𝑂𝑂𝑢𝑢𝑡𝑡_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑗𝑗𝑡𝑡. (1)

For example, if 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖

we assume that the𝑖𝑖𝑖𝑖shoe = 𝑋𝑋manufacturing
𝑖𝑖𝑖𝑖 /𝑄𝑄𝑖𝑖𝑖𝑖 , industry i
uses two intermediates j (leather and rubber with tariffs of 10% and 20%,
𝜃𝜃 𝜃𝜃 1/𝜃𝜃
[𝛼𝛼(𝐴𝐴𝑖𝑖 𝐾𝐾of𝑖𝑖 )0.8
𝑌𝑌𝑖𝑖 = shares
respectively) and value +and (1 0.2,
− 𝛼𝛼)(𝐵𝐵 𝑖𝑖 𝐿𝐿𝑖𝑖 ) ]
respectively, .then, using
the above formula, the input tariff faced by the shoe industry stands at
𝜃𝜃
12%. The input share,𝐿𝐿𝐿𝐿𝑖𝑖⍺=𝑖𝑗
1−
, is the𝛼𝛼(𝐴𝐴
share 𝑖𝑖 𝑘𝑘𝑖𝑖 )
of .input j in the total input cost
of industry i. We estimate the input shares in this study by using the
𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖𝑖𝑖 =Table
Input–Output Transactions 𝛽𝛽0 + 𝛽𝛽2003–04
1 𝑙𝑙𝑙𝑙𝐴𝐴𝑖𝑖𝑖𝑖𝑖𝑖 +for
𝛽𝛽2 𝑙𝑙𝑙𝑙𝑘𝑘
India 𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽3 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 +from
obtained 𝛽𝛽4 𝑇𝑇𝑗𝑗𝑗𝑗 + 𝑃𝑃𝑖𝑖 + 𝑌𝑌𝑡𝑡
126 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 4.6 Average Tariff for Manufacturing Industry

Groups Analyzed in the Study, 1997–2008

Year Average Output Tariff Average Input Tariff

1997 32.58 14.97
1998 33.86 15.83
1999 35.02 16.47
2000 34.90 16.60
2001 33.84 16.14
2002 30.94 14.81
2003 30.76 14.71
2004 30.87 14.73
2005 20.50 8.83
2006 17.96 8.51
2007 18.52 7.75
2008 14.90 5.78
Source: Authors’ estimates using the Trade Analysis Information System database and
Integrated Data Base accessed through the World Integrated Trade Solution software
of the World Bank.

the Central Statistics Office. Table 4.6 displays the average output and
input tariffs for the manufacturing sector in India during the period
1997–2008.

4.4.3 Controls

Worker-days lost. We suspect that the conditions of labor market

regulations may affect labor share adjustment due to trade liberalization.
Different states in India have different levels of labor market flexibility
because industrial relations fall under the concurrent subject in the
Indian Constitution. This allows state governments to make their own
amendments to the Industrial Disputes Act, which is the key regulation
that governs industrial relations in India. As a result, labor markets have
evolved differently across the various states in India (Besley and Burgess
2004). To measure state-level labor market frictions (condition), we
estimate an index named worker-days lost in strikes and lockouts per
industrial worker, which is the ratio of the total number of days lost in
strikes and lockouts to the total number of industrial workers in the
state. Data on worker-days lost have been obtained from the Indian
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 127

Labour Bureau, while data on the state-wise number of industrial

workers employed have been obtained from the aggregate ASI data.
Export intensity. Exporting patterns of industries can impact
employment and wages of the sector and hence labor share. To control
the same, we include a measure of export
𝐼𝐼𝑛𝑛_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓 𝑖𝑖𝑡𝑡= Σj⍺intensity of the
𝑖𝑖𝑗𝑗 𝑂𝑂𝑢𝑢𝑡𝑡_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓 𝑗𝑗𝑡𝑡. industry at
the two-digit level. We define export intensity as the ratio of exports
to domestic output, that is 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = 𝑋𝑋𝑖𝑖𝑖𝑖 /𝑄𝑄𝑖𝑖𝑖𝑖,, where Q is
the domestic output and X is gross exports. The impact can be either
𝑌𝑌𝑖𝑖 = [𝛼𝛼(𝐴𝐴
negative or positive depending 𝐾𝐾𝑖𝑖 )𝜃𝜃industry
on 𝑖𝑖the + (1 − 𝛼𝛼)(𝐵𝐵 𝑖𝑖 𝐿𝐿𝑖𝑖 ) ]
characteristics
𝜃𝜃 1/𝜃𝜃
. and
hence remains an empirical issue. 𝜃𝜃
𝐿𝐿𝐿𝐿𝑖𝑖 =
Factor intensity dummies. Labor 1 − share 𝑖𝑖 ) .
𝛼𝛼(𝐴𝐴𝑖𝑖 𝑘𝑘adjustment may vary across
industries depending on the factor intensity of their production. To
𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝑙𝑙𝑙𝑙𝐴𝐴𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝑙𝑙𝑙𝑙𝑘𝑘𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽3 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽4 𝑇𝑇𝑗𝑗𝑗𝑗 + 𝑃𝑃𝑖𝑖
analyze this, we use the Hinloopen and Marrewijk (2008) classification
to group industries on the basis of their factor intensities. Hinloopen
and Marrewijk’s classification corresponds to three-digit Standard
International Trade Classification (SITC) level, with 240 items classified
into five categories (number of items in each category in parentheses):
primary (83), natural resource intensive (21), unskilled labor intensive
(26), human capital resource intensive (43), technology intensive (62),
and unclassified (5). To match this classification with our panel data,
we construct a concordance table between SITC three-digit and NIC
three-digit levels to segregate plants based on the factor intensities of
their production.
Technology intensity dummies. In addition to factor intensity-
based classification, we also use the technology intensity classification
of industries based on the Organisation for Economic Co-operation
and Development (OECD) ISIC Rev 3. This classification primarily
corresponds to the two-digit level and classifies manufacturing
industries into four subgroups: high technology, medium-high
technology, medium-low technology, and low technology.3

4.5 Estimation Strategy

𝐼𝐼𝑛𝑛_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑖𝑖𝑡𝑡= Σj⍺𝑖𝑖𝑗𝑗 𝑂𝑂𝑢𝑢𝑡𝑡_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑗𝑗𝑡𝑡.
We start with a constant elasticity of substitution or CES production
function of the following form:
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = 𝑋𝑋𝑖𝑖𝑖𝑖 /𝑄𝑄𝑖𝑖𝑖𝑖 ,

𝑌𝑌𝑖𝑖 = [𝛼𝛼(𝐴𝐴𝑖𝑖 𝐾𝐾𝑖𝑖 )𝜃𝜃 + (1 − 𝛼𝛼)(𝐵𝐵𝑖𝑖 𝐿𝐿𝑖𝑖 )𝜃𝜃 ]1/𝜃𝜃 . (3)

𝐿𝐿𝐿𝐿𝑖𝑖 = 1 − 𝛼𝛼(𝐴𝐴𝑖𝑖 𝑘𝑘𝑖𝑖 )𝜃𝜃 .

𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝑙𝑙𝑙𝑙𝐴𝐴𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝑙𝑙𝑙𝑙𝑘𝑘𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽3 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽4 𝑇𝑇𝑗𝑗𝑗𝑗 + 𝑃𝑃𝑖𝑖 + 𝑌𝑌𝑡𝑡 + 𝜖𝜖𝑖𝑖𝑖𝑖
3
Since “Manufacturing n.e.c industry” at the two-digit level has been allocated to
more than one category, we leave it as “unclassified” for our analysis.
128 Trade Adjustment in Asia: 𝐼𝐼𝑛𝑛_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓
Past Experiences and Lessons Learned
𝑖𝑖𝑡𝑡= Σj⍺𝑖𝑖𝑗𝑗 𝑂𝑂𝑢𝑢𝑡𝑡_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑗𝑗𝑡𝑡.

𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = 𝑋𝑋 /𝑄𝑄 ,
𝑖𝑖𝑖𝑖 𝑖𝑖𝑖𝑖 𝑖𝑖𝑖𝑖
Assuming that labor is paid its marginal product, and following Bentolila
and Saint-Paul (2003), we get: 𝜃𝜃
𝑌𝑌 = [𝛼𝛼(𝐴𝐴 𝐾𝐾 ) + (1 − 𝛼𝛼)(𝐵𝐵 𝐿𝐿 ) ] . 𝜃𝜃 1/𝜃𝜃
𝑖𝑖 𝑖𝑖 𝑖𝑖 𝑖𝑖 𝑖𝑖
𝐼𝐼𝑛𝑛_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑖𝑖𝑡𝑡= Σj⍺𝑖𝑖𝑗𝑗 𝑂𝑂𝑢𝑢𝑡𝑡_𝑡𝑡𝑎𝑎𝑟𝑟𝑖𝑖𝑓𝑓𝑓𝑓𝑗𝑗𝑡𝑡.
𝐿𝐿𝐿𝐿𝑖𝑖 = 1 − 𝛼𝛼(𝐴𝐴𝑖𝑖 𝑘𝑘𝑖𝑖 )𝜃𝜃 . (4)
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = 𝑋𝑋𝑖𝑖𝑖𝑖 /𝑄𝑄𝑖𝑖𝑖𝑖 ,
By log-linearizing equation𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙(4) = 𝛽𝛽0putting
𝑖𝑖𝑖𝑖𝑖𝑖 and + 𝛽𝛽1 𝑙𝑙𝑙𝑙𝐴𝐴
in𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝑙𝑙𝑙𝑙𝑘𝑘𝑖𝑖𝑖𝑖𝑖𝑖
additional + 𝛽𝛽3 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖that
variables + 𝛽𝛽4 𝑇𝑇𝑗𝑗𝑗𝑗 + 𝑃𝑃𝑖𝑖 + 𝑌𝑌𝑡𝑡 +
𝜃𝜃 𝜃𝜃 1/𝜃𝜃
𝑌𝑌𝑖𝑖 = [𝛼𝛼(𝐴𝐴 𝑖𝑖 𝐾𝐾𝑖𝑖 ) + (1 − 𝛼𝛼)(𝐵𝐵 𝑖𝑖 𝐿𝐿 𝑖𝑖 ) ] .
can drive the labor share of income, we derive the following empirical
estimation model: )𝜃𝜃
𝐿𝐿𝐿𝐿𝑖𝑖 = 1 − 𝛼𝛼(𝐴𝐴𝑖𝑖 𝑘𝑘𝑖𝑖 .
𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝑙𝑙𝑙𝑙𝐴𝐴𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝑙𝑙𝑙𝑙𝑘𝑘𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽3 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 + 𝛽𝛽4 𝑇𝑇𝑗𝑗𝑗𝑗 + 𝑃𝑃𝑖𝑖 + 𝑌𝑌𝑡𝑡 + 𝜖𝜖𝑖𝑖𝑖𝑖𝑖𝑖 , (5)

where LSijt is the labor share (total wages divided by the net total sales
revenue) of plant i in industry j at time t; Aijt is the plant productivity; kijt is
the plant capital output ratio; Xijt are other plant-specific controls, such
as age, man-days lost in strikes and lockouts, state where the plant is
located, etc.; Tjt are industry-level trade openness related variables; and
Pi and Yt are plant and year dummies, respectively. Labor share (LS) in
this study is defined as the share of wages in the net sales revenue of the
plant. Plant age has been calculated based on the year of establishment as
reported by the plant managers. Plant capital has been estimated using
the perpetual inventory method following Balakrishnan, Pushpangadan,
and Babu (2000). Plant productivity has been estimated following the
technique of Levinsohn and Petrin (2004).

4.6 Results
The regression results of estimating equation (5) are displayed in Tables
4.7–4.11. The baseline results are shown in Table 4.4. In column (1) we
start with a simple model with both input and output tariff variables.
The coefficient of output tariff is found to be negative and highly
significant, suggesting that a decline in output tariffs led to a rise in labor
share of income. This result suggests that increased foreign competition
triggered firms to hire additional workers. This result hints toward a
Heckscher–Ohlin finding that India expanded the use of its relatively
abundant endowment, namely labor.
However, with respect to input tariffs, the coefficient is positive
and significant at 1%, suggesting that a fall in input tariffs led to a fall in
labor share. This suggests that firms substituted labor with other capital
inputs, such as machinery and equipment, when the policy with respect
to imports of inputs was relaxed.
In the subsequent models in columns 2–4, we introduce several
plant-level and industry-level controls. Note that the results with
respect to input and output tariffs in all the models remain consistent.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 129

In column (2), we introduce plant-level (lnk_int, lnp) and state-level

variables (lnmlpw). The coefficients of lnk_int and lnp are both negative,
as expected, and highly significant, at the 1% level. This result indicates
that a rise in capital input and a rise in productivity both lead to a decline
in the labor share of income. The coefficient of lnmlpw is positive and
highly significant. In the case of India, this variable is closely associated
with union strength; hence, a positive coefficient of lnmlpw is indicative
of a positive impact of union strength and further a higher labor share
of income. From the estimation results of column (2), we can say that
a 10% decline in output tariffs led to a 0.24% rise in the labor share of
income. On the other hand, a 10% decline in input tariffs led to a 0.12%
decline in the labor share of income for large manufacturing plants in
India. The coefficient of export_intensity is negative and significant, at
the 5% level. This reflects the inherent capital-intensive export bias in
the case of Indian manufacturing (Veeramani, Aerath, and Gupta 2018).
While our estimates show that there was a decline in the average
number of workers employed per plant as displayed in Table 4.7, this does
not necessarily imply that tariff reforms had a negative impact on plant-
level employment. To check for this, we run our regression by replacing
the dependent variable with number of workers in equation (5). All the

Table 4.7 Trade Liberalization and Labor Share

Model
(1) (2) (3) (4)
Variables ln_LS_w1 ln_LS_w1 ln_LS_w1 ln_LS_w1
out_tariff –0.0179*** –0.0244*** –0.0321*** –0.0208**
(0.00617) (0.00803) (0.00868) (0.00808)
in_tariff 0.0261*** 0.0125* 0.0149** 0.0144**
(0.00624) (0.00675) (0.00695) (0.00675)
lnk_int –0.0267*** –0.0292*** –0.0250***
(0.00700) (0.00785) (0.00705)
lnp –0.132*** –0.134*** –0.131***
(0.00737) (0.00839) (0.00740)
export_intensity –0.0710**
(0.0301)
lnmlpw 0.0128** 0.00960* 0.0142***
(0.00500) (0.00517) (0.00498)
continued on next page
130 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 4.7 continued

Model
(1) (2) (3) (4)
Variables ln_LS_w1 ln_LS_w1 ln_LS_w1 ln_LS_w1
lnage 0.0825***
(0.0134)
Constant –2.627*** –1.549*** –1.498*** –1.818***
(0.0176) (0.0706) (0.0782) (0.0832)
Year fixed effect Yes Yes Yes Yes
Observations 67,275 67,275 57,573 67,275
R-squared 0.007 0.066 0.069 0.068
Number of plants 13,353 13,353 11,859 13,353
Notes: Dependent variable is plant labor share. out_tariff is 1-year lagged output tariff; in_tariff is 1-year
lagged input tariff; lnk_int is plant capital intensity; lnp is plant productivity; lnage is plant age; lnmlpw is
man-days lost in strikes and lockouts; and export_intensity is export intensity of the industry group to which
the plant belongs. All values are in natural logarithms. Clustered standard errors are in parenthesis. All
regressions include firm and year fixed effects and constant.
*** p<0.01, ** p<0.05, * p<0.1.
Source: Authors’ calculations.

results in Table 4.8 indicate that a decline in both types of tariffs led to
a rise in employment. The coefficient of both input and output tariffs
is negative and highly significant across all models. The coefficients
of the other variables are as expected. A rise in capital intensification
(lnk_int) of the plant is associated with decline in employment. The rise
in productivity (lnp) has a negative and highly significant coefficient,
indicating that as total factor productivity grows less labor is employed.
Older plants tend to have higher levels of employment as reflected by
the positive and highly significant coefficient for the plant age variable
(lnage). The coefficients of man-days lost due to strikes and lockouts
(lnmlpw) and export intensity are statistically insignificant (export_
intensity).
In Table 4.9, we analyze how tariff liberalization had an impact on
various industries based on their factor intensity of production. For this,
we classify industries at the three-digit level into five subgroups. In
our regression model, however, we use four dummies: natural resource
intensive (FI_nat), labor intensive (FI_lab), technology intensive (FI_
tech), and human capital resource intensive (FI_hri). Using the four
dummies and interactions with tariff rates, we find that the different
subgroups adjust the labor share of income in very different ways.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 131

Table 4.8 Impact of Tariff Reforms on Employment

(Plant Fixed Effects Model)

Model
(1) (2) (3)
Variables employment employment employment
out_tariff –0.0445*** –0.0225*** –0.0436***
(0.00671) (0.00616) (0.00943)
in_tariff –0.0165** –0.0330*** –0.0185**
(0.00688) (0.00632) (0.00873)
lnk_int –0.347*** –0.362***
(0.00390) (0.0141)
lnp –0.303*** –0.320***
(0.00395) (0.0152)
lnage 0.368***
(0.00813)
lnmlpw 0.00603 –0.00269
(0.00463) (0.00544)
export_intensity 0.0305
(0.0329)
Constant 5.117*** 6.802*** 8.051***
(0.0192) (0.0478) (0.134)
Year fixed effect Yes Yes Yes
Observations 71,257 71,257 59,782
R-squared 0.030 0.179 0.146
Number of plants 13,945 13,945 12,287
Notes: Dependent variable is plant labor employment. out_tariff is 1-year lagged output tariff; in_tariff is
1-year lagged input tariff; lnk_int is plant capital intensity; lnp is plant productivity; lnage is plant age; lnmlpw
is man-days lost in strikes and lockouts; and export_intensity is export intensity of the industry group to
which the plant belongs. All values are in natural logarithms. Robust standard errors are in parentheses.
*** p<0.01, ** p<0.05, * p<0.1
Source: Authors’ calculations.

Across all specifications, the coefficient of the FI_lab dummy has

a positive and significant sign, as expected. Although positive in most
cases, the coefficient of the FI_tech and FI_hri dummies are insignificant.
In column (2) of Table 4.9, as we interact out_tariff and in_tariff with the
FI_lab dummy, the coefficients of both interaction terms are negative
and significant. This indicates that a decline in both kinds of tariffs led
132 Trade Adjustment in Asia: Past Experiences and Lessons Learned

to an increase in the labor share of income in the labor-intensive sector.

Columns (3) and (4) display the interaction of out_tariff and in_tariff
with the FI_tech and FI_hri dummies, respectively. Unlike the labor-
intensive sector, the interactions are positive and significant, indicating
that a decline in both input and output tariff rates led to a decline in
the labor share in the technology-intensive and human capital resource-
intensive sectors.

Table 4.9 Trade Liberalization, Labor Share, and Factor Intensity

of Production (Plant Fixed Effects Model)

Model
(1) (2) (3) (4)
Variables ln_LS ln_LS ln_LS ln_LS
out_tariff –0.0318*** –0.0123 –0.0279*** –0.0350***
(0.00832) (0.00859) (0.00841) (0.00882)
in_tariff 0.0143** 0.0284*** 0.0136** 0.0162**
(0.00674) (0.00692) (0.00681) (0.00675)
lnp –0.109*** –0.109*** –0.109*** –0.110***
(0.00307) (0.00308) (0.00308) (0.00308)
lnage 0.0900*** 0.0830*** 0.0850*** 0.0846***
(0.0134) (0.0134) (0.0134) (0.0134)
lnmlpw 0.0141*** 0.0116** 0.0139*** 0.0137***
(0.00496) (0.00496) (0.00497) (0.00498)
FI_nat –0.317*** 0.147 0.155 0.135
(0.110) (0.0977) (0.0967) (0.0962)
FI_lab 0.0382 0.649*** 0.0607 0.0363
(0.0971) (0.112) (0.0978) (0.0973)
FI_tech 0.0182 0.0251 –0.0712 0.0148
(0.0912) (0.0930) (0.0982) (0.0915)
FI_hri 0.0708 0.0701 0.0838 –0.0714
(0.0932) (0.0950) (0.0940) (0.103)
out_tariff * FI_nat 0.0967***
(0.0166)
in_tariff * FI_nat 0.0773***
(0.0123)
continued on next page
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 133

Table 4.9 continued

Model
(1) (2) (3) (4)
Variables ln_LS ln_LS ln_LS ln_LS
out_tariff * FI_lab –0.136***
(0.0154)
in_tariff * FI_lab –0.0772***
(0.00877)
out_tariff * FI_tech 0.0173
(0.0131)
in_tariff * FI_tech 0.0210**
(0.00824)
out_tariff * FI_hci 0.0265**
(0.0129)
in_tariff * FI_hci 0.0209**
(0.00925)
Year fixed effect Yes Yes Yes Yes
Observations 67,275 67,275 67,275 67,275
R-squared 0.070 0.074 0.068 0.068
Number of plants 13,353 13,353 13,353 13,353
Notes: Dependent variable is plant labor share in all regressions. out_tariff is 1-year lagged output tariff;
in_tariff is 1-year lagged input tariff; lnp is plant productivity; lnage is plant age; lnmlpw is man-days lost
in strikes and lockouts; FI_nat is the natural resource-intensive industry dummy; FI_lab is the labor-
intensive industry dummy; FI_tech is the technology-intensive industry dummy; and FI_hci is the human
capital resource-intensive dummy. All values are in natural logarithms. Clustered standard errors are in
parenthesis. All regressions include firm and year fixed effects and constant.
*** p<0.01, ** p<0.05, * p<0.1.
Source: Authors’ calculations.

In Table 4.10, we use another classification of industries to separate

the plants based on their technology intensity. This corresponds to the
OECD two-digit technology intensity classification, which comprises
five subgroups: low-technology industries, medium-low-technology
industries, medium-high-technology industries, high-technology
industries, and unclassified. The literature suggests that the labor share
should be affected negatively in the medium-technology-intensive
industries, where labor has an easy and high substitutability with capital.
In column (1), we interact the tech_low dummy with the corresponding
output and input tariffs. The interaction term with the output tariff is
134 Trade Adjustment in Asia: Past Experiences and Lessons Learned

negative and significant, suggesting that, as output tariffs fall, the labor
share rises in low-technology industries. The interaction term with
the input tariffs is positive but insignificant. In column (2), we interact
the tech_midlow dummy with output and input tariffs. Similar to low-
technology industries, the interaction term with output tariffs is negative
and significant. Also, the interaction with input tariffs is negative and
significant, indicating that a decline in input tariffs leads to a rise in the
labor share in medium-low-technology industries. In column (3), we
interact the tech_midhigh dummy with output and input tariffs. Both
coefficients are positive and significant, indicating that the labor share
falls as technology intensity rises. As we move to column (4), which
corresponds to the industries with the most advanced technology, both
the interactions remain positive with higher levels of significance. Overall,
these results suggest that as industries become more technology intensive,
tariff liberalization leads to a decline in the labor share of income.

Table 4.10 Trade Liberalization, Labor Share, and Technology

Intensity of Production (Plant Fixed Effects Model)

Model
(1) (2) (3) (4)
Variables ln_LS ln_LS ln_LS ln_LS
out_tariff –0.0193** 0.0157 –0.0433*** –0.0373***
(0.00802) (0.0106) (0.00885) (0.00819)
in_tariff 0.0192*** 0.0585*** 0.0203*** 0.00615
(0.00672) (0.00854) (0.00681) (0.00675)
lnp –0.117*** –0.116*** –0.117*** –0.117***
(0.00320) (0.00319) (0.00320) (0.00319)
lnage 0.0863*** 0.0882*** 0.0888*** 0.0811***
(0.0134) (0.0133) (0.0134) (0.0134)
lnmlpw 0.0140*** 0.0124** 0.0137*** 0.0135***
(0.00497) (0.00495) (0.00497) (0.00495)
tech_ low 0.112 –0.137** –0.0891 –0.0943
(0.159) (0.0660) (0.0657) (0.0659)
tech_ midlow 0.0298 0.510*** 0.0462 0.0304
(0.0769) (0.0872) (0.0761) (0.0767)
tech_ midhigh 0.108* 0.0796 –0.147** 0.111*
(0.0582) (0.0580) (0.0680) (0.0581)
continued on next page
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 135

Table 4.10 continued

Model
(1) (2) (3) (4)
Variables ln_LS ln_LS ln_LS ln_LS
tech_high –0.261*** –0.291*** –0.251*** –0.766***
(0.0555) (0.0557) (0.0551) (0.0724)
out_tariff * tech_low –0.0793*
(0.0444)
in_tariff * tech_low 0.0173
(0.0249)
out_tariff * tech_midlow –0.110***
(0.0129)
in_tariff * tech_midlow –0.0615***
(0.00790)
out_tariff * tech_ midhigh 0.0742***
(0.0119)
in_tariff * tech_midhigh 0.0138*
(0.00805)
out_tariff * tech_high 0.121***
(0.0152)
in_tariff * tech_high 0.0538***
(0.00922)
Year fixed effect Yes Yes Yes Yes
Observations 67,275 67,275 67,275 67,275
R-squared 0.072 0.078 0.074 0.077
Number of plants 13,353 13,353 13,353 13,353
Notes: Dependent variable is plant labor share in all regressions. out_tariff is 1-year lagged output tariff;
in_tariff is 1-year lagged input tariff; lnp is plant productivity; lnage is plant age; lnmlpw is man-days lost
in strikes and lockouts; tech_low is the low-technology-intensive industry dummy; tech_midlow is the
medium-low-technology-intensive industry dummy; tech_midhigh is the medium-high-technology-
intensive industry dummy; and tech_high is the high-technology-intensive industry dummy. All values are
in natural logarithms. XXX standard errors are in parenthesis. All regressions include firm and year fixed
effects and constant.
*** p<0.01, ** p<0.05, * p<0.1.
Source: Authors’ calculations.

In the case of India, labor laws fall under the concurrent list and
are hence controlled both by the center and the state. As a result, Indian
states vary significantly from each other in terms of the conditions of
their labor laws, which makes some states more flexible (employer
136 Trade Adjustment in Asia: Past Experiences and Lessons Learned

friendly) than others. To analyze whether state-level differences

in labor market flexibility affect the labor share of income, we ran
separate regressions for plants located in states with flexible labor laws
(column (1) of Table 4.11) and plants located in states with inflexible
labor laws (column (2) of Table 4.11). In the case of states with flexible
labor laws, the sign and significance of out_tariff remains similar to the
baseline results; in_tariff is, however, insignificant. In the case of states
with inflexible labor laws, the coefficients are positive and significant,
indicating that a decline in both types of tariff rates led to a decline in the
labor share of income. This reflects plant-level decisions in favor of less
labor-augmenting methods of production.

Table 4.11 Trade Liberalization, Labor Share,

and Labor Market Flexibility

Model
(1) (2)
Variables ln_LS ln_LS
out_tariff –0.0666*** 0.0215**
(0.0142) (0.0107)
in_tariff –0.0130 0.0377***
(0.0106) (0.00914)
lnk_int –0.0197 –0.0327***
(0.0126) (0.00904)
lnp –0.124*** –0.128***
(0.0134) (0.00935)
lnmlpw 0.0210** –0.00304
(0.00939) (0.00684)
Constant –1.397*** –1.743***
(0.127) (0.0911)
Year fixed effect Yes Yes
Observations 23,155 36,238
R-squared 0.063 0.068
Number of plants 4,579 7,149
Notes: Dependent variable is plant labor share in all regressions. out_tariff is 1-year lagged output
tariff; in_tariff is 1-year lagged input tariff; lnk_int is plant capital intensity; lnp is plant productivity;
and lnmlpw is man-days lost in strikes and lockouts. All values are in natural logarithms. Clustered
standard errors are in parenthesis. All regressions include firm and year fixed effects and constant.
*** p<0.01, ** p<0.05, * p<0.1.
Source: Authors’ calculations.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 137

4.7 Conclusion
This chapter analyzed the impact of input and out put tariff liberalization
on the labor share in the Indian manufacturing sector. We use data on
large Indian manufacturing plants for the period 1998–1999 to 2007–
2008. Overall, we find a decline in the labor share during the period of
our analysis. As to whether this implies that tariff liberalization made the
workers worse-off, our econometric analysis suggests that a reduction
in output and input tariffs had a differential impact on the labor share.
A fall in output tariffs led to a rise in the labor share. Given that labor
is abundant in India, this trend falls in line with the predictions of the
Heckscher–Ohlin model. In contrast, a fall in input tariffs led to a decline
in the labor share, which suggests that as accessibility to cheaper inputs
increased, Indian manufacturers substituted labor with capital.
We then extend our analysis to incorporate sector-level differences
across plants in factor intensity and technology intensity of production.
Segregating industries based on factor intensity and technology
intensity classifications yields a different picture. On the one hand,
we find that the overall decline in both input and output tariffs led to
a decline in the labor share in technology-intensive and human capital
resource-intensive sectors. On the other hand, it led to a rise in the labor
share for labor-intensive and low-technology plants. However, given
India’s manufacturing sector has a bias in favor of capital-intensive
manufacturing, the overall labor share on average has declined.
A plausible explanation of this capital-intensive manufacturing bias
is that India’s rigid labor laws discourage firms from hiring and firing
easily and hence employing labor more freely. To investigate if labor law
differences across states account for losses to workers, we run separate
regressions for plants based on their location. Our results suggest that
workers employed in states with flexible labor laws gained from tariff
reductions while workers employed in states with rigid labor laws
witnessed losses in labor share. These findings are crucial given that
India has a large unskilled and semi-educated population, with its
comparative advantage in labor-intensive sectors. Labor law rigidity
has likely done more harm than good to Indian workers. Further, gains
from trade liberalization were attenuated due to the capital-intensive
production bias in Indian manufacturing.
Our research has certain limitations that should be kept in mind
when interpreting and possibly generalizing the findings. First, our
sample only includes relatively large plants. These firms tend to be more
capital intensive and more productive. In a Melitz-type trade model,
these firms are more likely to benefit from trade liberalization compared
to smaller firms. Due to increased foreign competition induced by trade
138 Trade Adjustment in Asia: Past Experiences and Lessons Learned

opening, smaller, less productive firms might be forced to shrink or even

exit the market. Our results only reflect the adjustment that occurred
in large firms. Furthermore, our data at the plant level do not reveal
how plants are linked. Trade opening might have led to a reorganization
of activities across plants, but within the same firm. Firms might use
profits earned by one activity to subsidize other loss-making activities.
Such firm behavior might bias our estimation results, although, in the
long run, firms can only survive if they are able make profits.
The research results presented in this chapter answered several
important questions, but many remain. For example, our measurement
of the labor share is rather crude. It would be interesting to distinguish
the wage bill for skilled and unskilled labor. We have evidence for other
countries that trade opening benefited high-skilled labor more. Knowing
which group of workers benefited most from the trade opening in India
would be important for policy makers. In case high-skilled workers
benefited most, more investment in education may be warranted.
Furthermore, the plants in our sample are different not only in terms
of size but also in the number of products they produce. It would be
worthwhile testing whether the labor adjustment was different in firms
with multiple products compared to those selling few products. In case
of multiple products, labor could be reallocated more easily. Many other
pertinent questions can be answered using this rich dataset. We hope
that this chapter will prod other economists to undertake additional
research on the trade opening of India.
The Impact of Tariff Liberalization on the Labor Share in India’s Manufacturing Industry 139

References
Ahsan, R.N., and D. Mitra. 2014. Trade Liberalization and Labor’s Slice
of the Pie: Evidence from Indian Firms. Journal of Development
Economics. 108 (C). pp. 1–16.
Balakrishnan, P., K. Pushpangadan, and M.S. Babu. 2000. Trade
Liberalization and Productivity Growth in Manufacturing: Evidence
from Firm-Level Panel Data. Economic and Political Weekly. 35 (41).
pp. 3679–82.
Banerjee, A., and A. Newman. 2004. Notes for Credit, Growth, and Trade
Policy. Massachusetts Institute of Technology.
Banga, R., and A. Das. eds. 2012. Twenty Years of India’s
Liberalization: Experiences and Lessons. Geneva: United Nations
Conference on Trade and Development. http://unctad.org/en
/PublicationsLibrary/‌osg2012d1_en.pdf
Bentolila, S., and G. Saint-Paul. 2003. Explaining Movements in the
Labor Share. Contributions in Macroeconomics. 3 (1). pp. 1–33.
Besley, T., and R. Burgess. 2004. Can Labor Regulation Hinder Economic
Performance? Evidence from India. The Quarterly Journal of
Economics. 119 (1). pp. 91–134.
Bollard, A., P.J. Klenow, and G. Sharma. 2013. India’s Mysterious
Manufacturing Miracle. Review of Economic Dynamics. 16 (1).
pp. 59–85.
Cain, J., R. Hasan, and D. Mitra. 2012. Trade Liberalization and Poverty
Reduction: New Evidence from Indian States. In J. Bhagwati and
A. Panagariya, eds. India’s Reforms: How They Produced Inclusive
Growth. New York: Oxford University Press. pp. 91–185.
Das, D.K. 2003. Manufacturing Productivity under Varying Trade Regimes:
India in the 1980s and 1990s. Working Paper No. 107. New Delhi:
Indian Council for Research on International Economic Relations.
Goldar, B. 2012. Productivity in Indian Manufacturing in the Post-
Reform Period: A Review of Studies. In V. Kathuria, Rajesh Raj S.N.,
and K. Sen, eds. Productivity in Indian Manufacturing: Measurement,
Methods and Analysis. New Delhi: Routledge.
Gupta, P. 2016. Trade, Productivity and Markups: Analysis of Indian
Manufacturing Plants (PhD thesis). Mumbai: Indira Gandhi
Institute of Development Research.
Harrison, A.E., L.A. Martin, and S. Nataraj. 2012. Learning versus
Stealing: How Important Are Market-Share Reallocations to India’s
Productivity Growth? The World Bank Economic Review. 27 (2).
pp. 202–28.
Hasan, R., D. Mitra, and K.V. Ramaswamy. 2007. Trade Reforms, Labor
Regulations, and Labor-Demand Elasticities: Empirical Evidence
from India. The Review of Economics and Statistics. 89 (3). pp. 466–81.
140 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Hasan, R., D. Mitra, P. Ranjan, and A. Reshad. 2012. Trade Liberalization

and Unemployment: Theory and Evidence from India. Journal of
Development Economics. 97 (2). pp. 269-80.
Hinloopen, J., and C. van Marrewijk. 2008. Empirical Relevance of
the Hillman Condition for Revealed Comparative Advantage:
10 Stylized Facts. Applied Economics. 40 (18). pp. 2313–28.
Hsieh, C.T., and P.J. Klenow. 2009. Misallocation and Manufacturing
TFP in China and India. The Quarterly Journal of Economics. 124 (4).
pp. 1403–48.
Kannan, K.P., and G. Raveendran. 2009. Growth sans Employment:
A Quarter Century of Jobless Growth in India›s Organised
Manufacturing. Economic and Political Weekly. 44 (10). pp. 80-91.
Krishna, P., and G. Sethupathy. 2011. Trade and Inequality in India.
NBER Working Paper No. 17257. Cambridge, MA: National Bureau
of Economic Research.
Levinsohn, J., and A. Petrin. 2003. Estimating Production Functions
Using Inputs to Control for Unobservables. The Review of Economic
Studies. 70 (2). pp. 317–41.
Marjit, S., H. Beladi, and A. Chakrabarti. 2004. Trade and Wage Inequality
in Developing Countries. Economic Inquiry. 42 (2). pp. 295–303.
Mishra, P., and U. Kumar. 2005. Trade Liberalization and Wage
Inequality: Evidence from India. IMF Working Paper WP/05/20.
Washington, DC: International Monetary Fund.
Sivadasan, J. 2009. Barriers to Competition and Productivity: Evidence
from India. The BE Journal of Economic Analysis & Policy. 9 (1).
pp. 1935–62.
Thomas, J.J. 2013. Explaining the “Jobless” Growth in Indian
Manufacturing. Journal of the Asia Pacific Economy. 18 (4).
pp. 673–92.
Topalova, P. 2007. Trade Liberalization, Poverty and Inequality:
Evidence from Indian Districts. In A. Harrison, ed. Globalization
and Poverty. Chicago, IL: University of Chicago Press for National
Bureau of Economic Research. pp. 291–336.
Topalova, P., and A. Khandelwal. 2011. Trade Liberalization and Firm
Productivity: The Case of India. Review of Economics and Statistics.
93 (3). pp. 995–1009.
Veeramani, C., L. Aerath, and P. Gupta. 2018. Intensive and Extensive
Margins of Exports: What Can India Learn from China? The World
Economy. 41 (5). pp. 1196–222.
World Bank. 2019. World Development Indicators. https://datacatalog
.worldbank.org/dataset/world-development-indicators
 5
Export Boom,
Employment Bust?
The Paradox of Indonesia’s
Displaced Workers, 2000–2014
Rashesh Shrestha and Ian Coxhead

5.1 Introduction
Charles Dickens’ phrase “it was the best of times; it was the worst
of times” is for many Indonesian workers an apt summary of their
experience during the early 2000s. While the national economy and
especially its resource-exporting sectors enjoyed trade-driven growth
of unprecedented magnitude and duration, millions of blue-collar
workers and labor market entrants found themselves paradoxically
sidelined from well-paid jobs in manufacturing, and instead forced to
seek livelihoods in low-paid, low-skill service sector jobs. This happened
at a time when many Asian countries, led by the People’s Republic of
China, were enjoying (continued) expansion of manufacturing trade
by participating in global production networks, which in turn created
better employment opportunities for their less-skilled agricultural
workforces. For many Indonesians, on the other hand, the boom was
a period of stagnating real wages and diminished earnings prospects,
even as national income and spending surged ahead and overall
expectations for the future became increasingly bright. For workers,
the consequence of job displacement due to structural change would
have been particularly severe during this time.
The phenomenon of job displacement accompanied by earnings
losses is familiar from studies of developed country labor markets. A
substantial literature has explored the causes, duration, and implications
of job displacement in developed countries (e.g., Jacobson, LaLonde, and
Sullivan 1993; Kletzer 1998; Couch and Placzek 2010; Korkeamäki and

141
142 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Kyyrä 2014). In some developed countries, trade-related “downskilling”

(Modestino, Shoag, and Balance 2016) and declining real earnings,
especially for blue-collar workers, has become increasingly widespread,
and has been linked to competition from lower-cost manufacturers,
including in the People’s Republic of China (Autor et al. 2014; Autor,
Dorn, and Hanson 2016).1 Widespread job displacement has also been
a concern in Latin American countries undergoing major trade policy
adjustments, typically under highly adverse macroeconomic conditions
(Goldberg and Pavcnik 2005, 2007; McMillan and Rodrik 2011). In
most developing countries, however, job displacement has not been at
the forefront of issues studied by economists, likely because in these
countries structural change usually involves a transition to better jobs.
In contrast to the foregoing, Indonesian data reveal job displacement
in a setting that differs in one very visible way: It occurs not during a
negative trade shock or a period of crisis-induced macroeconomic
adjustment, but against a background of rapid economic expansion.
Studies of comparable “job displacement” in the United States (US) and
Europe dwell almost without exception on the aftereffects of negative
shocks, whether caused by recessions or by trade competition from
emerging economies such as the People’s Republic of China. Indonesia’s
manufacturing industries have also been impacted by external
competition (Coxhead 2007). But in addition, Indonesia—along with
Brazil and numerous other developing economies—has experienced
strong and sustained growth in global demand for its energy and natural
resource products. It is this resource export growth that has been the
dominant driver of structural change since about 2000.
There is in addition another subtler set of differences in the
Indonesian case. These arise from the fact that Indonesia, unlike even
the large Latin American economies, was (at least at the beginning of this
period) a low-income economy. While the rate of Indonesia’s economic
growth during the recent export boom was high, the boom induced
changes in the structure of economic activity—and thus of employment—
that were biased against high-wage jobs for blue-collar workers and
were sufficiently large to deny many poor Indonesians a share in the
proceeds of the boom. The paradox is solved by noting that during this
structural change, industries (largely in manufacturing) offering “good”
semi-skilled jobs, and especially industries in which formal employment
is widespread, contracted sharply relative to industries (largely services)
in which skills are seldom rewarded and in which regulation of wages
and employment conditions is almost totally lacking. Increased labor

1
This has directly contributed to current political backlash against globalization and
trade in the United States and Europe.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 143

market rigidity due to new regulations were likely contributing factors.

As a result, new labor market entrants and workers displaced from the
former types of jobs could still find employment. However, they could
do so only in sectors and occupations paying less, and offering few
prospects for promotion and no contractual security.
Each of these features of the Indonesian experience is salient to the
analysis we conduct, as will become clear. In this chapter, we explore
how workers who started out in formal employment fared several years
later, and how their transition out of the sector affected earnings. We
use the 2000, 2007, and 2014 rounds of the Indonesian Family Life
Survey (IFLS) data, research that tracks individual workers over time.
We find that transition out of formal employment, which is strongly
associated with manufacturing jobs, leads to substantial loss of earnings.
The results demarcate an episode of job displacement and earnings
losses during “the best of times” not previously examined in a rigorous
fashion. Building on our previous work examining the consequences of
Indonesia’s palm oil export boom for the structure of the labor market,
inequality, and educational incentives (Coxhead and Shrestha 2016), we
show that high growth does not automatically translate into positive
labor market changes for workers. One implication is that high growth
not matched by increased formalization or the creation of more formal
jobs could be a reason for the exacerbation of inequality in Indonesia.

5.2 Trade Shocks and Job Displacement

Given the seemingly strong symbiosis between the growth of labor-
intensive manufacturing and increases in income, it is not surprising that
development economists have expressed concern over two phenomena
that appear to indicate an ongoing structural change in this relationship.
One of these is “premature deindustrialization,” wherein (for reasons
including policy changes and loss of global competitiveness) the gross
domestic product (GDP) share of manufacturing peaks at a lower value
and at a lower level of income per capita than was the case in earlier
development experiences (Rodrik 2015). The underlying notion is that
economic growth led by labor-intensive manufacturing is preferable for
countries with a large pool of less-skilled labor. Loss of momentum in
labor-intensive manufacturing is unlikely to cause overall employment
to drop; rather, it is part of a structural shift in which job opportunities
in “good” (mainly formal) employment are replaced by the growth
of less productive, poorly paid, and insecure jobs, mainly in informal
services.
Concerns over job displacement have long been widespread in
developed economies, and a large literature has explored its causes,
144 Trade Adjustment in Asia: Past Experiences and Lessons Learned

duration, and implications (e.g., Jacobson, LaLonde, and Sullivan 1993;

Kletzer 1998). In early contributions to this literature, job displacement
was typically observed during macroeconomic downturns resulting
from the internal dynamics of the business cycle (Jacobsen, LaLonde,
and Sullivan 1993; Couch and Placzek 2010). Because business cycle
recessions are typically brief, the main policy concern was less with
temporary earnings drops and spells of unemployment, but rather
with persistent loss of individual earning power over longer periods
(Jacobson, LaLonde, and Sullivan 1993). Persistent losses from job
displacement were explained by loss of job-specific human capital and
loss of returns to job tenure (Carrington and Fallick 2014).
More recently, attention has shifted from business cycle
displacement to deeper structural causes associated with changes in
the relative competitiveness of domestic and foreign industries. Studies
of the effects of People’s Republic of China-related trade shocks on US
workers (Autor et al. 2014; Autor, Dorn, and Hanson 2016) have found
that US workers who are more exposed to a trade shock from the People’s
Republic of China have worse outcomes in terms of cumulative earnings
growth and employment. There is a substantial amount of switching
between jobs, industries, and sectors, but a surprisingly low rate of
geographical relocation. Because of this, trade-related job displacement
can have prolonged negative effects on welfare, especially among less-
skilled workers. Splitting their labor market sample by terciles of pre-
exposure earnings to capture the heterogeneity of impact on workers
with different earnings capabilities, Autor et al. (2014) find that workers
in the lowest tercile face a larger effect from exposure to a negative
shock than workers at the top end. Furthermore, this effect is driven by a
lower ability of low-earnings workers to adjust at the extensive margin,
that is, to exit from sectors with greater exposure and find jobs in less
exposed sectors. Thus, the capacity to recover from a negative labor
market shock is positively correlated with initial earnings. This result
is less surprising when we consider that variation in initial earnings is
itself a measure of individual education and ability.
The majority of the literature focuses on developed economy cases
in which labor markets are typically more complete, with lower search-
and-matching costs than in developing countries. Earnings losses
associated with job displacement are likely to be much more severe in
the context of a developing country, where the coexistence of formal and
informal labor market institutions greatly increases the significance of
job displacement. Displacement from formal employment could result
in large losses in individual welfare due to low earnings in informal
employment. The existing literature has focused on understanding the
characteristics of formal and informal workers, and in particular on
whether informality is voluntary or forced (see discussion in Gindling
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 145

2014). However, we know much less about how transition between

formality and informality affects individuals.
Likewise, the welfare implications of job displacement are arguably
more important in the case that unemployment insurance is lacking.
In addition, analysis of labor market adjustments is more complicated
because it typically takes place in an economy undergoing a secular
process of structural change in addition to short-run macroeconomic
shocks. Distinguishing these and their effects on labor demand and
returns to specific worker characteristics is an important task.

5.3 Indonesia’s Recent Trade

and Employment Trends
Indonesia is a relatively poor country—GDP per capita was just 31%
of the world average in 2017—but since 2001 its economy has grown at
5%–6% per year, about double the world average. At this rate, aggregate
labor demand growth should be correspondingly rapid, and with labor
force growth much slower at about 2% per year, real wages should rise.
However, growth has been accompanied by significant changes in the
structure of production and employment. Notably, manufacturing, a
prominent driver of growth in the 1990s, has been sluggish, while output
and employment in a wide range of services industries has expanded
(Aswicahyono, Hill, and Narjoko 2011).
Although Indonesia’s exports are now dominated by primary
commodities, manufacturing trade was a major source of job creation in
the decade prior to the Asian financial crisis. During the 1980s and early
1990s, the manufacturing industry (especially labor-intensive, export-
oriented sectors such as textiles, garments, and footwear) expanded
much faster than GDP. Between 1980 and 1991, the share of production
of these sectors in total manufacturing value added increased from 19%
to 30% (Sjöholm 1999). In 1996, just before the Asian financial crisis,
textiles and garments accounted for over 10% of Indonesian exports
(Figure 5.1). Subsequently, however, their share steadily declined as that
of primary commodities rose, as shown in Figure 5.2. While the total
value of their exports grew, their share in Indonesia’s total exports and
also in world exports both fell sharply. The commodity boom caused
substantial real exchange rate appreciation, rendering Indonesian
exports less competitive.2 Product market competition from low-cost

2
A more positive interpretation of the Indonesian experience in the 2000s follows
the Balassa–Samuelson hypothesis, in which productivity growth in tradable sectors
drives up real wages across the entire economy, combined with income-elastic
preference for nontradable services (Dornbusch 1988). However, there is no evidence
either of differential productivity growth or of rising real wages.
146 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 5.1 Indonesia’s Export and Import

Composition by Sector, 1996
Export Import
40

30
% Trade value in 1996

20

10

0
o ic l
Fo d P als
ot rod
H F ear
Mdes els
h in
M et ec
M ine als
Pl e ls
St ast llan
Te ne ub
Tr xtC las
Veans oth
ta rt
W ble
d

o ic l
Fo d P als
ot rod
H F ear
M es ls
h in
M et ec
M ine als
Pl sce als
St ast llan
Te ne ub
Tr xtC las
Veans oth
ta rt
W ble
d
Fo em ima

Fo em ima
oo

oo
isc ra

id ue
ge po

ge po
ac Sk

ac Sk
M El

M El
o iR

o iR
G

i r

G
i u
w

w
l

l
Ch An

Ch An

Source: United Nations International Trade Statistics Database (UN Comtrade).

Figure 5.2 Trends in Indonesia’s Exports

of Key Products, 1990–2016
(a) Total Value (b) Share in Total Domestic Exports
24 20
% of East Asia exports

23 15
Logged value in $

22 10

21 5

20 0
1990 1995 2000 2005 2010 2015 1990 1995 2000 2005 2010 2015
Year Year
Textile Wood products Footwear Textile Wood products Footwear

(c) Share in World Exports

6
% of world exports

1990 1995 2000 2005 2010 2015

Year
Textile Wood products Footwear

Source: United Nations International Trade Statistics Database (UN Comtrade).

Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 147

manufacturers in the People’s Republic of China further depressed what

had been an important source of employment growth in the 1990s. Both
of these channels led to specialization in nontradable sectors.3
From an employment perspective, labor-intensive manufacturing
is crucial to the development of nearly all emerging countries. Labor-
intensive sectors tend to have a higher employment elasticity of output
and, as such, their expansion leads to greater job creation. The mid-1990s,
as it happens, was the high-water mark of Indonesia’s manufacturing
job expansion. According to data in Aswicahyono, Hill, and Narjoko
(2011), employment growth in manufacturing, which had averaged 6%
per year in 1990–1996, fell to an average rate of just 0.9% per year in
2000–2008, less than half the overall employment growth rate. The
fastest rates of job growth in 2000–2008 were in construction (5.7%
per year), transport (3.9%), and other services (3.6%), as well as mining
(3.7%). Census data obtained from IPUMS International (Minnesota
Population Center 2017) show that the share of males and females aged
20–49 years working in manufacturing peaked in the 1990s, but in 2000,
following the 1997–1999 Asian financial crisis, it was once again below
its 1990 level. It recovered slightly between 2000 and 2010 but did not
regain the 1995 peak (Figure 5.3).
The precrisis manufacturing expansion was beneficial in terms not
only of the numbers of jobs created but also of their characteristics.
For given labor quality and production technology, growth of workers’
earnings depends heavily on complementarities between labor and
other factors of production, such as land and capital. For countries with
low-skilled labor, investment in the manufacturing sector, by increasing
the stock of capital, provides a direct path to higher labor productivity
and thus labor earnings.
In addition to higher unit earnings, labor-intensive manufacturing
provided an opportunity for many workers to be formally employed. In
Indonesia, as in other developing countries, formal employment means
a great deal but is quite rare among blue-collar workers. One recent
survey estimated that labor productivity in a median informal firm in

3
In the case of Indonesia, trade shocks may also interact with labor market regulations.
Indonesian labor market reforms introduced in 2003 are thought to have discouraged
expansion of formal jobs (Garnaut 2015; World Bank 2010). These reforms included
greater freedom to unionize and bargain collectively for wages, higher minimum
wages, stricter hiring and firing rules, and increased severance pay and long service
pay requirements upon job separation (Manning and Roesad 2007). Stringent labor
laws have been proposed as an explanation for low rates of formal sector job creation
in general (World Bank 2010). The labor regulations in Indonesia increased the costs
associated with both hiring and firing, making it harder for displaced workers to find
other formal jobs. If workers are unable to access formal jobs upon displacement,
labor market regulations such as these are likely to create segmentation in the labor
market and protect the lucky few who can maintain formal employment status.
148 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 5.3 Manufacturing Employment as Share of the

Population Aged 20–49 Years for Males and Females
12
% Manufacturing employees

10

4
1985 1990 1995 2000 2005 2010
Year
Males Females

Source: Author's calculation from IPUMS International data (MinnesotaPopulation Center 2017).

Indonesia is just 4.5% that in an average formal firm; labor productivity

in the two sectors converges only in the top few percentiles (Rothenberg
et al. 2016). Nearly all informal firms are tiny (under five workers); the
1990s manufacturing boom was dominated by growth in relatively large
firms. As such, it is no surprise that the share of Indonesia’s labor force
recorded as engaged in formal employment reached a peak of about 45%
immediately prior to the Asian financial crisis and has declined since
(World Bank 2010).
Other surveys allow us to study further the characteristics of the
jobs created by Indonesia’s manufacturing growth prior to the Asian
financial crisis. According to our analysis of the Survei Angkatan Kerja
Nasional (SAKERNAS), the Indonesian labor force survey, at the peak of
the manufacturing boom in 1997 over 5% of the Indonesian workforce
was engaged in the “textile, ready to wear clothes and leather” industry
and the “wooden commodities industry including furniture.”4 Among
younger workers (aged 15–29 years), the proportion was higher at 9%,
thus providing opportunities for Indonesia’s younger population to
engage in stable work outside agriculture. Geographically, these jobs
were highly concentrated, with over 50% of employment in these two
sectors located on Java. In terms of education, workers in these sectors

4
The corresponding industry codes are 32 and 33 in SAKERNAS 1997.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 149

were slightly more educated than the overall population, but not by
much. These sectors had slightly higher concentrations of workers with
junior high school-level schooling. They also had slightly better gender
ratios than overall nonagricultural work, so they were also crucial for
improved participation of women in formal work. By 2007, however, less
than 4% of workers were involved in these sectors.5 The concentration
of younger workers had also declined to 5.3%.
Studies on job transitions in Indonesia have by and large focused
on the secular movement of labor out of agriculture and into “modern”
sectors, such as manufacturing or urban services. Suryahadi, Hadiwidjaja,
and Sumarto (2012) and Suryadarma, Suryahadi, and Sumarto (2013)
both find that service sector growth made a substantial contribution
to poverty reduction in Indonesia in the 2000s. A few studies have
examined labor market responses to macroeconomic shocks, such as
the Asian financial crisis (e.g., Manning 2000), but these studies are
more descriptive than quantitative. Our own previous work (Coxhead
and Shrestha 2016) provides causal connections from trade shocks to
structural change in the labor market, notably the shift from formal to
less formal employment.
The fact that formal jobs in labor-intensive sectors provide
opportunities for low-skilled individuals to raise their earnings and
move out of poverty motivates our exploration of the consequences of
slower growth in high-productivity sectors and occupations, especially
in a developing country with a relatively large endowment of low-skilled
workers. The possible effects of slow growth include the relegation of
some less-skilled workers into low-productivity sectors due to lack of
opportunities, and/or reduced earnings and job security for workers
who get displaced by the shrinking of these sectors. The situation for
individual workers is much more difficult when slow job creation is
coupled with structural change that increases the rate of job separation
or reduces growth in formal jobs.
At the aggregate level, we can study patterns of job displacement
by tracking employment patterns for the same age cohorts across time.
For example, workers aged 20–24 years in 1985 would be aged 25–29 in
1990, 30–34 in 1995, and so on. By comparing the sectoral distribution
of employment of the same initial cohort, we can see how structural
change over time affects employment patterns. Stark evidence of
structural change can especially be found in data for younger workers,
as first-time job seekers are most likely to enter the sector that has been

5
The industry classification codes used in SAKERNAS 2007 differ from SAKERNAS
1997 as it uses three-digit classification codes rather than two-digit codes. The
relevant sectors have codes between 171 and 210 in 2007.
150 Trade Adjustment in Asia: Past Experiences and Lessons Learned

expanding. Thus, a growing manufacturing sector would attract more

young workers. We can observe this in repeated cross-sectional data by
focusing on successive cohorts of labor market entrants.
Figure 5.4 plots the share of manufacturing employment among
cohorts aged 20–24 years in the initial year (either 1985, 1990, 1995,
2000, or 2005) between 1985 and 2010. First, the share of new job market
entrants finding employment in manufacturing fell sharply after 1995.
Furthermore, each line in the figure tracks a cohort of individuals initially
aged 20–24 in the beginning year. These data show a rise in the share of
manufacturing employment in the 1990s followed by a sharp decline in the
early part of the 2000s within the same age cohort. For the male cohort
aged 20–24 in 1985, at least 12% worked in manufacturing until 1995, after
which the share fell to less than 10%. More than 13% of the 1995 urban
cohort started out in manufacturing, but by 2010 only 10% remained in
this sector. By 2005, only 8% of men aged 20–24 had manufacturing jobs,
although this number had increased slightly by 2010.
What can be inferred from these patterns? Overall, these figures
demonstrate that, since the late 1990s, not only were new entrants in the
labor market less likely than their predecessors to land manufacturing
jobs, but workers already holding manufacturing jobs were also
transitioning out of the sector. This trend in manufacturing jobs comes
despite growth in sector output, albeit at a rate much lower than in
the 1990s (Aswicahyono, Hill, and Narjoko 2011). Part of the reason
could also be that this cohort stayed in school to a greater extent than
previous cohorts, but lack of opportunity in manufacturing is likely to
be a major factor.

Figure 5.4 Trends in Shares of Male and Female

Manufacturing Employment by Age Cohort
(a) Males (b) Females
10
14
% Manufacturing employment
% Manufacturing employment

12 8

10
6

4
6
1985 1990 1995 2000 2005 2010 1985 1990 1995 2000 2005 2010
Year Year
1985 cohort 1990 cohort 1985 cohort 1990 cohort
1995 cohort 2000 cohort 1995 cohort 2000 cohort
2005 cohort 2010 cohort 2005 cohort 2010 cohort

Source: Authors’ computations using census data obtained from IPUMS (Minnesota Population
Center 2017).
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 151

5.4 Data and Estimation

To understand the impact of such displacement on worker earnings,
we require panel data tracking individual workers over time. The IFLS
provides such an opportunity to explore the Indonesian case. The IFLS
is a panel study that began in 1993, with follow-up rounds in 1997, 2000,
2007, and 2014 (for a detailed description of the surveys, see Strauss et
al. 2004, 2009). We perform two different analyses, making use of two
features of the IFLS data. The panel nature of the IFLS enables us to track
the evolution of labor market status of individual workers over time. In
addition, in a module concerning each worker’s displacement experience
over the past 5 years, added in 2007, respondents were asked to report
any termination from a salaried job in the previous 5 years. Although
the Indonesian structural change is one of declining manufacturing
competitiveness in textile, footwear, and wood products, unavailability of
detailed data (industry codes in the IFLS are only available at the one-
digit level) means that we use information on all formal workers.
The pattern over time in the IFLS employment data matches closely
that observed in national surveys. Panels (a) and (b) of Table 5.1 show
the distribution of workers by occupation and sector, respectively. In
terms of occupation, the proportions of production and semi-skilled

Table 5.1 Employment Characteristics

in Indonesian Family Life Survey Data, 1993–2014 (%)
(a) Distribution of Workers Aged 20–65 by Occupation

1993 1997 2000 2007 2014

Professional 4.59 5.28 4.81 5.09 5.37
Admin 1.92 0.24 0.37 0.32 0.41
Clerical 3.62 5.21 4.38 4.24 5.74
Sales 18.69 22.19 16.28 19.02 19.73
Service 6.76 5.09 14.71 15.26 15.72
Agriculture 38.90 33.37 32.83 30.07 25.13
Production 8.13 9.75 8.28 7.70 7.58
Semi-skilled 3.21 4.44 3.00 3.07 3.23
Laborer 12.32 12.54 13.52 13.29 15.01
Others 1.85 1.90 1.83 1.94 2.08
N 9,064 10,540 14,899 18,174 21,819
continued on next page
152 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.1 continued

(b) Distribution of Workers Aged 20–65 by Sector

1997 2000 2007 2014

Agriculture 33.56 33.06 30.25 25.76
Mining 0.62 0.55 0.64 1.30
Manufacturing 15.64 13.88 13.34 13.07
Utilities 0.51 0.34 0.32 0.51
Construction 5.58 4.50 4.78 5.00
Wholesale/retail trade 22.52 22.20 24.69 25.52
Transport 4.54 4.19 3.51 2.45
Financial services 0.85 0.78 0.89 4.70
Social services 16.18 20.50 21.57 21.68
N 10,540 14,896 18,174 21,819
Source: Authors’ calculation from various Indonesian Family Life Survey rounds. The 1993 survey did not
contain sector categories.

workers, who are mostly associated with the manufacturing sector,

remain unchanged or decline slightly between 1993 and 2014, after a
slight increase in 1997. Likewise, sectoral distribution moves toward
greater concentration in the services sector, and declines in agriculture
and manufacturing.
The core empirical task is to compare the earnings growth of
workers who remained formally employed to that of those who
moved out of formal employment, conditional on the observable
characteristics of workers and their initial jobs. Consider a labor
market in which there are two types of jobs, denoted by 0 and 1. In
the first period, all workers hold jobs of type 0. In the second period,
some of these workers are found to have moved to jobs of type 1. Each
worker possesses a set of general characteristics X with returns α,
and also an endowment of job type-specific human capital denoted by
vector A = (a0, a1).
The first-period earnings of a worker i are determined by:

Yi0 = β + X i′ α + θ0 a0 + ei0 . (1)

 change, both the returns to general characteristics,

Due to structural
α, and returns to specific human capital, a0, can change. In the second
̅ ′ ̅ j aj + eij .
period, worker i’s Yij = β +
earnings in X iα
job ̅j +
areθgiven by:

dYi = (β̅ − β) + X i′ (α
̅ − α) + (θ̅0 − θ0 )a0 + vi0 ,


Export Boom, Employment Bust?Y
The
i0 Paradox X i′ α + θDisplaced
= β +of Indonesia’s 0 a0 + Workers,
ei0 . 2000–2014 153

Yi0 = β + X i′′α
+ θ0 a0 + ei0 .
Yi0 = β + Xi α
+ θ0 a0 + ei0 .
 Yi0 =Yijβ=+β̅ X+ ′ ′ θ a ̅ + e .
 i α X+
iα
̅ + 0 0θj aj +i0eij . (2)
̅ ′
Yij = β̅ + Xi′ α ̅
̅ + θ̅ j aj + eij .
Yij = βif individual
Therefore, + Xi α ̅ + iθstays
j aj +inejob
ij . 0, the difference in earnings over
time is:
β̅ +=Xi′(β
Yij = dY ̅̅ +− θ̅β)
α j aj++Xe
′ .
̅ − α) + (θ̅0 − θ0 )a0 + vi0 ,
 i i ij(α
dYi = (β̅̅ − β) + X i′ (α
′
̅ − α) + (θ̅̅ 0 − θ0 )a0 + vi0 , (3)
dYi = (β − β)  + X i (α ̅ − α) + (θ0 − θ0 )a0 + vi0 ,
and if worker idY
switches
i =dY(β̅to−job,
̅ it+is:X ′ (α
β) ̅ − ̅ − α)̅ 0+−θ̅1θa01)a−
′ α) + (θ 0 +
θ0 av0i0 ,+ vi1 .
i = (β − β)i + X i (α
dYi = (β̅̅ Yi0− =β)β ++ X
X i′′α(α
̅ ̅. a − θ a + v .
+ θ−0 a0α)++ei0θ
dYi = (β − β)  + X ii′ (α ̅ − α) + θ̅11 a11 − θ00 a00 + vi1
i1 .(4)



The differential i =
dYearnings D (β̅= −
Yij = β̅ i+ Xgrowth
̅β) +−X i′θ(α
′ (θ1 a̅1 between
̅
α
̅a −) −α)(θ
+ θj aj + e0ij . 0 switchers
̅ 0 θ̅−1and
+ aθ1 0− )a θ a + v .
0 .0 0is thusi1
stayers
i
givenDby:= (θ̅ a − θ a ) − (θ̅ − θ )a .
Dii = (θ̅11a11 − θ00 a00 ) − (θ̅00 − θ00 )a00 .
 ̅̅1− ̅ |S
 DdYi i==E(a (θ(β 0a|S1iβ)−=+ θ1) =)−−
X0i′ a(α
̅
0 α)(θ
E(a 0 0 i−
+ 0 )a00. + vi0 , (5)
0θ)a
(θ̅0=−θ0)ǡ

E(a0 |Si =  1) = E(a0 |Si = 0)ǡ

The
E(a0second
|Si = 1) term = inE(a equation
0 |Si = (5) 0)ǡmeasures the change in returns
to workers
 who dYremain
i = (β̅ in
− job
β) +0.X i
′The
(α
̅ −first
α) +term
θ̅1 a1captures
− θ0 a0 +the vi1effects
. on
 E(a 0 |S i = 1) = E(a 0 |S i = 0)ǡ
workers who are displaced to job 1. For this group, not only do returns to

human capital change, but the value of their human capital may differ in

the new occupation.
Di = On(θ̅1 athe
1 −other
θ0 a0 )hand,
− (θ̅0 the
− θsecond
0 )a0 . term is hypothetical
for those displaced from job 0. We can estimate this term from data on

stayers if the unobserved characteristics are similar for switchers and
stayers, that is, E(a0 |Si = 1) = E(a0 |Si = 0)ǡ, where Si = 1 if the individual
switches employment and 0 otherwise.

In general, there are reasons to believe that switchers will be
different from stayers. If the transition is voluntary, Roy’s (1951) self-
selection model implies that individuals select into occupations in which
the returns to their skills are highest. Thus, the earnings potential of
switchers may be different than that of stayers. For example, those with
high a0 may be more likely to stay in job 0 if growth in returns to skills
is positive. These unobserved skills might drive both switching out of
formal employment and earnings conditional on switching. This means
that the observed earnings of switchers provides a biased estimate of the
potential earnings of nonswitchers.
On the other hand, if job changes are involuntary, workers with
a low a0 might be the first to be laid off from jobs of type 0, which
means that the expected a0 among switchers and stayers is different.
At this time—and in contrast to the much richer datasets available for
displacement studies on wealthy countries—we lack the data needed
154 Trade Adjustment in Asia: Past Experiences and Lessons Learned

to provide an adequate resolution to this selection issue.6 What we do

know, however, is that among workers who self-report displacement
from a formal job (defined here as earning a salary), the rate of formal
employment in a subsequent survey round is much lower than for
workers who do not switch, and median hourly earnings, while still
much higher than for workers who were never formally employed,
are 20%–30% lower than for workers who did not switch. Adding to
this the likely loss of job security, fringe benefits, and better-regulated
working conditions, it seems plausible that among blue-collar workers
with formal employment, only an exceptional few would voluntarily
choose to move to informal employment. We analyze these data in
greater detail later in section 5.5.
In translating these concepts into statistical analysis, we focus on
workers aged 20–53 years old who are employed formally in the baseline
year (2000 or 2007, analyzed separately) and analyze their labor market
outcomes in the next survey wave. The reason for focusing on those
aged 20–53 in the baseline year is to minimize the possibilities of job
transition through retirement after age 60. We still have to worry about
the issues related to the endogenous labor force participation of women.
We classify workers as formal if they are private employees (thus
excluding government workers) who work in firms comprising at least
five workers. Larger firm size is associated with greater productivity
and compliance with labor market regulation, including job stability
and benefits. However, a heavy concentration of small firms with low
productivity is a feature of developing countries, including Indonesia
(Hsieh and Olken 2014). In such a setting, displacement can easily
result in a large loss in earnings. In the IFLS, the highest rates of
formal employment among low-skilled occupations are found in the
manufacturing sector (Table 5.2).
We focus on formal workers as these are the individuals who are
most obviously vulnerable to displacement. In the data, formal workers
earn at a higher rate and are more likely to receive benefits such as
medical insurance (Table 5.3). This is true even if we focus only on
employees (thus excluding the self-employed) or on specific sectors
such as manufacturing. Therefore, movement out of formal jobs is more

6
Systematic selection of displaced workers is an important issue in the job
displacement literature. One way of tackling this issue is to include worker-specific
time trends to account for unobserved worker characteristics that evolve linearly
over time (Jacobson, LaLonde, and Sullivan 2005; Couch and Placzek 2010). The US
literature finds that point estimates on earnings losses are slightly smaller, but not
statistically different when using matching estimators that compare workers with a
similar ex ante probability of being displaced (Couch and Placzek 2010).
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 155

Table 5.2 Share of Formal Jobs by Occupation and Sector (%)

(a) Share of Formal Jobs by Occupation
2000 2007 2014
Professional 26 29 36
Admin 25 24 51
Clerical 41 44 52
Sales 11 12 13
Service 17 19 24
Agriculture 10 09 13
Production 39 39 48
Semi-skilled 45 53 58
Laborer 31 35 39
Others 30 27 37
N 14,899 18,174 21,819

(b) Share of Formal Jobs by Sector

2000 2007 2014
Agriculture 10 10 12
Mining 48 34 48
Manufacturing 43 46 56
Utilities 24 41 54
Construction 37 43 46
Wholesale/retail trade 11 12 17
Transport 17 20 32
Financial services 61 63 49
Social services 24 26 27
N 14,896 18,174 21,819
Note: A worker is classified as formal if they are private employees in firms with five or more workers.
Source: Authors’ calculation from various Indonesian Family Life Survey (IFLS) rounds. Sample includes
workers aged 20–65 years in each survey. The IFLS did not collect firm size information to measure
formality in 1993 and 1997.

likely to be involuntary. Furthermore, workers in informal employment

in the initial period are very likely to remain so in the subsequent
period as well. Between 2000 and 2007, only 10% of informal workers
transitioned into formal employment.
156 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.3 Characteristics of Formal Jobs—Earnings and Benefits

2000 2007
Informal Formal Informal Formal
Med. log hourly earnings 7.13 7.38 7.96 8.34
Job benefitsa
Housing benefits 4% 6% 4% 6%
Car 2 4 1 3
Transport allowance 9 22 7 17
Health expense 12 30 10 22
Insurance policy 04 22 8 24
Clinic 05 22 6 15
Credit 26 38 21 28
Number of observations 9,695 11,732
a
Benefits questions were only asked of employees (N=4,910 in 2000 and 6,646 in 2007).
Notes: Informal excludes government employees. Number of observations pertains to earnings data.
Source: Authors’ calculation from the Indonesian Family Life Survey 2000 and 2007.

The IFLS employment module asked individuals to report, among

other things, sector, occupation, hours worked per week, number
of weeks worked per year, the number of workers in their place of
employment, and yearly earnings. For our purposes, the main outcome
of interest is growth of hourly earnings. We calculate this by dividing
yearly salaries for wage workers (or yearly profits for self-employed
workers) by the total number of hours worked in a year. We use annual
rather than monthly earnings, because the latter could fluctuate on a
seasonal basis.
A major empirical challenge in the study of displacement is
constructing counterfactual earnings for displaced workers. Some
studies, especially those with richer datasets, have used the workers’
predisplacement wage trajectory and information on nondisplaced
workers (Carrington and Fallick 2014). However, due to selection,
choosing an appropriate comparison group is still a challenge. Couch
and Placzek (2010) use a panel estimator with individual fixed effects
and propensity score matching, but this is a data-intensive approach not
feasible in our study.
We use two different regression approaches. In the first, we regress
log earnings per hour on lagged log earnings, formality status, and
other control variables that measure workers’ human capital in terms
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 157

of education, gender, and tenure. The inclusion of lagged earnings,

while not uncontroversial econometrically, allows us to control for all
human capital rewarded in the market. In the second approach, we
regress change in log earnings over time on formality status, adding
other controls. Both approaches provide similar results but with slightly
different point estimates.
In the earnings regression, we control for workers’ variables
measured in 2000, including age (dummies for bins of age), educational
status (dummies for less than elementary, junior, senior, and tertiary),
and occupation in the baseline year. Standard errors are robust to
heteroskedasticity.
To address endogeneity due to the unobserved ability of workers
correlated with formality status in the subsequent year, we check the
robustness of our estimates across multiple specifications.

5.5 Results
5.5.1 Descriptive Statistics for the 2000–2007 Sample

In the 2000 survey, there are 2,750 formal workers who meet the
demographic criteria and report positive earnings (1,773 men and 977
women). This constitutes 25.7% of all workers in the age group. The rate
of formality varies greatly by sector and education level. Mining (54%),
manufacturing (49%), and financial services (62%) have the highest
rates of formal employment. Likewise, more educated workers are more
likely to hold formal jobs: their formal employment rate is 35%, against
just 19% among those with no more than primary education.7
Table 5.4 shows the distribution of formal workers by occupation,
sector, and education level for workers who meet our sampling
criteria. Most formal workers are performing low-skilled and semi-
skilled tasks—about 45% of formal workers fall within the last three
occupational categories, broadly described as “production and related
workers, transport operators, and laborers.” Taking advantage of the
two-digit classification of occupations in the IFLS, we find that the
largest shares of male formal workers are in construction (11.3% as
“stone-layers, carpenters, and other building workers”), transportation
(6.3% as “sea and land transportation workers”), and plantation

7
The informal employment rate among the highest educated workers in our dataset
is biased upward by inclusion of some government employees classed as being in
informal employment. Excluding government employees from the sample increases
the formality rate among tertiary educated workers from 30% to 46%.
158 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.4 Distribution of Formal Workers

by Occupation, Sector, and Education Level in 2000

Occupation % Sector % Education %

Professional 5.93 Agriculture 15.74 ≤ Primary 36.74
Clerical 0.51 Mining 1.35 Junior high school 16.08
Admin 9.17 Manufacturing 30.86 Senior high school 33.90
Sales 9.03 Utilities 0.36 ≥ College 13.28
Service 11.98 Construction 8.49
Agriculture 15.51 Wholesale/ 12.35
retail trade
Production 16.89 Transport 3.68
Semi-skilled 6.99 Financial services 2.48
Laborer 21.15 Social services 24.70
Note: Total number of observations is 2,749.
Source: Authors’ calculation from the Indonesian Family Life Survey 2000.

agriculture (6.2% as “agriculture and animal husbandry workers”).

For females, the most common formal occupation is plantation work
(17.4%), followed by sales (9.7%), tailoring (7.4%), and teaching (6.9%).
A relatively large share of formal work created in the agriculture sector
is explained by the development of large plantation estates for oil palm
production. It also seems that sources of formal jobs are quite distinct
for males and females, indicating some segregation in the labor market
by gender.
The other occupational category includes service workers (e.g.,
maids, barbers, and housekeepers). This is also confirmed by looking
at the sectoral distribution of these workers; the majority are employed
in manufacturing and social services. Within manufacturing, food
and beverage processing workers are the most prevalent, comprising
over 16% of formal workers in the manufacturing sector. In terms of
education, there seem to be two modes: those with primary or less
education and those with senior high school-level education. This
indicates heterogeneity in skill requirements within the formal sector,
with some formal work requiring little education. These low-skilled
formal jobs may contribute greatly to poverty reduction, but they are
also likely to be more vulnerable to a slowdown in economic growth.
The IFLS 2007 contains detailed labor market information for
2,140 of these individuals, of whom 1,892 report being employed.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 159

We lose some observations because entire households could not be

tracked and others because they had moved out of the household. As
a result, we have earnings information for 2,130 individuals, reducing
the sample by 23%.8
The main explanatory variable of interest is formal status in 2007. We
create a dummy variable taking the value 1 if displaced and 0 otherwise.
Unlike existing work on job displacement, which uses administrative
data, we do not know directly whether an individual changed employers
between 2000 and 2007.9 The results should be interpreted accordingly.
We estimate earnings loss associated with moving out of formal
employment.
Transition out of formality is large. Just over 42% of workers are
still classified as formal in 2007, while more than 50% are classified as
employees. The transitions are slightly different for males and females.
Men transition into self-employment and casual nonagricultural
employment to a greater extent than other job types, whereas women
transition into unemployment and unpaid family work.
Table 5.5 reports the share of formal workers in 2000 who were
still formal in 2007, by sector and education. Those in manufacturing
in 2000 had a slightly greater chance of staying formally employed than
in other sectors besides financial services (which is relatively small).
Formal workers in wholesale or retail trade and social services in 2000
are also more likely to be found in either informal or formal work in 2007.
In contrast, agricultural workers were more likely to transition from
formal to informal employment. In terms of educational achievement,
all those who transitioned out of formal employment had primary or
lower levels of education; formal employment rates were higher for all
higher education levels.
The heterogeneity within formal jobs is again apparent in the
variation in transition rates across different demographics.

8
Attrition could lead to some issues. The distribution of workers in the baseline
year, divided by those in and out of the sample in the subsequent year, shows
some divergence. This is reported in Appendix Table A5.1. Those not in the 2007
sample are more likely to come from the agriculture sector. Younger workers are also
more likely to be missing, possibly due to migration. A similar pattern of attrition is
evident between 2007 and 2014, shown in Appendix Table A5.2. Similarly, those with
junior- or college-level education tend to be missing from the sample. This may lead
to possible bias in the estimates, but the direction of this bias is a priori uncertain.
9
Each IFLS round contains a module on employment history for the previous 8
years based on recall data. Due to the lack of information necessary to construct our
formality measure, we do not use this information.
160 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.5 Likelihood of Staying Formally Employed

by Education Level and Sector in 2000

Sector in 2000 N % Formal Education N % Formal

Agriculture 340 31 ≤ Primary 841 38
Mining 30 33 Junior high school 333 38
Manufacturing 683 47 Senior high school 721 49
Utilities 9 44 ≥ College 244 47
Construction 188 40
Wholesale/retail 269 46
Transport 71 42
Financial services 40 60
Social services 507 41
Note: Total number of observations is 2,137.
Source: Authors’ calculation from Indonesian Family Life Survey 2000 and 2007.

5.5.2 Descriptive Statistics for the 2007–2014 Sample

We now examine trends for those formally employed at the later baseline,
2007. There are 3,569 such individuals in the sample, comprising 27.5%
of total workers fulfilling the demographic criteria. Similar to 2000,
formality rates are higher in manufacturing (54%), among those with
senior high school-level schooling (36%), and for those aged 20–29 years
(36%). The earnings differential between formal and informal workers
is slightly higher in 2007 than in 2000, ranging from a 0.26 log difference
in hourly earnings among those aged 20–29 to 0.46 among those
aged 40–53.
Table 5.6 presents the distribution of formal workers meeting our
sampling criteria by occupation, sector, and education level. Compared
with 2000 (Table 5.3), we note that the construction and wholesale
or retail trade sectors comprise a greater share of formal jobs in 2007.
Analyzing the distribution of formal workers at two-digit occupation
levels, we find that while construction still accounts for the largest share
of formal workers in 2007 (12.7%), the second largest formal occupation
is now sales (7%). For females, plantation work (12%), teaching (9.6%),
and sales (9.4%) are the top three formal occupations. The increase in
education composition is also noticeable. Formal workers comprise a
greater proportion of senior high school- and college-educated workers.
Once again, we lose a significant portion of this sample between
survey waves. Out of the 3,569 workers in 2007, 2,512 are still employed
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 161

Table 5.6 Distribution of Formal Workers

by Occupation, Sector, and Education Level in 2007

Occupation % Sector % Education %

Professional 7.23 Agriculture 13.17 ≤ Primary 27.29
Clerical 0.34 Mining 1.01 Junior high school 15.77
Admin 8.85 Manufacturing 29.48 Senior high school 38.64
Sales 11.24 Utilities 0.48 ≥ College 18.30
Service 13.51 Construction 9.53
Agriculture 12.75 Wholesale/retail 14.54
Production 14.35 Transport 3.25
Semi-skilled 7.82 Financial services 2.63
Laborer 21.49 Social services 25.92
Note: Total number of observations is 3,569.
Source: Authors’ calculation from Indonesian Family Life Survey 2007.

in 2014, and we do not have information on 734. Among those with job
status information, 57% are reported as working in the formal sector,
a much larger proportion than in the 2000–2007 sample.10 This could
partly be due to higher levels of reported formal employment overall in
2014: among males aged 20–53 years, 34.9% are formally employed, a
much higher share than in 2000 or 2007.
The importance of education in maintaining formal employment
status becomes clear by comparing the last two columns of Table 5.7.
Less-educated workers are less likely to continue in formal employment
than high educated workers. The difference between the formality rate
of the lowest and highest educated workers is over 15 percentage points.
In 2000–2007, this difference was just under 9 percentage points.
Our analysis suggests that attrition could lead to some potential
issues. In Appendix Tables A5.1 and A5.2, we report the shares of
worker status (unemployed, employed, missing) in the subsequent
year by sector and education levels of the workers for the baseline
years 2000 and 2007, respectively. We find that workers in skill-
intensive sectors, those with higher education, are missing to a greater
extent. For example, in both samples, over 30% of higher-educated
workers are missing from the sample. These workers are likely to have
maintained formal jobs had they been in the sample. This may lead to

10
In comparison, about 16% of workers transitioned from nonformal to formal work.
162 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.7 Likelihood of Staying Formally Employed

by Education Level and Sector in 2007

% %
Sector in 2007 N Formal Education N Formal
Agriculture 343 45 ≤ Primary 733 48
Mining 26 46 Junior high school 407 55
Manufacturing 740 63 Senior high school 955 64
Utilities 12 75 ≥ College 416 63
Construction 261 49
Wholesale/retail 350 59
Transport 84 49
Fin. services 58 53
Social services 637 64
Note: Total number of observations is 2,511.
Source: Authors’ calculation from the Indonesian Family Life Survey 2007 and 2014.

possible bias in the estimates, but the direction of this bias is a priori
uncertain.

5.5.3 Earnings Function Estimates

We estimate the earnings equations separately for 2000–2007 and for

2007–2014. The observations concern workers aged 20–53 years who
were formally employed in the base year (either 2000 or 2007) and have
non-zero earnings in the survey year (either 2007 or 2014). The variable
of interest is the indicator for formal employment in the later year. This
takes the value 1 if an individual is still in formal employment and 0
otherwise.
We use two different dependent variables. In one specification,
we use the log of earnings in the current year (2007 or 2014) as the
dependent variable. In this set of models, we also control for lagged
earnings (2000 for the 2007 model, 2007 for the 2014 model). In the
other specification, we use the change in log earnings over time as the
dependent variable.
The results are shown in Tables 5.8 and 5.9. In each table, we report
first a basic model without controls other than for the log of baseline
earnings, then add age, sex, education controls and region fixed effects,
and finally occupation controls. Choice of occupation is arguably
endogenous, so among the three models, the second is to be preferred
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 163

on a priori grounds. The last three columns of the tables show results in
first-difference form.
The results are quite consistent across models. The dummy for
formal employment in the survey year is very precisely estimated in
each case, and its coefficient values diminish only slightly with the
addition of controls. As we use a semilog specification with a dummy
variable, the elasticity of earnings in the survey year with respect to
formal employment status is calculated as eb – 1, where b is the coefficient
estimate. For values in the range of our estimates, the elasticities are
slightly larger than the coefficient estimates. In Table 5.8, model (2),
formal status is associated on average with per hour earnings 25%
higher than those of informal status (e0.224 – 1 = 0.25). In the differenced
version (model 5), the elasticity is 0.22. In Table 5.9, model (2), the same
elasticity is 0.4, or a 40% premium.
Comparing these results with those from previous studies, we can
draw a few tentative conclusions. First, the estimated magnitudes of
earnings loss are comparable with those from studies of involuntary
displacement. Studies of developed economies, with access to annual

Table 5.8 Impact of Displacement on Earnings in 2007

Hourly Earnings (logs) Difference

(1) (2) (3) (4) (5) (6)
Formal 0.260*** 0.224*** 0.210*** 0.198*** 0.172*** 0.164***
(0.0546) (0.0536) (0.0535) (0.0596) (0.0603) (0.0605)
Lagged log 0.467*** 0.300*** 0.288***
earnings per hour
(0.0366) (0.0352) (0.0347)
Constant 4.695*** 5.675*** 5.870*** 0.790*** 0.958*** 0.928***
(0.273) (0.258) (0.288) (0.0454) (0.109) (0.167)
Demographic No Yes Yes No Yes Yes
variables
Education No Yes Yes No Yes Yes
Occupation No No Yes No No Yes
Region No Yes Yes No Yes Yes
Observations 1,761 1,760 1,758 1,761 1,760 1,758
Notes: Robust standard errors in parenthesis. Sample includes workers in 2007 with positive earnings
who were aged 20–53 years and formally employed in 2000. Age dummies include 30–39 and 40–53
with 20–29 as base group. Region dummies include indicators for Java, Kalimantan, Sulawesi, and Papua,
with Sumatra as base group. Demographic variables include male dummy, age dummies, and education.
Occupation variables includes indicators for nine categories of occupation. * p<.1, ** p<.05, *** p<.01.
Source: Indonesian Family Life Survey 2000 and 2007.
164 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.9 Impact of Displacement on Earnings in 2014

Hourly Earnings (logs) Difference

(1) (2) (3) (4) (5) (6)
Formal 0.343*** 0.313*** 0.301*** 0.257*** 0.258*** 0.343***
(0.0646) (0.0642) (0.0643) (0.0679) (0.0697) (0.0646)
Lagged log 0.429*** 0.309*** 0.291*** 0.429***
earnings per hour
(0.0379) (0.0376) (0.0381) (0.0379)
Constant 5.278*** 5.848*** 6.419*** 0.616*** 0.705*** 5.278***
(0.305) (0.319) (0.361) (0.0607) (0.123) (0.305)
Demographic No Yes Yes No Yes Yes
variables
Education No Yes Yes No Yes Yes
Occupation No No Yes No No Yes
Province No Yes Yes No Yes Yes
dummies
Observations 2,336 2,336 2,336 2,336 2,336 2,336
Notes: Robust standard errors in parenthesis. Sample includes workers in 2007 with positive earnings
who were aged 20–53 years and formally employed in 2000. Age dummies include 30–39 and 40–53
with 20–29 as base group. Region dummies include indicators for Java, Kalimantan, Sulawesi, and Papua,
with Sumatra as base group. Demographic variables include male dummy, age dummies, and education.
Occupation variables includes indicators for nine categories of occupation. * p<.1, ** p<.05, *** p<.01.
Source: Indonesian Family Life Survey 2007 and 2014.

data, typically show a sharp earnings drop following displacement,

followed by a partial recovery over several subsequent years. Our
survey waves are 7 years apart, and we do not know with any precision
when in that interval each worker changed jobs or employment status.
Therefore, our results are best understood as an average of short- and
long-term impacts.
Second, displaced workers in developed economies typically have
access to at least partial income insurance through unemployment
benefits or other social safety net instruments. Because of this, estimates
of wage declines are likely to be greater than the change in actual income.
These mechanisms do not apply in Indonesia, or at least not through
official channels, although households may engage in less formal sharing
behaviors. Thus, our estimates are likely to be closer to actual income
changes than seen in developed economy studies.
Third, although we do not as yet have the means to decompose
our results, our estimates of earnings differences in the survey year
must reflect a combination of factors. Among these are losses due to
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 165

frictional or cyclical unemployment; losses due to reduced returns to

sector-specific human capital, and losses due to movement across the
extensive margin of formal labor market institutions. Because our
data are from an economy undergoing rapid expansion, the first type
of earnings loss is likely to be small relative to findings from developed
economies. Moreover, since most of the workers in our sample are blue-
collar wage earners with little formal education, we can speculate with
some confidence that the second type of loss is also small relative to
other studies. If so, we may hypothesize that the third institutional effect
is a more important driver of observed earnings differences. If this is
confirmed, it sharply refocuses attention on labor market policies, since
the more stringent of these may cause employers to limit their offers
of formal labor contracts (as has been argued for Mexico; see Hanson
2010). This is a topic for deeper investigation in the future.

5.6 Self-reported Displacement and Earnings

The foregoing analysis yields estimates of an average earnings effect
of movement from the formal to informal sector, but we are unable
to establish a rigorous chain of causality. In this section, we explore
an alternative approach. IFLS 2007 and 2014 asked some detailed
questions about an individual’s job history over the preceding 5 years
The surveys asked whether the individual held any salaried positions
over the previous 5 years and, if yes, whether he or she had been fired
or had quit. The survey also inquired about the year of the latest job
termination and the primary reason for termination.
Based on the responses to these questions, we can get one step
closer to identifying exogenous job displacement and its impact on
earnings. We create a categorical variable “displacement status” to
indicate the status of the workers: (1) never held a salaried position; (2)
held a salaried position that terminated due to firing;11 (3) held a salaried
job that terminated due to other reasons; and (4) held a salaried position
and never terminated. This variable is closer to the true notion of
displacement that has been used in the current literature as it captures
separation from an employer. The goal is to relate this variable to
differences in earnings.

11
The list of possible reasons included in the surveys are as follows: fired by the
company because the business was closed down/relocated/restructured; fired for
another reason; wage/salary was too low; nonconducive working environment;
refused to be relocated; prolonged sickness; marriage; childbirth; other family
reason. We consolidate workers experiencing displacement into “fired”
(first two reasons) and “other reasons.”
166 Trade Adjustment in Asia: Past Experiences and Lessons Learned

To keep these results comparable with the earlier work, we focus on

workers aged between 27 and 60 years in the survey year. This keeps the
sample’s age consistent with the earlier analysis. Second, our estimation
sample is limited to workers who are (1) currently formal and never
displaced (assuming they were also formal before),12 or (2) displaced
from a formal job, for which we use the same definition of formality as
before.13
One issue is that earnings information from previous jobs is asked
only of those who report being displaced, which means we cannot
compare earnings growth due to lack of information on the past earnings
for workers who are never displaced. We address this issue by merging
information from the previous survey for workers who appear in both
years. The second caveat is that workers are displaced at different
times within the previous 5 years. Due to the small sample, we pool all
workers who are displaced into the aforementioned categories without
distinguishing the timing of displacement.
Before exploring the earning patterns, we look at worker
characteristics and labor market status by displacement status for
all workers. These are shown in Tables 5.10 and 5.11. The first three
columns in each table report demographic characteristics, the next
column employment status, and the final column median earnings.
Our 2007 sample comprises 2,593 individuals,14 of whom almost 10%
reported displacement due to firing or business closure. Our 2014 sample
comprises 4,275 workers, of whom 220 (5%) reported being displaced
involuntarily and 1,040 reported voluntary job changes.
Tables 5.10 and 5.11 show that gender and education are highly
correlated with displacement status. The sample of workers reporting
involuntary displacement predominantly comprises men. Those with
tertiary education tend to have stable salaried jobs or voluntarily switch
employment. Only 14% of workers involuntarily displaced have tertiary-
level education compared to over 21% for the other two categories.
Furthermore, those experiencing displacement present less than 60%

12
This may not be exactly accurate as the size of these workers’ firms may have
expanded over time.
13
The workers who reported being displaced were asked about the size of their firm
and whether they worked in the government or private sector.
14
To arrive at the respective sample for each year, we start with the individuals who
appeared in the displacement module. We exclude workers who did not fall into
our age range or those who never held salaried jobs. If the worker reported being
displaced, we check to ensure that they were nongovernment workers employed at a
firm with at least five workers. If they reported never being displaced, we check their
current formality status and remove those currently with informal status.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 167

Table 5.10 Characteristics of Workers

by Displacement Category in 2007

Tertiary Median
Age Male Education Formal Earnings
Displaced last 5 years – fired 36.53 0.81 0.13 0.52 3,571.43
Displaced – other reason 34.54 0.74 0.25 0.50 3,900.00
Not displaced 36.45 0.67 0.20 1.00 4,813.16
N 2,593
Notes: Sample includes workers aged 27–60 years. For earnings, only those with non-missing earnings
information and nongovernment workers are included.
Source: Authors’ calculations from the Indonesian Family Life Survey 2007.

Table 5.11 Characteristics of Workers in 2014

by Displacement Category

Tertiary Median
Age Male Education Formal Earnings
Displaced last 5 years - fired 37.45 0.80 0.14 0.59 7,694.13
Displaced - other reason 35.25 0.73 0.22 0.58 9,209.04
Not displaced 37.50 0.63 0.21 1.00 9,558.82
N 9,987 9,263 8,470
Notes: Sample includes males aged 25–60 years. For earnings, only those with non-missing earnings
information are included.
Source: Authors’ calculations from the Indonesian Family Life Survey 2014.

formal employment in both years. The current formal employment rate is

similar for workers who were displaced, regardless of the stated reason.
Conditional on employment, median earnings are much higher than the
median earnings of workers who have never experienced displacement.
Comparing these statistics across years, we find a greater share
of formal workers reporting involuntary displacement in 2007.
Furthermore, the median earnings of those changing jobs voluntarily
are very similar to those never displaced in 2014. However, it is not clear
whether this suggests an improvement in the health of the Indonesian
labor market as the sample is highly self-selected and not representative
of the Indonesian economy in these years.
In Figure 5.5 we look at the cumulative distribution of log hourly
earnings by displacement status in 2007. While continued employment
dominates displacement at the lower end in terms of expected earnings,
168 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 5.5 Distribution of Log Hourly Earnings

by Displacement Status in 2007

1.0

0.8

0.6
CDF

0.4

0.2

0.0

0 5 10 15 20
Log hourly earnings
No salaried job Displaced - other reason
Displaced last 5 yrs - Fired Not displaced

CDF = cumulative distribution function.

Source: Authors’ calculation from the Indonesian Family Life Survey 2007.

there is a considerable overlap in the distribution at the upper tail of the

graph. This is illustrative of the two broad types of individuals who make
up the displaced sample, namely a very small number who left their
original employment for a better opportunity (or who were exogenously
displaced but “landed on their feet”), and those forced or sorted into
lower earnings work. Similar patterns are also evident in 2014, as shown
in Figure 5.6.

Figure 5.6 Distribution of Log Hourly Earnings

by Displacement Status in 2014

1.0

0.8

0.6
CDF

0.4

0.2

0.0

0 5 10 15 20
Log hourly earnings
No salaried job Displaced - other reason
Displaced last 5 yrs - Fired Not displaced

CDF = cumulative distribution function.

Source: Authors’ calculation from the Indonesian Family Life Survey 2014.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 169

We now analyze the impact of displacement status on earnings in a

regression setting. The dependent variable is the log of hourly earnings,
and the main explanatory variable is displacement status. As in the
previous section, we control for human capital variables, including
gender, age categories, and education, and lagged earnings information
from the preceding survey (2000 or 2007). Merging with previous
surveys inevitably leads to loss of some observations as they do not
appear in these surveys.
Tables 5.12 and 5.13 report the results from the 2007 and 2014
samples, respectively. In 2007, we find that those with stable formal jobs
had greater earnings compared to those involuntarily displaced. In this
case, the results are robust to the inclusion of additional controls and
alternative dependent variables. In 2014, the results are similar when
we use the estimation with lagged earnings as one of the controls, but
the statistical significance of the results disappears in the difference
model. This could again indicate some improvement in labor market
conditions and post-Asian financial crisis recovery. However, we still
need to interpret the results with caution due to the small sample size.

Table 5.12 Regression Results from

Self-Reported Displacement, 2007

(1) (2) (3) (4)

Displaced – other reason 0.225 0.150 0.261 0.223
(0.155) (0.157) (0.160) (0.163)
Not displaced 0.473*** 0.454*** 0.425*** 0.394***
(0.130) (0.131) (0.135) (0.138)
Log earnings 2000 0.494*** 0.333***
(0.0375) (0.0426)
_cons 4.344*** 5.175*** 0.649*** 0.762***
(0.310) (0.340) (0.132) (0.178)
Human capital variables No Yes No Yes
Region dummies No Yes No Yes
Observations 1,519 1,519 1,519 1,519
Notes: Robust standard errors in parenthesis. Sample includes workers in 2007 who were salaried 5
years before and aged 27–60 years. Age dummies include 40–49 and 50–60 with 27–39 as base group.
Region dummies include indicators for Java, Kalimantan, Sulawesi, and Papua, with Sumatra as base group.
Human capital variables include male dummy, age dummies, and education in 2007. * p<.1, ** p<.05,
*** p<.01.
Source: Indonesian Family Life Survey 2000 and 2007.
170 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 5.13 Earnings Growth and Displacement, 2014

(1) (2) (3) (4)

Displaced – other reason 0.320** 0.329** 0.292 0.265
(0.154) (0.149) (0.213) (0.214)
Not displaced 0.372** 0.414*** 0.241 0.246
(0.147) (0.143) (0.201) (0.203)
Log earnings 2007 0.195*** 0.139***
(0.0215) (0.0190)
_cons 7.185*** 7.142*** 0.747*** 1.074***
(0.226) (0.208) (0.198) (0.246)
Human capital variables No Yes No Yes
Region dummies No Yes No Yes
Observations 2,545 2,545 2,545 2,545
Notes: Robust standard errors in parenthesis. Sample includes workers in 2014 who were salaried 5 years
before and aged 27–60 years. Age dummies include 40–49 and 50–60 with 27–39 as base group. Region
dummies include indicators for Java, Kalimantan, Sulawesi, and Papua, with Sumatra as base group.
Human capital variables include male dummy, age dummies, and education in 2014. * p<.1, ** p<.05,
*** p<.01.
Source: Indonesian Family Life Survey 2007 and 2014.

