The Trade Structure in East Asia: Spiral Pattern of Development and Triangular Trade Structure (TTS) as a Regional Manufacturing Platform - from the White Paper on International Economy and Trade 2005

Date July 19, 2005
Speaker OKAMOTO Susumu(Former Deputy Director, Research and Analysis Division, Trade Policy Bureau, METI)
Moderator MIYAGAWA Tsutomu(Faculty Fellow, RIETI / Professor, Faculty of Economics, Gakushuin University)
Materials

Summary

The outline of my presentation is as follows: questions, trade structure in East Asian countries, hypothesis, quality of trade goods in East Asian countries, triangular trade structure (TTS), and conclusion.

In East Asia, the division of production processes across borders is highly developed, and the economic circumstances of East Asian countries lead me to the following two questions. First, how can we understand the economic dynamism and development in East Asia? Is the conventional wisdom, such as the "flying-geese pattern of development," still valid in East Asia? Second, how are trade relations in East Asia formed? Is the economic structure in East Asia closed or open to other regions, especially to developed countries like the United States and the European Union?

I will briefly explain the division of production processes across borders in East Asia. In Figure 1, I have classified all trade data into three main categories namely, primary goods, intermediate goods, and final goods, and five subcategories, Intermediate goods are goods used in the manufacturing process. Final goods are products you can buy.

This classification is compatible to the Input-Output (I-O) table as well as the System of National Account (SNA) categories, and therefore, I think it will be useful to future studies on the relationship between production and trade, et cetera.

The following graphs show the trade structure in each country, which is based on the aforementioned database. Japan's trade structure indicates that it exports intermediate and final goods, particularly parts and components, and capital goods. On the other hand, a growing share of Japan's imports consists of processed goods and parts and components. The graph also illustrates the decline in the export share of final goods, which is attributed to TTS.

China's trade structure is a real contrast to Japan's. Processed goods and parts and components account for a large share of China's imports, while capital goods and final goods account for a large share of its exports. This is because China imports processed goods to manufacture capital and consumption goods for export.

As you can see, Korea and Taiwan represent an intermediate trade structure somewhere between Japan and China. I assume these two economies are in the process of transforming from assembly-type production like China, to a comprehensive type of production like Japan. Hence they have a large share of processed goods and parts and components, which are composed of intermediate goods. They export final goods, but the share of final goods is declining. The countries of the Association of Southeast Asian Nations (ASEAN-4) have high shares of intermediate goods among both exports and imports. What makes Thailand unique is that a large share of its exports is final goods, I think, namely pick-up trucks. In addition, while parts and components account for a large share of imports and exports in Malaysia and the Philippines, processed goods account for a large share in Indonesia. These features of the ASEAN countries reflect the fact that processed goods such as textiles, pulp and paper in Indonesia are actively traded, as are parts and components such as electric machinery and consumer electric appliances in the Philippines and Malaysia.

The findings from the graphs are as follows. Firstly, the analysis indicates there are distinctive aspects in the trade structures of Japan, the newly industrializing economies (NIEs), China and ASEAN from the standpoint of the production stage. I assume the NIEs are "between" Japan and China. Secondly, the analysis also shows the possibility that East Asia will form a complementary economic region, with assembly production and intermediate goods and specialized production forming the backbone of the regional economy.

Here, I show the trade structure in East Asian countries. When we analyze the trade structure by considering the development of production networks in the region, I think we have to take into account the highly advanced division of production processes. This graph shows the international competitiveness of goods (where the international competitiveness index (ICI) = (exports - imports) / (exports + imports)). The x-axis indicates the international competitiveness of intermediate goods, and the y-axis indicates the international competitiveness of final goods.

The spiral pattern development model works in the following way. Industries in the third quadrant have weak international competitiveness in both intermediate and final goods, and will shift to the second quadrant because they have acquired technological knowledge or experience that allows them to assemble the products. However, they do not have sufficient technology to produce parts and components, and therefore they have to import key parts and components from outside. Hence industries have strong international competitiveness in final goods but weak international competitiveness in intermediate goods. Industries then shift to the first quadrant wherein they have strong international competitiveness both in intermediate and final goods because they have acquired a high enough technological level to produce parts and components. These industries also have a relatively low cost of labor. Then they move to the fourth quadrant because they lose their competitiveness in labor cost. The industries have acquired high incomes and they suffer from rising labor costs. Therefore, they lose their competitiveness in final goods. Lastly, the industries move back to the third quadrant because they have lost competitiveness in both intermediate and final goods. However, this does not necessarily mean the industries have entirely lost their competitiveness. They maintain their competitiveness in specific areas, particularly high-technology products.

