The bumpy road to low carbon development: Insights from changes In the innovation system of China’s photovoltaic Industry 1 Introduction Environmental scientists, energy experts and relevant enterprises assure us that the technological solutions needed for low carbon development are already available. This is good news as it implies that – at least technically – it is possible to prevent global warming beyond the ICC threshold. But what does _availability of low carbon technologies actually mean in the context of developing countries?
The mere fact that relevant technologies, for example renewable energy technologies, have been plopped by some enterprises or laboratories does not guarantee their diffusion and application. This is true for industrialized countries and even more so for the large emerging or the developing countries. In order to realize the goal of climate change mitigation within the short time left for doing so, low carbon technologies have to be rolled out much faster than it Is currently done.
In the context of the global climate change negotiations this dilemma between availability and application of technologies is regularly discussed with a focus on questions of technology ranches from developed countries to developing countries. While technology transfer may and does play an Important role In technology diffusion, another perspective on the dilemma is to address the role of the national innovation systems in enabling the development, adaptation and application of low carbon technologies in developing countries.
Three arguments support this perspective: First, technology transfer may occur Vela different venues (granting access to patent Information, foreign direct Investment, trade, licensing, Joint R etc. ) Implying different costs and risks for the owner of the technological know how. Creating win-win situations for technology transfer, or compensating the owners of technology for their costs if a win-win solution Is not found, may prove to be very difficult and time consuming.
Second, transfer of low carbon technologies even at zero costs for the developing countries would not ensure widespread application of these technologies in the developing countries. Innovation system research stresses the importance of co-evolution of economic, social, political and technological aspects of innovation. Last but not least, the observable shift In economic and political power from the Industrialized entries to developing Asia adds to doubts whether granting technologies to Asian developing countries, especially China is a smart move.
While it is acknowledged that climate change mitigation in China is of vital global interest, Chinese enterprises are competitors for _western’ enterprises in many global markets already. Thus, an undercurrent concern exists that technology transfer could add to the economic power shift by helping China to become a technological leader and exporter in low carbon technologies rather than really contributing to climate change mitigation in China. Against this background a more promising approach -? for the purpose of climate change mitigation – is to understand how to ensure application and diffusion of analyzes the development of the photovoltaic industry in China from an innovation system perspective. The innovation system approach is selected for its holistic perspective on innovation capacities and sector developments. The PA sector case is selected as one industry of importance for the energy shift envisaged in the context of climate changes.
China is a logical object of analysis due to her importance for limited change mitigation in general, and as a major force behind the ongoing =global shift’ in particular. Also the development of China’s PA sector highlights the argument of co-evolution developed in the innovation system literature: The global photovoltaic sector has experienced rapid growth and changes in recent years. Growth mainly originated from increasing demand in industrialized countries, supported by related government policies and financial incentives to promote the use of solar energy.
A major change in the sector has been China’s emergence as a producer and exporter of PA cells. Interestingly, mass PA cell production developed in China without a parallel development of PA installations. Thus, at least until 2009, the development of the PA industry in China confirmed the necessity to distinguish between technology availability and application. While the technological know-how to produce PA cells for export had been available in China, a comparatively large local market for PA installation did not emerge; the knowledge available did not contribute much to climate change mitigation within the country.
This paper argues that while the earlier successes of the Chinese PA industry abroad were mainly market driven, ejaculatory bottlenecks suppressed development of the home market. The situation has changed since 2009 due to the repercussions of the global financial crisis, an increasing emphasis on the importance of climate change mitigation and energy security in China, and also technological developments within the country. Overall, while global demand slowed, local demand increased.
These more recent developments actually highlight the impact of regulatory support (or the lack of it) for the development of the national PA sector and hence for the larger scale application of this low carbon technology in China. However, even though the Chinese government today does push PA technology use more than before, conflicting interests resulting from regional rivalry and technological path dependency still strongly influence the development of the sector in China, thus leaving some question marks concerning the future development of the sector.
