Indigenous innovation vs. teng-long huan-niao: policy conflicts in the development of China’s flat panel industry

1. Introduction

As labor costs have risen in recent years, China has been confronted with a daunting challenge to upgrade its industry. Two major policies have consequently been initiated, one by the central government, and the other by some local governments. The policy of “indigenous innovation (IIN in short),” initiated by the central government, was aimed at replacing the original model of factor-accumulation growth with innovation-driven growth. The catchword of the policy was to replace the “Made in China” label by the “Created in China” mark. Meanwhile, the policy of “teng-long huan-niao” (TLHN for short), which literally means emptying the bird cage to host new birds, was initiated by a few local governments in the coastal areas. The policy was aimed at attracting new investments from foreign countries to replace old ones that were no longer desirable. It was an attempt to restructure the local industry by restructuring foreign investments in the region. The goal of the two initiatives was the same, namely, to upgrade the existing industry to a higher level of value-added. However, IIN carries an ultimate policy intention of making China independent of foreign technologies. Along with it, a host of policy incentives were created in favor of domestic firms which own indigenous technologies. In a nutshell, IIN amounts to a “technology autonomy” policy which tends to restrict foreign investments or to constrain the behavior of foreign investors. By contrast, the TLHN policy carries with it a host of fiscal incentives to encourage foreign investment, albeit a new kind of foreign investment. IIN uses the domestic market as leverage to attract foreign investment provided that foreign investors can contribute to the grand objective of creating indigenous technologies. Most policy incentives, however, are designed to encourage R&D by indigenous firms to create their own technologies. By contrast, TLHN is built on the foundation of a regional disparity in China where some local governments have been offering superior institutions, infrastructure, or government services to attract foreign investment.

In this article, we study the Chinese flat panel industry to elucidate the policy conflicts between these two initiatives. The conflicts arise from the differences in their approaches to technology upgrading, and in their relationship with multinational corporations (MNCs). The IIN policy embraces an ambition of technology leapfrogging, while the TLHN policy endorses an incremental approach to improving local technologies. The former wants to contest MNCs, while the latter values a cooperative relationship with MNCs. Two distinctive policy regimes have given rise to two distinctive groups of firms that were supposed to be the vehicles for policy implementation. The results are overcrowded investment and a segmented market structure with very slow progress toward the objective of industry upgrading. When the central government intervened to make IIN a dominant policy, it encountered difficulties in aligning the interests of the upstream industry in which indigenous technologies were to be created with the interests of the downstream industry in which the technologies were to be applied. Because of path dependency in industrial development, a top-down policy such as IIN may actually impede the efforts of industrial restructuring at the regional level.

2. Indigenous innovation

The idea of “IIN” grew out of the experience of implementing a “market for technologies” policy following the economic reforms that were launched in 1978. “Market for technologies” is a policy to trade domestic market access for foreign technologies (He and Mu, 2012). It requires foreign investors to transfer certain technologies to China in exchange for access to Chinese markets. Foreign investors are typically required to enter into a joint venture with domestic firms, which in most cases are state-owned enterprises, and local joint-venture partners are expected to be the recipients of technologies targeted. To facilitate technology transfer, foreign investors are required to apply the targeted technologies in local production, to train local employees in the use of such technologies, or to transfer relevant technologies to local suppliers. The policy has been implemented in several industries, including telecommunications, automobiles, and color television (TV) receivers. Whether it has resulted in a real transfer of technologies has been debated, however¹ (Chu, 2011; Nam, 2011; Wei, 2011).

Expecting MNCs to transfer technologies is common among developing countries and China is no exception. Unlike other countries, China has one of the largest domestic markets in the world to leverage the policy. However, China shares with the other developing countries dissatisfaction with the results of technology transfer. There are two typical complaints by developing countries about the technology transfer from MNCs. First, technologies remain in the hands of expatriates or headquarters and are not transferred to local employees at all. Second, the technologies transferred are inappropriate in the sense that they do not fit local factor conditions (Fu et al., 2011). In China, there is a specific complaint by policy-makers that MNCs only transfer peripheral technologies, and not critical technologies (Ernst and Naughton, 2012). Without critical technologies, Chinese firms cannot operate independently without constantly seeking help from MNCs.

In October 2005, the fifth plenary meeting of the 16th Party Congress of the Chinese Communist Party passed the party platform for the 11th Five-Year Plan (2006–2010) in which science and education were placed as the cornerstone for the nation’s further development and IIN was slated as the major strategy for scientific development. In line with this strategy, in January 2006, in the “Medium- and Long-term Plan for the Development of Science and Technology,” the Chinese government laid out a new policy initiative for “IIN” aimed at creating locally owned technologies. IINs were defined as innovations that result in locally owned intellectual property rights. In a State Council document released later on, IINs were envisaged to take three forms: (i) innovations that genuinely originated in China, (ii) integrated innovations, meaning fusing together existing technologies in new ways, and (iii) re-innovations or assimilations and improvements upon imported technologies (State Council, 2007). Apparently, the objective of IINs goes beyond technology transfer, which normally means absorbing and practicing foreign technologies. “IIN” wants transferred foreign technologies to be used as a foundation to create new technologies, of which the intellectual properties are locally owned.

