Cross-Border knowledge pipelines and innovation performance of chinese firms: evidence from Zhangjiang in Shanghai

2.1 Cross-border knowledge pipelines

According to the previous paper, the concept of “global pipelines” has been used to refer to trans-local linkages across countries (Bathelt et al., 2004; Maskell, Bathelt & Malmberg, 2006; Grillitsch & Trippl, 2014). From the perspective of firm, we argue that cross-border knowledge pipelines can be defined loosely as the global linkages related to the knowledge acquisition strategies of MNEs to leverage and integrate dispersed ideas and technologies across nations. Cross-border knowledge pipelines provide diversified and non-redundant technologies for the recombinant and generation of newly valuable knowledge across national borders, in order to concentrate on the exploration and exploitation of external knowledge (Bathelt & Li, 2020).

Much of recent literature on cross-border knowledge pipelines stresses the differences between the formal organisation-based linkage and the relatively informal individual-based linkage (Moodysson, 2008; Yeung, 2009; Fitjar & Huber, 2015). In this present, we primarily focus on foreign R&D investment and transnational individuals represented by returnees as the two types of cross-border knowledge pipelines, and they often represent different learning modes and interactive mechanisms of knowledge creation.

To be specific, foreign R&D investment as the formal organisation-based global linkage is often the outcome of the firm-level strategic behaviour. They are conducted to achieve specific outcomes, and involve active interventions which are typically not indigenous but designed and maintained for short and single returns (Fitjar & Huber, 2015). Hence, foreign R&D investment as a kind of global linkage deliberately maintained by organisations is likely to serve for relatively short-time and one-way objectives. In terms of innovation activities, innovation emerging through foreign R&D investment tends to be characterised by strategic and relatively less technological diversity.

By contrast, the individual-based global linkages in the form of relatively informal personal ties and mobility, used to be created and held by transnational individuals. This type is more flexible in knowledge transferring, and ‘international personal networks exhibit a more robust positive relationship with innovation than international formal networks’ (Fitjar & Huber, 2015). Additionally, transnational individuals lead to innovation activities through trust-based interactions among heterogenous individuals (Trippl, Tödtling, & Lengauer, 2009). Such learning effect and knowledge outcomes brought by these creative individuals is emergent and more radical (Mudambi & Swift, 2009).

2.2 Foreign R&D investment, returnees and firms’ innovation performance

Foreign R&D investments for firms in developing countries are associated with their innovation strategies to generate disruptive innovation and keep pace with the industry leaders (Awate, Larsen, & Mudambi, 2015). In the context of developing countries, firm’s foreign R&D investment is driven primarily by the technological exploration rather than technological exploitation. In this respect, foreign R&D investment facilitates to obtain external innovation resources which are scarce and unavailable at domestic, and expand the firm’s knowledge base. Second, in the face of increasingly sophisticated technological changes, foreign R&D investment can help decrease the risk and uncertainty of innovation projects, responding quickly to fierce global competition and the future technical revolutions (Fu et al., 2021). More importantly, in the case of a lack of technical ability, these firms from developing countries could create business linkage with international technological leaders through their foreign R&D investment, providing the opportunities to leverage and learn from different types of international partners (Fan, 2011; Si & Liefner, 2014).

Meanwhile, increasing numbers of studies suggest that the importance of foreign R&D investment for firms in developing countries (Tang, Tang, & Su, 2019). Fan (2011) employed the case studies of the companies Huawei and ZTE and found that the R&D globalisation strategy as an important technological strategy is beneficial to tap into global resources and markets unavailable to firms from developing countries when they possess a certain level of technological capability. Using the patent data from different patent offices, Schaefer & Liefner (2017) analyse Huawei’s ability to reach better R&D performance through offshore R&D, and they find that foreign R&D investment outperform their domestic activities to help them produce advanced knowledge. Based on a questionnaire survey of 738 Chinese firms, Si, Liu & Zhang (2021) suggest that three types of foreign R&D investment in forms of cross-border R&D collaboration, setting up offshoring R&D centres and transnational merger and acquisition (M&A), all of which contribute to improving the firm’s innovation performance. Actually, contrary to these firms from developed countries, the process of accessing and applying external knowledge for firms in developing countries mainly extends from global to local contexts, and the vast domestic markets in developing countries are considered to provide for local firm natural testing grounds to integrate and apply innovation resources by conducting foreign R&D investment.

