Here - Beedie School of Business - Simon Fraser University

business.sfu.ca

Here - Beedie School of Business - Simon Fraser University

Dual-Edged Tools of Trade: How International Joint Ventures Help and

Hinder Capability Building of Chinese Firms

JING LI

Faculty of Business Administration

Simon Fraser University

8888 University Drive

Burnaby, BC

Canada V5A 1S6

Tel: +1 604 291-4568

Fax: +1 604 291-4920

Email: jingli@sfu.ca

URL: www.beedie.sfu.ca/profiles/JingLi

CHANGHUI ZHOU

Guanghua School of Management

Peking University

Beijing, China 100871

Tel: +86 10 6275-5089

Fax : +86 10 6275-1463

Email : czhou@gsm.pku.edu.cn

Journal of World Business

Both authors contributed equally to the paper. We thank the editors (Rosalie L. Tung and Zeynep Aycan),

the anonymous reviewer, Yongliang Huang, Alan M. Rugman, Edward J. Zajac, and seminar participants

at the University of Victoria and the Guanghua School of Management, Peking University for their

helpful comments and suggestions. We acknowledge financial support from the Natural Science

Foundation of China (Grant #70472023, to Zhou) and Simon Fraser University (Presidential Research

Grant).


Dual-Edged Tools of Trade: How International Joint Ventures Help and

Hinder Capability Building of Chinese Firms

Abstract

Forming international joint ventures (IJVs) with multinational corporations (MNCs) from

advanced economies has been widely adopted by firms in emerging economies as an

organizational approach to building up their innovative capabilities. In this paper, we

emphasize that such an approach has both advantages (knowledge transfer from MNCs) and

disadvantages (overdependence upon MNCs and reduction of innovation incentives) in the

capability building of indigenous firms. Utilizing a longitudinal dataset consisting of 474

industries in China during 1998–2002, we find supporting evidence for the co-existence of

both positive and negative impacts of IJVs. Specifically, we find that IJV presence has an

inverted U-shape impact on innovative capabilities of indigenous firms; furthermore, such

an impact becomes more salient in industries with low technology gap. We discuss the

implications of our findings for research, managerial practice, and government policy.

Key words: capability building, indigenous firms, emerging economies, international joint

ventures, China.

1


Innovative capabilities—the capabilities to create, change, or improve products and

processes (Bell & Pavitt, 1995; Lall, 1992)—have long been recognized as the main driver

of economic growth (Schumpeter, 1934), and as the ultimate source of competitive

advantage for firms and nations (Porter, 1990). As such, building up innovative capabilities

has been an important objective for indigenous firms in emerging economies (Amsden,

2001; Bell & Pavitt, 1995; Kim, 1997; Lall, 1987). As latecomers in the international

business arena, however, firms in emerging economies commonly lag far behind

multinational corporations (MNCs) in this dimension. These firms have realized that relying

solely on self-supported research and development activities to improve innovative

capabilities and to shorten the gap between such firms and MNCs is not only costly but also

slow. Therefore, firms in emerging economies have often adopted the practice of forming

international joint ventures (IJVs) with MNCs from advanced economies as an

organizational approach designed to accelerate the establishment of indigenous capabilities

(Lyles & Salk, 1996).

Research on IJVs in emerging economies has underscored the advantages of using

IJVs for capability building. Several studies, based on firm-level surveys or case analyses,

have provided supporting evidence for the transfer of valuable assets and knowledge such

as advanced technology, managerial skills, and marketing expertise from MNCs to their

local partners in IJVs (Hobday, 1995a; Lane et al., 2001; Luo, 2002; Lyles & Salk, 1996;

Mathews, 2002), and such knowledge transfer presumably contributes to the buildup of

innovative capabilities of indigenous firms.

2


What is much less understood, however, are the potential drawbacks of engaging in

IJVs, in terms of the effect of such engagement on the development of indigenous

capabilities. Some observations from the automobile industries in China and South Korea

have shown that collaborating with foreign firms in IJVs does not necessarily contribute to

or may even hinder the development of innovative capabilities of local partners (Gao, 2004;

Kim, 1997). Consequently, several questions naturally arise: What could be the potential

disadvantages of using IJVs to improve indigenous capabilities? If both upsides and

downsides of the impact of IJVs were considered, what would be the overall effect?

Furthermore, what factors may moderate the relative importance of the positive and

negative effects of IJVs? We intend to tackle these questions in the present paper. We

believe that by considering both sides of the impact of IJVs, we will be able to provide more

insights into the effectiveness of using IJVs as an organizational approach to the

improvement of indigenous capabilities.

Specifically, extending existing firm-level studies (Hobday, 1995a; Lane et al., 2001;

Lyles & Salk, 1996), we adopt an industry-level perspective to examine how the presence of

IJVs in an industry can affect innovative capabilities of indigenous firms in that industry.

Such a perspective enables us to add aggregate-level insights regarding the usefulness of

IJVs, by taking advantage of large secondary datasets available at an industry level.

In developing our theory, we first discuss the advantages of using IJVs to improve

innovative capabilities (knowledge transfer benefits from MNCs) and then point out the

potential negative impacts of using IJVs. As we will argue below, engaging in IJVs may

increase technological dependence of indigenous firms on their foreign partners, which in

3


turn leads to reduced incentives to innovate, and thus limits the development of innovative

capabilities. Given the positive and negative impacts of IJVs, we argue that there is likely

an optimal level of IJV presence in an industry (between too high and too low) that benefits

indigenous firms most; therefore, we propose an inverted U-shape hypothesis to predict the

impact of IJV presence on the innovative capabilities of indigenous firms. We further

examine the moderating effect of technology gap and hypothesize that the inverted U-shape

impact will be more salient in industries with low technology gap than in those with high

technology gap.

We conduct our empirical investigation in the context of China, which has proven to

be an appropriate research setting in which to study the impact of foreign direct investment

on the performance of indigenous firms (Buckley et al., 2002; Zhu & Tan, 2000). Our

sample is composed of a five-year (1998–2002) panel dataset with 474 industry sectors. The

findings based on this sample provide strong support for our hypotheses and therefore add

new insights to the understanding of conditions under which IJVs help build up indigenous

capabilities.

BACKGROUND AND HYPOTHESES

The existing international business literature suggests that the use of international

joint ventures can be an effective approach for indigenous firms in emerging economies to

the buildup of their innovative capabilities (Hobday, 1995a; Lane et al., 2001; Lyles & Salk,

1996; Mathews, 2002). For example, Hobday (1995a), using case analyses of several firms

in East Asia, presents evidence showing that technology transfer from developed countries

4


through IJVs provides opportunities for indigenous firms to adapt foreign technology to

local applications, and the technology adaptation process contributes to indigenous

capability building. Similarly, Mathews (2002) used several case studies to demonstrate that

forming IJVs with foreign firms can be an important catch-up strategy for latecomer firms

to gain competitive advantages.

In understanding the benefits of IJVs in improving indigenous capabilities, two

theoretical lenses are particularly useful. One such lens is transaction cost economics

(Williamson, 1985), from which perspective, when exploring markets in emerging

economies, MNCs are willing to bring about valuable firm-specific advantages to IJVs,

such as sophisticated technology, manufacturing skills, and managerial and marketing

expertise, due to control mechanisms and the profit- and risk-sharing structure of IJVs

(Beamish & Banks, 1987; Kogut, 1988). IJVs, as a hybrid organizational structure,

inherently facilitate collaboration between partnering firms and encourage the active

involvement of local partners, thereby creating a strong knowledge transfer effect on

indigenous firms.

Another theoretical perspective is the knowledge-based view (Grant, 1996), which

suggests that IJVs are an attractive vehicle for domestic companies to acquire tacit

knowledge through social and structural mechanisms that foster intimate interactions

between partners (Dhanaraj et al., 2004; Lane et al., 2001; Madhok, 1996, 1997). Lane et al.

(2001) and Lyles and Salk (1996), for example, conducted surveys of IJVs in Hungary and

found that with proper learning structures and processes, and with managerial assistance

5


from foreign partners, domestic Hungarian companies are able to acquire tacit knowledge

from foreign partners.

