Trade liberalization, forward‐looking firms, and welfare

• Date: 01 November 2017
• Published date: 01 November 2017
• Author: Kuo‐Feng Kao,Cheng‐Hau Peng
• DOI: http://doi.org/10.1111/roie.12293

ORIGINAL ARTICLE

Trade liberalization, forward-looking firms, and

welfare

Kuo-Feng Kao

1

|

Cheng-Hau Peng

2

1

Tamkang University, New Taipei City,

Taiwan

2

Fu Jen Catholic University, New Taipei

City, Taiwan

Correspondence

Cheng-Hau Peng, Department of Eco-

nomics, Fu Jen Catholic University, No.

510, Zhongzheng Rd, Xinzhuang Dis-

trict, New Taipei City 24205, Taiwan.

Email: chpon@mail.fju.edu.tw

JEL Classification: F12, F13

Abstract

We set up an oligopolistic model with two exporting firms

selling to a third market to investigate the welfare implica-

tions of trade liberalization when the exporting firms are

forward-looking. The results show that with cost asymmetry

trade liberalization encourages the exporting firms to engage

in tacit collusion, which may not only be detrimental to the

domestic welfare, but also to the consumer surplus of the

importing country. Moreover, we find that tacit collusion is

less sustainable if the government of the importing country

imposes a lower (higher) tariff on the more (less) efficient

exporting firm. If a nonforward-looking or a forward-

looking cost-efficient domestic firm exists in the importing

country, then trade liberalization also encourages tacit

collusion.

1

|

INTRODUCTION

Tariff protection has always been criticized as a policy that makes domestic consumers worse off by

forcing them to pay higher prices for imports. Therefore, if firms behave noncollusively, it is com-

monly believed that trade liberalization will favor consumers in importing countries owing to the more

intense rivalry between firms. This pro-competition effect may well characterize the reality in some

industries. In other industries, firms may have incentives to engage in tacit collusion.

In their empirical survey, Blume, Strand, and Färnstrand (2002) find that trade liberalization indu-

ces tacit collusion among exporting firms in European Union (E.U.) sugar markets. Bond and Syropou-

los (2008) also suggest a possible linkage between trade liberalization and international cartels. In

addition, the European Commission has reported cartel statistics from 1990 to 2014 in the European

Union.

1

We summarize these statistics in Table 1. The table shows that the number of cartel cases

increased significantly since 2000. Nevertheless, data compiled by the World Bank show that the

weighted mean applied tariff of manufactured products in the European Union was around 6 percent in

1990 but later dropped to 2 percent in 2000 and was steadily around 2 percent after 2000.

2

Although

Rev Int Econ. 2017;25:999–1016. wileyonlinelibrary.com/journal/roie V

C2017 JohnWiley & Sons Ltd

|

999

Received: 10 January 2014

|

Revised: 2 December 2016

|

Accepted: 10 March 2017

DOI: 10.1111/roie.12293

this provides no direct evidence of a causal relationship between trade liberalization and tacit collusion,

it seems to be appropriate to investigate the possibility of such a relationship.

We also observe that some exporting firms have fixed or raised their export prices despite the

fact that trade costs have been gradually decreasing around the world. One example is the dynamic

random access memo ry (DRAM) industry i n the European Unio n. Micron was the firs t to squeal

on the rest of the cartel. In 2010, the European Commission fined 10 chipmakers (including Nanya

and Samsung) a total of e331.3 million for keeping the prices of DRAM artificially high between

1998 and 2002 by sharing “secret information”among themselves. A second example is that

involving Unilever and P&G who in 2011 were fined a total of e315.2 million by the European

Commission for agreeing with Henkel KGaA, the German maker of Persil, to fix the prices of the

detergent in eight countries over a 3-year period.

3

These observations motivate us to investigate

how an importing co untry’s trade liberalization affects the incentives of exporting firms to collude

when selling to that market.

In this paper, we employ an oligopolistic setting with two exporting firms selling to an importing

country that lowers its tariffs.

4

The results show that trade liberalization encourages the forward-

looking exporting firms to engage in tacit collusion. This tacit collusion may raise the market price in

the third country, which is pernicious to its consumers and welfare. Such a result is robust regardless

of the competition mode of the firms, the existence of a domestic competitor, and whether the import-

ing country adopts the most favored nation clause (MFN, i.e., a uniform tariff) or discriminatory

tariffs.

This paper is closely related to the literature on the nexus between trade policies and the collusive

behavior of firms in repeated games. Most studies in this strand of the literature consider an intra-

industry model with firms selling a homogeneous product based on the framework of Brander and

Krugman (1983).

5

Examples are Pinto (1986), Schultz (2002), Haufler and Schjelderup (2004), and

Bond and Syropoulos (2008), among others. Pinto (1986) extends Brander and Krugman (1983) by

considering repeated interactions among quantity-setting firms in both their domestic and export mar-

kets. He finds that if trade costs are negligible, no exports result in a strong Nash equilibrium for the

two colluding firms. Schultz (2002) investigates the incentives for identical domestic firms to collude

in an export cartel. He shows that the formation of an export cartel improves the monitoring of collu-

sion, thereby shortening the time needed to detect the deviation. In addition, when the production tech-

nology of firms exhibits constant returns to scale, this will lead to firms that deviate being subject to

greater punishment and result in the collusion becoming more stable. Haufler and Schjelderup (2004)

analyze the effects of international commodity tax policies on the stability of collusive agreements

between firms, and also find that reducing the tax stabilizes the collusion among firms. Bond and Syro-

poulos (2008) show that the relationship between collusive conduct and trade costs may be nonmono-

tonic depending on the magnitude of the trade costs. Furthermore, if trade costs are below the

threshold level, trade liberalization may reduce total output and lower welfare.

In a departure from the above-mentioned papers, we consider a three-country exporting model sim-

ilar to the models in Brander and Spencer (1985), Gatsios (1990) and Hwang and Mai (1991), and

assume that exporting firms produce differentiated products with different marginal costs. Brander and

TABLE 1 Cartel statistics in the European Union

1990–1994 1995–1999 2000–2004 2005–2009 2010–2014

No. of cases 10 10 30 33 30

Amount of fines (e) 344,282,550 270,963,500 3,157,348,710 7,920,497,226 7,613,104,579

1000

|

KAO AND PENG