International joint ventures and technology diffusion: Evidence from China

• Author: Ruosi Lu,Chao Yang,Qing Liu
• Date: 01 January 2020
• DOI: http://doi.org/10.1111/twec.12809
• Published date: 01 January 2020

146

|

wileyonlinelibrary.com/journal/twec World Econ. 2020;43:146–169.

© 2019 John Wiley & Sons Ltd

Received: 6 August 2018

|

Revised: 27 March 2019

|

Accepted: 15 April 2019

DOI: 10.1111/twec.12809

ORIGINAL ARTICLE

International joint ventures and technology

diffusion: Evidence from China

QingLiu1

|

RuosiLu2

|

ChaoYang2

1Renmin University of China, Beijing, China

2University of International Business and Economics, Beijing, China

Funding information

Fundamental Research Funds; Renmin University of China, Grant/Award Number: 19XNL005

KEYWORDS

China, FDI, international joint ventures, technology diffusion

1

|

INTRODUCTION

Do international joint ventures (JVs) diffuse more technology to host country firms than wholly‐for-

eign‐owned firms (WFOs)? If yes, is it because JVs bring better technology or easier technology

access to the host country? The role of ownership structure of FDI in technology diffusion is at the

centre of FDI policies in many countries (e.g. China, India, Indonesia, Brazil, Mexico, Korea, Turkey,

the Philippines and Thailand (Karabay 2010; Holmes, McGrattan, & Prescott 2015) as well as in the

debates of international investment negotiations. One underlying premise for this kind of ownership

restriction policy is that local participation helps reveal and diffuse technology to domestic firms to

generate spillover effects (Blomstrom & Sjoholm 1999). However, there is little evidence supporting

this premise (Javorcik & Spatareanu 2008).

Theoretically, FDI (either JVs or WFOs) exerts two opposite intra‐industry impacts on host coun-

try firms (Aitken & Harrison 1999). One is the negative competition effect due to the loss of market

share and diminishing scale economies, which will decrease the productivity of domestic firms. As

long as they operate in the host country market, the negative impacts would persist regardless of the

ownership structure of FDI projects. The other is the positive technology diffusion effect, which could

be very different for JVs and WFOs for several reasons. First, JVs may bring more (e.g. Van Reenen

& Yueh 2012) or less (e.g. Desai, Fritz Foley, & Hines, 2009) technology to host countries than

WFOs. Second, JVs may diffuse technology to host country firms more effectively than WFOs. For

example, domestic and foreign‐shared ownership helps reveal the proprietary technology of foreign

investors to domestic firms (Blomström & Sjöholm, 1999; Van Reenen & Yueh, 2012); local partners

within JVs may use the knowledge acquired from the foreign investors in their other operations not

involving the foreign shareholders; when local partners are in charge of labour hiring, they may have

less ability or incentive to limit knowledge leakage or employee turnover, which facilitates technology

diffusion to host country firms (Javorcik & Spatareanu, 2008); JVs are more likely to outsource input

|

147

LIU etaL.

production to local suppliers and thus help transfer more technology to local suppliers (Javorcik, 2004;

Javorcik & Spatareanu, 2008; UNCTAD, 2001). Therefore, we expect that JVs are different from

WFOs as far as the net spillover effect is concerned, with the JVs possibly exerting more positive

technology diffusion effect.1

This paper attempts to investigate the international technology diffusion effect through FDI by em-

ploying Chinese data. We deviate from the traditional FDI spillover literature (thus contributing to this

literature) by explicitly taking into account the ownership structure of FDI projects. We distinguish

FDI inflows in China into China–foreign equity joint ventures (JVs) and wholly‐foreign‐owned firms

(WFOs) and investigate the potentially differentiated impacts of JVs and WFOs on Chinese domestic

firms. Our empirical evidence shows that the ownership structure of FDI has profound implications

for international technology diffusion. JVs facilitate more technology diffusion to Chinese firms com-

pared to WFOs. According to our preferred model specification, a one‐standard‐deviation increase of

the presence of JVs promotes Chinese firms’ TFP by 1.92%, while a one‐standard‐deviation increase

of the presence of WFOs decreases Chinese firms’ TFP by 7.99%.

The positive effect of JVs and the negative effect of WFOs on Chinese firms are robust to a battery

of checks. We show that our baseline results are not sensitive to alternative measures of firm produc-

tivity, and alternative definitions of WFO and JVs; To rule out the potential sample selection problem,

we run a separate regression considering a firm's entry and exiting. In addition to identifying horizon-

tal spillover effects, we also test how JVs and WFO spill over vertically; ownership structure within

domestic firms and the location of FIEs are also considered in other regression analyses. The differen-

tiated effects of JVs and WFOs on local firms also show heterogeneities as indicated in the literature,

which lends further support to the reliability of our findings and has policy implications; that is, the

benefits from JVs are not unconditional. The differentiated impacts of JVs and WFOs are causal, as

shown by our instrumental variable estimation, where the instrument is constructed by employing

China's FDI policy change. We also check the potential mechanisms behind the differentiated impacts

of JVs and WFOs. We find that JVs perform better than WFOs in TFP, R&D activities, employee

training and new product development, and we show that the superior performance of JVs is not due

to the potential complementarity between foreign investors and Chinese local partners. We also find

that, relative to WFOs, the diffusion effect of JVs is more sensitive to institutions (e.g. local financial

development and intellectual property rights (IPRs) protection) that affect the learning/access cost of

local firms. These results may imply that JVs bring better technology than WFOs to China, and it is

easier for Chinese firms to learn from JVs than from WFOs.

Our paper contributes to several strands of literature. First, technology is a key factor determining

long‐term economic growth and the quid pro quo FDI policy has been widely employed by develop-

ing countries to promote technology transfer from developed to developing countries (Karabay 2010;

Holmes, McGrattan and Prescott 2015). However, few empirical studies have evaluated the interna-

tional technology diffusion effect of international joint ventures (Van Reenen and Yueh 2012). Rising

research interests on this ownership structure issue are demonstrated by Van Reenen and Yueh (2012)

and Holmes, McGrattan, and Prescott (2015), both of which investigate the technology transfer from

multinational to China–foreign joint ventures. A recent paper by Jiang, Keller, Qiu, & Ridley (2017)

looks at the impact of forming joint ventures with foreign firms on the innovation of the parent firms

of Chinese local partners. Different from these papers, we look at the differentiated spillover impact

1 In empirical analyses, we could not separate the positive and the negative sides of the FDI impact. We estimate the net effect

of JVs and the net effect of WFOs on Chinese firms. With this notification, for succinctness we call the spillover effect

technology diffusion effect when the spillover effect is positive, and competition effect when the spillover effect is negative,

without causing misunderstanding.