How do R&D networks change? The upgrading of innovation capabilities in emerging market firms. Insights from China’s wind energy sector

ABSTRACT

Innovative activities are increasingly decentralized and globally dispersed, which provides new upgrading opportunities for emerging market firms. However, little is known about how latecomer firms (re-)organize their research and development (R&D) over time as conditions for upgrading change. This paper systematically maps the R&D networks of China’s lead firms in the wind turbine industry. The empirical findings reveal that latecomer firms not only exploit but increasingly co-create vanguard knowledge in global R&D networks through organizational diversification. Taking an evolutionary perspective, the paper extends our understanding of the changing nature of upgrading mechanisms and provides new insights into the reorganization of innovation processes in an era of technological change.

KEYWORDS:

• Latecomer firms
• upgrading of innovation capabilities
• global R&D networks
• wind energy
• China

JEL CODES:

• O31 (Innovation and Invention: Processes and Incentives)
• O32 (Management of Technological Innovation and R&D)
• O33 (Technological Change: Choices and Consequences)
• Q20 (Renewable Resources and Conservation)
• Q55 (Technological Innovation)
• L10 (Market Structure
• Firm Strategy
• and Market Performance)
• L60 (Industry Studies Manufacturing)
• D83 (Search
• Learning
• Information & Knowledge)

Acknowledgments

The author is grateful for helpful and constructive comments on earlier drafts of this paper from the following persons: Stine Haakonsson, Susana Borrás, Rasmus Lema, Max von Zedtwitz, Primoz Konda and Jan Giese.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 ‘Open Innovation’ (Chesbrough 2003, 2017; Chesbrough 2006) and the ‘Organizational Decomposition of Innovation’ (Schmitz and Strambach, 2009) constitute two differently framed perspectives on the same underlying phenomenon.

2 Digital technologies allow firms to access external knowledge and collaborate with external partners at an unprecedented scale (Veugelers et al., 2010; Curley and Salmelin 2018). In addition, given that open innovation occurs in multidisciplinary and cross-organizational R&D networks (Ritala et al., 2017), it increases the likelihood of generating more radical innovation (Chesbrough 2006; Lee 2010).

3 Decentralization and dispersion describe activities related to the decomposition of innovation, with the first highlighting the organizational and the second the geographical dimensions.

4 It is acknowledged that the terms market-based and science-based partners are not complete substitutes for loose/fixed links to production, as they characterize the types of actors rather than the outcome of a particular collaboration (e.g., new knowledge created through inter-firm collaborations or purely commercial links with universities). In addition, science-based partners such as universities can become market partners when they spin off and create separate entities primarily concerned with commercial industry practice.

5 Upgrading is herein defined as enhancing a latecomer firm’s innovation capabilities through effective learning; it does not refer to the notion of upgrading as defined in the global value chains literature, i.e., increasing a firm’s benefits by participating in global value chains.

6 For example, China’s value-added content in ICT and electronics has increased substantially over the last decade (OECD, 2018b).

7 The terms ‘technology transfer,’ as originally used by Lema and Lema (2012), and ‘technology transmission’ (Haakonsson and Slepniov, 2018) imply by convention knowledge flows from OECD to emerging market firms. This study understands conventional upgrading as largely unidirectional and unconventional upgrading as a largely multidirectional learning process.

8 Expanding the innovation space is important both to absorb knowledge in the home market and to gain access to vanguard knowledge in international markets.

9 GOL licensed first from Jacobs Energie/REpower (DE) and later from Vensys (DE), where it acquired a majority stake of 70% in 2008. MNG licensed from aerodyn (DE) and General Electric (US).

10 E.g., GOL with Hebei Electric Equipment (CN), CSR Electric (CN), and Beijing Tianrun (CN); MNG with Nanjing Gearbox Corporate (CN) and ABB (CH).

11 E.g., GOL with Xinjiang Agriculture University (CN), the National Wind Technology Centre (CN), and Delft University (NL); MNG with the Guangdong Wind Power Research Institute (CN), Xiamen University (CN), and Risø-DTU (DK).

12 GOL as Xinjiang Wind Energy Company and MNG as Ming Yang Electric.

13 No patents were filed during the first phase as GOL and MNG were buying complete turbine designs through licensing agreements; ENV only became operational at the end of the first phase.

14 E.g., Xiexin Wind Power (CN), China Machinery and Equipment (CN), and Yiwu Tianrun Wind Power (CN).

15 E.g., with CELEC in Ecuador, Mainstream Renewable Power in Chile, and InterEnergy and UEP Penome in Panama.

16 Referring to the Rendsburg-based (DE) aerodyne Energiesysteme GmbH.

17 This window of opportunity was opened as the Chinese government required conventional power plant constructors and operators to invest a certain percentage (ca. 12%) into renewable energies.

18 ‘Hybrid’ refers to the combination of renewable energy technologies (often wind and solar PV) and storage solutions.

19 E.g., ProtectWise, Vidder, Onion ID, PubNub, and Baffle in the field of cloud security/computing and Orbital Insight and ZingBox in the field of big data analytics.

20 E.g., Eco 5 (DE), Delta Energy (DE), DNV GL (NO), University Twente (NL), Frauenhofer IWES (DE).

21 E.g., Tsinghua University (CN), National Renewable Energy Laboratory (US), Frauenhofer Institute (DE), University of Stuttgart (DE), Technical University of Denmark (DK).

22 State Grid Qinghai Electric Power Company, Innovation Centre for Industrial Big Data, and Tsinghua University.

23 Oxford PV (GB), an Oxford University spin-off, for solar PV, Best Blades (DE) for blade design, and SaltX (SE) for large-scale energy storage.

24 Global partners include aerodyn (DE), Frauenhofer Institute (DE), and ECN (NL).

25 Besides GOL and MNG, other board members are GE Renewable Energy (since 2019), ENERCON (since 2019), ACWA Power (since 2019), and Shell (since 2018).

26 The inverted u-shape describes the rapid market/technology growth in early development stages (e.g., due to conventional upgrading mechanisms such as technological licensing), followed by a rapid decline in market/technology growth (e.g., due to market restrictions or expiring access to technological designs through licensing agreement).

27 Licensing strategies between the firms varied considerably in terms of technology (1) control and (2) novelty. GOL exerted a relatively high level of control due to its majority stake in Vensys and licensed a novel technology (PMDD) that was rarely used by the incumbent firms. In turn, MNG did not acquire stakes in aerodyn and licensed the less novel slow-rotating hybrid drive. ENV did not license any technology, which is why it focused on developing new digital technology in the absence of technological path-dependencies.

28 Supervisory control and data acquisition (SCADA) comprises both hardware (e.g., sensors) and software components for the remote and real-time supervision and control of energy plants.

Additional information

Funding

The author gratefully acknowledges the financial support from the Otto Mønsteds Fond (grant Nr. 19-70-0028).