With reference to the discussion on technology transfer, the paper suggests to put less stress on the question whether technologies and property rights may be infringed in the process of technological transfer. Instead it would be helpful to have a closer look on the sectional innovation system with respect to low carbon development or, as we call it, the sustainability-oriented innovation system. L If the national innovation system clearly demonstrates a propensity to support the use of low carbon technologies, cooperation in the development of the necessary regulatory framework and institutions may be at least as 1 See Oldenburg et al. Paper prepared for Globulins 2010. 3 important and promising for ensuring low carbon technology usage as technology transfer in the narrow sense. The paper is organized as follows: Chapter 2 presents a he production value chain of PA power production as a background for the in-depth analysis of China PA sector innovation system presented in Chapter 4. The paper ends with a discussion on the _ lessons learned’ both for the understanding of China’s PA sector and low carbon development and the framing of _technological transfer’ in climate change policies. Review of the literature The underlying argument of this paper is that the idea of technology transfer being a forceful instrument to promote low carbon development may be misleading. This is not to deny that the transfer of technology could be helpful. But the argument here is that a more complex set of factors is needed for the successful application and use of technologies. This argument is first of all based on insights developed in the innovation system literature.
The idea of innovation systems was originally developed in order to better understand the differences between nations in terms of innovative capacity and competitiveness (Exquisite, C. 1997). Related analysis showed the complex interrelations between enterprises, public and private actors in science, technology development and innovation, and governments as well as the importance of the institutional environment. Technological development and innovation have to be understood as a complex, messy, non-linear process related to and embedded in economic, social and political developments (Savviest 2005).
This embeddings of innovation and technology does not allow for simplistic ideas of -?technology transfers. Rather it is to be expected that technologies developed in one society and institutional environment, once _transferred’, have to be adapted to and embedded into the environment of the receiving country. The existence or creation of a certain knowledge base is as necessary as an enabling business environment (World Bank 2010). For example, the PA power development in Germany has thrived following not only the development of PA cell and grid connection technologies as such, but also due to the _feed-in law for renewable energies.
Thus, the attempt to copy the success of PA technology in Germany may less rely on the transfer of the technology as such, but more on copying the attempt to develop a supportive regulatory framework. As institutions such as laws and regulations have to be compatible to the institutional environment of the respective country, again, simply copying the German feed-in law would probably not be sufficient (Saving, J. ; Flailing, C. 2004) Another important insight of the innovation systems literature is – amongst others – that innovation is a relative concept.
Innovation can mean _ new to the world’ inventions, but more often than not an innovation is rather the application and adaptation of knowledge in a different environment. It is new to the environment, but not necessarily new to the world. This 4 is especially true in the context of developing countries (ref. ). Innovation is rarely radical but often incremental. In addition, as literature on _disruptive innovation’ has shown, the interesting innovations may result from attempts to redefine a technology y looking for -?cheaper, easier-to-use alternatives to existing products or services] that target previously ignored customers. Willis, R. Et al. 2007: 4). Hence, in terms of innovation for low carbon development and their fast diffusion and application, the most promising advances may not come from innovations at the technological frontier as such, but from innovations at the _cost frontier’ that allow to apply and sustainability transitions and systems innovation for low carbon development clearly stress the point that a change in production and consumption modes at a large scale s necessary to achieve the turn to low carbon development.
This would be accelerated by the availability of cheap alternatives to fossil fuels, more so perhaps than cutting edge innovation in some specific high tech realms (Ref. ). A special line of research in innovation systems is sectored innovation systems (Malaria 2004) which explore characteristics of innovation systems related to a specific sector. The reference point of such analysis are the specific knowledge, technologies, inputs and demand within a sector.
These features may translate into specific actors constellations and institutions. Different sectors within one country thus may develop certain differences in the respective innovation system. Different from national innovation systems, sectored innovation systems do not necessarily end at the national borders, depending on the characteristics of the sector and the involved production value chain (Sofia et al 2008: 4).
While studies in sectored innovation system generally speaking focus on the interrelations and networks between actors, studies on renewable energy and low carbon development related sectors, emphasize the importance of policy and regulation as the -?single most important river for innovation in the energy real (Burp et al. 2008:73). Foster et al. (2010:249) stress that -?without proper institutional and market frameworks to operate and maintain renewable systems long term, they eventually fail (see also EIA 2008).