In short, these aforementioned firm-level studies have emphasized the positive

benefits that indigenous firms can obtain by forming IJVs with foreign firms. Adding to an

aggregate industry level, these studies seem to indicate that IJV presence in an industry will

have a positive linear effect on the buildup of indigenous capabilities; that is, the higher the

level of IJV presence, the more benefits that indigenous firms will obtain.

Another line of research which is relevant to our paper points to several

industry-level studies in the foreign direct investment (FDI) literature that have considered

the effect of foreign presence on the productivity of indigenous firms (e.g., Blomstrom,

1986; Buckley et al., 2002; Caves, 1974, 1996; Kokko, 1994; Liu et al., 2000). 1 These

studies indicate that, at an industry level, we should consider not only the direct technology

transfer benefits that IJV presence brings to indigenous firms that are involved in IJVs, but

also the indirect knowledge spillover benefits to firms that are not involved in IJVs. Indirect

spillover benefits are obtained mainly through informal and nonmarket mechanisms such as

demonstration effects, the movement of skilled workers, and supplier-buyer linkages (Caves,

1996; Eden et al., 1997). Since the knowledge spillover benefits further strengthen the

positive impact of IJV presence, these FDI studies suggest predictions similar to those of

the firm-level joint venture studies; that is, IJV presence will have a positive linear effect on

the capability building of indigenous firms.

1

Here “foreign presence” refers to the presence of foreign firms in an industry including both IJVs and

wholly owned subsidiaries. This line of research does not distinguish between different types of foreign firms.

6


Extending the previous studies, we are interested in exploring both the advantages

and disadvantages of IJV presence in affecting indigenous capability building. As we will

argue below, IJV presence could also have a significant negative impact on indigenous

capability building—instead of the linear relationship indicated by the existing literature, we

argue for an inverted U-shape relationship.

Impact of IJV Presence: An Inverted U-shape Hypothesis

Different from previous studies, we contend that knowledge transfer through IJV

presence can be a double-edged sword. Central to our argument is the fact that, in addition to

the benefits of IJV presence in transferring knowledge, the negative impact of IJV presence

not only exists but also depends on the level of IJV presence in an industry. Specifically, we

argue that IJV presence may gradually induce indigenous firms’ reliance on foreign partners’

sophisticated technology and well-established brands, such that engaging in IJVs reduces

indigenous firms’ incentives to exert efforts on innovative activities. Moreover, these

negative effects are amplified as IJV presence becomes high in an industry.

Industry reports have provided some evidence of the downsides of using IJVs to

improve indigenous capabilities (Gao, 2004; Kim, 1997). Gao (2004) describes the situations

of Chinese automobile companies, who tend to regard their joint venture partners more as

collaborators than competitors, and consequently feel less competitive pressure to perform

innovative activities and to develop new products in their independently owned subsidiaries.

For example, First Auto Works, one of the largest automobile companies in China, still does

not own any successful self-developed sedans, although they have been engaged in joint

ventures with two foreign automobile companies for about two decades. By contrast, small

7


and newly established automakers such as Chery and Geely—despite the fact that they

seldom have joint ventures with foreign multinational companies—have already developed

their own car models, successfully promoted their own brands, and aggressively expanded to

overseas markets (Gao, 2004). 2 A similar case has been documented in Kim (1997), about a

joint venture formed by Korea’s Daewoo Motor Company and General Motors. That joint

venture was partly responsible for the slow improvement in Daewoo’s innovative capabilities

(Kim, 1997, p. 115).

Therefore, the main disadvantage of using IJVs to improve innovative capabilities is

that it may increase the tendency of indigenous companies to rely on their foreign partners

and thereby reduce their innovative activities. In our view, this disadvantage can be severe

because, as suggested in the studies on capability building of firms in emerging markets (Bell

& Pavitt, 1993, 1995; Dutrenit, 2000; Figueiredo, 2001; Kim, 1997; Lall, 1987, 1992, 1993),

improving innovative capabilities is ultimately an endogenous and accumulative process,

which requires strong incentives to innovate, as well as substantial endogenous efforts.

Seen from an industry perspective, we argue that both the benefits and downsides of

IJV presence amplify as the level of IJV presence increases. On the one hand, there is more

knowledge transfer from MNCs to indigenous firms, directly and indirectly, resulting in

improvement in indigenous capabilities, while, on the other hand, indigenous companies

increasingly rely on MNCs, which in turn hinders their capability building. So what will be

2

We acknowledge that although these newly established automakers do not directly engage in joint ventures

with foreign automakers or benefit from direct knowledge transfer, they apparently benefit from the foreign

companies by indirectly imitating their technology and marketing skills. Chery was even accused of infringing

on GM’s intellectual property in developing one of its most popular models in China.

8


the net benefits (the benefits minus the downsides) that IJV presence may generate for

indigenous firms as IJV presence increases?

First, consider the situation in which the level of IJV presence is very low in an

industry. In this situation, the negative impact of IJV presence on innovation incentives is

insignificant, while the benefits of knowledge transfer are also low, suggesting that the net

benefits that indigenous companies obtain are likely limited. As IJV presence increases from

such a low level, we argue that the additional benefits of technology transfer are likely to be

greater than the additional costs of technology dependence, and as a result, the net benefits of

IJV presence likely increase. This is because, as IJV presence increases from a low level, the

additional benefits are gleaned not only by indigenous firms that get involved in IJVs

(through direct knowledge transfer), but also by many of those that are not involved in IJVs

(through indirect knowledge spillovers), while the additional costs are incurred only by those

engaged in IJVs.

Conversely, when the level of IJV presence becomes too high in an industry, although

the knowledge transfer benefits will become more significant, technological dependence of

domestic firms upon foreign partners will increase considerably as well, which will severely

weaken the net benefits of IJV presence. We argue that, as IJV presence increases from such

a high level, the additional costs of technology dependence will likely exceed the additional

benefits gained from knowledge transfer, and as a result, the net benefits of IJV presence will

likely decrease. On the one hand, since firms have already benefited greatly from direct and

indirect knowledge transfer provided by high IJV presence, they may not obtain much

additional benefit from further increase in IJV presence. On the other hand, there is no

9


slowdown in the increase of additional costs associated with technology dependence; as more

firms form IJVs with MNCs, their technology dependence leads to substantial reductions in

endogenous efforts.

The above discussion suggests that since both knowledge transfer from MNCs and

endogenous efforts are important in the capability buildup of indigenous firms, there is likely

an optimal level of IJV presence—somewhere between too high and too low—which benefits

indigenous firms most in improving their innovative capabilities. Below this optimal level, as

IJV presence increases, so do the net benefits of IJV presence due to increasing benefits

related to knowledge transfer, while above this optimal level, the net benefits of IJV presence

may decrease due to increasing costs associated with technology dependence. Hence, an

inverted U-shape relationship may exist between IJV presence and indigenous firms’

innovative capabilities. Formally,

H1: IJV presence in a local industry in emerging economies will have an inverted U-shape

effect on innovative capabilities of indigenous firms in that industry.

The Moderating Effect of Technology Gap

An important factor that may moderate the impact of IJV presence on innovative

capabilities of indigenous firms is the technology gap between foreign and indigenous firms,

a factor that may well reflect the absorptive capacity of indigenous firms (Kokko, 1994; Liu

et al., 2000). Cantwell (1989) and Lane et al. (2001) have pointed out that the ability of

indigenous companies to catch up with advanced foreign technology and to develop new

products and technology is contingent on their existing absorptive capacity (Cohen &

10


Levinthal, 1990). Since higher technology gap implies lower absorptive capacity, this view

suggests that indigenous firms have greater ability to absorb knowledge from IJVs in

industries with low technology gap than in those with high gap. Empirical studies have

presented evidence supporting this view (Buckley et al., 2002; Liu et al., 2000). For

example, Liu et al. (2000) examined knowledge acquisition in U.K. firms and found that

indigenous firms with low technology gap are more capable of absorbing foreign

knowledge and, accordingly, they benefit more from the presence of foreign investment in

the local industry sectors.

