Placing Technological Innovation in Globalising China: Production ...

48(14) 2999–3018, November 2011

Placing Technological Innovation in Globalising China: Production Linkage, Knowledge Exchange and Innovative Performance of the ICT Industry in a Developing Economy George C. S. Lin, Cassandra C. Wang, Yu Zhou, Yifei Sun and Yehua Dennis Wei [Paper first received, January 2010; in final form, November 2010]

Abstract This study critically examines the relevance of the perceived notions of localised production linkages, knowledge spillover and external technology transfer to the experiences of the growth of the ICT industry in China. The research is based on a major firm-level survey conducted in China’s three most important mega urban regions—Beijing, Shanghai-Suzhou and Shenzhen-Dongguan—where the bulk of the Chinese ICT industry is located. The results of the survey showed a distinct landscape of ICT industrial production in which each of the Chinese regions has functioned as the site of capital investment from different sources for different strategic interests. Despite a marked regional variation in ownership, industrial structure, market orientation and technological investment, firms in all regions have invariably reported internal development as the main source of core technology. A negative relationship existed between the level of technological innovation and external orientation in both capital investment and export production. No evidence has been found to verify the hypothesis that a higher level of technological innovation would co-exist with stronger production linkages and knowledge exchanges with both local firms and foreigninvested enterprises. A further analysis of the firms with different technological performance has highlighted the significance of regional setting, ownership, ability of capital mobilisation and corporate strategy and management in the process of technological innovation. George C. S. Lin is in the Department of Geography, The University of Hong Kong, Pokfulam Road, Hong Kong. E-mail: [email protected]. Cassandra C. Wang (corresponding author) is in the Department of Earth Sciences, Zhejiang University, 38 Zheda Road, Hangzhou, China, 310027. E-mail: [email protected]. Yu Zhou is in the Depertment of Earth Science and Geography, Vassar College, Box 465, Poughkeepsie, New York, NY 12604 0465, USA. E-mail: [email protected]. Yifei Sun is in the Department of Geography, California State University, Northridge, 18111 Nordhoff St, Northridge, California, 91330, USA. E-mail: [email protected]. Yehua Dennis Wei is in the Department of Geography, University of Utah, 260 Central Campus Drive, Salt Lake City, Utah, 84112, USA. E-mail: [email protected]. 0042-0980 Print/1360-063X Online © 2011 Urban Studies Journal Limited DOI: 10.1177/0042098010396232

3000 GEORGE C. S. LIN ET AL.

Introduction In recent decades, the importance of technological innovation to sustained regional economic growth in the context of intensified global competition has received great scholarly attention. Ever since Schumpeter (1934) brought up technological innovation as one of the basic factors for economic development, the dynamism of innovation has never ceased to intrigue researchers (Fagerberg et al., 2005). Thus far, much of the scholarly effort has been devoted towards understanding the relationship between technological innovation and regional economic development. In contrast, relatively less is known about the internal mechanism of innovation. As Fagerberg et al. have observed In spite of the large amount of research in this area during the past fifty years, we know much less about why and how innovation occurs than what it leads to (Fagerberg et al., 2005, p. 20).

Existing research on the process of inno vation has initially been concerned with interfirm linkages which are believed to be central to collective learning and know ledge spillover in an industrial cluster. The dynamics of technological innovation in the developing world have been treated separately because technological upgrading in developing countries has often been seen as dependent upon technology transfer from the developed world. While the notions of localised knowledge spillover and external technology transfer have contributed to the understanding of the dynamism of innovation in the developed and developing world, there exist important theoretical issues that require clarification and further investigation. The two popular notions of knowledge spillover and technology transfer have both been centred around the production and diffusion of knowledge and technology. They tend to overlook other important factors

that affect the formulation of strategies by individual firms to invest and engage in innovative activities. This paper examines the growth dynamics and regional variation of China’s ICT industry based on the results of a major firm-level and cross-regional questionnaire survey conducted in 2006-07 in Beijing, ShanghaiSuzhou and Shenzhen-Dongguan—the three most important city-regions that harbour the bulk of the ICT industry in the country. Against the backdrop of the continuing theoretical debate over localised knowledge spillover and external technology transfer, the main concern of this research is with the ways in which localised interaction and external connection with foreign-invested enterprises (FIEs) have contributed to the pursuits, practices and performance of technological innovation of the individual firms in different regions. The remainder of this paper is organised in four parts. It begins with a critical evaluation of the competing interpretations of the dynamism of technological innovation in developing countries. This is followed by a clarification of our own research design and methodology. Attention is then turned to the pattern and processes of industrial clustering and technological innovation identified from the three key city-regions. A comparative study is conducted for the firms with different innovative performance to probe further into the factors that explain why certain firms have turned out to be more innovative than others. Important findings of the research are highlighted and their theoretical implications discussed in the final section.

Understanding the Dynamics of Technological Innovation in Developing Countries: Localised Knowledge Spillover or External Technology Transfer? Over the past two decades, the dynamics of uneven regional industrial growth in the

ICT INDUSTRY IN CHINA 3001

context of intensified global competition have been the subject of extensive documentation and heated debates. A great amount of ink has been spilled to shed light over why certain regions are able to maintain and enhance economic growth while others have suffered from a decline. Whereas the earlier approach utilised the concept of comparative advantages and focused on an optimal provision of key production factors, the current intellectual trend, popularised by such concepts as the ‘new economic geography’ and ‘new regionalism’, is to emphasise the formation of competitive advantages and industrial clustering in which technological innovation and knowledge production are believed to be the key to winning intensified global competition. Among many other things, it is widely recognised that co-location of related firms could facilitate co-operation and competition, produce mutual trust, stimulate collective learning and bring about localised knowledge spillover, all of which will contribute to the growth of a regional knowledge-based economy (Fan and Scott, 2003; Porter, 1990). For the dynamics of technological innovation in advanced economies, much of the scholarly attention has been directed towards the concept of industrial clusters and localised knowledge spillover because it is generally believed that innovation depends to a large extent on the process of knowledge acquisition and accumulation (Döring and Schnellenbach, 2006). Influenced by the endogenous growth theory, the prevailing view is that the characteristics of partial excludability and non-rival knowledge allow firms to benefit from the spilled knowledge without paying any fees (Fischer, 2006, p. 101). It has been recognised that the knowledgebase of individual firms is limited and that external heterogeneous knowledge is an important complement for a firm to achieve technological innovation. Localised knowledge spillover could cut down the cost of knowledge search and scientific discovery

