Heterogeneous effects of trade liberalisation on firm‐level markups: Evidence from China

• Author: Feixiang Chen,Xunyong Xiang,Chun‐Yu Ho,Wen Yue
• Published date: 01 August 2017
• Date: 01 August 2017
• DOI: http://doi.org/10.1111/twec.12516

SPECIAL ISSUE ARTICLE

Heterogeneous effects of trade liberalisation on

firm-level markups: Evidence from China

Xunyong Xiang

1

|

Feixiang Chen

2

|

Chun-Yu Ho

2,3

|

Wen Yue

4

1

Institute of Industrial Economics, Jinan University, Guangzhou, China

2

Antai College of Economics and Management, Shanghai Jiao Tong University, Shanghai, China

3

Department of Economics, University at Albany, State University of New York, Albany, NY, USA

4

School of Business, Jiangnan University, Jiangsu, China

1

|

INTRODUCTION

Over the past few decades, trade liberalisation has become an important part of the development

strategies of many countries to achieve more efficient allocation of resources. Theoretical works

have shown that trade liberalisation affects firm markups (Arkolakis, Costinot, Donaldson, &

Rodr

ıguez-Clare, 2015; Bernard, Eaton, Jensen, & Kortum, 2003; Edmond, Midrigan, & Xu, 2015;

Feenstra & Weinstein, 2010; Mayer, Melitz, & Ottaviano, 2014; Melitz & Ottaviano, 2008).

Tougher import competition induced by the reduction in output tariffs exerts downward pressure

on firm prices (which is known as the pro-competition effect); additionally, access to more vari-

eties of intermediate inputs and higher quality inputs induced by the reduction in input tariffs helps

firms to reduce costs (which is known as the cost-reduction effect). Therefore, it is important to

account for both channels to understand the overall impact of trade liberalisation on firm markups.

Our empirical analysis is based on the information covered in firm-level production data, highly

disaggregated transaction-level product trade data and tariff data at the eight-digit Harmonised Sys-

tem (HS) level for China from 2000 to 2006. To explore the relationship between firm markups

and output and input tariffs, we follow a three-step empirical procedure. First, we estimate firm-

level markups following De Loecker and Warzynski (2012) for firms across 28 two-digit manufac-

turing industries with firm-level production data. Second, we construct firm-level output tariffs and

input tariffs with transaction-level product trade data and tariff data. Third, we examine how firm

markups respond to reductions of firm-level output and input tariffs, which represent distinct

shocks to domestic firms.

Our results show that firm markups decrease as output tariffs decrease, which is consistent with

the pro-competition effects of reducing output tariffs. Firm markups increase as input tariffs

decrease, which is consistent with the cost-reduction effect of reducing input tariffs. We further

examine the heterogeneity of the markup effect of trade liberalisation. First, we find that the cost-

reduction effect of input tariffs on firm markups is weaker for industries with a higher industry

concentration because competition from new firm entry induced by a lower input tariff offsets the

cost-reduction effect of input tariff reduction on markup. Second, we find that both output tariffs

DOI: 10.1111/twec.12516

World Econ. 2017;40:1667–1686. wileyonlinelibrary.com/journal/twec ©2017 John Wiley & Sons Ltd

|

1667

and input tariffs have weaker effects on processing trade firms and state-owned enterprises (SOEs).

Finally, our results are robust to the use of an instrumental variable approach to control for the

endogeneity of tariffs and sample selection.

This paper contributes to two strands of literature. First, previous studies empirically explor e

the relationship between firm markups and import competition. Some papers report reduced

markups when firms are exposed to more import competition (Harrison, 1994; Konings, Van

Cayseele, & Warzynski, 2005; Konings & Vandenbussche, 2005; Krishna & Mitra, 1998; Levin-

sohn, 1993). Nonetheless, previous studies usually use import penetration but not trade tariffs as

a proxy for trade liberalisation. Recently, De Loecker, Goldberg, Khandelwal, and Pavcnik

(2016) found that output tariff declines lead to more import competition and thus lower mark-

ups, and input tariff declines lead to a reduction in marginal cost and thus higher markups. Our

study contributes to this literature in several aspects. First, our paper fills a gap in the literature

by examining how trade liberalisation affects firm markups in China, which has the second-lar-

gest economy and the largest trade volume in the world. Second, to the best of our knowledge,

our study is the first to show that the decline of input tariffs not only has a cost-reduction effect

but also a competition effect by inducing more firms to enter the market. In sum, the effect of

input tariffs on markups is mixed. Finally, we show that the effect of trade liberalisation on

firm-level markups depends on industry concentration, trade regime (processing versus ordinary

trade) and firm ownership.

In a closer relationship to our work, two works are developing concurrently with ours.

1

Fan

et al. (2015) show that input trade liberalisation increases firm-product-level markups. Our work

differs from theirs in focusing on the effects of input and output trade liberalisation on firm-

level markups and exploring the effects of industry concentration and firm ownership on such

a relationship. Liu and Ma (2015) show that there is a positive effect of output tariffs on

firm-level markup, which is consistent with our results. Their focus is to show that input tariffs

only affect firm-level markups for importers and that the effect is stronger for less-competitive

industries. Our work differs from theirs in showing that: (i) input tariff reduction induces more

firm entry in more-concentrated industries, which reduces the cost-reduction effect of input tar-

iff reduction; and (ii) there are heterogeneous effects of trade liberalisation across firm owner-

ship types.

Second, our study contributes to the growing literature that examines the effect of trade liberali-

sation on firm-level productivity in developing countries. Examples include Schor (2004) for Bra-

zil, Amiti and Konings (2007) for Indonesia, Nataraj (2011) and Topalova and Khandelwal (2011)

for India, and Ge, Lai, and Zhu (2011), Hu and Liu (2014) and Yu (2015) for China. However,

because of data limitations, the previous works employ revenue productivity deflated by industry-

specific deflators. Thus, the productivity measure is likely to capture both technical efficiency as

well as markups (De Loecker, 2011; De Loecker & Warzynski, 2012; Katayama, Lu, & Tybout,

2009). Our work extends the literature in showing the importance of markups as a part of the rise

in revenue productivity during trade liberalisation.

The remainder of this paper is organised as follows. Section 2 outlines our empirical models.

Section 3 presents our data and descriptive statistics. Section 4 presents the empirical results. Sec-

tion 5 examines the heterogeneity in our results. Section 6 concludes and discusses policy implica-

tions.

1

Since we began our work in 2014, Fan, Li, and Luong (2015) has been developing a concurrent study. We found Liu and

Ma (2015) joining the literature during our revision for a second revision request from the journal in 2016.

1668

|

XIANG ET AL.