To examine how technology gap moderates the inverted U-shape relationship

between IJV presence and innovative capabilities, we first consider the situation where

technology gap is high. Under this condition, although there is more room or need for

acquiring additional knowledge and skills, indigenous firms may have difficulty in

obtaining benefits from MNCs due to the indigenous firms’ low absorptive capacity (Lane

et al., 2001). It follows that, in industries with high technology gap, the benefits of IJV

presence are likely limited, and an increase in IJV presence may not lead to a salient

increase in innovative capabilities, which implies that the first half of the inverted U-shape

curve is likely “flat.” Meanwhile, when technology gap is high, incentives to innovate may

not play a very important role in improving innovative capabilities because firms, with low

absorptive capacity, are less likely to build up innovative capabilities, even if they have

strong incentives to innovate. It follows that when IJV presence becomes high, its negative

effects (costs associated with reduction in innovation incentives) are also likely limited; an

increase in IJV presence may not lead to substantial decrease in innovative capabilities,

11


which suggests that the second half of the inverted U-shape curve is also likely flat. Hence,

when technology gap is high, innovative capabilities of indigenous firms are not very

sensitive to changes in IJV presence; that is, the inverted U-shape relationship may not be

salient.

On the contrary, we argue that this inverted U-shape relationship becomes more

pronounced when technology gap is low. On the one hand, indigenous firms with low gap

from MNCs are better able to absorb knowledge from MNCs, suggesting that an increase in

IJV presence likely leads to salient increases in innovative capabilities of indigenous firms.

Thus, the first half of the inverted U-shape curve is likely “steep.” On the other hand, in

low-gap industries, incentives to innovate play an important role in developing indigenous

capabilities because firms are already equipped with the technological competence to

innovate (Kim, 1997). It follows that high IJV presence that may significantly reduce firms’

innovation incentives can result in pronounced negative effects on indigenous capability

building. Hence, once IJV presence becomes high, the net benefits of IJV presence may

drop quickly, which implies that the second half of the inverted U-shape curve may be steep

as well.

To sum up, innovative capabilities of indigenous firms in low-gap industries are

more sensitive to changes in the level of IJV presence, compared with firms in high-gap

industries. Specifically, as IJV presence increases from a low level, the net benefits of IJV

presence will increase faster in low-gap industries due to advanced absorptive capacity.

Once IJV presence passes its optimal level, the net benefits of IJV presence will also

12


decrease more quickly due to higher opportunity costs associated with technology

dependency. Hence, we reach the following hypothesis.

H2: The inverted U-shape effect of IJV presence on innovative capabilities of indigenous

firms will be more salient in industries with low technology gap than in those with high

technology gap.

Sample and Data

METHODS

To test our hypotheses, we constructed a longitudinal sample of 474 industries in

China during 1998–2002, using China’s Annual Census of Industrial Enterprises (hereafter

referred to as “the Census data”). The Census data, compiled by the National Bureau of

Statistics of China, consist of detailed information about a company’s operational profile,

such as total production output, total output accounted for by new products, number of

employees, type of ownership, accumulated capital investment, etc. Articles that use data

publications by the National Bureau of Statistics of China (such as China Statistical

Yearbook and Industrial Census) have been published in leading academic journals such as

Journal of International Business Studies and China Economic Review (e.g., Buckley et al.,

2002; Li et al., 2001; Lin, 2000). Since our study is based on industry-level information, we

aggregated firm-level data to industry level according to the four-digit industry codes in the

Census data, which are similar to the SIC codes used in the U.S. (Buckley et al., 2002). For

the period 1998–2002, the Census dataset covers 606 industry sectors, some of which have

no foreign-investment presence. Given the purpose of this paper, we excluded all the

sector-year observations in which IJV presence is nil. As a result, our final sample includes

13


an imbalanced panel dataset of 474 industry sectors in China during 1998–2002. To check

the reliability of the Census data, we have performed several small subsample tests (detailed

in the appendix); the test results add support to the reliability of the Census data.

Statistical Model

In this study, we adopted standard panel-data regression models. Contingent on the

different assumptions on the properties of the unobserved industry-specific characteristics,

there are two types of estimation models: fixed-effects and random-effects models (Greene,

2003). We employed both models to analyze the data and performed Hausman tests to

examine which estimation model is the correct specification for our study.

Theoretical Variables

Dependent variable. In this study, the dependent variable is innovative capabilities of

indigenous firms in industry i at year t. Note that we define indigenous firms as those with no

foreign investment. We measure innovative capabilities using product innovation, as proxied

by the share of market value of new products in all products produced by indigenous firms in

industry i at year t. 3 In the Census data, new products are defined as “those developed and

produced for the first time in China in the past three years, including those adopting

completely new technological principle or design, or those substantially improved compared

with existing products in terms of structure, materials, and design such that these new

products have significant improvement in functionality or performance (Zhu, 2000).”

3

Scholars researching innovations have adopted different measures for innovative capabilities. For example,

some use questionnaires to ask managers or industry experts directly about the innovativeness of companies

(Bell, 2005; Subramaniam & Youndt, 2005), while others use the frequency with which one company’s

patents are cited (Phene et al., 2006). In our study, since our data are from an industry-level secondary dataset,

we do not have any questionnaire results or patent citations which we can use as an alternative measure for

firms’ innovative capabilities.

14


Although we are not able to distinguish whether the new products are more market-oriented

or technology-based, and whether they represent incremental or radical innovations, using the

share of market value of new products can to some extent capture the important role of new

products in an industry, which further reflects the innovativeness of an industry. Note that the

market value of new products provided in the Census data is not sales of new products but is

calculated by using the total number of new products (sold or not) multiplied by their current

market prices (Zhu, 2000). Arguably, our measure of innovative capabilities represents an

improvement over measures based on percentage of sales of new products (Sen & Egelhoff,

2000), because large sales of new products may not represent advanced innovative

capabilities but simply reflect favorable market conditions or sophisticated marketing skills.

IJV Presence. Our main theoretical variable is the presence of IJVs in an industry.

Following the FDI literature (Buckley et al., 2002; Liu et al., 2000), we employed alternative

measures to construct the variables on IJV presence: We measured IJV Presence_Emp as the

share of the employment of IJVs in the total employment of an industry; IJV Presence_Cap is

the share of the capital of IJVs in the total capital of an industry. We included this measure

with a one-year lag, and thus the scope of analysis for the dependent variable is between

1999 and 2002.

Technology Gap. In order to test H2, we are interested in how technology gap

moderates the impact of IJV presence on the capability building of indigenous firms. We

calculated technology gap as the ratio of the average productivity of foreign-invested

enterprises (including IJVs and wholly owned subsidiaries) to that of indigenous firms in an

industry (Kokko, 1994; Liu et al., 2000).

15


Control Variables

WOS Presence. This variable captures the impact of the presence of wholly owned

subsidiaries (WOSs) of foreign firms in an industry on capability building of indigenous

firms. We expect WOS presence to positively influence the capability building of

indigenous firms mainly through knowledge spillovers and enhancement of market

competition (Caves, 1996; Chung et al., 2003). Similar to the measure for IJV presence, we

measured WOS presence as the share of the employment or capital of WOSs in the total

employment or capital of an industry. Note that we implemented this measure with a

one-year lag.

Market Concentration and Market Growth. We used these two variables to control

the effect of industry structure. We employed a standard Herfindahl measure (Raghunathan,

1995) for market concentration by utilizing firm-level information for local market sales

contained in the Census data. For a given industry with n firms, we calculated the Herfindahl

measure as follows:

Herfindahl =

n

∑[ n

j=

1


j=

1

Sales

j

Sales

Note that market concentration was included with a one-year lag. Market Growth is

measured by

n

n

⎛ ⎞ ⎛ ⎞

⎜ Σ Sales j ⎟ ⎜ Σ Sales j ⎟ −1

.

⎝ j=

1 ⎠ ⎝ j=

1 ⎠

it

Market concentration may have a negative impact on innovative capabilities because

firms may feel less competitive pressure to innovate in more concentrated industries

(Davies & Caves, 1987), while it may also have a positive effect because firms in more

it−1

j

]

2

.

16


concentrated industries may have more available resources for innovative activities. Market

growth may increase firms’ incentives to innovate in order to gain competitive advantages

in a growing market, thus leading to rapid development in their innovative capabilities.