Indeed, if the technologies transferred from foreign countries to apply in a local industry become a foundation for IINs, the local industry will upgrade itself over time and the host country can cede its dependence on foreign technologies. Japan and Korea both went through this kind of process and were considered role models of the import-substitution industrial development policy (Amsden, 1989; Ranis, 1990). However, if the critical technologies are not transferred, absorbed, or cannot be regenerated by the local industry, the technology-recipient country will continue to depend on foreign technologies. As the industry evolves, the old technologies become obsolete and a new round of import substitution policies is needed to obtain new technologies from the advanced countries. The technology-recipient country is therefore trapped in a state of technology dependency.

Empirical studies have shown that whether foreign technologies will be transferred and absorbed by local firms depends on many things, including absorptive capacity and the nature of local market competition (Cohen and Livinthal, 1989; Blomstrom et al., 1994; Keller, 1996). A permissive market environment, which allows inefficient firms to survive due to a lack of competitive pressures, discourages technology transfer by MNCs and slows down the learning processes of local firms. Japan and Korea succeeded because they dispelled this permissive environment (Kim, 1997). IIN goes beyond a successful technology transfer and absorption. It is a process by which local firms master the adopted technologies and thereupon introduce new technologies (Amsden, 1989). It comes out as the product of a system of local institutions, encompassing firms, universities, industrial research organizations, government policies, etc., and cannot be explained by a single factor (Nelson and Rosenberg, 1993).

There are multiple means of transferring foreign technologies, and the means may affect the end purpose of generating IINs based on transferred technologies. However, the success models also appear to be multiple. For example, until the mid-1980s, Taiwan had depended mostly on foreign direct investment and technology licensing to acquire foreign technologies, whereas Korea had depended on importing capital goods, especially in the form of turnkey plants, as the major source of foreign technologies (Westphal et al., 1985; Kim, 1993). Japan, up until the early 1970s, had restricted imports and foreign direct investments to force technology licensing from the advanced countries (Odagiri and Goto, 1993).

Alternative approaches to technology transfer may have been bound by the distinctive local institutions of the respective countries, but it is hard to judge which approach is more conducive to IINs. Taiwan, Korea, and Japan have all had strong local institutions to transform foreign technologies into indigenous technologies. In the case of Taiwan, government-sponsored research institutions were the backbone of this transformation. In the case of Korea, big business conglomerates played the key role, while government-sponsored research organizations were auxiliary. In the case of Japan, big businesses often teamed up with small firms to absorb, diffuse, and recreate imported technologies in a cooperative way (Okada, 1999). However, in all three cases, imitation and learning preceded innovation, and there was a common policy to invest aggressively in local technology bases. The R&D expenditure as a percentage of gross domestic product (GDP) started to accelerate in the 1980s for Japan, and in the 1990s for Korea and Taiwan. Today the three countries boast some of the highest R&D expenditures and numbers of patents granted in the world. All have built up a capacity for IIN but, at the same time, they have continued to import foreign technologies for local production (Hobday, 1995; Mathews, 2006; Park and Lee, 2006).

The advantage of technology acquisition through licensing or importing capital goods as opposed to foreign direct investment is that the acquirer can freely decide how to use, modify, or add value to the acquired technologies to suit local conditions. By doing so, imported technologies can be transformed into indigenous technologies. In the case of Korea, this was achieved through reverse engineering, assimilation, and immense local R&D efforts (Kim, 1992, 1993). During the planned economy era from 1949 to 1978, China also acquired many foreign technologies from the advanced countries, first from the Soviet Union and later from Germany, Japan, and the United States. Technology acquisition took the form of government-arranged technology transfer programs, prototype copying, or the purchase of complete plants. The approach was similar to that of Korea in the sense that foreign direct investment (FDI) was conspicuously rejected in favor of technology imports. However, owing to the lack of absorptive capacity and the insistence on self-reliance, the internal diffusion of acquired technologies was limited, not to mention local innovations that were based on imported technologies (Heymann, 1975). China placed much more emphasis on technological self-reliance than Korea, which had limited the scope of technology assimilation. Compared with Korea, China also benefited less from foreign-trained scientists and engineers as a conduit of technology transfer.² The current policy of IIN indeed has its roots in the planned economy era where technological self-reliance was stressed. Although the current policy did not rule out making use of foreign technologies, IINs were considered by policy-makers to be an alternative to importing foreign technologies (Zhou et al., 2007: 29). However, given that China is a latecomer in terms of technology advancement, indigenous technologies are more likely to be diffused if they are compatible with the global technology ecosystem (Liu and Cheng, 2011). This means that the indigenous technologies to be pursued should be an enhancement of, rather than a substitute for, foreign technologies.

3. Teng-long huan-niao

TLHN was a policy phrase first used by local government officials in Jiangsu and Guangdong provinces (Asia Business Council, 2011). The two provinces are the most developed provinces in China today, this being largely attributable to successful foreign investment policies implemented by the local governments since the economic reforms began in 1978. Most foreign investments in these provinces operated in the export-oriented labor-intensive industries. As labor costs rose rapidly, these industries were losing their export competitiveness. In order to sustain the growth momentum, they needed to attract new investments that could bring higher value to the region. Therefore the theme of TLHN is to abandon old investors (old birds) and to welcome new ones (new birds).

