Foreign direct investment productivity premium and foreign affiliates' heterogeneity: A comparison between advanced and emerging market overseas investments in the EU

• Date: 01 October 2019
• Published date: 01 October 2019
• DOI: http://doi.org/10.1111/twec.12837
• Author: Filippo Reganati,Rosanna Pittiglio

3030

|

wileyonlinelibrary.com/journal/twec World Econ. 2019;42:3030–3064.

© 2019 John Wiley & Sons Ltd

Received: 15 February 2018

|

Revised: 6 June 2019

|

Accepted: 11 June 2019

DOI: 10.1111/twec.12837

ORIGINAL ARTICLE

Foreign direct investment productivity premium

and foreign affiliates' heterogeneity: A comparison

between advanced and emerging market overseas

investments in the EU

RosannaPittiglio1

|

FilippoReganati2

1University of Campania ”Luigi Vanvitelli”, Caserta, Italy

2Sapienza University of Rome, Rome, Italy

KEYWORDS

advanced market multinationals, emerging market multinationals, productivity differentials, total factor productivity, type of

foreign direct investment

1

|

INTRODUCTION

The theory on multinational enterprises (MNEs) strongly suggests that foreign‐owned firms (FOFs) are

more productive than their domestic counterparts. The reason being that MNEs are presumed to pos-

sess firm‐specific intangible assets (i.e., superior production technology, intellectual property such as

patents or brands, or internal organisational superiority), which are transferred to their affiliates at lower

marginal cost due to their public good character. Such an allegedly superior performance of FOFs has

been widely documented in the empirical research (Aitken & Harrison, 1999; Arnold & Javorcik, 2009;

Bertrand & Zitouna, 2006; Caves, 1996; Conyon, Girma, Thompson, & Wright, 2001, 2002; Doms &

Jensen, 1998; Girma, 2005; Girma & Görg, 2007a, 2007b; Harris & Robinson, 2003; Swenson, 1993).

For example, in the Swedish manufacturing sector, Karpaty (2007) estimates that the productivity ad-

vantage of FOFs over domestically owned firms is from 2% to 7%. Arndt and Mattes (2010) find that

FOFs have a productivity advantage of about 6% over domestic multinationals in Germany.

However, it should be noted that the overwhelming majority of empirical studies in the literature

analyse the effects of foreign direct investment (FDI) on productivity by comparing domestically and

foreign‐owned firms combined in a single “foreign ownership” variable, without paying attention to

the country source of FDI. Instead of focusing on the existence of productivity heterogeneity between

groups of firms, some recent studies have analysed the differences in productivity within the group of

FOFs.1 In particular, Benfratello and Sembenelli (2006) find that US firms in Italy are not only sig-

nificantly more productive than their domestic counterparts but also more productive than other FOFs.

1 In the literature, scholars have mainly focused on heterogeneity between groups by internationalization mode: domestic

firms versus exporters versus firms involved in inward and outward FDI.

|

3031

PITTIGLIO and REGanaTI

Using German data, Arndt and Spies (2012) find that affiliates from geographically close countries

with a high level of freedom to trade and good availability of credit outperform their competitors from

countries that are further away or countries with poorer institutions. For the US, Chen (2011) shows

that firms owned by advanced market multinationals (AFOFs) experience an increase in labour pro-

ductivity of up to 13% compared with domestically owned firms, while firms owned by emerging

market multinational enterprises (EFOFs) experience 23% lower labour productivity gains than do-

mestically owned firms. By breaking down all FOFs according to the origin country (US, European

countries and other extra‐European countries), Weche Gelübcke (2013) finds that within the popula-

tion of foreign firms, US affiliates in Germany stand out as having the highest productivity.

Given the upsurge of investments by emerging market multinationals (EMNEs) in advanced mar-

kets, the productivity‐improving role of FDI on FOFs is particularly relevant.2 In fact, some peculiar-

ities of EMNEs might call into question the theory regarding the inevitable transfer of specific

advantages for MNE affiliates. Several scholars (Li, Li, & Shapiro, 2012; Luo & Tung, 2007) have

stressed that EMNEs, instead of utilising capabilities already on hand, expand overseas in search of

capabilities that are not available in their home markets. In such a case, the transfer of technological

assets across borders from the parent might be irrelevant.

The activities of EMNEs have attracted the attention of many scholars, who were mainly inter-

ested in understanding their determinants for internationalisation (Blomkvist & Drogendijk, 2016;

Buckley, Clegg, Cross, Liu, Voss, & Zheng, 2007; Giuliani, Gorgoni, Guenther, & Rabellotti, 2013;

Kolstad & Wiig, 2012), and whether existing theories were able to explain their investment patterns

(Cuervo‐Cazurra, 2012; Mathews, 2006; Ramamurti, 2009). Little research, however, has focused on

the productivity differences between EFOFs and AFOFs located in advanced countries.3 To the best

of our knowledge, only two studies (Clegg & Voss, 2012; Sanfilippo, 2015) have extended this anal-

ysis to emerging countries. By analysing Chinese overseas FDI in the EU, Clegg and Voss (2012) find

that Chinese foreign‐owned firms are not so different in terms of productivity from those of leading

third‐country investors, such as the US and Japan. Focusing on a sample of foreign affiliates and

domestic MNEs in Europe, Sanfilippo (2015) finds not only the existence of a productivity gap be-

tween EMNEs' affiliates and others, but also a significantly larger difference in more sophisticated

industries, in both the services and manufacturing.