Hence, the predicted sign of market concentration is uncertain, while that of market growth

is positive.

IJV Age and IJV State Equity. Since our study focuses on the effect of international

joint ventures in an industry, we also included two control variables that characterize IJVs: 1)

IJV Age, measured by the average of the age of all IJVs in an industry; and 2) IJV State

Equity, measured by the total equity investment by state-owned enterprises (SOEs) in IJVs

divided by the total equity investment in IJVs in an industry. 4 We expect IJV age to

positively affect innovative capabilities of Chinese firms because engaging in IJVs for a

longer time period may contribute to more knowledge transfer. We expect IJV state equity to

negatively influence innovative capabilities of indigenous firms because SOEs may have

lower absorptive capacity to understand and assimilate knowledge transferred from MNCs

(Buckley et al., 2002), and they may also lack strong incentives to conduct innovative

activities, compared with privately owned enterprises (Tan & Tan, 2005).

Other variables. Following the literature (Buckley et al., 2002; Liu et al., 2000), we

include Capital-Labor Intensity, Firm Size, and Labor Quality, as they are all expected to

positively influence the capability building of indigenous firms. Specifically, we measured

capital-labor intensity by the logarithms of the ratio of the total capital to total employment

4

We also calculated two other variables for robustness tests: variance of IJV age and variance of IJV state

equity. No matter which sets of measures for IJV age and IJV state equity were used, we found similar

regression results.

17


in an industry; firm size is the logarithms of the average number of employees in an industry;

and labor quality is an index of the average wage rate in an industry. In addition, we

calculated SOE Weight as the share of employment or capital of state-owned enterprises in

the total employment or capital of an industry (Buckley et al., 2002). We expected SOE

weight to negatively influence innovative capabilities of indigenous firms because, as

mentioned earlier, SOEs may have low absorptive capacity or incentives in acquiring

knowledge from foreign firms. Finally, we also included three year dummies for 2000, 2001,

and 2002 in the regressions.

RESULTS

Table 1 summarizes the top 20 industry sectors, sorted by IJV presence (based on

capital measures) in 2002. The industry ranking, based on employment, is only slightly

different. Note that our sample consists of very specific industry sectors instead of more

aggregate ones, which provides an advantage in scrutinizing the impact of IJV presence.

Table 2 presents the correlation matrix and descriptive statistics of the variables in

the analysis. It shows that in our sample, the average IJV presence is 10.8 percent in

employment and 21.5 percent in capital. On average, technology gap between indigenous

and foreign firms is 2.5; that is, the average productivity of foreign-invested enterprises is

2.5 times that of indigenous enterprises.

We used both fixed-effects and random-effects models to analyze our data, and the

large values of Hausman tests suggest that the fixed-effects models provide better

specifications for our study. Therefore, Tables 3–5 report the results based on the

fixed-effects models. Results in Table 3 are based on the whole sample of 474 industry

18


sectors and 1,680 industry-year observations. Models 1 and 2 use employment-based

measures for IJV presence, WOS presence, and SOE weight, whereas models 3 and 4 use

capital-based measures. Models 2 and 4 consistently show that IJV presence has an inverted

U-shape relationship with innovative capabilities of indigenous firms. Therefore, H1 is

strongly supported; that is, there is likely an optimal level of IJV presence, and a moderate

level of IJV presence has a higher effect on innovative capabilities than either a low or a

high level of IJV presence.

To investigate the moderating effect of technology gap, we implemented subsample

tests and summarized the results in Tables 4 and 5. We used the median of technology gap

(median = 1.951) to split the whole sample into one with high technology gap (324

industries and 840 observations) and the other with low technology gap (320 industries and

840 observations). We derived the results in Table 4 using employment measures of IJV

presence, WOS presence, and SOE weight, whereas those in Table 5 use capital measures.

In both tables, we based models 1 and 2 on the sample with high technology gap and

models 3 and 4 on the sample with low technology gap. Both tables show that in the

low-technology-gap sample, IJV presence has an inverted U-shape effect on innovative

capabilities, while it does not have any significant effect in the high-technology-gap sample.

These results strongly support H2; that is, an inverted U-shape impact of IJV presence is

more salient in industries with low technology gap.

Tables 3, 4, and 5 also show some interesting results on the control variables. Table

3 shows that WOS presence has a positive effect on innovative capabilities of indigenous

firms. This is consistent with the literature, which emphasizes that WOS presence benefits

19


indigenous firms through knowledge spillovers and the enhancement of market competition

(Caves, 1996). As suggested in Tables 4 and 5, these benefits of WOS presence are more

salient in industries with low technology gap, probably because indigenous firms in these

industries are more capable of competing with foreign companies and can thus benefit more

from WOS presence.

Table 3 also indicates several important points: 1) Market concentration has no

effect on innovative capabilities, probably because its positive and negative effects, as

specified earlier, offset each other. 2) Market growth has a positive effect on innovative

capabilities, which is consistent with our prediction. 3) Unexpectedly, we did not find a

significant positive effect of IJV age on innovative capabilities of indigenous Chinese firms.

A plausible explanation is that, although engaging in IJVs for a longer time period may

contribute to more knowledge transfer, a long-term association with foreign partners may

also facilitate technology dependency upon foreign firms, which in turn limits the learning

incentives of the Chinese partners. 4) IJV state equity does not have a negative effect on

indigenous capabilities, as we have expected. Some plausible explanations include that

SOEs may have improved their ability to absorb knowledge from their foreign partners, as

well as their incentives to conduct risky innovative activities over time, due to the SOE

reforms carried out by the Chinese government (Tan & Tan, 2005). 5) Consistent with the

results for IJV state equity, we did not find a negative effect of SOE weight on innovative

capabilities of firms, either; instead, SOE weight has a significant positive effect, which

further strengthens our speculations that SOEs in China may have improved their incentives

or ability to become innovative. 6) Tables 4 and 5 suggest that Labor quality has no effect

20


when technology gap is high but has a significant positive effect when technology gap is

low, which implies that enhancing labor quality is more important in developing innovative

capabilities when technology gap is low.

We performed several robustness checks on the results in Tables 3, 4, and 5. First,

instead of using median split to divide the whole sample into high-technology-gap and

low-technology-gap subsamples, we employed the average of the technology gap as a cutoff

point to obtain the subsamples. We found that the results remain the same as those in Tables

4 and 5; thus, H2 is strongly supported. Second, to test the reliability of our data and the

robustness of the results, we ran regressions by using randomly selected subsamples. The

idea was that, if there were obvious inconsistencies in our data, regressions based on a

randomly selected sample might not generate consistent results. Specifically, we randomly

chose 95 percent of the observations from our data sample as a subsample, and the new

regressions showed consistent results and therefore strong support for all the hypotheses in

our study. Since the results based on capital or employment measures are similar, Table 6

only reports those on capital measures.

DISCUSSION AND CONCLUSION

As latecomers in the international business arena, indigenous firms in emerging

markets such as China are climbing a learning curve. In this learning process, a widely

adopted practice, among others, is the use of international joint ventures in the hope of

obtaining advanced technology and managerial capabilities from foreign multinational

companies, so as to improve organizational effectiveness (Lyles & Salk, 1996). In this paper,

21


we particularly view the use of IJVs as an organizational approach that indigenous firms in

emerging economies employ to improve their innovative capabilities. Adopting such a

perspective, we further see both benefits and potential downsides in the use of the IJV

approach by indigenous firms in emerging markets. Our argument is rooted predominantly

in perspectives theorizing the learning dynamics at the firm level, viewed through such

lenses as transaction cost economics, knowledge-based view, absorptive capacity, and

technology dependency. Our underlying logic is that, if this argument holds at the firm level,

the pattern of learning dynamics in question should be more likely to surface at an

aggregate industry level.