Compared with the IIN policy which uses the privilege to access the domestic market as the leverage to attract foreign investment, TLHN uses local factors, namely, land and labor, as the leverage. Local governments in Guangdong and Jiangsu have been able to attract foreign investment by offering cheap land combined with cheap local labor, including migrant workers from other provinces, to generate good benefits for foreign investors. In fact, foreign-owned firms depend heavily on migrant workers to keep their labor costs down. This model started to crumble when it became increasingly costly to hire and to retain migrant workers. Increasing costs of migrant workers did not fall on the employers only, but were also shouldered by the local government in the form of social security provisions, which have become increasingly difficult to escape from.³ Therefore the TLHN policy had a double-purpose of increasing the land value and decreasing the dependence on migrant workers. For example, TLHN initiated by Guangdong called for a double-transformation at the same time: industry transformation and labor transformation (Guangdong Provincial Government, 2008). Industry transformation was aimed at attracting investments with higher value-added. Labor transformation was aimed at reducing the number of migrant workers in the industrial heartland of Guangdong, i.e., the Pearl River Delta. The actual policy actions of the Guangdong Provincial Government include (i) raising the entry barriers for labor-intensive industries; only higher value-added investment projects will be allowed; (ii) encouraging the existing industries to diversify away from the Pearl River Delta and into the rural areas of Guangdong; and (iii) encouraging skilled workers to migrate to the Pearl River Delta to support the new industries which are expected to be skill-intensive. In other words, land in the well-developed areas is to be reallocated to the projects that promise to generate higher-value products. Low value-added production projects are to be pushed to the peripheral areas of the province. Along with this transformation, unskilled labor is to be pushed back to the rural areas while skilled labor is to be drawn to the central cities. Therefore, an industrial transformation is to be accompanied by a labor reshuffling.

Presumably, high-value production is based on high technologies. Therefore, TLHN shares the vision of IIN in terms of technology upgrading. In fact, the two policies have sometimes been mixed together in actual implementation. For example, in Guangdong, foreign-owned labor-intensive operations are allowed to stay in the Pearl River Delta if they establish R&D divisions, and selected state-owned enterprises in the region are subsidized for their R&D expenditures. However, the two policies are also conceptually distinctive. IIN wants technologies to be indigenous, meaning locally owned, or at least locally controlled. By contrast, TLHN is not concerned with the ownership or control of the technologies, as long as they are applied in local production. China is not alone in its concern over what is actually locally owned or controlled in a globalized world. Whether ownership matters to economic welfare was an issue debated in the United States in the late 1980s. Robert Reich, a central figure of the debate, has argued that what really matters is value created by the people that make up the country rather than the ownership of corporations, technologies, or production (Reich, 1991).

Reconciliation between the two policies can be found if IIN serves as a supplemental incentive to TLHN. If, on the other hand, IIN dominates TLHN, policy conflicts arise. Because IIN leverages the domestic markets, it is necessary to limit market entry so that domestic markets can be allocated to potential beneficiaries. Companies achieving IINs are also awarded policy preferences such as a priority in government procurement and outright financial subsidies. Meanwhile, TLHN has no stake in the domestic markets. In fact, for historical reasons, most foreign investors under the TLHN regime are export-oriented. Limiting market entry has never been a policy option. On the contrary, a clustering of competing companies in the same region may even create external benefits (Marshall, 1920). In fact, the Guangdong government has explicitly indicated that industrial clustering is an integral part of its TLHN initiative (Guangdong Provincial Government, 2008). Industry clusters are also conducive to local innovations (Porter, 1998), but the causation runs from clustering to innovation, and not the other way around.

In implementing the TLHN policy, local officials in Guangdong and Jiangsu identified the old “birds” (existing foreign-invested companies) to be chased out of the cage and encouraged them to relocate to less-developed regions within the province. Foreign investors were compensated for the release of the land that they originally occupied under long-term leasing contracts. They might have even been subsidized for the moving expenses (Asia Business Council, 2011). The empty land was then re-contracted with a new investor, presumably at a higher price. Therefore, the TLHN policy also enhances land value, from which many local governments derive their major fiscal revenues. In that sense, TLHN was implemented by the local government for their own fiscal purpose. The policy also takes advantage of tax concessions (a 15% business income tax as opposed to a 25% normal rate) offered by the central government to high-tech industries. It may also bank on the IIN initiative of the central government if the latter policy reinforces its primary objective of land reallocation. Whether new investors contribute to IIN is a secondary issue.

Because TLHN seeks to lease the land at a higher price, the local government cannot offer cheap land as an investment incentive anymore. Meanwhile, since labor costs in the TLHN-implementing regions are already higher than in the rest of the country, the only advantage that local governments in these regions can leverage is the existing industry base. In other words, new investments that complement the existing industry base have a better chance of being the new “birds.” In terms of the industrial policy, TLHN is analogous to a backward-integration development strategy adopted by many developing countries. Given the fixed resources, backward integration of a certain industry will necessarily crowd out other industries in the region. Therefore, TLHN should also lead to a more concentrated industry structure. By selecting new investments, TLHN also selects industries. Given the path-dependency nature in technology development, TLHN seeks to upgrade the technology level in accordance with the existing technology base in the region. Because MNCs dominate the local technology base in these regions, TLHN will not succeed without the cooperation of MNCs.

By contrast, IIN seeks to break away from the existing path of technology development and embarks on a new course in which local firms are in the driver’s seat. In order to do so, a lot more R&D resources need to be expended and local firms have to take charge. When foreign technologies are needed, MNCs are coerced into cooperation with local firms. For example, in the case of the telecommunications industry, China adopted a unique standard to force foreign equipment makers to cooperate with indigenous firms, using local markets as the leverage (Yan, 2007; Liu and Cheng, 2011). In implementing the IIN policy, market entry is limited so as to create a privilege for the market access and only selected firms are awarded the privilege. IIN is therefore not only an industry-targeting policy but a firm-selection process as well.