and hence to a large extent reduce the risk resulting from the uncertainty of innovation (Audretsch and Feldman, 2003). The localised knowledge spillover resulting from the clustering of firms is therefore identified as a major driving force for technological innovation and regional growth (Kesidou and Romijn, 2008). For technological innovation in the developing world, a plethora of studies exists to highlight the importance for developing countries to tap in and gain access to international knowledge production as a means to enhance their competitiveness in the global economy (Kesidou and Romijn, 2008; Lin, 2009). The role played by multinational corporations and the effects of technology transfer have long preoccupied the research agenda concerning technological development in the developing countries that are seen as ‘late-comers’ in the process of innovation (Kim and Nelson, 2000). Despite the popular notion that MNCs have little incentive to transfer their core technology to the developing countries, recent research has shown that successful technological catch-up could be achieved if developing countries adopted sound policies towards investment in human capital, exports and industrial development (Lin and Wang, 2009; Wei et al., 2011; Zhou, 2008; Zhou et al., 2011). While the theories of localised knowledge spillover and technology transfer have shed significant light over the dynamics of technological innovation in developed and developing economies, a critical evaluation would identify a number of conceptual and methodological issues that require further clarification and interrogation. First, the two conceptual frameworks have both stressed the importance of interactions among the firms within or outwith a region while paying little attention to the nature, attributes and characteristics of the firms themselves as active agents and actors in the process of technological innovation. Recent studies have shown


that the nature and attributes of firms (size, ownership, market orientation, etc.) have implications for technological innovation no less significant than the regional environment that surrounds the firms (Beugelsdijk, 2007). It has also been observed that the success or failure of technological innovation of a firm is, to a great extent, dependent upon the firm’s capability to find a coupling between its production and a niche in the market at which the firm has its competitive advantages (Brown and Fai, 2006). Hence, attention should be directed more to the functioning of the firm, its internal organisation and its positionality in global production networks and commodity chains (Humphrey and Schmitz, 2002; Bathelt et al., 2004; Gereffi et al., 2005; Zhou et al., 2011). Secondly, the theories of knowledge spill over and technology transfer have shared a common emphasis placed on the production, exchange and diffusion of knowledge and technology without taking into consideration the role played by capital mobilisation to finance research and development. A precondition for innovation is a long-term financial commitment by which many firms have tended to be easily intimidated. The financial constraints upon firm-level technological innovation have been highlighted in a number of studies (Hyytinen and Toivanen, 2005). Finally, the two notions of knowledge spillover and technology transfer have tended to assume that all firms in a cluster are tempted by the benefits of innovation and would therefore like to engage in research and development activities. Explanations for why some firms are more innovative than others have been made in terms of the different capability of the firms to gain access to knowledge without paying adequate attention to the motivations of firms and the opportunity costs involved. In addition to financial constraints, there are many strategic considerations under which firms are reluctant to enter the innovation cycle (Meeus and Oerlemans, 2005).

It is not unusual to see that the return from imitation becomes higher and certainly safer than direct investment in innovation because the latter is plagued with risks and uncertainty and could easily end up with market failure (Christensen, 1997). The controversy and limitations of the extant literature identified warrant further investigations of the actual practices in different world regions. Against this conceptual and theoretical backdrop, this study is conducted to examine several important issues that are highly relevant to on-going debates and enquiries concerning the dynamics of technological innovation. What are the structural and spatial characteristics of China’s ICT industry? How innovative are the Chinese ICT firms and how does the innovative performance of these firms vary from region to region? Why are some ICT firms in some places more innovative than others in other places? To what extent does localised know ledge spillover contribute to the innovative performance of the firms and of the regions? To what extent do the linkages with foreigninvested enterprises contribute to technology transfer and hence better innovative performance of the firms and of the region? Are there other important factors that have significantly affected different innovative performance among firms and among regions, and what are they?

Data and Methodology The purpose of this study is essentially to probe into the internal dynamics of the growth of China’s ICT industry in different regional contexts. To accomplish this objective and to address the specific issues raised in the foregoing section, two working hypotheses are made. First, there exists a significant and positive relationship between the extent of localised interactions among firms within a region and the innovative performance of the firms. Secondly, there exists a significant and


Figure 1. The ICT manufacturing sector in China’s three mega-urban regions, 2007. Note: This map is derived from the calculation of the data for the above-scale ICT manufacturing enterprises which include all state-owned enterprises and those non-state-owned enterprises whose annual sales revenue is larger than 5 million yuan. Source: MII (2008).

positive relationship between the extent of interactions with foreign-invested enterprises (FIEs) and the innovative performance of Chinese ICT firms.1 The data used in this research are derived from a major firm-level questionnaire survey we conducted during 2006–07 in the three most important mega urban regions in China—namely, Beijing, Shanghai-Suzhou and Shenzhen-Dongguan. These three regions occupy a tiny portion of the total territory of China, but they contributed

more than half of China’s ICT manufacturing employment and nearly 60 per cent of the national manufacturing output and export in 2004, the year when the first national economic census was conducted. The latest statistics show that Beijing, Shanghai, Jiangsu and Guangdong contributed nearly 70 per cent of China’s ICT manufacturing employment and produced 72 per cent of the national manufacturing output and 80 per cent of export in 2007 (Figure 1; MII, 2008). Our sample was drawn from the database