At the industry level, we have chosen to focus on the impact of two specific

contingencies, namely, the level of IJV presence and technology gap, on the building of

innovative capabilities of indigenous firms. Accordingly, we examined these effects by

utilizing data on 474 industries in China during 1998–2002. As our results have shown,

there is clear evidence that the presence of IJVs in an industry has an inverted U-shape

effect on product innovations of indigenous firms. There is also evidence for the moderating

effect of technology gap; that is, the inverted U-shape effect of IJV presence is more likely

to exist in industries with low technology gap than in those with high technology gap. These

findings are reliable, as well as impressive, because the evidence emerged out of

information pertinent to the whole population of Chinese enterprises across several years,

and such a longitudinal industry-level study can minimize the potential problem of sample

selection bias that commonly exists in firm-level studies.

22


Therefore, our findings have substantive meanings and add new insights into the

existing literature on the learning effect of using IJVs in emerging markets (Lane et al.,

2001; Lyles & Salk, 1996), as well as the literature on capability building of latecomer

firms (Hobday, 1995a; Kim, 1997; Lall, 2000; Mathews, 2002). First of all, our finding on

the inverted U-shape effect of IJV presence suggests the use of IJVs may have both positive

and negative impacts on the innovative capabilities of indigenous firms. Clearly, this offers

a strong point of departure from previous studies, which have emphasized only the positive

effects of using IJVs (Hobday, 1995a; Lane et al., 2001; Lyles & Salk, 1996; Mathews,

2002). Our findings particularly highlight the potential drawbacks in the use of IJVs; that is,

a propensity for indigenous firms to grow reliant on joint ventures for innovative

capabilities.

Furthermore, our findings on the moderating effect of technology gap suggest that

paying attention to the absorptive capacity of indigenous firms is important in scrutinizing

the impact of foreign direct investment (including IJVs) on host-country economies

(Buckley et al., 2002; Lane et al., 2001; Liu et al., 2000). Consistent with the existing

literature, our study indicates that if absorptive capacity is low, indigenous firms may

experience difficulty in assimilating and integrating knowledge from their joint ventures

with foreign MNCs; conversely, high absorptive capacity may facilitate the learning of

indigenous firms. Interestingly, beyond the existing studies, our results also suggest that in

the case of high absorptive capacity, the downsides of using IJVs may become more

pronounced because high absorptive capacity also implies high opportunity costs of relying

on MNCs.

23


Our findings have important implications for both managers of indigenous firms and

host-country governments in emerging markets. To managers of indigenous firms, the

implications of our study are mainly twofold. First, if they are attracted to the benefits of

using IJVs to acquire knowledge, then they should also keep in mind that such benefits may

come with costs. One such cost for indigenous firms is the reduction in innovative

incentives due to the growing reliance on foreign partners. The aggregate evidence

presented in this paper suggests that such tendency may well exist. Therefore, local

managers should be vigilant in curbing themselves from growing reliant on foreign MNCs.

Local managers should be aware that minimizing such reliance is especially important when

their companies are already equipped with technological competence, because the

opportunity costs of relying on their foreign partners can be significant.

The divergence of the technology development paths of two Chinese home

appliance companies (Haier Group and Beijing Snowflake Electric Appliance Group) can

be used to illustrate these points (Palepu, Khanna, & Vargas, 2006). Both firms strategized

to learn from foreign firms in their early stage of development in the early 1990s: Haier

entered a joint venture with Mitsubishi, and Beijing Snowflake with Whirlpool. During the

collaborations with foreign partners, Haier continued to strive for independent technological

competence and innovative capabilities through R&D investments, while Beijing Snowflake

formed technological dependence upon its foreign partner. The result is that Haier has

become the second largest refrigerator manufacturer in the world, behind Whirlpool, while

Beijing Snowflake has failed to succeed. This particular example reflects the importance of

24


developing independent technological capabilities while engaging in joint ventures with

foreign firms.

Our paper by no means implies that indigenous firms should give up their links with

foreign firms in IJVs; instead, we urge that indigenous firms should take full advantage of

these links while making purposeful, continuous and effective efforts to improve

independent innovative capabilities. These implications share some insights with the

previous literature on the improvement of latecomer companies. By studying some

electronics companies in the four East Asian “dragon” countries (South Korea, Taiwan,

Hong Kong, and Singapore), Hobday (1995b) concluded that subcontracting and original

equipment manufacture (OEM) mechanisms enabled latecomer firms in these countries to

improve manufacturing and design technology, since the needs of export consumers drive

the pace of learning. However, to overcome technology dependency on OEMs and licensing,

South Korean manufacturers invested heavily in their in-house R&D spending, acquired

overseas high-technology firms, and formed technology partnerships with leading foreign

companies on an equal-technology basis (Hobday, 1995b; Kim, 1997). Several studies have

shown that it took latecomer electronics companies in the four dragons about 30 years to

transform from having the capability of simple assembly to embracing independent

innovative capabilities (Choung, Hwang, Choi, & Rim, 2000; Hobday, 1995b). As China

has opened its market for FDI for 27 years since 1979, it seems about time for Chinese

companies to transfer from a stage of technology dependency upon foreign partners in IJVs

(technology-using stage) to a new stage of independent technology improvement

25


(technology-generating stage), which makes endogenous innovation efforts important for

the Chinese firms.

The second implication for the local managers is that they should understand how

learning is related to the level of absorptive capacity of their own firms. It is always

important for indigenous firms to improve their learning ability in order to absorb

knowledge from international joint ventures and their foreign partners. Our recent interview

with a senior manager in Shenyang Brilliance Jinbei Automotive Co., which has a joint

venture with BMW, a prestigious German automaker, has provided some insights to

illustrate this point. When commenting on learning from BMW, the manager said, “it is like

a Chinese girl who does not speak English who marries an American, so it is very difficult

for knowledge acquisition to occur.” Hence, without an appropriate level of absorptive

capacity, the learning benefits of engaging in IJVs may be very limited.

Overall, we would advise managers in indigenous firms that the following three

activities should be well managed for the purpose of learning and catching up: actively

engaging in joint ventures, continuously improving absorptive capacity, and firmly

maintaining strong incentives to innovate. As engaging in joint ventures brings in learning

opportunities, whether latecomer firms are able to catch up with the technology frontier

depends on their learning ability and innovation incentives. If indigenous firms are only

satisfied with short-term profits in joint ventures and tend to depend on the existing

technology and products supplied by their foreign partners, involvement in IJVs with

foreign firms will actually hinder the improvement of indigenous capabilities.

26


As for governments in emerging economies, we would suggest that policies

encouraging the use of IJVs may be effective in improving indigenous capabilities, but only

to a certain degree. Governments in emerging economies often require foreign firms that

wish to enter their countries to establish IJVs with indigenous firms, in the hope that

indigenous firms will benefit significantly from knowledge transfer from foreign firms

(Beamish, 1993; Moran, 1998). However, our study suggests that knowledge acquisition of

indigenous firms varies across industries, contingent on the technology gap between

indigenous and foreign firms. Hence, government policies towards IJVs should also depend

on the technology level of different industries. In industries with high technology gap, we

find that learning from IJVs is likely marginal. The implication for government policy is

that relying solely on foreign technology transfer in IJVs is not effective; instead,

governments should employ complementary policies to help latecomer industries acquire

technology assistance from a variety of sources, so as to improve the absorptive capacity of

latecomer industries.

Our study also shows that in industries with low technology gap, international joint

ventures tend to play a positive role in building up indigenous capabilities, but the benefits

decrease substantially once the presence of IJVs reaches a certain high level. This suggests

that a high level of IJV presence in the local economies may result in indigenous firms’

reliance on IJVs, and such reliance is definitely destructive to the capability improvement of

indigenous firms. Therefore, it is important for the host-country government to adopt a

dynamic perspective in enacting and adjusting their policies toward the use of IJVs, so as to

27


take advantage of knowledge transfer in IJVs while minimizing the tendency of technology

dependence.

The Chinese government has set a good example in this regard. China required the

establishment of IJVs in many industries before its accession to the WTO in 2001. There is

no doubt that firms with an appropriate level of absorptive capacity have benefited from

knowledge transfer from their foreign partners. Upon joining the WTO, the Chinese

government committed to gradually reducing the strict requirements on the use of IJVs and

allowing more wholly owned subsidiaries for inward foreign direct investment in China.