IIN and TLHN also differ in terms of their relations with the existing industries. In order to establish an indigenous technology, which is untested in the market by definition, existing industries must bear the risks of using it. This can happen only if the industries operate in a permissive market which confers excessive profits. For example, in the case of the flat panel industry to be discussed in the following section, the indigenous technology is to be promoted at the expense of panel-using industries, such as TV receivers, computers, tablets, and smart phones, unfortunately all of which operate under cutthroat competition in the domestic market. There are no excessive profits to absorb the learning costs. By contrast, TLHN evolves from an export-oriented development strategy whereby the role of the policy is to enhance the export competitiveness of the industry existing in the region. TLHN aims at bringing new capabilities to complement the local export industry, or building new strengths on the existing strengths. We can imagine that it will be more difficult for IIN to succeed because it has to realign the industry interests, whereas TLHN only reinforces the existing interest structure.

4. Policies to develop the flat panel industry

China has used an import substitution policy to successfully nurture its color TV receiver industry. China applied a high tariff rate to color TV receivers beginning in the early 1980s. Within the tariff wall, China imported related technologies from Japan and other countries, and by the late 1990s, domestic TV makers, such as TCL, Konka, Skyworth, and Changhong, had captured most of the domestic market shares from foreign brands. Total production of color TV receivers reached 38 million sets in 2001, accounting for a 23% share of the world total. In the process of industrial development, it is believed that China created some indigenous technologies through its own R&D efforts by local firms and state-sponsored research institutions (Gao and Tisdell, 2004). However, roughly at the same time, a technology revolution in the TV receiver industry started to take hold: flat panels made of liquid crystal and plasma began to replace the traditional cathode ray tubes (CRT) as the means of image display. As it later turned out, the liquid crystal display (LCD) panels came to dominate the plasma display panels (PDP) in terms of commercial applications. LCD technology was first developed and commercialized by Japan. It was later diffused to Korea and Taiwan through technology licensing arrangements and autonomous R&D investments in the two countries. These three countries virtually monopolized the production of LCD panels in the world. As a result of the technological revolution, Chinese market shares of color TV receivers, which were once firmly held by domestic producers, started to be chipped away by Korean and Japanese brands made of flat panels. Lacking a domestic flat panel industry, local Chinese TV makers had to import flat panels from Korea, Taiwan, or Japan. The Chinese authorities considered the absence of a flat panel industry to be the main cause of the dwindling market shares in the TV receiver industry and some kind of industrial policy was desirable.

A series of policies were initiated, beginning with the “11th Five-Year Plan” which covered the period 2006–2010. In 2006, a sub-plan for the development of the information industry (xinxi chanye changqi fazhan gangyao 2006) within the 5-year planning period was released. The sub-plan called for the promotion of the digital TV industry, which uses flat panels to display images, with the aim being to “develop technologies and products with indigenous intellectual property rights to meet the basic demand in the domestic markets.” The wording of the document reflects an IIN initiative and suggests an import-substitution nature of the policy. The flat panel display was not explicitly mentioned in the plan, although it is understood that flat panels are essential components of the digital TV receivers. In line with this plan, beginning September 15, 2006, tax rebates on value-added tax (VAT) paid on the intermediate goods that went into the production of flat panel TV receivers were increased from 13% to 17% upon exporting.

In 2008, the Subprime Mortgage Financial Crisis that originated in the United States hit the Chinese economy hard. The Chinese authorities launched a “Consumer Electronics Going to the Villages” (jia-dian-xia-xiang) program as part of the stimulus package aimed at boosting domestic demand amid the global recession. The program offered a subsidy to rural consumers for the purchase of consumer electronics, including TV receivers. The subsidy was not particularly designed for the promotion of the digital TV industry or flat panels, as traditional TV receivers were also eligible for the subsidy. Nevertheless, it was a program designed to assist the domestic TV makers, as their market share was falling owing to foreign competition.

A more concrete policy initiative regarding the development of the flat panel industry was launched in 2009. In February 2009, the State Council issued the “Plan to Adjust and Promote the Electronics-Information Industry” (dianzi xinxi chanye tiaozheng yu zhenxing jihua 2009), with a focus on the flat panel industry, calling for “a comprehensive program to overcome the bottlenecks of the flat panel industry, improve its production system, make use of global resources, strengthen cross-strait industrial collaboration, and achieve technological breakthroughs.” It is interesting to note that the technological resources in Taiwan were noticed and conspicuously targeted by the plan. The State Council encouraged local governments to provide financial support to strategic investment projects in the flat panel industry. The relevant projects could also seek financial support from the central government, which was a strong incentive for the local government to implement the said policy.

Following this policy instruction, the National Development and Reform Commission (NDRC) issued an ordinance in the same month to spell out the strategic investment projects that would qualify for the financial subsidy from the central government. The financial subsidy could take the form of an investment subsidy or interest subsidy. The amount of the subsidy could go up to 200 million renminbi (RMB), beyond which the subsidy could be supplanted by direct equity participation.⁴ Investments in the following four categories were eligible for a financial subsidy: (i) large screen (32-inch or wider) color TV receivers made of flat panel displays, (ii) flat panel display [including thin-film-transistor (TFT) LCD, PDP, organic light-emitting diode (OLED)] modules beyond the sixth generation, (iii) upstream materials for the flat panels, and (iv) machinery and equipment used in the production and testing of flat panels. The ordinance mentioned that the objective of the program was to establish a self-sustained capacity for the development of the local color TV receiver industry.

