The innovative performances of China’s automobile industry shall be evaluated by quantitative indexes of enterprises and qualitative indexes of managers. Hagedoorn and Cloodt (2003) proposed that innovative performances reflect all achievements by innovative activities and final values by strategy and R&D in a generalized and special way, respectively. In the same line knowledge-based view and absorptive capacity theory are utilized to judge the role of distributed innovation to the automobile industry. It is widely accepted that knowledge management from different sources on which distributed innovation dependent as different sources provide knowledge as users, suppliers resulted in innovation and competitive advantage as Flynn et al. (2003); Zack et al. (2009); Huang (2011); Lau and Bruton (2011); Jin et al. (2012); Kang et al. (2014); Iqbal et al. (2018); Qi and Chau (2018); Zand et al. (2018); Abubakar et al. (2019); Hammoud (2020). It is believed that this performance includes efficiency and performance of innovations. The innovative performances of China’s automobile industry can be evaluated in a subjective or objective way. The objective way emphasizes enterprise profit, including sales volume, return on investment and financial indexes, while the subjective way focuses on judgment on enterprise performance of the respondent, including evaluation of innovation capability and consumer satisfactory of new products. In either way, the conclusions are neither accurate nor convincing. Hence, some scholars believed that subjective and objective factors shall be combined to make the results convincing. We also believe that innovative performances of China’s automobile industry shall be evaluated based on both subjective and objective factors.

Chesbrough et al. (2018) distribution innovation is the process of managing the flows of knowledge across the organizational boundaries using pecuniary and non-pecuniary mechanisms in line with the organizational business model. In the same line Bertello et al. (2022) stress the need of the distributed innovation for the firms to achieve the competitive advantage and survive in this competitive market. The study proposed that knowledge beyond the organizational boundaries influence the organizational performance. Cummings (2006) claimed that distributed innovations are achieved by execution of ideas, tasks or procedures by employees all over the world. Coombs and Metcalfe (2002) pointed out that distributed innovation refers to distributed realization of technologies and related capabilities required by innovation networks in different organizations and other knowledge production institutions. Owing to rapid globalization, researchers prefer to investigate distributed innovation from perspectives of era and resources. Dooley and O’Sullivan (2007) proposed that distributed innovation embedment lies within the entire supply chain of its organization and cooperative and project innovations induced by enterprise alliances have positive effects on enterprise performances such as innovations, thus enhancing its competitiveness in the global market. Yan et al. (2017) defined distributed innovation as collaborative innovation activities of enterprises during technological innovations. Specifically, enterprises achieve resource sharing and knowledge management by integrating internal and external resources in virtue of their advantages such as geographical dispersion. Liu et al. (2010) believed that distributed innovations are innovative activities by enterprises with cooperative relationship (e.g., upstream and downstream) based on resource sharing and shared network platform. Compared with centralized innovations, distributed innovations are characterized by resource sharing, regional distribution, simultaneity, coordination and cooperation, which enable excellent innovative performances of enterprises. Chen et al. (2017) divided distributed innovations into three dimensions: distributed structure, distributed cognition, distributed cooperation. These dimensions enable growth and profit of enterprises in various aspects. In other words, the positive effects of distributed innovation embedment on innovative performances of enterprises are strengthened.

Marlow et al. (2008) proposed the concept of knowledge collaboration and defined it as a process where multiple parties with knowledge resources participate in knowledge activities, with the aim of knowledge innovation. It is also proposed that knowledge collaboration is a dynamic process that can effectively enhance organizational competitiveness and innovative performance, thus achieving a collaborative effect for enterprises.

Previous studies of distributed innovation reveal that knowledge collaboration embedment has influences on innovative performances of enterprises. For instance, Hameed et al., (2018) examined the driver of the distributed innovation among the Malaysian small and medium enterprises and they concluded that R&D, external knowledge and internal innovation are the main drivers of the distributed innovation. Li et al. (2020) highlighted the effective use of organizational ambidexterity is significantly dependent upon the distributed innovation. They concluded that firms can only take the advantage of organizational ambidexterity is dependent how firms incorporate the knowledge outside the organizational boundaries. Oborn et al. (2019) highlighted the role of digital technologies in the development of distributed innovation and found that distributed innovation resulted in the better performance of the firms. With rapid development of knowledge economy, scholars divided distributed innovation into distributed structure, distributed cognition and knowledge collaboration. The distributed structure includes member distribution and project modularization within the distributed innovation network. The former refers to the geographical dispersion of members, while the latter enables close correlation of internal factors in the project with team task as the core part. The distributed cognition mainly refers to trust between members in the distributed network and similarity of distributed cognition. The knowledge collaboration consists of transfer and sharing of knowledge and innovation management, namely vertical and horizontal expansions of knowledge. Participation of China’s automobile enterprises in distributed innovation includes strengthening of geographic location and project modularization in distributed structure, strengthening of trust and cognitive inclination of members in distributed cognition, and strengthening of innovation management in knowledge collaboration, which is beneficial to their innovative performances. Despite its irreplaceable role in domestic economy, China’s automobile industry exhibits low overall competitiveness. Especially, China’s automobile manufacturers are significantly inferior to top players in the world in terms of innovative performances. Hence, significant improvements of innovative performances of China’s automobile industry are urgently needed. Distributed innovation embedment is beneficial to product innovation, management innovation and technological innovation in China’s automobile industry by reducing innovation cost and improving innovative performance.