We envision that such policy change in China will actually further propel the innovative

capabilities of the indigenous firms in the long run because, as the technology gap between

foreign and indigenous firms has been substantially narrowed, Chinese companies are able

to face strong foreign competition. A more open market will reduce technology dependency

and encourage more endogenous efforts on the part of Chinese firms. We believe that such a

gradual open-market policy—which enables latecomer companies to take advantage of

technology transfer benefits from IJVs, while reducing the companies’ technology

dependence in the long run—can be useful for other emerging economies such as India and

Russia.

Finally, we hope that future studies will address a number of possible extensions of

our study. First, since one potential limitation of our study is that our empirical data were

from the period 1998 –2002, future studies could use more recent data to examine the

robustness of our results. Second, subsequent research could more fully examine the

underlying mechanisms that may influence the impact of IJV presence on indigenous

28


capability. For example, R&D investments and innovation of IJVs may affect knowledge

transfer from foreign to local partners because local partners may benefit from the joint

innovation process. Third, to fully capture the effectiveness of using IJVs as an

organizational approach to the improvement of indigenous capabilities, one could compare

the impact of IJVs on indigenous companies to other factors that contribute to innovative

capabilities, such as endogenous R&D efforts, capital-goods imports, and technology

licensing. Future studies could also extend the current industry-level analysis to firm-level

examination by directly comparing the capability development of local companies involved

in IJVs with those not involved in IJVs. Such comparison could provide more insight into

how to maximize the benefits of IJVs while minimizing the potential costs.

29


References

Amsden, A. (2001). The rise of "the rest": Challenges to the West from late-industrializing

economies. New York and Oxford: Oxford University Press.

Beamish, P. W., & Banks, J. (1987). Equity joint ventures and the theory of the multinational

enterprise. Journal of International Business Studies, 18(2): 1-16.

Beamish, P. W. (1993). The characteristics of joint ventures in the People's Republic of

China. Journal of International Marketing 1(2), 29-48.

Bell, G. G. (2005). Clusters, networks, and firm innovativeness. Strategic Management

Journal, 26(3): 287-295.

Bell, M., & Pavitt, K. (1993). Technological accumulation and industrial growth: Contrasts

between developed and developing Countries. Industrial & Corporate Change, 2(2):

157-210.

Bell, M., & Pavitt, K. (1995). The development of technological capabilities. In I. U. Haque

(Ed.), Trade, Technology and International Competitiveness. Washington, DC: World

Bank.

Blomstrom, M. (1986). Foreign investment and productive efficiency: The case of Mexico.

Journal of Industrial Economics, 35(1): 97-110.

Buckley, P. J., Clegg, J., & Wang, C. (2002). The impact of inward FDI on the performance

of Chinese manufacturing firms. Journal of International Business Studies, 33(4):

637-655.

Cantwell, J. A. (1989). Technological Innovation and the Multinational Corporation. Oxford:

Basil Blackwell.

Caves, R. E. (1974). Multinational firms, competition, and productivity in host-country

markets. Economica, 41(162): 176-193.

Caves, R. E. (1996). Multinational enterprises and economic analysis. New York; Cambridge,

MA: Cambridge University Press.

Choung, J.-Y., Hwang, H.-R., Choi, J.-H., & Rim, M.-H. (2000). Transition of latecomer

firms from technology users to technology generators: Korean semiconductor firms.

World Development 28(5), 969-982.

Chung, W., Mitchell, W., & Yeung, B. (2003). Foreign direct investment and host country

productivity: the American automotive component industry in the 1980s. Journal of

30


International Business Studies, 34(2): 199-218.

Cohen, W. M., & Levinthal, D. A. (1990). Absorptive capacity: A new perspective on

learning and innovation. Administrative Science Quarterly, 35(1): 128-152.

Davies, S., & Caves, R. E. (1987). Britain’s Productivity Gap. Cambridge: Cambridge

University Press.

Dhanaraj, C., Lyles, M., Steensma, H., & Tihanyi, L. (2004). Managing tacit and explicit

knowledge transfer in IJVs: The role of relational embeddedness and the impact on

performance. Journal of International Business Studies, 35(6): 428-443.

Dutrenit, G. (2000). Learning and knowledge management in the firm: From knowledge

accumulation to strategic capabilities. Cheltenham, UK and Northampton, MA, USA:

Edward Elgar.

Eden, L., Levitas, E., & Martinez, R. J. (1997). The production, transfer and spillover of

technology: Comparing large and small multinationals as technology producers. Small

Business Economics, 9: 53-66.

Figueiredo, P. N. (2001). Technological learning and competitive performance. Cheltenham,

UK ; Northampton, MA: Edward Elgar.

Gao, P. (2004). Shaping the future of China's auto industry. McKinsey Quarterly(3): 123-126.

Grant, R. M. (1996). Toward a knowledge-based theory of the firm. Strategic Management

Journal, 17(Special issue): 109-122.

Greene, W. H. (2003). Econometric analysis. Upper Saddle River, N.J.: Prentice Hall.

Hobday, M. (1995a). Innovation in East Asia: the challenge to Japan. Aldershot: Edward

Elgar.

Hobday, M. (1995b). East Asian latecomer firms: Learning the technology of Electronics.

World Development 23(7), 1171-1193.

Kim, L. (1997). Imitation to Innovation: The dynamics of Korea's technological learning.

Boston: Harvard Business School Press.

Kogut, B. (1988). Joint ventures: Theoretical and empirical perspectives. Strategic

Management Journal, 9(4): 319-332.

Kokko, A. (1994). Technology, market characteristics, and spillovers. Journal of

Development Economics, 43(2): 279-293.

31


Lall, S. (1987). Learning to industrialize: The acquisition of technological capability by India.

Macmillan, London.

Lall, S. (1992). Technological capabilities and industrialization. World Development, 20(2):

165-186.

Lall, S. (1993). Promoting technology development: The role of technology transfer and

indigenous effort. Third World Quarterly, 14(1): 95-108.

Lall, S. (2000). Technological change and industrialization in the Asian newly industrializing

economies: Achievements and challenges. In L. Kim, & R. R. Nelson (Eds.),

Technology, Learning and Innovation: Experiences of Newly Industrializing

Economies. Cambridge, U.K.: Cambridge University Press.

Lane, P. J., Salk, J. E., & Lyles, M. A. (2001). Absorptive capacity, learning, and

performance in international joint ventures. Strategic Management Journal, 22(12):

1139-1161.

Li, J., Lam, K., & Qian, G. (2001). Does Culture Affect Behavior and Performance of Firms?

The Case of Joint Ventures in China. Journal of International Business Studies, 32(1):

115-131.

Lin, J. Y. (2000). WTO accession and China's agriculture. China Economic Review, 11(4):

405-408.

Liu, X., Siler, P., Wang, C., & Wei, Y. (2000). Productivity spillovers from foreign direct

investment: Evidence from UK industry level panel data. Journal of International

Business Studies, 31(3): 407-425.

Luo, Y. (2002). Stimulating exchange in international joint ventures: An attachment-based

view. Journal of International Business Studies, 33(1): 169-181.

Lyles, M. A., & Salk, J. E. (1996). Knowledge acquisition from foreign parents in

international joint ventures: An empirical examination in the Hungarian context.

Journal of International Business Studies, 29(2): 154-174.

Madhok, A. (1996). The organization of economic activity: Transaction costs, firm

capabilities, and the nature of governance. Organization Science, 7(5): 577-590.

Madhok, A. (1997). Cost, value and foreign market entry mode: The transaction and the firm.

Strategic Management Journal, 18(1): 39-61.

Mathews, J. A. (2002). Competitive advantages of the latecomer firm: A resource-based

32


account of industrial catch-up strategies. Asia Pacific Journal of Management, 19:

467-488.

Moran, T. H. (1998). Foreign direct investment and development: the new policy agenda for

developing countries and economies in transition. Washington, DC.: Institute for

International Economics.

Palepu, K., Khanna, T., & Vargas, I. (2006). Haier: Taking a Chinese company global.

Harvard business School Cases.

Phene, A., Fladmoe-Lindquist, K., & Marsh, L. (2006). Breakthrough innovations in the U.S.

biotechnology industry: the effects of technological space and geographic origin.

Strategic Management Journal, 27(4): 369-388.