In line with this ordinance, the tariffs and value-added tax concessions on imported inputs for the flat-panel industry were extended for another 3 years.⁵ The machinery and equipment to be used in the production and testing of flat panels was exempted from tariffs and value-added tax. The upstream materials were exempted from tariffs only. The above measures led to a plethora of investment proposals around the country, mostly orchestrated by the local governments in combination with the locally initiated economic stimulus programs designed to cope with the ongoing Subprime Financial Crisis at the time. It was not the first time that a policy initiative from the central government led to competing investments at the local level. Overinvestments often occurred in China when policy incentives were strong and technological barriers were low (European Chamber, 2009). To stem an apparent overcrowded investment wave, the NDRC, together with the Ministry of Industry and Information, issued an ordinance in January 2010, depriving the local government of the authority to approve the new-generation LCD panel projects. All related investment projects were to be authorized by the central government from that date on. It was a clear signal that the central government wanted to restrict market entry so as to more fully exploit the effects of the import substitution policy. The NDRC also wanted to raise the technological barriers so as to forestall overinvestment. At the time of issuance, two LCD manufacturing projects, both approved by the local governments, had already commenced plant construction. One was invested by BOE, a company closely tied to the local government of Beijing, and the other by TCL as part of a consortium with the Shenzhen government in Guangdong Province. In response to the impending door-closing policy announcement, at least five local governments scrambled to put together five separate investment proposals to compete for a license from the NDRC. It was not clear how many licenses would be awarded by the NDRC at the time, but the number was understood to be small. All five projects were initiated by the local government as part of a consortium with some multinational firms that had been active in its jurisdiction: the city of Guangzhou with LG (Korea), the city of Suzhou with Samsung (Korea), the city of Hefei with Xincheng (Taiwan), the city of Chengdu with Chimei (Taiwan), and the city of Nanjing with Sharp (Japan). To win the policy favor from the central government, a local state-MNC alliance, which has been a distinctive feature of China’s economic development (Wu, 1997), was reinforced. The local state needed the MNCs to provide their technological credentials, while the MNCs needed the local government to win the policy competition.

In October 2010, the State Council issued a policy guideline for the promotion of seven strategic industries that were targeted by the 12th Five-Year Plan (2011–2015); the new-generation information technology (IT) industry was one of them. New display devices, including TFT-LCD, were identified as a part of the new-generation IT industry by the State Council. The policy guideline called for the application of the following policy instruments to promote these strategic industries: financial support from the government in the form of equity participation and R&D subsidies, tax incentives to encourage innovation, investment, or consumption, and financial support from the banking industry. The guideline provided a legal basis for the local government to engage in direct investment or to provide financial subsidies to the flat panel industry.

In December 2010, the NDRC approved two flat panel investment projects among the five proposals submitted: the Guangzhou project involving LG, and the Suzhou project involving Samsung. The other three projects were withheld, but not rejected outright. Presumably, they could still win approval in the future if their investment plans are modified to demonstrate more merits, or if the approved projects do not materialize or live up to the expectations of the authorities. In fact, because of the global recession that ensued in 2011, the plant construction in relation to both approved projects has been postponed until 2013.⁶ However, the investment projects by BOE and TCL both went ahead as scheduled, with each of them adopting the most advanced 8.5th generation LCD technologies (capable of producing 55-inch panels). BOE commenced production in 2011, and TCL followed in 2012. Soon after their commercial operations, the Chinese government raised the import tariff on large-screen LCD panels from 3% to 5% in April 2012 to better protect domestic products from import competition.

Increased tariffs on flat panels raised the production costs of domestic TV manufacturers, who had already been squeezed by the ongoing recession. In June 2012, the NDRC launched a new version of the consumption subsidy program to aid the consumer electronics industry. This time, the subsidy was extended to urban as well as rural consumers, but only “energy efficient” products would qualify. TV receivers were the main target of the subsidy, and only those made of flat panels were certified as energy efficient. The subsidy program lasted for 1 year, with a total budget of 26.5 billion RMB expended.

5. Policy conflicts

Both BOE and TCL, which championed the large-screen LCD production in China, were offshoots of the import substitution policy in the post-reform period. Both were founded by local governments under the import substitution regime, but were converted into private ownership in later years. BOE was established by the Beijing government in 1993 to make electronic components that feed TV displays, e.g., CRT. TCL was founded by the government of Huizhou city in Guangdong Province in 1981 to produce color TV receivers. Both contributed to the import substitution course of the Chinese TV receiver industry. After firmly establishing its position in the industry, BOE went through a series of acquisitions to expand its business scope and to enhance its industry position. In 2003, it acquired a large sum of shares of a Taiwan-based company, VOC, to become the controlling shareholder of the company. VOC was the largest contract manufacturer of computer monitors in the world at the time. VOC later launched a brand of its own for computer monitors sold in China, and subsequently acquired the monitor and TV divisions of Philips of the Netherlands. Becoming a controlling shareholder of VOC was a stepping stone for BOE to enter the flat panel industry, as VOC provided a downstream outlet for its products. In 2003, BOE acquired the existing TFT-LCD production facilities owned by Korea’s Hynix and made them a learning platform. In 2005, BOE built its own fifth generation TFT-LCD (suitable for 27-inch panels) plant in Beijing, mainly to feed the computer monitor industry in China. In 2009, BOE built a sixth generation TFT-LCD plant (suitable for 37-inch panels) in Hefei, allowing it to begin offering LCD panels to the local TV industry. In 2011, it built China’s first 8.5th generation TFT-LCD plant (suitable for 55-inch panels) in Beijing, giving it the capacity to substitute imported flat panels of all sizes. In 2012, it was invited by the Hefei government to be a joint venture partner of the Hefei-Xincheng project, which had failed to win a license from the NDRC, and started building another 8.5th generation TFT-LCD plant. The commencement of production was scheduled to take place in 2014.