5.7 Conclusions
Indonesia is a developing economy which by virtue of its specialization
in natural resources, minerals, and labor-intensive manufactures is
especially vulnerable to shocks from the global market. In the 2000s,
Indonesia experienced rapid growth and equally rapid structural
change, largely as a consequence of global market trends. Both the overall
growth of the economy and price-induced changes in the structure of
production led to large changes in the vitality and composition of labor
demand. Textbook models of economic growth and trade predict that
greater openness and more growth should increase labor demand and
productivity in low-income, labor-abundant economies. Yet in Indonesia,
an export boom and rapid GDP growth in the decade after 2000 was
accompanied by real wages that were flat on average, as well as declining
earnings for a large number of workers. This is likely because the source
of growth was not low-skilled manufacturing, as was the case before the
Asian financial crisis, but exports of natural resources.
We have explored these seemingly paradoxical trends using
individual employment data from the IFLS. We hypothesized that
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 171

observed trends in wages and earnings may be connected to involuntary

changes in sector or occupation. We found that the earnings of workers
with informal jobs who had held formal jobs in a previous survey round
were significantly lower relative to those of workers who remained
in the formal market, and that this effect appears to be much larger
than any effect due to changing returns to sector-specific skills. This
distinction adds an important developing country dimension to the job
displacement literature.
Our findings in this research add one piece to the puzzle of the
causes for a startling increase in inequality in Indonesia during the same
decade. In 2003–2013, Indonesia’s Gini coefficient for individual income
inequality rose one third, from 0.32 to 0.43 (Yusuf, Sumner, and Rum
2014). This rise has many possible causes, both related to income changes
and fiscal and other policies, but a formal decomposition of changes in
the Gini index has yet to be conducted. However, the magnitude of the
shift away from formal employment and the earnings drop experienced
by workers so displaced are undoubtedly a strong contributing factor.
The phenomenon of job displacement and what appears as “jobless
growth” during an economic boom may be a uniquely Indonesian
paradox, but the conditions in which they become possible are broadly
shared in the developing world. While displacement to informal
employment is almost certainly part of the explanation, the constraint
of working with a dataset that is designed for other purposes means that
the task of establishing a rigorous causal connection remains incomplete.
At an individual level, more work is needed to identify characteristics
that may predict job displacement. At the labor market level, the
apparent displacement of workers from formal sector jobs has several
possible causes. These include secular changes in economic structure,
trade-induced changes in industry-level activity and employment,
and domestic policy innovations. Identifying these and distinguishing
between them is an important task for future work.
172 Trade Adjustment in Asia: Past Experiences and Lessons Learned

References
Aswicahyono, H., H. Hill, and D. Narjoko. 2011. Indonesian
Industrialization: A Latecomer Adjusting to Crises. WIDER
Working Paper. No. 2011/53. Helsinki: United Nations University
World Institute for Development Economics Research.
Autor, D. H., D. Dorn, G. H. Hanson, and J. Song. 2014. Trade Adjustment:
Worker-Level Evidence. Quarterly Journal of Economics. 129 (4).
pp. 1799–860.
Autor, D. H., D. Dorn, and G. H. Hanson. 2016. The China Shock: Learning
from Labor Market Adjustment to Large Changes in Trade. Annual
Review of Economics. 8 (1). pp. 205–40.
Carrington, W. J., and B. C. Fallick. 2014. Why Do Earnings Fall with Job
Displacement? Federal Reserve Bank of Cleveland Working Paper
14-05.
Couch, K. A., and D. W. Placzek. 2010. Earnings Losses of Displaced
Workers Revisited. American Economic Review. 100 (1). pp. 572–89.
Coxhead, I. 2007. A New Resource Curse? Impacts of China’s Boom on
Comparative Advantage and Resource Dependence in Southeast
Asia. World Development 35 (7). pp. 1099–119.
Coxhead, I., and R. M. Shrestha. 2016. Could a Resource Export Boom
Reduce Workers Earnings? The Labor Market Channel in Indonesia.
Bulletin of Indonesian Economic Studies. 52 (2). pp. 185–208.
Dornbusch, R. 1988. Real Exchange Rates and Macroeconomics:
A Selective Survey. NBER Working Paper. No. 2775. Cambridge,
MA: National Bureau of Economic Research.
Garnaut, R. 2015. Indonesia’s Resources Boom in International
Perspective: Policy Dilemmas and Options for Continued
Strong Growth. Bulletin of Indonesian Economic Studies. 51 (2).
pp. 189–212.
Gindling, T. 2014. Self-Employment in the Developing World. World
Development. 56 (C). pp. 313–31.
Goldberg, P. K., and N. Pavcnik. 2005. Trade, Wages, and the Political
Economy of Trade Protection: Evidence from the Colombian Trade
Reforms. Journal of International Economics. 66 (1). pp. 75–105.
____. 2007. Distributional Effects of Globalization in Developing
Countries. Journal of Economic Literature. 45 (1). pp. 39–82.
Hanson, G. 2010. Why Isn’t Mexico Rich? NBER Working Paper.
No. 16470. Cambridge, MA: National Bureau of Economic Research.
Hsieh, C.-T., and B. Olken. 2014. The Missing “Missing Middle.” Journal
of Economic Perspectives. 28 (3). pp. 89–98.
Jacobson, L. S., R. J. LaLonde, and D. G. Sullivan. 1993. Earnings Losses of
Displaced Workers. American Economic Review. 83 (4). pp. 685–709.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 173

____. 2005. Estimating the Returns to Community College Schooling for

Displaced Workers. Journal of Econometrics. 125 (1). pp. 271–304.
Kletzer, L. G. 1998. Job Displacement. Journal of Economic Perspectives.
12 (1). pp. 115–36.
Korkeamäki, O., and T. Kyyrä. 2014. A Distributional Analysis of
Earnings Losses of Displaced Workers in an Economic Depression
and Recovery. Oxford Bulletin of Economics and Statistics. 76 (4).
pp. 565–88.
Manning, C. 2000. Labour Market Adjustment to Indonesia’s Economic
Crisis: Context, Trends and Implications. Bulletin of Indonesian
Economic Studies 36 (1). pp. 105–36.
Manning, C., and K. Roesad. 2007. The Manpower Law of 2003 and
Its Implementing Regulations: Genesis, Key Articles and Potential
Impact. Bulletin of Indonesian Economic Studies. 43 (1). pp. 59–86.
McMillan, M., and D. Rodrik. 2011. Globalization, Structural Change, and
Productivity Growth. NBER Working Paper. No. 17143. Cambridge,
MA: National Bureau of Economic Research.
Minnesota Population Center. 2017. Integrated Public Use Microdata
Series, International: Version 6.5 [dataset]. Minneapolis: University
of Minnesota.
Modestino, A. S., D. Shoag, and J. Balance. 2016. Downskilling: Changes
in Employer Skill Requirements over the Business Cycle. John
F. Kennedy School of Government, Harvard University: Faculty
Research Working Paper Series RWP16-014.
Rodrik, D. 2015. Premature Deindustrialization. Journal of Economic
Growth. 21 (1). pp. 1–33.
Rothenberg, A. D., A. Gaduh, N. E. Burger, C. Chazali, I. Tjandraningsih,
R. Radikun, C. Sutera, and S. Weilant. 2016. Rethinking Indonesia’s
Informal Sector. World Development. 80 (C). pp. 96–113.
Roy, A. D. 1951. Some Thoughts on the Distribution of Earnings. Oxford
Economic Papers. 3 (2). pp. 135–46.
Sjöholm, F. 1999. Productivity Growth in Indonesia: The Role of
Regional Characteristics and Direct Foreign Investment. Economic
Development and Cultural Change. 47 (3). pp. 559–84.
Strauss, J., K. Beegle, B. Sikoki, A. Dwiyanto, Y. Herawati, and F. Witoelar.
2004. The Third Wave of the Indonesia Family Life Survey (IFLS3):
Overview and Field Report. WR-144/1-NIA/NICHD.
Strauss, J., F. Witoelar, B. Sikoki, and A. Wattie. 2009. The Fourth Wave
of the Indonesian Family Life Survey (IFLS4): Overview and Field
Report. WR-675/1-NIA/NICHD.
Suryahadi, A., G. Hadiwidjaja, and S. Sumarto. 2012. Economic Growth and
Poverty Reduction in Indonesia before and after the Asian Financial
Crisis. Bulletin of Indonesian Economic Studies. 48 (2). pp. 209–227.
174 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Suryadarma, D., A. Suryahadi, and S. Sumarto. 2013. Sectoral Growth

and Job Creation: Evidence from Indonesia. Journal of International
Development. 25 (4). pp. 549–61.
World Bank. 2010. Indonesia Jobs Report: Towards Better Jobs and
Security for All. World Bank Working Paper 56348. Washington,
DC.
Yusuf, A. A., A. Sumner, and I. A. Rum. 2014. Twenty Years of Expenditure
Inequality in Indonesia, 1993–2013. Bulletin of Indonesian Economic
Studies. 50 (2). pp. 243–54.
Export Boom, Employment Bust? The Paradox of Indonesia’s Displaced Workers, 2000–2014 175

Appendix A5

Table A5.1 Distribution of Workers in 2000 across Sector

and Schooling by Their Presence in the 2007 Sample (%)
Sector
in 2000 Unemployed Employed Missing Education Unemployed Employed Missing
Agriculture 9.95 68.75 21.30 Primary 10.10 73.17 16.73
or less
Mining 2.70 78.38 18.92 Junior 10.63 64.71 24.66
high school
Manufacturing 12.63 68.00 19.36 Senior 8.37 68.99 22.64
high school
Utilities 0.00 90.00 10.00 College 5.75 61.10 33.15
Construction 4.29 76.39 19.31
Wholesale/ 11.80 67.55 20.65
retail trade
Transport 0.99 69.31 29.70
Financial 2.94 55.88 41.18
services
Social services 6.49 68.29 25.22
Note: Rows sum to 100%.
Source: Authors’ calculation from IFLS 2000 and 2007.

Table A5.2 Distribution of Workers in 2007 across Sector

and Schooling by Their Status in the 2014 Sample (%)
Sector
in 2007 Unemployed Employed Missing Education Unemployed Employed Missing
Agriculture 12.13 72.98 14.89 Primary or 10.68 75.36 13.96
less
Mining 0.00 72.22 27.78 Junior high 9.59 72.29 18.12
school
Manufacturing 11.22 70.44 18.35 Senior high 9.21 69.25 21.54
school
Utilities 0.00 70.59 29.41 College 5.82 63.71 30.47
Construction 3.82 76.76 19.41
Wholesale/ 10.60 67.44 21.97
retail trade
Transport 6.03 72.41 21.55
Financial 8.51 61.70 29.79
services
Social services 7.03 68.86 24.11

Source: Authors’ calculation from IFLS 2007 and 2014.

 PART III
Firm-level
Adjustments in Asia
 6
Firm Adjustment to Trade
Policy Changes in East Asia
Dionisius Narjoko and Shujiro Urata

6.1 Background
Trade and investment liberalization has been one of the key features of
economic policy in many developing countries since the 1990s. A new
understanding of the benefits of international trade triggered unilateral
tariff reductions from countries throughout the world. As a result, the
global economy in the early 21st century has seen significantly reduced
barriers, creating much larger trade volumes between countries. This
has promoted globalization, as the increasingly borderless countries
have nurtured the growth of production networks between countries.
It has also made exports an engine of growth and a strategy to foster
industrialization.
Economic literature on international trade closely follows
globalization, and research has consistently produced more evidence
on the benefits of globalization. Undertaking economic analysis of
globalization has been facilitated by access to more sophisticated or
detailed data (i.e., microdata at firm or plant level). In this context,
the recent theoretical literature on heterogeneous firms and trade has
emphasized a couple of new mechanisms through which changes in
trade policy (trade liberalization) increase aggregate productivity and
welfare. While this development has revolutionized our view of how an
economy responds to trade and trade policy changes, our understanding
is still only partial.
This chapter reviews some recent studies on the subject of firms
in a globalized economy to enable us to understand more about how
firms respond to globalization or changes in trade and investment
liberalization. It focuses on presenting or explaining the underlying
mechanisms through which the effects are realized. The studies
summarized in this chapter generally confirm the positive impact

179
180 Trade Adjustment in Asia: Past Experiences and Lessons Learned

of trade liberalization on productivity or the spectrum of measures

reflecting productivity, such as product quality, firm size, or skill
intensity. The positive impact goes through various channels, including
competition and industry dynamics, exporting and innovation decisions,
and production or investment decisions.
This chapter is organized by broad topics commonly adopted by
studies in the literature: productivity, competition, product dynamics,
technology and innovation, and product fragmentation. Table 6.1
provides a summary of key empirical findings organized by these topics.
The chapter concludes with a section on policy implications.

6.2 Productivity
Voluminous amounts of research have addressed the impact of
globalization on productivity. While the benefits of globalization
on productivity gains across sectors are relatively clear and well-
documented, little is known about the impact at the plant or firm level.
There is more variation on the impact when using more disaggregated
or micro-level data.
Recent theoretical developments in international trade allow us to
understand more about what happens regarding productivity change
within an industry when trade and investment liberalization occurs.
Departing from the standard trade models, the new wave of trade
models recognizes the impact of firm heterogeneity, particularly in
terms of productivity, within an industry (Pavcnik 2002). These models
point to the importance of firm dynamics (i.e., entry, exit, and growth of
the survivors) in shaping both aggregate- and plant-level productivity
change. In an environment with heterogeneous firms, trade and
investment liberalization induces the entry of more capable firms, forces
less-productive firms to exit, and triggers a reallocation of market share
toward more productive firms. The disappearance of less-productive
firms is reflected by an increase in the level of industry productivity (or
“between” firms’ productivity growth).
Trade and investment liberalization encourages firms to adopt
new technology to ensure their survival, either in domestic or foreign
markets. Firms, however, perceive such encouragement differently, as
some firms choose to adopt the new technology but others do not. In
other words, there is variation between firms, even within the same
industry, in responding to liberalization.
A new wave of theoretical developments underlines the importance
of firm, or plant, heterogeneity in shaping firms’ productivity within
an industry, pioneered by Melitz (2003). This developed from growing
evidence that the variation of exporting firms cannot be derived from
 Table 6.1 Summary of Key Findings

Productivity Exporting Competition

Competition, measured typically by dynamics of There is robust evidence on the self-selection Trade liberalization has a positive impact on
firms (i.e., firm entry, exit, and growth), increases hypothesis to exporting in the literature. One competition. One argument is due to the
productivity and improves resource reallocation implication of the hypothesis is the significant mechanism of imports as a competitive discipline,
(e.g., Liu 1993; Liu and Tybout 1996; Olley and difference between exporters and nonexporters. whereby greater trade inhibits domestic firms to
Pakes 1996; Aw, Chen, and Roberts 2001; and Bernard, Jensen, and Lawrence (1995) and conduct anticompetitive practices. Erdem and
Narjoko 2012). Bernard and Jensen (1999) documented that Tybout (2003) showed that trade liberalization
Aw, Chen, and Roberts (2001) found that exporters in manufacturing in the United States negatively affects the price-cost margins of
in a newly industrialized economy in Asia, (US) are larger, more productive, and more capital firms, which was also found by Harrison (1994)
new manufacturing firms have lower average intensive; pay higher wages; and employ more and Krishna and Mitra (1998) for the case
productivity than incumbents, although skilled workers than nonexporters. Sjoholm and of manufacturing in Côte d’Ivoire and India,
productivity varies significantly across the firms. Takii (2003) also observed that exporting plants respectively.
They also found that the more productive in Indonesian manufacturing are larger and more Further impacts are positive performance and
entrants survive, and their productivity converges productive; the labor productivity of these plants stronger innovation outcome, as was found by
to the level of incumbents. was about twice as high as nonexporting plants. Pavcnik (2002) for Chile and Amiti and Konings
Narjoko (2012) found a positive relationship Another implication is that firms prepare for (2007) for Indonesia. Trade also positively affects
between firm entry and industry productivity exporting. Bernard and Jensen (1999) found that innovation, since trade liberalization stimulates
growth in Vietnamese manufacturing. Rapid exporters in US manufacturing are more efficient, competition, which forces firms to become more
trade and investment liberalization occurring in are larger, and grow faster several years before efficient and productive through innovation.
Viet Nam since the early 1990s has substantially they become exporters. For the manufacturing Fernandes and Paunov (2009) showed that
reduced the cost of establishing private industry in the Republic of Korea, Aw, Chung, trade liberalization stimulates product quality
enterprises and of exporting, and it has triggered and Roberts (2000) found that the average upgrading in Chilean manufacturing, while Bloom,
rapid growth in a number of firms entering the productivity of continuing exporters and new Draca, and Van Reenen (2010) found a similar
country’s manufacturing and services sectors. entrants as exporters is significantly higher than relationship between trade liberalization and
There was a reallocation of resources across exiting exporters and nonexporters. innovation in the People’s Republic of China using
firms within Viet Nam's manufacturing toward patent, information technology (IT), research and
the more productive firms, which has resulted in development (R&D), and total factor productivity
higher industry-level productivity growth. (TFP) as the indicators. Aldaba (2012) found
Firm Adjustment to Trade Policy Changes in East Asia

for manufacturing in the Philippines that trade

liberalization increases competition in the domestic
181

market, and this forces firms to increase their R&D.

continued on next page
Table 6.1 continued
182

Productivity Exporting Competition

Choi and Hahn (2013) examined the relationship The alternative hypothesis, i.e., learning
between trade liberalization and productivity by exporting, is growing in terms of evidence
at firm and product level. They found that the collected by cases around the world. For
increase in intermediate input variety via trade economies in East Asia, for example, a Japanese
reduces the cost of R&D, and hence induces new case study (Ito 2011) showed that first-time
product introduction and TFP improvement. At exporters increased their R&D expenditure
the product level within firms, they found that the immediately after they exported, although the
increase in imported intermediate input increases increase varies by export market destinations.
the extent of product switching within firms, defined A study in the Republic of Korea (Hahn and
as simultaneous product adding and dropping. Park 2011) showed that exporting promotes the
The finding suggests the existence of a “creative creation of new products, while an Australian
destruction” process within firms, which implies a study (Palangkaraya 2011) showed that exporters
better reallocation of resources. in the services sector increase their process
innovation activities.
Product Dynamics Technological Change and Innovation International Production Networks
The heterogeneous firm theory has become International trade or foreign direct investment International production networks (IPNs) began
more advanced by adopting models with (FDI) plays a role in promoting R&D to generate to be developed as multinational enterprises
multiproduct firms. It produces predictions on innovation. Engagement in exporting stimulates (MNEs) adopted a fragmentation strategy, under
the optimal solution concerning dynamics of R&D activities of exporters and increases which they break up an entire production system
product portfolio within a firm. Theories have exporters’ productivity, within the learning- into various processes or production blocs, which
Trade Adjustment in Asia: Past Experiences and Lessons Learned

been developed to predict the impact of trade by-exporting hypothesis. Ito (2011) found are then relocated to different countries where
liberalization on product scope of a firm (e.g., that the decision to export by new Japanese a particular process can be undertaken most
Feenstra and Ma 2008; Eckel and Neary 2010; exporters increases their R&D spending. Hahn efficiently. IPNs have been formed by connecting
and Bernard, Redding, and Schott 2011). and Park (2011) found evidence to support the or linking the production blocs located in
role of innovation in the learning-by-exporting different countries.
hypothesis—that is, a statistically significant
positive impact of exporting on product creation.
Product creation here is defined to involve strong
innovation activities.
continued on next page
Table 6.1 continued
Product Dynamics Technological Change and Innovation International Production Networks
All these predict that trade liberalization As for FDI, firm-specific advantages of Critical to IPNs is liberalization of trade and
reduces product scope, which was evident in MNES—in the form of knowledge-based assets, FDI, and evidence of this is strong in cases of
Baldwin and Gu (2009); Bernard, Redding, and managerial know-how, quality of the workforce, Southeast Asian countries. Kohpaiboon and
Schott (2011); and Mayer, Melitz, and Ottaviano and marketing and branding—are expected to Jongwanich (2013) found that firms participating
(2014). These studies suggest that dropping promote R&D activity in the host countries and in IPNs are more active in R&D activities than
products is the most immediate (and easy) hence, generate innovation. Kohpaiboon and those not participating. Aldaba (2017) showed
response to fiercer competition resulting from Jongwanich (2013) found that foreign investment that the expansion of the global value chain index
trade liberalization. The evidence is, however, not encourages firms to commit investment in R&D. in electronic industries in the Philippines is closely
yet robust. Qiu and Zhou (2013) found increased They found that the investment tends to be related to the opening of intermediate-input
product scope as an impact of trade liberalization imported—embodied in imported capital goods— sectors of the industries, as well as privatization
in manufacturing in the People’s Republic rather than invested in R&D in host countries. and fiscal incentives provided for MNEs invested
of China. Hahn, Ito, and Narjoko (2016) in a Regardless, a positive impact is still observed, in economic zones, a key element of FDI
comparative study of three countries (Japan, the including evidence that the presence of MNEs liberalization.
Republic of Korea, and Indonesia) found that firm stimulates locally owned firms to conduct R&D IPNs change the production structure in the
product scope increases, rather than decreases, activities. Kuncoro (2011), meanwhile, found medium and longer term, especially because
with export participation. that foreign ownership of firms in Indonesian they increase the demand for skilled workers in
Trade liberalization improves product quality. manufacturing determines the R&D decisions of participating developing countries, due to greater
Hayakawa, Matsuura, and Takii (2015), using a the firms but not the scale of the R&D investment. use of more technology-intensive imported input
case study of Indonesian manufacturing, found and of more advanced technology embodied
that reduction in input tariffs generally boosts in imported capital goods such as machineries.
quality upgrading, whereas the decrease in output Kohpaiboon and Jongwanich (2013) showed that
tariffs does not have a significant impact. This is engagement in IPNs increases the demand for
consistent with the view that imported inputs are skilled workers, albeit only in firms that are already
high in quality. skill intensive. Thangavelu (2013) found that firms
in Viet Nam participating in IPNs restructure their
production methods by installing machines with
Firm Adjustment to Trade Policy Changes in East Asia

more advanced technology, suggesting a higher

demand for skilled workers for these firms.
183

Source: Authors’ compilation.

184 Trade Adjustment in Asia: Past Experiences and Lessons Learned

a random sample, since not all firms within an industry export. Eaton,
Kortum, and Kramarz (2004), for example, highlight this for French
manufacturing, while Helpman, Melitz, and Yeaple (2004) did so for the
data on manufacturing in the United States (US).
Melitz built a theoretical model that takes into account the
importance of productivity differences across firms in an imperfect
competition setting. As explained and summarized by Helpman
(2006), Melitz’s model predicts that firm dynamics created by trade
liberalization reduce the productivity threshold for any firm to export,
implying that any firm now has a higher probability of exporting
compared with the situation before the liberalization. At the same time,
however, trade liberalization increases the productivity threshold for
the survival selection of any operating firm. This means that only more
productive firms survive after the trade liberalization. Industry output
is hence reallocated to these survivors. What we should ideally observe
then is a situation where the overall industry productivity improves.
The Melitz model has been extended by including technology
adoption and innovation to reflect technology upgrading by firms.
Some of these models are Bustos (2011), Yeaple (2005), and Ekholm and
Midelfart (2005). The Bustos model overall predicts that only a fraction
of firms—that is, firms with an intermediate level of productivity—
respond to trade liberalization by upgrading their technology (Helpman
2006). This comes as a result of both the coexistence of firms within
the industry with different levels of productivity and the existence of
different types of technology adopted by firms in the industry. Less-
productive firms, meanwhile, continue to use traditional technology.
It is important to mention the existence of a closely related strand
of literature that examines the relationship between firm dynamics and
economic performance. Certain theoretical works, in particular Jovanovic
(1982) and Hopenhayn (1992), model the interrelationship between
entry–exit and firm heterogeneity in terms of productivity. These models
detail how competitive struggle, reflected by firm dynamics (i.e., entry,
exit, and growth), affect productivity growth. Empirical studies on this
issue include Olley and Pakes (1996), Liu (1993), Liu and Tybout (1996),
and Aw, Chen, and Roberts (2001). Aw and her coauthors, for example,
found that new manufacturing firms in a newly industrialized economy
in Asia have lower average productivity than incumbents, although
productivity varies significantly across the firms. They also found that
the more productive entrants survive, and their productivity converges
to the level of incumbents (Aw, Chen, and Roberts 2001).
More recent studies from research projects run by the Economic
Research Institute for ASEAN and East Asia (ERIA) provide more
evidence on the positive impact of globalization on productivity
 Firm Adjustment to Trade Policy Changes in East Asia 185

and, more importantly, provide more knowledge on the underlying

mechanisms creating the impact.
Taking the heterogeneous firm theory as the basis, Narjoko (2012)
examined whether trade and investment liberalization in Viet Nam
improved industry productivity by improving resource allocation across
firms within industries. This study is motivated by the observation that
Viet Nam underwent rapid trade and investment liberalization during
the 1990s and experienced a massive firm entry in the 2000s. The
study asked whether trade and investment liberalization contribute to
the entry of firms, whether more firm entry is associated with greater
industry productivity growth, and whether the productivity level before
trade reforms matters for the extent of the productivity growth.
The study establishes a positive relationship between firm entry and
industry productivity growth in Viet Nam’s manufacturing. The rapid
trade and investment liberalization occurring in Viet Nam since the early
1990s, which has substantially reduced the cost of establishing private
enterprises and of exporting, seems to have triggered rapid growth in
the number of firms entering the country’s manufacturing and services
sectors. This finding suggests a reallocation of resources across firms
within Viet Nam’s manufacturing toward the more productive firms,
which has resulted in higher industry-level productivity growth.
Narjoko further examined the within-sector impact of firm entry.
Plotting the change in the distribution of productivity growth over
time, there is evidence that many firms have become more productive.
The productivity improvements, however, vary across firms. The study
shows that the entry of firms lowered the productivity of firms located
at the bottom of the distribution but increased the productivity of firms
located at the center of the distribution. It suggests that the increase in
productivity, as a result of the high entry rate, only applies to the firms
that have already acquired some intermediate level of productivity
before trade reform.
Hahn and Choi (2013) examined the effect of trade liberalization
on plant total factor productivity (TFP) growth and within-plant
across-product reallocation behavior in manufacturing in the Republic
of Korea during 1991–1998. They took the variety-based endogenous
growth models, which suggest that the increase in intermediate input
variety via trade reduces the cost of research and development (R&D),
and hence induces new product introduction and TFP improvement.
They examined whether the increase in imported intermediate input
variety increased plant TFP growth and the extent to which products
are switched (simultaneously added or dropped).
Hahn and Choi showed some evidence that tariff liberalization in
the Republic of Korea contributed to the growth of input variety during
186 Trade Adjustment in Asia: Past Experiences and Lessons Learned

the period studied. They found that plants belonging to industries with
higher variety growth in imported intermediate inputs experienced
higher productivity growth.
Hahn and Choi further elaborated the variety–productivity
relationships by testing the relationship between the imported
intermediate variety and product switching. Product switching, defined
as simultaneously adding and dropping products, can be understood
as part of a continuous process of “creative destruction” within plants.
Active product-switching behavior can enhance the resource allocation
process within firms and thereby improve their production efficiency.
The empirical results support the hypothesis, suggesting that the
increase in imported intermediate variety has a positive impact on
stimulating product switching by domestic plants.

6.3 Exporting
One of the most immediate implications of the Melitz (2003) approach,
commonly known as heterogeneous firm theory, is that it is easier for
firms to engage in the international market after trade liberalization.
Existing exporters can expand their export sales, and some firms start
to export for the first time.
Consistent with this prediction is the self-selection hypothesis,
which existed before Melitz’s heterogeneous firm theory. This is
based on the presumption that participating in export markets brings
additional costs, which usually involve high fixed costs. These include
transport costs and expenses related to establishing distributional
channels and production costs in adapting products for foreign tastes
(Bernard and Jensen 1999). Trade liberalization in export-destination
countries reduces the total costs of firms exporting to these countries, in
addition to providing more access markets. This is reflected in Melitz’s
framework by a reduced threshold for firms to export.
Both Melitz’s framework and self-selection theory imply that
exporters and nonexporters are different. Studies support this, and
exporters are considered better performers. For developed countries,
Bernard, Jensen, and Lawrence (1995) and Bernard and Jensen (1999)
documented that exporters in US manufacturing are larger, more
productive, and more capital-intensive; pay higher wages; and employ
more skilled workers than nonexporters. Aw and Hwang (1995) and
Berry (1992) observed a similar finding for developing countries.
Sjoholm and Takii (2003) also observed that exporting plants are larger
and more productive; the labor productivity of these plants was about
twice as high as nonexporting plants, and this difference seemed to
increase during the 1990s.
 Firm Adjustment to Trade Policy Changes in East Asia 187

The essence of self-selection means that firms prepare for exporting.

Supporting evidence for this hypothesis exists (e.g., Bernard and Jensen
1999; Clerides, Lach, and Tybout 1998; Aw, Chung, and Roberts 2000;
Hallward-Driemeier, Iarossi, and Sokoloff 2002). Bernard and Jensen
found that exporters in US manufacturing are more efficient and larger,
and they grow faster several years before they become exporters. For
the manufacturing industry in the Republic of Korea, Aw, Chung, and
Roberts (2000) found that the average productivity of continuing
exporters and new entrants as exporters is significantly higher than
exiting exporters and nonexporters.
Melitz’s heterogeneous firm theory more recently introduced a self-
selection mechanism and analyzed the effects of liberalized trade (e.g.,
Melitz 2003; Bernard et al. 2007). In these models, trade liberalization
raises aggregate productivity by inducing resource reallocation across
firms—that is, the contraction and exit of low-productivity firms as well
as the expansion and entry into export markets of high-productivity
firms—even if there is no change in firm-level productivity.
The self-selection hypothesis focuses on action before exporting. The
difference in performance between exporters and nonexporters can also
be explained by actions after exporting. Participating in export markets
creates a learning effect for firms, as exporters gain access to technical
expertise, including product design and method, from their foreign
buyers (Aw, Chung, and Roberts 2000). The learning process accumulates
knowledge acquired by firms and increases the productivity of exporters
over time, widening the performance gap between exporters and
nonexporters. This is often termed the learning-by-exporting hypothesis.
The more recent ERIA project provides evidence supporting
the learning-by-exporting hypothesis in terms of a firm’s innovation
responses after engaging in exporting. A Japanese case study (Ito 2011)
showed that first-time exporters increased their R&D expenditure
immediately after they exported, although the increase varies by export
market destinations. A study of the Republic of Korea (Hahn and Park
2011) showed that exporting promotes the creation of new products,
while an Australian study (Palangkaraya 2011) showed that exporters in
the services sector increase their process innovation activities. All these
studies show that the innovation response improves performance of the
exporters.

6.4 Competition
Globalization increases competition in the domestic market and triggers
dynamism in the survival and creation of new firms. Innovation links
competition and firm dynamics.
188 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Competition and innovation have a mixed relationship. The

most recent theoretical framework suggests an inverted U-shaped
relationship between competition and innovation (Aghion et al. 2002).
The framework correlates firms’ market power with their level of
innovation. In this framework, firms facing intense competition will
innovate more, as innovation serves as a method to escape from the
fierce competition. In contrast, on the other end of the spectrum, firms
facing weak competition do not have the incentive to innovate because
firms with market power do not need to win in competing with other
firms in the market. Evolution in the competitive struggle that moves
between these two extremes creates the inverted U-shaped relationship.
Innovation goes up when the market is very competitive, but greater
innovation generates market power for some firms; this reduces the
incentive to innovate, resulting in less innovation.
This theoretical framework has been reinforced by empirical
evidence from Aghion et al. (2002) and Aghion and Burgess (2003).
However, several studies find that the inverted U-curve relation is not
generally applied in several countries. Creusen et al. (2006) did not find
an inverted U-curve relation, although the relation between competition
and innovation was found to be positive. Hopman and Rojas-Romagosa
(2010) found a negative relationship between competition and
innovation, as well as insufficient evidence on the inverted U-curve
relation.
Meanwhile, an extensive amount of studies have found strong
evidence regarding the positive effect of trade on competition. The most
prevalent argument on this relationship is that trade fosters competition
and constrains domestic firms in conducting anticompetitive activities
(Cadot, Grether, and de Melo 2000). This is known as the “imports as
competitive discipline” hypothesis, which has found robust empirical
evidence. For instance, Erdem and Tybout (2003) have shown that trade
liberalization negatively affects the price-cost margins of firms. This was
reinforced by Harrison (1994), who found the same evidence in the Côte
d’Ivoire, and Krishna and Mitra (1998) in India. Trade liberalization is
also found to increase productivity, as exhibited by several empirical
studies, such as Pavcnik (2002) for Chile and Amiti and Konings (2007)
for Indonesia. Trade also positively affects innovation since trade
liberalization stimulates competition, which forces firms to become
more efficient and productive through innovation. Earlier work by
Aghion and Burgess (2003) showed the positive effect of reduced trade
barriers on the economic performance of firms close to the technological
frontier. Fernandes and Paunov (2009) presented evidence that trade
liberalization stimulates product quality upgrading using Chilean
manufacturing data, while Bloom, Draca, and Van Reenen (2010) found
 Firm Adjustment to Trade Policy Changes in East Asia 189

a similar relationship between trade liberalization and innovation in the

People’s Republic of China using patents, information technology (IT),
R&D, and TFP as the indicators.
Recently, more evidence was gathered for the Association of
Southeast Asian Nations (ASEAN) and East Asian economies from
ERIA’s microdata research projects on the impact of globalization.
Aldaba (2012), among others, examined the impact of competition on
innovation for manufacturing firms in the Philippines, using firm-level
panel data over 1996–2006. She examined the impact of trade barrier
removal on innovation activities and questioned whether an increase in
competition increased the innovation activities.
Aldaba found that trade reforms (i.e., reductions in tariff and/or
nontariff barriers), conducted several times in the Philippines from
the 1990s to the 2000s, have had a strong impact on the country’s
manufacturing sector by increasing competition in domestic markets.
The tariffs are found to be positively related to the price-cost margin.
This is the finding from the first step of Aldaba’s econometric estimation.
The second step of the estimation revealed that profitability is negatively
related to R&D expenditure. In other words, higher competition
stimulates R&D. Thus, overall, trade liberalization positively affects
R&D through the product market competition channel. All these
findings are generally the same even after she controls for firm entry and
exit, which are proxies for the industry selection impact arising from
competition. Further, from the results of her estimation in the “mixed”
sector (i.e., a broad sector group that consists of mostly exporting and
importing industries), she found that the net-entry variable is negatively
related to profitability. Together with a negative relationship between
profitability and R&D expenditure, this indicates that as more firms exit
(presumably the inefficient ones), the surviving firms tend to engage in
R&D to outcompete the new firms entering the market.
Another example is the study by Nguyen et al. (2011), which
examined the determinants of innovation by Viet Nam’s small
and medium-sized enterprises (SMEs) in the context of increased
competition resulting from rapid trade expansion in the 2000s. The
authors used data for 2007 and 2009 from the Viet Nam SME Survey.
The years of the data are chosen to capture the period when Viet Nam
experienced rapid trade liberalization. Unlike the approach taken by
other studies, Nguyen and his coauthors used information on pricing
strategies to capture the extent of competition among firms. The use of
this information was driven by the availability of information in the data.
Nguyen and his coauthors found moderately important effects of
competition, both domestic and international. Specifically, matching the
price of competitors has a positive impact on product innovation using
190 Trade Adjustment in Asia: Past Experiences and Lessons Learned

the 2007 data and on product improvement using the 2009 data. As for
the impact of international competition, they found that pressure from
foreign firms—in terms of the price set by them—improves all kinds of
innovation activities (i.e., product innovation, product modification, and
process innovation) by Viet Nam’s SMEs. The finding differs slightly
when the study uses the 2009 data. The authors not only addressed
the globalization impact through the competition channel but also
tested whether linkages with foreign firms help SMEs to increase their
innovation activities. They found rather convincing evidence for this,
using both years of the data and examining other innovation activities.

6.5 Product Dynamics

The literature on heterogeneous firms has gone on to consider the
models of multiproduct firms, motivated by an observation that trade
is now dominated by firms producing (and trading) more than one
product. Research for these models is also developing because of the
greater availability of product-level data by firm or plant. Theory based
on the multiproduct model suggests that trade liberalization changes
firms’ product portfolios and increases productivity.
Bernard, Redding, and Schott (2011) developed a model that interacts
firm-level productivity with firm-product-specific expertise, which
allows a firm to endogenously choose the range of products it exports.
The general equilibrium setting of the model results in a prediction for
adjustment at both industry and firm level. The adjustment at industry
level is the general result of the heterogeneous firm model, which
predicts that inefficient firms will exit the exporting market. Adjustment
at firm level—across a product range—enables firms with greater ability
(or productivity) to produce more products, extending the scope of
products that the firm can produce.
Trade liberalization pushes the firm to focus on its “core competence”
resulting from the change in focus of the firm on producing only higher-
expertise products because of the much higher export opportunities of
these products. This is reflected in the dropping of the lower-, or lowest-,
expertise products from the range of products for export. Unlike the
prediction of the other models, these products are still produced but
sold only in domestic markets. Thus, the scope of the products of the
firm increases as the firm becomes more productive over time. The
decision of the firm to drop its lowest-expertise products raises the
productivity of the surviving products and increases the overall firm-
level productivity. The model hence predicts a monotonic relationship
between productivity and product scope, as illustrated in Figure 6.1.
 Firm Adjustment to Trade Policy Changes in East Asia 191

Figure 6.1 Theoretical Prediction of the Relationship

between Productivity and Product Scope

τ
Note: N is the number of products (product scope) and τ is the level of productivity.
Source: A. W. Bernard, S. J. Redding, and P. K. Schott. 2011. Multiproduct Firms and Trade
Liberalization. The Quarterly Journal of Economics. 126 (3). pp. 1271–318.

Eckel and Neary (2010) built a model that recognizes (i) the
“cannibalization effect,” which is defined as the impact coming from
the internalization of demands within the firm across products the
firm produces; and (ii) “flexible manufacturing” (reflecting flexible
technology in machineries), which allows firms to produce a range of
products containing the firm’s core competence. Eckel and Neary’s
model predicts that globalization makes a firm “leaner and meaner” in
its product scope, which means that its range of products is pruned to
focus on its core competence.
Feenstra and Ma (2008) built a model with a similar prediction to
the one built by Eckel and Neary, that is, a firm produces (at the end) only
within the range of its core competence. The Feenstra and Ma model
can say more about what happens in the process. That is, the lowered
costs, caused by trade liberalization or a more open trade regime,
expand the range of products produced by the firm. The cannibalization
effect becomes unbearable for the firm when the market size grows,
however, since globalization forces the firm to start dropping products.
This results in a leaner product scope, as predicted by Eckel and Neary
(2010).
All the theoretical mechanisms above point to a reduced product
scope (i.e., product rationalization) because of trade liberalization. This
is evident in the studies conducted by Baldwin and Gu (2009), Bernard,
Redding, and Schott (2011), and Mayer, Melitz, and Ottaviano (2014),
192 Trade Adjustment in Asia: Past Experiences and Lessons Learned

suggesting that dropping products is the most immediate (and easy)

response to fiercer competition resulting from trade liberalization.
However, robust evidence of this is not yet available. Qiu and Zhou
(2013), for example, found increased product scope as an impact of
trade liberalization in the People’s Republic of China’s manufacturing.
Another study with similar results is Hahn, Ito, and Narjoko (2016).
Hahn and his coauthors examined the impact of exporting on
product portfolio upgrading, using plant-and-product level data from
Indonesia, Japan, and the Republic of Korea. The upgrading is defined
technically by the increase in the attributes of a product of a firm (or
a plant). The analysis was conducted in two steps: (i) the relationship
between exporting and product scope is examined, and (ii) after
measuring the attributes for products, the relationship between product
dynamics (product adding or dropping) and product attributes is
examined. The second step addresses whether changes in the product
extensive margin (product adding or dropping) reflect the resource
reallocation from products with lower product attributes to those with
higher product attributes.
The results provided evidence that changes in the product
composition are associated with the exporting activity. Evidence also
shows that plants’ total product scope increases, rather than decreases,
with export participation, though the results are not statistically
significant. These results are broadly in line with several recent empirical
studies that find that trade liberalization causes firms to add products
and expand product scope (Iacovone and Javorcik 2010; Berthou and
Fontagne 2013; and Qiu and Yu 2014). With respect to product portfolio
upgrading, Hahn, Ito, and Narjoko (2016) found that added products
have higher or better product attributes than dropped products.
More specific on the impact on quality of the product, more recent
studies point to an improvement in product quality as a result of trade
liberalization, which should be able to be traced back to improved
productivity.
Hayakawa, Matsuura, and Takii (2015) examined the effect of
tariff reductions on firms’ quality upgrading in Indonesia’s apparel
industry. The empirical results suggest that the reduction in input
tariffs generally boosts quality upgrading, whereas the decrease in
output tariffs does not have a significant impact. The results also
suggest that the positive impact of input tariff reduction on quality
upgrading is greater, particularly for firms importing intermediate
inputs. This is consistent with the view that imported inputs are high
in quality. These results show that imported products, especially
imported intermediate inputs, are an important factor for productivity
growth. The reduction in imported input prices due to tariff reduction
 Firm Adjustment to Trade Policy Changes in East Asia 193

encourages firms to increase imports of foreign materials, resulting in

an upgrade of output quality.
The positive impact extends to nonimporters, suggesting the
presence of positive technology spillovers. Local suppliers learned from
the increased foreign inputs and improved the quality of the inputs
they use. The improvement thus may boost quality upgrading by the
nonimporters.

6.6 Technological Change and Innovation

Innovation has been widely recognized as a key factor in generating
industrial development and promoting sustainable economic growth. As
in many innovation-based endogenous growth models, firms’ innovation
activity drives productivity growth as does the introduction of new
products or varieties (e.g., Romer 1990; Grossman and Helpman 1991). In
an open economy setting, international trade or foreign direct investment
(FDI) also play a role in promoting R&D to generate innovation.
Regarding the role of international trade, recent literature points to
the engagement in exports, which would help stimulate the R&D activities
of exporters and increase exporters’ productivity, as a mechanism within
the framework of the learning-by-exporting hypothesis. As for the role
of FDI, firm-specific advantages of multinational enterprises (MNEs)—
in the form of knowledge-based assets, managerial know-how, quality of
the workforce, and marketing and branding—are expected to promote
R&D activity in the host countries and hence generate innovation.
Therefore, competition has been strong among developing countries to
attract R&D-intensive FDI through fiscal incentives and high-quality
infrastructure at subsidized prices (Athukorala and Kohpaiboon 2010).

6.6.1 Exporting and Innovation

More evidence on the impact of exporting (as a response to trade
liberalization elsewhere) is identified for East Asian countries, such as
those highlighted in studies from the microdata project of ERIA. Ito
(2011) addressed the role of innovation in the context of the learning-
by-exporting hypothesis. She asks whether learning by exporting has
an effect on innovation and, subsequently, whether and how the impact
of exporting on innovation affects productivity. Ito attempts to find
answers to these questions by examining the behavior and performance
of first-time exporters in Japanese manufacturing. The study, therefore,
not only seeks evidence for the positive impact of learning by exporting
on innovation, but also moves deeper to find insights on the source of
the learning by exporting.
194 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Ito found that first-time exporters are able to increase their sales
and employment growth more than firms serving domestic markets.
More importantly, the decision to begin to export promotes innovation,
as first-time exporters record an increase in R&D intensity and volume.
Going deeper into the mechanism of learning by exporting, the
study examined whether there are differences in the performance of
innovation and other variables, which arise from engaging in exporting
to different destinations. The evidence showed that starting to export
to North America or Europe has larger positive effects on productivity
than starting to export to Asia. This difference is also observed for other
performance variables (i.e., sales and employment growth), innovation
variables, and some characteristics of the firms. This finding is ascribed
to differences in absorptive capacity, in that first-time exporters to
North America or Europe have greater absorptive capacity than those
exporting for the first time to Asia.
Hahn and Park (2011) used a rich combination of plant- and
product-level manufacturing data from the Republic of Korea in their
investigation. Unlike the previous studies, however, they adopt a
different approach in defining product innovation. They use plant-and-
product matched data to distinguish two types of product innovations:
those that are new to the plant (termed “product addition”) and those
that are new to the country’s economy (termed “product creation”). The
former tends to capture imitation by domestic competitors or the process
of domestic knowledge diffusion, while the latter reflects product cycle
phenomenon or international knowledge spillover. Product creation
could mean product addition, although this does not necessarily work
the other way around.
Hahn and Park found evidence to support the learning-by-
exporting hypothesis for the role of innovation in the export–
productivity relationship. Using propensity score matching, they
found a statistically significant positive impact of exporting on product
creation. They cannot, however, infer the existence of this relationship
when innovation is defined by product addition; the impact of exporting
on product addition is not statistically significant, although it shows
the same (positive) sign. The study was not able to find evidence to
support the selection hypothesis. More specifically, it could not find
any significant effect of innovation (for both product creation and
addition) on exporting. The investigation was extended by using the
vector autoregressive (VAR) method. This route is taken to examine
the dynamic interdependence between export and innovation, as well
as productivity. The key results are consistent with the key finding that
exporting significantly affects product creation. The finding from the
VAR indicates that this impact is quite persistent; it takes more than
 Firm Adjustment to Trade Policy Changes in East Asia 195

5 years for the impact on product creation to die out. The VAR results
also show that productivity significantly and positively affects both
exporting and product creation.
Palangkaraya (2011) investigated the direction of the causality
between exporting and innovation using firm-level data from Australian
SMEs. His investigation also looks at the direction of causality for the
group of new exporters and new innovators, to ensure the robustness of
the results. The sample of the study comprises not only manufacturing
firms, but also enterprises in the services and other nonmanufacturing
sectors. This offers distinct added value to the research, considering the
lessons from the usual samples from the manufacturing sector may not
be valid for the other sectors.
Palangkaraya found evidence that the relationship between
exporting and innovation runs in both directions: both reflecting the
self-selection and learning-by-exporting hypothesis. However, this only
appears for process innovation in the services sector, not for product
innovation and not in the manufacturing or other nonmanufacturing
sectors. The investigation also finds that the positive two-way
relationship varies across industries. Palangkaraya attributes all these
results to the uniqueness of the innovation characteristics of SMEs
and the importance of services in the Australian economy. Process
innovation matters more than product innovation, because SMEs are
usually financially constrained and product innovation is arguably
substantially more expensive than process innovation.

6.6.2 Foreign Direct Investment and Innovation

FDI plays a role in promoting R&D through the knowledge and

technology brought by MNEs to host countries. FDI liberalization,
therefore, is expected to be positively related to the extent of innovations.
The following section presents the findings of a few studies emanating
from the ERIA research project on the topic.
Kohpaiboon and Jongwanich (2013) examined the roles of MNEs
and exporting in determining the decision to carry out R&D, as well
as the intensity of R&D activities, in firms in the Thai manufacturing
sector, using the most recent (2006) industrial census data. Unlike the
other studies, which measure different types of R&D in their total value
terms, this study disaggregated R&D activities into three categories: (i)
R&D leading to improved production technology, (ii) R&D leading to
product development, and (iii) R&D leading to process innovation. The
study examines not only the direct effect of MNEs on R&D activities, but
also the indirect effect of MNEs on the presence and intensity of R&D in
locally owned plants (termed “R&D spillovers”).
196 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Kohpaiboon and Jongwanich found that globalization, through

exporting and FDI, can play a role in encouraging firms to commit to
R&D investment. The role played by FDI, however, seems to be different
from the role of exporting. The study found that the R&D propensity of
MNE affiliates is lower than that of locally owned firms. This suggests
that MNE affiliates in Thailand prefer to import technology from their
parent companies rather than investing in R&D in the host country
(Thailand). Nonetheless, this does not mean that no effect arises from
MNE presence on firm R&D propensity and intensity. In fact, the study
found that the presence of MNEs stimulates locally owned firms to
conduct R&D activities.
Kuncoro (2011) examined the globalization determinants of the
decision to invest in R&D and the intensity of R&D expenditure, of
medium-sized and large manufacturing firms in Indonesia. The study
considers export participation, foreign investment, and trade protection
as the variables that represent globalization. In addition, it looked at the
impact of the spatial concentration of MNEs on a firm’s R&D investment
decisions and expenditure. The author uses data from the mid-1990s to
the mid-2000s in his empirical investigation.
The study found that being an exporter significantly affects
a firm’s decision to invest in R&D, as well as the extent of a firm’s
R&D expenditure. Foreign ownership was found to be an important
determinant only for the R&D investment decision, but not for the
amount of R&D expenditure the firm commits. In terms of testing the
potential R&D spillover effect arising from the concentration of MNEs
in a location, the study found that R&D activities tend to be higher in
big urban areas, not in a specialized or agglomerated location. In the
interpretation of the findings related to foreign ownership and the
presence of MNEs, Kuncoro asserts that a critical mass of MNEs may
be needed in a location or agglomeration area for these MNEs to have a
meaningful impact in terms of innovation or R&D performance.

