Second draft for The World Bank-IPS
Research Project on the Rise of China and India
The Coming Age of China-plus-One:
The Japanese Perspective on East Asian Production Networks
February 18, 2006
Masahisa FUJITA
IDE-JETRO and KIER-KyotoUniversity
Nobuaki HAMAGUCHI*
RIEB-KobeUniversity
______
* Corresponding author: E-mail address
- Introduction
In this paper, which focuses on China, we examine the factors affecting the recent reorganization of industrial production networks in East Asia, paying specific attention to the role of foreign direct investment (FDI) and Japanese multinational firms (MNFs). We also present some policy suggestions for the continued development of affected countries.
Over the last two decades, FDI has been rapidly growing in East Asia. Most notably, China alone received about US$ 60 billion in 2004, representing nearly 10% of the total world FDI flow. China attracts global attention as a production platform by offering a huge pool of low cost labor as well as a growing number of skilled laborers. It also offers a huge consumer market with a rapidly growing number of middle class. Undoubtedly, economic linkages with China have been playing an important role in the recent economic growth of many countries around the world. We will explore in detail what factors are likely to contribute to the continuing migration of industries to China.
However, followingthe old adage, there is an emerging concern for “putting all the eggs in one basket”. In particular, Japanese MNFs have started to pursuea strategy of diversification of risk concerning the concentration of their production activities in China. We discuss what could be the countervailing factors mitigating the concentration of industrial activities in China, leading to a more even distribution of industries across Asia and elsewhere.
This paper is structured as follows. First, in Section 2, we present a rough picture of current production networks in East Asia, showing the pattern of specialization across countries. Then, in Section 3, we suggest that the concentration of manufacturing industries in China can be characterized as a cumulative process of FDI, exports, and income growth. It is also pointed out that this phenomenon is geographically concentrated, leading to the self-organization of industrial clusters in China. In Section 4, from the perspective of Japanese MNFs, we discuss emerging risk factors in China which are gaining in relevance to the future development of the regional production network in Asia. The final section concludes the paper with the discussion of policy implications in the long-run.
- Integrated Production Networks in East Asia
For the past two decades, East Asia has deepened intra-regional trade[1]. An appropriate measure for examining this issue is the intra-regional trade share, which represents the share of international trade (exports and imports) within the region in terms of the total trade by all countries in that region with all countries in the world. It is apparent from Figure 1 that the intra-regional trade share is the highest for the EU15, hovering at around the 60% mark. This is not surprising, as the situation of many rich neighboring countries with small lands cooperating in unison over the past half-centuryhas made regional integration considerably deeper and wider. On the other hand, the intra-regional trade of both East Asia and North America has shown steady growth in the past couple of decades. In particular, except during the period of the Asian financial crisis in 1997-98, the intra-regional trade share of East Asia has been progressively increasing and approaching that of EU15, while becoming significantly higher than that of NAFTA.
Figure 1. The intra-regional trade share (export + import) of each region, 1980-2003
Moreover, one of two lines in the bottom part of Figure 1 shows the change in the intra-regional trade share of ASEAN-10, whereas the other line shows that of China-Japan-Korea region (North East Asia). We can see from the figure that the intra-regional trade share of each sub-region of East Asia is much lower than that of the whole of East Asia (which consists of ASEAN 10 and China-Japan-Korea plus Taiwan and Hong Kong). For example, in 2003, the intra-trade share of ASEAN 10 was 22.2 % and that of China-Japan-Korea was 25.8%, whereas that of East Asia was 52.4%. This indicates that although neither ASEAN alone, nor China-Japan-Korea alone represents a sufficiently integrated region, the two sub-regions together constitute an integrated region with a strong interdependency. We can also see from Figure 1 that over the last several years, the intra-trade share of China-Japan-Korea is increasing faster than that of ASEAN 10. This is due to the rapid growth of China in recent years.
Next, Table 1 shows the recent trend in the relative changes of economic powers among East Asian countries in terms of GDP and world trade shares. The top block of the table indicates that East Asia as a whole increased its world GDP share from 20.5% in 1990 to 23.4% in 2000, which then returned to 20.1%, coinciding with the rise of China and the decline of Japan. In contrast, the share of NIES3 and ASEAN remains the same over the same period. As a consequence, Japan’ s GDP share in East Asia declined from 71.2% in 1990 to 56.7% in 2004 whereas that of China increased from 9.3% to 19.9%, implying that Japan remains in the dominant position despite of the rapid catch-up of the latter over the two decades.