TCL is China’s largest TV receiver manufacturer today. It also emerged from the import-substitution waves of the 1980s, with the initial technologies being licensed from Japan. The market share of TCL, like those of other domestic brands, has declined in recent years as the flat panel TV receivers gradually replaced the traditional CRT-based TV receivers. In 2009, TCL teamed up with the government of Shenzhen to found a flat panel company named China Star and began constructing the LCD plants. It was reported that the Shenzhen government owned 30% of the shares, and that TCL owned the rest.⁷ Compared with BOE, whose technologies were obtained through acquisitions and learning, TCL jump-started its plants by recruiting a group of Taiwanese engineers (more than 200 in number) who had previously worked for Taiwan’s major LCD companies. Commercial production of large-screen panels commenced in 2012 despite the fact that TCL had no previous experience in the LCD industry. Whereas BOE built its technology base by manufacturing small-screen panels which fed the local computer industry, TCL depended on turnkey plants and ex-engineers of MNCs to leapfrog into the production of large-screen panels. The process of technology learning has now just begun. Unless technology learning is inconsequential to production efficiency, TCL will have to bear the burden of large capital investment coupled with low production efficiency for some time to come.

In contrast to BOE and TCL, which were offshoots of China’s import substitution industrialization, the group of investment projects which won the approval of the central government originated in regions which have been pursuing export-led growth. Both Guangzhou and Suzhou are heartlands of China’s export-oriented electronics industry. The two city governments wanted to attract new investment projects that promise to provide new momentum for growth in the region. They are used to working with MNCs. Indeed, local governments like those in Guangzhou and Suzhou are strategic partners of MNCs, and have created a set of local institutions that minimize the risks for MNCs in their cross-border operations (Yeung, 1998). In so doing, they have taken advantage of the policy incentives offered by the central government and have exploited the local resources, notably labor, available in the rest of China. In the case of the flat panel industry, these local governments again attempted to take advantage of the policy incentives offered by the central government. In addition to Guangzhou and Suzhou, the city of Kunshan also received approval from the NDRC to upgrade an existing sixth generation TFT-LCD panel plant (suitable for 37-inch panels) built by a joint venture between the city government and a Taiwan-based MNC. The project proposed to upgrade the existing facility to at least the eighth generation (capable of making 52-inch panels) by inviting a major Taiwanese LCD company to be the new joint-venture partner.

The import-substitution group of BOE and TCL has sought to produce flat panels to replace panels that would otherwise be imported. The benefits of import substitution are supposed to come from the vertical integration of the industry, which allows technologies to be shared between upstream and downstream producers. Therefore, in the document concerning the promotion of the flat panel industry, Plan to Adjust and Promote the Electronics-Information Industry 2009 (dianzi xinxi chanye tiaozheng yu zhenxing jihua 2009), issued by the State Council, “improving the supply chains of the domestic TV industry” was repeatedly mentioned as one important policy objective of the plan. In fact, before this policy initiative was launched, major suppliers of LCD panels from Japan, Korea, and Taiwan had already set up a number of factories in various regions to produce LCD modules for domestic TV producers in China. Sometimes the factories were joint ventures between foreign panel producers and local TV makers. The panels were imported from Japan, Korea, or Taiwan to be incorporated with locally produced frames, backlights, and electronic controlling devices to form display modules for the local TV receiver industry. The factories were usually located close to the assembly lines of TV manufacturers so as to facilitate coordination between the module suppliers and their users, in addition to saving transportation costs.

Therefore, the proposed establishment of local production of flat panels did not bring additional benefits to local TV assemblers in terms of transportation or coordination costs. The only benefit had to come from the technology spillover effects. In other words, Chinese policy-makers had presumed that MNCs would not share “critical” technologies with their downstream users, and hence it was necessary to nurture domestic producers with indigenous technologies. Indigenous technologies would be shared because they were locally owned. Therefore the main rationale of the policy intervention under the name of IIN could be understood as a policy to ensure that indigenous technologies were created and shared. It was reported that the reason why the Guangzhou and Suzhou projects won the NDRC approval was that their joint-venture partners, LG and Samsung, both agreed to transfer core technologies to China.

The challenge to this policy, however, was the following: indigenous technologies are likely to be inferior to foreign technologies, at least in the early stages of technology development, and will necessarily undermine the competitiveness of the downstream manufacturers. Domestic TV makers will not adopt flat panels made of indigenous technologies unless they are compensated. However, the policy incentives are provided to panel producers rather than downstream users. Misaligned incentives between downstream users and upstream suppliers will impede innovation even if the users are forced to purchase the upstream products. It has been widely demonstrated in the literature that users are instrumental to the innovation of producers (Hippel, 1976, 1977). Concentrating the policy design on the enhancement of local supply capacity to the neglect of demand-side factors could spell trouble for China’s panel industry.

Despite the government support, BOE has been operating at a loss since it began making LCD panels in 2005.⁸ For TCL, which makes its own TV receivers, the disadvantage of indigenous technologies can be internalized through vertical integration. However, it is doubtful that other domestic TV makers will adopt TCL-made flat panels so as to enable TCL to learn better technologies as there is no guarantee that they will benefit from such learning.