6.7 International Production Networks

International production networks (IPNs) began to be developed as
MNEs adopted a fragmentation strategy, under which they break up an
entire production system into various processes or production blocs,
which are then relocated to different countries where a particular
process can be undertaken most efficiently. IPNs have been formed by
connecting or linking the production blocs located in different countries.
The extent or degree of fragmentation depends mostly on the
cost of establishing and managing production blocs and the cost of the
 Firm Adjustment to Trade Policy Changes in East Asia 197

service link that connects production blocs. The cost of establishing and
managing production blocs depends largely on the labor cost, the quality
of infrastructure (including the supply of electricity, transportation,
and communication services), openness to foreign firms, and others,
while the cost of the service link depends on the cost of international
transportation and communication services, which are affected by the
international trade policies of the countries involved.
The expansion of IPNs has been aided by the liberalization of trade
and FDI policies implemented by Southeast Asian economies, as the
governments of these countries recognized the beneficial impacts of
hosting MNEs with extensive IPNs.1 IPNs bring not only export sales
and import procurement networks, which enable host countries to
import high-quality intermediate and capital goods, but also technology,
which contributes to an improvement in productivity. Since rapid and
extensive development of IPNs is partly due to MNEs’ response to the
liberalization of trade and FDI policies, we examine the impacts of IPNs
on firm behavior and the development of industries in East Asia.
The importance of FDI and an open trade regime is confirmed by
Kohpaiboon and Jongwanich (2013), who found that firms participating
in IPNs are more active in R&D activities than those not participating.
The dynamism of industries engaged in IPNs required firms populating
the industries to keep the industries competitive in international
markets.
Aldaba (2017) provided more evidence on the role of trade and
investment liberalization for participation of firms in IPNs. The
Philippine electronic industry has transformed to become deeply
integrated within networks of industries in East Asia. Analyzed using the
global value chain (GVC) participation index and length, Aldaba showed
that the Philippines increased participation in the backward linkage of
GVC over time.2 The share of foreign inputs in Philippine electronic
exports (looking backward along the value chain) increased from 8.5%
in 1995 to 32.5% in 2000 and 34.4% in 2008. The trend is the same for
the forward GVC participation of the sector. The share of domestically

1
IPNs involving Southeast Asian countries were triggered by the currency
appreciation of industrialized East Asian economies in the 1980s. The appreciation
of the yen in the latter half of the 1980s prompted a massive outflow of Japanese
FDI by Japanese MNEs, which adopted the fragmentation strategy and relocated
production processes from Japan to Southeast Asian economies.
2
The GVC participation index is defined as the share of foreign inputs (backward
participation) and domestically produced inputs used in third countries’ exports
(forward participation), expressed as a percentage of gross exports (De Backer and
Miroudot 2013).
198 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 6.2 Global Value Chain Participation Index

of Electronics Industry in Selected Economies, 2009
50
45
40
35
30 32.4
25
20
15 17.2 13.0
13.6
12.9
10 14.3
14.3 14.8 4.3
5 7.4 1.5
5.3 6.8 5.4 4.7 2.1
0 2.7 0.8 3.2 0.9 1.7
0.2
es

nd

sia

re

es

sia

am
a
re

ic

pa
in
o
in

at
la

ay

ne
ex

N
Ko

ap

Ch

Ja
pp

St
ai

al

do
M

et
ng
Th

of
ili

d
of

Vi
In
te
Si
Ph

ic

ic

ni
bl

bl

U
pu

pu
Re

Re
's
le
op
Pe

Backward Forward

Source: Organisation for Economic Co-operation and Development (OECD) Global Value Chains
indicators – May 2013, https://stats.oecd.org/Index.aspx?DataSetCode=GVC_INDICATORS
(accessed 21 August 2018).

produced inputs used in third-country exports (looking forward along

the value chain) increased from 2.2% in 1995 to 8.4% in 2000 and 16.2%
in 2008. Reflecting this, the GVC participation of the industry in the
Philippines is among the highest in the region, as indicated in the cross-
country comparison of the GVC index in Figure 6.2.
Aldaba (2017) further explained that the development of the
Philippine electronics industry to the level seen in the 2000s can be
attributed to reforms to liberalize the investment and trade regime that
took place since the 1990s. Development of the industry is argued to have
been the result of both privatization of economic zone management and
fiscal incentives applied exclusively for investment in economic zones,
both of which have been elements of FDI liberalization since the 1990s.
The Philippine Economic Zone Authority granted significant incentives
for investment in the electronics industry, such as tax- and duty-free
importation of capital goods and intermediate inputs, and defining the
electronics industry as a preferred area of investment from 1988 to 1994
and from 2006 to 2007.
The electronics industry has benefited from trade liberalization
through tariff cuts or the removal of import restrictions for the import
of material inputs and finished goods. Liberalization of the import
regime for material inputs contributes directly to competitiveness by
reducing the price of final products produced domestically, while the
 Firm Adjustment to Trade Policy Changes in East Asia 199

liberalization of finished goods affects indirectly by improving efficiency

as a result of greater competition from imports.
IPNs change the production structure in the medium and longer
term. The basic proposition is that the networks increase the demand
for skilled workers in participating developing countries. Feenstra and
Hanson (1996, 1997) predict this to come from greater usage of imported
intermediate inputs, which are typically skill-intensive inputs performed
by MNEs in IPNs. They argue that while IPNs shift production blocs that
are unskilled or less technology-intensive from developed countries,
they are still considered skill- or technology-intensive production blocs
from developing countries’ perspective.
In the context of Feenstra and Hanson’s “outsourcing” or
“production sharing” theory, skill-biased technological change is
another explanation. Skill-biased technological change argues that
the new technology embodied in imported capital goods—through a
more open trade regime or an increase in FDI as a result of investment
liberalization—increases the demand for skilled workers (in host
countries). In other words, the technical changes induced by trade
and FDI liberalization have some effect (i.e., the “bias”) to increase the
demand for skilled workers.
Kohpaiboon and Jongwanich (2013) provided some support for the
predicted higher demand for skilled workers for firms participating in
IPNs. They found that engagement in IPNs increases the demand for
skilled workers, but this only applies to firms that are already skill-
intensive. Thangavelu (2013) found that firms in Viet Nam participating
in IPNs restructure their production methods by installing machines
with more advanced technology, suggesting higher demand for skilled
workers for these firms.

6.8 Policy Implications

This chapter presents several key topics on the responses of firms to
globalization, in responding to a more open trade or investment regime
between countries. All these have policy implications, and the discussion
below presents some of these.
First, most previous studies suggest that a country should continue
with ongoing trade liberalization and maintain a relatively open trade
regime. Strong domestic market competition drives firms to engage in
innovation-enhancing activities, through the ability of the competition
to create a contestable market situation. A liberalized trade regime could
be even more beneficial in the framework of a deepened integration of
a country in Southeast Asia or East Asia. Some studies underline this
in the context of linking firms to established IPNs in these regions.
200 Trade Adjustment in Asia: Past Experiences and Lessons Learned

They found a positive relationship between participation in production

networks and increased R&D activities by firms.
To complement trade liberalization, a reduction in trade costs
(commonly done via trade facilitation reform) should be a high
priority on the policy agenda for countries that have yet to join IPNs.
Improving trade-related infrastructure is likely to be an important
ingredient of policy. In many developing countries, transport costs
remains a key bottleneck. Poor transport infrastructure raises transport
costs and isolates markets. Such isolated markets may also feature
minimal competition, and this will worsen within-country poverty and
distribution issues.
Second, it is necessary to ensure that the forces of competition are
at work in domestic markets. Some of the dynamic gains from trade
are realized through reallocation across firms and industries, and
even across products within firms. It is therefore necessary to focus
on the elimination or reduction of existing regulations, such as entry
regulations, strong employment protection, and business regulations
based on firm size, which inhibit the reallocation of resources by market
forces. In cases where proper institutions or markets are lacking, such
as bankruptcy laws and procedures, building and improving these
institutions or markets should be a top priority.
Third, policies to promote exports encourage firm innovation. Thus,
policies to assist firms to export more, as well as to cause more firms
to engage in exports, seem warranted. Several findings on the positive
relationship between exporting and innovation activities and/or
performance support this. Among others, and perhaps most importantly,
is evidence of the positive effect of learning-by-exporting on exporters’
innovation—for instance, exporting encourages the creation of new
products as well as the expansion of export markets over time.
Fourth, policies for stronger foreign participation in industrialization
should be encouraged. The justification for this comes mostly from
evidence of the impact of R&D spillovers on domestically owned firms.
For example the presence of MNEs encourages locally owned firms
to gain technological knowledge and capability from various possible
channels, such as demonstration and the competition effect. From a
macro and practical perspective, encouraging a higher presence of
foreign ownership or MNE units requires a policy to sustain excellent
infrastructure quality, both physical and institutional. The logic is clear;
MNEs would consider investing in host countries if they are able to
operate efficiently, and one of the key factors is supportive infrastructure.
It is also important to achieve and/or maintain a stable macroeconomic
environment to attract MNEs.
 Firm Adjustment to Trade Policy Changes in East Asia 201

It is useful to comment here that a unique characteristic of

countries in East Asia and Southeast Asia is that they are very flexible
and welcoming to the evolving production networks between countries
orchestrated by MNEs. Very open trade and investment regimes, with
the help of sizable fiscal incentives sometimes, plus the typically flexible
labor market, seem to have strongly facilitated formation of IPNs within
East Asia and Southeast Asia. This marks a significantly different model
of industrialization to that adopted by other regions in the world, such
as the one typically adopted by countries in Latin America.
Fifth, findings from the research suggest that globalization seems
to benefit not only large firms but also SMEs. While this is encouraging,
if one considers affirmative action policies for SMEs in the context of
increased globalization in a country’s economy, the more important
question perhaps is how to devise policies that could harness the benefits
of globalization. Conceptually, the policies should be to equip SMEs to
learn more about process innovation, rather than product innovation,
from using globalization forces. This approach is sensible given the
natural disadvantages of SMEs, vis-à-vis their larger counterparts, in
terms of financial resources and economies of scale. Further, given the
usual assistance-type policy for SMEs, export promotion policies for
SMEs in general would be most effective if they were integrated with
policies to promote SMEs’ innovation activities, which in this case
should focus more on process innovation activities.
202 Trade Adjustment in Asia: Past Experiences and Lessons Learned

References
Aghion, P., N. Bloom, R. Blundell, R. Griffith, and P. Howitt. 2002.
Competition and Innovation: An Inverted U Relationship. NBER
Working Paper Series No. 9269. Cambridge, MA: National Bureau
of Economic Research.
Aghion, P., and R. Burgess. 2003. Liberalization and Industrial
Performance: Evidence from India and the UK. In E. Zedillo, ed. The
Future of Globalization: Explorations in Light of Recent Turbulence.
London: Routledge. pp. 557–92.
Aldaba, R. M. 2012. Trade Reforms, Competition, and Innovation in the
Philippines. ERIA Discussion Paper Series No. 5. Jakarta: Economic
Research Institute for ASEAN and East Asia.
____. 2017. The Philippines in the Electronics Global Value Chain:
Upgrading Opportunities and Challenges. In L. Y. Ing and F. Kimura,
eds. Production Networks in Southeast Asia (Routledge-ERIA Studies
in Development Economics). London: Routledge. pp. 161–84.
Amiti, M., and J. Konings. 2007. Trade Liberalization, Intermediate
Inputs, and Productivity: Evidence from Indonesia. American
Economic Review. 77 (5). pp. 1611–38.
Athukorala, P., and A. Kohpaiboon. 2010. Globalization of R&D by
US-based Multinational Enterprises. Research Policy. 39 (10).
pp. 1335–47.
Aw, B. Y., X. Chen, and M. J. Roberts. 2001. Journal of Development
Economics. 66 (1). pp. 51–86.
Aw, B. Y., S. C. Chung, and M. J. Roberts. 2000. The World Bank Economic
Review. 14 (1). pp. 65–90.
Aw, B. Y., and A. R. Hwang. 1995. Productivity and Export Market: A Firm-
Level Analysis. Journal of Development Economics. 47 (2). pp. 313–32.
Baldwin, J., and W. Gu. 2009. The Impact of Trade on Plant Scale,
Production-Run Length, and Diversification. In T. Dunne, J.B.
Jensen, and M. Roberts, eds. Producer Dynamics: New Evidence from
Micro Data. Chicago: University of Chicago Press. pp. 557–92.
Bernard, A. B., and J. B. Jensen. 1999. Exceptional Exporter Performance:
Cause, Effect, or Both? Journal of International Economics. 47 (1).
pp. 1–25.
Bernard, A. B., J. B. Jensen, and R. Z. Lawrence. 1995. Exporters, Jobs,
and Wages in U.S. Manufacturing: 1976–1987. Brookings Papers on
Economic Activity, Microeconomics. pp. 67–119.
Bernard, A. W., J. B. Jensen, S. J. Redding, and P. K. Schott. 2007. Firms in
International Trade. Journal of Economic Perspective. 21 (3). pp. 105–30.
Bernard, A. W., S. J. Redding, and P. K. Schott. 2011. Multiproduct Firms
and Trade Liberalization. The Quarterly Journal of Economics.
126 (3). pp. 1271–318.
 Firm Adjustment to Trade Policy Changes in East Asia 203

Berry, R. A. 1992. Firm (or Plant) Size in the Analysis of Trade and
Development. In G. K . Helleiner, ed. Trade Policy, Industrialization, and
Development: New Perspectives. Oxford: Clarendon Press. pp. 46–88.
Berthou, A., and L. Fontagne. 2013. How Do Multiproduct Exporters
React to a Change in Trade Costs? Scandinavian Journal of
Economics. 115 (2). pp. 326–53.
Bloom, N., M. Draca, and J. Van Reenen. 2010. Trade Induced Technical
Change? The Impact of Chinese Imports on Innovation, IT and
Productivity. NBER Working Paper Series. No. 16717. Cambridge,
MA: National Bureau of Economic Research.
Bustos, P. 2011. Trade Liberalization, Exports, and Technology
Upgrading: Evidence on the Impact of MERCOSUR on Argentinean
Firms. American Economic Review. 101 (1). pp. 304–40.
Cadot, O., J-M. Grether, and J. de Melo. 2000. Trade and Competition
Policy: Where Do We Stand? Journal of World Trade. 34 (3). pp. 1–20.
Choi, Y.-S., and C. H. Hahn. 2013. Effects of Imported Intermediate
Varieties on Plant Total Factor Productivity and Product Switching:
Evidence from Korean Manufacturing. Asian Economic Journal.
27 (2). pp. 125–43.
Clerides, S., S. Lach, and J. R. Tybout. 1998. Is “Learning-by-Exporting”
Important?: Micro-Dynamic Evidence from Colombia, Mexico and
Morocco. Quarterly Journal of Economics. 113 (3). pp. 903–47.
Creusen, H., B. Vroomen, H. van der Wiel, and F. Kuypers. 2006.
Dutch Retail Trade on the Rise? Relation Between Competition,
Innovation and Productivity. CPB Document. No. 137. The Hague:
CPB Netherlands Bureau for Economic Policy Analysis.
De Backer, K., and S. Miroudot. 2013. Mapping Global Value
Chains. OECD Trade Policy Paper. No. 159. Paris: OECD Publishing.
Eaton, J., S. Kortum, and F. Kramarz. 2004. Dissecting Trade: Firms,
Industries, and Export Destinations. American Economic Review.
94 (2). pp. 150–54.
Eckel, C., and J. P. Neary. 2010. Multiproduct Firms and Flexible
Manufacturing. Review of Economic Studies. 77 (1). pp. 188–217.
Ekholm, K., and K. Midelfart. 2005. Relative Wages and Trade-induced
Change in Technology. European Economic Review. 49. pp. 1637–63.
Erdem, E., and J. Tybout. 2003. Trade Induced Technical Change? The
Impact of Chinese Imports on Innovation, IT and Productivity.
NBER Working Paper Series. No. 16717. Cambridge, MA: National
Bureau of Economic Research.
Feenstra, R., and H. Ma. 2008. Optimal Choice of Product Scope
for Multiproduct Firms under Monopolistic Competition. In
H. Helpman, D. Marin, and T. Verdier, eds. The Organization of
Firms in a Global Economy. Cambridge, MA: Harvard University
Press. pp. 173–99.
204 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Feenstra, C. and G. Hanson. 1996. Globalization, Outsourcing, and Wage

Inequality. NBER Working Paper Series. No. 5424, Cambridge, MA:
National Bureau of Economic Research.
____. 1997. Foreign Direct Investment and Relative Wages: Evidence
from Mexico’s Maquiladoras. Journal of International Economics.
42 (3-4). pp. 371–93.
Fernandes, A., and C. Paunov. 2009. Does Tougher Import Competition
Foster Product Quality Upgrading? Policy Research Working Paper.
No. 4894. Washington, DC: World Bank.
Grossman, G., and E. Helpman. 1991. Trade, Knowledge Spillovers, and
Growth. European Economic Review. 35 (2–3). pp. 517–26.
Hahn, C. H., and Y-S. Choi. 2013. Trade Liberalization and the Wage
Skill Premium in Korean Manufacturing Plants: Do Plants’ R&D and
Investment Matter? In C. H. Hahn and D. A. Narjoko, eds. Impact of
Globalization on Labor Market. ERIA Research Project Report 2012,
No. 4. Jakarta: Economic Research Institute for ASEAN and East
Asia. pp. 15–37.
Hahn, C. H., K. Ito, and D. Narjoko. 2016. Exporting and Upgrading
of Product Portfolio: Evidence from Korea, Japan, and Indonesia.
Asian Economic Journal. 30 (4). pp. 349–73.
Hahn, C. H., and C.-G. Park. 2011. Direction of Causality in Innovation-
Exporting Linkage: Evidence on Korean 79 Manufacturing. In C.H.
Hahn and D. Narjoko, eds. Globalization and Innovation in East
Asia. ERIA Research Project Report 2010, No. 4. Jakarta: Economic
Research Institute for ASEAN and East Asia. pp. 79–115.
Hallward-Driemeier, M., G. Iarossi, and K.L. Sokoloff. 2002. Exports
and Manufacturing Productivity in East Asia: A Comparative
Analysis with Firm-level Data. NBER Working Paper Series. No.
8894. Cambridge, MA: National Bureau of Economic Research.
Harrison, A. 1994. Productivity, Imperfect Competition and Trade
Reform: Theory and Evidence. Journal of International Economics.
36 (1-2). pp. 53–73.
Hayakawa, K., T. Matsuura, and S. Takii. 2015. Does Trade Liberalization
Boost Quality Upgrading? Evidence from Indonesian Plant-Product-
Level Data. In C.H. Hahn, and D. Narjoko, eds. Trade Policy Changes
and Firm Adjustment: A Search for the Underlying Mechanisms.
ERIA Research Project Report 2014, Jakarta: Economic Research
Institute for ASEAN and East Asia.
Helpman, E. 2006. Trade, FDI, and the Organization. Journal of
Economic Literature. 44. pp. 589–630.
Helpman, E., M. Melitz, and S. Yeaple. 2004. Export versus FDI
with Heterogeneous Firms. American Economic Review. 94 (1).
pp. 300–16.
 Firm Adjustment to Trade Policy Changes in East Asia 205

Hopenhayn, H. A. 1992. Entry, Exit, and Firm Dynamics in Long Run

Equilibrium. Econometrica. 60 (5). pp. 1127–50.
Hopman, C., and H. Rojas-Romagosa. 2010. The Relation Between
Competition and Innovation: Empirical Results and Implementation
into Worldscan. CPB Memorandum. The Hague: CPB Netherlands
Bureau for Economic Policy Analysis.
Iacovone, L., and B. S. Javorcik. 2010. Multi-product Exporters: Product
Churning, Uncertainty and Export Discoveries. The Economic
Journal. 120 (544). pp. 481–99.
Ito, K. 2011. Sources of Learning-by-Exporting Effects: Does Exporting
Promote Innovation?” In C. H. Hahn, and D. Narjoko, eds.
Globalization and Innovation in East Asia. ERIA Research Project
Report 2010, No. 4. Jakarta: Economic Research Institute for ASEAN
and East Asia. pp. 20–67.
Jovanovic, B. 1982. Selection and the Evolution of Industry. Econometrica.
50 (3). pp. 649–670.
Kohpaiboon, A., and J. Jongwanich. 2013. Global Production Sharing and
Wage Premium: Evidence from Thai Manufacturing. In C.H. Hahn
and D. Narjoko, eds. Impact of Globalization on Labor Market. ERIA
Research Project Report 2012, No. 4. Jakarta: Economic Research
Institute for ASEAN and East Asia. pp. 135–64.
Krishna, P., and D. Mitra. 1998. Trade Liberalization, Market Discipline
and Productivity Growth: New Evidence from India. Journal of
Development Economics. 56 (2). pp. 447–62.
Kuncoro, A. 2011. Globalization and Innovation in Indonesia: Evidence from
Micro-Data on Medium and Large Manufacturing Establishments. In
C. H. Hahn and D. Narjoko, eds. Globalization and Innovation in East
Asia. ERIA Research Project Report 2010, No. 4. Jakarta: Economic
Research Institute for ASEAN and East Asia. pp. 193–224.
Liu, L. 1993. Entry-Exit, Learning, and Productivity Change: Evidence
from Chile. Journal of Development Economics. 42 (2). pp. 217–42.
Liu, L., and J. R. Tybout. 1996. Productivity Growth in Chile and
Colombia: The Role of Entry, Exit, and Learning. In M. J. Roberts
and J. R. Tybout, eds. Industrial Evolution in Developing Countries:
Micro Patterns of Turnover, Productivity, and Market Structure.
New York: Oxford University Press. pp. 73–103.
Mayer, T., M. Melitz, and G. Ottaviano. 2014. Market Size, Competition,
and the Product Mix of Exporters. American Economic Review.
104 (2). pp. 495–536.
Melitz, M. J. 2003. The Impact of Trade on Intra-Industry Reallocations
and Aggregate Productivity. Econometrica. 71 (6). pp. 1695–725.
Narjoko, D. 2012. Policy Reforms, Firm Entry, and Labor Productivity
Change: Learning from Vietnamese Manufacturing. In S. Urata,
206 Trade Adjustment in Asia: Past Experiences and Lessons Learned

C. H. Hahn, and D. Narjoko, eds. Economic Consequences of

Globalization: Evidence from East Asia. Routledge–ERIA Studies in
Development Economics. London: Routledge. pp. 103–26.
Nguyen, N. A., P. M. Nguyen, D. N. Nguyen, and D. C. Nguyen. 2011.
Trade Liberalization and Innovation Linkages Micro-evidence
from Vietnam SME Surveys. In C. H. Hahn and D. Narjoko, eds.
Globalization and Innovation in East Asia. ERIA Research Project
Report 2010, No. 4. Jakarta: Economic Research Institute for ASEAN
and East Asia. pp. 315–40.
Olley, G. S., and A. Pakes. 1996. The Dynamics of Productivity in the
Telecommunications Equipment Industry. Econometrica. 64 (6).
pp. 1163–97.
Palangkaraya, A. 2011. The Link between Innovation and Export:
Evidence from Australia’s Small and Medium Enterprises. In
C. H. Hahn and D. Narjoko, eds. Globalization and Innovation in East
Asia. ERIA Research Project Report 2010, No. 4. Jakarta: Economic
Research Institute for ASEAN and East Asia. pp. 103–40.
Pavcnik, N. 2002. Trade Liberalization, Exit, and Productivity
Improvement: Evidence from Chilean Plants. Review of Economic
Studies. 69 (1). pp. 245–76.
Qiu, L. and M. Yu. 2014. Multiproduct Firms, Export Product Scope,
and Trade Liberalization: The Role of Managerial Efficiency. ERIA
Discussion Paper Series. DP-2014-06. Jakarta: Economic Research
Institute for ASEAN and East Asia.
Qiu, L., and W. Zhou. 2013. Multiproduct Firms and Scope Adjustment in
Globalization. Journal of International Economics. 91 (1). pp. 142–53.
Romer, P. 1990. Endogenous Technological Change. Journal of Political
Economy, 98 (5). pp. S71–102.
Sjoholm, F., and S. Takii. 2003. Foreign Networks and Exports: Results
from Indonesian Panel Data. ICSEAD Working Paper Series. Vol.
2003–33. Kitakyushu, Japan: International Center for the Study of
East Asian Development.
Thangavelu, S. 2013. Trade, Technology, Foreign Firms and Wage
Gap: Case of Vietnam Manufacturing Firms. In C. H. Hahn and
D. Narjoko, eds. Impact of Globalization on Labor Market. ERIA
Research Project Report 2012, No. 4. Jakarta: Economic Research
Institute for ASEAN and East Asia. pp. 107–33.
Yeaple, S. 2005. A Simple Model of Firm Heterogenity, International
Trade, and Wages. Journal of International Economics. 65 (1).
pp. 1–20.
 7
The Rise of the
People’s Republic of China
and Its Competition Effects
on Innovation in Japan1
Nobuaki Yamashita and Isamu Yamauchi

7.1 Introduction
As a reaction to import competition from low-wage economies, firms
in developed economies would respond by upgrading their innovative
activities, leading to so-called defensive skill-biased innovation. In this
chapter, we examine this “defensive innovation” hypothesis, which was
first discussed in Wood (1994) and subsequently formalized in Thoenig
and Verdier (2003). In a broader context, the effect of competition on the
rate of innovation has been one of the most studied areas in the literature
(e.g., Aghion et al. 2005). In the study most relevant to our work, Bloom,
Draca, and Van Reenen (2016) found that a large sample of European
firms increased a wide range of their innovative activities (patenting,
research and development [R&D] expenditures, computer use, and total
factor productivity growth), driven by intensified competition from the
People’s Republic of China. This innovation was conducted within-firm.2

1
This study is conducted as a part of the Mobility of Knowledge and Innovation
Performance project undertaken at the Research Institute of Economy, Trade and
Industry (RIETI). This study utilizes the microdata of the questionnaire information
based on the Basic Survey of Japanese Business Structure and Activities, which is
conducted by Japan's Ministry of Economy, Trade and Industry, and the Kikatsu
Oyako converter, which is provided by RIETI.
2
Amiti and Khandelwal (2013) find that increased import competition (measured by
a decline in tariffs) spurs an economy’s export quality (measured by market share) in
the market in the United States (US).

207
208 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Building on the foundation set by the previously mentioned

studies, this chapter examines the causal effect of intensified import
competition from the People's Republic of China on the innovative
activities of a panel of Japanese firms for the period 1994–2005. We
focus on patent usage data as an indicator of innovative outputs. Unlike
other studies using patent statistics, this study adds to the literature by
exploring strategic patent usage as a response to import competition
from a low-wage economy (People’s Republic of China). While it is
generally acknowledged that patent statistics are meaningful proxies
for firm-level innovation, firm-level patenting serves as more than just
an indicator of knowledge capital output (Nagaoka, Motohashi, and
Goto 2010). Well-known inventor surveys3 have revealed that many of
the patents are not used to introduce new products into the market;
instead, they are used as effective strategic instruments to “block” other
competitors from innovating or imitating. Boldrin and Levine (2013)
present a case involving Microsoft, a market incumbent with a stockpile
of patents blocking Google in the smartphone market.
Studying innovative firms’ responses to import competition from
the People’s Republic of China provides an interesting and excellent
testing ground for the following reasons: First, over the past decades,
the People’s Republic of China has emerged as a pivotal assembly-export
economy of high-tech products (mainly, electronics), importing parts
and components from other developed economies and exporting final
products (including the famous iPhone). Accordingly, the country’s
export bundle has dramatically changed from labor-intensive goods
to high-tech products, exerting considerable competitive pressures on
firms in developed economies. Second, many of their exports compete at
lower cost margins than most high-tech products. For instance, a study
by Schott (2008) found that the People’s Republic of China’s export
similarity index has become closer to that of Organisation for Economic
Co-operation and Development (OECD) economies, but the unit prices
of the People’s Republic of China’s exports have been consistently lower
than those of OECD economies.
The finding suggests that the People’s Republic of China’s import
competition leads Japanese firms to expand their innovative activities,
as found by Bloom, Draca, and Van Reenen (2016). The expansion is
partly driven by an increase in firms’ numbers of unused patents, which
reflect the strategic use of intellectual property protection.

3
For example, the survey jointly conducted by the Research Institute of Economy,
Trade and Industry (RIETI) in Japan and Georgia Tech of investors in the US.
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 209

7.2 T
 he Rise of the People’s Republic of China
in World Trade
Figure 7.1 depicts the rise of the People’s Republic of China in world
exports for 1990–2011. In 1990, the People’s Republic of China’s exports
accounted for a tiny share (around 3%) of world exports. Since then, the
country’s share has gradually increased. In particular, its export growth
has risen since the early 2000s. In the second half of the 2000s, the
country achieved formidable export expansion by overtaking Germany’s
position as the world’s largest exporter, accounting for more than 10% of
world exports. The People’s Republic of China’s export share has been
growing without any disruptions, while the world shares of Japan, the
United States, and Germany have not grown during the same period. At
the same time, the country has become an important economy in the
world import market. While the United States still accounts for the bulk
of world imports (around 15%–20% in world imports), its share has
gradually been declining since 2000. By contrast, the People’s Republic
of China’s share has steadily increased to close to 10% in 2011.

Figure 7.1 The Rise of the People’s Republic of China

in World Trade, 1990–2011 (% in total exports)
14

12

10

0
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010

People’s Republic of China Germany Japan US

US = United States.
Source: United Nations International Trade Statistics Database (UN Comtrade).

With the rise of the People’s Republic of China in world trade, its
specialization has dramatically changed, as well. Figure 7.2 depicts the
share of relatively more capital- and technology-intensive products (e.g.,
electrical machinery and household electric appliances) as compared to
210 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 7.2 Structural Changes in the People’s Republic

of China’s Export Product Composition, 1990–2011
(% in total exports)
50
45
40
35
30
25
20
15
10
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Textiles and toys Electrical machinery and household electric appliances

Source: United Nations International Trade Statistics Database (UN Comtrade).

more labor-intensive products (e.g., textiles and toys). There has been a
notable shift of comparative advantage from more labor-intensive products
toward more capital- and technology-intensive products. In 1992, textiles
and toys accounted for approximately 45% of the People’s Republic of
China’s total exports. However, this share continuously declined and
dropped to close to 20% in 2011. On the other hand, the export share of
electrical machinery and household appliances doubled from less than 15%
in 1992 to 30% in 2011. In this product category, the export composition
is highly concentrated in information and communication technology
products. Other important product categories include office machines and
telecommunications sound equipment (including mobile phones).
Based on the income-weighted export bundle of goods from
the People’s Republic of China, some commentators argue that the
technological capability of the country is showing signs of rapidly
converging toward the technological frontier of developed OECD
economies and is now directly competing with them in the export
market. However, this should be interpreted cautiously. Allowing for
intraproduct specialization, it is known that the People’s Republic of
China’s export specialization still rests largely on the labor-intensive
assembly stage rather than specialization in technological content
(Athukorala and Yamashita 2006). In other words, the country’s
comparative advantage still rests on a labor-intensive segment in high-
tech products, even though these products are exported from the
country (a final assembly economy). This explains why Schott (2008)
observes that the unit price of the People’s Republic of China’s export
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 211

bundles are at the lower end of the price range, as compared to those of
OECD economies (the price competitiveness coming from the People’s
Republic of China’s lower labor costs). In sum, the bulk of the country’s
exports are mass-market commodities assembled with relatively low
unit costs using imported high-tech parts and components from other
industrial economies (notebook computers, mobile phones, etc.).
Table 7.1 displays first the top and bottom eight industries by
degree of import competition from the People’s Republic of China in
1994 (the beginning of the estimation period).4 In the textile industry,
where firms are considered to have comparative advantage, the degree
of import competition was already strong in 1994: of Japan’s import of
textile products, 49% came from the People’s Republic of China. That
share continued to increase, reaching 77% in 2005. More strikingly, the
largest increase in the People’s Republic of China’s share of Japanese
imports is in office and service industry machines, rising from 19% in
1994 to 76% in 2005. Correspondingly, in the industries in which the
People’s Republic of China’s share increased, there was a decline in the
share of Asian newly industrializing economies and the United States.
In the bottom eight industries, an increase in the People’s Republic of
China’s share is palpable, with strong growth in electronic equipment
and semiconductor devices. Production networks between Japan and
the People’s Republic of China may explain an expansion in import in
those high-tech industries.

Table 7.1 Change of Import Competition by Source Economies

in Japan's Manufacturing Industry, 1994 and 2005

1994
People’s
Republic Asian
of China NIEs SEA US
Manufacturing, total 11.4 15.9 10.2 25.7
Top eight sectors in 1994
Coal products 68.9 13.2 0.0 2.7
Textile products 48.7 15.1 8.0 5.6
Miscellaneous ceramic, stone, 34.4 19.1 3.0 13.6
and clay products
Rubber products 33.4 18.3 10.1 15.7
continued on next page

4
Data from 1990 are used in an experimental stage, but the order-import completion-
exposed industries are roughly the same in 1994.
212 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 7.1 continued

1994
People’s
Republic Asian
of China NIEs SEA US
Leather and leather products 26.5 19.9 5.8 5.2
Electrical generating, transmission, 24.4 24.1 19.5 19.6
distribution, and industrial apparatus
Pig iron and crude steel 23.7 4.0 3.0 7.0
Office and service industry machines 19.4 16.5 21.6 22.1
Bottom eight sectors in 1994
Chemical fibers 1.2 48.9 2.7 26.4
Petroleum products 1.0 22.4 12.0 6.2
Electronic equipment and electric 0.6 3.1 0.6 63.9
measuring instruments
Pulp, paper, and coated and glazed paper 0.5 1.7 0.9 40.7
Semiconductor devices 0.4 41.7 8.2 49.1
and integrated circuits
Printing, plate making for printing, 0.4 26.0 1.0 64.7
and bookbinding
Tobacco 0.1 0.0 0.0 95.3
Motor vehicles 0.0 0.3 0.0 27.7

2005
People’s
Republic Asian
of China NIEs SEA US
Manufacturing, total 28.6 12.8 10.9 15.2
Top eight sectors in 1994
Coal products 92.2 1.5 0.0 0.5
Textile products 76.5 3.5 4.0 2.0
Miscellaneous ceramic, stone, 60.4 5.0 3.8 9.6
and clay products
Rubber products 58.4 6.9 17.2 5.5
Leather and leather products 46.5 1.8 2.7 2.0
Electrical generating, transmission, 47.2 8.5 17.2 10.2
distribution, and industrial apparatus
Pig iron and crude steel 29.7 6.7 1.5 1.7
Office and service industry machines 76.2 8.2 7.7 2.7
continued on next page
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 213

Table 7.1 continued

2005
People’s
Republic Asian
of China NIEs SEA US
Bottom eight sectors in 1994
Chemical fibers 13.9 34.2 15.5 13.3
Petroleum products 2.8 21.1 12.8 2.5
Electronic equipment and electric 10.5 3.5 4.0 38.8
measuring instruments
Pulp, paper, and coated and glazed paper 7.3 6.5 13.4 33.5
Semiconductor devices 7.9 48.2 19.1 18.9
and integrated circuits
Printing, plate making for printing, 13.5 11.2 4.5 23.3
and bookbinding
Tobacco 0.6 0.1 0.1 89.6
Motor vehicles 1.4 1.9 0.9 8.8

Change 1994–2005
People’s
Republic Asian
of China NIEs SEA US
Manufacturing, total 17.2 –3.1 0.7 –10.5
Top eight sectors in 1994
Coal products 23.3 –11.7 0.0 –2.2
Textile products 27.8 –11.6 –4.0 –3.6
Miscellaneous ceramic, stone, 26.0 –14.1 0.9 –4.1
and clay products
Rubber products 25.0 –11.5 7.2 –10.2
Leather and leather products 20.0 –18.0 –3.1 –3.3
Electrical generating, transmission, 22.8 –15.6 –2.3 –9.4
distribution, and industrial apparatus
Pig iron and crude steel 6.0 2.7 –1.5 –5.3
Office and service industry machines 56.7 –8.4 –13.9 –19.4
Bottom eight sectors in 1994
Chemical fibers 12.7 –14.7 12.8 –13.1
Petroleum products 1.8 –1.4 0.8 –3.7
Electronic equipment and electric 9.9 0.4 3.4 –25.1
measuring instruments
continued on next page
214 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 7.1 continued

Change 1994–2005
People’s
Republic Asian
of China NIEs SEA US
Pulp, paper, and coated and glazed paper 6.8 4.8 12.6 –7.1
Semiconductor devices 7.5 6.5 10.9 –30.3
and integrated circuits
Printing, plate making for printing, 13.1 –14.8 3.5 –41.4
and bookbinding
Tobacco 0.6 0.1 0.0 –5.8
Motor vehicles 1.4 1.7 0.9 –18.9
NIE = newly industrializing economy, SEA = Southeast Asia, US = United States.
Source: Japan Industrial Productivity 2013 database.

7.3 Data and Variables

7.3.1 Firm-Level Patent Data

Patent statistics as an indicator for innovative outputs have recently

become widely available to researchers because of the significant
progress made in data accessibility (e.g., United States National Bureau
of Economic Research patent data, Japan Patent Office, PATSTAT
Worldwide Patent Statistical Database). Patent statistics carry important
invention-related information such as bibliographic data, backward and
forward citations, technology fields, name of inventor, and usefulness.
However, it has been well-documented from survey-based studies that
not all patents are in use (but are “sleeping”). In Japan, approximately
60% of pharmaceutical patents are reportedly not currently in use
(Nagaoka, Motohashi, and Goto 2010).5 Rather, firms obtain patents as a
defensive blocking mechanism in response to technology competition.6
“Blocking” patents might protect a firm’s once-exclusive market as it
becomes commercialized. This project, for the first time in the literature,

5
More generally, it is more common in the discrete technology industries. In the
pharmaceutical industry, R&D can take as long as 10–15 years before new drugs can
be introduced into the market. Hence, there is a substantial number of patents for
drugs that are still in the process of R&D and not yet in the market.
6
It is important to note that those unused patents may simply reflect firms that lack
the internal assets to commercialize or are searching for licensees.
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 215

empirically relates this unexploited nature of patent holdings to import

competition from a low-wage economy.
For this purpose, we extracted the relevant data from a Japanese
firm-level survey—the Basic Survey of Japanese Business Structure and
Activities, conducted by the Ministry of Economy, Trade and Industry7—
covering the period 1994–2005.
The firm-level patent usage data are then merged with industry-
level exposures to the People’s Republic of China’s import competition,
resulting in a unique dataset for the following aspects. First, it provides
a panel dataset of patent usage as it relates to competitive pressures. The
available surveys tend to report single-year responses, only depicting
the static nature of patent usage.8 Using a panel of firm-level data offers
the perspective of within-firm variations of patent usage in response to
import competition. Second, the data period is long enough to cover the
People’s Republic of China’s changing comparative advantage from more
labor-intensive to more skill- and technology-intensive goods. Third,
using panel data allows firm-specific effects to be included, because
(unobserved) managerial skills (assuming time-variant intrafirm
elements) can be controlled, along with industry and year fixed effects.
Clearly, in a cross-sectional setup, this cannot be controlled.
Based on firm-level information, we created the patent usage
variables as shown in Table 7.2. In short, for each firm, we count the
patents owned (PAT), the patents in use (USE) and patents that are
not in use (NON-USE). Within PAT-USE, we have information for the
number of patents based on internal inventions (DEV), and the number
of patents that are licensed out (LICENSE). These variables form the
dependent variables in the regression analysis that follows.

7
This survey is governed by the Japanese Statistics Act, and failure to reply results
in a fine. The survey sample is restricted to firms that have more than 50 employees
and capital of more than 30 million yen. It collects firms’ accounting information
(sales, employment, employment compensation, the number of establishments,
R&D spending, exports, and imports). The industry classification is available at a
three-digit level. For our purpose of analyzing the impact of import competition,
however, we restricted the sample to only manufacturing firms. All individual firms
are assigned unique identifiers, making it possible to track the operations of the same
firms over time (the panel data).
8
Motohashi (2008) uses the data from the Survey of Intellectual Property Activities
by the Japan Patent Office conducted in 2001 in order to classify patent usage. It
was found that some of the patents are withheld by firms wishing to use them (or
license them out) in the future. Others are kept because a firm needs them for future
licensing negotiations. This practice is common in the electronics industry where
cross-licensing occurs more frequently.
216 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 7.2 Patent Usage and Variable Definitions

Variable Symbol Brief Explanation and Definitions

Patent owned PAT The count of patents owned (including those
purchased and cross-licensed) reported by
a firm in a given fiscal year. This includes the
cumulative count of patents owned by firms,
not just patents for which application has
been made in a given year.
Use (including USE Those patents currently in use.
licensed out)
In-house DEV Patents based on internal inventions that
inventions are in use.
–use NON-USE Defined as PAT minus USE, including
blocking and future commercial use/
negotiation.
Licensed out LICENSE Total count of patents that are licensed out.
Domestic and international segregation is
available, as well as the amount of money
received.
Source: Authors.

It is important to note several limitations. First, the patent statistics in

our data are a patent pool: all patents in which the firms have ownership.
Empirical work that uses patent statistics collected from a patent office
normally covers those patents for which an application has been made,
as well those patents that have been granted to the firm. In our data, all
patents are presumably those granted (because the survey question asks
how many patents a firm owns, rather than patents that have been applied
for or are being granted). Since patent applications can indicate firms’
innovative efforts, our measure may underestimate them.
Second, our patent data are simply the count. Other studies
employing patent statistics usually weight the patent count to its
(backward and forward) citations, thus controlling for patent quality.
The higher-quality (or sometimes more basic) inventions attract more
forward citations than lower-quality inventions (sometimes referred
to as “patent thickness”). Without the ability to link our data on firm-
owned patents with the citation information, we are unable to account
for this quality dimension.
Third, our data do not adjust for the depreciation rate of outdated
patents. It is appropriate to adjust for this, because some firm-held
patents can become obsolete. However, with no identification of the
grant (or application) date of each patent, the deprecation rate cannot
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 217

be applied in our data. We therefore look at the growth rate of each

patent usage (rather than a simple count), hoping to minimize the bias
coming from the nondepreciation of the patents.

7.3.2 Japan Industrial Productivity Data

Industry-level variables used in the regression analysis are mainly

sourced from Japan Industrial Productivity (JIP) Database (JIP 2013)
stored in the online database in the Research Institute of Economy,
Trade and Industry in Japan.9 The JIP dataset is organized at the three-
digit industry level (52 manufacturing industries).

Import Competition from the People’s Republic of China

We use the value of imports originating from the People’s Republic of
China as a share of total world imports as a measure of the exposure to
the country’s import competition in a given JIP industry j:

Chinese imports𝑗𝑗
CHM𝑗𝑗 = . (1)
Imports𝑗𝑗

We also employChinese imports𝑗𝑗 method of constructing import

the conventional
CHM𝑗𝑗 =
penetration by normalizing the 𝑗𝑗People’s Republic of China’s import on
Imports
domestic absorptionCHM
(i.e., domestic
Chinese imports𝑗𝑗 + imports
absorption = value added
𝑗𝑗 =
– exports).10 (Value Added𝑗𝑗 + Imports𝑗𝑗 − Exports𝑗𝑗

Chinese imports𝑗𝑗
CHM𝑗𝑗 = . (2)
(Value Added𝑗𝑗 + Imports𝑗𝑗 − Exports𝑗𝑗
∆ln(PAT)𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑖𝑖 + 𝛼𝛼𝑗𝑗𝑗𝑗 + 𝛽𝛽1 CHM𝐽𝐽𝐽𝐽−5 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖 .

Instrumental Variable
While our (PAT)𝑖𝑖𝑖𝑖𝑖𝑖 =for
∆lnmotivation 𝛼𝛼𝑖𝑖 the
+ 𝛼𝛼𝑗𝑗𝑗𝑗 + 𝛽𝛽1 CHM
empirical 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖to. estimate the
𝐽𝐽𝐽𝐽−5 + is
analysis
causal effects of the People’s Republic of China’s import competition
on patent outputs, we encounter a possible endogeneity problem. Firm-
level innovative activity for reasons other than import competition

9
See the appendix for further details on the JIP Database 2013 (http://www.rieti
.go.jp/en/database/JIP2013/).
10
“Value added” is defined as the difference between gross output and intermediate
inputs. Gross output is measured as the sum of industry shipment, revenues from
repairing and fixing services, and revenues from performing subcontracting works.
Intermediate inputs are defined as the sum of raw materials, fuels, electricity, and
subcontracting expenditure.
218 Trade Adjustment in Asia: Past Experiences and Lessons Learned

may also shape trade flows, altering the degree of import competition
in the industry (e.g., more innovative firms might opt to do more
offshoring to the People’s Republic of China in order to facilitate their
innovative home operations). For the same reason, the reverse causality
is also a possibility. Imports from the People’s Republic of China may
be correlated with industry-wide technology shocks (to some degree,
industry-specific fixed effects may take care of this concern, but it might
not be sufficient). This makes ordinary least squares (OLS) estimators
biased and inconsistent.
We use a measure of the People’s Republic of China’s (labor)
productivity as an instrument for the endogenous import variables
in the technology equation. This implied volatility strategy extracts
any exogenous variations affecting the People’s Republic of China’s
export supply capacity, while indirectly affecting the level of innovative
activity only through the intensified import competition in Japan. This
instrument is inspired by the use of an instrument in other studies.
Autor et al. (2015) used the exposure to the People’s Republic of China’s
import competition of eight developed economies11 as instruments
to measure the exposure of the United States to People’s Republic of
China’s imports. The motivation for their implied volatility strategy was
to extract supply-side productivity elements in the People’s Republic of
China’s export performance. However, as the authors pointed out, their
instrument faces a validity challenge, whereby industry technological
changes among those developed economies must be separate incidents.
In other words, the technological diffusions must be limited across those
high-income economies. In our implementation of the implied volatility
strategy, we directly use the productivity measure (labor productivity) of
the People’s Republic of China’s industries, which undoubtedly has been
behind the surge in the country’s export growth yet is indirectly related
to firm-level innovative activity. These data are extracted from the China
Industrial Productivity (CIP) Database.12 There is no strict industry
correlation between CIP and JIP industries, so we arbitrarily assigned
the corresponding CIP manufacturing industries to 52 JIP industries.

7.4 Empirical Specification and Results

We use the following linear specification to relate firm-level patent
growth (for different patent usages separately) to the exposure of the
People’s Republic of China’s import competition in industries:

11
Australia, Denmark, Finland, Germany, Japan, New Zealand, Spain, and Switzerland.
12
The CIP Database 2015 is available at http://www.rieti.go.jp/en/database/CIP2015
/index.html.
 Chinese imports𝑗𝑗
CHM𝑗𝑗 =
(Value Added𝑗𝑗 + Imports𝑗𝑗 − Exports𝑗𝑗
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 219

∆ln(PAT)𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑖𝑖 + 𝛼𝛼𝑗𝑗𝑗𝑗 + 𝛽𝛽1 CHM𝐽𝐽𝐽𝐽−5 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖 . (3)

where subscripts i, j, and t denote firm, industry, and time. For each firm
i, we have the count of patents owned (PAT), the count of patents in use
(USE), and patents that are not in use (NON-USE). Within the group
USE, we also have information broken down by patents that are based
on in-house inventions (DEV). We also have the count of patents that are
licensed out (LICENSE). These variables form the dependent variables
separately in the regression analysis that follows.
The dependent variable is the 5-year (log) change in the patent
usage categories as an indicator of firms’ innovative activity. An
explanatory variable, CHMit-5, is in level for the period t–5. This linear
specification slightly differs from that in Bloom, Draca, and Van Reenen
(2016), who use the 5-year log changes in both dependent (technology)
and explanatory (exposure to the People’s Republic of China’s import
competition) variables. The formulation of equation (3) is preferred in
our data; it is intuitively more appealing because creating technology
(and filing for patents) requires more time.13 This specification tests
the subsequent firms’ innovative reaction to the People’s Republic of
China’s import competition experienced in the period t–5.14 Aghion et al.
(2005) and Amiti and Khandelwal (2013) also use a specification similar
to equation (3).
The baseline specification also includes both firm fixed effects (αit )
and industry-year fixed effects (αj ), to purge invariant shocks common
in the respective dimensions (such as the unobserved managerial
techniques within firms), as well as an industry-specific propensity
to patent. It has been concretely reported that some industries are
intrinsically prone to produce more patents than other industries
because of effective patent enforcement (chemical and pharmaceutical).
We also form the patent production function to include other
explanatory variables (in log form), which are drawn from the knowledge
production function that treats patents as knowledge output and other
firm characteristics as knowledge inputs: employment, age of firm, and
R&D ratio to sales (R&D intensity).

13
Growth rate is also preferred for a technical reason. Our data on the patent count
include the cumulative number of patents in which firms claim ownership. Hence,
by using growth rate, we only account for newer patents while discounting older
patents.
14
Even using the same specification as Bloom, Draca, and Van Reenen (2016), it turns
out that the estimation results are quite similar. This goes to show the persistent
impact of the People’s Republic of China’s import competition on the technology
variables.
220 Trade Adjustment in Asia: Past Experiences and Lessons Learned

7.5 Results
Table 7.3 presents the benchmark results. We ran a set of regressions in
OLS with firm and industry-year fixed effects. To aid the interpretation
of the main results, descriptions of key variables are presented in the
appendix (Tables A7.1 and A7.2). Column (1) of Table 7.3 indicates that
import competition from the People’s Republic of China provides an
overall inducement for more innovative activities among Japanese firms,
although its estimated effect seems to be relatively smaller than the one
found in Bloom, Draca, and Van Reenen (2016): a 10-percentage-point
increase in the People’s Republic of China’s import competition would
result in a 0.37% increase in firm-level patents. Across the results, there
is a visible position effect of import competition, with the exception of
PAT-DEV and LICENSE.15
The most interesting finding is that the People’s Republic of China’s
import competition also generates more unused patents (NON-USE). It
appears that the estimated coefficient is consistently larger than the one

Table 7.3 Import Competition from the People’s Republic of China

and Patent Usage, 1994 and 2005
OLS
(1) (2) (3) (4) (5)
PAT USE DEV NON-USE LICENSE
CHMjt-5 0.037*** 0.013* –0.019*** 0.026*** –0.001
(0.008) (0.007) (0.006) (0.007) (0.001)
Constant –0.552*** –0.353* 0.359** –0.447** 0.106***
(0.191) (0.198) (0.169) (0.171) (0.033)
Firm fixed effects Yes Yes Yes Yes Yes
Industry-year Yes Yes Yes Yes Yes
fixed effects
R-squared 0.394 0.346 0.342 0.298 0.289
N 35,200 35,200 35,200 35,200 35,200
OLS = ordinary least squares
Notes: *** denotes 1% significance; ** denotes 5% significance; * denotes 10% significance. Estimation is by
ordinary least squares with standard errors clustered by industry. The dependent variable is in 5-year log
differences of each patent usage type. Imports from the People’s Republic of China as a fraction of total
industry imports represent an explanatory variable. All columns include a full set of firm and industry-year
fixed effects.
Source: Authors’ calculation.

15
In fact, it is puzzling to see that the intensified import competition from the People’s
Republic of China would actually lower the rate of in-house invention patents, while
it has no statistically significant impact on patents designed for licensing out.
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 221

estimated for a USE equation (column (2) in Table 7.3): a 10-percentage-

point increase in the People’s Republic of China’s import competition
would result in a 0.26% increase in unused patents (versus a 0.13%
increase in patents that are in use). Notwithstanding a reservation about
the limitation in this variable (i.e., not all unused patents are used for the
purpose of “blocking”), the evidence suggests that Japanese firms would
undertake more defensive reactions to the increased import competition.
The regression result indicates that lowering the People’s Republic of
China’s import competition would trigger more patents based on in-house
inventions ( judging from the negative sign in a DEV regression, column
(3) in Table 7.3) In addition, the People’s Republic of China’s import
competition has no statistically significant effects on patents designed for
licensing out (LICENSE), as shown in column (4).
These results and associated interpretations are reinforced upon
taking an instrumental approach (Table 7.4). In the first-stage regression
(not shown), labor productivity has a statistical significance that is on
the level of the import competition.16 The estimated coefficients in all
regressions now show larger effects as compared to the OLS estimates.17

Table 7.4 Import Competition from the People’s Republic of China

and Patent Usage (Implied Volatility Regressions), 1994 and 2005
Instrumental Variable
(1) (2) (3) (4) (5)
PAT USE DEV NON-USE LICENSE
CHMjt-5 0.113*** 0.064*** –0.073*** 0.077*** –0.009***
(0.004) (0.004) (0.004) (0.004) (0.001)
Constant –1.017*** –0.524*** 0.366*** –0.828*** 0.065**
(0.090) (0.091) (0.094) (0.098) (0.030)
Firm fixed effects Yes Yes Yes Yes Yes
Industry-year Yes Yes Yes Yes Yes
fixed effects
N 35,200 35,200 35,200 35,200 35,200
Notes: *** denotes 1% significance; ** denotes 5% significance; * denotes 10% significance. Estimation is by
ordinary least squares with standard errors clustered by industry. The dependent variable is in 5-year log
differences of each patent usage type. Imports from the People’s Republic of China as a fraction of total
industry imports represent an explanatory variable. All columns include a full set of firm and industry-year
fixed effects.
Source: Authors’ calculation.