In regards to another important form of cross-border knowledge pipelines, the benefits of transnational individuals related to firm’s innovation can be attributed to the following aspects: on the one hand, transnational individuals can transfer effectively tacit and complex knowledge critical to firm’s innovation, and are hence typically viewed as ideal carriers of this type of knowledge (Kogut and Zander, 1992). There is increasing recognition of the critical roles of transnational individuals in transferring and acquiring external knowledge, especially for ‘know-how’ and ‘know-who’ types of knowledge, particularly in developing countries where firms are characterised by insufficient knowledge base and less absorptive capacity for technological learning (Filatotcheva, Liu, Lu, & Wright, 2011; Bathelt & Li 2020). On the other hand, transnational individuals possess cross-cultural background and experience, and thus have advantages in coping with challenges from different organisational and institutional contexts to integrate external knowledge. Innovation activities increasingly depend on interactive learning. Social capital theory emphasises the role of relational capital in acquiring knowledge externally through individualized networks. Transnational individuals represented by returnees can help to access external knowledge based on their relational networks (Liu et al., 2010).

The voluminous body of work on transnational individuals shows the importance of returnees for external knowledge sourcing. For instance, Fu, Woo, & Hou (2016) have stressed that it is the returnees that help Chinese solar panel firms have become global leaders. Wang (2012) also pointed out that returnees have played an important role in the development of the strategic emerging industries such as renewable energy, electrical cars and biotechnology in China. Using the survey data, Liu et al., (2010) found that returnee firms are more innovative than non-returnee firms, and returnees are an important channel for international knowledge spillovers based on unique networks resources and plentiful social interaction.

Hence, we propose the following hypotheses:

H1a: Foreign R&D investment positively affects firm innovation performance.

H1b: Returnees positively affect firm innovation performance.

2.3 The interaction effect of foreign R&D investment and Returnees

The interplay especially for the complementary relationship between different types of cross-border knowledge pipelines has been stressed by recent literature (Bathelt & Li, 2020). With rapidly changing and complex technology, people increasingly realised the importance of external knowledge sources in firms’ innovation management practices. However, to benefit from external knowledge, one must cope with the potential barriers of cross-border knowledge pipelines involving connecting, sense-making, and integrating of cross-border knowledge (Mathews, 2002; Bathelt, Cantwell, & Mudambi, 2018). In this respect, transnational individuals not only directly affect firms’ innovation but also indirectly affect the innovation performance of Chinese firms when they engage in internationalization activities to build cross-border knowledge pipelines (Wu & Coe, 2022).

On the one hand, transnational individuals comprise a vital channel for gaining tacit knowledge of international markets, facilitating the efficiency of foreign R&D investments (Liu & Buck, 2007; Fitjar & Huber, 2015). Introducing foreign technology is one thing but then being able to use it fully is another (Fu, Woo, & Hou, 2016), that is to say, the ease of mastering foreign technological knowledge is associated with the firm’s absorptive capacity which represents the ability of a country to identify, assimilate and exploit knowledge from the environment (Cohen & Levinthal, 1989). A firm can absorb new foreign technology largely depends on its absorptive capacity. Prior studies demonstrated the significant role of individuals in affecting diffusion and absorption of foreign knowledge, for transnational individuals are characterised by their cross-cultural advantages to bridge transnational contexts (Liu, Lu, & Choi, 2014; Si, Zhang, & Teng, 2021). What’s more, foreign technologies developed in other countries mismatch with the economic and social conditions of developing countries, which requires transnational individuals to adapt this new foreign technology to suit local conditions. Therefore, only with the successful internalization of the tacit knowledge of the foreign innovation can the foreign innovation be employed to reach its potential. In short, there is synergy between foreign R&D investments and transnational individuals.