developed by China’s State Statistical Bureau (CSSB) on the basis of the 2004 first national economic census. The sample size was predetermined with the target of a 5 per cent sampling rate based on the national database. We then proportioned the sample size based on the sectorial and regional information obtained from the national database and made necessary adjustments to ensure a sample size larger than 30 for all sub-categories except semi-conductors. We designed our structured questionnaires in accordance with our research questions, hypotheses and objectives. Two sets of questionnaires were made: one for hardware manufacturing (manufacturing of computer/communication equipment, code nos SIC 401 and 404; semi-conductors, code nos SIC 4052 and 4053 and electronic parts, code nos 4051 and 406) and the other for software design (code no. SIC 62). The questionnaires were then administered by a professional survey company which is an offshoot from and still maintains a close affiliation with the China State Statistical Bureau. The survey of Shenzhen and Dongguan was conducted as a pilot during October and November 2006, and that was followed by subsequent surveys of the sampled firms located in Beijing, Shanghai and Suzhou in the spring of 2007. A total of 1023 valid responses to the questionnaires were received, including responses from 633 hardware firms and 390 software companies. Our analysis takes two steps, starting with an identification of the diverse regional trajectories of technological development and then probing further into the firm-level dynamism of technological innovation. A common objective that runs through our analysis at both regional and firm levels is to interrogate how localised knowledge spillover and connections with foreign-invested enterprises have contributed to the technological innovation of the firms and regions in a rapidly developing economy.

A Diverse and Deviant Geography of ICT Production: Technological Innovation, Localised Production Networks and External Technology Transfer The ICT industrial landscape emerging in China is characterised not simply by a remarkable regional variation that could be easily anticipated but, more importantly, by different matching between the interests of the Chinese regions and FIEs, different strategies and approaches adopted towards R&D and, ultimately, a different dynamism of technological innovation. First, each region has formed its own kind of public–private partnership and has developed its own linkages with different kinds of foreign investor. In an alternative perspective of the multinationals, each of these Chinese regions has been placed in a different position on the agenda of global accumulation by dispossession. Figure 2 summarises the ownership structure of the ICT firms we have surveyed from the three regions. Beijing is dominated by domestic capital and has the highest presence of individual firms that are privately owned (getijingji). Zhongguancun—China’s Silicon Valley—has been the birthplace of non-governmental and privately owned high-tech firms, many of which are off-shoots from state-funded research and educational institutions (Zhou, 2008). The Shanghai-Suzhou mega urban region has the highest proportion of firms chosen by multinational corporations as their branches in China (Sun, 2002; Lin and Wang, 2009; Wei et al., 2011). The ShenzhenDongguan region, as the prime production site of outsourcing from Hong Kong, has the highest percentage of firms established and owned by Hong Kong and Taiwan (Wang and Lin, 2008; Wang et al., 2010). In other words, these three regions have been situated differently on the national and international agenda of ICT industrial development—Beijing as the national hub of R&D, Shanghai-Suzhou

ICT INDUSTRY IN CHINA 3005 100%

80% Others HK/TW owned enterprises

60%

Foreign-owned enterprises Private-owned enterprises State-owned enterprises

40%

20%

0% Beijing

Shanghai-Suzhou

Shenzhen-Dongguan

Figure 2. Ownership type of the surveyed firms, by regions, 2007. Notes: state-owned enterprises are non-corporate economic entities registered in accordance with the Regulation of the People’s Republic of China on the Management of Registration of Legal Enterprises, where all assets are owned by the state; private-owned enterprises refer to economic units invested or controlled by natural persons who hire workers for profit-making activities; foreign-owned enterprises refers to all industrial enterprises registered as jointventure, co-operative, sole (exclusive) investment industrial enterprises and limited liability corporations with foreign funds; HK/TW-owned enterprises refer to all industrial enterprises registered as the joint-venture, co-operative, sole (exclusive) investment industrial enterprises and limited liability corporations with funds from Hong Kong and Taiwan (see CSSB, 2008, pp. 546–547). Source: authors’ survey.

as a branch of foreign technological operations in China and Shenzhen-Dongguan as the production site to accommodate those relocated from Hong Kong (Zhou et al., 2011). The results of our questionnaire survey in the three regions have revealed different regional strategies and approaches dealing with R&D. Table 1 summarises the input in R&D by employment and expenditure among the three regions. The firms we surveyed in Beijing reported the highest proportion of the labour force and highest share of expenditure devoted to R&D. Firms in Shanghai-Suzhou took the second place and those from Shenzhen-Dongguan were third. Obviously, firms in these three regions have been positioned differently in the global

pipeline of ICT industrial production and technological innovation. Despite their different relationship with foreign and domestic enterprises, all the firms we surveyed in the three regions have provided a similar response when they were asked about where they had obtained their core technology. Internal development has been identified by all surveyed firms across the three regions (95 per cent in Beijing and over 60 per cent in the other two regions) as the most important source of core technology although the degree of importance varied slightly from region to region. The dominance of internal development in technological innovation has been outstanding for the firms in Beijing. By comparison, imports of foreign technology


Table 1. R&D input of the surveyed firms, by regions

Beijing Shanghai-Suzhou Shenzhen-Dongguan

Percentage share of R&D employment in total employment Minimum Maximum Mean S.D. N

0 0 80 100 29.67 12.9 23.54 19.37 100 260

0 67 4.51 6.67 266

Percentage of R&D expenditure among total expenditure Minimum 0 0 Maximum 90 95 Mean 32.05 11.96 S.D. 21.93 12.78 N

92 262

0 60 11.51 12.23 242

R&D worker’s monthly income (yuan) Minimum Maximum Mean S.D. N

1000 1000 8000 9500 4518.48 3212.56 1372.25 1201.31 92 207

1500 10000 3846.61 1564.93 191

Source: authors’ survey.

with or without modification through use have played some important role in the development of core technology for the firms in the Shanghai-Suzhou and Shenzhen-Dongguan mega urban regions. For China to move along the path of technological upgrading, the key will still be indigenous R&D rather than external technology. It is interesting to note from the responses to our survey that universities and research institutions were not identified as an important source of core technology, even in Beijing where China’s most prestigious universities and R&D institutes are clustered. Other than training high-tech talents, Chinese universities and research institutions do not seem to have contributed what they

should to the technological upgrading of the ICT industry. How then did the firms of different regions differentiate among themselves in technological innovation, and why? Table 2 shows the results of responses from the firms in the three regions to our questions about the performance of innovation in two indicators—namely, share of new products in the total sales revenue and percentage of firms with one or more granted invention patents. In both indicators, firms in Beijing reported a record significantly ahead of those in the other two regions. The one-way ANOVA test revealed that there exists a statistically significant regional difference in the share of