16
A full set of tests needs to be carried out to establish the validity of instruments.
17
In Bloom, Draca, and Van Reenen (2016), similar results were obtained.
222 Trade Adjustment in Asia: Past Experiences and Lessons Learned

In Table 7.5, the empirical specification follows a form of the

conventional knowledge production function, treating patents
as knowledge outputs. Even after we control for relevant firm
characteristics, the People’s Republic of China’s import competition
remains positive and statistically significant. With respect to firm
size (measured by the number of employees), it indicates that smaller
firms obtain more patents, and older firms (in terms of the age of the
firm) engage in more innovative activity (interestingly, the estimated
coefficients for firm characteristics remain much larger than a variable
of capturing the level of import competition from the People’s Republic
of China): a 10-percentage-point decrease in employment leads to a 4.4%
increase in innovative activity. Other than a PAT regression, we found
the effect of the People’s Republic of China’s competition to be positive
and statistically significant in a NON-USE regression (column (4)).
Conditioned on relevant firm characteristics, the People’s Republic of

Table 7.5 Import Competition from

the People’s Republic of China and Patent Usage
(OLS with Firm-level Characteristic Controls), 1994 and 2005

(1) (2) (3) (4) (5)

PAT USE DEV NON-USE LICENSE
CHMjt-5 0.031*** 0.010 –0.012* 0.021*** –0.000
(0.008) (0.007) (0.006) (0.007) (0.001)
Log(Emp)it-5 –0.445*** –0.113 0.023 –0.532*** 0.029
(0.096) (0.098) (0.106) (0.098) (0.028)
Log(Age)it-5 0.517*** 0.326*** –0.617*** 0.372*** –0.089***
(0.109) (0.110) (0.139) (0.134) (0.029)
Log(R&D)it-5 –0.253*** –0.372*** –0.696*** –0.032 –0.069**
(0.056) (0.101) (0.118) (0.075) (0.026)
Constant 0.276 –0.714 2.564*** 1.307** 0.266
(0.597) (0.589) (0.556) (0.634) (0.165)
R squared 0.398 0.349 0.350 0.301 0.291
N 35,164 35,164 35,164 35,164 35,164
OLS = ordinary least squares.
Notes: *** denotes 1% significance; ** denotes 5% significance; * denotes 10% significance. Estimation is
by OLS with standard errors clustered by industry. The dependent variable is in 5-year log differences of
each patent usage type. Imports from the People’s Republic of China as a fraction of total industry imports
represent an explanatory variable. All columns include a full set of firm and industry-year fixed effects.
Source: Authors’ calculation.
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 223

China's import competition would produce patents of a more defensive

nature (unused patents) among Japanese firms.
Table 7.6 sequentially introduces the import competition indicators
from other economies. We introduced import competition from Asian
newly industrializing economies separately for those from high-
income OECD economies (including the United States and high-wage
European economies).18 Overall, the main results remain the same.
Increased import competition would make Japanese firms pursue more
patenting, while import competition from other high-wage economies
has no statistical significance. These findings conform to those found in
Bloom, Draca, and Van Reenen (2016). The theoretical intuition drawn
from the trapped-factor model is that import competition from high-
wage economies is not a substitute for old products, which do not create
incentives for innovation. There is positive and statistically significant
effect on non-use of patents (NON-USE), while in other regressions the
sign for CHM changes or loses statistical significance as compared to
the benchmark estimation.

Table 7.6 Import Competition from

the People’s Republic of China and Patent Usage
(OLS with Other Import Competition Variables), 1994 and 2005

(1) (2) (3) (4) (5)

PAT PAT USE USE NON-USE
CHMjt-5 0.036*** 0.033*** 0.015*** 0.007 0.025***
(0.008) (0.009) (0.005) (0.009) (0.008)
NIEjt-5 –0.003 0.005 –0.004
(0.008) (0.008) (0.005)
Highjt-5 –0.007 –0.008**
(0.006) (0.004)
Constant –0.494** –0.144 –0.471* 0.126 –0.351*
(0.238) (0.366) (0.233) (0.336) (0.200)
R-squared 0.394 0.395 0.346 0.347 0.298
N 35,200 35,200 35,200 35,200 35,200
continued on next page

18
In an experimental stage, import competition from other low-wage economies (such
as those in mainland Southeast Asia) was included, but it turns out that it is not
important, and does not change the estimated coefficient for CHM.
224 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 7.6 continued

(6) (7) (8) (9) (10)
NON-USE DEV DEV LICENSE LICENSE
CHMjt-5 0.023*** –0.021*** –0.019** –0.002* 0.000
(0.007) (0.005) (0.008) (0.001) (0.001)
NIEjt-5 –0.007 –0.002
(0.007) (0.001)
Highjt-5 –0.005 –0.000 0.002*
(0.008) (0.005) (0.001)
Constant –0.170 0.512** 0.365 0.154*** –0.003
(0.408) (0.214) (0.390) (0.032) (0.077)
R-squared 0.298 0.342 0.342 0.290 0.290
N 35,200 35,200 35,200 35,200 35,200
OLS = ordinary least squares.
Notes: *** denotes 1% significance; ** denotes 5% significance; * denotes 10% significance. Estimation is
by OLS with standard errors clustered by industry. The dependent variable is in 5-year log differences of
each patent usage type. Imports from the People’s Republic of China as a fraction of total industry imports
represent an explanatory variable. All columns include a full set of firm and industry-year fixed effects.
Source: Authors’ calculation.

7.6 Conclusion
This chapter examined the impact of the People’s Republic of China’s
import competition on the innovation responses of a panel of Japanese
manufacturing firms for the period 1994–2005. Based on the unusually
detailed firm-patent dataset, we have uncovered several heterogeneous
dimensions of the impact of innovation in the case of import competition
from the People’s Republic of China. First, we found that, while increased
imports from the People’s Republic of China have induced Japanese firms
to take out more patents, they are mostly of lower quality (i.e., patents
with zero-forward citation). This was inferred as evidence suggesting
that Japanese firms have increased the defensive nature of patents
in order to protect their core inventions. This is similar to a strategy
followed by firms in “continuous” technology-intensive industries in
the field of information and communication technology; to build up
the patent fence in order to deter new entrants in the technology field.
This finding coincides with a sector that has been subject to intensified
import competition from the People’s Republic of China over the past
2 decades.
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 225

Second, when the sample of firms is split into globally engaged firms
with positive importing (and exporting) activity and domestic-oriented
firms, the former group has responded positively to import competition
from the People’s Republic of China by increasing its R&D intensity. Our
interpretation of this result is that Japanese firms (and presumably more
innovative firms) have built up their innovation capacity while moving
away from low-cost manufactured goods, in which the People’s Republic
of China has more comparative advantages. In contrast, such effects are
consistently muted for firms with a domestic market focus. These types
of firms are completely insulated from the People’s Republic of China’s
import competition.
226 Trade Adjustment in Asia: Past Experiences and Lessons Learned

References
Aghion, P., N. Bloom, R. Blundell, R. Griffith, and P. Howitt. 2005.
Competition and Innovation: An Inverted U Relationship. Quarterly
Journal of Economics. 120 (2). pp. 701–28.
Amiti, M., and A. Khandelwal. 2013. Import Competition and Quality
Upgrading. Review of Economics and Statistics. 95 (2). pp. 476–90.
Athukorala, P., and N. Yamashita. 2006. Production Fragmentation and
Trade Integration in a Global Context. North American Journal of
Economics and Finance. 17 (4). pp. 233–56.
Autor, D., D. Dorn, G. Hanson, and J. Song. 2015. Trade Adjustment:
Worker Level Evidence. Quarterly Journal of Economics. 129 (4).
pp. 1799–860.
Bloom, N., M. Draca, and J. Van Reenen. 2016. Trade Induced Technical
Change? The Impact of Chinese Imports on Innovation, IT and
Productivity. Review of Economic Studies. 83 (1). pp. 87–117.
Boldrin, M., and D. K. Levine. 2013. The Case against Patents. Journal of
Economic Perspectives. 27 (1). pp. 3–22.
Motohashi, K. 2008. Licensing or Not Licensing? An Empirical Analysis
of the Strategic Use of Patents by Japanese Firms. Research Policy.
37 (9). pp. 1548–55.
Nagaoka, S., K. Motohashi, and A. Goto. 2010. Patent Statistics as an
Innovation Indicator. In B. H. Hall and N. Rosenberg, eds. Handbook
of the Economics of Innovation, Volume 2. Amsterdam: Elsevier.
pp. 1083–127.
Schott, P. K. 2008. The Relative Sophistication of Chinese Exports.
Economic Policy. 23 (53). pp. 5–49.
Thoenig, M., and T. Verdier. 2003. A Theory of Defensive Skill Biased
Innovation and Globalization. American Economic Review. 93 (3).
pp. 709–28.
Wood, A. 1994. North-South Trade, Employment and Inequality: Changing
Fortunes in a Skill-Driven World. Oxford: Clarendon Press.
The Rise of the People’s Republic of China and Its Competition Effects on Innovation in Japan 227

Appendix A7
Table A7.1 Descriptive Statistics

PAT NON-USE USE DEV LICENSE CHM

Year Mean Mean Mean Mean Mean Mean
1994 109.1 77.2 35.0 27.9 0.4 10.3
1995 107.2 75.2 32.2 27.8 0.3 12.5
1996 121.1 86.7 34.6 30.7 0.4 13.7
1997 98.6 64.3 34.4 91.7 0.6 15.1
1998 106.4 68.1 38.3 101.2 0.5 16.0
1999 115.2 73.3 41.9 108.8 0.6 16.8
2000 50.2 36.6 24.1 40.7 0.9 18.2
2001 123.3 75.4 47.9 38.9 0.8 20.5
2002 124.3 78.1 46.3 38.8 0.7 23.0
2003 142.5 91.9 50.5 41.4 0.7 24.1
2004 141.9 88.7 53.2 43.1 2.1 26.5
2005 130.9 81.0 49.9 43.2 2.3 28.9

PAT USE NON-USE LICENSE # of Firms

Year Sum Sum Sum Sum Sum
1994 46,908 16,118 30,790 200 6,374
1995 49,417 16,300 33,117 79 6,637
1996 53,485 17,650 35,835 300 6,614
1997 53,352 17,600 35,752 800 6,464
1998 52,119 17,200 34,919 228 6,513
1999 55,909 18,692 37,217 247 6,447
2000 43,166 9,800 33,366 344 6,340
2001 50,000 39,726 10,274 938 6,415
2002 47,000 24,670 22,330 301 6,269
2003 48,061 20,155 27,906 350 5,764
2004 47,166 43,000 4,166 8,930 6,088
2005 42,662 34,000 8,662 10,000 5,937
Source: Authors’ calculation.
228 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table A7.2 Descriptive Statistics for Variables Used in Regressions

Variable Unit Obs Mean Std. Dev. Min Max

PAT Count 75,862 113.8 1,102.1 0 55,909
USE Count 75,862 40.4 433.2 0 43,000
NON_USE Count 75,862 74.5 819.2 0 42,662
DEV Count 75,862 53.1 653.8 0 55,909
LICENSE Count 75,862 0.8 49.3 0 10,000
Emp. total Unit 75,862 629.0 2,424.1 50 80,500
R&D Value in yen 75,862 888.4 10,266.2 0 527,359
expenditures
Est. year Year 75,857 1,951.0 111.0 0 2,006
CHM Percentage 75,862 18.6 16.2 0.02 98
Source: Authors’ calculation.
 8
Trade Reform, Managers,
and Skill Intensity:
Evidence from India1
Pavel Chakraborty

8.1 Introduction
International trade economists have long been interested in
understanding the distributional implications of globalization or trade
liberalization or product market competition. One of the crucial aspects
of such distributional effects, which have received a lot of attention,
especially from the 1990s onward, is how such forces divide the labor pie
into skilled (or nonproduction) and unskilled (or production) workers.
In other words, does an increase in trade participation or exposure to
international markets result in an increase in returns for skilled or less
skilled workers?
The theoretical underpinning of such an important empirical
question originates from the predictions of a well-known theorem in
international trade: Stolper–Samuelson. In a model with two factors, it
states that “for a rise in the relative price of a good, it will lead to a rise in
the return to that factor which is used most intensively in the production
of the good, and conversely, to a fall in the return to the other factor”
(Stolper and Samuelson 1941). For example, let us denote skilled and
unskilled labor as two factors. Now, as countries reduce trade barriers,
the Stolper–Samuelson theorem predicts a rise in unskilled-labor wages
and a fall in skilled-labor wages in developing countries (as they have
a big pool of relatively less skilled workers). The opposite is true in
the case of skill-rich countries. In other words, the theorem points out
that exposure to international trade or world markets can significantly

1
This chapter is based on Chakraborty and Raveh (2018).

229
230 Trade Adjustment in Asia: Past Experiences and Lessons Learned

affect the distribution of resources within the country and can generate
substantial distributional conflict.
To investigate whether this is the case, a significant number
of studies have tried to establish a causal link between the effects
of competitive forces (in the form of trade liberalization) and skill
premium or some other measure of wage inequality between skilled
and unskilled workers: (i) Argentina (Galiani and Sanguinetti 2003;
Bustos 2011); (ii) Brazil (Pavcnik et al. 2004; Gonzaga, Menezes-Filho,
and Terra 2006; Menezes-Filhoz and Muendler 2011; Araújo and Paz
2014; Krishna, Poole, and Senses 2014); (iii) Chile (Beyer, Rojas, and
Vergara 1999); (iv) the People’s Republic of China (Chen, Yu, and Yu
2017); (v) Colombia (Attanasio, Goldberg, and Pavcnik 2004; Goldberg
and Pavcnik 2005); (vi) India (Chamarbagwala 2006; Kumar and Mishra
2008; Chamarbagwala and Sharma 2011; Mehta and Hasan 2012);
(vii) Indonesia (Smith et al. 2002; Amiti and Davis 2012; Amiti and
Cameron 2012); (viii) Mexico (Feenstra and Hanson 1997; Revenga 1997;
Harrison and Hanson 1999; Feliciano 2001; Verhoogen 2008; Frías,
Kaplan, and Verhoogen 2009, 2012); (ix) Morocco (Currie and Harrison
1997); (x) Turkey (Krishna, Mitra, and Chinoy 2001); (xi) Viet Nam
(McCaig 2011); and (xii) Latin American countries (Behrman, Birdsall,
and Székely 2000; Haltiwanger et al. 2004).
The primary reason for such an overwhelming number of studies
focusing on developing or emerging economies is that during the last 3
decades or so, many developing countries, most notably Latin American
countries in the 1980s and early 1990s, India in the early 1990s, and the
People’s Republic of China joining the World Trade Organization in 2001,
underwent a significant trade liberalization process that substantially
increased their exposure to international markets. The main conclusion
that emerges from these studies is that the skill premium rose in
developing countries due to exposure to international trade. This is
puzzling in a Heckscher–Ohlin context as developing countries have a
comparative advantage in producing low-skill-intensive goods.
A handful of researchers have also investigated the demand for
different kinds of workers in the case of developed countries, but between
exports and non-exporters: (i) France (Biscourp and Kramarz 2007); (ii)
Germany (Baumgarten 2013); (iii) Hungary (Koren, Csillag, and Kollo
2019); (iv) Portugal (Martins and Opromolla 2010); and (v) the United
States (Feenstra and Hanson 1996; Bemard and Jensen 1997). Analysis
across this set of countries finds strong evidence that an exporter wage
gap, conditional on workers’ skill levels, contributed to the growth in
wage inequality. This finding is consistent with recent heterogeneous-
firm trade models that feature an exporter wage premium as well as
variability of the premium with respect to increasing trade liberalization.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 231

Given this background, one issue that is currently at the center of

economic debates regarding the dynamics of the labor market is how
trade reform or exposure to international market(s) or product market
competition affects a firm’s demand for managers,2 which in turn affects
productivity and performance.3 The literature on firms’ managerial
practices or demand for managers originates from a seminal paper by
Garicano (2000).
He asks a simple question: What does a firm do? A firm solves
problems. Problems arise during different stages of production, and
managers solve not-so-common problems, whereas nonmanagers
take care of routine problems. The demand for managers rises as the
ratio of not-so-common problems increases. Garicano (2000) argues
that this happens when a firm invests in technological deepening (of
the production function). In other words, managerial inputs act as
complements to technological inputs. Therefore, with greater adoption
of technology (or technological inputs), the demand for skilled labor
(or managers) increases. Caliendo and Rossi-Hansberg (2012) use this
framework to show that participation in export markets also increases
the demand for managers, as firms increasingly face new sets of not-so-
common problems.
In a related context, Acemoglu (2003) develops a model to analyze
the impact of international trade on wage premiums. He shows that
wage inequality can also happen through skill-biased technical change
because of increased international trade. And this may explain the rise
in wage inequality without a rise in the relative prices of skill-intensive
goods (both in the United States and less developed economies), which
is the usual intervening mechanism in standard trade models.
Putting these two issues together, I argue that trade reform—
or in my case a drop in input tariffs—can induce firms to adopt more
technologically intensive inputs. Adoption of high-tech inputs can
increase the demand for managers. On the other hand, managers make
up a sizeable proportion of skilled workers. Therefore, I hypothesize
that skill intensity may be a complementary channel through which
trade may result in an increase in the demand for managers.
Adopting the case of India, I empirically study this nexus in a
developing economy, through which we can unravel new dynamics that
emphasize the distinctive features of such an economy in this context.

2
See Chakraborty and Raveh (2018), Chen (2017), Caliendo and Rossi-Hansberg
(2012), and Marin and Verdier (2008, 2014).
3
Studies that link firm organization and managerial practices to firm performance
and productivity include Garicano and Rossi-Hansberg (2004, 2006), Bloom and Van
Reenen (2007, 2010), Bloom et al. (2013), and Bloom et al. (2014), among others.
232 Trade Adjustment in Asia: Past Experiences and Lessons Learned

All the previous studies investigating similar issues (skill intensity

or premium) focus on either and/or both (i) differential returns for
production and nonproduction workers; and/or (ii) skilled and unskilled
workers, where the workers are sorted according to the number of years
of education. I extend and complement the literature by focusing on
one niche aspect of the group of skilled workers: the managers. To see
whether managers can possibly represent skilled workers, I compute a
simple correlation between managerial compensation (by aggregating
firm-level data to industry) and the ratio of skilled workers or skilled
intensity (nonproduction workers/total number of employees). The
correlation coefficient is 0.56, suggesting that managers can fairly
represent the skilled workers group.
Managers are the section of workers who manage or are associated
with the production activities of a firm in the dataset that I exploit for this
chapter.4 The primary focus of this chapter is to investigate the effect of
trade liberalization, as compared with changes in tariffs, on the demand
for managers relative to nonmanagers, and in addition explore whether
skill intensity can act as a complementary channel.5 While previous studies
examined components related exclusively to the managerial side, such as
wages and bonuses (e.g., Cunat and Guadalupe 2009), very little attention,
if any, has been given to the inclusion of the nonmanagers’ side to consider
relative terms and within-firm inequality.6 I study the causal link and try
to identify the underlying mechanisms through which it operates.
I start by presenting a simple link between trade (both exports
and imports) and the relative demand for managers in a sample of
Indian firms for 1990–2011 (Figure 8.1).7 All three measures increase
steadily throughout the period exhibiting a correlation of 0.85 between
managerial compensation and exports, and 0.89 with imports. The surge
in trade is a consequence of the 1990s trade liberalization exercise in

4
I exclude any manager who is associated with any kind of administrative duties in a
firm, such as a human resources manager.
5
Managers are defined as any workers who manage at least one other worker (or
who is the sole worker in the firm), with nonmanagers accounting for the remaining
balance. I will further discuss this in detail in the empirical part.
6
An exception is Ma and Ruzic (2018). They study the impact of globalization on
executives’ income shares in firms in the United States using conditional correlations.
In contrast, I try to establish a causal link and empirically identify the underlying
channel, while examining a more general definition of managers.
7
The figure presents the yearly average of the share of total trade (exports plus
imports) in gross value added and the share of managerial compensation in total
labor compensation for a representative Indian manufacturing firm over 1990–2011.
I proxy for the relative demand for managers using the latter. I discuss both the
measures in more detail in the empirical part.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 233

Figure 8.1 Trade and Managerial Compensation, 1990–2011

Exports-Imports and Managerial Compensation
Indian Manufacturing Firms: 1990–2011

5,000 0.10

Export and Import (₹ million)

0.08

Managerial compensation
4,000

3,000 0.06

2,000 0.04

1,000 0.02

0 0.00
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
Year
Imports Managerial Compensation Exports

Note: The figure presents the average trade (exports and imports) values and the average
compensation share of managers, 1990–2011 (ρ = 0.85).
Source: Author’s calculations.

India, which I discuss further later; the increase in the compensation

share of managers is what I aim to investigate. I seek to understand
whether there is indeed a causal relation between the two. To
apprehend further whether such is the case, I divide the sample of firms
into importing and nonimporting firms and plot the relative demand
measure in Figure 8.2. The figure indicates that the surge (in the share of
managerial compensation) is almost an exclusive feature of the former
types. This motivates a focus on tariffs. To test whether, and how, the
latter creates a causal effect, I exploit the exogenous nature of India’s
1990s trade reform to study a rich dataset on Indian manufacturing
firms that uniquely disaggregates labor compensation to managers and
nonmanagers over a period of 2 decades.
I find a remarkably robust, persistent, and economically meaningful
negative effect that, in line with the findings in the initial analysis, is
entirely driven by input tariffs. The benchmark estimations indicate
that a 10% decrease in input tariffs increases the share of managerial
compensation (as well as their number) by approximately 0.5%–3.5%.
This effect is robust to considering various controls, specifications,
and estimation techniques. These results point to a quality-upgrading
mechanism reminiscent of Caliendo and Rossi-Hansberg (2012),8

8
Studying a model of heterogeneous firms with knowledge-based hierarchies,
they show that trade liberalization increases the number of management layers
in exporting firms, as managers can solve more efficiently problems arising from
increasing output than workers for whom costly knowledge needs to be acquired.
234 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 8.2 Managerial Compensation,

Importers and Nonimporters, 1990–2011
Managerial Compensation: Importers and Nonimporters
Indian Manufacturing Firms: 1990-2011
Managerial Compensation / Total Compensation
Nonimporter Importer
0.10

0.05

0.00
1990 1995 2000 2005 2010 1990 1995 2000 2005 2010
Year

Note: The figure presents the average compensation share of managers for importing and
nonimporting firms, 1990–2011.
Source: Author’s calculations.

adjusted to an importing-based economy: Firms import intermediate

inputs of higher quality and greater variety. These, together with the
products they produce, are embedded with new knowledge that in turn
increases the relative demand for workers with skills to manage that.9
Next, in investigating whether skill intensity can act as a
complementary channel resulting in demand for managers, my results
suggest that the phenomenon is particularly significant for firms below
halfv of the size distribution. In other words, as small and medium-sized
firms start to import high-quality intermediate inputs (as a result of the
trade reform) without any prior knowledge of how to use them in the
production technology, this results in a higher demand for managers
(relative to the top half of the size distribution) with suitable knowledge
to use them. This increase in the demand for managers materializes in
those sectors where there is a complementary effect of the drop in input
tariffs and a higher ratio of skill intensity.
This chapter is primarily related to the literature on trade
liberalization and the demand for skill or skill intensity in developing

9
Interestingly, this is in contrast to previous studies that examined developed
economies, pointing to a product market competition mechanism (e.g, Cunat and
Guadalupe 2009; Bloom, Sadun, and Van Reenen, 2010), hence emphasizing the
extent to which the case of a developing economy may present different dynamics
and provide new insights.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 235

economies. As discussed before, neoclassical trade theory, via the

Stolper–Samuelson theorem, predicts that trade liberalization increases
the demand for the abundant factor, which is expected to translate to an
increased relative demand for low-skill labor in developing economies.
Several studies, however, have documented the opposite (Goldberg
and Pavcnik 2007). Various explanations have been offered, including
trade-induced skill-biased technical change (Acemoglu 2003), credit
constraints (Bonfatti and Ghatak 2013), improved exports (Zhu and
Trefler 2005), import composition (Raveh and Reshef 2016), and
quality upgrading (Verhoogen 2008). I check whether skill intensity
can potentially be cited as one of the channels through which India
experiences an increase in the demand for managers due to the trade
liberalization episode.
Second, the chapter also contributes to literature regarding
offshoring and wage inequality. Feenstra and Hanson (1996, 1997)
show that purchasing an input from a foreign source can replace a task
previously done by a worker and therefore can lower wages. On the other
hand, the ability to use foreign inputs can raise wages (Grossman and
Rossi-Hansberg 2008). Autor, Levy, and Murnane (2003); and Ebenstein
et al. (2014) in the case of the United States; and Hummels et al. (2014)
for Danish firms show that workers whose occupations involve routine
tasks experience large wage drops with offshoring. I also find a similar
result: increased use of foreign inputs results in a demand for managerial
workers, but in the case of a developing country.
Last, the chapter is also closely related to the literature on how the
adoption of technologically intensive inputs induces a skill wage gap
between and within firms. However, as pointed out by Card and DiNardo
(2002), a central issue regarding such an association is the problem of
identifying a causal link between the adoption of new technology and
a rise in skilled workers’ wage. I use the drop in tariffs on intermediate
inputs because of the trade liberalization exercise in India in the 1990s
to identify a causal effect of technology adoption (in terms of adoption
of high-quality technologically intensive intermediate inputs) on wage
inequality.

8.2 Literature Review

As discussed earlier, a growing body of academic and policy debates on
the merits and demerits of liberalization have centered on the internal
distributional consequences and how they affect labor markets. This
section presents all the other evidence from India on the impact of
trade reform on wages. India offers an interesting case for studying the
236 Trade Adjustment in Asia: Past Experiences and Lessons Learned

effects of trade reform for a couple of reasons. First, the magnitude of

trade liberalization was very big (Kumar and Mishra 2008). The average
tariff drop in manufacturing was more drastic than trade liberalization
episodes in Latin American countries (e.g., Brazil, Colombia, and
Mexico). In addition to tariffs, India has also reduced nontariff barriers
since 1991. Second, trade reforms in India were exogenous and were in
response to a severe balance-of-payments crisis in 1991. The objective of
reducing trade barriers was based on the International Monetary Fund’s
conditionalities for assistance. Therefore, policy makers had less room
to cater to special lobby interests.
The first paper to investigate the issue of wage inequality in India
was by Chamarbagwala (2006). She uses data from the Employment and
Unemployment Schedule of the National Sample Survey Organization
for four rounds—1983–84, 1987–88, 1993–94, and 1999–2000—to
investigate India’s skill wage gap and gender wage differential during
the 2 decades that coincide with the economic liberalization in India.
Using a nonparametric methodology, she argues that economic
liberalization contributed to the widening of the skill wage gap. In other
words, demand for skilled labor increased, mostly due to skill upgrading
within industries. On the other hand, the paper shows that international
trade in manufacturing goods benefitted skilled men, but hurt skilled
women, whereas outsourcing of services generated a demand for both
male and female skilled workers. Dutta (2007) uses the same dataset
for the same time period but estimates wage regression models using
the augmented Mincer earnings equation controlling for human capital,
industry affiliation, and various other characteristics. She also concludes
that trade reforms have substantially increased wage inequality as the
relative wages of the unskilled workers fell considerably.
Kumar and Mishra (2008) use household survey data from the
Employment and Unemployment Schedule for the same four rounds
to estimate the effect of a drop in tariffs on industry wage structure.
However, since manufacturing is largely located in urban areas, they
focus their attention on workers only in urban areas. In contrast to the
abovementioned studies on India, they find that trade liberalization
has led to a decrease in wage inequality between skilled and unskilled
workers in India. This is because the magnitude of tariff reductions was
relatively larger in sectors with a higher proportion of unskilled workers.
Azam (2010), using microdata for 1983–2005, investigates the
role of the demand and supply of skilled workers in explaining the
rise in skill premium in India. The paper presents the following
findings. First, the tertiary (college)-secondary (high school) wage
premium increased in India during the 1990s and 2000s, and this
increase differs across age groups. Increases in wage premiums have
 Trade Reform, Managers, and Skill Intensity: Evidence from India 237

been driven mostly by younger age groups, while older age groups did
not experience any significant increase. Second, the increase in wage
premium was due to demand shifts in favor of workers with a tertiary
education, mainly between 1993 and 2004. He argues that the growth
rate of the demand for tertiary-educated workers relative to secondary-
educated workers was fairly stable in the 1980s and 1990s. This is due
to the increase in the relative supply of tertiary workers during 1983–
1993, which negated the demand shift. As a result, the wage premium
did not increase much. However, between 1993 and 1999, the growth
rate of the relative supply of tertiary workers decelerated and became
virtually stagnant between 1999 and 2004. This resulted in an increase
in the wage premium.
Chamarbagwala and Sharma (2011) investigate the relationship
between industrial delicensing, trade liberalization, and skill upgrading
during the 1980s and 1990s among manufacturing plants in India.
They use Annual Survey of Industries data to test whether industrial
delicensing during the 1980s and 1990s played a role in skill upgrading
(as measured by the employment and wage bill shares of white-collar
workers). Using both difference-in-differences as well as regression
discontinuity techniques, they find two important results. First,
industrial delicensing during the 1980s increased the relative demand
for skilled workers via capital- and output-skill complementarities;
second, trade liberalization did not play a major role in raising the
relative demand for skilled labor during the 1990s.
Last, Mehta and Hasan (2012) examine the effects of trade and
services liberalization on wage inequality in India. Their main finding
is that labor reallocations and wage shifts due to services reforms are
many times larger than those of manufacturing goods liberalization.
Additionally, the paper also highlights that (i) a large proportion (30%–
66%) of the increase in wage inequality is due to changes in industry
wages and skill premiums that cannot be empirically linked to trade
liberalization; and (ii) the bulk of the effects of trade liberalization do
not remain in interindustry wage shifts and skill premiums but are
subsumed by general equilibrium effects.
Overall, the evidence is mixed. The majority of the studies find
trade liberalization to have increased skill premium, whereas others
ascertain no effect. This chapter does not make any effort to investigate
the direct effect of trade reform on skill premium but looks at whether
trade reform affects the demand for managerial workers, where skill
intensity acts as an intermediary channel. In doing so, I find that
skill intensity can possibly be termed a complementary channel (in
increasing the demand for managerial workers) for firms below half of
the size distribution.
238 Trade Adjustment in Asia: Past Experiences and Lessons Learned

8.3 Firm-level Data

The firm-level data that I primarily use for this chapter are based on the
PROWESS database, constructed by the Centre for Monitoring Indian
Economy (CMIE). The PROWESS database contains information
on approximately 27,400 publicly listed companies, all within the
organized sector, of which almost 11,500 are in the manufacturing
sector. I examine firms belonging to the Indian manufacturing sector.
Firms are placed according to the five-digit 2008 National Industrial
Classification (NIC) level and are reclassified to the 2004 NIC level to
facilitate matching with the industry-level tariffs. The database reports
direct measures on a vast array of firm-level characteristics, including
sales, disaggregated trade components (imports and exports), research
and development (R&D) expenditures, technology transfers, production
factors employed, gross value added, assets, ownership, and others. In
addition, it covers both large and small enterprises; data for the former
types are collected from balance sheets, whereas the latter is based on
the CMIE’s periodic surveys of smaller companies.
PROWESS presents several features that make it particularly
appealing for the purposes of this study. It is in effect a panel of firms,
enabling their behavior to be studied over time. The (unbalanced) sample
covers up to 8,000 firms, across 108 (four-digit NIC) manufacturing
industries that belong to 22 (two-digit NIC) larger ones,10 over 1990–
2011, thereby covering the 1990s trade reform.
The unique feature of the dataset upon which the analysis is mainly
based is that it disaggregates compensation data to those received
by managers and nonmanagers, with a further disaggregation of
compensation to wages and bonuses. Specifically, the division is done
at three levels: nonmanagers, directors, and executives, with the last
two comprising the managers group. While the definition of the former
is that they do not manage other employees, directors are defined as
managers without executive powers, as opposed to executives, who
do possess such responsibilities. Executives include, for instance, the
chief executive officer, chief financial officer, and chairperson, whereas
directors cover positions such as divisional managers. In effect, directors
are considered middle management, whereas executives are the top
management. While there may be scope for subjective interpretation
of this distinction by firms, it does not affect the main analysis where I

10
In terms of composition, approximately 20% of the firms in the dataset are registered
as chemical and pharmaceutical industries, followed by food products and beverages
(13.74%), textiles (10.99%), and basic metals (10.46%).
 Trade Reform, Managers, and Skill Intensity: Evidence from India 239

aggregate executives and directors. These features enable me to study

the relative demand for managers and through that trace down the
underlying channel that affects it.
Table 8.1 presents a conditional correlation matrix of the share of
managerial compensation with exports and imports (with imports also
divided into four different categories: import of raw materials, import
of capital goods, import of stores and spares, and import of finished
goods). Column (1) shows that the total imports of a firm and share of
managerial compensation are significantly correlated at the 5% level.
Columns (2)–(5) divide total imports into the four categories. The
numbers indicate that the correlation is strongest in the case of import
of capital goods (0.03) followed by import of raw materials (0.01), with
no significance in the case of import of stores and spares and of finished
goods. I also do not find any significant correlation between the exports
of a firm (column (6)) and managerial compensation. Nonetheless, these
numbers are merely suggestive and not conclusive, unless we control for
any other policy effects and firm- and industry-level attributes.

Table 8.1 Correlation Matrix—Imports, Exports,

and Managerial Compensation

Import Import
Import of Import of Stores of
Total Capital of Raw and Finished Total MComp/
Imports Goods Materials Spares Goods Exports TComp
(1) (2) (3) (4) (5) (6) (7)
Total Imports 1.00
Import of 0.14* 1.00
Capital Goods
Import of 0.70* 0.02* 1.00
Raw Materials
Import of 0.12* 0.02* 0.16* 1.00
Stores and
Spares
Import of 0.71* 0.00 0.008* 0.00 1.00
Finished Goods
Total Exports 0.97* 0.009* 0.75* 0.11* 0.63* 1.00
Managerial 0.01* 0.03* 0.01* 0.002 –0.002 0.001 1.00
Compensation/
Total
Compensation
Notes: Numbers denote correlation coefficients. * denotes significance at 5% level.
Source: Author’s calculations.
240 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 8.3 Managerial Compensation,

across Industries, 1990–2011
Managerial Compensation: By Industry Groups
Indian Manufacturing Firms: NIC 2004 two-digit level
Managerial Compensation/Total Compensation
0.04

0.03

0.02

0.01
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Industry Group NIC 2004 two-digit level

NIC = National Industrial Classification.

Note: The figure presents the average compensation share of managers across the 2004 NIC
two-digit level industries, 1990–2011.
Source: Author’s calculations.

The dataset provides much variation across firms and industries in the
compensation characteristics of managers compared to nonmanagers. For
instance, Figure 8.3 plots the average share of managerial compensation
in total labor compensation across two-digit industries for 1990–2011.11 It
goes from a low of approximately 0.5% to a high of around 4%, and the
difference across industries is clearly observed. This is also noted when
measuring changes over time: Averaging annual changes over the same
period, I observe that while in some industries the average annual rate
of change is around 10%, in others it can get higher than 200%, thereby
providing quite large differences. When this translates to the firm level,
such variation will be even more prominent.
Last, the dataset has a relatively wide coverage, accounting for more
than 70% of the economic activity in the organized industrial sector,
and 75% (95%) of the corporate (excise duty) taxes collected by the
Government of India (Goldberg et al. 2010). In terms of trade, it covers
approximately 40%–45% of India’s total export and import activity,
presenting a reasonably good aggregate picture of India’s trade position.
In addition, it has been used in previous similar studies, providing
some reassurance of its relevance and applicability to the particular

11
Note that all industry-level categorizations done throughout the chapter are based
on the 2004 NIC classification.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 241

issues studied.12 All variables are measured in millions of Indian rupees,

deflated to 2005 using the industry-specific Wholesale Price Index, and
are outlined in the appendix. Table 8.2 presents descriptive statistics for
all variables.13
Before proceeding to the regression analysis, it is imperative to clear
up an important initial implication of the patterns outlined in Figure 8.2.
The trend observed in terms of the increase in managerial compensation
could be simply due to an administrative reclassification of workers.
To show that this is not the case, I divide the sample of firms into four
different quartiles by size (assets) and plot the share of managerial
compensation for both importing and nonimporting firms across these
four quartiles in Figure 8.4. A similar trend is observed across firms of all
sizes: it is the importing firms for which the share is rising significantly.
There is no plausible reason to argue that the firms across the size
distribution in India are reclassifying their workers from nonmanagers
to managers as the trade liberalization kicks in. This encourages me to
look for an effect of trade reform where I use both firm-level import
ratios and industry-level tariffs.

Table 8.2 Descriptive Statistics

Mean Median Std Dev. Min. Max.

Panel A: Dependent Variables
Managerial 0.02 0 0.07 0 1
Compensation/
Total Compensation
Managerial 1.31 0 169.65 0 66,315.1
Compensation
Nonmanagerial 95.53 14.4 631.83 0 47,619.5
Compensation
Managers 1.56 1 0.72 1 7
continued on next page

12
See, for example, Goldberg et al. (2010), Topalova and Khandelwal (2011), Ahsan
(2013), Ahsan and Mitra (2014), and De Loecker et al. (2016).
13
One pattern described in Table 8.1 deserves further comment. As reported, the
maximum figures of various measures normalized using gross value added (GVA)
can reach relatively high values. This is a feature of the definition of GVA (see the
appendix), and occurs in cases of high purchases and low sales, such as in initial
investments. All results are robust to omitting observations with GVA-normalized
figures higher than 1; nonetheless, we maintain the full sample in the main analyses
for the purposes of exploiting its full extent.
242 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 8.2 continued

Mean Median Std Dev. Min. Max.
Panel B: Firm-/Industry-level Determinants – Explanatory Variables
Total Imports/GVA 0.89 0.04 39.63 0 7,323.5
Import of Raw Materials/ 0.68 0.15 10.25 0 1,142.67
GVA
Import of Capital Goods/ 0.40 0.02 12.66 0 1,192
GVA
Import of Stores and 0.059 0.01 0.58 0 40.45
Spares/GVA
Import of Finished 5.65 0.04 149.59 0 7,323.5
Goods/GVA
Technology Adoption/ 0.07 0 9.77 0 2,163
GVA
Capital Employed 8.82 1.76 128.57 0 16,789
Productivity 0.48 0.42 0.34 0 5.50
GVA 1,181.05 127.48 16,000.95 0.086 103,1605
Skill Intensity 0.26 0.25 0.07 0.04 0.71
Factories 3,870.49 3,304 3,021.15 15 13,893
Management Technology 2.49 2.48 0.42 0 3.17
Input Tariffs 73.02 48.83 49.40 17.34 202.02
Output Tariffs 75.93 50 57.14 14.5 298.07
GVA = gross value added.
Notes: Annual data at the firm level, covering 1990–2011. Monetary values are in real Indian rupees (million).
“Managerial Compensation/Total Compensation” is the share of managerial compensation in total labor
compensation. Compensation is the sum of wages and bonuses. With regard to managers, it is the sum
for executives (top management) and directors (middle management), whereas for nonmanagers, it is all
other employees. “Managers” is the total number of managers. “Total Imports” is Imports of Raw Materials
+ Imports of Capital Goods + Imports of Stores and Spares + Imports of Finished Goods. “Technology
Adoption” is Research and development expenditure + Royalty payments for technical know-how. “Capital
Employed” is the amount of capital employed. “Productivity” is a measure for firm productivity computed
following the Levinsohn and Petrin (2003) methodology.a “GVA” is gross value added, defined as Total sales
– Total raw material expenditure. “Skill intensity” is the ratio of nonproduction workers to total employees
at the 2004 National Industrial Classification (NIC) three-digit level. “Factories” is the number of factories
at the 2004 NIC three-digit level. “Management technology” is the management quality score obtained
from Bloom and Van Reenen (2010) at the 2004 NIC two-digit level.b “Tariffs (input and output)” are at
the industry level (NIC 2004 four-digit level).
a J. Levinsohn and A. Petrin. 2003. Estimating Production Functions Using Inputs to Control for
Unobservables. The Review of Economic Studies. 70 (2). pp. 317−341.
b N. Bloom and J. Van Reenen. 2010. Why Do Management Practices Differ across Firms and Countries?
Journal of Economic Perspectives. 24 (1). pp. 203–24.
Source: Author’s calculations.
 Figure 8.4 Managerial Compensation, across Size Distribution, 1990–2011

1st Quartile 2nd Quartile

Nonimporter Importer Nonimporter Importer
0.15
0.10

0.10
0.05

0.05

0.00 0.00

1990 1995 2000 2005 2010 1990 1995 2000 2005 2010 1990 1995 2000 2005 2010 1990 1995 2000 2005 2010
Year Year

3rd Quartile 4th Quartile

Nonimporter Importer
Nonimporter Importer
0.10
0.10

Managerial Compensation / Total Compensation

0.05 0.05

0.00 0.00

1990 1995 2000 2005 2010 1990 1995 2000 2005 2010 1990 1995 2000 2005 2010 1990 1995 2000 2005 2010
Year Year

Note: The figure presents the average compensation share of managers for importing and nonimporting firms across size distribution, 1990–2011.
Trade Reform, Managers, and Skill Intensity: Evidence from India

Source: Author’s calculations.

243
 244 Trade Adjustment in Asia: Past Experiences and Lessons Learned

8.4 T
 rade Reform and the Relative Demand
for Managers
8.4.1 Preliminary Analysis
I start by testing the general association between trade and demand for
managers through a firm-level analysis, using the data described earlier.
Specifically, I use direct firm-level measures of trade, via import and
export penetration, to see which form of trade flow is associated with
demand for managerial workers. I consider the following equation for
firm i, at time t:

𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑗𝑗
( ) = 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇/𝐺𝐺𝐺𝐺𝐺𝐺)𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝑓𝑓𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑖𝑖𝑖𝑖𝑖𝑖 (1)
𝑗𝑗
= 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇/𝐺𝐺𝐺𝐺𝐺𝐺)𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝑓𝑓𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 + 𝜀𝜀𝑖𝑖𝑖𝑖
𝑖𝑖𝑖𝑖
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀where Mcomp is the managers’ total compensation, Tcomp is total 𝑗𝑗
( ) = 𝛿𝛿𝑖𝑖𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀
+ 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇/𝐺𝐺𝐺𝐺𝐺𝐺)𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝑓𝑓𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 + 𝜀𝜀𝑖𝑖𝑖𝑖 𝑗𝑗
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇labor
𝑖𝑖𝑖𝑖𝑖𝑖 compensation,
( ) Trade = 𝛿𝛿𝑖𝑖 is
+ either total imports
𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇) 𝑗𝑗𝑗𝑗−1 + or𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓
exports, and GVA 𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 + 𝜀𝜀𝑗𝑗𝑗𝑗
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇
is the gross value added 𝑖𝑖𝑖𝑖𝑖𝑖 of a firm i at industry j in year t. firmcontrols is
= 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇)𝑗𝑗𝑗𝑗−1 + 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑖𝑖𝑖𝑖−1It+includes
a vector of firm-level characteristics. 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 firm + 𝜀𝜀𝑗𝑗𝑗𝑗age, age squared,
𝑗𝑗

𝑖𝑖𝑖𝑖 and R&D intensity [(R&D expenditure + Royalty payment for technical
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀
( know-how)/GVA]. δi and η𝑗𝑗𝑗𝑗−1
) = 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇) t are +firm and time fixed
𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 effects. 𝑗𝑗 refers to
𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 + 𝜀𝜀𝑗𝑗𝑗𝑗
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇interactions
𝑖𝑖𝑖𝑖𝑖𝑖 between industry fixed effects and time trends. It controls
for other types of shocks (such as a change in labor policy or availability
of more finance) at the industry level, which vary over time and may
affect the compensation share of managers. I cluster standard errors at
the firm level.
βT is our coefficient of interest. It is the empirical association
between normalized imports, or exports, and the relative demand for
managers. In effect, the equation examines the determinants of the
relative demand for managers, measured through the wage bill share of
managers. Results appear in Table 8.3.
Starting with imports (Imp/GVA), column (1) presents the
benchmark setting. As can be seen, the coefficient of interest is positive
and significant. In addition, the magnitude is economically meaningful. A
1% increase in the GVA share of total imports increases the compensation
share of managers by approximately 0.1%. In column (2), as I replace
imports with exports (Exp/GVA), the effect vanishes. In other words,
I do not find any effect of exports on the relative demand for managers.
Column (3) uses both exports and imports. The significant effect of
imports on the demand for managerial workers continues, with no
effect from exports. Interestingly, this particular result depicts different
 Trade Reform, Managers, and Skill Intensity: Evidence from India 245

Table 8.3 Imports, Exports, and Relative Demand for Managers

MComp/TComp
(1) (2) (3) (4) (5)
(Imp/GVA)t-1 0.010*** 0.010***
(0.001) (0.001)
(Exp/GVA)t-1 0.002 0.0002 0.0003 0.001
(0.001) (0.001) (0.002) (0.002)
(ImpInput/GVA)t-1 0.008***
(0.001)
(ImpNInput/GVA)t-1 0.001
(0.003)
(ImpRaw/GVA)t-1 0.006***
(0.002)
(ImpCap/GVA)t-1 0.010***
(0.003)
(ImpStoSpa/GVA)t-1 0.002
(0.006)
(ImpFin/GVA)t-1 0.001
(0.003)
Firm Controlst-1 Yes Yes Yes Yes Yes
R-Squared 0.16 0.16 0.17 0.19 0.42
N 73,045 73,045 73,045 73,045 73,045
Firm FE Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes
(four-digit)
*Year Trend
FE = fixed effect, GVA = gross value added.
Notes: Columns (1)–(5) use the share of managerial compensation in the total compensation (MComp/
TComp) of a firm as the dependent variable. Total compensation is the sum of compensation to managers
and compensation to nonmanagers, where compensation to managers is the sum of compensation of all
the management levels and compensation to nonmanagers is the compensation to all other employees.
Imp/GVA is the GVA share of total imports (Import of raw materials + Import of capital goods + Import of
stores and spares + Import of finished goods) of a firm. Exp/GVA is the GVA share of total exports of a firm.
ImpInput/GVA is the GVA share of imports of capital goods and raw materials of a firm. ImpNInput/GVA is
the GVA share of imports of stores and spares and finished goods of a firm. ImpRaw, ImpCap, ImpStoSpa,
and ImpFin are import of raw materials, capital goods, stores and spares, and finished goods. GVA is the
gross value added of a firm, defined as Total sales – Total raw material expenditure. Firm Controls include age
of a firm, age squared, TechAdop/GVA and size of a firm. TechAdop/GVA measures the level of technology
adoption, defined as Research and development expenditure + Royalty payments for technical know-how,
normalized by GVA. I use “Assets” as the size indicator. All the dependent variables are in natural logarithm,
measured in millions of Indian rupees, deflated to 2005 using the industry-specific Wholesale Price Index.
Numbers in parenthesis are robust clustered standard errors at the firm level. Intercepts are not reported. *,
**, and *** denote 10%, 5%, and 1% level of significance, respectively.
Source: Author’s calculations.
246 Trade Adjustment in Asia: Past Experiences and Lessons Learned

dynamics than those presented in previous studies that emphasize

the role of exports in developed economies (e.g., Caliendo and Rossi-
Hansberg 2012), implying that the case of a developing economy may
provide a new perspective on this.
Next, I exploit the classification of imports into several categories
in columns (4) and (5). In column (4), I put together import of capital
goods and raw materials and denote it as “import of inputs” (ImpInput/
GVA), whereas I sum import of stores and spares and finished goods and
classify it as “import of noninputs” (ImpNInput/GVA). The estimates
show that the aggregate effect of imports on managerial compensation is
completely driven by import of inputs. Column (5) regresses managerial
compensation on exports and all the separate components of imports:
import of capital goods (ImpCap/GVA), import of raw materials
(ImpRaw/GVA), import of stores and spares (ImpStoSpa/GVA), and
import of finished goods (ImpFin/GVA). Estimates demonstrate that the
import of capital goods and import of raw materials are significantly and
positively correlated with the share of managerial compensation, with
the effect being higher in the case of capital goods.
To understand whether skill intensity can be termed one of the
complementary channels for the increase in demand for managers,
I interact the skill intensity ratio with the several import penetration
ratios in Table 8.4. I define skill intensity as the ratio of nonproduction
workers to total employees of an industry. This ratio is constructed
at the three-digit level 2004 NIC. Columns (1)–(3) interact the skill
intensity ratio with (Imp/GVA), (ImpInput/GVA), (ImpNInput/GVA),
(ImpRaw/GVA), (ImpCap/GVA), (ImpStoSpa/GVA), and (ImpFin/GVA),
respectively. The estimates do not show any evidence of the interaction
effect of import ratios and skill intensity on the increase in the demand
for managers.