Porter, M. (1990). The competitive advantage of nations. New York: Free Press.

Raghunathan, S. P. (1995). A refinement of the entropy measure of firm diversification:

Toward definitional and computational accuracy'. Journal of Management, 5:

989-1002.

Schumpeter, J. A. (1934). The theory of economic development. Cambridge, MA: Harvard

University Press.

Sen, F. K., & Egelhoff, W. G. (2000). Innovative capabilities of a firm and the use of

technical alliances. IEEE Transactions on Engineering Management, 47(2): 174-183.

Subramaniam, M., & Youndt, M. A. (2005). The influence of intellectual capital on the types

of innovative capabilities. Academy of Management Journal, 48(3): 450-463.

Tan, J., & Tan, D. (2005). Environment-strategy co-evolution and co-alignment: A staged

model of Chinese SOEs under transition. Strategic Management Journal, 26:

141-157.

Williamson, O. E. (1985). The economic institutions of capitalism: firms, markets, relational

contracting. New York; London: Free Press; Collier Macmillan.

Zhu, G., & Tan, K. Y. (2000). Foreign direct investment and labor productivity: New

evidence from China as the host country. Thunderbird International Business Review,

42(5): 507-528.

Zhu, Z. (2000). Interpretations of Main Measures in China's Statistical Report System.

Beijing, China: China Statistics.

33


Appendix: Reliability Check of the Census Data in China

First, to examine the reliability of employment information in the Census data, we matched

our Census data with the IPO data from the database at China Center for Economic

Research (CCER) at Peking University. We used the Chinese name of a company to match

the two datasets. Since the IPO data includes a very limited number of observations with

information on company employment, we had only 35 matched observations. We then

calculated a percentage measure to capture the difference between the two datasets; that is,

DIFF = absolute value of [(employee number at the IPO data – employee number at the

Census data)/employee number at the Census data]. We found that DIFF varies between 0

and 3.7%, with an average at 1.45%. We believe that such difference is reasonably small

because employment in the IPO data is reported at a specific date when a company went

public, while employee number at the Census data is reported at the end of a fiscal year.

More importantly, when adding up the firm-level employee information, we found DIFF to

be very small, at 0.25%. This adds some validity to our study, which is based on aggregate

industry-level information.

Second, since we have only 35 matched observations on employment, we matched the stock

market data at CCER with our data on a common variable—total assets. As a result, we

found 320 matched observations according to a company’s Chinese name. We also

calculated a percentage measure, DIFF1 = absolute value of [(total assets in the stock

market data – total assets at the Census data)/total assets at the Census data]. We found that

DIFF1 varies between 0 and 1.1%, with a mean at 0.34%. When adding up the firm-level

data to an aggregate level, we found that DIFF1 is only 0.015%, which further lends support

to our study based on industry-level information.

34


Table 1. Top 20 Industries by IJV Presence in China in 2002 5

Industry Sectors IJV Presence

1 Containers Manufacturing 0.816

2

3

Water-wheel engines

Manufacturing

Instrument, Meter, Stationery and

Office Machine Maintenance

0.760

0.751

4 Car Manufacturing 0.739

5 Hair-Cutting Equipment 0.726

6 Other Flavoring Products 0.690

7

Auto Instrument, Meter

Manufacturing

0.670

8 Carbonated Soft Drinks 0.669

9 Electronic Musical Instruments 0.608

10

Household Laundry Equipment

Manufacturing

0.588

11 Timing Instruments 0.572

12 Air Conditioner Manufacturing 0.549

13

14

Other Silk Spinning And Silk

Fabrics

Industrial Technique Glass

Manufacturing

0.545

0.543

15 Airplane Repair 0.532

16 Integrated Circuits 0.529

17 Automobile Body Manufacturing 0.515

18 Steam Engine Manufacturing 0.505

19

Commercial, Food and Beverage,

and Service Industry Machinery

Manufacturing

0.499

20 Other Synthon Manufacturing 0.493

5 IJV presence is measured by the share of capital of IJVs in the total capital of the industry.


Table 2. Correlation Matrix and Descriptive Statistics

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 Innovative Capabilities 1

2 IJV Presence_Emp 0.051 1

3 IJV Presence_Cap 0.094 0.697 1

4 WOS Presence_Emp 0.046 0.263 0.023 1

5 WOS Presence_Cap -0.056 0.232 -0.023 0.807 1

6 Technology Gap 0.066 -0.144 0.012 -0.157 -0.112 1

7 Market Concentration 0.142 0.038 0.007 0.107 0.023 0.118 1

8 Market Growth 0.083 0.023 0.031 -0.003 -0.004 0.039 -0.027 1

9 Capital-Labor Intensity 0.262 0.182 0.283 -0.101 -0.103 0.207 0.161 0.108 1

10 Labor Quality 0.007 0.004 0.067 -0.033 -0.004 0.002 0.063 0.109 0.02 1

11 Firm Size 0.321 -0.063 -0.083 -0.041 -0.164 0.093 0.092 0.044 0.036 -0.053 1

12 SOE Weight_Emp 0.164 -0.295 -0.16 -0.324 -0.365 0.339 0.077 -0.001 0.2 0.044 0.421 1

13 SOE Weight_Cap 0.128 -0.306 -0.187 -0.305 -0.346 0.313 0.078 -0.012 0.181 0.021 0.397 0.869 1

14 IJV Age 0.032 0.012 -0.034 -0.046 -0.01 -0.069 -0.045 -0.022 0.013 -0.106 0.123 -0.021 0.001 1

15 IJV State Equity 0.052 -0.071 0.034 -0.246 -0.259 0.141 0.006 -0.015 0.176 0.006 0.147 0.334 0.286 0.068 1

Mean 0.081 0.108 0.215 0.088 0.155 2.486 0.055 0.194 3.949 0.077 5.53 0.304 0.303 6.887 0.105

S.D. 0.107 0.09 0.147 0.117 0.147 2.257 0.073 0.339 0.581 0.145 0.577 0.213 0.226 1.902 0.093

Note: IJV Presence_Emp refers to the presence of international joint ventures in an industry in terms of employment;

IJV Presence_Cap refers to the presence of international joint ventures in an industry in terms of capital;

WOS Presence_Emp refers to the presence of wholly owned subsidiaries in an industry in terms of employment;

WOS Presence_Cap refers to the presence of wholly owned subsidiaries in an industry in terms of capital;

SOE Weight_Emp refers to the weight of state owned enterprises in an industry in terms of employment;

SOE Weight_Cap refers to the weight of state owned enterprises in an industry in terms of capital.

IJV State Equity refers to the share of equity investment by state owned enterprises in all IJVs in an industry.

36


Table 3. Results of Fixed-Effects GLS Regression on the Whole Sample 1998–2002

Dependent Variable: Innovative Capabilities of Indigenous Chinese Firms

Based on

employment measures

Based on

capital measures

Model 1 Model 2 Model 3 Model 4

IJV Presence

0.048

(0.034)

0.135**

(0.059)

0.033*

(0.019)

0.124***

(0.046)

IJV Presence^2

-0.168*

(0.091)

-0.145**

(0.067)

WOS Presence

0.071**

(0.029)

0.072**

(0.029)

0.056***

(0.021)

0.053***

(0.021)

Technology Gap

-0.000

(0.001)

0.000

(0.001)

-0.001

(0.001)

-0.001

(0.001)

Market Concentration

0.028

(0.038)

0.033

(0.038)

0.029

(0.038)

0.042

(0.038)

Market Growth

0.011***

(0.004)

0.012***

(0.004)

0.011***

(0.004)

0.011***

(0.004)

Capital-labor Intensity

0.006

(0.007)

0.006

(0.007)

0.006

(0.007)

0.005

(0.007)

Labor Quality

0.003

(0.010)

0.004

(0.010)

0.002

(0.010)

0.002

(0.010)

Firm Size

-0.023**

(0.010)

-0.023**

(0.010)

-0.022**

(0.010)

-0.022**

(0.010)

SOE Weight

0.028*

(0.018)

0.028*

(0.018)

0.039***

(0.013)

0.039***

(0.013)

IJV Age

0.000

(0.001)

0.000

(0.001)

0.000

(0.001)

0.000

(0.001)

IJV State Equity

0.027

(0.023)

0.028

(0.023)

0.028

(0.023)

0.027

(0.023)

Year 2000

0.005

(0.004)

0.005

(0.004)

0.006

(0.003)

0.006

(0.003)

Year 2001

0.010**

(0.004)

0.010**

(0.004)

0.011***

(0.004)

0.011***

(0.004)

Year 2002

0.009*

(0.005)

0.009*

(0.005)

0.010**

(0.004)

0.011**

(0.004)

Observations/Industries 1680/474 1680/474 1680/474 1680/474

F value 3.38 3.38 3.87 3.93

Within R 2 0.0382 0.0409 0.0434 0.0472

Between R 2 0.0157 0.0078 0.0196 0.0126

Overall R 2 0.0028 0.0005 0.0057 0.0025

Note: Values in parentheses are standard errors. *** p < 0.01; ** p < 0.05; * p < 0.10.