Table 2. Innovative performance by the share of new products out of total sales revenue and by invention patents

Beijing Shanghai-Suzhou Shenzhen-Dongguan

Percentage share of new products out of total sales revenue Minimum 0 0 Maximum 100 100 Mean 40.84 14.21 S.D. 35.80 23.10

0 100 16.45 25.31

N

89 263

214

Percentage of firms with granted invention patents

38 20

14

N

99 256

263


new products out of total sales revenue among these three mega urban regions (F = 36.074, P = 0.000). On average, the new products developed by firms in Beijing contributed 41 per cent of total sales revenue, significantly higher than those in Shanghai-Suzhou (14 per cent) and Shenzhen-Dongguan (16 per cent). This is not, of course, surprising given the fact that the firms in Beijing committed a higher input (employment and expenditure) than those in the other two regions in R&D as reported earlier. A higher input in innovative activities should naturally give rise to a better performance. What is interesting, however, is a peculiar relationship, or a mis-match, between the level of technological innovation on one side and the extent of localised production networks, technological linkages among firms within the region and linkages with foreign-invested enterprises (FIEs) on the other side.2 To test our hypotheses concerning the importance of localised production networks, knowledge spillover and linkages with FIEs to the regional performance of technological innovation, we have included relevant questions in our questionnaires and

the responses have shown a rather intriguing pattern that deviates from our original expectation. Table 3 shows the results of our questionnaire survey of the ICT firms in the three mega urban regions concerning the extent of their production linkages with local firms and FIEs. Although the firms in Beijing outperformed their counterparts in the other regions in technological innovation, they reported the least extent of local production networking and linkages with FIEs. For example, the share of local purchases reported by the firms in Beijing was only 37 per cent, significantly lower than what was reported by those in Shanghai-Suzhou (47 per cent) and those in Shenzhen-Dongguan (51 per cent). A similar pattern is observed in the percentage of local customers, percentage of sales revenue from local sub-contractors and percentage of sales revenue from local contractors. Responses to our questions concerning the importance of production linkages with FIEs have shown almost an identical pattern. Firms in Beijing with the highest level of innovation among the three regions reported the least share of FIE customers, percentage of imported parts,


Table 3. Results of ANOVA test on production relations with local firms and FIEs Mean Shanghai- Shenzhen- Significance Indicators Beijing Suzhou Dongguan level of ANOVA Percentage share of local purchases (2 hrs) in domestic purchases Percentage of local customersa Percentage of sales revenue from local sub-contractors Percentage of sales revenue from local contractors Share of FIE customers Percentage of imported parts Share of domestic purchases from FIEs Share of FIEs from major parts purchase Percentage of sales from foreign contractors

37.15 46.62 51.30 0.003**

2.47 3.15 2.85 0.000** 38.34 50.37 58.22 0.021* 51 55.87 63.96 0.168 7.07 32.52 29.86 0.000** 25.51 42.74 35.89 0.000** 25.23 41.18 55.43 0.000** 34.71

42.77

35.43

0.042*

5.77 15.97 18.80 0.001**

Percentage of local customers is not measured by the actual percentage but by the ratio interval: 1: 0; 2: 1–25; 3: 26–50; 4: 51–75; 5: 76–100. a

Notes: ** the mean difference is significant at the 0.01 level; * the mean difference is significant at the 0.05 level. Source: authors’ survey.

share of domestic purchases from FIEs, share of FIEs from major parts purchases and percentage of sales from foreign contractors (Table 3). It appears that the results of our survey did not provide any convincing evidence to verify the two hypotheses we made on the basis of the conventional wisdom of localised production networks. If localised production linkages with domestic and foreign firms did not appear to be the factor explaining regional variation in technological innovation, did localised knowledge exchange contribute to a better technological performance? An analysis of the results of our questionnaire survey did not provide any evidence strong enough to suggest a positive and decisive relationship.

Tables 4 and 5 summarise the results of our survey concerning technological linkages with local and foreign firms within the same region. Despite some regional variation, the majority of the firms across the three regions reported the formation of partnerships or co-operation with domestic and foreign firms in the process of technological development as either non-existent or unimportant. The majority of these firms also reported no or very little technology transfer or seeking technology advice from domestic and foreign firms within the region. This pattern is consistent with the one identified in the foregoing section in which internal development was reported as the main source of core technology. It is also consistent with the


Table 4. Technology linkages with local firms

Beijing

Shanghai-Suzhou Shenzhen-Dongguan

Frequency Percentage Frequency Percentage Frequency Percentage

Importance of domestic alliance in technology development None or not 73 73.3 189 70.8 221 83.1 important OK 4 4.3 28 10.5 15 5.6 Important 7 7.3 40 15.0 24 9.0 Very important 15 15.3 10 3.7 6 2.3 Total

99 100.0 267 100.0 266 100.0

Importance of domestic co-operation in technology development None or not 68 71.6 178 66.7 216 81.2 important OK 1 1.1 32 12.0 11 4.1 Important 5 5.3 44 16.5 31 11.7 Very important 21 22.1 13 4.9 8 3.0 Total

95 100.0 267 100.0 266 100.0

Frequency of domestic technology transfer None or very few 86 86.0 208 77.9 250 84 A few 8 8.0 45 16.9 14 5.3 Frequent 5 5.0 14 5.2 2 .8 Very frequent 1 1.0 0 0 0 .0 Total

100 100.0 267 100.0 266 100.0

Frequency of technology advice from domestic enterprises None or very few 58 58.0 175 65.6 212 80 A few 8 8.0 68 25.5 40 15.1 Frequent 28 28.0 23 8.6 11 4.2 Very frequent 6 6.0 1 .4 2 .8 Total

100 100.0 267 100.0 265 100.0


findings of several recent studies (Sun, 2002; Wang and Lin, 2008; Zhou et al., 2011). It does not lend any support, however, to the hypotheses we made in line with the popularly perceived notions of localised production network and localised know ledge spillover.