Table 8.4 Imports, Exports, Relative Demand

for Managers, and Skill Premium

MComp/TComp
(1) (2) (3)
(Imp/GVA)t-1 × SkillIntt-1 0.003
(0.005)
(ImpInput/GVA)t-1 × SkillIntt-1 –0.002
(0.007)
(ImpNInput/GVA)t-1 × SkillIntt-1 –0.016
(0.021)
continued on next page
 Trade Reform, Managers, and Skill Intensity: Evidence from India 247

Table 8.4 continued

MComp/TComp
(1) (2) (3)
(ImpRaw/GVA)t-1 × SkillIntt-1 0.006
(0.013)
(ImpCap/GVA)t-1 × SkillIntt-1 0.013
(0.029)
(ImpStoSpa/GVA)t-1 × SkillIntt-1 0.003
(0.006)
(ImpFin/GVA)t-1 × SkillIntt-1 0.002
(0.006)
Firm Controls t-1 Yes Yes Yes
R-Squared 0.61 0.61 0.61
N 73,045 73,045 73,045
Firm FE Yes Yes Yes
Year FE Yes Yes Yes
Industry FE (four-digit)*Year Trend Yes Yes Yes
FE = fixed effect, GVA = gross value added.
Notes: Columns (1)–(3) use the share of managerial compensation in the total compensation (MComp/
TComp) of a firm as the dependent variable. Total compensation is the sum of compensation to managers
and compensation to nonmanagers, where compensation to managers is the sum of compensation of all
the management levels and compensation to nonmanagers is the compensation to all other employees.
Imp/GVA is the GVA share of total imports (Import of raw materials + Import of capital goods + Import
of stores and spares + Import of finished goods) of a firm. Exp/GVA is the GVA share of total exports of a
firm. ImpInput/GVA is the GVA share of imports of capital goods and raw materials of a firm. ImpNInput/
GVA is the GVA share of imports of stores and spares and finished goods of a firm. ImpRaw, ImpCap,
ImpStoSpa, and ImpFin are import of raw materials, capital goods, stores and spares, and finished goods.
GVA is the gross value added of a firm, defined as Total sales – Total raw material expenditure. SkillInt
is the skill intensity of an industry, defined as the ratio of nonproduction workers to total employees of
an industry at the 2004 National Industrial Classification three-digit level. Firm Controls include age of
a firm, age squared, TechAdop/GVA, and size of a firm. TechAdop/GVA measures the level of technology
adoption, defined as Research and development expenditure + Royalty payments for technical know-how,
normalized by GVA. I use “Assets” as the size indicator. All the dependent variables are in natural logarithm,
measured in millions of Indian rupees, deflated to 2005 using the industry-specific Wholesale Price Index.
Numbers in parenthesis are robust clustered standard errors at the firm level. Intercepts are not reported. *,
**, and *** denote 10%, 5%, and 1% level of significance, respectively.
Source: Author’s calculations.

8.4.2 Causal Inference

India’s Trade Reform

Prior to the 1990s, India was one of the most trade-restrictive
economies in Asia, with high tariff and nontariff barriers. In
1991, following a balance-of-payments crisis, India turned to the
248 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 8.5 Tariff Reform in India,

Manufacturing Industries, 1990–2011
Trade Reform: Input and Output Tariffs (Levels)
Indian Manufacturing Firms: 1990–2011
175

150

125
Tariff Levels

100

75

50

25

0
1990 1995 2000 2005 2010
Year
Input Tariffs Output Tariffs

Note: The figure presents the average output and input tariffs across the 2004 National Industrial
Classification four-digit level, 1990–2011.
Source: Author’s calculations.

International Monetary Fund for assistance. The latter conditioned

such assistance on the implementation of a major adjustment
program. A major part of the adjustment program was to abandon
the restrictive trade policies. As a result, average tariffs fell by more
than half between 1990 and 1996 (Topalova and Khandelwal 2011).
Nontariff barriers experienced a similar drop between the late 1980s
and the mid-1990s (Goldberg et al. 2010). Figure 8.5 plots average
tariff levels (both input and output) across manufacturing industries.
Starting at around 150 in 1990, the average tariff level dropped to less
than a tenth of that by 2011. These major tariff changes form the key
policy measure I plan to exploit.
One major advantage with the tariff liberalization program is that
it did not seem to have targeted industries within the manufacturing
sector in a way that was related to prereform conditions (Goldberg et
al. 2010). This establishes the plausibly exogenous nature of the reform.
Next, there is much variation in the tariff changes across industries.
The four-digit industry-level average annual decreases in tariffs range
from as low as 2% to as high as 25%, with a mean of 6% and a standard
deviation of approximately 2.5% (Chakraborty and Raveh 2018).
The tariff data are derived from the Trade Analysis Information
System (TRAINS) and World Integrated Trade Solution (WITS) tariff
database, at the Harmonized System six-digit level. These output
tariffs are passed through India’s input–output matrix for 1993–1994
 Trade Reform, Managers, and Skill Intensity: Evidence from India 249

to construct input tariffs. Next, both the input and output tariffs are
then concorded to the four-digit 2004 NIC level using the Debroy and
Santhanam (1993) concordance table. The tariffs are then matched with
the firm-level data.
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀
( ) = 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇/𝐺𝐺𝐺𝐺𝐺𝐺)𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝑓𝑓𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 +
Empirical𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇
Strategy𝑖𝑖𝑖𝑖𝑖𝑖and Results
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 I estimate the following reduced-form equation to understand 𝑗𝑗 the effect
) = 𝛿𝛿𝑖𝑖 of
+ changes
𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇/𝐺𝐺𝐺𝐺𝐺𝐺)
in tariffs on the + 𝑓𝑓𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖
relative
𝑖𝑖𝑖𝑖𝑖𝑖−1 demand for + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 + 𝜀𝜀𝑖𝑖𝑖𝑖
managers:
𝑖𝑖𝑖𝑖−1
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑖𝑖𝑖𝑖𝑖𝑖
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑗𝑗
( ) = 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇)𝑗𝑗𝑗𝑗−1 + 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 +
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑖𝑖𝑖𝑖𝑖𝑖 (2)
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑗𝑗
) = 𝛿𝛿𝑖𝑖 + 𝛽𝛽𝑇𝑇 ln⁡(𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇)𝑗𝑗𝑗𝑗−1 + 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑖𝑖𝑖𝑖−1 + 𝜂𝜂𝑡𝑡 + 𝜃𝜃𝑡𝑡 + 𝜀𝜀𝑗𝑗𝑗𝑗
𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑖𝑖𝑖𝑖𝑖𝑖
where ln(Tariff)jt-1 is the natural logarithm of tariff levels corresponding
to industry j at period t-1. I use both output and input tariffs. The
remaining notation follows that described previously. I follow Moulton
(1990) to cluster standard errors at the industry level.
I start by using both input and output tariffs; results are
reported in Table 8.5. Column (1) regresses the share of managerial
compensation on lagged output tariffs, a number of firm controls (firm
age, age squared, R&D intensity, and assets of a firm), firm fixed effects,
year fixed effects, and interactions of industry fixed effects and year
trends. The estimate shows that a drop in output tariffs or increase
in product market competition significantly increases the share of
managerial compensation. I additionally use lagged value of dependent
variable in column (2); output tariffs continue to significantly affect
managerial compensation. In column (3), I use both input and output
tariffs. Including both input and output tariffs concurrently, the
results show that the effect of output tariffs drops to 0 and a drop in
tariffs on intermediate inputs now explains the rise in managerial
compensation. Column (4) adds the lagged value of the share of
managerial compensation. The previous finding continues: no effect
of output tariffs and drop in tariffs on intermediate inputs explain the
demand for managers.
This result (when using the input and output tariffs concurrently)
also provides some insights into the potential underlying mechanism.
While a decrease in output tariffs may stiffen product market (import)
competition (Amiti and Konings 2007), a decrease in input tariffs increases
the technological complexity of the production process. The latter is a
feature of the higher quality and variety of imported inputs (Acemoglu
and Zilibotti 2001; Eaton and Kortum 1996; Goldberg et al. 2010).
250 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 8.5 Output Tariffs, Input Tariffs,

and Relative Demand for Managers

MComp/TComp
(1) (2) (3) (4)
OutTariffst-1 –0.006** –0.005** –0.00002 –0.00006
(0.003) (0.002) (0.003) (0.003)
InpTariffst-1 –0.010** –0.008**
(0.005) (0.004)
(MComp/TComp)t-1 0.260*** 0.263***
(0.020) (0.021)
Firm Controlst-1 Yes Yes Yes Yes
R-Squared 0.15 0.18 0.14 0.18
N 70,369 70,369 70,369 70,369
Firm FE Yes Yes Yes Yes
Year FE Yes Yes Yes Yes
Industry FE (four-digit)* Year Trend Yes Yes Yes Yes
FE = fixed effect, GVA = gross value added.

Notes: Columns (1)–(4) use the share of managerial compensation in the total compensation (MComp/
TComp) of a firm as the dependent variable. InpTariffs/OutTariffs is input (output) tariffs at the four-digit
National Industrial Classification 2004 level. Firm Controls include age of a firm, age squared, TechAdop/
GVA, and size of a firm. TechAdop/GVA measures the level of technology adoption, defined as Research and
development expenditure + Royalty payments for technical know-how, normalized by GVA. I use “Assets”
as the size indicator. All the dependent variables are in natural logarithm, measured in millions of Indian
rupees, deflated to 2005 using the industry-specific Wholesale Price Index. Numbers in parenthesis are
robust clustered standard errors at the firm level. Intercepts are not reported. *, **, and *** denote 10%, 5%,
and 1% level of significance, respectively.
Source: Author’s calculations.

The dominating effect of input tariffs suggests that the observed

increase in the relative demand of managers is triggered by changes in
the production technologies rather than by a stronger competition in
the final goods market. I continue my following analysis focusing on
input tariffs.
Benchmark results. The benchmark results are presented in
Table 8.6. In this table, I look at both intensive (price of managers or
compensation of managers) and extensive (number of managers) margins
of managerial demand in a firm. I start with the former. Columns (1)–(4)
show that a 10% drop in input tariffs increases the relative managerial
compensation of a firm by 0.8%–2.3%. In other words, the higher the
usage of imported foreign inputs, the higher the demand for managers.
Column (2) presents a dynamic version of Equation (2), providing a
similar result.
 Table 8.6 Input Tariffs, Relative Demand for Managers, and Skill Premium—Benchmark Results

Non-
No. of MComp/ MComp/
MComp/TComp Managers GVA GVA
(1) (2) (3) (4) (5) (6) (7) (8)
InpTariffst-1 –0.015*** –0.023** –0.007 –0.0002 –0.014*** –0.312*** –0.010*** –0.004
(0.003) (0.010) (0.005) (0.003) (0.005) (0.104) (0.002) (0.004)
(MComp/TComp)t-1 0.657***
(0.031)
InpTariffst-1 –0.008***
×Importer (0.003)
InpTariffst-1 –0.013***
×Importer Input (0.001)
Importer Input 0.067***
(0.006)
InpTariffst-1×SkillIntt-1 –0.005+
(0.003)
SkillIntt-1 0.013
(0.013)
Firm Controlst-1 Yes Yes Yes Yes Yes Yes Yes Yes
R-Squared 0.12 N/A 0.12 0.62 0.61 0.58 0.31 0.43
N 70,369 70,369 70,369 70,369 70,369 27,975 70,369 70,369
Trade Reform, Managers, and Skill Intensity: Evidence from India

continued on next page

251
Table 8.6 continued
Non-
No. of MComp/ MComp/
MComp/TComp Managers GVA GVA
(1) (2) (3) (4) (5) (6) (7) (8)
Firm FE Yes Yes Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes Yes Yes
Industry FE (four- Yes Yes Yes Yes Yes Yes Yes Yes
digit)* Year Trend
FE = fixed effect, GVA = gross value added.
Notes: Columns (1)–(5) use the share of managerial compensation in the total compensation (MComp/TComp) of a firm as the dependent variable. Columns (6)–(8) use number
of managers, GVA share of managerial compensation, and GVA share of nonmanagerial compensation as the dependent variables, respectively. InpTariffs"is input tariffs at the 2004
National Industrial Classification (NIC) four-digit level. Importer is a dummy variable that takes a value of 1 if a firm imports (either import of raw materials, capital goods, stores and
spares, or finished goods). ImporterInput"is a dummy variable if a firm imports either capital goods or raw materials. SkillInt"is the skill intensity of an industry, defined as the ratio of
nonproduction workers to total employees of an industry at the 2004 NIC three-digit level. Firm Controls include age of a firm, age squared, TechAdop/GVA, and size of a firm. TechAdop/
GVA measures the level of technology adoption, defined as Research and development expenditure + Royalty payments for technical know-how, normalized by GVA. I use “Assets”
as the size indicator. All the dependent variables are in natural logarithm, measured in millions of Indian rupees, deflated to 2005 using the industry-specific Wholesale Price Index.
Numbers in parenthesis are robust clustered standard errors at the firm level. Intercepts are not reported. +, *, **, and *** denote 12%, 10%, 5%, and 1% level of significance, respectively.
Source: Author’s calculations.
252 Trade Adjustment in Asia: Past Experiences and Lessons Learned
 Trade Reform, Managers, and Skill Intensity: Evidence from India 253

Column (3) introduces an interaction term of input tariffs and an

importer dummy (it takes a value of 1 for a firm that is an importer). The
estimates show that the entire effect is concentrated for firms that are
importing. In column (4), to understand the source of the previous effect
according to the type of importing firm, I create an additional dummy
where it takes a value of 1 if an individual firm is importing production
units (capital goods and raw materials) and interacts with input tariffs.
Likewise, in column (3), the results demonstrate that the magnitude of
the effect increases by more than 1.5 times and the interaction term is
significant at the 1% level. This gives additional support to the results
shown in Table 8.3. Firms importing more intermediate inputs as a result
of a drop in input tariffs (due to trade reform) require more managers
to manage those inputs in order to utilize them in their production
processes.
Column (5) introduces our key variable: skill intensity.14 As before,
I measure skill intensity through the three-digit industry-level ratio
of nonproduction workers to all employees, with the standard skill
intensity measure being used in the literature.15 This measure is
obtained from Ghosh (2014) (1990–2000) and the Indian Annual Survey
of Industries (2001–2011). Previous studies indicate that globalization
increases the demand for skill in developing economies (Goldberg and
Pavcnik 2007). This, in turn, may affect the demand for managers. The
main coefficient of interest, βT, remains stable. Unlike the interaction
terms between import ratios and skill intensity, the interacted effect
of input tariffs and skill intensity explains the increase in demand for
managers, but only at the 12% level of significance. In other words,
InpTariffs × SkillInt indicates that there is an apparent differential effect
across industries’ benchmark skill intensity levels: The higher the drop
in input tariffs, the higher the increase in the demand for skills, and thus

14
At first glance, it may suggests that skill and managers might be correlated through the
standard definition of skill in the literature, which considers nonproduction workers
or otherwise those in white-collar occupations. Note, however, that this definition,
while also covering managers, includes various additional occupations that are not
necessarily managerial positions. For instance, in the cases of Berman, Bound, and
Griliches (1994) and Zhu and Trefler (2005), skilled workers are defined as holding
the following positions within the manufacturing sector: manager, professional,
technician, and clerical worker. Indeed, managers represent a subset of that, though
the other professions can fall under the nonmanagers classification.
15
Proxying skill intensity by “nonproduction” is nontrivial, though this is common
practice by necessity given data limitations. This measure is adopted by various
studies on trade liberalization and skill in developing countries (e.g., Raveh and
Reshef 2016; Zhu and Trefler 2005). In addition, Berman, Bound, and Griliches
(1994) show that the production or nonproduction worker classification is a good
proxy for skilled and unskilled workers.
254 Trade Adjustment in Asia: Past Experiences and Lessons Learned

the rise in managerial compensation. In other words, the demand for

managers tends to rise in those sectors where there is a complementary
effect of a drop in input tariffs and a higher ratio of skill intensity. This
particular finding gives some possible indication of an increase in the
skill premium through the import of high-quality intermediate goods by
firms, which we investigate in column (8). On the other hand, the skill
intensity variable is positively correlated with the demand for managers,
but not significantly.
Column (6) uses the number of managers as the dependent variable.
A drop in input tariffs also significantly affects the extensive margin of
managers. I use the GVA share of managerial compensation in column
(7). The benchmark result continues to be the same: a drop in tariffs
increases the price of managers.
Lastly, column (8) exploits compensation of the other category of
workers (i.e., nonmanagers) as the dependent variable. The reason to
look at this category is to understand whether there is an opposite or
differential effect of trade reform across different categories of workers.
The point estimate shows that trade reform (or drop in input tariffs) has
no effect on the demand for nonmanagers. In other words, the effect of
tariff liberalization on nonmanagerial compensation is indistinguishable
from 0. Based on these results, it is difficult to conclusively claim that
higher demand for managers (as a result of trade reform) led to an
increase in skill premium in India (although we find some evidence of
skill intensity being one of the complementary channels).
Additional channels. Having identified the main effect, I now
consider other possible complementary channels that may affect the
demand for managers in Table 8.7. In each case, I focus on two points:
first, the role of the additional control as an intermediate channel, by
examining its direct effect, and second, via its effect through the main
variable of interest (input tariffs).
I start with the potential connection between managers and capital
employed. The key variable, intermediate inputs, is a flow measure
of incoming equipment. The stock value of capital, which includes
nonequipment stock as well, may also affect the demand for managers. For
instance, capital-intensive production processes may involve automation
and hence less problem solving and less demand for managers than labor-
intensive production technologies. To test the role of capital intensity, I
add firms’ GVA share of capital employed in column (1). Its direct effect is
indistinguishable from 0, providing no evidence that the stock of capital
is correlated with the compensation share of managers. Importantly, the
result does not change relative to the benchmark case, indicating that
the effect of the incoming flow of equipment on the relative demand
for managers holds regardless of whether the firm is relatively capital
intensive. Notably, the interaction of this measure with input tariffs does
 Trade Reform, Managers, and Skill Intensity: Evidence from India 255

Table 8.7 Input Tariffs, Relative Demand for Managers,

and Skill Premium—Additional Channels
MComp/TComp
Capital Total Factor Management
Employed Productivity Factories Technology
(1) (2) (3) (4) (5)
InpTariffst-1 –0.003* –0.00004 –0.009* –0.013*** –0.035***
(0.002) (0.004) (0.005) (0.004) (0.010)
InpTariffst-1 × 0.001 0.005**
CapEmpt-1 (0.001) (0.002)
InpTariffst-1 × –0.002 0.014***
TFPt-1 (0.003) (0.004)
InpTariffst-1 × –0.001** –0.002**
Factoriest-1 (0.000) (0.001)
InpTariffst-1 × 0.004 –0.029
MTt-1 (0.004) (0.019)
CapEmpt-1 0.003 0.007
(0.004) (0.008)
TFPt-1 0.016* 0.042***
(0.009) (0.016)
Factoriest-1 0.014*** 0.019***
(0.005) (0.006)
Firm Controlst-1 Yes Yes Yes Yes Yes
R-Squared 0.58 0.63 0.12 0.12 0.08
N 69,704 46,286 70,369 68,856 45,337
Firm FE Yes Yes No No No
Industry FE No No Yes Yes Yes
Year FE Yes Yes Yes Yes Yes
Industry FE Yes Yes Yes Yes Yes
(four-digit)
*Year Trend
FE = fixed effect, GVA = gross value added, TFP = total factor productivity.
Notes: Columns (1)–(7) use the share of managerial compensation (MComp/TComp) as the dependent
variable. InpTariffs is input tariffs at the four-digit National Industrial Classification (NIC) 2004 level. “apEmp
is the amount of capital employed by a firm. TFP is total factor productivity at the firm level estimated using
Levinsohn and Petrin (2003).a Factories is the number of factories at the NIC 2004 three-digit level. MT is an
index of Management Quality at the 2004 NIC two-digit level and has been sourced from Bloom and Van
Reenen (2010).b Firm Controls include age of a firm, age squared, TechAdop/GVA, and size of a firm. TechAdop/
GVA measures the level of technology adoption, defined as Research and development expenditure + Royalty
payments for technical know-how, normalized by GVA. I use “Assets” as the size indicator. All the dependent
variables are in natural logarithm, measured in millions of Indian rupees, deflated to 2005 using the industry-
specific Wholesale Price Index. Numbers in parenthesis are robust clustered standard errors at the firm level.
Intercepts are not reported. *, **, and *** denote 10%, 5%, and 1% level of significance, respectively.
a J. Levinsohn and A. Petrin. 2003. Estimating Production Functions Using Inputs to Control for
Unobservables. The Review of Economic Studies. 70 (2). pp. 317−341.
b N . Bloom and J. Van Reenen. 2010. Why Do Management Practices Differ across Firms and Countries?
Journal of Economic Perspectives. 24 (1). pp. 203–24.
Source: Author’s calculations.
256 Trade Adjustment in Asia: Past Experiences and Lessons Learned

not point to any kind of systematic differential effects across capital

intensity levels.
Next, I test for the effect of productivity. Previous research shows
that trade liberalization increases firm productivity (e.g., Topalova and
Khandelwal 2011). Higher productivity may increase the demand for
managers due, for instance, to its potential effects on organizational
design (Garicano 2000). To test whether it also acts as a complementary
channel, I add a measure of productivity in column (2). I adopt the
Levinsohn and Petrin (2003) methodology to construct firm-level
total factor productivity.16 The estimated coefficients indicate that the
interaction effect of input tariffs and productivity is also not associated
with the relative demand for managers.
Third, despite controlling for firm assets, I follow Acemoglu et al.
(2007) and Bloom, Sadun, and Van Reenen (2010) to dig deeper into
the potentially important effect of size on the demand for managers, by
testing an additional related measure: the number of factories and plants
at the three-digit industry level. I add this measure in column (3); the
estimated coefficient indicates that the main result is robust to this
addition. The estimate also shows that a drop in input tariffs induced the
establishment of more factories, which consequently led to an increase
in managerial compensation, as local knowledge is important.
Last, an additional potential determinant relates to management
technology. In a recent study, Chen (2017) makes a connection
between trade liberalization and management technology. If better
management technology requires a higher volume and quality of
managers, it may represent a viable channel. To potentially test for this,
I proxy management technology through the cross-country industry
management survey carried out by Bloom and Van Reenen (2010).
Surveying a large number of firms in various manufacturing industries in
India (among other countries) throughout 2004, the authors construct
a measure for management quality in different sectors. This index is a
number between 1 and 5, with 5 representing the best quality. Estimates
in column (4) indicate that input-trade liberalization does not have any
systematically different effect on the relative demand for managers
across industries’ level of management technology.
In column (5), I include all the additional controls and their
interactions with the input tariffs. This is a relatively demanding
specification in terms of potential multicollinearity. However, the
primary coefficient of interest remains negative and significant, similarly
to the benchmark estimates.

16
The method controls for the potential simultaneity in the production function by
using firms’ raw material inputs as a proxy for the unobservable productivity shocks.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 257

Firm characteristics. I now take a step further and look into several
other firm- and industry-level characteristics to investigate which
type(s) of firm or industry characteristic(s) is (are) driving the main
result. An additional purpose is to check whether there is any kind of
stronger evidence of skill intensity as a complementary channel for any
subsample of firms that got masked in the aggregate results. The results
are presented in Table 8.8.
I start by investigating the role of the size of a firm in column (1),
more specifically whether the increase in the relative demand for
managers is concentrated in one section of firms or differs across the
size distribution. I divide the firms according to their size. I use the total
assets of a firm as the size indicator. I use the following method: If the
total asset of a firm is below the 25th percentile of the total assets of
that industry, that firm belongs to the first quartile. Likewise, if a firm’s
total asset falls between the 25th and 50th, 50th and 75th, or is greater
than the 75th percentile, it falls into the second, third, or fourth quartile,
respectively. Since firms could move across quartiles over time, I use
the average rank of the firms for the period of analysis. In order to find
out the required effect, I interact the input tariffs with the respective
quartiles.
The estimates reveal some interesting facts. All firms, except the
big ones, show significant evidence of skill intensity as an additional
channel due to a drop in input tariffs, with the effect being highest for
the smallest firms. This is intuitive. As firms import more high-quality
intermediate goods, due to trade reform, they require more managers.
As a result, skill intensity acts as an additional channel through which
demand for managers or managerial compensation rises. This is highest
in the case of small firms, as they did not have any exposure before using
these high-quality foreign intermediate inputs. On the other hand, the
interaction terms of input tariffs and quartile dummies are significant
across the size distribution, suggesting that skill intensity is not a
channel (through which there is a rise in the demand for managers) for
the big firms or the firms belonging to the fourth quartile.
Columns (2) and (3) divide the sample into exporters and
nonexporters in order to understand whether there is any kind of
premium attached to an exporting firm. As the results demonstrate,
the effect of a drop in input tariffs on the demand for managers is
observed for both exporters and nonexporters. However, the effect is
stronger in the case of the exporting firms. Also, the evidence of skill
intensity as a channel for the rise in demand for managers is stronger
and greater for exporting firms than for nonexporters. The results point
to an interesting outcome. The rise in the demand for managers or for
a set of skilled workers is not only restricted to the group of exporters,
 Table 8.8 Input Tariffs, Relative Demand for Managers, and Skill Premium—Firm Characteristics
258

MComp/TComp
Size Export Orientation End Use Ownership
Non- Final Intermediate
Exporters exporters Goods Goods Domestic Foreign
(1) (2) (3) (4) (5) (6) (7)
InpTariffst-1 –0.021** –0.009+ –0.007 –0.017** –0.022*** –0.018
(0.008) (0.006) (0.006) (0.007) (0.007) (0.019)
InpTariffst-1 × SkillIntt-1 –0.012** –0.007* –0.004 –0.008+ –0.013*** –0.010
(0.005) (0.004) (0.004) (0.005) (0.004) (0.013)
InpTariffst-1 × Qr1×SkillIntt-1 –0.020***
(0.007)
InpTariffst-1 × Qr2×SkillIntt-1 –0.015**
(0.007)
InpTariffst-1 × Qr3 × SkillIntt-1 –0.018***
(0.006)
InpTariffst-1 × Qr4 × SkillIntt-1 –0.007
(0.006)
Trade Adjustment in Asia: Past Experiences and Lessons Learned

InpTariffst-1 × Qr1 –0.054***

(0.010)
InpTariffst-1 × Qr2 –0.052***
(0.009)
continued on next page
Table 8.8 continued
MComp/TComp
Size Export Orientation End Use Ownership
Non- Final Intermediate
Exporters exporters Goods Goods Domestic Foreign
(1) (2) (3) (4) (5) (6) (7)
InpTariffst-1 × Qr3 –0.054***
(0.010)
InpTariffst-1 × Qr4 –0.039***
(0.009)
Firm Controlst-1 Yes Yes Yes Yes Yes Yes Yes
R-Square 0.17 0.20 0.08 0.10 0.12 0.12 0.22
N 70,369 37,325 33,044 31,815 38,554 65,777 4,592
Firm FE Yes Yes Yes Yes Yes Yes Yes
Year FE Yes Yes Yes Yes Yes Yes Yes
Industry FE (four-digit)*Year Trend Yes Yes Yes Yes Yes Yes Yes
FE = fixed effect.
Notes: Columns (1)–(7) use share of managerial compensation (MComp/TComp) as the dependent variable. InpTariffs is input tariffs at the 2004 National Industrial Classification
(NIC) four-digit level. SkillInt is the skill intensity of an industry, defined as the ratio of nonproduction workers to total employees of an industry at the 2004 NIC three-digit level.
Quartiles (Qri ) are defined according to the total assets of a firm. A firm belongs to the first quartile (Qr₁) if the assets of that firm are below the 25th percentile of the total assets of
that industry to which the firm belongs and so on. Firm Controls include age of a firm, age squared, TechAdop/GVA, and size of a firm. TechAdop/GVA measures the level of technology
adoption, defined as Research and development expenditure + Royalty payments for technical know-how, normalized by GVA. I use “Assets” as the size indicator. All the dependent
variables are in natural logarithm, measured in millions of Indian rupees, deflated to 2005 using the industry-specific Wholesale Price Index. Numbers in parenthesis are robust
Trade Reform, Managers, and Skill Intensity: Evidence from India

clustered standard errors at the firm level. Intercepts are not reported. +, *, **, and *** denote 12%, 10%, 5%, and 1% level of significance, respectively.
Source: Author’s calculations.
259
260 Trade Adjustment in Asia: Past Experiences and Lessons Learned

but rather it spans across the entire set of manufacturing firms. This is
unlike the other cases, where the change in skill premium because of
trade reform concentrates only on the exporters. In the case of India,
the results suggest that the entire sector of manufacturing firms has
undergone a change in their technological production processes.
Next, I categorize firms according to the end use of their goods:
final and intermediate. The former comprises consumer nondurable
and consumer durable goods, whereas the latter includes intermediate,
basic, and capital goods. I follow Nouroz (2001) and match the firm-
level dataset with the input–output classification. Columns (4) and (5)
present the required result. The point estimates show us that the effect
of the trade liberalization on the demand for managers is significant only
in the case of the intermediate goods sector; similarly for the evidence
of skill intensity.
Lastly, I investigate the ownership structure of an Indian
manufacturing firm. I divide the sample of firms into two different
groups: domestic (which includes both private and public firms) and
foreign. The coefficients of interest in columns (6) and (7) tell us that
the main result is entirely driven by the change in the managerial
compensation ratio in the domestic firms, more so for the privately
owned ones. While it is not entirely unexpected that privately owned
firms have undergone a change in their production processes due to
the adoption of high-quality foreign inputs, it is nevertheless surprising
to see that only the domestic firms are the main drivers of change in
the overall change in the demand for managers observed and not the
multinationals.

8.4.3 Discussion of Results and Policy Relevance

Let us first summarize the main results of the empirical analysis and
provide further interpretations. The key finding is that a drop in input
tariffs, or increased use of imported intermediate inputs, increases the
compensation (intensive) and number (extensive) of managers, with no
effect on nonmanagerial workers. The effect is acute: (i) across firm-
size distribution; (ii) whether a firm is an exporter or not; (iii) in firms
producing intermediate goods; and (iv) in privately owned domestic
firms. In addition, the results show some evidence of skill intensity as an
additional channel, but only in the case of firms below the halfway point
of the size distribution.
Two key questions arise. First, how may these findings be important
for understanding the distributional effect, in terms of compensation of
these two different kinds of workers (managers and nonmanagers), of
trade policies? In particular, is the increase in wage gap between these
 Trade Reform, Managers, and Skill Intensity: Evidence from India 261

two categories of workers solely due to an increase in the adoption of

skill-biased technological inputs (due to a fall in input tariffs), or is there
a simultaneous fall in the supply of skilled labor, which accentuated the
wage premium? Second, what is the role of the government in responding
to changes in the demand for more skilled workers through the supply
of managerial skills and other types of skills? In order to address these
questions, I draw on previous related research on India and consider a
possible conceptual framework that can fit my findings into this broader
picture.
I start by addressing the former. Input (output) tariffs relate to
imported inputs (final goods). Goldberg, Khandelwal, and Pavcnik
(2013) point out that because of the drop in input tariffs, due to the
trade liberalization episode in India, imports of intermediate inputs
saw the highest increase, of almost 300%; and the vast majority of
the inputs are imported from the countries in the Organisation for
Economic Co-operation and Development (OECD). Table 8.9 lists the
top destinations in India’s percentage of imported capital. It shows that
India imports around 82% of capital goods from OECD destinations.

Table 8.9 Import of Capital Goods—

Top 10 Destinations

Rank Trading Imported Capital

(1) (2)
1 United States 20.14
2 Japan 16.80
3 Germany 16.73
4 United Kingdom 6.60
5 Singapore 4.98
6 France 4.96
7 Italy 4.63
8 Switzerland 3.10
9 Republic of Korea 2.18
All Other 19.88
Total 100
Note: Numbers in the table represent the share of capital goods imported by
India from different destinations.
Source: I. Kandilov, A. Leblebicioglu, and R. Manghnani. 2016. Tariffs on Imported
Capital and Firm-Level Investment in the Indian Manufacturing Sector, Mimeo,
North Carolina State University.
262 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Eaton and Kortum (2001) argue that the capital goods produced in
the OECD countries are of high quality and R&D intensive. Thus, an
increased use of imported inputs by a firm upgrades the technological
intensiveness of the production technology it uses and therefore
requires managers to cope with the new knowledge, thereby increasing
their relative demand. Realizing the main effect is completely driven by
the input side and hence implies that a quality upgrading channel is at
work, operating via input-tariff liberalization.17
Previous studies on both developing and developed economies
pointed to an export-based quality-upgrading channel (Caliendo and
Rossi-Hansberg 2012) or product market competition (e.g., Cunat and
Guadalupe 2009; Bloom, Sadun, and Van Reenen 2010). Verhoogen
(2004) finds strong support for this hypothesis in the case of Mexico.
Greater exports as a result of the peso crisis resulted in better-quality
products being produced by the exporters. Since higher-quality products
require a higher proportion of skilled workers, the relative demand
for and returns to skilled labor increased. This chapter shows how a
developing economy can present different dynamics regarding this.
Several hypotheses other than the “quality-upgrading” channel can
be put forward to explain this rise in demand for managers and skill
intensity. Some relate to economic reforms in general and not specifically
to trade reform in driving the returns to skilled labor. According to this
hypothesis, developing countries may experience higher returns to
skilled-labor-intensive occupations such as professional, managerial
jobs as a result of reforms that generate demand for individuals who
can implement these reforms. The above results suggest that in India
external sector reforms may have created more white-collar jobs.
Empirical evidence is mixed: Cragg and Epelbaum (1996) find support
for this hypothesis for pre-NAFTA Mexico, while Attanasio, Goldberg,
and Pavcnik (2004) find no changes in the occupational returns between
1986 and 1998 in Colombia.
Further, outsourcing or global production sharing has also been
identified as one of the reasons to explain the rise in skill intensity
or premium and demand for skilled labor in developing economies.
Feenstra and Hanson (1996, 2003) argue that trade liberalization by the
developing countries allows their counterparts (developed countries)
to transfer the production of intermediate goods and services. These

17
To the extent that a higher demand for managers is associated with better
management practices, these patterns are consistent with those documented by
Bloom et al. (2016). They find that better-managed firms in the People’s Republic
of China and the United States use more imported inputs, and specifically more
expensive and higher-quality inputs.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 263

activities are skill intensive, which results in a greater demand for

and returns to skilled labor. Therefore, the import of intermediate
goods can benefit skilled workers in a developing economy, more so
for firms that had the least exposure before the reform. Feenstra and
Hanson (1997) find empirical support for this hypothesis for the case
of Mexico.
The final one relates to skill-biased technical change. Wood (1995)
argues that greater competition from foreign firms may induce domestic
firms in a developing economy to either engage in R&D or to adopt
new and advanced technologies in order to secure their market share
in the domestic and international markets. Because of technology-skill
complementarities, adoption of modern technologies raises the demand
for and returns to skilled labor. He called this “defensive innovation.”
Harrison and Hanson (1999) and Attanasio, Goldberg, and Pavcnik
(2004) found empirical support for this hypothesis for Mexico and
Colombia, respectively. In the case of Ghana, Gorg and Strobl (2002)
come to a similar conclusion (to mine): an increase in the relative wages
of skilled labor (in my case, managers) brought about by skill-biased
technological change induced through imports of technology-intensive
capital goods. However, Pavcnik (2003) rejects the skill-biased technical
change hypothesis for Chilean plants.
My analysis of managerial compensation of Indian manufacturing
firms’ documents large demand shifts toward managerial workers but
does not find significant evidence of a shift away from nonmanagerial
workers. I find that skill intensity played an important role in widening
the wage gap between managerial and nonmanagerial workers in
India between 1990 and 2011, but for small and medium-sized firms.
These demand shifts were for both exporters and nonexporters as
well as firms producing intermediate goods. The results also suggest
that demand for managerial workers was primarily within industries
during this period in India. This finding provides strong evidence for
all the hypotheses discussed above: skill intensity as a result of external
sector reforms played a major role in the creation of managerial jobs,
thereby generating demand for skilled and managerial labor.
I now focus on the second question concerning the demand and
supply factors influencing the role of skill intensity. Some of the existing
literature (Dutta 2005; Kijima 2006; Chamarbagwala 2006) in India
observes that the increase in wage inequality during 1983–1999 was
mainly attributable to an increase in the returns to skills (as captured by
educational attainment). However, Azam (2010) argues that the driving
forces that led to the increase in wage premium for high-skilled workers
(tertiary graduate workers) have not been fully explored. It is imperative
to understand the determinants of the wage premium, as for policy
264 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 8.10 Employment Share, by Educational Status

1987 1993 1999 2004

(1) (2) (3) (4)
Below Primary 21.35 17.96 14.63 14.57
Primary 13.61 9.99 8.40 9.70
Middle 13.87 14.47 14.46 14.52
Secondary 29.47 31.14 32.72 31.50
Graduate and Above 21.71 26.45 23.79 29.71
Note: Numbers in the table indicate the share of regular employed male workers in urban India
in the age group 23–57 years.
Source: M. Azam. 2010. India’s Increasing Skill Premium: Role of Demand and Supply. The BE
Journal of Economic Analysis and Policy. 10 (1). pp. 1−26.

makers it is important to know whether the increase in wage premium

is driven by demand or supply, since the policy responses differ for these
two scenarios. In addition, changes in wage premium have important
implications for the evolution of wage inequality, and hence overall
income inequality.
In order to understand whether the supply or demand factor played
a role, I first look at the change in employment shares of different
types of workers between 1987 and 2004. Table 8.10 divides workers
according to educational status. The table uses data from four schedules
(1987–1988, 1993–1994, 1999–2000, and 2004–2005) of the Employment
and Unemployment Schedule administered by the National Sample
Survey Organization. It shows that the employment share of workers
with a graduate degree went up from 22% to 30% between 1987 and
2004. Though the supply of workers with a graduate degree and above
increased between 1987 and 1999, it ceased to grow between 1999 and
2004. For the workers with primary and below primary education,
the employment shares fell from 14% to 10% and from 21% to 15 %,
respectively, between 1987 and 2004. However, the shares increased
a little between 1999 and 2004. The employment share of secondary
graduates also declined between 1999 and 2004.
Next, I look at the wage premiums. Figure 8.6 plots the wage
premiums for graduate and secondary urban male workers. The figure
shows two important things. First, the wage premium for graduate
workers (calculated as the difference in mean log hourly wages between
regular male workers with a tertiary or graduate degree and those with
a secondary degree) increased from 0.37 to 0.52, whereas for secondary
workers (wage premium calculated with respect to workers with
 Trade Reform, Managers, and Skill Intensity: Evidence from India 265

Figure 8.6 Wage Premium, Tertiary

and Secondary Degree Workers, 1987–2004
Wage Premium
Tertiary and Secondary Workers

0.55

0.50
Tariff Levels

0.45

0.40

0.35
1987 1993 1999 2004
Year
Tertiary-Secondary Secondary-Less than Secondary

Note: The figure presents the wage premium for workers with a graduate degree and below a
graduate degree, 1987–2004.
Source: M. Azam, M. 2010. India’s Increasing Skill Premium: Role of Demand and Supply. The BE
Journal of Economic Analysis and Policy. 10 (1). pp. 1−26.

below a secondary degree), it almost remained the same. Second, most

of the increase in wage premium occurred in the 1990s, while it was
relatively stable in the 1980s. Table 8.11 breaks down the wage premium
estimates for all these years by different age categories. For the age
group 23–27 years, the wage premium increased between 1987 and 1993,

Table 8.11 Tertiary-Secondary Wage Premium, by Age Group

Age
Group 23–27 28–32 33–37 38–42 43–47 48–52 53–57
Year (1) (2) (3) (4) (5) (6) (7)
1987 0.36 0.35 0.35 0.32 0.42 0.41 0.53
(0.04) (0.03) (0.03) (0.03) (0.04) (0.04) (0.06)
1993 0.41 0.37 0.40 0.37 0.39 0.43 0.33
(0.05) (0.05) (0.04) (0.05) (0.05) (0.06) (0.08)
1999 0.33 0.48 0.45 0.46 0.51 0.44 0.36
(0.04) (0.04) (0.05) (0.04) (0.04) (0.04) (0.10)
2004 0.63 0.55 0.42 0.59 0.52 0.45 0.45
(0.05) (0.05) (0.06) (0.05) (0.05) (0.05) (0.06)
Notes: The table entries are wage differential in mean log hourly wages between a tertiary graduate worker
and secondary graduate worker. Numbers in parenthesis are robust standard errors.
Source: Azam, M. 2010. India’s Increasing Skill Premium: Role of Demand and Supply. The BE Journal of
Economic Analysis and Policy. 10 (1). pp. 1−26.
266 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Figure 8.7 Wage Premium, Different Age Groups,

Tertiary and Secondary Degree Workers, 1987–2004
Wage Premium, Different Age Groups
Tertiary and Secondary Workers
0.60

0.55

0.50

0.45

0.40

0.35
1987 1993 1999 2004
Year
23–32 48–57

Note: The figure presents the wage premium for workers with a graduate degree and below a
graduate degree for different age groups, 1987–2004.
Source: M. Azam, M. 2010. India’s Increasing Skill Premium: Role of Demand and Supply. The BE
Journal of Economic Analysis and Policy. 10 (1). pp. 1−26.

declined between 1993 and 1999, and then increased sharply between
1999 and 2004. The wage premium for the 28–32 age group increased
continuously during 1987–2004. However, the wage premium for older
age groups, 48–52 and 53–57 years, remained about the same between
1987 and 2004. This shows that the overall increase in wage premium of
tertiary graduate workers between 1987 and 2004 was mostly driven by
younger age groups. Figure 8.7 plots the wage gap between tertiary and
secondary degree workers for the age groups 23–32 (younger group)
and 48–57 (older group). It also shows similar trends.
Putting all these together, it can be argued that, as the relative
supply of tertiary workers (or workers with a graduate degree or above)
slowed down during the 1990s and 2000s, coupled with the increase
in demand for them as the firms started to use more technologically
intensive inputs due to a drop in tariffs, it led to a rise in wage premium.
Further, this wage premium is particularly high for workers belonging
to the 23–32 age group.