Table 1. World share in GDP and trade
Japan / NIES3 / ASEAN / China / East AsiaGDP* / 1990 / 14.6% / 2.4% / 1.6% / 1.9% / 20.5%
2000 / 15.1% / 3.1% / 1.8% / 3.4% / 23.4%
2004 / 11.4% / 2.8% / 1.8% / 4.0% / 20.1%
Exports** / 1990 / 8.3% / 6.2% / 4.0% / 1.8% / 20.4%
2000 / 7.5% / 8.2% / 6.7% / 3.9% / 26.3%
2004 / 6.3% / 7.6% / 6.0% / 6.6% / 26.4%
Imports** / 1990 / 6.6% / 5.8% / 4.5% / 1.5% / 18.4%
2000 / 5.8% / 7.8% / 5.8% / 3.4% / 22.8%
2004 / 4.9% / 7.1% / 5.3% / 6.0% / 23.3%
Note: NIES3 includes Hong Kong, South Korea, and Taiwan. Singapore is included in ASEAN.
（Source）*IMF, World Economic Outlook Database （ and **IMF, International Financial Statistics (CD-ROM) 70..D and 71..D
In contrast with GDP share, trade shares provide a significantly different picture of East Asia. As the second and third blocks of Table 1 shows, the world export share of East Asia grew from 20.4% in 1990 to 26.4% in 2004. However, its composition has changed dramatically over the same period. World share of both exports and imports of NIES3, ASEAN, and China grew much faster than that of GDP. As a consequence, they occupy about the same world share as Japan. All four countries/regions of East Asia have significantly smaller import shares than export shares. This means that East Asia as a whole has built up an export-platform to the rest of the world over the past quarter-century.
Figure 2. Composition of Exports and Imports of Asian Countries by Types of Goods
(2003)
Unit: million US$
(Source) Author’s own calculation based on METI (2005) Figure 2-3-1 and IMF Direction of Trade Statistics (CD-ROM).
The trade pattern is characterized by an intra-regional division of labor. We can observe from Figure 2 that China is by far the biggest net exporter of consumer goods while importing parts, processed materials and primary goods. On the other hand, Japan is the major supplier of capital goods and parts in a complementary relationship with China. Korea and Taiwan have similar characteristics toJapan, although they are still net exporters of consumer goods. Singapore is specialized in capital goods related to information technology and processed materials derived from petrochemicals. Other ASEAN countries are net exporters of consumer goods while also showing comparative advantages in exports of capital goods (computers in the Philippines and Malaysia and automobiles in Thailand) or processed and unprocessed primary commodities in Indonesia.
Figure 3. The triangular trade pattern in Asia-Pacific.
(Source) Fujita (2005)
Figure 3 represents a rough picture of the trade pattern in the Asia-Pacific region with the focus on East Asia and the US. Japan, Asian NIEs, China and ASEAN together have developed a highly integrated production system. Intermediate goods move back and forth, while consumer goods assembled in China and ASEAN (as well as in Japan and NIES) are exported to the US (as well as to the rest of the world (ROW))[2]. In particular, China has become the largest export-platform to the ROW.
Here, it must be pointed out that a large share of China’s trade activity is conducted by affiliates of multinational firms (MNFs). For example, according to the Chinese official data[3], 57% of both exports and imports were mediated by MNFs in 2004. We examine the evolution of this production system, with a focus on the role of Japanese MNFs, in the next section.
3. Agglomeration to China
The formation of the regional production network in East Asia was influenced by active foreign direct investment (FDI) by Japanese firms. Japanese FDI intensified in the late 1980s due to the appreciation of the yen which followed the Plaza Accord of 1985. Such investment was first hosted by ASEAN countries with low cost unskilled labor, enabling cost reduction in the final assembly for export. In the 1990s, China started to gain more attention with its economic liberalization and rapid growth. In Table 2, we can observe that the share of FDI in developing economies in the world increased from 17.2% in 1990 to 36.0% in 2004. The FDI in developing economies was slightly diversified as the share of Asia declined from 63.3% to 54.7%. Within Asia, there was clear tendency towards concentration inChina, which represented only 15.4% in 1990 (illustrating concentration inSoutheast Asia) but sharply increasing to 47.5% in 2004. Thus we can roughly depict a picture of fractal concentration where a half of FDI for developing economies was invested in Asia which is, in turn, redirected towardChina. Then, as we can see below, FDI in China is concentrated in the costal regions.