Firm-level Technological Dynamism: Production Linkages, Knowledge Exchange or Other Factors? The analysis in the foregoing section aggregated firms at a regional level for


Table 5. Technology linkages with foreign firms

Beijing

Shanghai-Suzhou Shenzhen-Dongguan

Frequency Percentage Frequency Percentage Frequency Percentage

Importance of foreign alliance in technology development None or not 83 83.0 204 76.9 223 83.8 important OK 1 1.0 22 8.3 9 3.4 Important 13 13.0 25 9.4 25 9.4 Very important 3 3.0 14 5.3 9 3.4 Total

100 100.0 265 100.0 266 100.0

Importance of foreign co-operation in technology development None or not 79 81.4 186 69.6 211 79.3 important OK 0 0 24 9.0 12 4.5 Important 1 1.0 45 16.9 30 11.3 Very important 17 17.5 12 4.5 13 4.9 Total

97 100.0 267 100.0 266 100.0

Frequency of foreign technology transfer None or very few 91 91.0 223 83.5 247 92.8 A few 4 4.0 30 11.2 14 5.3 Frequent 3 3.0 11 4.1 5 1.9 Very frequent 2 2.0 3 1.1 0 0 Total

100 100.0 267 100.0 266 100.0

Frequency of technology advice from foreign enterprises None or very few 70 70.0 189 70.8 220 82.7 A few 3 3.0 50 18.7 32 12.0 Frequent 22 22.0 24 9.0 11 4.1 Very frequent 5 5.0 4 1.5 3 1.1 Total

100 100.0 267 100.0 266 100.0


comparison. The results only demonstrated the importance of a regional political and institutional environment, rather than localised production networks or external technology transfer, to technological innovation in China. Such a regional comparison needs to be cross-checked with and supplemented by a firm-level analysis. To do this, we divided our sampled firms into the two groups of

innovative and non-innovative firms using the number of granted invention patents as a yardstick. Innovative firms are those that have been granted at least one invention patent. We then systematically compared these two groups of firms in terms of their extent of production linkages and knowledge exchange with both local firms and FIEs. The results of comparison are listed in Tables 6 and 7.


Table 6. T-test results on the relations with local firms between innovators and non-innovators Mean

Non-innovators Innovators

T-value P-value

Production linkages Share of local purchases (2 hrs) in domestic purchases Percentage of local customersa Percentage of local sub-contractors in 2 hrs Percentage of local contractors

47.88

45.14

0.785

0.433

2.92 55.62

2.95 –0.281 0.779 43.58 2.205* 0.030

60.72

51.78 1.412 0.160

Knowledge exchangeb Importance of domestic alliance in technology development Importance of domestic co-operation in technology development Frequency of technology transfer from domestic firms Frequency of technology advice from domestic firms Frequency of personnel exchange with domestic firms Frequency of information exchange with domestic firms

0.69

1.22

–4.021**

0.000

0.79

1.49 –4.886** 0.000

0.45

0.83 –5.018** 0.000

0.71

1.34 –5.436** 0.000

0.80

1.52 –5.567** 0.000

0.77

1.61 –5.971** 0.000

Percentage of local customers is not measured by the actual percentage but by the ratio interval: 1: 0; 2: 1–25; 3: 26–50; 4: 51–75; 5: 76–100. a

The importance or frequency is ranked from 0 to 4, with 0 meaning no such linkages and 4 referring to most important or most frequent. b

Notes: * the mean difference is at the 0.05 significance level; ** the mean difference is at the 0.01 significance level. Source: authors’ survey.

Firms in the two groups reported a different extent of production linkages with local firms as indicated in the different share of local purchases in domestic purchases, percentage of local customers, percentage of local contractors and local sub-contractors. These differences are not, however, statistically significant according to the results of the T-test. The only exception lies in the percentage of local sub-contractors in which the two groups of firms showed a difference that is statistically significant. Even

here, the result of comparison took us by surprise: non-innovative firms reported a percentage of local sub-contractors higher than that for the innovative firms. In fact, non-innovative firms reported an extent of local production linkages slightly higher than that for the innovative firms in three of the four indicators (Table 6). This peculiar pattern suggested that the local production linkages built by these firms had little to do with innovative activities and that innovative firms have managed to obtained their


Table 7. T-test results on relations with FIEs between innovators and non-innovators Mean


T-value P-value

Production linkages Percentage share of foreign customers Percentage of imported parts Share of domestic purchases from FIEs in China Share of FIEs in key parts purchases Percentage of sales from foreign contractors

28.92 36.58 45.35

22.21 2.248** 0.025 37.48 –0.275 0.783 41.61 1.187 0.236

38.21 15.84

37.87 0.091 0.928 13.68 0.707 0.480

Knowledge exchangea Importance of foreign alliance in technology development Importance of foreign co-operation in technology development Frequency of foreign technology transfer Frequency of technology advice from foreign firms Frequency of personnel exchange with foreign firms Frequency of information exchange with foreign firms

0.63 1.04 –3.049** 0.003 0.76

1.36 –4.225** 0.000

0.40 0.64

0.77 –4.274** 0.000 1.16 –4.435** 0.000

0.71

1.28 –4.568** 0.000

0.69

1.43 –5.436** 0.000

The importance or frequency is ranked from 0 to 4, with 0 meaning no such linkages and 4 referring to most important or most frequent. a

Notes: * the mean difference is at the 0.05 significance level; ** the mean difference is at the 0.01 significance level. Source: authors’ survey.