37


Table 4. Results of Fixed-Effects GLS Regression on the Subsamples 1998–2002

Dependent Variable: Innovative Capabilities of Indigenous Chinese Firms

High-tech-gap sample Low-tech-gap sample

Model 1 Model 2 Model 3 Model 4

IJV Presence_Emp

0.101

(0.072)

0.166

(0.122)

0.079*

(0.046)

0.229***

(0.079)

IJV Presence_Emp^2

-0.165

(0.250)

-0.252**

(0.109)

WOS Presence_Emp

-0.071

(0.063)

-0.083

(0.066)

0.180***

(0.038)

0.192***

(0.038)

Technology Gap

-0.001

(0.001)

-0.001

(0.001)

0.004

(0.008)

0.004

(0.008)

Market Concentration

0.100

(0.069)

0.112

(0.071)

-0.091

(0.057)

-0.089

(0.057)

Market Growth

0.010

(0.006)

0.009

(0.006)

0.008

(0.006)

0.010

(0.006)

Capital-Labor Intensity

0.013

(0.012)

0.013

(0.012)

0.005

(0.013)

0.007

(0.013)

Labor Quality

-0.012

(0.013)

-0.012

(0.013)

0.041**

(0.019)

0.042**

(0.019)

Firm Size

-0.028

(0.017)

-0.027

(0.017)

-0.015

(0.014)

-0.017

(0.014)

SOE Weight_Emp

0.004

(0.026)

0.004

(0.026)

0.082***

(0.029)

0.080***

(0.029)

IJV Age

0.001

(0.002)

0.001

(0.002)

0.000

(0.002)

0.000

(0.002)

IJV State Equity

0.007

(0.032)

0.007

(0.032)

-0.009

(0.044)

-0.008

(0.044)

Year 2000

0.003

(0.005)

0.003

(0.005)

0.009*

(0.006)

0.009*

(0.006)

Year 2001

0.012*

(0.007)

0.012*

(0.007)

0.008

(0.006)

0.007

(0.006)

Year 2002

0.007

(0.009)

0.007

(0.009)

0.012*

(0.007)

0.011

(0.007)

Observations/Industries 840/324 840/324 840/320 840/320

F value 3.10 2.92 3.49 3.65

Within R 2 0.0796 0.0804 0.0881 0.0977

Between R 2 0.0432 0.0351 0.0018 0.0009

Overall R 2 0.0203 0.0142 0.0014 0.0008

Note: Values in parentheses are standard errors. *** p < 0.01; ** p < 0.05; * p < 0.10.

38


Table 5. Results of Fixed-Effects GLS Regression on the Subsamples 1998–2002

Dependent Variable: Innovative Capabilities of Indigenous Chinese Firms

High-tech-gap sample Low-tech-gap sample

Model 1 Model 2 Model 3 Model 4

IJV Presence_Cap

0.033

(0.032)

0.113

(0.074)

0.077***

(0.028)

0.235***

(0.074)

IJV Presence_Cap^2

-0.122

(0.101)

-0.271**

(0.119)

WOS Presence_Cap

-0.011

(0.038)

-0.018

(0.038)

0.136***

(0.030)

0.138***

(0.030)

Technology Gap

-0.001

(0.001)

-0.001

(0.001)

-0.006

(0.008)

-0.006

(0.008)

Market Concentration

0.098

(0.068)

0.116*

(0.069)

-0.089

(0.057)

-0.075

(0.057)

Market Growth

0.011*

(0.006)

0.011*

(0.006)

0.004

(0.006)

0.004

(0.006)

Capital-Labor Intensity

0.019

(0.012)

0.018

(0.012)

0.012

(0.013)

0.012

(0.013)

Labor Quality

-0.014

(0.013)

-0.014

(0.013)

0.041**

(0.019)

0.037**

(0.019)

Firm Size

-0.027*

(0.016)

-0.029*

(0.016)

-0.011

(0.014)

-0.013

(0.014)

SOE Weight_Cap

-0.008

(0.020)

-0.007

(0.020)

0.105***

(0.021)

0.102***

(0.021)

IJV Age

0.001

(0.002)

0.001

(0.002)

0.001

(0.002)

0.001

(0.002)

IJV State Equity

0.013

(0.032)

0.012

(0.032)

0.021

(0.044)

0.013

(0.043)

Year 2000

0.001

(0.005)

0.001

(0.005)

0.010**

(0.005)

0.010**

(0.005)

Year 2001

0.009

(0.006)

0.009

(0.006)

0.009*

(0.006)

0.008

(0.006)

Year 2002

0.002

(0.008)

0.003

(0.008)

0.012**

(0.006)

0.011**

(0.006)

Observations/Industries 840/324 840/324 840/320 839/320

F value 2.93 2.83 4.66 4.73

Within R 2 0.0755 0.0782 0.1142 0.1233

Between R 2 0.0273 0.0210 0.0002 0.0001

Overall R 2 0.0104 0.0064 0.0009 0.0005

Note: Values in parentheses are standard errors. *** p < 0.01; ** p < 0.05; * p < 0.10.

39


Table 6. Results of Fixed-Effects GLS Regression on a Randomly Selected Sample 1998–2002

Dependent Variable: Innovative Capabilities of Indigenous Chinese Firms

Whole sample

High-tech-gap

sample

Low-tech-gap

sample

Model 1 Model 2 Model 3

IJV Presence_Cap

0.128***

(0.048)

0.112

(0.075)

0.243***

(0.078)

IJV Presence_Cap^2

-0.150**

(0.069)

-0.117

(0.102)

-0.278**

(0.124)

WOS Presence_Cap

0.056*

(0.022)

-0.015

(0.039)

0.144***

(0.031)

Technology Gap

-0.001

(0.001)

-0.001

(0.001)

-0.002

(0.008)

Market Concentration

0.036

(0.041)

0.091

(0.071)

-0.078

(0.061)

Market Growth

0.010**

(0.004)

0.010

(0.006)

0.002

(0.007)

Capital-Labor Intensity

0.005

(0.008)

0.020

(0.012)

0.014

(0.014)

Labor Quality

0.002

(0.010)

-0.015

(0.013)

0.044**

(0.020)

Firm Size

-0.022**

(0.010)

-0.024*

(0.017)

-0.013

(0.015)

SOE Weight_Cap

0.038***

(0.014)

-0.016

(0.020)

0.113***

(0.022)

IJV Age

0.000

(0.001)

0.000

(0.002)

0.001

(0.002)

IJV State Equity

0.026

(0.024)

0.019

(0.033)

0.003

(0.045)

Year 2000

0.005

(0.004)

0.000

(0.005)

0.010**

(0.006)

Year 2001

0.010***

(0.004)

0.007

(0.007)

0.009

(0.006)

Year 2002

0.010**

(0.005)

0.000

(0.008)

0.012**

(0.007)

Observations/Industries 1602/472 799/314 803/319

F value 3.35 2.41 4.83

Within R 2 0.0432 0.0715 0.1338

Between R 2 0.0146 0.0124 0.0004

Overall R 2 0.0035 0.0025 0.0009

Note: Values in parentheses are standard errors. *** p < 0.01; ** p < 0.05; * p < 0.10.

40

More magazines by this user
Similar magazines