Transnational individuals enable firms to overcome the liability of being outsiders when firms conduct internationalization activities like foreign R&D investments. Transnational individuals may compensate for geographical distances and create institutional proximity to innovation opportunities through their international networks (Saxennian, 2006). Transnational individuals, especially for returnees, may connect diverse knowledge to help Chinese firms engage in foreign R&D investment, further improving the positive effect of foreign R&D investment on innovation. These returnees returned to home countries not only brought back new knowledge in science and technology, but also brought back ideas for advanced commercial production, playing an important role in improving the innovation performance of firms in developing countries (Rauch & Trindade, 2002). Saxenian (2002, 2006) in the 1980s observed that local firms in Israel, Taiwan, and India have greatly benefited from these cross-region collaborations with important customers and partners in Silicon Valley, where transnational technological communities play particularly significant roles. In this regard, transnational individuals play the role to overcome frictions due to different knowledge bases and institutional contexts (Bathelt & Li, 2020). We propose that the moderating role of transnational individuals should be taken into account when examining the effect of knowledge spillovers from foreign R&D investments. Hence, we give the following hypothesis:

H2: Returnees play a positive moderating role in the relationship between foreign R&D investment and firm innovation performance.

Figure 1:

Concept Framework – Foreign R&D Investment, Transnational Individuals, and Cross-Border Knowledge Pipelines

2.3 The moderating effect of ownership

The process of building cross-border knowledge pipelines is highly associated with firm’s internationalization activities. Much of literature have highlighted the importance of institutional factors in explaining the success of internationalization and innovation activities of firms in developing countries (Meyer, 2018). Innovation is viewed as the outcome of knowledge recombination in a new way. The likelihood of accessing less redundant knowledge will be increased when firms operate internationalization activities across countries. Simultaneously, recombinatory costs such as capital expenditures, coordination costs, and integration costs will increase due to the lack of interdependence among countries (Kim, Lampert & Roy, 2020). Institutional advantages facilitate to enhance their recombinant ability and reduce the liability of foreignness they face, which offset deficits in traditional ownership.

Benefiting from cross-border knowledge pipelines requires firm’s recombinant capability, while the recombinant capability of firms largely depends on the firm-specific advantages (FSAs). For those firms in developing countries, their FSAs are typically more associated with the endowment of institution advantages rather than these traditional assets in technology and managing experience (Meyer, 2018). More specifically, firm’s ownership is considered as the proxy for institutional factor at firm-level that shapes organizational decisions and the structure of organization interaction (Rialp-Criado & Komochkova, 2017). The advantages of FSAs for those in developing countries provide preferential access to resources (Hennart 2012; Narula 2012), notably to financial resources (Morck, Yeung & Zhao, 2008), but also to network relationships with major innovative players. Rialp-Criado and Komochkova (2017) indicate that access to financing is the most severe obstacle to firm’ business development, as exemplified by more than 10 % of private Chinese SMEs. In contrast to private firms, state-owned firms often possess more closer political connections with government, indirectly providing a favourable institutional environment for firms’ innovation internationalization activities.

From the perspective of knowledge application, institution advantages are reflected in providing large market to apply overseas knowledge. The innovation systems of developing countries are immature due to the limited absorptive capacity and weak institutions (Françoso & Vonortas, 2022). Most of innovation resources are concentrated in public sector like universities and public research organizations. Actually, innovation is not solely the R&D processes, but also is the commercialization processes (Kim & Pennings, 2009). Such policies and unique technological resources because of closer-ties with government allow firms to obtain more opportunities not only in accessing more various and more superior overseas R&D sources, but also more importantly in the aspect of market to transform R&D resources.

State-owned firms with higher government-associated relationship enjoy preferential access to friendly regulation, financial sources, and important information related to the foreign business environment, and hence theoretically possess more abundant resource and more significant incentives to help firm build cross-border knowledge pipelines, like investing foreign R&D and employing returnees (Chang et al., 2006). Therefore, we argue that state-owned firms are more likely to achieve higher innovation performance by building cross-border knowledge pipelines. Building upon these arguments, we hypothesise the following:

H3a: State ownership positively moderates the effect of foreign R&D investment on a firm’s innovation performance.