Graveyard et al (2010) integrate this perspective into an analysis of the global integration of the solar PA sector along the production value chain. China’s growing importance in global markets and increasing influence in issues of global concern has also increased the interest in China’s economy and the factors defining the amounts competitiveness and its potential to become a _technological superpower’ (Grounds 2005). The promulgation of a ‘National medium and long- term development program for science and technology, 2006-2020’ in 2005 only additionally nurtured related research interest in China and abroad (MOST 2005, Cacao et al. 006). In 2008, the COED has published a large study on China’s innovation policies, attesting China impressive progress on the road to a _ knowledge economy (COED 2008). The main weaknesses of China’s innovation system identified were the legacy of the planned economy still reflected by the relative importance of Tate-owned enterprises, rather weak linkages between the private business sector on the one hand and public research institutions/ universities on the other (Lieu 2009: 138, Landfall/Gu 2006).
Also the propensity of Chinese enterprises to invest in R tends to be rather weak which has been associated with the still weak protection of intellectual 5 property rights and strong competition within an environment of weak competition rules, I. E. A lack of an antitrust law (COED 2008, Lie 2008). 2 With regards to the development of China’s renewable energy innovation system in general and the hotfooting sector in specific, there has been an increase in studies published outside China in recent years.
These have been induced by legislation and industrial policies as well as the sudden rise of Chinese enterprises as competitors in the global wind and solar power markets (see for example The Climate Group 2010). The market or the thrust of Chinese competition in terms of technological development and energy mix targets (I. E. Lieu et al. 2010, Marino 2006). An exception is an COED study of 2009 which analyzes policies for CEO-innovation, including to a certain extent alarm energy.
Also Marigold Fox/ Pearson (2007)) look into the role and technological know how of Chinese manufacturers in PA cell production from an innovation system research perspective. Naturally there is a vast body of Chinese language literature on China’s innovation system, renewable energy sector development as well as PA sector developments. Again the majority of this literature is not theoretic, but either business oriented market research or shorter reports in popular media reporting on sector developments.
This latter type of sources is actually of great value for this paper, as these reports reflect rather well the internal discussions in China on policy options and sector problems, sometimes being more open than academic articles on the subject. The number of Chinese academic books and articles reflecting more recent sector developments is actually not that large and often rather descriptive. A weakness shared by all publication on the PA sector related publications is the scarcity of reliable data.
This is naturally true for data of 2009 and 2010 for which no official statistics are yet available, but also for earlier data, as renewable energies eave for a long time not been part of the official energy statistics. The standards for assessing the PA sector statistically do not seem well elaborated, yet, Just as the delineation of different types of use of solar energy (solar water heating, solar thermal, on-grid and off-grid PA power, building integrated solar power production, stand alone PA power stations etc. Are sometimes mixed and sometimes separated in statistical accounts. Given the developments in recent years, some of the studies (and prognoses) on China’s PA sector have already been outdated by factual placements. But apart from this, most studies on China’s PA power sector – both published in Chinese or in English – fail to highlight the embeddings of the sector’s development into the overall economic policy framework and industrial policy logics of China more general.
The interesting questions (and presumably answers) related to the PA power sector innovation system arise from a more closer look into major regulatory shifts and their impact on sector development, from a comparison with other sectors (for example wind) and from a look at the different administrative bevels and how national policies are translated in local policies and strategies. This paper concentrates on the 2 After a preparatory phase of about 20 years, the State Council finally adopted an Antitrust Law 2007 which took effect in August 2008 (COED AAA). First question that is the insight drawn from major policy shifts, especially the interesting consequences for the more general debates on technology transfer. Some analysis into the role of the regions will be included as the different regional strategies in reaction to central level policies may serve as a proxy to assess the impact of central level policies at a time when statistics are not yet available. An in- depths comparison with the wind sector will not be undertaken though in principal the wind sector is an important reference point to highlight the modest approach to PA sector development taken by the Chinese government until 2009. 3 Technologies and PVC of PA power generations to look at the sector from a production value chain perspective. The production value chain of the PA industry differs slightly depending on the raw materials used. The core of the PA energy production value chain is the solar cell. Solar cells are composed of various semiconductors materials, but over 95% of all the solar cells produced worldwide are composed of the semiconductor material Silicon (S’).