granted invention patents by means other than local production linkages. If local production linkages cannot explain why some firms turned out to be more innovative than others, can we turn to the factor of knowledge exchange as a possible source of innovation? Here, the results of comparison appear to be closer to normal theoretical expectation, but overall the evidence found looks rather weak. As already presented in the foregoing section, the overwhelming majority of the firms we surveyed reported either the formation of partnerships, co-operation and exchange with other local firms in the process

of technological innovation either not to exist or to be unimportant. This unenthusiastic response applied to the majority of firms in all regions and it did not change when the firms were reclassified into innovative or non-innovative. Nonetheless, innovative firms did show a higher frequency than the non-innovative firms in giving an ‘importance’ response to our questions concerning knowledge exchange with other local firms (Table 6). Although an overwhelming majority of the firms, innovative or non-innovative, did not consider localised knowledge exchange to be important to their technological development, innovative


firms appear to have a stronger desire and more positive attitude towards knowledge exchange than the non-innovative ones. A further analysis of production linkages and knowledge exchange with FIEs has identified a pattern similar to the one already discussed. As shown in Table 7, innovative and non-innovative firms reported different extents of production linkages with FIEs, but these differences are not statistically significant. The share of foreign customers is the only area that sets the two groups apart far enough to warrant our attention. However, non-innovative firms actually reported a higher share of foreign customers than their innovative counterparts, which deviated from our original expectation. No evidence is found here to link stronger production linkages with FIEs with a better performance of the firms in technological innovation. As for the factor of knowledge exchange with FIEs, innovative firms gave a more positive response than noninnovative firms to our questions concerning the importance of forging partnerships, cooperation and technology transfer with FIEs. However, the majority of firms, regardless of where they are located or how innovative they are, did not consider knowledge exchange with FIEs to be an important factor contributing to firm-level technological innovation (see Table 5). This pattern is consistent with the findings of several recent studies that the core technology of Chinese ICT firms comes primarily from R&D within the firm and that a stronger presence of FIEs did not lead to a higher level of technological innovation in Chinese regions (Sun, 2002; Zhou et al., 2011). If production linkages and knowledge exchange with local firms or FIEs did not seem to be significant enough to explain why some firms are more innovative than others, what then are the factors that really make a difference in the process of technological innovation for the Chinese ICT firms? We probed further into the responses to our questionnaires and have obtained some important

insights into the special dynamism of technological innovation in the Chinese anomaly. Table 8 compares the internal characteristics of innovative and non-innovative firms. Our analysis focuses on those attributes that are considered to be statistically significant according to the results of the T-test. First, the regional setting appears to make a significant difference to the innovative performance of the firms. A chi-squared test confirmed that there exists a significant regional difference between innovative and non-innovative firms among these three mega urban regions (chi-squared = 26.061, P = 0.000). Not surprisingly, Beijing—the national hub of China’s high-tech industry—hosted the largest share of innovative firms (38 per cent). This was followed by the Shanghai-Suzhou mega urban region (20 per cent) and Shenzhen-Dongguan corridor of export production (14 per cent). Although there is little ‘local buzz’ through which localised production networks and knowledge spillover would supposedly bring about innovation for Chinese firms, a ‘local base’ with a rich accumulation of human resources, an innovative cultural tradition and institutional support such as that characteristic of Beijing’s Zhongguancun—China’s Silicon Valley—have clearly been conducive to technological innovation. Given the special nature of the Chinese political economy, the strategic position held by the region in the national economy and its coupling with the interests of the state have also been instrumental in the incubation and growth of innovative activities. Conversely, a regional setting dominated by FIEs and functioning simply as an outlet of labour-intensive manufacturing in the global pipeline has proved to be less favourable to technological innovation (Sun, 2002; Wei et al., 2011; Zhou et al., 2011). Secondly, ownership of the firms is found to be an important factor setting the innovative firms apart from the non-innovative ones (chi-squared = 14.919, p = 0.037). Although


the Chinese economy has undergone profound market transition through which the private sector and FIEs have been provided with greater space to grow, key projects with strategic national interests have remained owned and directly controlled by the state. As a consequence, state-owned enterprises (SOEs) have been found to outperform other firms in technological innovation at this particular historical juncture. Among the firms that responded to our survey, those owned by the state reported the highest share of new products out of the total sales revenue (39.4 per cent). The one-way ANOVA test revealed that there exists significant difference in the share of new products in the total sales revenue among ownership types of the firms. We use the Scheffe test further to examine the pattern and the results show that state-owned enterprises have produced a share of new products in total sales revenue significantly higher than that generated by either the subsidiaries of foreign-owned enterprises or the firms owned by the Chinese industrialists in Hong Kong and Taiwan. In the case of the US, it has been well documented that federal spending in the military industry in the 1960s was instrumental in the growth of the risky high-tech sector. In the case of China, government support has been equally instrumental, if not more so, to the innovative performance of the ICT firms, and investment in the SOEs has clearly been a direct support of the government. This does not mean that the private sector and FIEs will lag behind forever. As the Chinese economy continues to ‘grow out of the plan’, it is foreseeable that the non-state sector will play a growing part in the ICT industrial production and innovative activities. Thirdly, the ability of the firms to mobilise capital, including venture capital and floating assets, stands out as another important factor explaining why some firms are more innovative than others. Of the two groups of firms, those with a better performance in innovation were able to mobilise capital and pay their

workers, particularly R&D workers, a monthly salary significantly higher than that paid by the non-innovative firms (Table 8). Given the fact that innovative activities are usually long-term and risky investments that may not bring about instant returns, the mobilisation of venture capital has naturally been crucial to the pursuit of technological innovation. In the case of China, capital mobilisation of the firms has remained highly dependent upon the support of the central and local governments who control the banking sector. It is therefore not surprising to see that those firms with a better economic and financial partnership with the central or local government have done a better job in technological innovation. For innovative firms, the share of government purchase in their total sales (13 per cent) has been significantly higher than that for the non-innovative firms (4 per cent). This pattern is consistent with the one identified from ownership. Finally, the determination of the firms to embark upon the long and risky journey of R&D has made a significant difference in bringing about innovation. A comparison of the two groups of firms suggests that the innovative firms have allocated a portion of capital and personnel to R&D activities significantly higher than that committed by the non-innovative firms (Table 8). Moreover, innovative firms devoted a larger share of workers for marketing purposes and managed to occupy a larger share of the top tier of the national market for their core products. For instance, over 35 per cent of the innovative firms were able to occupy the top five positions in the national market for their core products, whereas only 16 per cent of the innovative firms were able to do this. This pattern suggests that the corporate strategies and managerial skills of the firms have obviously played a role no less important, if not greater, than such external forces as production linkages or knowledge spillover in the process of technological innovation.