H3b: State ownership positively moderates the effect of returnees on a firm’s innovation performance.

3.1 Data

The empirical analysis employed a dataset of 4,685 Chinese firms which registered in Zhangjiang National Innovation Demonstration Zone. Zhangjiang was established in 1992 and consist of a group of 22 high-tech parks in Shanghai at the end of 2015 (Figure 2). In 2011, Zhangjiang was formally granted by the central government as a national experiment and demonstration zone for indigenous innovation. Over the last two decades, Zhangjiang has taken aggressive actions in encouraging MNEs to explore and exploit global R&D resources, and attracted high-quality talent from the whole world, aiming to build a regional innovation ecosystem in connection with global innovation networks (Zeng et al., 2011).

As an important demonstration area to serve the‘indigenous innovation’ strategy of the Chinese government, Zhangjiang have great advantages in accessing extensive global innovation sources and take especially aggressive action in exploiting cross-border R&D resources and transnational individuals (Zeng et al., 2011). By 2011, Zhangjiang attracted 58 persons in the central “thousands of talents” programme, accounting for 67 % of the total in Shanghai and 7 % of the national total (ZJHP, 2012), and the percentage of R&D in GDP increased by 3.11 % (Zhang, 2015).

The unique dataset in this study is collected by the Administrative Committee of Zhangjiang in Shanghai. As the only management office authorized by government, the Administrative Committee of Zhangjiang is responsible to investigate the development of Zhangjiang every year, and this dataset is based on the 2015 Zhangjiang Annual Sampling investigation. The advantage of this dataset lies in covering detail firm-level information, including firm’s basic characteristic and innovation outcome, as well as firm’s internationalization behaviours such as exporting, foreign R&D investment, the employment of transnational individuals. Table 2 offers an overview of the basic characteristic of firms in Zhangjiang.

3.2 Variables

3.2.2 Independent Variables

To test the above hypotheses, different types of cross-border knowledge pipelines were captured, namely foreign R&D investment and transnational individuals represented by returnees. The key variables of this study contain foreign R&D investment and transnational individuals represented by returnees. To be specific, foreign R&D investment is constructed as a dummy variable. It is set as the value of ‘1’ when a firm operates any form of foreign R&D investment, such as transnational merger and acquisition (M&A), offshoring R&D, international technology alliance, and ‘0’ otherwise. Returnees are those who have gone abroad to study and obtain a degree among employees. They can be high-end experts with leading technology, senior executives of firms, or professional and technical personnel with special skills. Following previous research, returnee is measured by the ratio of returnees to the total number of employees.

Figure 2:

The Landscape of Zhangjiang Park

3.3 Methods

Given the character of the dependent variables, we used the tobit regression to analysis for our estimation model. Keeping line with prior literature, it is preferable for employing the tobit regression to test our hypotheses since the dependent variable is censored (Tobin, 1958; Yi, Hong, Hsu, & Wang, 2017; Teng et al., 2020). Specifically, the equation of regression models as follows:

Innovation performance _i = α + βForeignR&D_i + λTransindividual _i + δX _i + ε, (1)

The variable Innovation performance _i indicates innovation performance of the firm i in Eq. (1). X_i is a vector of the interaction terms (Returnee × Foreign R&D, Ownership x Foreign R&D, Ownership x Returnee).

Table 3 provides detailed information on the descriptive statistics of the indicators and correlation matrix of all variables. We conducted a test on the potential issue of collinearity among independent variables. According to the empirical findings presented in Table 3, there is no multicollinearity in our model since the highest Variance Inflation Factor (VIF) among explanatory variables is only 2.79, far below the critical point of 10 (Belsley et al., 1980). Regression analyses further identify interrelationships among judgement variables.