Three cell types are distinguished according to the type of silicon used: incontestable, polycrystalline and amorphous silicon cells. To produce a incontestable silicon cell, absolutely pure semiconductors material is necessary. Incontestable rods are extracted from melted silicon and then sawed into thin plates. This production process guarantees a relatively high level of efficiency in solar energy transformation. In contrast, the production of polycrystalline cells is more cost-efficient.
In this process, liquid silicon is poured into blocks that are subsequently sawed into plates. During solidification of the material, crystal structures of varying sizes are formed, at whose borders defects emerge. As a result of this crystal defect, the polycrystalline solar cell is less efficient in transforming solar radiation into energy. If a silicon film is deposited on glass or another substrate material, this is a so-called amorphous (a-Is) or thin film cell. Silicon is a raw material that is not scarce as such, but has to be reified for PA cell use.
In the early times of solar cell production, purified silicon for PA cell use was taken from scrap that came out of silicon production for chips/ integrated circuits in the electronics industry. With growing global demand for polycrystalline silicon for PA use, special production facilities became necessary as the amount of scrap left over from the electronics industry was no more sufficient. Purified silicon is first formed into wafers which are used for producing the solar cells. Finally, solar cells are integrated into modules.
Thin film cells are mostly based n silicon, but other semiconductor materials (for example copper indium discipline (CICS, GIGS) and cadmium telluride Cite) are used or tried as 3 Policy support for wind energy development became substantial already in 2003 (Ghana 2010). 4 See Graveyard 2010, Foster et al. 2009, pa Group 2009, El-reroute et al. 2009, yang 2003) 7 well. Thin film modules are constructed by depositing extremely thin layers of photosensitive materials onto a low-cost backing such as stainless steel, glass or plastic.
Crystalline silicon and thin film technologies to produce modules for PA yester differ in the complexity and energy intensity of the production process, the environmental impact of the production process (or costs to limit these impacts), and in their efficiency of translating sunlight into energy. For example, the production costs of thin film cells are – generally speaking – lower due to the lower material costs. However, the efficiency of amorphous cells is much lower than that of the crystalline silicon cell types.
Because of this, amorphous cells are primarily used in low power equipment (watches, pocket calculators) or as facade elements. The technological reorient, as reflected by patenting rates, is concentrated in material sciences to improve thin film efficiency. In addition, advances in material science have opened new avenues for PA technology development such as annotate and dye-sensitizes approaches (which can be painted on to surfaces) (Graveyard et al. 2010; see also Mean et al. 2009). At the upstream end of the production value chain, in the case of conversion into wafers requires substantial investment and technical knowledge.
As a result, the number of enterprises engaged at this level of the production value chain is rather small. Cell and module production on the other hand is less knowledge and investment intensive, hence the number of producers is much larger. Further down the production value chain, crystalline silicon modules and thin film modules do not differ much. The additional PA systems components (Balance of System components, ASS) such as inverters and battery materials are important components in the PVC of either technology for PA power production.
Service and installation of PA systems are usually in the hand of smaller local businesses. Solar energy produced by PA modules can be used in stand-alone technologies, I. E. Ells integrated in electric or electronic products in order to produce the energy needed for this specific product. Small grid applications are also stand alone solutions as the solar energy is fed into a small local electricity network, but not connected to a larger grid which also relies on input from other energy sources.
Current discussions on the use and potential of solar energy often refer to PA power fed into a larger electricity grid. On-grid PA power can be produced in specific (larger) PA power plants or by PA power installation connected to existing buildings (on roof tops etc. ). Grid connection f PA power production implies considerable technological know how in grid management as PA power input is not produced on demand but depending on solar radiation. Hence this may imply technological upgrading of existing grid infrastructure.
This paper ultimately is interested in grid connected production of PA power in China but refers to other aspects of solar energy use were necessary for putting PA power development into perspective. 8 4 China’ PA sector development 4. 1 Early development of the Chinese PA sector The roots of the photovoltaic industry in China date back to the sass when the first attempts in research were undertaken. Until the end of the sass, interest in photovoltaic research was mainly nourished by the potential use of photovoltaic power production in space.