The reason I call this model "spiral pattern development" is because the industry moves from quadrants one, four, three and two as they develop their technology or add value to their products. The third axis in the center represents the spiral pattern. I would say this is not just a circular development but a spiral pattern of development. The y-axis represents final goods and the x-axis represents intermediate goods. Japan exports more intermediate goods than it imports, while it imports more final goods than it exports.

With regard to steel, nonferrous metal and metal products, Indonesia recently shifted from the second to the first quadrant and it now enjoys strong competitiveness in both intermediate and final goods. On the other hand, while Japan was in the first quadrant in 1980, it has since moved to the fourth quadrant. In other words, Japan lost competitiveness in final goods. As for chemicals, no country is placed in the first quadrant, which is called the overall production-type area. However, complementarity is formed between the countries in assembly production, such as China and Indonesia, and those in intermediate goods and specialized production such as Japan and Korea. These countries export intermediate goods to the world; and China and Indonesia import such intermediate goods from the world and export final goods to the world. With electric machinery, not many dynamic changes are observed compared to other industries. Here, almost all countries are placed in the first quadrant. This means East Asian countries have strong competitiveness in electric machinery for both intermediate and final goods. This clearly shows East Asia's dominant share of electric machinery in the world market. China, however, is placed in the assembly-type area, so it enjoys competitiveness in the assembly process in production. With regard to transport machinery, Thailand had marginal competitiveness in 1990 but later moved to the second quadrant. Also, Japan and Korea are both in the first quadrant. In miscellaneous goods and toys, both Japan and Korea lost competitiveness in final goods, while China is in a strong position.

Looking at the international competitiveness index by country, in this case, Japan, the interesting point is that many industries have strong competitiveness in intermediate goods. As such, most industries are on the right side of the chart. That there are no industries in the second quadrant also means that Japan has no competitiveness in the assembly process in manufacturing. On the other hand, many industries in China are at the top, meaning they have strong competitiveness in final goods. China also does not have any industries in the fourth quadrant, meaning it does not have any industries specializing in intermediate goods. This indicates that there is a big time lag in development between Japan and China. Korea is in between Japan's and China's competitiveness structure. Thailand is also in the middle but specialized industries have a strong position in the chart in such areas as consumer electronic appliances and precision machinery.

The findings from the charts are as follows: Japan, the NIEs, China and ASEAN have different strengths so that complementarity in the region as a whole is observed. The second point is very interesting. Complementarity is not static but is dynamically changing along with economic growth as well as the level of production technology.

I will now briefly explain about the quality of trade goods. This index represents the ratio of export unit prices to import unit prices. If the index is higher than one, the country exports products with higher added value compared to import goods. If the index is less than one, the country imports higher added-value products and exports lower added-value products. In 2003, more than 65% of Japan's exports were products with high added value. On the other hand, China's share of items with an index of more than one decreased from 1995 to 2003 as China focused on relatively cheap goods. There is thus a big contrast between Japan and China. In addition, ASEAN has a comparatively high proportion of items with an index higher than one. This may indicate that ASEAN countries export relatively high-priced goods, mainly for trade in the region.

Moving on to TTS, this figure shows the trade volume between Japan/NIEs and China/ASEAN. It indicates that the export of intermediate goods from Japan/NIEs to China/ASEAN has increased. China/ASEAN's exports of intermediate goods to Japan/NIEs also rose. In addition, China/ASEAN's exports of final goods, particularly consumption goods, to the U.S./EU surged. And Japan/NIEs' exports of both intermediate and final goods to the U.S./EU increased.

These graphs indicate three things: First, reciprocal trade of intermediate goods between Japan/NIEs and China/ASEAN has risen. Second, there was significant growth in exports of final goods, particularly consumption goods, from China/ASEAN to U.S./EU. Third, there was steady growth in the export of intermediate and final goods from Japan/NIEs to the U.S./EU.

A TTS has formed among these three regions. Japan/NIEs produce parts through high-technology, capital-intensive processes and export them to China and ASEAN. China and ASEAN, in turn, assemble such parts into final goods, capital goods or consumption goods, and export them to final markets such as the U.S. and EU, which consume such goods. Moreover, trade value in TTS has increased ((intermediate goods exports of Japan/NIEs to China/ASEAN) + (final goods exports of China/ASEAN to U.S./EU)). The share of TTS in total trade value has also more than doubled. Therefore, TTS has been strengthened.

Here is the change in the triangular trade index for industry. In almost all industries except food and transport machinery, TTS has intensified. Hence, I conclude that TTS has formed across industries.

This graph illustrates the trend of local content in China and the ASEAN region. As the index increases, local content also rises. Although the local content in the region has increased, TTS has formed.

