Local Government and Firm Innovation in China:
The Case of the Clean Energy Sector
Margaret Pearson, University of Maryland, Department of Government and Politics
Paper presented at the China Initiative Research Seminar, Watson Institute, Brown University, November 12, 2015.
Note to readers: while any feedback on this draft paper is welcome, I am particularly interested in the utility and formulation of the taxonomy suggested in the conclusion. Also, please do not cite or quote without explicit permission.
I. Introduction[1]
How do local governments contribute to innovation and, byextension, development in China? While scholars disagree as to whether local governments are helpful or harmful, there is a consensus that they are relevant.[2] The link between local governments and growth is well-considered in scholarship. Similarly, literature on innovation by Chinese firms is extensive,[3] as is the question of how the PRC central government policy may foster innovation. (Contributions to these literatures are cited throughout this paper). Yet the influence of local government on firm innovation is curiouslyunderexamined in studies of innovation and industrial upgrading in China. A major study of China’s innovation capacity by the OECD, for example, largely fails to mention the role of local government (OECD 2007).[4] The role played by local government might be addressed in several ways. One approach would be to look at the efficiency and innovation outcomes of funds spent by local governments, in other words, to “followthe (local state) money” and track pro-innovation outcomes such as increase in patents or measures tied to market success. While this approach would have many advantages, the lack of a clear (or transparent) trail and other design and measurement issues at the local level would pose substantial difficulties. Alternatively, one might examine variation in innovative outcomes among in-country regions to try to assess what local politicalfactors (among others) lead to greater or lesser innovation in a sector (e.g., Chen 2013, 2015; Breznitz and Murphree 2011; see Rithmire 2014). This papertakes a complementary approach. It considers how local governments in China typically do or do not foster an innovative ecosystem for firms. Under what circumstances do local governments create an environment that can enable firm-level upgrading or industry success, thereby contributing to the center’s goal of an innovative economy? In what ways might the ecosystem presented by local governments be a hindrance to firm innovation? Thus, rather than directly assessing whether innovation has occurred (a project that itself is subject to substantial debate about how “innovation”should be valued and defined), I examine how the local government ecosystem affects industries subject to industrial policy in areas Beijing hopes will be innovative.[5] I draw on scholarship on PRC central-local relations and local officials’ behavior, as well as scholarship on “innovation systems,” to show patterns of behavior that affect industry structures and other firm outcomes in a given locality.
The empirical focus is local government influence on the development of clean energy industries, specifically solar cells(PVs) and electric vehicles (EVs). Development of the clean energy sector has been a key theme of discussions on China’s response to environmental degradation and climate change. Yet clean energy has been not just an environmental policy; it is also, and perhaps foremost, an industrial policy. Despite high-level emphasis on sustainable development, the implementation of China’s clean energy industry – as a part of China’s push for innovation as well as sustainable development - has followed in the well-trod footsteps of Chinese industrial policies. It has been designated as one of the PRC’s “Strategic Emerging Industries” (SEIs) and fully incorporated into the Five-Year Plan process.[6] While this is a natural and expected process in the Chinese policy context, the envelopment of the sectors by industrial policy processes leaves a distinctive – and, I conclude, often deleterious - mark on the trajectory of China’s clean energy industry.
More optimistically, it is clear that new energy industries in China have successfully upgraded, both in terms of cost and process, and perhaps product.[7] Many Chinese manufacturing firms, including in new energy, have shown strong ability to upgrade, especially at the lower price and quality end of the market.[8] Yet there is also quite a bit of variation in innovation by new energy firms. In electric vehicle sedans, for example, Chinese firms has been very slow to convert the country’s strong auto industry to be able to supply a nascent EV market. (Bernstein 2013) Consistent with this variation, we can imagine that local governments create both helpful and harmful impacts. Drawing on analysis of these varying outcomes in the empirical cases of solar cell and electric vehicles industries, Iconclude by suggesting a simpletaxonomy of how local governments treat industries in sectors in which China’s central government hopes to foster innovation.
Background note on innovation literatures
Parameters of the relationship between governments and firm innovation are set squarely in broad debates (in economics, business and policy literatures) characterized by two dominant strands, roughly categorized as market-based and state-supported. Market-based theories on successful innovation depict the process primarily as a bottom-up one, originating with entrepreneurs, and often pivoting off firms’ access to and position in the global or regional value chain.[9] While recognizing that it is desirable for states to provide a positive environment for technology development (through proper policy guarantees of, e.g., protection of physical and intellectual property, regulation, etc.), this model emphasizes that the main locus of innovation rests in society and particularly among non-state economic actors. Private capital markets including venture capital, moreover, provide the most efficient means of allocating resources to inherently risky ventures.