8.5 Conclusion
This chapter investigates the effect of India’s trade liberalization
episode in the form of a drop in tariffs on the demand for managerial
workers during 1990–2011. Additionally, I check whether the
demand for managers can be explained through a widely researched
 Trade Reform, Managers, and Skill Intensity: Evidence from India 267

phenomenon, an increase in skill intensity. The study uses detailed data

on compensation for Indian manufacturing firms and shows that a drop
in input tariffs—and not output—significantly increases the demand for
managers. A 10% drop in input tariffs increases the share of managerial
compensation by 0.5%–3.5 %. The trade-induced demand shifts toward
managerial workers find some support for quality upgrading, sharing of
production activities, or the skill-biased technical change hypothesis,
even though it is not possible to decompose the demand increase for
managerial workers into its exact source.
The results also show that one possible channel for an increase in
the compensation share of managers for firms below the halfway point
of the size distribution may be through an increase in skill intensity. On
the other hand, the estimates do not show any kind of demand shift away
from nonmanagerial workers. Further analysis shows that the shortage
in the supply of skilled workers during the late 1990s and 2000s, coupled
with the increase in demand for these workers, led to an increase in
wage premium for these workers. And this wage premium was highest
for the workers belonging to the 23–32 years age cohort. This suggests
that the demand for skill in the Indian economy was not solely due to
an increase in the use of intermediate inputs but also to changes within
the economy that were not related to trade. Dutta (2005) decomposes
the wage regression functions to highlight that the industry affiliation of
workers can also explain about a quarter of the wage inequality.
In India, low mobility between industries and a lack of transferable
skills prevent workers from moving out of industries with declining
relative wages in response to trade reform. All the results put together
and this characteristic of the labor market in India suggest the current
need to increase labor market flexibility through labor market and
other institutional reforms. However, these reforms would also need to
be supplemented by adequate provisions for social protection. Safety
net programs for workers affected by trade reforms are necessary to
minimize the short-run adjustment costs faced by workers from which
there was a demand shift. There is also a need for a coherent strategy for
social protection such as the rationalization of severance pay schemes, a
movement toward insurance mechanisms covering both the organized
and unorganized sectors, and skill development programs for workers.
268 Trade Adjustment in Asia: Past Experiences and Lessons Learned

References
Acemoglu, D. 2003. Patterns of Skill Premia. The Review of Economic
Studies. 70 (2). pp. 199−230.
Acemoglu, D., P. Aghion, C. Lelarge, and J. Van Reenen. 2007. Technology,
Information and the Decentralization of the Firm. The Quarterly
Journal of Economics. 122 (4). pp. 1759−99.
Acemoglu, D., and F. Zilibotti. 2001. Productivity Differences. The
Quarterly Journal of Economics. 116 (2). pp. 563−606.
Ahsan, R. 2013. Input Tariffs, Speed of Contract Enforcement, and the
Productivity of Firms in India. Journal of International Economics.
90 (1). pp. 181−92.
Ahsan, R., and D. Mitra. 2014. Trade Liberalization and Labor’s Slice
of the Pie: Evidence from Indian Firms. Journal of Development
Economics. 108 (1). pp. 1−16.
Amiti, M., and L. Cameron. 2012. Trade Liberalization and the Wage
Skill Premium: Evidence from Indonesia. Journal of International
Economics. 87 (2). pp. 277–87.
Amiti, M., and D. R. Davis. 2012. Trade, Firms, and Wages: Theory and
Evidence. The Review of Economic Studies. 79 (1). pp. 1−36.
Amiti, M., and J. Konings. 2007. Trade Liberalization, Intermediate
Inputs, and Productivity: Evidence from Indonesia. The American
Economic Review. 97 (5). pp. 1611−38.
Araújo, B. C., and L. S. Paz. 2014. The Effects of Exporting on Wages:
An Evaluation Using the 1999 Brazilian Exchange Rate Devaluation.
Journal of Development Economics. 111. pp. 1−16.
Attanasio, O., P. Goldberg, and N. Pavcnik. 2004. Trade Reforms and
Wage Inequality in Colombia. Journal of Development Economics.
74 (2). pp. 331−366.
Autor, D., F. Levy, and R. J. Murnane. 2003. The Skill Content of Recent
Technological Change: An Empirical Exploration. The Quarterly
Journal of Economics. 118 (4). pp. 1279−333.
Azam, M. 2010. India’s Increasing Skill Premium: Role of Demand and
Supply. The BE Journal of Economic Analysis and Policy. 10 (1).
pp. 1−26.
Baumgarten, D. 2013. Exporters and the Rise in Wage Inequality:
Evidence from German linked Employer−Employee Data. Journal
of International Economics. 90 (1). pp. 201−17.
Behrman, J. R., N. Birdsall, and M. Székely. 2000. Economic Reform and
Wage Differentials in Latin America. IADB Research Department
Working Paper. No. 435. Washington, DC: Inter-American
Development Bank.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 269

Bemard, A. B., and J.B. Jensen. 1997. Exporters, Skill Upgrading, and the
Wage Gap. Journal of International Economics. 42 (1−2). pp. 3−31.
Berman, E., J. Bound, and Z. Griliches. 1994. Changes in the Demand for
Skilled Labor within US Manufacturing: Evidence from the Annual
Survey of Manufactures. The Quarterly Journal of Economics.
109 (2). pp. 367−97.
Beyer, H., P. Rojas, and R. Vergara. 1999. Trade Liberalisation and
Wage Inequality. Journal of Development Economics. 59 (1).
pp. 103−23.
Biscourp, P., and F. Kramarz. 2007. Employment, Skill Structure and
International Trade: Firm-level Evidence for France. Journal of
International Economics. 72 (1). pp. 22–51.
Bloom N., B. Eifert, A. Mahajan, D. McKenzie, and J. Roberts. 2013. Does
Management Matter? Evidence from India. The Quarterly Journal
of Economics. 128 (1). pp. 1−51.
Bloom, N., L. Garicano, R. Sadun, and J. Van Reenen. 2014. The Distinct
Effects of Information Technology and Communication Technology
on Firm Organization. Management Science. 60 (12). pp. 2859−85.
Bloom, N., K. Manova, S. Sun, J. Van Reenen, and Z. Yu. 2016. Managing
Trade: Evidence from China and the US, Mimeo.
Bloom, N., R. Sadun, and J. Van Reenen. 2010. Does Product Market
Competition Lead Firms to Decentralize? The American Economic
Review Papers and Proceedings. 100. pp. 434−8.
Bloom, N., and J. Van Reenen. 2007. Measuring and Explaining
Management Practices Across Firms and Countries, The Quarterly
Journal of Economics. 122 (4). pp. 1351−1408.
____. 2010. Why Do Management Practices Differ across Firms and
Countries? Journal of Economic Perspectives. 24 (1). pp. 203–24.
Bonfatti, R., and M. Ghatak. 2013. Trade and the Allocation of Talent with
Capital Market Imperfections. Journal of International Economics.
89 (1). pp. 187−201.
Bustos, P. 2011. The Impact of Trade Liberalization on Skill Upgrading:
Evidence from Argentina, Mimeo.
Caliendo, L., and E. Rossi-Hansberg. 2012. The Impact of Trade on
Organization and Productivity. The Quarterly Journal of Economics.
127 (3). pp. 1393−467.
Card, D., and J. DiNardo. 2002. Skill-Biased Technological Change and
Rising Wage Inequality: Some Problems and Puzzles. Journal of
Labor Economics. 20 (S4). pp. S733−S783.
Chakraborty, P., and O. Raveh. 2018. Input-Trade Liberalization, and the
Demand for Managers: Evidence from India. Journal of International
Economics. 111. pp. 159−76.
270 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Chamarbagwala, R. 2006. Economic Liberalization and Wage Inequality

in India. World Development. 34 (12). pp. 1997−2015.
Chamarbagwala, R., and G. Sharma. 2011. Industrial De-licensing, Trade
Liberalization, and Skill Upgrading in India. Journal of Development
Economics. 96 (2). pp. 314−36.
Chen, C. 2017. Management Quality and Firm Hierarchy in Industry
Equilibrium. American Economic Journal: Microeconomics. 9 (4).
pp. 203−44.
Chen, B., M. Yu, and Z. Yu. 2017. Measured Skill Premia and Input
Trade Liberalization: Evidence from Chinese Firms. Journal of
International Economics. 109. pp. 31−42.
Cragg, M. I., and M. Epelbaum. 1996. Why Has Wage Dispersion Grown
in Mexico? Is It the Incidence of Reforms or the Growing Demand
for Skills? Journal of Development Economics. 51 (1). pp. 99−116.
Cunat, V., and M. Guadalupe. 2009. Globalization and the Provision
of Incentives inside the Firm. Journal of Labor Economics. 27 (2).
pp. 179−212.
Currie, J., and A. Harrison. 1997. Sharing the Costs: Trade Reform and
Labor Market Adjustment in Morocco. Journal of Labor Economics.
15 (S3). pp. S44−S71.
De Loecker, J., P. Goldberg, A. Khandelwal, and N. Pavcnik. 2016. Prices,
Markups and Trade Reform. Econometrica. 84 (2). pp. 445−510.
Debroy, B., and A. T. Santhanam. 1993. Matching Trade Codes with
Industrial Codes. Foreign Trade Bulletin. XXIV (1). pp. 5−27.
Dutta, P. V. 2005. Accounting for Wage Inequality in India. Indian
Journal of Labour Economics. 48 (2). pp. 273−95.
____. 2007. Trade Protection and Industry Wages in India. Industrial and
Labor Relations Review. 60 (2). pp. 268−86.
Eaton, J., and S. Kortum. 1996. Trade in Ideas Patenting and Productivity
in the OECD. Journal of International Economics. 40 (4). pp. 251−78.
____. 2001. Trade in Capital Goods. European Economic Review. 45 (7).
pp. 1195−1235.
Ebenstein, A., A. Harrison, M. McMillan, and S. Phillips. 2014. Estimating
the Impact of Trade and Offshoring on American Workers Using the
Current Population Surveys. The Review of Economics and Statistics.
96 (4). pp. 581−595
Feenstra, A. D., and G. H. Hanson. 1996. Foreign Investment, Outsourcing
and Relative Wages. In R. C. Feenstra, G. M. Grossman, and D. A.
Irwin, eds. Political Economy of Trade Policy: Papers in Honor of
Jagdish Bhagwati. Cambridge, MA: MIT Press. pp. 89−127.
____. 1997. Foreign Direct Investment and Relative Wages: Evidence
from Mexico’s Maquiladoras. Journal of International Economics.
42 (3−4). pp. 371−93.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 271

____. 2003. Global Production Sharing and Rising Inequality: A Survey

of Trade and Wages. In E. Choi and J. Harrigan, eds. Handbook of
International Trade. Malden, MA: Blackwell. pp. 146−85.
Feliciano, Z. 2001. Workers and Trade Liberalization: The Impact of
Trade Reforms in Mexico on Wages and Employment. Industrial
and Labor Relations Review. 55 (1). pp. 95−115.
Frías, J. A., D. S. Kaplan, and E. Verhoogen. 2009. Exports and Wage
Premia: Evidence from Mexican Employer-Employee Data,
Mimeo.
____. 2012. Exports and Within-Plant Wage Distributions: Evidence
from Mexico. The American Economic Review. 102 (3). pp. 435–40.
Galiani, S., and P. Sanguinetti. 2003. The Impact of Trade Liberalization
on Wage Inequality: Evidence from Argentina. Journal of
Development Economics. 72: 497−513.
Garicano, L. 2000. Hierarchies and the Organization of Knowledge in
Production. Journal of Political Economy. 108 (5). pp. 874−904.
Garicano, L., and E. Rossi-Hansberg. 2004. Inequality and the
Organization of Knowledge. The American Economic Review. 94 (2).
pp. 197−202.
____. 2006. Organization and Inequality in a Knowledge Economy. The
Quarterly Journal of Economics. 121 (4). pp. 1383–435.
Ghosh, S. 2014. Manufacturing Sector in India: Role of External
Economies and Industrial Policy, Mimeo, Jawaharlal Nehru
University.
Goldberg. P. K., A. Khandelwal, and N. Pavcnik. 2013. Variety In, Variety
Out: Imported Input and Product Scope Expansion in India,
In J. Bhagwati and A. Panagariya, eds. Reforms and Economic
Transformation in India. New York: Oxford University Press.
pp. 168−99.
Goldberg, P. K., A. Khandelwal, N. Pavcnik, and P. Topalova. 2010.
Imported Intermediate Inputs and Domestic Product Growth:
Evidence from India. The Quarterly Journal Economics. 125 (4).
pp. 1727−1767.
Goldberg, P. K., and N. Pavcnik. 2005. Trade, Wages, and the Political
Economy of Trade Protection: Evidence from the Colombian Trade
Reforms. Journal of International Economics. 66 (1). pp. 75−105.
____. 2007. Distributional Effects of Globalization in Developing
Countries. Journal of Economic Literature. 45 (1). pp. 39−82.
Gonzaga, G., N. Menezes-Filho, and C. Terra. 2006. Trade Liberalization
and the Evolution of Skill Earnings Differentials in Brazil. Journal of
International Economics. 68 (2). pp. 345−67.
Gorg, H., and G. Strobl. 2002. Relative Wages, Openness, and Skill-biased
Technological Change in Ghana, Mimeo.
272 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Grossman, G., and E. Rossi-Hansberg. 2008. Trading Tasks: A Simple

Theory of Offshoring. The American Economic Review. 98 (5).
pp. 1978−97.
Haltiwanger, J., A. Kugler, M. Kugler, A. Micco, and C. Page. 2004. Effects
of Tariffs and Real Exchange Rates on Job Reallocation: Evidence
from Latin America. Policy Reform. 7 (4). pp. 191−208.
Harrison, A., and G. Hanson. 1999. Who Gains from Trade Reform?
Some Remaining Puzzles. Journal of Development Economics. 59 (1).
pp. 125−54.
Hummels, D., R. Jorgensen, J. Munch, and C. Xiang. 2014. The Wage
Effects of Offshoring: Evidence from Danish Matched Worker-Firm
Data. The American Economic Review. 104 (6). pp. 1597−629.
Kandilov, I., A. Leblebicioglu, and R. Manghnani. 2016. Tariffs on
Imported Capital and Firm-Level Investment in the Indian
Manufacturing Sector, Mimeo, North Carolina State University.
Kijima, Y. 2006. Why Did Wage Inequality Increase? Evidence from Urban
India 1983−99. Journal of Development Economics. 81. pp. 97−117.
Koren, M., M. Csillag, and J. Kollo. 2019. Machines and Machinists:
Importing Skill-Biased Technology, Mimeo, Central European
University.
Krishna, P., D. Mitra, and S. Chinoy. 2001. Trade Liberalization and Labor
Demand Elasticities: Evidence from Turkey. Journal of International
Economics. 55 (2). pp. 391−409.
Krishna, P., J. P. Poole, and M. Z. Senses. 2014. Wage Effects of Trade
Reform with Endogenous Worker Mobility. Journal of International
Economics. 93 (2). pp. 239−52.
Kumar, U., and P. Mishra. 2008. Trade Liberalization and Wage
Inequality: Evidence from India. Review of Development Economics.
12 (2). pp. 291−311.
Levinsohn, J., and A. Petrin. 2003. Estimating Production Functions
Using Inputs to Control for Unobservables. The Review of Economic
Studies. 70 (2). pp. 317−341.
Ma, L., and D. Ruzic. 2018. Globalization and Top Income Shares, Mimeo.
Marin, D., and T. Verdier. 2008. Power inside the Firm and the Market:
A General Equilibrium Approach. Journal of the European Economic
Association. 6 (4). pp. 752–88.
____. 2014. Corporate Hierarchies and International Trade: Theory and
Evidence. Journal of International Economics. 94 (2). pp. 295−310.
Martins, P. S., and L. Opromolla. 2010. Exports, Imports and Wages:
Evidence from Matched Firm-Worker-Product Panels, Mimeo.
McCaig, B. 2011. Exporting Out of Poverty: Provincial Poverty in Vietnam
and US Market Access. Journal of International Economics. 85 (1).
pp. 102−13.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 273

Mehta, A., and R. Hasan. 2012. The Effects of Trade and Services
Liberalization on Wage Inequality in India. International Review of
Economics and Finance. 23. pp. 75−90.
Menezes-Filhoz, N., and M-A. Muendler. 2011. Labor Reallocation in
Response to Trade Reform, Mimeo.
Moulton, B. R. 1990. An Illustration of a Pitfall in Estimating the Effects
of Aggregate Variables on Micro Units. The Review of Economics and
Statistics. 72 (2). pp. 334−8.
Nouroz, H. 2001. Protection in Indian Manufacturing. New Delhi:
MacMillan Publishers India Ltd.
Pavcnik, N. 2003. What Explains Skill Upgrading in Less Developed
Countries? Journal of Development Economics. 71 (2). pp. 311−28.
Pavcnik, N., A. Blom, P. Goldberg, and N. Schady. 2004. Trade
Liberalization and Industry Wage Structure: Evidence from Brazil.
World Bank Economic Review. 18 (3). pp. 319−44.
Raveh, O., and A. Reshef. 2016. Capital Imports Composition,
Complementarities, and the Skill Premium in Developing Countries.
Journal of Development Economics. 118: 183−206.
Revenga, A. 1997. Employment and Wage Effects of Trade Liberalization:
The Case of Mexican Manufacturing. Journal of Labor Economics.
15 (S3). pp. S20−S43.
Smith, J. P., D. Thomas, E. Frankenberg, K. Beegle, and G. Teruel.
2002. Wages, Employment and Economic Shocks: Evidence from
Indonesia. Journal of Population Economics. 15 (1). pp. 161−93.
Stolper, W. F., and P. A. Samuelson. 1941. Protection and Real Wages. The
Review of Economic Studies. 9 (1). pp. 58−73.
Topalova, P., and A. Khandelwal. 2011. Trade Liberalization and Firm
Productivity: The Case of India. The Review of Economics and
Statistics. 93 (3). pp. 995−1009.
Verhoogen, E. 2004. Trade, Quality Upgrading and Wage Inequality in
the Mexican Manufacturing Sector: Theory and Evidence from an
Exchange-Rate Shock, Mimeo, University of California, Berkeley.
Verhoogen, E. A. 2008. Trade, Quality Upgrading, and Wage Inequality
in the Mexican Manufacturing Sector. The Quarterly Journal of
Economics. 123 (2). pp. 489−530.
Wood, A. 1995. How Trade Hurt Unskilled Workers. Journal of Economic
Perspectives. 9 (3). pp. 57−80.
Zhu, S. C., and D. Trefler. 2005. Trade and Inequality in Developing
Countries: A General Equilibrium Analysis. Journal of International
Economics. 65 (1). pp. 21−48.
274 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Appendix A8

Data
I use an annual-based panel of Indian firms that covers up to 8,000+
firms, across 108 industries within the manufacturing sector, over
1990–2011 (with the exception of specific cases, where specified). Unless
otherwise specified, variables are based on data from the PROWESS
database of the Centre for Monitoring Indian Economy (CMIE). All
monetary-based variables are measured in millions of Indian rupees,
deflated to 2005 using the industry-specific Wholesale Price Index.
All industry-level cases are based on the 2004 National Industrial
Classification (NIC).

Variable Definitions
o Managerial Compensation/Total Compensation: Share
of managerial compensation in total labor compensation;
compensation defined as the sum of all salaries, and additional
bonuses.
o Total Managers: Total number of managers in a firm. This is
the sum of the total number of managers at the top and middle
management level.
o Managerial Compensation: Total managerial compensation of a
firm. This is the sum of all the management layers put together.
o Nonmanagerial Compensation: Total nonmanagerial
compensation of a firm. This is the sum of compensation of all
nonmanagerial workers.
o Input/Output Tariffs: Input/output tariffs at the four-digit
industry level, obtained from Ahsan and Mitra (2014) for
1990–2003, with the balance collected from Chakraborty and
Raveh (2018).
o Imp/GVA: Share of total imports in gross value added.
o ImpRaw/GVA: Share of raw material imports in gross value
added.
o ImpCap/GVA: Share of capital imports in gross value added.
o ImpSto/GVA: Share of stores and spares imports in gross value
added.
o ImpFin/GVA: Share of final goods imports in gross value added.
o Exp/GVA: Share of total exports in gross value added.
o GVA: Gross value added, defined as the difference between total
sales and expenditures on raw materials.
 Trade Reform, Managers, and Skill Intensity: Evidence from India 275

o Skill Intensity: The three-digit industry-level ratio of

nonproduction workers to all employees, obtained from the
Indian Annual Survey of Industries (2001–2011) and from Ghosh
(2014) (1990–2000).
o Capital Employed: Total amount of capital employed by a firm.
o Productivity: Total factor productivity at the firm level is
computed using the Levinsohn and Petrin (2003) methodology.
o Factories: The three-digit industry-level number of factories/
plants.
o Management Technology: The four-digit industry-level
management quality score in 2004, obtained from Bloom and
Van Reenen (2010); the score is between 1 and 5, with 5 denoting
the highest quality.
o Technology Adoption: Share of research and development
expenditure and royalty payments for technical know-how in
gross value added.
o Assets: Total assets of a firm. This is an indicator of size.
o Age: Age of a firm in years.
o Ownership: This indicates whether a firm is domestically owned
or foreign owned.
 9
Multiproduct Firms,
Tariff Liberalization,
and Product Churning in
Vietnamese Manufacturing
Ha Thi Thanh Doan

9.1 Introduction
Multiproduct firms are the dominant players in international production
and trade (Bernard, Jensen, and Schott 2010; Goldberg et al. 2010a).
Moreover, these firms are active in alternating their combination of
product varieties. In fact, Bernard, Jensen, and Schott (2010) have
documented a frequent change in the product mix in the United States
(US), where almost 50% of multiproduct firms change their product mix
every 5 years. Indeed, firms’ adjustment in product scope constitutes
one important layer of firm heterogeneity (Nocke and Yeaple 2006).
Understanding firms’ product adjustment is crucial for several
reasons. First, changes in the commodity mix of manufacturing firms
affect firms’ output and productivity, through which they exert an impact
on the economy’s aggregate growth. For example, Bernard, Jensen, and
Schott (2006) have demonstrated that the contribution to output growth
of a product margin outweighs that of firm entry and exit. Goldberg et al.
(2010a) have observed a similar phenomenon in Indian manufacturing,
where changes in firms’ product mix contributed to as much as 25%
of output expansion. In this regard, the changing of product lines is
a nontrivial channel of resource reallocation within firms. Second,
switching production activities has important implications for the
structural shift across sectors. For instance, a shift away from resource-
based and primary products to more capital-intensive products, a source
of industrial upgrading, will induce the economy to move to the next
stage of the industrialization process.

276
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 277

Why some firms diversify their production is not a new question

in the industrial organization literature. For instance, Penrose (1955)
has suggested that product diversification provides firms with greater
opportunities for market expansion, which can be limited if they only
manufacture a single product. Recent studies on international trade and
firm heterogeneity, however, have proposed a different approach. Most
of the theoretical models on firms’ responses to trade at the product
level predict that product dropping is popular among all multiproduct
firms (Eckel and Neary 2010; Mayer, Melitz, and Ottaviano 2014).
Competition pressure instigates firms to narrow down their product
range by dropping peripheral products and reallocating resources to
their core competencies, defined as the product with the largest cost
advantage compared to other products of the firm. Just as the least
productive single-product firms are swept out of the market due to
competition, the least productive product for each multiproduct firm
should also be dropped.
However, several studies suggest a more heterogeneous picture,
where an adjustment in product scope is contingent on the firm’s
position in the productivity distribution, firm size, or ownership type
(Qiu and Zhou 2013; Lopresti 2016). Lopresti (2016), for example,
examined changes in the product structure of US firms following
the Canada–US Free Trade Agreement of 1989. Utilizing Bayesian
econometric techniques, the author found that heterogeneity exists
in firms’ response conditioning regarding their engagement in global
markets. In particular, more domestically oriented firms narrow down
their product range, while more internationalized firms either add
more products or do not respond to tariff reduction. Nevertheless, the
adjustment is mixed when sales are used as an additional dimension
of firm heterogeneity. Given these inconsistent theoretical findings, a
conclusion remains an empirical matter.
This research adds to the growing literature on firm–product
dynamics by investigating product turnover in Viet Nam's
manufacturing, a developing country with impressive economic
growth and a high level of trade openness. I utilize the Vietnam
Enterprise Survey covering the 2010–2015 period. The research
objectives are threefold. I first present several stylized facts about
multiproduct firms, including their presence in manufacturing, their
relative performance compared to single-product enterprises, and
the frequency of product turnover. I then utilize the decomposition
framework in Goldberg et al. (2010a) to examine the contribution of
the extensive and intensive firm–product margin to aggregate output
growth. Finally, I link product refocusing to trade liberalization as one
of the most significant policy reforms during this period. In particular,
278 Trade Adjustment in Asia: Past Experiences and Lessons Learned

this chapter addresses two questions. First, does a reduction in tariffs

impact firms’ product scope? Second, do responses vary depending on
firms’ trade status and ownership types?
To the best of my knowledge, this is the first study on Viet Nam.
This chapter is closely related to Goldberg et al. (2010a), who
examined product turnover in response to a reduction in tariffs in
Indian manufacturing. However, my study deviates from Goldberg et
al. (2010a) in two important respects. First, I consider the potential
differences in scope decisions depending on firms’ ownership. In
Viet Nam, there exists a large gap in competitiveness and efficiency
among multinational enterprises (MNEs), state-owned enterprises
(SOEs), and small and medium-sized enterprises (SMEs). SMEs
account for over 90% of firms and make an important contribution to
job creation. However, this sector has low competitiveness and limited
innovation and internationalization activities (Trinh and Doan 2018).
Facing financial and managerial constraints, it is possible that these
firms have limited flexibility to adjust their product mix. Foreign
investors, on the other hand, are larger, more productive, and the main
exporters.1 Therefore, it is likely that MNEs are more proactive in
product adjustment. Given the country’s heavy dependence on MNE
exports, MNEs’ internal resource reallocation is expected to exert a
nonnegligible impact on aggregate trade and industrial performance.
The third group, SOEs, tend to behave differently from MNEs and
SMEs, as profit maximization may not be their business target. This
implies that the core-competency argument does not necessarily apply
to SOEs.
In addition, I take into account differences in a firm’s response
to trade depending on its export status. More diverse output markets
allow exporters to better cope with increased competition in one
market, while their productive performance encourages them to take
advantage of better market access to expand their scope. Lopresti (2016)
has shown that domestic-oriented firms become leaner in response to
trade shocks. In contrast, firms with a greater share of foreign sales
expand. Baldwin and Gu (2009) have found that trade liberalization
induces nonexporting firms to narrow down their scope, but there is no
significant effect on exporters. Although I do not have data on exports
by product, trade status could reveal potential heterogeneity according
to firms’ engagement in the international market.
From a policy perspective, this study can contribute in the following
ways. First, to the extent that changes in product mix account for a

1
The foreign direct investment sector accounts for 50% of output and approximately
70% of export turnover in 2016.
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 279

nontrivial fraction of aggregate growth, a study on multiproduct firms

can shed light on another important channel for enhancing allocative
efficiency. While better resource reallocation is crucial for any economy,
productivity improvement is currently one of the top priorities for
policy makers in Viet Nam. As one of Asia’s fastest-growing economies,
Viet Nam has lifted itself out of poverty and achieved the status of a
lower-middle-income country. However, impressive economic growth
during the last 2 decades primarily originates from an extraordinary
structural shift from agriculture and considerable labor expansion.
The contribution of productivity, the third component of growth,
remains limited (World Bank 2017). Second, examining the product
scope decision can also facilitate understanding of the changes in
the commodity composition of production observed at the aggregate
level. According to Nguyen (2015), the contribution of resourced-
based industries to overall manufacturing output has fallen markedly.
For example, the ratio of output of the chemical products industry
plunged from 7% to just 0.1%, while that of processed food fell from
32.4% to 24.2% over the 1995–2009 period. There has been a shift to
more capital-intensive industries, such as electronics and computing.
I expect, therefore, that this study can contribute to the discussion on
industrial upgrading and sustainable growth.

9.2 Literature Review

This study is related to the literature on multiproduct firms and trade
liberalization. On the theory side, most models predict that more
competitive markets stimulate firms to drop their least profitable
product and refocus on the product with the largest cost advantage, or
the core product. Eckel and Neary (2010) have constructed a model in
which globalization affects both the extensive and intensive margin of
multiproduct firms through a competition effect and a cannibalization
effect. Adjustment of internal demand linkages, or the cannibalization
effect, allows firms to improve productivity by becoming leaner. In
contrast, competition implies a decline in product variety. Bernard,
Jensen, and Schott (2010) have extended Melitz’s (2003) model by
allowing firms to produce multiple products. The theoretical model
suggests that severe competition in more liberalized industries drives
the least productive firms and the least profitable products of firms out
of the market. Mayer, Melitz, and Ottaviano (2014) assume that firms
face a product ladder. Productivity or quality is negatively associated
with the number of varieties produced. Tougher competition results
in lower markups across products, rendering firm sales skewed toward
core competences.
280 Trade Adjustment in Asia: Past Experiences and Lessons Learned

However, Qiu and Zhou (2013) have predicted product scope

expansion for more productive firms. They argue that if we relax the
assumption of a fixed fee for the introduction of each new variety and
allow the fees to increase steeply, highly productive firms can still earn a
profit by expanding their product scope. Dhingra (2013) has argued that
the varieties produced by one firm are more substitutable than varieties
across firms. Product expansion then reduces demand for existing
products within the firm.
Inconclusive theoretical predictions suggest the essential role
of empirical analysis. Baldwin and Gu (2009) have found that tariff
reduction leads small firms to narrow down product scope, whereas
large firms do not. Moreover, nonexporters drop products, whereas
the impact on exporters is not significant. The authors argue that
once firms enter the export market, they are more affected by factors
other than tariffs, including learning-by-exporting, competition in the
export market, and opportunities for better market access. Goldberg
et al. (2010a) have identified a nonnegligible impact of changes in
product mix on changes in output in Indian manufacturing. Trade
liberalization (proxied as tariffs), however, does not have a significant
impact on a firm’s extensive margins. They postulate that strict
industrial regulations in India may limit firms’ flexibility in shedding
existing product lines. Iacovone, Rauch, and Winters (2013) have
found import competition from the People’s Republic of China result
in a fall in sales and number of products in the case of Mexican firms.
The impact is highly heterogeneous across extensive and intensive
margins. Smaller plants and more marginal products are negatively
affected. In contrast, large firms and core products do not seem to
be affected. Moreover, large firms benefit from access to cheaper
imported intermediate inputs from the People’s Republic of China.
Arkolakis and Muendler (2010) have investigated the case of Brazilian
exporters and demonstrated that the firm-product extensive margin is
heterogeneous across firm sizes. Liu (2010) has noted that Canadian
multiproduct firms are more likely to refocus on their core products
in response to trade liberalization. The author constructed indices of
product relatedness and demonstrated that the weaker the linkages
between marginal products and the core product, the more likely it
is that peripheral products are dropped. Goldberg et al. (2010b) have
examined another aspect of within-firm reallocation, asking whether
exposure to trade liberalization affects the input allocation of firms.
The empirical results showed a positive impact of a lower input tariff
on the introduction of new products thanks to better access to new
intermediate inputs.
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 281

9.3 Data Source

My primary data source is the Vietnam Enterprise Survey (VES)
provided by Viet Nam’s General Statistics Office. Data have been collected
annually since 2000, and the VES is by far the most comprehensive
dataset available on Viet Nam’s firms; it is the main source of firm-level
statistics in the formal agriculture, industry, and service sectors.
The VES includes a general questionnaire covering basic statistics at
the firm level, including ownership, assets and liability, employment, sales,
capital stock, and industry code from January to December of a particular
year. The survey covers all SOEs and foreign direct investment (FDI)
without any firm size threshold. As for domestic private firms, however, a
certain threshold is applied.2 All formal firms with employment size above
the threshold are included, while firms below the threshold are chosen
by random sampling. Since 2010, the VES has also provided information
on total exports and imports.3 There is a consistent and unique tax code
assigned to each firm, which allows us to track the firm across years.4
Apart from the general module, the General Statistics Office also
designs industry-specific modules to survey the activities of each sector.
For manufacturing, production data are provided at the plant level.
The data comprise the list of products, the quantity produced for each
product, the unit of measurement, the value of sales, and product codes,
among others. The General Statistics Office applies an internal product
classification developed based on the Viet Nam Standard Industrial
Classification (VSIC) version 2007, European Classification of Products
by Activity 2008, United Nations Central Product Classification 2.0, and
Harmonized System 2007. Products are classified at eight digits, where
the first five digits correspond to VSIC 2007. Under this classification,
there are approximately 2,400 products in the manufacturing sector.

9.3.1 Variables

The key variables for our analysis are product codes and product sales.
Product sales are deflated by the producer price index (PPI) at the

2
The threshold varies across years, provinces, and sectors. For example, in 2015, the
threshold goes up to 100 in certain sectors for firms located in Ha Noi and Ho Chi
Minh City. On the other hand, the maximum threshold for 2008 is only 10. For the
census years (2006, 2011, and 2016), all formal firms were included.
3
Before 2010, trade status is only available for a few years.
4
A detailed description of the firm-level dataset is provided in Ha and Kiyota (2014).
282 Trade Adjustment in Asia: Past Experiences and Lessons Learned

two-digit sectoral level. Due to a change in product classification in

2010, my analysis is limited to the 2010–2015 period. In addition, we
also utilize information on firms’ unique identification to construct the
panel, and firms’ industry as indicated in the general module. Value-
added deflated by PPI and employment data are used to compute labor
productivity.
Given that the production module is at the plant level whereas the
general module is at the firm level, I aggregate all data in the production
module to firm level for consistency. As the production decision is made
at the firm level, an analysis at the firm level is also more appropriate
(Bernard, Jensen, and Schott 2010). Furthermore, I only focus on the
manufacturing products of firms.
To complement the firm-product data, I use tariff data from the
World Integrated Trade Solutions (WITS) database at four-digit
International Standard Industrial Classification (ISIC) revision 3.
I match ISIC with VSIC codes based on a concordance table provided
by the General Statistics Office. I utilize effectively applied tariff, which
is defined as the lowest available tariff. I favor trade-weighted tariff over
simple average tariff, as the former can capture the relative importance
of each industry’s import share.
To account for the impact of trade liberalization on access to
imported intermediate inputs, I also measure input tariff following
Amiti and Konings (2007) as follows:
p

inst = ∑ asp ∗ out pt , (1)

1

where inst and outpt denote input tariff of downstream sector s and
output tariff of two-digit upstream sector p, respectively. asp denotes
imported input coefficients,∆Yit = defined
∑ ∆Yas the value of intermediate
ijt + ∑ ∆Yijt ,
imports from sector p over total j∈C output of sectors.
j∈E
5
To compute input
coefficients, we utilize the Organisation for Economic Co-operation
and Development’s Inter-country Input–Output Table (ICIO) edition
2016. ICIO provides annual information on inter-industry and across-
country trade transactions∆Y for 63 ∑
t = countries
i(∑j∈A ∆Y including
ijt + ∑j∈D Viet Nam
∆Yijt + ∑over theijt + ∑j∈F ∆Yijt ).
j∈R ∆Y
1995–2011 period. Industrial classification is based on ISIC Rev. 3 at the
two-digit level. Accordingly, 34 sectors are covered.
I favor the use of ICIO over Viet Nam’s domestic input–output table
for two reasons. First, ICIO = β1 out
Yjtadopts ISIC+classification,
theit−1 β2 init−1 + βwhich
3 X jt−1can
+ βbe
4 HHIit + αs + 𝑎𝑎𝑡𝑡

5
Note that we can only measure input tariff at the two-digit sectoral level due to data
availability.
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 283

matched directly with output tariff data from WITS. Second, and more
importantly, ICIO contains information on imported intermediates,
which is not available in the domestic table. To better capture the impact
of tariff changes on a firm’s adjustment along the supply chain, it is more
appropriate to measure the imported input coefficient than the domestic
input coefficient. Although the database is available for 1995–2011,
I only use ICIO for 2011, assuming that the structure of the economy is
relatively stable across 2010–2015.

9.4 A Profile of Multiproduct Firms

This section documents the characteristics of multiproduct firms and the
pervasiveness of product churning in Viet Nam’s manufacturing during
a 6-year period from 2010 to 2015. Following Goldberg et al. (2010a),
I define sector and industry at the two- and four-digit levels of VSIC 2007,
respectively. Product classification is defined at the eight-digit level.
Table 9.1 illustrates the presence of multiproduct firms in my sample.
I include four groups of firms: firms that produce only one product,
firms that produce at least two products, firms that operate in more
than one four-digit industry, and firms with activities spread across
two-digit sectors. Two features stand out. First, Viet Nam’s firms are
relatively specialized. On average, only 19% of firms produce more than
one product. An average multiproduct firm manufactures 2.6 products.
The proportions of multiple-industry and multiple-sector firms are

Table 9.1 Frequency and Output Share of Firms

Single Multiple Multiple Multiple

Product Product Industry Sector
Whole sample
Share of firms 0.81 0.19 0.07 0.05
Share of output 0.59 0.41 0.24 0.20
Average number of products, industries 1 2.62 1.45 1.28
or sectors per firm
FDI
Share of firms 0.81 0.19 0.07 0.04
Share of output 0.56 0.44 0.28 0.24
Average number of products, industries 1 2.73 1.39 1.25
or sectors per firm
continued on next page
284 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 9.1 continued

Single Multiple Multiple Multiple
Product Product Industry Sector
SOE
Share of firms 0.53 0.47 0.25 0.19
Share of output 0.26 0.74 0.50 0.46
Average number of products, industries 1 2.93 1.80 1.50
or sectors per firm
Domestic private
Share of firms 0.82 0.18 0.07 0.05
Share of output 0.72 0.28 0.13 0.08
Average number of products, industries 1 2.58 1.45 1.28
or sectors per firm
FDI = foreign direct investment, SOE = state-owned enterprise.
Notes: FDI sector includes 100% foreign-invested firms and joint ventures of which the share of foreign
capital exceeds 50% of total legal capital. Sector and industry are defined at the two- and four-digit levels
of the Viet Nam Standard Industrial Classification 2007, respectively.
Source: Author’s calculations from the Vietnam Enterprise Survey data.

even smaller, accounting for 7% and 5% of firm share, respectively. The

figure is significantly lower than that reported in Bernard, Jensen, and
Schott (2010) on the US and in Goldberg et al. (2010a) on India. Both
studies documented a share of around 40% of multiproduct firms. The
difference, however, is not surprising as in Viet Nam over 90% of firms
are micro, small, and medium-sized firms with limited technological
capability and low competitiveness.
Second, multiproduct firms tend to be larger. Despite the modest
firm share, they contribute to 41% of total output, which is similar
to the US and India, where the output share of multiproduct firms
is also double that of firm share. Third, there exists heterogeneity
across ownership types. Contrary to the overall trend, we observe the
prevalence of multiproduct firms in the SOE sector. They constitute
nearly 50% of total SOEs and account for 74% of output. The average
number of products is also higher than the overall, reaching 2.93. In
contrast, the FDI and domestic private sectors show a similar structure,
closely in line with the overall trend.6 One possible explanation for the
specialization of MNEs is their exploitation of economies of scope. On

6
It should be noted that SOEs account for a minority of my sample. Therefore, it is
likely that the overall trends are driven by domestic private firms and FDI.
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 285

Table 9.2 Superiority of Multiproduct Firms

Multiple Multiple Multiple

Product Industry Sector
Output 1.131 1.067 0.98
Export probability 0.161 0.143 0.153
Labor productivity 0.278 0.262 0.224
Employment 0.704 0.705 0.675
Capital intensity 0.25 0.22 0.185
Notes: Sector and industry are defined at the two- and four-digit level of the Viet Nam Standard Industrial
Classification 2007, respectively. Each column reports the regression result of firms’ characteristics
according to status: multiproduct, multi-industry, and multisector. We use a dummy variable on the right-
hand side to indicate each status. Industry-fixed effects are also included. All estimates are significant at
the 5% level.
Source: Author’s calculations from the Vietnam Enterprise Survey data.

the other hand, small capacity may limit domestic private firms in terms
of diversifying their product portfolio.
Studies on multiproduct firms highlight the premium in terms of
performance of more diversified enterprises. Firms face fixed costs
when expanding their scope. Just as more productive firms self-
select into export markets, only better-performing firms will choose
to become multiproduct firms. I check if this is also the case for Viet
Nam by looking at the relative characteristics of multiproduct firms
compared to their single-product counterparts. Table 9.2 documents
the characteristics of multiproduct firms. I find consistent evidence
within the existing literature regarding their superiority. In particular,
Viet Nam’s multiproduct firms are more productive; they have higher
labor productivity (0.27 log points), produce larger output, employ
more workers, and are more capital-intensive. They are also more active
in international markets, being 16% more likely to export. In short,
multiproduct firms outperform single-product firms.
Having examined the frequency and overall performance of
multiproduct firms, I now turn to the product structure of these firms.
Table 9.3 presents the sales distribution of products within firms. It
is clear that the distribution is highly skewed, meaning that a large
proportion of firm sales is generated from few primary products, which
is indicative of the core-competency hypothesis.
The average sales share of the largest product decreases from 74%
to 42% as the firm’s production increases from 2 to 10 or more. However,
even for firms with a large number of products, sales of the “core”
product account for at least 42% of total manufacturing sales.
286 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 9.3 Sales Distribution across Products

Number of Products Produced by the Firm

Rank of sales in 1 2 3 4 5 6 7 8 9 10+
descending order
100 74 63 57 53 48 45 43 43 42
26 26 25 24 24 23 22 21 21
11 13 13 14 14 14 13 12
6 7 8 8 9 9 8
3 4 5 6 6 6
2 3 4 4 4
2 2 3 3
1 1 2
1 +1
1
Note: The columns indicate number of products; the rows indicate the sales share of each product in firms’
total manufacturing sales.
Source: Author’s calculations from the Vietnam Enterprise Survey data.

9.5 Firm’s Adjustment of Product Scope

and Aggregate Output Growth
The existing literature suggests the importance of product churning for
aggregate economic outcome. To investigate the issue, I begin this section
by documenting the dynamics of product adjustment. I classify firms’
activities into one of four mutually exclusive groups. The “No activity”
category includes firms that do not change their product line in the period
of study. “Add” refers to firms that produce new products in period t that
are not in their product line in period t–1. “Drop” means that firms stop
producing a product in period t, which was produced in period t–1. Finally,
“Add and drop” includes firms that alternate their product mix by both
adding and dropping. I focus on changes in product structure of the firm
over time. Therefore, in this section I only use a subsample of continuing
firms that appear in the sample throughout the whole period.
Table 9.4 shows the share of firms that alternate their product mix
over 1-year, 3-year, and 6-year periods. A balanced panel is used for this
analysis. The main findings from Table 9.4 are threefold. First, product
churning is pervasive among Viet Nam’s manufacturing firms. Sixty
percent of all firms adjust their product range over a 6-year period.
The corresponding numbers for 3-year and 1-year periods are 50% and
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 287

Table 9.4 Frequency of Product Turnover

Percentage of Firms (Unweighted)
6-Year Period 3-Year Period 1-Year Period
Single Multiple Single Multiple Single Multiple
All Product Product All Product Product All Product Product
No activity 40.0 50.6 13.0 49.4 57.7 21.8 63.6 71.3 36.6
Add only 7.3 7.9 5.9 5.9 5.8 6.4 5.0 4.7 6.4
Drop only 8.8 NA 31.0 6.6 NA 28.5 5.7 NA 25.5
Add and drop 43.9 41.4 50.2 38.1 36.5 43.3 25.7 24.0 31.5
Percentage of Firms (Weighted by Sales)
6-Year Period 3-Year Period 1-Year Period
Single Multiple Single Multiple Single Multiple
All Product Product All Product Product All Product Product
No activity 34.5 46.1 16.5 45.9 56.9 29.4 64.2 77.1 48.5
Add only 8.7 8.6 8.8 15.1 10.0 22.6 12.3 7.0 18.9
Drop only 10.3 NA 26.3 7.6 NA 18.9 6.5 NA 14.4
Add and drop 46.5 45.3 48.4 31.5 33.1 29.1 17.0 16.0 18.3

NA = not available.
Note: No activity means that the firm’s product mix does not change between two consecutive periods.
A product is added if it was produced in period t but not in the previous period. Similarly, a product is dropped
if it was produced in period t–1 but not in period t. The statistics are computed on a balanced panel.
Source: Author’s calculations from the Vietnam Enterprise Survey data.

40%, respectively. When I weigh our sample by firm sales, the number
changes slightly, with 65% of firms changing their product mix over the
whole period. The annual pattern, while less pervasive, also shows a
high level of product turnover, with 40% of firms changing their product
mix. Furthermore, I observe that multiproduct firms are more active in
adjusting their product scope compared to single-product firms. Over
80% of the former group add and/or drop some products within 6 years.
In addition, product dropping is much more popular than product
adding. Firms that only add products account for less than 10% of the
unweighted sample.
To further investigate the pattern of product churning, I categorize
firms by ownership types. The FDI sector includes 100%-foreign-
invested enterprises and joint ventures in which foreign capital accounts
for at least 51% of total legal capital. The domestic private sector covers
the rest of our sample. We conjecture that the behavior of these groups
is heterogeneous along the product dimension given their performance
gap. Table 9.5 reports the results.
288 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 9.5 Product Turnover of Multiproduct Firms

by Ownership Type

Domestic Domestic
MNEs SOEs Private MNEs SOEs Private
Percentage of Multiproduct Firms: Percentage of Multiproduct Firms:
Unweighted, 6-Year Period Weighted by Sales, 6-Year Period
No activity 15.37 20.57 10.97 13.64 22.39 13.88
Add only 6.77 2.13 5.74 10.07 3.45 7.44
Drop only 35.12 24.82 29.66 23.39 31.6 29.94
Add and 42.74 52.48 53.64 52.9 42.56 48.74
drop
MNE = multinational enterprise, SOE = state-owned enterprise.
Note: No activity means that the firm’s product mix does not change between two consecutive periods.
A product is added if it was produced in period t but not in the previous period. Similarly, a product is
dropped if it was produced in period t–1 but not in period t. The statistics are computed on a balanced panel
of multiproduct firms only.
Source: Author’s calculations from the Vietnam Enterprise Survey data.

The left panel of Table 9.5 presents results without output weight,
while the right panel includes output weight. The figures suggest that
compared to SOEs, MNEs and domestic private firms are more active in
adjusting their product mix: 85% of MNEs and 90% of domestic private
firms change their product portfolio over a 6-year period. Moreover,
albeit modest compared to the other three activities, the ratio of product
adding is larger for MNEs and domestic private firms than for SOEs.
One may be concerned that the low percentage of product adding
could originate from coding or reporting errors. However, if that is the
case, one should also expect a lack of evidence on product dropping. My
statistics demonstrate the opposite. Furthermore, if firms deliberately or
mistakenly dropped some products in the survey, it is likely that the gap
between total manufacturing sales reported in the general module and
total product sales from the production module would be remarkable.
I have compared the two datasets and found a good match. Third, if
the list of products were not reported correctly, missing information
on sales and quantity produced would probably constitute an issue. My
database, on the contrary, provides detailed information on sales and
physical output of each product with negligible numbers of missing
values. Therefore, it is expected that the number adequately reflects the
actual pattern of product churning.
Changes in the product mix make a nontrivial contribution to
changes in the output of incumbents. To account for the sources of
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 289

p
output growth, I decompose growth of gross sales into two components:
inst = ∑ asp ∗ out pt ,
changes in the product mix and changes due to existing products.
I define these two sources as1 extensive margin and intensive margin.
Growth of output can then be expressed as

∆Yit = ∑ ∆Yijt + ∑ ∆Yijt , (2)

j∈C j∈E
p
where Y denotes
inst = ∑ asp ∗output
out pt (sales);
, and i, j, and t denote firm, product, and
time, respectively. C represents the set of continuing products (intensive
1
margin), and E represents
∆Yt =the∑iset of products
(∑j∈A ∆Yijt + ∑that only
j∈D ∆Yijt appear
+ ∑j∈Rin ijt + ∑j∈F ∆Yijt ).
∆Yeither
period t or period t–1.
Following Goldberg et al. (2010a), I further decompose the net
extensive
∆Yit = ∑ margin into
∆Yijt + ∑the ∆Y contribution
, of added products (A) and
dropped products (D). = βijt1 out it−1
YjtSimilarly, the +netβintensive
2 init−1 + margin
β3 X jt−1consists
+ β4 HHIofit + αs + 𝑎𝑎𝑡𝑡 +
j∈C j∈E
two components: the fall (F) and rise (R) of individual product sales.
Then aggregate output growth can be computed as

∆Yt = ∑i(∑j∈A ∆Yijt + ∑j∈D ∆Yijt + ∑j∈R ∆Yijt + ∑j∈F ∆Yijt ). (3)

Table 9.6 presents the decomposition. Two major findings stand

out. = β1 out
Yjt First, the it−1 + β2 init−1
contribution of +
theβ3intensive
X jt−1 + βmargin,
4 HHIit +
orαthe
s + change
𝑎𝑎𝑡𝑡 + ujtin
sales of individual products, exceeds that of the extensive margin. On
average, out of a growth in output of 6.4 percentage points, 6 percentage
points are from the intensive margin. Product churning only contributes
to 0.4 percentage points. Second, on both the extensive and intensive

Table 9.6 Contribution of Product Turnover to Output Growth

Extensive Margin Intensive Margin

Gross Rising Falling
Sales Net Addw Drop Net Products Products
2010
2011 1.5 2.2 13.9 –11.7 –0.7 7.4 –8.1
2012 6.9 –0.8 12.7 –13.5 7.6 18.6 –11.0
2013 15.7 0.6 11.0 –10.4 15.1 24.8 –9.7
2014 1.6 0.5 9.4 –9.0 1.1 13.4 –12.3
2015 6.4 –0.5 11.7 –12.2 6.8 16.2 –9.3
Source: Author’s calculation based on the Vietnam Enterprise Survey data.
290 Trade Adjustment in Asia: Past Experiences and Lessons Learned

margin, product adding or growing products makes a significant

contribution to the net increase. In the case of the intensive margin,
the growth is large enough to offset the negative impact of shrinking
products, leading to high overall output growth. For the extensive
margin, however, the negative impact of product dropping is too large
to be compensated by product adding. The net extensive margin is
thus relatively small. This observation is consistent with my previous
analysis, where product dropping is prevalent.

9.6 Trade Liberalization and Product Turnover

p
The literature on international trade and firm heterogeneity emphasizes
inst = ∑
product asp ∗ out
churning , important channel of resource reallocation as
asptan
a result of free trade. While the theoretical predictions and empirical
1
evidence do not provide a clear-cut picture of the direction of impact,
most studies suggest a relationship between product dropping and
trade liberalization. Given the high rate of product dropping found in
the = ∑ ∆Ysections
∆Yitprevious ijt + ∑ and∆Yijtthe
, substantial trade reform that Viet Nam’s
economy has experienced, it is then natural to ask if the relationship
j∈C j∈E
holds in the case of Viet Nam. To shed light on this issue, this section
examines the links between reduced trade costs and firms’ extensive
margin.
∆Yt = ∑iIn (∑particular, I ask whether firms in industries with larger
j∈A ∆Yijt + ∑j∈D ∆Yijt + ∑j∈R ∆Yijt + ∑j∈F ∆Yijt ).
tariff changes experience product churning. I follow Baldwin and Gu
(2009) to estimate the following equation on continuing firms:

Yjt = β1 out it−1 + β2 init−1 + β3 X jt−1 + β4 HHIit + αs + 𝑎𝑎𝑡𝑡 + ujt (4)

where Yjt represents the number of product varieties of firm j in four-

digit industry i at time t; outit-1 measures lagged output tariff of industry j
at time t; init-1 is lagged input tariff of sector s at time t; and Xjt is a vector
of firm-specific characteristics, including employment, lagged export
status, lagged export share over total output, and interaction terms
between ownership type and trade variables. I include the Herfindahl–
Hirschman concentration index HHI to capture competition at the
industry level. Further, at is the year dummy and αs the unobserved two-
digit sector s fixed effect.
The main research question here is whether changes in the output
tariff affect the number of products of the firm, controlling for the input
tariff, export status and export intensity, and ownership structure of the
firm. The choice of control variables is based on the literature. The input
tariff, for example, has been widely used in studies on trade liberalization
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 291

and firm productivity.7 For the literature on product turnover, the input
tariff is included in the analysis of Goldberg et al. (2010a, 2010b), among
others. While the reduction in output tariffs could intensify competition
pressure, a lower input tariff provides access to more intermediate inputs
varieties. For a developing country with limited technological capacity
such as Viet Nam, it is possible that advanced technology embodied in
more advanced imported intermediates lowers the cost of innovation
and encourages the development of new products, contributing to
aggregate output growth.
In addition, as Lopresti (2016) suggests, the impact of trade
liberalization on product scope depends on the extent of a firm’s
participation in the international market. A more globalized firm,
defined as one with larger export sales over total output, tends to add
more product or keep the product portfolio unchanged in response
to lower trade costs. On the other hand, a more domestically oriented
firm drops its product when facing international competition. To check
whether this observation holds for Viet Nam, I include in my estimation
export intensity, defined as the ratio of export turnover to a firm’s total
revenue, and its interaction term with the output tariff.
Table 9.7 reports the regression results. Columns (1)–(5)
demonstrate the relationship between the number of products and
tariffs in level. Columns (6)–(10) examine the determinants of firms’
product extensive margins. Several findings are worth mentioning. First,
in level, a higher output tariff is associated with a smaller number of
products. The coefficient on the output tariff is negative and significant.
Firms in industries with a lower output tariff are more likely to produce
more products. One possible explanation is the competition effect,
where firms diversify to reduce competition pressure. This finding is
consistent with Dang (2017), who finds that import competition from
the People’s Republic of China stimulates Viet Nam’s firms to introduce
new products. The economic magnitude is small, however. Second, a
higher input tariff is associated with a broader product range. I do not
find evidence of expansion of product scope thanks to better access to
more imported intermediates. One reason could be that, via access to
more technologically advanced materials, firms are more likely to invest
in quality upgrading of existing products.

7
See, for example, Amiti and Konings (2007), Topalova and Khandelwal (2011), and
Bas (2012).
292 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 9.7 Tariff Reduction and Product Churning

(1) (2) (3) (4) (5)

Variables Number of Products Produced by Firms
Output tariff –0.003*** –0.003*** –0.004*** –0.004*** –0.003***
(0.001) (0.001) (0.001) (0.002) (0.001)
Input tariff 0.071** 0.078*** 0.077*** 0.073***
(0.027) (0.028) (0.028) (0.028)
Export dummy 0.038** 0.040**
(0.016) (0.016)
Export * output tariff 0.000 0.000
(0.001) (0.001)
SOE * output tariff 0.003
(0.002)
FDI * output tariff 0.001
(0.002)
Export intensity 0.041
(0.036)
Export intensity * 0.000
output tariff
(0.003)
∆output tariff

∆input tariff

Export * ∆output tariff

SOE * ∆output tariff

FDI * ∆output tariff

Export intensity *
∆output tariff

Employment 0.053*** 0.054*** 0.054***

(0.011) (0.011) (0.011)
HHI 0.021 0.021 0.024
(0.072) (0.072) (0.072)
continued on next page
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 293

Table 9.7 continued

(1) (2) (3) (4) (5)
Variables Number of Products Produced by Firms
Constant 1.602*** 1.490*** 1.200*** 1.204*** 1.200***
(0.148) (0.156) (0.165) (0.165) (0.164)
Observations 42,908 42,908 42,905 42,905 41,508
R-squared 0.004 0.004 0.006 0.006 0.005
Number of firms 7,294 7,294 7,294 7,294 7,257

(6) (7) (8) (9) (10)

Variables Change in Number of Products
Output tariff

Input tariff

Export dummy 0.023*** 0.023***

(0.008) (0.008)
Export * output tariff

SOE * output tariff

FDI * output tariff

Export intensity 0.042***

(0.015)
Export intensity * output
tariff

∆output tariff –0.001 –0.001 –0.001 –0.001 –0.001

(0.001) (0.001) (0.001) (0.001) (0.001)
∆input tariff –0.010 –0.011 –0.011 –0.016
(0.017) (0.017) (0.017) (0.018)
Export * ∆output tariff –0.001 –0.001
(0.001) (0.001)
SOE * ∆output tariff –0.000
(0.001)
continued on next page
294 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Table 9.7 continued

(6) (7) (8) (9) (10)
Variables Change in Number of Products
FDI * ∆output tariff –0.000
(0.001)
Export –0.002
intensity*∆output tariff
(0.003)
Employment 0.010 0.010 0.009
(0.007) (0.007) (0.007)
HHI 0.005 0.005 0.006
(0.049) (0.049) (0.050)
Constant 0.276** 0.279** 0.219* 0.219* 0.227*
(0.124) (0.124) (0.129) (0.129) (0.129)
Observations 35,591 35,591 35,588 35,588 34,430
R-squared 0.002 0.003 0.003 0.003 0.003
Number of firms 7,247 7,247 7,247 7,247 7,196
FDI = foreign direct investment, HHI = Herfindahl–Hirschman Index, SOE = state-owned enterprise.
Note: Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1.
Source: Author’s calculation from the Vietnam Enterprise Survey data.