Table 8. T-test results on comprehensive strengths and R&D strategies between innovators and non-innovators Mean


T-value P-value

Profit in 2005a 1.96 2.04 –0.832 0.44 Profit in 2004a 2.18 2.18 –0.005 0.996 Profit in 2003a 2.24 2.24 0.078 0.938 Year of establishment 1998 1998 –1.134 0.257 Total assets (million yuan) 118.28 611.30 0.958 0.340 Employment (persons) 492 1686 1.587 0.115 Workers monthly income (yuan) 1611 2057 4.888** 0.000 R&D workers monthly income (yuan) 3621 4291 3.015** 0.003 Percentage of R&D expenditure 13 20 4.012** 0.000 among total expenditure Percentage share of R&D workers 12 19 2.747** 0.006 in total employees Percentage share of professional 14.20 18.21 –2.877** 0.005 managers in total employees Percentage share of marketing workers 7.61 13.91 –4.841** 0.000 in total employees Rank of market share of core productsb 2.6 2.1 4.934** 0.000 Government purchase as percentage 3.62 12.98 –3.963** 0.000 of total sales of the firm a

Profit is measured by several intervals: 1: profit margin is larger than 10, 2: profit margin is 5–10, 3: profit margin is 1–4, 4: profit margin is smaller than zero. b

1: top 5; 2: top 6–10; 3: below top 10.

Note: ** p value is at 0.01 level. Source: authors’ survey.

Conclusion and Discussion One of the controversial issues that has attracted considerable attention from researchers and policy-makers in recent years has been the dynamism of technological innovation in the process of sustained urban and regional development in the era of intensified global competition. After a brief episode of fascination with the seemingly irresistible global market forces, the new intellectual trend is to turn to some localised production networks that have embedded and glued footloose economic activities in an increasingly slippery space. The prevailing view is

that the competitive advantages of cities and regions rest upon the cultivation of some dense localised production networks, mechanisms of knowledge spillover and industrial clustering which are believed to be conducive to technological innovation and sustained economic growth. Recent theoretical attempts to connect ‘local buzz’ with ‘global pipeline’ and evaluate the ‘strategic coupling’ of the imperatives of global capital accumulation with local assets and conditions have provided new and useful insights (Bathelt et al., 2004; Yeung, 2009; Zhou et al., 2011; Wang et al., 2010). It remains unclear whether or not and how actual practices in different world regions


have played out of the logics of production networks. This study investigates the patterns and processes of industrial production and technological innovation in the ICT manufacturing sector in China. Our analyses of the data gathered from a large survey of the firms in the three major mega urban regions of the country have identified a distinct landscape of industrial production and technological innovation characterised by both various regional trajectories of growth and a peculiar dynamism of innovation that deviates from normal theoretical expectation. Each of the three regions in our study has been developed into a site of industrial production by different kinds of capital for different purposes. Beijing is characterised by the dominance of domestic capital, strong enterprise-government connections and orientation to the domestic market. The Shanghai-Suzhou region has been chosen by multinational corporations to develop their branches in China. The Shenzhen-Dongguan region has been turned into an outlet of production outsourcing by the Hong Kong industrialists who share a common cultural root and enjoy geographical proximity. Instead of a singular territorialisation of global capital in China or a sequential movement of technological diffusion from one region to another, what we have found is essentially a juxtaposition of multiple and different kinds of capital territorialisation taking place at the same time but in different locales, each involving different actors and agents with different strategic interests and hence different dynamism of industrial production and innovation. Despite a marked regional variation in industrial structure, ownership, market orientation and technological development, firms in all regions have invariably identified internal R&D as the main source of core technology. A cross-regional comparison of the firms has revealed a negative relationship between the presence of foreign-invested enterprises

and export production on one hand and the level of technological innovation on the other. This is the result of the nature of the Chinese ICT industrial firms which have occupied primarily the low-end position of manufacturing with low production costs as the main comparative advantages. It is also the result of the fact that domestic and foreign firms are not ‘sleeping in the same bed’ of technological innovation and, even when they did ‘sleep in the same bed’ they were ‘having different dreams’. Our firm-level surveys and subsequent interviews have suggested that many firms tended to see foreign and even domestic firms in the same region as rivals and competitors from whom information should be kept as commercial secrets rather than helpful partners for technological advice, knowledge exchange or co-operation. Our research has not found sufficient evidence to verify the two hypotheses we made concerning the relationships between technological innovation, production linkages and knowledge exchanges with both local firms and foreign-invested enterprises. Firms in Beijing with the highest level of innovation of all have surprisingly reported the least extent of production linkages and knowledge exchanges with local or foreign firms. A further comparison of the firms with different levels of innovation revealed no significant contribution made by localised production linkages to technological innovation. Instead, the nature and attributes of the firms are found to be the significant factors that set apart innovative from non-innovative firms. Important attributes of the firms included their regional setting, ownership, ability to mobilise venture and floating capital, and corporate strategy and management. The roles played by the state and enterprise–government affiliations are also significant determinants of the level of technological innovation demonstrated by indigenous Chinese firms. The peculiar Chinese case has raised important questions concerning the relevance and


adequacy of the popular notions of the ‘new economic geography’ and ‘new regionalism’ that overemphasise the explanatory power of localised production networks at the expense of the firms themselves as actors and agents in the process of technological innovation. Notes 1. We follow the Chinese official definition of foreign-invested enterprises (FIEs) which refer to the firms officially registered as sole (exclusive) foreign-owned ventures, joint ventures, co-operative ventures or foreign-funded corporations of limited liability (see China State Statistical Bureau, 2008, p. 547). 2. It should be noted that our usage of patent registration as a measurement of innovation has set a standard higher than normal expectation. A lower percentage of firms with granted invention patents in the Shanghai-Suzhou and Shenzhen-Dongguan regions does not deny the existence of technological upgrading through the setting up of new production lines, procurement of intermediary products and introduction of quality control.