The dominant competing theoretical approach to the market-based model of innovation is, of course, a statist approach that places government policy at the center. One strand of the statist approach, the developmental state literature, draws its inspiration from Gershenkron’s (1962) argument with regard to late 19th century industrial development that state institutions can usefully aid industries in their efforts to acquire capital and technology.[10] A highly capable bureaucracy at the nation’s administrative center works with major industrial firms to formulate a policy to leapfrog developmental stages that typically had been followed in Western advanced countries. A second strand of the statist approach focuses on the state’s rolein ameliorating market failure, i.e., when market mechanisms do not adequately incentivize the funneling of resources into knowledge creation and innovation.[11] Governments can therefore play a positive role not just in establishing institutions (rights to intellectual property, etc.) but also in directing public financing to firms, reducing the risk they face from failed innovation efforts. (Hall and van Reenen 2000) Similarly, governments may help produce demand for products, largely through demand-side subsidies or purchases. A third strand, the literature on national innovation systems, goes even further to suggest the benefit of more extensive government contributions, such as funding public research and development institutions, and expansive funding for public education. The statist literature pays less attention to the problem of political failure, when political concerns of government officials produce other incentives that may hinder innovation. (I find such political failure important in the cases analyzed below.)
Each of these literatures suggests clear policy prescriptions, most of which are not mutually-exclusive. In turn, China’s own innovation-related policies over the past three decades represents each of these sets of policy prescriptions; indeed, innovation-promotion policy in China actually is quite diverse, combining efforts to foster some combination of top-down industrial policy and bottom-up market driven factors. The emphasis of the past decade on “commercialization” and the core role of technology firms in innovation recognizes that market-based sources will be key, including firms already deeply engaged in global value chains. At the same time, the Chinese government clearly sees a role for the state in innovation.[12] As in the developmental state model, the party has taken pains to develop – and has been largely successful (despite tendencies for corruption) at developing - a highly capable bureaucracy at both the central and local levels.[13] Beijing pays massive attention to industrial development, using instruments consistent with the developmental state, market failure, and national innovation system strands of statist policy literature. Such attention is at the core of the Strategic Emerging Industries initiative.
Local governments play a key role in China’s SEI policies. Local state involvement revolves around local governments’ efforts to use the institutional infrastructure of industrial policy, and to respond to the center’s signals to use industrial policy instruments. At the same time, local officials attempt to respond to conditions they face on the ground. These may include “market conforming” efforts directed at helping firms build on existing capabilities found within enterprises. Alternatively, their efforts may be directed at the political benefits local officials stand to gain from showing a loyal response to the center’s signals to produce “innovation,” especially enhancing the prospects for cadre promotion. Still further, local officials are keen to have development of these industries serve other economic needs, particularlyemployment and taxes.[14] While the consequence of local officials’ desire to meet their political and development goals need not always be negative, this dynamic can provide a breeding ground for political failure in industries Beijing hopes will be innovative.
The local government - new energy industry nexus: the key roles of industry characteristics and Beijing’s signals
The two sectors examined here, solar cells and electric vehicles, are subject to Chinese industrial policy, but they differ in industry-leveloutcomes. In brief, the low end of the EV sector (low speed vehicles) has been quite dynamic, despite Beijing’s efforts to discourage it, whereas the high end of the EV sector (passenger sedans and buses) has been relatively moribund. The solar panel sector has seen substantial cost innovation, but also tremendous waste. Exploring these differences is instructive for analysis of the local government role. Ifocus on two clusters of variables, both of which set the context for local officials, and that appear in sync with different outcomes: industry characteristics (inter-related qualities of barriers to entry, fragmentation of the market and presence of strong incumbents) and strength of signals sentby Beijing. In terms of industry characteristics, the solar panel industry in its developing years was characterized by relatively low technological barriers to entry, and despite the presence of some large firms (such as Suntech, Trina and Yingli), also contained many small and very local firms. This sector also has been characterized by deep ties to the global value chain. Given extensive market competition, the main type of innovation has been in terms of cost and pace of production (Nahm and Steinfeld 2013). The electric vehicle sector is divided into large vehicles (sedans and buses) and small low-speed vehicles. The former have depended heavily on large incumbent auto companies in China, including joint ventures. In this part of the EV market, technological and sometimes protectionist barriers to entry are high; for example, many of the main technologies are foreign (especially Japanese and Korean, involving high IP costs), and in general the cost of a key component – lithium-based batteries – is high. These cost and market entry barriers have not successfully been reduced. (Bernstein 2013) The low-speed and e-bicycle segment of the EV market are quite different, and have more in common with the solar cell market although from the start they have produced mainly for the domestic market. Firms (such as Kandi and Shifeng, as discussed below) have built on manufacturing and cost advantages in their regions, and lack of foreign competition, to supply the low end of the market. This has allowed consumers who cannot afford sedans to upgrade from bicycles and motorcycles to small vehicles. It has led in some cities to new business models, such as promoting a rental and battery switching model. (These in turn are being tried for EV sedans.) Neither technology nor cost barriers to entry are as high as in the sedan sector.