For the local government that follows the export-led growth model, it makes little difference whether the technologies are indigenous or foreign. It only matters if the technologies are new and high so as to qualify for tax incentives or financial subsidies offered by the central government. As long as the new and high technologies can be applied in the region to generate a new momentum for growth, the objective of sustaining local economic growth will be achieved. Whether these technologies are transferred to the local joint-venture partners or whether they are shared with downstream users is not a concern of the local government. Even if the related investment projects are to be attained at the expense of other regions, the local government will not hesitate to support them.

However, as previously mentioned, for these regions to compete with other regions in attracting foreign investment, it is useful to leverage the local industry already in existence. Both the Guangzhou and Suzhou governments used the local electronics industry as their leverage. The local electronics industry, which is export-oriented, is very sensitive to costs. Any protective policy, such as market entry restrictions or import barriers, tends to increase its costs. Local components or parts made of indigenous technologies will not be accepted if they result in inferior products or higher costs. Therefore it is very difficult for the export-oriented industries, such as the electronics industries in Guangdong and Jiangsu, to align the interests of downstream users with the policy of promoting upstream materials made of indigenous but inferior technologies.

Before the central government intervened in the flat panel industry, local governments in Kunshan, Nanjing, Hefei, and Shenzhen, among others, had been providing financial subsidies to attract foreign investment in LCD production with no strings tied to indigenous technologies. A small production facility applying sixth generation LCD technologies was invested in by Taiwan-based MNCs in Kunshan and Shenzhen, as well as a Japan-based MNC in Nanjing. The LCD panels produced there were mainly fitted into computers, monitors, and TV receivers for export purposes. This policy was essentially an attempt to renew the export enclaves of the respective regions. The success of this policy has hinged on the potential agglomeration effects that are believed to exist in an industrial cluster, and the agglomeration effects are indeed taking shape, albeit at a slow pace. For example, in the case of Kunshan, the flagship LCD project, Long-Teng, has attracted investments from some MNCs to produce upstream materials, such as glass substrates and polarizers. In fact, this cumulative approach was the essence of the TLHN policy whereby new flagship investment projects were established to ignite a new round of the agglomeration process.

When, in 2010, the central government announced its policy to restrict the LCD investment in favor of IIN, it forced the local governments that had been pursuing the TLHN policy to switch their course. They had to join the IIN bandwagon to compete for the policy bounty because they worried about the crowding-out effects. Central government-endorsed investment projects may monopolize the financial resources available to the industry, not to mention the fiscal incentives built in the central government budget. However, the investment projects they were competing for were the eighth or higher generation LCD panels, as mandated by the NDRC, that are suited to large-screen TVs rather than computers and monitors. No major domestic TV makers are located in their regions to absorb the forthcoming products. In order to make a backward linkage to TV production, cross-region integration is necessary. Unfortunately, there is no mechanism to facilitate cross-region vertical integration in China, which carries with it a redistribution of regional interests, as the downstream industry has to bear the costs of technology learning for the upstream industry located in another region. This leads to a tension between the central- and local-government policies.

Regional competition has been an important driver of economic growth in China since the economic reforms began in 1978 (Xu, 2011). However, local-state championed regional competition has often resulted in segmentation of the markets, thereby preventing the formation of industrial clusters, which are believed to be the foundation for technology spillovers. For example, in the semiconductor industry, despite the fact that China has secured the most advanced fabrication technologies of the industry, it has not been able to foster a cluster of the kind in Korea or Taiwan because of market segmentation (Kaza et al., 2011). One reason for the central government to intervene in the investment rush of the LCD industry is to prevent over-competition between regions. However, because all investment proposals submitted for approval are tied to a specific region and bound by the regional interests, picking the winners also results in policy preferences for specific regions. Without a redistribution mechanism to address the conflicts of regional interests, it is hard to expect any of the selected projects to become a vehicle for cross-region industry integration. Even with the endorsement of the central government, regional champions may never become national champions.

According to the official documents, the main purpose in cultivating indigenous technologies is said to be to enhance the overall competitiveness of China’s local TV receiver industry. However, the dysfunctional distribution of the costs and benefits across different firms and regions is likely to thwart the working of the policy. Indeed, an indigenous technology may generate positive spillover effects in the long run, but the effects are likely to be negative in the short run. Because the long-term winners are not the same short-term losers, without a functional policy to address the redistribution problem, the long-term benefits may never be realized. The importance of taking care of the redistribution problem in the short run can be illustrated by Taiwan’s experience in the 1970s. At that time, Taiwan attempted to develop its petrochemical industry with a traditional import substitution policy whereby the downstream users in the textile and plastics industries had to bear the costs of technology learning of the upstream producers of petrochemical materials. To make sure that this burden was not excessive, the downstream users were given the opportunity to import foreign-made materials if the local materials were expensive to the extent that their export competitiveness was threatened (Chen and Ku, 1999). This “safeguard” measure allowed the petrochemical industry to grow without destroying its fundamental customer base. As a small economy depending on export markets, Taiwan could not afford to nurture an upstream industry at the expense of its downstream exports, neither could Korea (Krueger, 1990). The Chinese policy makers probably thought otherwise because they had a large domestic market.

In China, the problem of integrating the upstream with the downstream industries is further complicated by geography. Given the fact that downstream users of TV manufacturers and upstream flat panel producers are often located in different regions, the expected spillover effects may be limited even if the indigenous technologies are indeed created, as spillover effects tend to decay with distance (Audretsch and Feldman, 1996). If each region is to form a self-contained, vertically integrated cluster to fully realize the spillover effects, the overinvestment problem becomes inevitable.