The 6th Five-Year Plan (FYI) (1980-1985) was the first to include photovoltaic science and technology projects. A *round solar collaborative group’ was established in 1983, as was the China Optoelectronic Technology Center, the first specialized research center. Research in crystalline silicon cells and related technologies was then encouraged during the 7th FYI (1986-1990) and the first regional Renewable Energy Association was established in Annum in 1992. Still, the cake-off of China’s photovoltaic industry only started during the 9th FYI (1996-2000).
This was backed by a development strategy for renewable energies in the period from 1996 to 2010 (ref. ), the China Light Project (1997) and support by the World Bank and the KEF for the centralization of renewable energies in China, a project that started in 1999. It was also reflected, amongst others, in the founding of associations for renewable (and new) energies in all major provinces in the years after 1998. During the sass and early in the 21st century, the industry focus was on solar consumer goods production (like garden lamps etc. And electrification of remote rural areas.
The 2002 state policy -?Send electricity to the village” spurred progress in photovoltaic power generation to a certain extent, though at a still low of which was used by telecommunication and industry, 51 per cent was used as electricity in rural and remote areas, 9 per cent for solar powered consumer goods and only 4 per cent was fed into the power grid. Also in 2003, China became the largest producer of solar consumer goods globally. By 2010, eleven PA consumer goods producers were listed at foreign stock exchanges, reflecting China’s growing importance in the global market for PA cells and modules.
It has been the fast rise of China’s solar cell producers that recurrently resulted in media headlines of China becoming the world’s leading solar cell producer, a _threat’ for other countries’ PA industry, a =global green tech leader’, and that – as a result – green tech trade wars may be around the corner (Stokes 2010). This short summary seemingly describes an easy to grasp success story of China’s photovoltaic sector. But, hidden in the account is an important paradox of solar energy development in China: While the production f PA cells and modules thrived, PA energy generation did lag behind.
As can be seen from table 1, annual output of PA cells grew rapidly since 2002, while PA power annual installation only managed to surpass the one time high threshold of 2002 (resulting from the rural electrification policy) again in 2008. In terms of global PA cell production China had a 32,7 per cent share of the global market, whereas 5 China’s total energy consumption in 2003 was about 19000 GO (Pu, Gong 2008: 214. ). 9 annual PA power installation only accounts for 0,7 per cent (2008) of global installed PA power (PA Group 2009, Robber’s 2010).
Table 1: Annual output of PA cells and annual PA power installation in China 2002-2009 2002 2004 2005 2006 2007 2008 2009 Annual output of PA cells (Map) 10 50 200 370 1087 2589 4676 Annual PA power installation (MAW) 20,3 5 20 45 160 Source: China’s PA Industry Report 2006-2007; European Photovoltaic Industry energy is large and the potential for using solar energy huge, the prognoses for future PA power installation in the mainland remained low: As lately as 2007, the National Development and Reform Commission (ENDER) still envisaged a cumulative PA power installation for 2020 of only MOMMA.
This target roughly equaled the PA ewer producing capacity of the PA cells produced in China in the year 2008 alone. It could hardly pass as an ambitious target. Figure 1: Annual and cumulative installed PA power in China source: yang (2009), ENDER 2007 110100100010000100000Annual pa power incalculableness pa power installedPrognoses2010 predicted target for 20202008 predicted target for 2010 2007 predicted target for 2020 10 Since 2009, the official target for cumulative power installation for the year 2020 has been corrected upward by a dramatic degree.
Currently the less ambitious target discussed is 20 GO installed power, while a more daring target of 30 GO is also mentioned sometimes (21st Century Economic Herald (chin. ), 4. 8. 2010). This shift does already and will further influence the development of the PA sector in China. As it is not only a numerical target, but supported by a sudden spur in PA power installation related policy-making at all administrative levels, it underlines the hypothesis that certain incentives and regulations are necessary in order to ensure deployment and use of low carbon technologies (ref. ). 4. Upgrading along the PVC Chinese enterprises entered the PA power industry at that level in the global PVC where they possessed the greatest competitive advantage, I. . The comparatively low cost production of solar cells and module assembly. This cell and module industry experienced a rapid take-off in the early sass when two major enterprises, Sunsets Power and Twinge Youngling New Energy Resources, began production. Between 1997 and 2005, while global production in PA cells grew at an average annual rate of 36 per cent, the annual growth of production capacity in China nearly doubled this, growing 70 per cent (Marino 2006).