Third, the change in the output tariff does not have any significant
impact on the extensive margin. One possibility is the increasing
numbers of nontariff measures in Viet Nam (Ing, Cordoba, and Cadot
2016), some of which can be used with protectionist intent. If this is the
case, the rise in nontariff measures can partly offset the impact of tariff
reduction. Although it is desirable to incorporate nontariff measures
in the analysis, distinguishing between protective and nonprotective
measures is not a simple task. I shall leave this issue for further study.
Fourth, exporters produce more products and are more likely to
add products. Coefficients on both export dummy and export intensity
are positive and significant in both specifications. One observation is
that exporters’ extensive margin does not seem to be affected by tariffs.
Coefficients of the interaction term between export status, including
export intensity and export dummy, and output tariff, both in level and
difference, do not show any significance. It is possible that once firms
enter the export market, market diversification reduces the potential
impact of the domestic market’s competition on these firms.
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 295

Fifth, there is no significant impact of ownership on product

churning. This result confirms findings from the previous analysis that
the three groups of firms are not markedly different in terms of product
turnover.
Several implications can be drawn from the regression analysis.
First, my result further confirms the potential positive contribution of
exporters to aggregate growth through product adding. Second, there is
a need to consider other factors of trade policy reform, particularly the
incidence of nontariff measures to capture another important aspect of
trade liberalization. This study suggests that aside from common driving
factors in the literature, the pattern of product turnover is heterogeneous
across countries, depending on the regulatory environment and the
competitiveness of firms, among other factors. This implication calls for
careful country-specific analysis.

9.7 Conclusion
Here I have studied multiproduct firms in Viet Nam. The major findings
are as follows. First, multiproduct firms are larger, more capital-
intensive, more productive, and more likely to export. Second, while the
share of multiproduct firms in Viet Nam is smaller than that found in the
US and India, Viet Nam’s multiproduct firms are active in the market.
Approximately 60% of firms adjust their product scope within a 6-year
period. Third, the contribution of firms’ product extensive margin to
aggregate output growth is limited due to the prevalence of product
dropping, which offsets the positive impact of product adding to output
growth. Most output growth during the period is thus generated by the
intensive margin.
Turning to the link between tariff reduction and product shedding, I
did not detect any significant impact. However, I found the important role
of exporters in product adding, suggesting the potential contribution of
exporters to aggregate growth through the channeling of product scope
expansion. Contrary to expectations, this analysis offers limited support
regarding the heterogeneity of product turnover across ownership types.
While I find that SOEs are more likely to spread economic activities
across products and industries, there are limited difference in terms of
product churning among FDI, SOEs, and the domestic private sector.
The analysis provides several policy implications. First, as product
adding contributes positively to aggregate output growth, firms should
be encouraged to diversify their product range. This could be done
through enhancing innovation, for example, through technology transfer
and the enhancement of interfirm linkages and exports. Diversification
296 Trade Adjustment in Asia: Past Experiences and Lessons Learned

also supports firms in reducing competitive pressure. Second, as multi-

industry and multisector firms account for only 5% and 7% of firm share,
respectively, whereas most product shedding occurs within narrowly
defined categories, it is less likely that product churning or industry
switching can represent a significant source of industrial upgrading
toward more capital-intensive sectors. Therefore, rather than aiming
at expansion across industries, a feasible policy option with respect to
existing firms is to promote investment in process innovation to further
increase the quality of existing products or expansion to closely related
products, through which the intensive margin can be boosted.
Multiproduct Firms, Tariff Liberalization, and Product Churning in Vietnamese Manufacturing 297

References
Amiti, M., and J. Konings. 2007. Trade Liberalization, Imported
Intermediate Inputs, and Productivity: Evidence from Indonesia.
American Economic Review. 97 (5). pp. 1611–38.
Arkolakis, C., and M.A . Muendler. 2010. The Extensive Margin of
Exporting Products: A Firm-level Analysis. NBER Working Paper.
No. w16641. Cambridge, MA: National Bureau of Economic
Research.
Baldwin, J., and W. Gu. 2009. The Impact of Trade on Plant Scale,
Production-Run Length and Diversification. In T. Dunne,
J. Bradford Jensen, and M.J. Roberts, eds. Producer Dynamics: New
Evidence from Micro Data. Chicago, IL: University of Chicago Press.
pp.557–92.
Bas, M. 2012. Input-Trade Liberalization and Firm Export Decisions:
Evidence from Argentina. Journal of Development Economics. 97 (2).
pp. 481–93.
Bernard, A. B., J.B. Jensen, and P. K. Schott. 2006. Survival of the Best
Fit: Exposure to Low-Wage Countries and the (Uneven) Growth
of US Manufacturing Plants. Journal of International Economics.
68 (1). pp. 219–37.
____. 2010. Survival of the Best Fit: Exposure to Low-Wage Countries
and the (Uneven) Growth of US Manufacturing Plants. American
Economic Review. 100 (1). pp. 70–97.
Dang, D. A. 2017. The Effects of Chinese Import Penetration on Firm
Innovation: Evidence from the Vietnamese Manufacturing Sector.
WIDER Working Paper. No. 2017/77. Helsinki: World Institute for
Development Economic Research.
Dhingra, S. 2013. Trading Away Wide Brands for Cheap Brands. American
Economic Review. 103 (6). pp. 2554–84.
Eckel, C., and J. P. Neary. 2010. Multi-product Firms and Flexible
Manufacturing in the Global Economy. The Review of Economic
Studies. 77 (1). pp. 188–217.
Goldberg, P. K., A. K, Khandelwal, N. Pavcnik, and P. Topalova. 2010a.
Multiproduct Firms and Product Turnover in the Developing
World: Evidence from India. The Review of Economics and Statistics.
92 (4). pp. 1042–49.
____. 2010b. Imported Intermediate Inputs and Domestic Product
Growth: Evidence from India. The Quarterly Journal of Economics.
125 (4). pp. 1727–67.
Ha, D. T. T., and K. Kiyota. 2014. Firm-Level Evidence on Productivity
Differentials and Turnover in Vietnamese Manufacturing.
The Japanese Economic Review. 65 (2). pp. 193–217.
298 Trade Adjustment in Asia: Past Experiences and Lessons Learned

Iacovone, L., F. Rauch, and L. A. Winters. 2013. Trade as an Engine

of Creative Destruction: Mexican Experience with Chinese
Competition. Journal of International Economics. 89 (2). pp. 379–92.
Ing, L. Y., S. F. de Cordoba, and O. Cadot. 2016. Non-Tariff Measures
in ASEAN. Jakarta: Economic Research Institute for ASEAN and
East Asia (ERIA). http://www.eria.org/publications/non-tariff
-measures-in-asean/.
Liu, R. 2010. Import Competition and Firm Refocusing. Canadian Journal
of Economics/Revue canadienne d’économique. 43 (2). pp. 440–66.
Lopresti, J. 2016. Multiproduct Firms and Product Scope Adjustment in
Trade. Journal of International Economics. 100 (C). pp. 160–73.
Mayer, T., M. J. Melitz, and G. I. Ottaviano. 2014. Market Size,
Competition, and the Product Mix of Exporters. American Economic
Review. 104 (2). pp. 495–536.
Melitz, M. J. 2003. The Impact of Trade on Intra-industry Reallocations and
Aggregate Industry Productivity. Econometrica. 71 (6). pp. 1695–725.
Nguyen, T. K. 2015. Manufacturing Exports and Employment Generation
in Viet Nam. Southeast Asian Journal of Economics. 3 (2). pp. 1–21.
Nocke, V., and S. Yeaple. 2006. Globalization and Endogenous Firm
Scope. NBER Working Paper. No. w12322. Cambridge, MA: National
Bureau of Economic Research.
Penrose, E. 1955. Limits to the Growth and Size of Firms. The American
Economic Review. 45 (2). pp. 531–43.
Qiu, L. D., and W. Zhou. 2013. Multiproduct Firms and Scope Adjustment
in Globalization. Journal of International Economics. 91 (1). pp. 142–53.
Topalova, P., and A. Khandelwal. 2011. Trade Liberalization and Firm
Productivity: The Case of India. Review of Economics and Statistics.
93 (3). pp. 995–1009.
Trinh, L. Q., and H. T. T. Doan. 2018. Internationalization and the
Growth of Vietnamese Micro, Small, and Medium Sized Enterprises:
Evidence from Panel Quantile Regressions. Journal of Asian
Economics. 55 (C). pp. 71–83.
World Bank. 2017. Viet Nam 2035: Toward Prosperity, Creativity, Equity,
and Democracy. Washington, DC.
 Index
Figures, notes, and tables are indicated by f, n, and t following the
page number.

A Balakrishnan, P., 122, 128

Acemoglu, D., 231, 256 balance-of-payments crises, 8, 236, 247
AFC (Asian financial crisis), 25, 145, Balassa–Samuelson hypothesis, 145n2
147–49 Baldwin, J., 191, 278, 280, 290
Africa, correlations between income and bargaining power, 6, 19–20, 28, 47–49, 123
tariffs in, 67–68, 71, 72f, 78 Bas, M., 27
Aghion, P., 188, 219 Basic Survey of Japanese Business
Ahsan, R., 32, 46, 47, 123 Structure and Activities, 207n1, 215,
Aldaba, R. M., 189, 197–98 215n7
Amiti, M., 25, 29, 40, 45, 188, 207n2, 219, Beladi, H., 123
282 Bentolila, S., 128
Appleton, S., 33 Berman, E., 253nn14–15
Arkolakis, C., 280 Bernard, A. B., 186, 187, 276, 279, 284
Artuc, E., 34 Bernard, A. W., 190, 191
ASEAN (Association of Southeast Asian Berry, R. A., 186
Nations), 189 Bhagwati, J., 36, 49
Asia. See also specific regions and countries bias. See also skill-biased technological
correlations between income and tariffs change
in, 67–69, 70f capital-intensive production, 129, 137
economic growth in. See economic in markup estimations, 23
growth of OLS estimators, 218
financial crisis in, 25, 145, 147–49 in production function estimations, 24
first-time exporters to, 194 in productivity estimations, 23
income and ROW tariff changes in, sample selection, 25
74–75, 75t status quo, 17
international production networks in, Bloom, N., 188–89, 207, 208, 219, 220, 223,
197, 197n1, 201 256
labor mobility costs in, 34 Boldrin, M., 208
poverty in, 1, 49 Bollard, A., 122
trade opening in. See trade Bound, J., 253nn14–15
liberalization Brandt, L., 25–26
Asian financial crisis (AFC), 25, 145, Brazil
147–49 firm-product extensive margin in, 280
Association of Southeast Asian Nations income inequality in, 45
(ASEAN), 189 informal employment in, 32
Aswicahyono, H., 147 labor market adjustments in, 5
Attanasio, O., 45, 68, 262, 263 resource export growth in, 142
automation, 5, 254 trade liberalization in, 68
Autor, D. H., 144, 218, 235 Buckingham, W., 87
Aw, B. Y., 184, 186, 187 Burgess, R., 188
Azam, M., 236–37, 263 Bustos, P., 184

B C
Babu, M. S., 122, 128 Cain, J., 39, 40, 49, 122
Bacchetta, M., 1 Caliendo, L., 231, 233

299
300 Index

Cameron, L., 45 Creusen, H., 188

Canada–US Free Trade Agreement Currie, J., 30
(1989), 277
cannibalization effect, 191, 279 D
capital-intensive production bias, 129, 137 Dang, D. A., 291
Card, D., 235 Das, D. K., 117
Causa, O., 27 Davis, D. R., 29
Centre for Monitoring Indian Economy Debroy, B., 248–49
(CMIE), 238, 274 defensive innovation, 207, 263
Chakrabarti, A., 123 De Loecker, J., 24–26
Chakraborty, Pavel, 8, 229 developed countries. See also specific
Chamarbagwala, R., 236, 237 countries
Chan, K. W., 87 correlations between income and tariffs
Chen, B., 45, 256 in, 69, 70–71f
Chen, X., 184 exporters vs. nonexporters in, 186, 230
China Industrial Productivity (CIP) job displacement in, 141–42
Database, 218, 218n12 labor market polarization in, 5
Choi, Y.-S, 185–86 labor mobility costs in, 34
Chung, S. C., 187 manufacturing sector in, 2
CMIE (Centre for Monitoring Indian developing countries. See also specific
Economy), 238, 274 countries
Colombia comparative advantage of, 230
income inequality in, 45 employment quality in, 18–19, 31
poverty in, 41 exporters vs. nonexporters in, 186
trade liberalization in, 27, 32, 68 imports and, 6, 8
comparative advantage job displacement in, 142, 144, 171
of developing countries, 230 labor market adjustments in, 3–5
gains from trade through, 37 labor share of income in, 46
import competition and, 210, 211, 225 protectionist policies in, 3
in labor-intensive sectors, 120, 137, 210 skilled workers in, 6, 199, 229, 262
specialization and, 49, 82 tax collection in, 10
trade liberalization and, 1–2 trade liberalization in, 17, 179
compensation. See income transport infrastructure in, 200
competition Dhingra, S., 280
in domestic markets, 200 DiNardo, J., 235
firm-level, 9–10, 187–90 displaced workers. See job displacement
globalization and, 187, 189, 190, 279 Dix-Carneiro, R., 93
import. See import competition Doan, Ha Thi Thanh, 9, 276
innovation and, 5, 8, 187–90, 207–8, Draca, M., 188–89, 207, 208, 219, 220, 223
224–25 Du, Y., 92n5
markups and, 279 Dutt, P., 29, 30, 35, 267
patents and, 8, 208, 215, 218–24, 220–24t Dutta, P. V., 236
perfect, 20, 22, 45
productivity and, 26, 28 E
R&D in response to, 189, 225, 263 earnings. See income
skill-biased technological change and, East Asia
5, 263 exporting in, 193
tariffs and, 3, 24–27 globalization in, 189
trade liberalization and, 188 international production networks in,
Couch, K. A., 156 197, 201
Coxhead, Ian, 7–8, 141 labor mobility costs in, 34
Cragg, M. I., 262 poverty in, 1
creative destruction, 186 public works programs in, 36
 Index 301

Eaton, J., 184, 262 Europe and European Union

Ebenstein, A., 235 enlargement of, 4
Eckel, C., 191, 279 first-time exporters to, 194
economic growth innovation in response to Chinese
export-oriented policies for, 1, 86, 179 competition, 207
formality in labor market and, 7–8 job displacement in, 142
income inequality and, 17–18 exports
innovation and, 193 advantages of engaging in, 9
poverty and, 1, 17–18, 36–38, 279 composition by sector, 145, 146f
trade liberalization and, 17–18, 36–37, 82 in economic growth strategies, 1, 86, 179
Economic Research Institute for ASEAN firm-level, 186–87
and East Asia (ERIA), 184–85, 187, 189, import competition and quality of,
193, 195 207n2
education informality and, 32–33
formal employment and, 157–61, 158t, innovation and, 187, 193–95, 200
161–62t intensity of, 127
hukou system and, 86 labor demand and, 30, 48
income and, 236–37, 264–65, 265–66f learning-by-exporting hypothesis, 187,
inequality in, 18 193–95, 200, 280
investment in, 138 managerial demand and, 231, 244–46,
job displacement and, 166, 167t 245–47t, 257
poverty and, 40 in manufacturing sector, 118, 119t, 145,
share of employment by, 264, 264t 186–87
Ekholm, K., 184 poverty and, 69
elasticity of labor demand, 48, 123 product dynamics and, 192, 209–10, 210f
empirical general equilibrium analysis, regional adjustments in, 102–3t, 103–4
41–43 resources, 142, 170
employment. See also formal self-selection hypothesis for, 186–87, 195
employment; labor market world shares of, 209, 209f
adjustments; unemployment
displaced workers. See job displacement F
educational status and share of, 264, Fan, H., 26n2
264t FDI. See foreign direct investment
empirical evidence on, 29–31 Feenstra, R. C., 45, 191, 199, 235, 262–63
globalization and, 49 females. See gender differences
import changes and, 91, 91f Fernandes, A. M., 27, 188
informal. See informal employment Fiess, N., 33
in manufacturing. See manufacturing firm-level adjustments
sector competition and, 9–10, 187–90
productivity and, 28, 130 exports and, 186–87
public works programs, 20, 36, 39, 50 imports and, 26, 45
quality of, 18–19, 31, 154, 156t international production networks,
regional adjustments in, 83, 101–3, 196–201, 197n1
102–3t literature review, 8, 179–80, 181–83t
self-selection model of, 153 markups and, 25–26
skilled. See skilled workers monopoly power and, 19–20, 28, 47, 85
tariffs and, 88–89, 96–100, 97t, 129–30, 131t policy implications of, 199–201
trade shocks and, 82–83, 108, 110 product dynamics and, 190–93, 191f, 276
unskilled. See unskilled workers productivity and, 25, 31–33, 122, 148, 180,
Epelbaum, M., 262 184–86
Erdem, E., 188 tariff levels impacting, 125
ERIA. See Economic Research Institute technological, 180, 184, 193–96
for ASEAN and East Asia flexible manufacturing, 191
302 Index

foreign direct investment (FDI) participation index, 197–98, 197n2, 198f

global value chains and, 1 skill-biased technological change in, 6
innovation and, 193, 195–96 Goldberg, P. K., 24, 32, 36, 41, 45, 68, 89,
liberalization of, 195, 197–99 261–63, 276–78, 280, 283–84, 289, 291
multiproduct firms and, 284, 287, 295 Gorg, H., 263
poverty reduction and, 41 Griliches, Z., 253nn14–15
productivity and, 122 gross domestic product (GDP)
R&D activities and, 196 in econometric models, 73–74
skill-biased technological change and, in hukou analysis, 95, 95n10, 106
199 in premature deindustrialization, 143
formal employment trade-to-GDP ratios, 38, 117
economic growth and, 7–8 well-being of population and, 1
by education level, 157–61, 158t, 161–62t as World Development Indicator, 69
gender differences in, 157–58, 160 Grossman, G. M., 21
job displacement and, 142–44, 159, 167, growth accounting, 22
167t, 171 Gu, W., 191, 278, 280, 290
by occupation and sector, 154, 155t, Gupta, Prachi, 7, 115, 122
157–58, 158t, 160, 161t GVCs. See global value chains
productivity and, 25, 32, 148
quality of, 31, 154, 156t H
trade liberalization and, 9, 31–33 Hadiwidjaja, G., 149
transition to informal employment, 159, Hahn, C. H., 185–86, 192, 194–95
160t Hakobyan, S., 68
Fugazza, M., 33 Hall, Robert, 22
Han, J., 42
G Hanson, G. H., 45, 199, 235, 262–63
Garicano, L., 231 Harmonized System six-digit (HS6)
GDP. See gross domestic product product level, 85, 90, 93, 248
gender differences Harrison, A. E., 22, 27, 30, 188, 263
in formal employment, 157–58, 160 Hasan, R., 3, 30, 35, 39–40, 48–49, 122–23,
in job displacement, 159, 166, 167t 237
in manufacturing sector, 147, 148f, 150, Hayakawa, K., 192–93
150f Heckscher–Ohlin model, 28, 37, 123, 128,
in skill-based income inequality, 236 137, 230
Ghosh, A., 47, 253 Helble, M., 1, 7, 115
Gini coefficient, 19, 38, 43, 171 Helpman, E., 21, 46, 184
Global Attitudes Survey, 1 heterogeneous firm theory, 185–87, 190
globalization Hill, H., 147
competition and, 187, 189, 190, 279 Hinloopen, J., 127
distributional impact of, 3, 201, 229 Hollweg, C. H., 35–36
employment and, 49 Hopenhayn, H. A., 184
Global Attitudes Survey on, 1 Hopman, C., 188
labor share of income and, 46 HS6 (Harmonized System six-digit)
multiproduct firms and, 279 product level, 85, 90, 93, 248
political backlash against, 142n1 hukou (household registration) system,
productivity and, 180, 184–85 81–112
product scope impacted by, 191 data collection on, 92, 94–95
R&D investment and, 196 description of, 85
trade liberalization in promotion of, 179 discrimination on basis of, 86–87
unskilled workers and, 123 effects of abolishment, 105–11, 106–9t,
global value chains (GVCs) 110f
foreign direct investment and, 1 empirical specification and results,
import competition and, 4 95–100
labor market adjustments and, 2, 4, 5 friction measure, 82–83, 87, 95–100
 Index 303

gains from trade and, 104f, 105, 110–11, imports as competitive discipline
110f hypothesis, 188
heterogeneity in granting practices, 87 import substitution, 1, 17, 30, 48, 85, 116
input tariff cuts and, 96–100, 97–98t income
internal geography considerations, by age group, 265–66, 265t, 266f
111–12, 111t correlations with tariffs, 67–71, 70–72f
migration control through, 7, 9, 86, 94n9 econometric model for, 73–74, 73t
reform efforts, 87–88 education and, 236–37, 264–65, 265–66f
rural/urban distinctions, 86 empirical evidence on, 29–31, 44–46
trade shocks and, 82–83, 108, 110 in formal vs. informal employment,
types of registration, 87 18–19
welfare quantifications, 100–105, inequality in. See income inequality
102–3t, 104f job displacement and, 144, 154, 154n6,
Hummels, D., 235 162–69, 163–64t, 168f, 169–70t
Hwang, A. R., 186 labor share of. See labor share of income
of managers. See managerial
I compensation
Iacovone, L., 280 productivity and, 18, 28
Ianchovichina, E., 84 in recovery from trade shocks, 144
ICIO (Inter-country Input–Output regional adjustments in, 78–79, 101–3,
Table), 282–83 102–3t
IFLS. See Indonesian Family Life Survey ROW tariff changes and, 74–78, 75–78t
IMF. See International Monetary Fund of skilled workers, 45, 236
import competition of unskilled workers, 45, 123, 236
comparative advantage and, 210, 211, income inequality
225 economic growth and, 17–18
export quality and, 207n2 empirical evidence on, 43–46, 122
global value chains and, 4 labor share of income and, 46
informality and, 33 measures of, 19, 171
innovation and, 5, 8, 207–8, 224–25 offshoring and, 235
job displacement in, 28 skill-based, 231, 236, 237, 263–64
measurement of, 217–18, 217n10 tariffs and, 8, 43–45
patents and, 8, 208, 218–24, 220–24t Theil index of, 44
poverty and, 39 India
productivity and, 20–21 balance-of-payments crisis in, 8, 236,
skill-biased technological change and, 5 247
by source economies, 211, 211–14t comparative advantage in labor-
tariff cuts and, 104 intensive sectors, 120, 137
imports economic growth and openness in, 116t,
capital goods, 261–62, 261t 117
composition by sector, 145, 146f elasticity of labor demand in, 48
developing countries and, 6, 8 empirical general equilibrium analysis
duty-free policies on, 84, 85 for, 42
employment and changes in, 91, 91f employment in, 3, 25, 29–32, 264, 264t
firm-level, 26, 45 import of capital goods by, 261, 261t
licensing requirements for, 85 income inequality in, 38, 43, 236–37,
managerial demand and, 244–46, 263–64
245–47t labor market flexibility in, 126, 135–36,
in manufacturing sector, 118, 119t, 145 136t, 267
nontariff barriers to, 84–85, 117–18, 236 labor mobility costs in, 34
poverty and, 41 labor share of income in, 46–48, 115,
for processing purposes, 27, 84 120, 120t, 128–29
regional adjustments in, 102–3t, 103–4 managers in. See managerial
tariffs and, 89t, 90–91, 261–62 compensation; managerial demand
304 Index

manufacturing in. See manufacturing trade liberalization and, 188–89

sector input tariffs
markups in, 22–25, 188 employment and, 96–100, 97t, 130, 131t
multiproduct firms in, 276, 278, 280, income inequality and, 45
284 labor market adjustments to, 96–100,
nontariff barriers in, 117–18, 236, 248 97–98t
poverty in, 38–41, 49, 68, 122 labor share of income and, 47, 115,
productivity in, 22–23, 25 128–29, 132–34, 137
tariffs in, 117–18, 117–18f, 125–26, 126t, liberalization of, 100, 116, 137
248, 248f managerial demand and, 249–57,
trade liberalization in, 7–8, 22–25, 37, 250–52t, 255t, 258–59t, 267
115–17, 235–37, 247–49 in manufacturing sector, 125, 126t
WTO dispute with US, 118 markups and, 24–26
Indonesia measurement of, 282, 282n5
displaced workers in. See job poverty and, 40, 41
displacement procompetitive effects of, 24
employment trends in, 147–52, 148f, product churning and, 290–91, 292–94t
150f, 151–52t product dynamics and, 192–93
exports and imports by sector, 145, 146f productivity and, 3, 24–27, 32, 122
firm-level productivity in, 25, 148 trade shocks and reduction in, 92–94
formal vs. informal employment in, 7–9, input trade liberalization, 45, 88, 94–96,
147–48 256
income inequality in, 45, 171 intellectual property. See patents
labor market reforms in, 147n3 Inter-country Input–Output Table
labor mobility costs in, 34 (ICIO), 282–83
manufacturing in. See manufacturing International Monetary Fund (IMF), 17,
sector 22–23, 116, 236, 247–48
poverty in, 40, 41, 149 international production networks
product dynamics in, 192–93 (IPNs), 196–201, 197n1
resource export growth in, 142, 170 Ito, K., 192–94
structural changes in, 142–43, 170
in World Trade Organization, 25 J
Indonesian Family Life Survey (IFLS), Japan
143, 151–52t, 151–58, 159n9, 165, 170 firm-level patent data, 214–17, 215nn7–8,
inequality. See income inequality 216t
informal employment first-time exporters in, 193–94
classification of, 92n5 import competition for, 211, 211–14t,
economic growth and, 7–8 217
productivity and, 25, 31–33, 148 industrial productivity data, 217–18
quality of, 31 Japan Industrial Productivity (JIP)
trade liberalization and, 9, 31–33, 36 Database, 217, 218
transition from formal employment, Jensen, J. B., 186, 187, 276, 279, 284
159, 160t job displacement, 141–71. See also
innovation. See also patents; research and employment
development attrition of workers in sample, 159n8,
competition and, 5, 8, 187–90, 207–8, 161–62, 175t
224–25 characteristics of workers, 166, 167t
defensive, 207, 263 data and estimations for, 151–57
in domestic policy, 171 descriptive statistics on, 157–62, 158t,
economic growth and, 193 160–62t
exporting and, 187, 193–95, 200 education level and, 166, 167t
foreign direct investment and, 193, empirical challenges in study of, 156
195–96 formal employment and, 142–44, 159,
technological, 184 167, 167t, 171
 Index 305

gender differences in, 159, 166, 167t man-days lost and, 126–27
in import competition, 28 in manufacturing sector, 115, 126–37,
income and, 144, 154, 154n6, 162–69, 129–36t
163–64t, 168f, 169–70t productivity and, 129
labor market reforms and, 147n3 tariffs and, 46–48, 115, 128–34, 129–33t,
literature review, 141–45 137
patterns of, 149–50 technology intensity and, 127, 133–34,
self-reported, 165–69, 167t, 168f, 169–70t 134–35t, 137
structural changes related to, 141–44, trade adjustments in, 126, 127
149 Latin America. See also specific countries
trade shocks and, 142–45 correlations between income and tariffs
Jongwanich, J., 195–97, 199 in, 67–68, 71, 71f, 78
Jovanovic, B., 184 income inequality in, 45–46
industrialization model in, 201
K job displacement in, 142
Kamal, F., 47 manufacturing sector in, 2
Kambhampati, U., 29–30 Lawrence, R. Z., 186
Khandelwal, A., 2–3, 23–24, 122, 125, learning-by-exporting hypothesis, 187,
207n2, 219, 261 193–95, 200, 280
Kim, E., 27 Lederman, D., 34
Kis-Katos, K., 40 Levine, D. K., 208
Klenow, P. J., 122 Levinsohn, J., 27, 128
Kohpaiboon, A., 195–97, 199 Levinsohn–Petrin approach, 23, 24, 256
Konings, J., 25, 40, 188, 282 Levy, F., 235
Kortum, S., 184, 262 Liang, Z., 33
Kovak, B., 68, 88, 93 Liu, L., 184
Kramarz, F., 184 Liu, R., 280
Krishna, P., 22–23, 29–30, 44, 46, 49, 122, Lopresti, J., 277, 278, 291
188 Lovely, M. E., 47
Kumar, U., 44, 123, 236
Kuncoro, A., 196 M
Ma, H., 191
L Ma, L., 232n6
labor-demand elasticities, 48, 123 males. See gender differences
labor market. See employment managerial compensation
labor market adjustments descriptive statistics, 239–40t, 241, 242t
challenges in analysis of, 145 firm-level data, 238–42, 274
global value chains and, 2, 4, 5 importer vs. nonimporter, 233, 234f,
input tariff cuts and, 96–100, 97–98t 242, 243f
literature review, 4–6 by industry groups, 240–41, 241f
in manufacturing sector, 2–5, 138 trade value and, 232–33, 233f
mobility costs and, 9, 34–36 variable definitions, 241, 274–75
spatial adjustments, 83 managerial demand
labor share of income capital intensity and, 254, 256
declines in, 120, 120t, 121, 129, 129–30t exports and, 231, 244–46, 245–47t, 257
defined, 120, 128 firm characteristics and, 257–60,
estimation strategies for, 127–28 258–59t
export intensity and, 127 imports and, 244–46, 245–47t
factor intensity and, 115, 127, 130–32, preliminary analysis, 244–46, 245–47t
132–33t, 137 productivity and, 256
globalization and, 46 results summary and policy relevance,
income inequality and, 46 260–66
labor market flexibility and, 126, 135–36, skill intensity and, 246, 246–47t,
136t 251–52t, 253–60, 255t, 258–59t
306 Index

tariff changes and, 249–57, 250–52t, Mehta, A., 237

255t, 258–59t, 267 Melitz, M. J., 180, 184, 186, 187, 191, 279
technology and, 231, 256, 262 men. See gender differences
trade liberalization and, 232–35 Mendoza, A., 79
man-days lost, 126–27 Mexico
manufacturing sector, 115–38 defensive innovation in, 263
delicensing within, 237 export quality in, 262
demographics of employees in, 120, 121t, income inequality in, 45
147–50, 148f, 150f poverty in, 41
displaced workers in. See job trade liberalization in, 30
displacement Midelfart, K., 184
exports in, 118, 119t, 145, 186–87 Mishra, P., 44, 123, 236
factor intensity in, 115, 127, 130–32, Mitra, D., 6, 17, 22–23, 29–30, 32, 35–40,
132–33t, 137 43, 46–49, 122–23, 188
flexible manufacturing in, 191 MNEs. See multinational enterprises
formal vs. informal, 32–33, 143, 147–48 monopoly power, 19–20, 28, 47, 85
gender differences in, 147, 148f, 150, 150f Monte, F., 110
imports in, 118, 119t, 145 most-favored nation status, 93, 96n12
income/sales ratio in, 120–21, 121t Motohashi, K., 215n8
labor market adjustments in, 2–5, 138 Moulton, B. R., 249
labor market flexibility in, 126, 135–36, Muendler, M. A., 280
136t multinational enterprises (MNEs)
labor share of income in, 115, 126–37, firm-specific advantages of, 193
129–36t in international production networks,
multiproduct firms in. See multiproduct 196–97, 199
firms as multiproduct firms, 278, 288
output growth in, 116t, 117 R&D activities of, 195–96
plant-level data, 123–24, 124t, 130, 131t specialization of, 284
premature deindustrialization of, 143 multiproduct firms, 276–96
product dynamics in, 190–93, 191f aggregate output growth, 289–91, 289t,
productivity in, 120–22, 185 295
R&D investment in, 195–96, 238 cannibalization effect and, 279
tariff rates in, 118, 118f, 125–26, 126t, data collection on, 281–83
248, 248f dynamics of product adjustment,
Marchand, Ural, 41–42 286–90, 287–89t
Marjit, S., 123 export status of, 278
markups frequency of product turnover, 286–87,
competition and, 279 287t
estimation of, 21–23 globalization and, 279
in monopoly power measurement, 19 literature review, 279–80
reduction of, 20, 21 ownership types, 278, 284, 287–88, 288t,
surplus shifts and, 18 295
tariffs and, 24–27 policy considerations, 295–96
trade liberalization and, 22–28, 188 product churning by, 9, 283, 286–96,
Marrewijk, C. van, 127 292–94t
Martin, W., 84 productivity of, 280, 285
Matsuura, T., 192–93 product scope in, 190, 191f, 276–77
Matusz, S. J., 35 profile of, 283–85, 283–86t
Mayer, T., 191, 279 sales distribution across products, 285,
McCaig, B., 3, 32–33, 41, 68–69 286t
McLaren, J., 68 tariff changes impacting, 290–95, 292–94t
medium-sized enterprises. See small and variables for analysis, 281–83
medium-sized enterprises Murnane, R. J., 235
 Index 307

N strategic use of, 208, 214, 224

Narjoko, D., 8, 147, 179, 185, 192 usage variables, 215–19, 216t, 227–28t
Nataraj, S., 25, 31–32 Paunov, C., 188
Nayyar, G., 79 Pavcnik, N., 3, 27, 32–33, 36, 41, 45, 68, 89,
Neary, J. P., 191, 279 188, 261–63
Nguyen, N. A., 189–90 Penrose, E., 277
Nguyen, T. K., 279 People’s Republic of China
Nicita, A., 69 comparative advantages for, 210, 211,
North American Free Trade Agreement 225
(NAFTA), 4, 68 competition from, 8, 207–8, 211, 211–14t,
Nouroz, H., 260 217–25, 220–24t
emergence as economic power, 81
O empirical general equilibrium analysis
offshoring, 5, 218, 235 for, 42
Olarreaga, M., 6–7, 67, 69 employment in, 33, 36, 91–92, 91f, 92n5
Olley, G. S., 184 exports in, 187, 209–10f, 209–11
Olley–Pakes approach, 25 firm-level markups in, 25–26
ordinary least squares (OLS) estimators, hukou system in. See hukou system
218, 220, 222–24t import effects in, 89t, 90–91, 91f
Organisation for Economic Co-operation income inequality in, 18, 43, 45
and Development (OECD) integration into world economy, 3, 105
export similarity index, 208 internal migration within, 7, 81, 82,
imports and, 223, 261–62 85–86, 94n9
Inter-country Input–Output Table, labor market adjustments in, 96–100,
282–83 97–98t
technology intensity classification by, labor mobility costs in, 34, 35
127, 133 labor share of income in, 47
Ottaviano, G., 191, 279 local labor markets in, 91–92
output tariffs manufacturing sector in, 192
employment and, 96, 97, 130, 131t nontariff barriers in, 84–85
labor share of income and, 47, 115, poverty in, 18, 38, 42
128–29, 132–34, 137 tariffs in, 88–91, 88f, 89t, 96–100, 97–98t
liberalization of, 116, 137 total factor productivity in, 3, 25–27
managerial demand and, 249, 250t trade liberalization in, 25–27, 37, 84–85
in manufacturing sector, 125, 126t in World Trade Organization, 4, 25, 82,
markups and, 24–26 84–85, 88
poverty and, 40, 41 perfect competition, 20, 22, 45
procompetitive effects of, 3, 24–27 Petrin, A., 128
product churning and, 290–94, Pew Research Center, 1
292–94t Philippines
productivity and, 3, 24–27, 31–32, 122 competition and innovation in, 189
outsourcing, 32, 199, 236, 262–63 electronics industry in, 197–99
labor mobility costs in, 34, 35
P Piermartini, R., 6–7, 67, 79
Pakes, A., 184 Placzek, D. W., 156
Palangkaraya, A., 195 Poole, J. P., 46
Park, A., 92n5 Porto, G., 6–7, 34, 41, 43, 67, 69
Park, C.-G., 194–95 poverty
patents economic growth and, 1, 17–18, 36–38, 279
competition and, 8, 208, 215, 218–24, education level and, 40
220–24t exports and, 69
depreciation rate of, 216–17 foreign direct investment and, 41
firm-level data, 214–17, 215nn7–8, 216t imports and, 41
308 Index

service sector growth and, 149 competition and, 189, 225, 263
tariffs and, 39–41, 68 costs of, 185
trade liberalization and, 17–19, 39–43, exporting and, 9, 187, 193, 194
49, 122 international production networks and,
unskilled workers and, 37 200
premature deindustrialization, 143 investment in, 21, 193, 195–97
product churning, 9, 283, 286–96, 292–94t in manufacturing sector, 195–96, 238
production sharing. See outsourcing in pharmaceutical industry, 214n5
productivity. See also total factor spillovers of, 200
productivity rest-of-world (ROW) tariffs, 67–79
employment and, 28, 130 data analysis, 69–71, 70–72f
estimation of, 21–23 econometric model and results, 72–74,
of exporters vs. nonexporters, 186, 187 73t
of firms, 25, 31–33, 122, 148, 180, 184–86 factors influencing effects of, 78–79
foreign direct investment and, 122 income and changes in, 74–78, 75–78t
globalization and, 180, 184–85 literature review, 68–69
import competition and, 20–21 overview, 67–68
income and, 18, 28 Ricardo, D., 37
labor share of income and, 129 Roberts, M. J., 184, 187
managerial demand and, 256 Rodrik, D., 48
in manufacturing sector, 120–22, 185 Rojas-Romagosa, H., 188
of multiproduct firms, 280, 285 Rossi-Hansberg, E., 110, 231, 233
procompetitive effects of, 26, 28 ROW tariffs. See rest-of-world tariffs
relationship with product scope, 190, Roy, A. D., 153
191f Ruzic, D., 232n6
tariffs and, 3, 24–27, 31–32, 122
technological upgrading and, 180, 184 S
trade liberalization and, 18, 22–28, 122, Sadun, R., 256
180, 184–88 Saint-Paul, G., 128
product switching, 186 salary. See income
protectionism, 3, 20, 116, 294 Santhanam, A. T., 249
PROWESS database, 238, 274 school. See education
public works programs, 20, 36, 39, 50 Schott, P. K., 190, 191, 208, 210–11, 276,
Pushpangadan, K., 122, 128 279, 284
self-selection hypothesis for exports,
Q 186–87, 195
Qiu, L., 192, 280 self-selection model of employment, 153
Senses, M., 46
R Seshan, G., 43
Ramaswamy, K. V., 48 Sethupathy, G., 44, 122
R&D. See research and development SEZs (special economic zones), 84, 85
Ranjan, P., 35, 36 Sharma, G., 122, 237
Rauch, F., 280 Shin, J., 30, 47, 48
Ravenga, A., 30 Shrestha, Rashesh, 7–8, 141
Redding, S. J., 110, 190, 191 Sivadasan, J., 122
Republic of Korea Sjoholm, F., 186
elasticity of labor demand in, 48 skill-biased technological change
employment in, 30 competition and, 5, 263
exporting in, 187, 194–95 foreign direct investment and, 199
labor mobility costs in, 34 in global value chains, 6
labor share of income in, 47 income inequality and, 231
total factor productivity in, 27, 185–86 trade liberalization and, 20, 235
research and development (R&D). See skilled workers. See also managerial
also innovation compensation; managerial demand
 Index 309

definitions of, 253n14 output. See output tariffs

demand for, 6, 9, 37, 199, 231 poverty and, 39–41, 68
in hukou system, 87 preliberalization levels and changes in,
imported inputs as substitutes for, 45 88f, 89–90
income of, 45, 236 product churning and, 290–95, 292–94t
industrial delicensing and, 237 product dynamics and, 192–93
labor market polarization and, 5 productivity and, 3, 24–27, 31–32, 122
Stolper–Samuelson theorem on, 229–30 ROW. See rest-of-world tariffs
trade liberalization and, 229–30, 262 unemployment rates and, 3, 30, 122–23
Slaughter, M. J., 48 taxation, 10, 31, 39, 79, 241, 281
small and medium-sized enterprises technology. See also skill-biased
(SMEs), 189, 190, 195, 201, 278 technological change
social protection policies, 17–18, 20, 36, in firm-level adjustments, 180, 184,
50, 267 193–96
Solow, Robert, 22 intensity and labor share of income, 127,
Song, L., 33 133–34, 134–35t, 137
South Korea. See Republic of Korea of international production networks,
Sparrow, R., 40 197
special economic zones (SEZs), 84, 85 managerial demand and, 231, 256, 262
state-owned enterprises (SOEs), 278, 281, Markov process for, 23
284, 287–88, 295 spillovers, 122, 193
status quo bias, 17 upgrading, 5, 180, 184
Stolper–Samuelson theorem, 21, 28, Thangavelu, S., 199
229–30, 235 Theil index of income inequality, 44
Strobl, G., 263 Thoenig, M., 207
Sumarto, S., 149 Topalova, P., 2–3, 23–24, 39, 40, 68, 122,
Sundaram, A., 32, 49 125
Suryadarma, D., 149 total factor productivity (TFP)
Suryahadi, A., 149 employment and, 130
estimation of, 22
T tariffs and, 3, 26–27
Takii, S., 186, 192–93 trade liberalization and, 25–27, 185
tariffs trade adjustments
average and weighted average, 117, 117f costs of, 19, 34–36, 123
correlations with income, 67–71, 70–72f firm-level. See firm-level adjustments
employment and, 88–89, 96–100, 97t, heterogeneity of, 82
129–30, 131t labor market. See labor market
HS6 product level, 85, 90, 93, 248 adjustments
hukou abolishment and impact on, in labor share of income, 126, 127
108–9t, 108–11, 110f policy responses to, 35–36
imports and, 89t, 90–91, 261–62 regional, 101–3, 102–3t
income inequality and, 8, 43–45 Trade Analysis Information System
input. See input tariffs (TRAINS) database, 93, 125, 248
internal geography and gains from trade liberalization
reduction of, 111–12, 111t adjustments to. See trade adjustments
labor share of income and, 46–48, 115, comparative advantage and, 1–2
128–34, 129–33t, 137 competition and, 188
managerial demand and, 249–57, as conditionality for IMF loans, 22–23
250–52t, 255t, 258–59t, 267 distributional impact of, 3, 10, 37, 229
in manufacturing sector, 118, 118f, economic growth and, 17–18, 36–37, 82
125–26, 126t, 248, 248f empirical general equilibrium analysis
markups and, 24–27 of, 41–43
most-favored nation status, 93, 96n12 employment and. See employment
objectives of, 6, 68 exogeneity of, 88–90
310 Index

exports and. See exports exporters vs. nonexporters in, 186, 187
globalization promotion through, 179 job displacement in, 142
hukou system and. See hukou system labor market polarization in, 5
imports and. See imports labor mobility costs in, 34
income and. See income; income manufacturing sector in, 2, 3
inequality multiproduct firms in, 276, 284
innovation and, 188–89 WTO dispute with India, 118
input, 45, 88, 94–96, 256 unskilled workers
labor-demand elasticities and, 48, 123 demand for, 37
labor share and. See labor share of globalization and, 123
income in hukou system, 87
managerial demand and. See managerial imported inputs as complements for, 45
demand income of, 45, 123, 236
markups and, 22–28, 188 poverty and, 37
motivations for, 17, 179 Stolper–Samuelson theorem on, 229–30
multiproduct firms and. See trade liberalization and, 229–30
multiproduct firms Ural, B. P., 32, 39, 40
poverty and, 17–19, 39–43, 49, 122 Urata, S., 8, 179
productivity and, 18, 22–28, 122, 180,
184–88 V
regional adjustments to, 101–3, 102–3t value chains. See global value chains
shocks from. See trade shocks Van Reenen, J., 188–89, 207, 208, 219, 220,
skill-biased technological change and, 223, 256
20, 235 vector autoregressive (VAR) method,
summary of responses to, 57–64 194–95
tariffs and. See tariffs Verdier, T., 207
trade shocks Verhoogen, E., 262
adjustment process following, 19, 34–35 Viet Nam
domestic-oriented firms and, 278 competition and innovation in, 189–90
employment and, 82–83, 108, 110 empirical general equilibrium analysis
hukou system and, 82–83, 108, 110 for, 43
income level and recovery from, 144 international production networks in,
input tariff reductions and, 92–94 199
internal migration in reaction to, 82 labor market adjustments in, 5
job displacement following, 142–45 manufacturing in. See manufacturing
policy responses to, 35–36 sector
WTO accession and, 100 multiproduct firms in. See multiproduct
TRAINS (Trade Analysis Information firms
System) database, 93, 125, 248 nontariff barriers in, 294
Trefler, D., 253n14 poverty in, 41, 43, 68–69, 279
Tybout, J. R., 184, 188 product churning in, 9, 283, 286–96,
292–94t
U productivity in, 185, 279
unemployment. See also employment trade liberalization in, 185
frictional, 165 Vietnam Enterprise Survey (VES), 277, 281
multisectoral search model of, 35 W
social protection policies for, 36, 50 wages. See income
tariffs and, 3, 30, 122–23 Winters, L. A., 280
trade liberalization and, 30, 35 women. See gender differences
United States Wood, A., 207, 263
Canada–US Free Trade Agreement, 277 workers. See employment
correlations between income and tariffs World Bank, 116
in, 68 World Integrated Trade Solutions
elasticity of labor demand in, 48 (WITS) database, 248, 282, 283
 Index 311

World Population Policies, 83, 112 Y

World Trade Organization (WTO) Yamashita, N., 8, 207
Integrated Data Base of, 125 Yamauchi, I., 8, 207
labor market adjustments and, 4 Yeaple, S., 184
preconditions for accession to, 84–85 Yu, M., 3, 26–27, 45
pre-WTO tariff changes following Yu, Z., 45
accession, 88f, 89–90
trade liberalization and, 17, 25, 82 Z
trade shocks and accession to, 100 Zhou, W., 192, 280
US-India dispute, 118 Zhu, S. C., 253n14
Wu, Y., 92n5 Zi, Y., 7, 81–83
 Trade Adjustment in Asia
Past Experiences and Lessons Learned

Asia’s economic success over the past four decades has been built on a strategy of
export promotion coupled with trade opening. Poverty rates throughout the region
have fallen dramatically, especially in countries that have succeeded in integrating
into regional or global value chains. However, this economic success has been
accompanied by structural changes, such as the need for workers to change roles,
sectors, and sometimes regions. Faced with increased foreign competition, firms
have been forced to reorganize and quickly adopt new technologies.
Despite the importance of this adjustment process, relatively little empirical evidence
exists. This volume aims to close this gap by providing new insights into how Asia’s
labor markets and firms have adjusted to trade opening. Written by leading trade
economists with expertise in the region, the publication shows that trade opening has
led to a more efficient allocation of capital and labor, but this has been accompanied
by significant adjustment costs. The book sheds light on the effects of trade on
workers and firms, with the aim of improving understanding of the adjustment
process and contributing to the debate on how to make globalization work for all.

About the Asian Development Bank Institute

ADB Institute, located in Tokyo, is the think tank of the Asian Development Bank, an
international financial institution. ADBI aims to be an innovative center of excellence
for the creation of rigorous, evidence-based knowledge that can be implemented as
new actionable policies by developing and emerging economies, so as to contribute
to achieving a prosperous, inclusive, resilient, and sustainable Asia and the Pacific.

About the World Trade Organization

The World Trade Organization (WTO), located in Geneva, deals with the global
rules of trade between nations. At its heart are the WTO agreements, negotiated
and signed by the bulk of the world’s trading nations and ratified in their
parliaments. The goal of the WTO is to help the producers of goods and services,
exporters, and importers conduct their business as smoothly and freely as possible.

Marc Bacchetta is a Chief of Section in the Economic Research and Statistics

Division of the World Trade Organization in Geneva.
Matthias Helble is an Economist in the Economic Research and Regional
Cooperation Department of the Asian Development Bank in Manila, and an ADBI
Adjunct Fellow at the Asian Development Bank Institute in Tokyo.

Asian Development Bank Institute World Trade Organization

3-2-5 Kasumigaseki Centre William Rappard
Chiyoda-ku Rue de Lausanne 154
Tokyo, 100-6008 Japan CH-1211 Geneva 21, Switzerland
Tel +81 3 3593 5500 Tel +41 22 739 5111
www.adbi.org www.wto.org