Table 1 provides some indicators of the performance of China’s flat panel industry in recent years. It can be seen that the industry shipment expanded by about four times in 2009–2012, with China’s share in the world market increasing from a negligible amount to 6.6% in 2012. With more production capacity now in place, China’s share will almost surely break the 10% mark in 2013. If capacity continues to grow under government subsidies, China will soon overtake Japan as the third largest flat panel producer in the world. However, the trade statistics suggest that increasing domestic production may not necessarily serve the purpose of import substitution. The trade data show that, as China’s domestic flat panel TV receiver industry grew, imports and exports of flat panels increased at the same time. This suggests a nature of intra-industry trade where domestic panels, presumably lower-end products, are exported in exchange for higher-end products, a phenomenon also seen in China’s other overinvested industries such as steel. The exports went to countries like Vietnam, Malaysia, Turkey, and Russia, while the imports continued to flow from Korea, Taiwan, and Japan. If the flat panel industry follows the same development course of the other overinvested industries, the Chinese flat panels will soon flood the world market but the premium products that are needed domestically will continue to be imported. To depart from this course, the “indigenous” technologies now possessed by BOE and TCL will have to be upgraded to catch up with the world frontiers. Whether they can do so depends on how fast they learn from production. The key, in our opinion, is whether they can align their interests with those of the downstream users that manufacture high-quality products in China, especially those located in their regions.

6. Concluding remarks

In this article, we study two policies that are related to technology upgrading, namely, IIN and TLHN. IIN is a top-down policy with an aspiration for technology leapfrogging, while TLHN is a bottom-up policy which adopts a cumulative approach to technology advancement. The study suggests that the cumulative approach is superior because it allows China to leverage global technology resources, and avoids the conflicts of interest between the upstream suppliers and downstream users. Misaligned incentives between the upstream suppliers and downstream users, which are evident in the implementation of the IIN policy, make downstream users reluctant to support local products. The result is a fragmented industry structure unsuited for technology learning. The same phenomenon can be observed in the semiconductor industry where the upper-tier Chinese integrated circuits (IC) design houses preferred to work with foreign semiconductor foundries. In fact, superior technologies offered by foreign foundries were good catalysts for IINs attained by local IC designers (Ernst and Naughton, 2012). The lesson of the flat panel industry seems applicable to industries in which a key component determines the competitiveness of the final product.

When implementing the IIN policy, the Chinese authorities insisted on applying the frontier and critical technologies. In the case of the flat panel industry, only investment projects employing the newest generation of technologies were to be approved. This forced a technology jump in the local industry, which may be counterproductive to technology learning. Learning, which is incremental in nature, has been shown to be an important process of forming indigenous technologies in other East Asian countries. Instead of incremental learning, any attempt to straddle the technology gap entails extra risks in R&D investment and the resulting technologies may not create a competitive industry. Moreover, by promoting new investment projects that are believed to carry the newer technologies with priority access to the domestic market, the opportunity for the existing projects that adopt the older-generation technologies to upgrade themselves may have been foreclosed.

The issue of “appropriate technology,” which has been discussed in the development literature (Schumacher, 1973; Eckaus, 1977), is also relevant in this policy experiment. As a developing country, the Chinese market has a substantial demand for substandard flat panels which can be used in low-end monitors or TV receivers. This special segment of the market, which may not exist elsewhere, allows China’s local flat panel producers that operate with the older-generation technologies to survive, even though their production efficiency is not on a par with that of their competitors in Korea and Taiwan. With the newest-generation technologies, the panels produced are too big to fit into this segment of the market where demand is concentrated in small screens. An indigenous technology can hardly be developed without taking note of the idiosyncratic local conditions.

China has benefited enormously from joining the forces of globalization. Whereas TLHN represents an initiative to continue riding the waves of globalization, IIN is an attempt to counter the globalization forces. In order to change China’s position in the global production system as a global factory to the position of a global innovator, the authorities believe that indigenous technologies are indispensable. However, an attempt to acquire indigenous technologies by leapfrogging the current technological bases, which are intricately tied to the global system, implies a system competition with the rest of the world. This may deny China the apparent benefits of globalization, such as trade in intermediate goods, therefore disrupting its market-led development course (Ernst, 2010). Whether a country can win a system competition depends on the network effects that the system engenders (Katz and Shapiro, 1994). The network effects are most pronounced if the system is complementary to the genuine forces of globalization. Japan has had many good indigenous technologies but has failed many times in the system competition because the Japanese systems failed to generate enough network effects (Funk, 2002). The Chinese authorities apparently believe that China can outdo Japan because it commands a bigger domestic market than Japan. Network effects are, however, determined in the global market instead of in the domestic market.

Philosophically speaking, the regions in pursuit of the TLHN policy are regions that have followed the East Asian development model. The essence of the East Asian model is “resources leverage” (Mathews and Cho, 2002). Countries like Korea and Taiwan have all taken a cooperative approach to advance their technologies by leveraging international technology resources. In doing so, they minimize the investment costs and the risk of failure in R&D. The idea of IIN, meanwhile, is an attempt to break away from this model which the authorities believe has locked China in a state of technological dependency. Again, the Chinese authorities are of the opinion that China can trace an independent path of technology development because its domestic market is large. The authorities forget the fact that this large market is segmented, and therefore without realigning the regional interests, it is difficult for a top-down policy like IIN to work. When the central government decided which regions would lead the drive in IIN, such policies were likely to invite defensive actions from the losing regions, which were often allied with MNCs. Hence, any attempt to foster linkages between regions will be resisted.

References