4.1 Effects of foreign R&D investment and returnees

Table 4 reports the empirical results of the impact of different types of cross-border knowledge pipelines on firms’ innovation. The result of Model 2 in Table 4 shows that both foreign R&D investment and returnees have statistically significant and positive relationships with firms’ patent innovation (β = 0.0587, p < .01; β = 0.2657, p < .001, respectively), implying that foreign R&D investment facilitates to enhance firms’ innovation performance. This type of cross-border knowledge pipeline in forms of foreign R&D investment could prove to be effective and important for firms’ patent innovation. Foreign R&D investment can provide firms with diversified and complementary knowledge from dispersed geographical locations. Actually, foreign R&D investment have often been severed as a major driver of technological upgrading in less developed countries especially at the initial stage of development (Yang, 2015). The transnational technical communities consist by returnees create potential windows of opportunity for firms in developing countries by providing privileged linkages to world-class innovation resources (Saxenian, 2002). Jons (2009) argued that this type of brain circulation contributed significantly to the reintegration of Germany into the international scientific community after the Second World War. In terms of patent innovation, H1a and H1b is therefore supported, which is consistent with the finding in most prior research (Hsu et al., 2015; Schaefer & Liefner, 2017; Si et al., 2021).

Besides, Model 2 and Model 3 in Table 5 report the impact of diffident types of cross-border knowledge pipeline on firm’s product innovation. We find that foreign R&D investment exerts no significant influence on the production of innovation, while returnees exerts negative influence on the production of innovation (β = –5.3780, p < .005). The establishment and management of global pipelines involve differences in the cultural and institutional contexts in which the firms operate, not least because of considerable uncertainties and high investments (Maskell, Bathelt & Malmberg, 2006). In contrast to patent innovation, the relationship between cross-border knowledge pipelines and product innovation especially for returnees are sensitive to the cognitive distance caused by cultural and institutional contexts. It shows that returnees as transnational individuals have more advantages in technological innovation rather than the innovation of business model or market innovation.

4.2 The interaction of foreign R&D investment and returnees

To examine the interaction of foreign R&D investment and returnees, we empirically examine the moderating effect of returnees on the relationship between foreign R&D investment and firms’ innovation performance. According to the empirical results shown in model 5 of Table 4 and Table 5, we find that returnees have a positive and statistically significant moderating effect on the relationship between foreign R&D investment and firms’ patent innovation (β = 0.5521, p < .001). Returnees help overcome the liabilities of foreignness, liabilities of outsiders, and liabilities of origin for latecomer firms, facilitating to improve the effective of foreign R&D investment related to patent innovation (Solheim & Fitjar, 2016). In essence, innovation activities are highly associated with a culturally embedded process. Yang and Liao (2010) identified three interrelated dimensions of ‘embeddedness, including societal embeddedness, network embeddedness, and territorial embeddedness’. Returnees can be understood as the concept of communities of practice to some extent, which “contributes to the solution of practical problems and bridges the gap between theoretical and practical knowledge” (Moodysson, 2008).

4.3 Moderating Role of State Ownership

Model 6 and Model 7 in Table 4–5 separately present the results concerning on the moderating effects of firm’s ownership. We find that state ownership is not equally positive for the effect of different cross-border knowledge pipelines on firms’ innovation performance. According to the empirical result in Model 6, there is a negative and statistically significant moderating effect of state ownership on the relationship between foreign R&D investment and firms’ patent innovation (β = –0.1295, p < .001), while state ownership exerts no significantly moderating effect on the relationship between cross-border knowledge pipelines and firms’ product innovation, no matter for foreign R&D investment or returnees. Hence, H3a and H3b is not supported. This implies that state ownership hinders the positive effects of foreign R&D investment on firms’ innovation performance to some extent, which is line with the finding in recent some literature like Pereira et al., (2021).

Although state ownership has many advantages to help improve the effective of cross-border knowledge pipelines, absorptive capacity act as a more important moderator for the relationship between foreign R&D investment and firm’s innovation. We argue that it is likely to be associated with the business model of state-owned firms in China, which always serves as the agent of government decision-making rather than seeking commercial benefits, and the process of learning the tacit knowledge required in using the foreign technology fully is made easier by strong absorptive capability represented by returnees.