The main purpose of this paper is, therefore, to investigate the existence of productivity heteroge-

neity among FOFs based in the EU and, in particular, to analyse to what extent EMNEs' FDI differs

from that of advanced market multinationals (AMNEs). More specifically, the research questions

we want to answer are the following: Do foreign affiliates from emerging countries have a different

productivity level to foreign affiliates from advanced countries? And, if so, to what extent is this pro-

ductivity gap related to the type of investment (i.e. horizontal vs. vertical FDI)?

Finding an answer to these questions is particularly important for both researchers and policymak-

ers. From the theoretical point of view if, on the one hand, wide‐ranging literature has emphasised the

productivity‐improving role of FDI on foreign affiliates, on the other, the issue of the link between

motivation and the productivity effects of FDI has so far received scant attention. From a policy per-

spective, being aware of the existence of productivity heterogeneity within foreign affiliates seems to

be extremely important when seeking an efficient and effective means of conducting FDI attraction

policies.

2 In 2013, outward FDI (OFDI) from EMNEs accounted for 39% of global OFDI, compared with only 12% at the beginning of

the 2000s (UNCTAD, 2013).

3 Some studies (Buckley et al., 2014; Chari et al., 2012) have analysed the effects of AMNEs' acquisitions on the performance

of target firms in developed countries.

3032

|

PITTIGLIO and REGanaTI

In order to address these issues, we have used firm‐level data for a sample of FOFs investing in

the EU over the period 2006–14. The empirical analysis reported in the paper is first carried out using

both unconditional means and Kolmogorov–Smirnov tests of stochastic dominance among the two

groups of firms, and then regression analyses to examine whether the productivity gaps are indepen-

dent of any structural and firm‐specific characteristics.

By distinguishing between different types of FDI, our main contribution to the existing literature lies

in demonstrating the importance of the nature and motivation of foreign direct investment in affecting

the performance of FOFs and, in particular, in determining the productivity gaps between EFOFs and

AFOFs. We believe that the issue of whether performance differences exist within the group of FOFs is

extremely relevant for policymakers, who have to ascertain the country from which it would be better to

attract FDI. Indeed, by emphasising the benefits arising from inward FDI for the host countries, there has

been an overall tendency towards implementing liberalised policies to attract FDI. However, the growing

flows of outward FDI (OFDI) from emerging market countries might trigger some changes in the policy

stances of advanced countries. This point could be particularly relevant when FDI from emerging coun-

tries is aimed to acquire knowledge capabilities that could be moved to their home countries.

The remainder of the paper is organised as follows. In Section 2, we discuss the theoretical back-

ground behind the reasons for productivity differences between foreign affiliates. Section 3 discusses

the characteristics of our sample of firms and gives some descriptive statistics. Sections 4, 5 and 6

outline the econometric model and the main empirical results. Section 7 provides some robustness

analyses, and finally, Section 8 ends with some concluding remarks.

2

|

THEORETICAL BACKGROUND AND HYPOTHESES

2.1

|

The determinants of productivity differences among AFOFs and

EFOFs

In this section, we discuss how, in theory, EFOFs have a different productivity level from AFOFs.

Based on the commonly held view that MNEs transfer technological assets across borders to their own

foreign affiliates, and the evidence from advanced market FDI flows to emerging markets, foreign

ownership has been associated with improvements in firm productivity.

The traditional theory on MNEs (Dunning, 1988) strongly suggests that FOFs are more productive

than their domestic counterparts, the reason being that MNEs are presumed to possess firm‐specific

assets—unique products and production processes or intangibles, such as trademarks or reputations

for quality—that they can transfer to their affiliates at low marginal cost to overcome the disadvantage

of operating in a foreign environment (i.e., the asset‐exploiting perspective). The knowledge‐capital

model (Markusen, 2002) also makes similar deductions pointing out that MNEs have access to firm‐

specific assets that can translate into operating with superior technology or better management strat-

egy and know‐how. More recently, models with heterogeneous firms (Helpman, Melitz, & Yeaple,

2004) indicate that only the most productive firms can undertake FDI because doing business abroad

entails costs and risks that are higher than in the domestic market; only the most productive and inno-

vative firms will find it profitable to engage in foreign production. As a result, foreign investor firms

have to be more productive than their domestically owned competitors.

However, in the case of OFDI from emerging to advanced markets, the productivity‐improving role

of technology transfers is no longer so obvious. Recent literature (Athreye & Kapur, 2009; Deng, 2009;

Luo & Tung, 2007; Mathews, 2006; Rui & Yip, 2008) has argued that EMNEs lack intangible resources

such as technology, know‐how and brand names, and therefore invest abroad in order to obtain strategic