This graph indicates the unit price change in TTS for Japan's exports of intermediate goods to China, and China's exports of final goods to the U.S. If the rate of increase in unit price of Japan's exports to China exceeds the rate of increase in unit price of Japan's exports to the world, the number of comparative change rises above one. In 2000 to 2003, 52.4% of Japan's exports to China had a comparative change of more than one. Therefore, I think I can say that the added value of Japan's exports to China has risen. The ratio of exports with a comparative change of more than one has decreased for China, but it is still more than 50%, and the rate of increase exceeds that of the rest of the world.

Figure 15 demonstrates that there is a difference between exports from Japan to the U.S. and exports from China to the U.S. It indicates the unit price of Japan's exports to the U.S. exceeds that of China's exports to the U.S. The conclusion is that while both Japan's and China's exports to the U.S. have risen, the same items have different added values for each good traded in these two trade relationships. Thus, Japan and China have different trade goods.

Looking at the comparison of the unit prices of Japan's exports to China and China's exports to Japan, you can see that the unit prices of close to 90% of goods in 2003 were higher for Japan's exports to China than those of China's exports to Japan. Even when Japan and China trade products in the same category, such products have different added values. Japan's exports have a much higher added value than those of China.

In conclusion, as you can see from the graphs, the observed spiral pattern of development suggests that regions where the production network has spread across borders such as East Asia require analysis with respect to production stage. Secondly, the complementarity in East Asia across industries not only exists but changes dynamically rather than statically. Thirdly, while the complementarity in the region has increased, TTS have formed between East Asia and U.S./EU. Fourthly, the development of TTS as we saw in the change in unit prices is a function of both development of supporting industries in East Asia and increasing depth in the capital-intensive industrial bases. Lastly, TTS is not exclusive to the East Asian region but is becoming a source of dynamism that driving the growth of the world economy by promoting trade relations with U.S./EU.

Questions and Answers

Q: I could not exactly understand your statement in Figure 13: "As local production of intermediate goods advances, TTS strengthens." Why does China's increasing use of local content for consumer electric appliances lead to strengthened TTS? The trade in intermediate goods from Japan to China should decrease as a result of more local production of parts and equipment. Secondly, with regard to Figure 4 on page 18, have you found any specific industries in particular countries that fall under pattern five?

A: Figure 13 assumes that the relationship between domestic production and domestic consumption is unchanged. There are four elements to think about: exports, imports, internal consumption, and internal production. Since I did not want to go deeply into an internal production survey for each country, I decided to refer to the relationship between imports and exports of intermediate goods and final goods. We saw in Figure 12 that TTS has been strengthening. We also saw in Figure 11 that TTS has been strengthening in terms of volume and ratio. Using these three graphs, I wanted to show that although local content in China has been increasing, TTS has been strengthening as well. This implies that the technological level of local production has risen and that the added value of trade or unit price of trade goods is also rising. I wanted to show all the elements of the trade structure.

As for your second question, I turn to the textile industry. The U.S. textile industry has entered the third quadrant, and Japan is following the U.S. movement. How do we interpret this change? The consensus is that the industry has specific products. We found that the Japanese textile industry produces very technologically advanced products that cannot be produced by other countries. In such small, niche areas the Japanese textile industry has a dominant share. Therefore, this does not mean industries in the third quadrant have entirely lost their international competitiveness. Industries can compete in the domestic and world markets with specialized products as well as brand names. In many cases, developing countries like China are suffering from poor brand power.

Q: This division of labor or division of production which you describe is through the development of so-called business networks in the region. To what extent does this fit with what you might call national aspirations? How long can this production pattern persist given that countries presumably desire to produce finished goods eventually? Finally, the White Paper seemed to suggest that there were many problems in China or potential problems. Were you suggesting that one cannot rely on the Chinese economy being a stable link in the production chain?

A: Your first question is exactly the question that I have in my mind. I think each industry has distinctive features. For instance, in the textile industry, the U.S. is ahead in the spiral pattern of development, followed by Japan, Korea, Taiwan and Indonesia. I assume Japan will enter the third quadrant in the near future, and Korea and Taiwan will shift to the fourth quadrant. The industry has sort of a predictable future. Perhaps we have to think about technology, the level of labor skills, or brand names. Of course the complementarity is dynamically changing. So far, I do not think Japan's dominance in, say, transport machinery is harmed by other countries. I am not so worried about whether each industry is in the fourth quadrant or entering the third quadrant.

As for the 2005 White Paper, what we wanted to say is that we have to construct a win-win relationship between Japan and other East Asian countries, particularly China. China is a very important country not just to East Asia but also to the Japanese economy. The conclusion of the research is that the Chinese role in the manufacturing process is extremely important and that it is not a threat to the Japanese economy. We need to think about the relationship between Japan and China, taking into consideration the significant differences in competitiveness between the two countries.