Table 2. FDI inflow to Developing Asia
(Unit: Million US$)
1990 / 2004World <W> / 207,878 / 648,146
- Developing Economies <D> / 35,736 / 233,227
- Developing Asia <A> / 22,614 / 127,545
- China <C> / 3,487 / 60,630
D/W (%) / 17.2% / 36.0%
A/D (%) / 63.3% / 54.7%
C/A (%) / 15.4% / 47.5%
(Source) UNCTAD, World Investment Report FDI Database,
Table 3 shows that Japanese FDI in Asia at the beginning of the 1990s was concentrated in ASEAN, and in particular in Singapore, Malaysia, and Thailand. Since then the focus of interest has shifted to mainland China. Japanese FDI in China is mostly concentrated in three areas, which are, in terms of relative importance, Bohai Rim （Beijing, Tianjin, Hebei, Shandong, Liaoning）, Yangzi River Delta (Jiangsu, Zhejiang, Shanghai), and Hong Kong – Guangdong (Pearl River Delta). Recently, Vietnam is being increasingly considered as an alternative location for investment toGuangdong because the wages of unskilled labor have started to rise. In terms of the rest of ASEAN, the growth of the number of affiliates of Japanese firms has been moderatein Indonesia and the Philippines and has even declined in Malaysia and Singapore. However, Thailand stands out as an exceptional case, as it has kept attracting Japanese FDI, stimulated by the recent development of the automobile industry there. Finally, the growth of Japanese FDI in India is still modest, although financial investment in the Indian capital market is already heated[4].
Table 3. Number of Japanese Firms’ Foreign Affiliates in Asia
Location / 1990 / 1994 / 2000 / 2004China / 315 / 1061 / 2432 / 4041
BohaiBay / 141 / 404 / 815 / 1039
YanziRiver Delta / 77 / 384 / 1060 / 2187
Pearl River Delta / 47 / 152 / 310 / 525
Others / 50 / 121 / 247 / 290
Hong Kong / 793 / 1022 / 1112 / 1121
Taiwan / 727 / 812 / 891 / 909
Korea / 399 / 404 / 496 / 640
Singapore / 743 / 961 / 1129 / 1067
Malaysia / 509 / 709 / 881 / 805
Thailand / 766 / 983 / 1342 / 1512
Indonesia / 292 / 439 / 676 / 698
Philippines / 171 / 234 / 426 / 453
Vietnam / 1 / 21 / 174 / 220
India / 71 / 81 / 168 / 193
(Source) Toyo Keizai Inc., Overseas Japanese Companies Data.
Due to time lags between the investment boom in ASEAN and that in China, Japanese firms tended to consider the strategies for the two economies separately. With stronger regional integration, however, there is a growing concern now that operations in East Asia should be reorganized, especially when a company has duplicate functions within the region. For example, Asaka (2005) claims that with the highly developed international logistic chain and expanding web of the free trade agreements, Japanese firms are facing the need to establish a more unified “Greater Asia Strategy”. At the same time, Japanese firms are revaluating their investment within Japan, while their interest in India is rapidly growing. Thus, a substantial reorganization of the supply chain of Japanese MNFs is expected to occur on a wider regional scale in Asia.
In this regard, the FY2005-result of the Survey on Overseas Business Operations by Japanese Manufacturing Companies[5], conducted annually by the Japan Bank for International Cooperation (JBIC), brought more detailed insights about the future outlook. According to the survey, 85 firms out of the sample-set of 590 firms expressed their intention to reduce some foreign operations in the next 3 years because of thefollowing motivations: to transfer business units back to Japan (10 firms); to withdraw from the present business there (38 firms); and to transfer business units to other foreign countries (37 firms). Among the 37 firms that intend to change one foreign location to another, there are 21 production units and 16 sales/administrative units. The largest part of rearrangement strategies are considered within the East Asian production network. For eleven production units and seven sales/administrative units, their next location will be China. As far as production units are concerned, the distribution of locations where the reduction/withdrawal might occur is: 5 from Malaysia; 3 each from Taiwan and Indonesia; 2 each from Hong Kong, Singapore, China, and EU15; and 1 each from North America and Mexico. Most firms moving from Taiwan and Hong Kong may go to mainland China, while firms in Singapore will move to Malaysia. Four out of eight firms intending to relocate from either Malaysia or Indonesia declare China as an alternative location, although they also cite Thailand, Vietnam and Indonesia (if the origin is Malaysia) as possible options. At the same time, there are two cases which are considering moving out of China to Myanmar and Indiain search of cheaper labor costs. These figures suggest that while the East Asian strategy of Japanese MNFs will continue to be focused on China, the redefinition of operations in ASEAN will be another key issue. The latter point will depend on the deepening of intra-ASEAN integration and the progress of specialization there.