In more recent years (2005-2008), global production growth outside China was about 33 per cent annually, while China’s PA cell production still grew about 58 per cent. As a result, China’s market share in global PA cell production reached 33 per cent in 2008 and 38 per cent in 2009 (Reverser 2010). Until the mid sass PA cell production in China was mainly for domestic use, but this changed completely afterwards. By 2009 around 95 per cent of PA cells were produced for exports (Lie 2010: 38).
In terms of production capacity, four of the ten largest PA cell production companies globally today are of Chinese origin (Sunsets, Youngling Green Energy, JAG Solar and Train Solar) (Reverser 2010). Most Chinese PA enterprises today are vertically integrated along the core part of the reduction value chain, I. E. They produce wafers as well as cells and modules. Until very recently, though, China did not have any substantial or competitive production of silicon feedstock (Lie, D. 2009).
Globally the market for mono- and polycrystalline silicon used to be dominated by seven large firms located mainly in Japan, the US and Europe. Chinese producers had to import about 95 per cent of the purified silicon used in their PA cell production (Lie 2010: 38). Given the importance of imports of raw materials and the large share of production for exports, China’s role in the global PA reducer. As such the PA industry fitted into the popular image of China as _factory of the world’.
From the Chinese perspective the PA sector thus is another example of an industry that developed without ownership over the core technological know (He 2010). The lack of core technologies for the production of crystalline silicon usable for PA production as well as the lack of know-how to produce high quality balance of systems 1 1 components such as power inverters was still mentioned as the major concern of the industry by Chinese industry and research representatives interviewed by the author n early 2009 (see also Mum, Lieu 2009:178).
The industry feared that PA cells produced in China would not be competitive in the long term due to the dependence on imported silicon feedstock. However, a change of this situation was already under way. In reaction to the import dependence for silicon feedstock as well as the fast rising prices for purified silicon during the first decade of the new century, and in anticipation of a PA market expansion related to legislation in many industrialized countries supporting renewable energies, numerous Chinese enterprises started to invest in production lines for crystalline silicon.
Statistical data on China’s polycrystalline silicon production differ in detail, but they all reflect the same development. In 2005, while national production only amounted to 60 or 80 tons, the demand for the PA cell production was above 1600 tons. In 2008, the national production had expanded to 4000 to 5000 tons but was lower than the national demand of that year. However production capacity was already said to be around 20000 tons (Shogun Dianna ABA (ZED), 21. 05. 2010), indicating a huge growth in production capacity (but also problems in producing high quality polycrystalline Alison) (Lie, D. 009). At the same time, according to official estimations, production lines were in the pipeline that could add another 100000 tons of production capacity. By mid 2009, with these capacities gradually entering the market, discussions about overcapacity in crystalline silicon production dominated the local market discourse (ZED, 21. 05. 2010). The boost in crystalline silicon production was a move to lessen the dependence on imports and to thereby secure and enhance the competitiveness of the crystalline silicon Chinese PA cell and module production in the booming global racket.
This was supported by related industrial policies formulated in the course of the 10th FYI (ENDER 2007/4, ) that encouraged the development of expertise on PA power technologies not yet available in China, including silicon purification processes and Balance of System components, via research and development as well as cooperation with foreign enterprises. At that time the policy support for the industry was hardly driven by energy and climate change related policies or by local consumer demand. Actually, skepticism concerning the potential and advantages of expanding PA power use prevailed in China.
However, supporting silicon purification and BOSS components technologies still fitted well into the main policy agenda. At least since 2006, one overall target of economic as well as SIT policies was the development of _ indigenous innovation’ in order to increase Chinese ownership in core technologies (ZED, 8. 12. 2009). Global market prices for polycrystalline silicon plunged in the first half of 2009 due to the expansion of capacities within and outside China. This development contributed to price declines for crystalline silicon PA cells