Q: Your hypothesis is very interesting. But how do you reconcile your hypothesis with traditional international trade theory? For example, if the intermediate goods Japan specializes in are capital intensive, that means the Japanese economy is accumulating more capital than any other country in the region.

Secondly, if Japan is exporting higher value-added intermediate goods to China, and China is exporting lower value-added final products to the U.S., that means Japan is taking advantage of China's weaker currency. It seems to me that even if the yuan appreciates, we can continue to export value-added intermediate goods to China.

A: I have to say the difference between the hypothesis I raised here and conventional theories is that the latter considers economic development by country and industry. The question I had at the beginning of this research project was whether this was the right approach to understand economic development in a region like East Asia, where the division of the production process is highly integrated across borders. If you stick to conventional models you can only think about development in terms of industry. However, the situation we face is as follows. For instance, you do not know whether Japan's textile exports are intermediate or final goods. China imports and exports textiles, but you do not know whether those goods are intermediate or final goods. You do not know the degree to which the production process has spread across borders. I thought it was better to break down the competitiveness of trade goods into intermediate goods and final goods in order to understand whether imports or exports are intermediate or final goods. From my research, I came up with this international competitiveness chart. And you can see that countries or industries have moved.

Your second question is very difficult. I do not think anybody has a correct answer to that. Chinese industries are acquiring advanced technologies and it is just a matter of time until China catches up. The question is how Japan will find a path to compete with China. Right now, the Japanese economy has an advantage in many industries. Miscellaneous goods and toys in which China has a strong advantage are not a big problem, at least for me, because they are not sophisticated products. I would say Japan has to focus on certain industries or goods to compete in the world as well as in the East Asian market. I think that is a decision to be made by companies or private institutions. China imports parts and components at a lower price because of its currency, but as a result, it has to sell in the U.S. market at a higher price. I do not know of any specific evidence that shows such a currency movement would benefit the Chinese economy or the East Asian economy, including the Japanese economy.

Q: If all Asian currencies were to appreciate together, would that have a different effect than if the China currency appreciated alone?

A: I think the very simple argument is that if the East Asian market accounts for say 30% of the total trade, and the rest of the world 70%, then that 30% would be affected. However, it is a little more complicated with the triangular trade structure because China or ASEAN imports parts and components and intermediate goods from the rest of East Asia and exports to the U.S. and the EU. We have to think about the relationship between China and the rest of Asia, as well as between China and the U.S. and the EU.

Q: Are many of the exporting firms in China Japanese affiliates? Would the behavior of Japanese and Chinese firms be different if the yuan appreciated?

A: Yes, I think so. And it is not just Japanese companies. There are fully-owned European and U.S. companies exporting from China as well. Their behavior is different, but it is not just about Japan-China relations. It is about China and the U.S. and China and the EU because many foreign countries are going into the Chinese market, not just Japanese companies.

Q: I would like to know the cause of this spiral pattern of development. What is the key factor that led to the new stage? Is R&D investment in each country a crucial factor? Or is it foreign direct investment (FDI)?

A: My impression is that FDI is very important, particularly in East Asia. For instance, Thailand's automobile industry rose from the third quadrant to the second quadrant because of FDI by Japanese automobile manufacturers. The first and foremost factor is FDI, and I think the second reason is R&D, because every industry in every country wants to grow and develop its skills and technologies. There are many ways to do so, including introducing foreign companies' technology, hiring highly skilled engineers from foreign companies, or buying the patent for technologies from foreign companies.

Q: What sort of policies apply to each industry?

A: I think the policies that apply to each industry depend on the phase of development of each industry or country. For instance, Japan's textile industry is approaching the third quadrant. This means it has lost its competitiveness, even in intermediate goods like processed goods or parts and components. Here, R&D is extremely important because the Japanese textile industry cannot go back to quadrants one or two due to its higher labor costs compared to those of other East Asian countries. Therefore, the only way that Japan can compete in the domestic and world markets is by using its technological advantage. Patents are also important.

The government can help the Japanese textile industry by developing patent agreements with China and other East Asian countries. In some cases, developing a brand name, conducting R&D, and developing new technologies does not benefit private industries. Therefore, the government must support or subsidize the development of new technologies. Hence there is some space for the government to assist industries in the fourth quadrant. In the first or second quadrant, however, the Japanese government does not have many ways to help the industry. I think we have to prepare a market where industries compete on equal footing. The competitiveness conditions must not be different for Japanese, European and U.S. companies. This is true not only for strong industries like automobiles but for weak industries like textiles as well.

*This summary was compiled by RIETI Editorial staff.