It should be noted that the continuing migration of firms into China exhibits a circular causation phenomenon. In other words, the attractiveness of China as the destination of FDI has been enhanced by growth and agglomeration itself, making the growth trend even stronger. One force in the self-organizing agglomeration of economies is the rapidly growing consumer market. Panel (a) of Figure 4 plots the Chinese provinces evaluated by their growth rate of exports during 2004-2005 on the vertical axis and the share of the multinational firms (MNFs) in exports in 2005. A cluster analysis (in Ward’s method) identified four groups with respect to the growth of exports and the presence of MNFs. As shown by Panel (b), the provinces with high export growth and more than 50% share of MNFs in exports (Group I) constitute three economic cores in the costal area. Then, provinces of Group II with 20-40% MNF share are located the adjacent regions which can be seen as a sprawl of Group I. Inland provinces of Group III still demonstrate high export growth but the contribution of MNFs is less than 20%. Finally, for the provinces of the periphery regions (Group IV), both export growth and presence of MNFs are quite low.
Figure 4. Mapping of the MNF-led Export Growth
(a) (b)
(Source) Author’s own calculation based on Tables 18-10 & 18-12 of China Statistical Yearbook 2005
Next, a comparison between the Panel (b) and Figure 5 suggests that the population of higher income households that lead consumption in domestic markets is concentrated in the area occupied by provinces from Group I. It is possible to interpret this phenomenon in the following way. In general, productivity is higher in export oriented MNFs than local firms allowing the workers in MNFs to obtain higher wages. With the population earning higher wages, the MNFs are able to expand sales in the local market. Then, the scale economy from increased production size enables the MNFs to achieve even higher productivity. Thus, the circular causality through the interaction of the market size and the scale economy may have been playing a key role in the formation of industrial clusters.
As we observed in Table 1, it is in these regions that Japanese FDI has concentrated since the 1990s. With this data analysis, we can roughly confirm that the MNF-led export growth performed as a catalyst for uneven regional development in China. Here, the case of Chongqing may be seen as an exceptional case. Chongqing belongs to Group III in Figure 4 and has less than 20% of MNF share in exports, however, the government has designated the city as a gateway to the West and invested heavily to establish an inland growth pole.
Figure 5. Regional distribution of the higher income households*
＊Number of households earning greater than 6,000 yuan in large metropolitan areas or 4,000 yuan in medium size cities, which in total correspond to about 10% of total urban households.
(Source) METI (2005) Figure 2-1-87.
Next, the second cause of the spatial agglomeration of Japanese FDI in China is the backward and forward linkages between final goods manufactures and their suppliers. Essentially, the establishment of final assembly plants has attracted specialized suppliers of parts and components, whose diversity and accumulation, in turn, has enhanced the productivity of final production. Table 4 shows that final good producers and parts suppliers are mutually reinforcing the concentration of Japanese FDI in the electric and electronics equipment industry in three regions. In this table, each number represents the number, by year of establishment, of the producers of intermediate products classified as “electric and electronics equipment”, while each company name shows a final good manufacturer. In particular, localization in Yangzi Delta region has been intensified recently. In the automobile sector, reported in Table 5, a large number of auto-parts suppliers have also been located in Yangzi Delta region, although there is no Japanese automobile plants there, for export-oriented production as well as to attend to the market created by other multinational firms[6]. More recently, Japanese FDI in the automobile sector is rapidly increasing on the Bohai Rim and Southern coastal area, where big Japanese automakers have established passenger car assembly plants since entry deregulation in the late 1990s. As these new comers are gaining market share against the incumbent producers, supported by the clustering of parts suppliers nearby, the geography of automobile production in China is changing.