In summary, the following hypotheses are proposed:

Burgelman et al. (1996) claimed that technological innovation capability serves as the foundation of technological innovation strategy of enterprises. Leonard-Barton (2003) believed that technological innovation capability is a core competitiveness of enterprises. The innovation capability is essential for sustainable development of enterprises as it has positive effects on their innovative performances and competitiveness. Innovations are originated from R&D. Indeed, enhancement of enterprise innovation depends heavily on the improvement of its R&D capability. Hence, distributed innovation embedment in the R&D network plays a key role in enhancement of R&D capability.

Owing to limited resources and capabilities, enterprises have to rely on distributed innovation systems to enhance R&D capability. A distributed innovation system is both a collaborative process of different organizations within the distributed innovation network and an interactive process of mutual adaptation and interactions between these organizations and the external environment. The distributed structure of distributed innovation system delivers a R&D network with peers in the automobile industry or upstream/downstream enterprises by modularization of complex system and strategy alliance. In this way, the R&D capability of an enterprise can be strengthened. The distributed cognition of distributed innovation system emphasizes common interest-based distributed cognition and trust, as well as dependence of upstream and downstream enterprises, resulting in a large-scale economy of China’s automobile industry. The knowledge collaboration is indeed the core part of distributed innovation system. R&D is originated from collaboration of vertical processing and horizontal modularization of knowledge. Creativity, knowledge and technical guidance can be obtained from collaborations, resulting in an effective R&D system.

In summary, the following hypotheses are proposed:

Drucker (2014) regarded enterprise innovation as a tool from the perspective of function. Specifically, innovation is regarded as a tool owned by enterprise and appropriate utilization of this tool inspires novel creativity from existing resources, thus achieving increased profits. R&D activities play a key role in enterprise competitiveness. However, Mr. Zhu Sen, the executive vice president of China Machinery Industry Federation, said that “most of the existing technologies in China’s automobile industry are imported, while independent R&D is still at an early stage.” Zhang et al. (2008) also insisted the dominant role of R&D capability in competitiveness of China’s automobile industry. Unfortunately, most investments by China’s automobile manufacturers have been used to expand the production capability instead of R&D, while foreign competitors have made considerable investments in R&D. In the market economy, enterprises are direct beneficiaries of R&D activities and R&D is supposed to attract great attention by enterprises. Enterprises’ investment in R&D serves as the objective foundation of ownership of core technologies by China’s automobile industry. Additionally, investment in R&D allows enterprises to surpass competitors in product innovation, technological innovation and management innovation, thus achieving high-end products and services.

In summary, the following hypothesis is proposed:

The proposed model (see Figure 1) includes distributed innovation embedment (e.g., knowledge collaborative embedment, distributed cognition embedment, distributed structure embedment, R&D capability, innovative performances of China’s automobile industry). Questionnaires were designed and optimized according to professional suggestions. Each variable is reflected by several items and each item is evaluated by the Likert five-level scale (1 = significantly over-low, 5 = significantly over-high).

Innovative performances of China’s automobile industry are affected by distributed innovation embedment, including knowledge collaborative. This study further explores the significance of distributed innovation (e.g., knowledge collaboration embedment, R&D capability) to enterprise development. Hence, innovative performances of China’s automobile industry, distributed innovation and R&D capability were measured. Innovative performances of China’s automobile industry have been investigated based on sales volume, market shares, financial indexes, evaluation of innovation capability by managers and consumer satisfactory on new products. Considering differences in strategy and management of enterprise innovation, we measured innovative performances of enterprises from perspectives of product innovation, product update, technological innovation and managers, which are denoted by IP1, IP2, IP3 and IP4, respectively. Six items were designed for this measurement as Tong (2011).