The second dimension, the strength and nature of industrial policy signals – meaning primarily the level of priority in Beijing’s strategic hierarchy - also varies across these sectors.[15] The central government has pressed a “green energy” agenda that includes both solar and EVs as new energy strategic industries. As noted, this agenda is implemented primarily through the channels of China’s industrial policy, and also have been seen as venues for possible “indigenous innovation” (especially in the case of EV sedans). Local officials can benefit from showing responsiveness to Beijing’s SEI agenda. Despite these similarities, in practice Beijing has been much less supportive of low-speed EVs and bikes than EV sedans. The central government’s treatment of the solar industry has been more supportive than for low-end EVs, and for example has made limited subsidies for solar available, but in some respects has signaled its displeasure with this industry over runaway investment, particularly after export markets collapsed.[16] But, as is well known, local governments havebeen more keen than the center to subsidize the establishment of new solar panel firms for purposes of local industrial development than for innovation. Similarly, despite electric vehicles’ place as a specified SEI industry, the NDRC actively tried to squelch the development of the low-speed segment. For solar panels and low-speed EVs, local governments had more leeway to respond in ways that were market-conforming. In contrast, signals for higher-end electric sedans and public vehicles were strong,including extensive subsidies to consumers and encouragement of local governments to provide demand and infrastructure; in other words, Beijing was much more active in addressing market failure in this arena. Local governments were quite keen to respond to Beijing’s signal, and to help create the market, even as they also wished to leverage the opportunities new auto manufacturing locations would have for local growth and employment. But while needing to respond to Beijing’s signals, producers of EV sedans and buses had less opportunity to be market conforming, as there was not a market! As a result of these varying dynamics, at least in part, key differences in the local ecosystems emerge.
The remainder of this paper is organized as follows. The second section discusses the ways in which Chinese local governments might ideally provide an environment that fosters innovation, including being well positioned to play positive roles in overcoming market failure. The third section dives into what are essentially mini-case studies in the solar and electric vehicles industries that illustrate how local governments have in fact served to enable firms in innovation-related industries. It then considers examples of problematic local behavior, notably problems of wasted local investment of land and funds, extensive local protectionism and fragmentation of efforts, and problematic firm responses to subsidies and lack of adequate private financing. The final section concludes with suggesting a simple taxonomy as to how local governments may provide (or not) an innovative ecosystem.
II. Intersection Points for Local Governments in Promoting Innovative Industries
What role might local governments ideally play in fostering new and innovative industries? How might pro-innovation incentives set by the center be channeled down to local governments to produce industrial innovation and upgrading? Scholars of innovation have identified a clear role for local governments. The “regional innovation systems” approach (a subset of Nelson’s classic concept of “national innovation systems”) emphasizes that innovation – as a collective enterprise – is frequently best served when collaboration is promoted between governments, firms, and research organizations such as universities and institutes. Much as occurs with industrial production clusters, the co-location of these actors may create synergies. (Asheim and Gertler 2005; Sagar and Zwaan 2006) In theory, regional proximity can open the door for local governments to help coordinate the circulation of knowledge as well as in the promotion of strong systemic relationships between firms and a given region’s knowledge infrastructure. Moreover, to the extent that different regions contain distinctive “regional cultures,” it may be crucial for local institutions (including governments) to help coordinate central policy.[17]
Local governments also often channel economic resources from the national government to the cluster or firm. (Asheim and Gertler 2005) Yet the vision of local-government-as-coordinator is not limited to implementing policy from the national level. Local coordination can help overcome lack of trust that may be inherent between actors – actors in competing firms, institutes, and funding organizations. (Powell and Grodal 2005) Coordination can direct resources to where they are most effectively used, and prevent wasteful duplication. Local governments also are crucial for setting a proper local policy environment, particularly for market entrance and exit of firms, but also in tax policy (e.g., tax incentives related to technology zones), promotion of effective allocation schemes for funds (including public and private lending) and land, among other standard policy instruments.[18]