Acknowledgements The work described in this paper has been sponsored by the grants obtained from the Research Grants Council of the Hong Kong Special Administrative Region, China (HKU 7666/05H and HKU 747509H), the National Science Foundation of the US (NSF BCS 0552237, 0552265, 0757615), the Ford Foundation (10851022) and the Mrs Li Ka Shing Fund. The authors wish to thank Tong Xin, Du Debin, Chen Wen and Ma Xiaojiao for assistance with data collection.

References Audretsch, D. B. and Feldman, M. P. (2003) Knowledge spillovers and the geography of innovation, prepared for The Handbook of Urban and Regional Economics, Vol. 4. (http://www.core.ucl. ac.be/staff/thisseHandbook/audretsch:feldman. pdf; accessed on 16 October 2008). Bathelt, H., Malmberg, A. and Maskell, P. (2004) Clusters and knowledge: local buzz,

global pipelines and the process of knowledge creation, Progress in Human Geography, 28(1), pp. 31–56. Beugelsdijk, S. (2007) The regional environment and a firm’s innovative performance: a plea for a multilevel interactionist approach, Economic Geography, 83(2), pp. 181–199. Brown, S. and Fai, F. (2006) Strategic resonance between technological and organisational capabilities in the innovation process within firms, Technovation, 26(1), pp. 60–75. Christensen, C. M. (1997) The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. Boston, MA: Harvard Business School Press. CSSB (China State Statistical Bureau) (2008) Zhongguo Tongji Nianjian 2008 [China Statistical Yearbook 2008]. Beijing: China Statistics Press. Döring, T. and Schnellenbach, J. (2006) What do we know about geographical knowledge spillovers and regional growth? A survey of the literature, Regional Studies, 40(3), pp. 375–395. Fagerberg, J., Mowery, C. D. and Nelson, R. R. (Eds) (2005) The Oxford Handbook of Innovation. Oxford: Oxford University Press. Fan, C. C. and Scott, A. J. (2003) Industrial agglomeration and development: a survey of spatial economic issues in east Asia and a statistical analysis of Chinese regions, Economic Geography, 79(3), pp. 295–319. Fischer, M. M. (Ed.) (2006) Innovation, Networks, and Knowledge Spillovers. Berlin: Springer. Gereffi, G., Humphrey, J. and Sturgeon, T. J. (2005) The governance of global value chains, Review of International Political Economy, 12(1), pp. 78–104. Humphrey, J. and Schmitz, H. (2002) How does insertion in global value chains affect upgrading in industrial clusters?, Regional Studies, 36(9), pp. 1017–1027. Hyytinen, A. and Toivanen, O. (2005) Do financial constraints hold back innovation and growth? Evidence on the role of public policy, Research Policy, 34(9), pp. 1385–1403. Kesidou, E. and Romijn, H. (2008) Do local knowledge spillovers matter for development? An empirical study of Uruguay’s software cluster, World Development, 36(10), pp. 2004–2028.


Kim, L. and Nelson, R. R. (Eds) (2000) Technology, Learning, and Innovation: Experiences of Newly Industrializing Economies. Cambridge: Cambridge University Press. Lall, S. (1993) Promoting technology development: the role of technology transfer and indigenous effort, Third World Quarterly, 14(1), pp. 95–108. Lin, G. C. S. (2009) Scaling-up regional development in globalizing China: local capital accumulation, land-centred politics, and reproduction of space, Regional Studies, 43(3), pp. 429–447. Lin, G. C. S. and Wang, C. (2009) Technological innovation in China’s high-tech sector: insights from a 2008 survey of the integrated circuit design industry in Shanghai, Eurasian Geography and Economics, 50(4), pp. 402–424. Meeus, M. and Oerlemans, L. (2005) Innovation strategies, interactive learning and innovation networks, in: S. Casper and F. van Waarden (Eds) Innovation and Institutions, pp. 152–189. Cheltenham: Edward Elgar. MII (Ministry of Information Industry) (2008) Statistical Yearbook of Chinese Electronic Information Industry 2007, digital version. Beijing: Publishing House of Electronics Industry,. Porter, M. E. (1990) The Competitive Advantage of Nations. New York: Free Press. Schumpeter, J. A. (1934) The Theory of Economic Development, transl. by R. Opie. Cambridge, MA: Harvard University Press. Sun, Y. (2002) Sources of innovation in China’s manufacturing sector: imported or developed in-house?, Environment and Planning A, 34, pp. 1059–1072.

Sunley, P. (2008) Relational economic geography: a partial understanding or a new paradigm?, Economic Geography, 84(1), pp. 1–26. Teece, D. J. (1986) Profiting from technological innovation: implications for integration, collaboration, licensing and public policy, Research Policy, 15, pp. 285–305. Wang, C. C. and Lin, G. C. S. (2008) The growth and spatial distribution of China’s ICT industry: new geography of clustering and innovation, Issues & Studies, 44(2), pp. 145–192. Wang, C. C., Lin, G. C. S. and Li, G. (2010) Industrial clustering and technological innovation in China: new evidence from the ICT industry in Shenzhen, Environment and Planning A, 42(9), pp. 1987–2010. Wei, Y. H. D., Lu, Y. Q. and Chen, W. (2009) Globalizing regional development in Sunan, China, Regional Studies, 43(3), pp. 409–427. Wei, Y. H. D., Zhou, Y., Sun, Y. and Lin, G. C. S. (2011) Production and R&D networks of foreign ventures in China: implications for technological dynamism and regional development, Applied Geography, doi:10.1016/j.apgeog .2010.06.008. Yeung, H. W. C. (2009) Regional development and the competitive dynamics of global production networks: an east Asian perspective, Regional Studies, 43(3), pp. 325–351. Zhou, Y. (2008) The Inside Story of China’s Hightech Industry: Making Silicon Valley in Beijing. Lanham, MD: Rowman and Littlefield. Zhou, Y., Sun, Y., Wei, Y. H. D. and Lin, G. C. S. (2011) Decentering ‘spatial fix’: patterns of territorialization and regional technological dynamism of ICT hubs in China, Journal of Economic Geography, 11(1), pp. 119–150.