R&D capability was measured from perspectives of ratio of investment in R&D, patent quantities, R&D employees, which are denoted by RDA1, RDA2 and RDA3, respectively. Six items were designed for this measurement as Xie et al. (2013).

The variables of distributed innovation embedment were measured by methods reported elsewhere. Specifically, the variables are divided into categories of distributed structure embedment, distributed cognition embedment and knowledge collaborative embedment, which are denoted by DI1, DI2 and DI3, respectively. The distributed structure embedment focuses on geological distributions and project modularized allocation; the distributed cognition embedment focuses on trust, cognitive learning and cooperation among members; the knowledge collaborative embedment focuses on knowledge transfer, knowledge sharing and innovation management. Sixteen items were designed for this measurement as Li (2013).

Questionnaires were designed according to the measurement variables and sent to representative automobile manufacturers in China, including FAW-Volkswagen, Dongfeng Motor Corporation, Dongfeng Peugeot Citroen Automobile Company Ltd. (DPCA), Renault S.A., Volvo and Dongfeng Yueda Kia (DYK). One hundred eighty-seven questionnaires were allocated via e-mail and 117 of them completed by respondents from automobile manufacturers participated in distributed innovation were regarded as effective ones. The study used non-probability sampling technique for the sample selection. All respondents are managers and engineers who have a thorough understanding of the enterprise and the industry. The manufacturers involved are mainly located in Hubei, Shanghai, Beijing, Chongqing, Guangzhou and Tianjin. Additionally, 90% of respondents have distributed innovation as a part of their enterprise strategy. In summary, samples are widely distributed and representative, thus meeting the requirements.

In order to ensure the validity of this study, validity and reliability of samples shall be verified before data analysis. In this study, validity and reliability of samples were verified by factor analysis and internal consistency test, respectively. As shown in Table 1, Cronbach’s α coefficients of all measuring variables exceed 0.7, exhibiting good scale reliability. Data analysis by KMO and Bartlett’s test reveals that variables are suitable for factor analysis as the KMO value is 0.797 and the Chi-Square significance is 0.000. The accumulated factor contribution rates of measuring variables are no less than 0.652, indicating good validity of the scale. In summary, the proposed questionnaire scale shows high reliability and validity.

Since reliability and validity of samples have been demonstrated to be reasonable, a theoretical model is proposed according to the hypotheses (see Figure 2) to investigate the correlations of variables. AMOS17.0 was used for verification as Coombs et al. (2003); Leonard-Barton (2003); Li and Jin (2010); Alexy and Leitner (2011); Liu et al. (2011); Peng (2012); Wencong and Guilong (2013); Osipitan (2014); Chen et al. (2017).

Model verifications of different variables demonstrate that fit indexes of the proposed theoretical model meet the requirements. Meanwhile, model verifications indicate that the goodness of fit index (GFI), the comparative fit index (CFI) and the root mean square error approximation (RMSEA) are 0.942, 0.973 and 0.035, all of which meet the requirements, demonstrating good fitting of the model and practical data. Hence, the results are statistically significant. Table 2 lists the results of hypothesis test. The path coefficients, corresponding hypotheses and p values in the theoretical model are summarized to obtain path coefficients and verification results.

As shown in Table 2, all path coefficients (ð, μ, ξ, ɤ) have statistical significances and Hypotheses H1a, H1b, H1c, H2a, H2b, H2c and H3 are valid. Hence, Hypotheses H1, H2 and H3 are valid, indicating that all hypotheses are verified. Additionally, hypothesis test demonstrates that distributed innovation embedment can influence innovative performances of China’s automobile industry directly and indirectly via R&D capability.

Hypotheses H1a is valid. According to H1a, China’s automobile enterprises shall pay more attention to project modularized allocation and geographical distribution of 4S stores in order to increase their market shares.

Hypotheses H1b is valid. According to H1b, learning cognition and identification of corporate culture by employees have positive effects on innovative performances of China’s automobile industry.

Hypotheses H1c is valid. According to H1c, sharing, transfer and creation of key resources, namely knowledge, have significant effect on success rate and frequency of new product development and automobile industry shall pay more attention to knowledge management to develop intelligent enterprises and creative employees.

Hypotheses H2 is valid. According to H2, distributed innovation has positive effects on R&D capability of China’s automobile industry.

Hypotheses H3 is valid. According to H3, enhanced R&D capability has positive effects on technical content, success rate and frequency of new product development in China’s automobile industry.

In summary, China’s automobile enterprises shall enhance their innovative performances and competitiveness by participating in distributed innovation and improving R&D capability.