Abstract

Transport advantage, assembly hub under global value chains, technology advantage, and home-market effect are the four main factors contributing to the emergence of trade hubs in East Asia. Using a region-specific trade “hub-ness” measure, we examine the evolution of trade hubs in the East Asia. A China-Japan twin-hub pattern was found in the 1990s, but the twin hub began to divide in the beginning of the 21st century because of a rising China and declining Japan. We also show that ASEAN as an integrated region may lead to the emergence of another trade hub in East Asia, because of the enlarged ASEAN home-market-size effect, for the highly fragmented automatic data processing industry.

1.  Introduction

Japan and China have been the two leading economic powers in East Asia since the 1990s. Based on the linkages of free trade agreements (FTAs), Baldwin (2004) finds that Japan and China are two hubs surrounded by a group of “spoke” countries in East Asia; that is, a Japan-China twin-hub trade pattern has emerged since the early 2000s. Theoretically, why and when a country develops into a trade hub should differ. Thus, it is worth investigating the long-run evolution of a twin-hub trade pattern, which is one of the main objectives of this paper.

Japan has been the leading country for industry development in East Asia after World War II, starting with traditional labor-intensive industries such as textiles and consumer electronics. After successfully moving up to capital-intensive industries in the 1970s, its labor-intensive industries were shifted to Taiwan, South Korea, Hong Kong, and Singapore, or the four dragons, leading them to become the first-tier industrializing Asian economies (FIAE hereafter). In turn, the FIAE successfully moved up to the capital-intensive industries in the 1980s and shifted the traditional sectors to the second-tier industrializing Asian economies (SIAE, hereafter), comprising the ASEAN-4 (Indonesia, Malaysia, the Philippines, and Thailand). The information and communication technology (ICT) industry embedded with higher technology also followed a similar path of the so-called flying-geese pattern.

Hence, before China's integration into the global economy, East Asia's industrial development was led by Japan, followed by the FIAE, and later the ASEAN-4. China opened up to the global market in the late 1980s and soon became the third-tier IAE in the traditional labor- and capital-intensive industries. However, for the highly technology-intensive integrated circuit (IC) industry, China appears to follow a different pattern of development. China has in fact surpassed the ASEAN-4 to catch up in the IC industry, and it has developed into a major production base within the global value chain (GVC) in East Asia, resulting in a leapfrogging pattern.

Obviously, these paths reflect the importance of Japan's main hub status for trade and industrial development in Asia. Japan remains the major producer and exporter of upstream technology-intensive, key components and manufacturing equipment for the other East Asian countries. Subsequently, the FIAE, ASEAN-4, and China sequentially developed into a manufacturing base for their respective tier leaders.

However, China's evolving engagement with the world market is unique, in terms of its scale and impact on the development order in East Asia and even in the world. Initially, China used its abundant cheap labor to replace the ASEAN-4 as the new assembly base for Japan and the FIAE with the opening up of its economy in the early 1980s. The Asian financial crisis in 1997 marked another milestone for China's development path. By then, the ASEAN-4 had not been able to succeed fully in upgrading to the more advanced capital-intensive industries and had started to lose their low-cost advantage to China in the traditional sectors. Consequently, China developed into a big assembly base not only for Japan and the FIAE in East Asia but also for many multinational companies (MNCs) from Europe and the United States. China's accession into the World Trade Organization (WTO) in 2001 provided a better investment environment that is complemented by a potentially large domestic market to develop traditional sectors, leading to an impressive double-digit economic growth of China for decades. China then further attracted more midstream and upstream capital-intensive and even technology-intensive investments from the MNCs. Consequently, China became the factory of the world, and its trade volume increased because of both its comparative export advantage and increasing imported parts and components from midstream and upstream industries. In 2009, its trade volume surpassed Germany and it became the second-largest trading country in the world. By 2013, China surpassed the United States and became the largest trading country in the world. The position of being the manufacturing base in East Asia and the world's factory have not only boosted China's industrial upgrading, but also enhanced its new status as a regional trade hub in East Asia.

Japan's and China's respective roles of trade in East Asia have developed into a so-called twin-hub pattern, initially addressed by Baldwin (2004) based on the configuration of FTA-related “spoke and hub” linkages among the East Asian economies. Although an FTA-induced double-hub pattern appears intuitive, the de facto trade hub is in fact market driven. In other words, the emergence of double hubs led by Japan-China is essentially market driven by production networks and the huge sizes of their respective domestic markets, rather than by the FTA linkages.1 By comparing their industrial development and main trade products, China and Japan hubs are quite different. Thus, the aim of this paper is to discuss the theoretical reasons for the formation of trade hubs, especially for East Asia. A weighted region-specific “hub-ness” measurement is extended from Baldwin (2004) to examine the evolution of trade hubs in the East Asian region in the last few decades. The likely effect of an integrated ASEAN region on the trade hub pattern in East Asia is also be discussed in this paper.

The rest of this paper is organized as follows: Section 2 provides the theoretical reasons for the formation of a trade hub, and the long-run path of trade dependency on the Chinese and Japanese markets among East Asia countries. In Section 3, a regional hub-ness measurement is introduced and applied to the East Asian region. Specifically, we illustrate the phenomenon of a double hub in East Asia that emerged in the early 1990s. However, the twin hubs began to divide after 2001, with a rising China and a declining Japan. The likely effect of an effectively integrated ASEAN like the ASEAN Economic Community on the twin trade hubs in East Asia are also estimated in this section. The last section concludes.

2.  East Asia's trade hubs

2.1  The formation of trade hubs

The formation of trade hubs can be attributed to the following factors: transportation hub, manufacturing-assembly base, technological advantage, and home-market effect. The transportation hub is analyzed by Krugman (1993) in a three-country case, or countries A, B, and C. Country B can become the trade hub for countries A and C simply because of its cost advantage in transportation. The total trade volume of country B will be inflated because of the addition of redirected trade flows between countries A and C. For example, Singapore's status as a regional trade hub of ASEAN can be partly attributed to its role as a transportation center of ASEAN (Kuo 2017).

Being a manufacturing-assembly base can be another reason to become a trade hub. In vertical production chains, a country can utilize its cheap labor cost to assemble parts and components from other countries for re-exporting to third-country markets. That is, a manufacturing-assembly base is most likely to be created when a country opens up to trade and foreign direct investments with cheap labor; examples are Taiwan, South Korea, Hong Kong, and Singapore in the 1970s, the ASEAN-4 in the 1980s, and particularly China after its accession into the WTO in 2001.

Another reason for being a trade hub is the size of the domestic market, usually represented by population or GDP. The bigger the size of the domestic market, the more likely the country can lower the average cost of production in sectors with scale economies, thus leading to a more than proportional share of the country in the global market, under free trade. This is the so-called home-market effect (HME) as raised by Krugman (1979, 1980). That is, according to the HME, China is more likely to become a trade hub than any other smaller economies of the FIAE or the ASEAN-4.

Being an assembly base under GVC is another reason for being a trade hub. With a huge amount of cheap labor and its gradual opening to the world economy, China has replaced the FIAE as the main labor-intensive manufacturing-assembly base in East Asia since its open-door policy in 1979. Unlike the FIAE in the 1970s and 1980s, China's domestic market is big enough to attract a tremendous volume of exports to China that it re-exports to the world market after assembly operations. As a result, China has become a trade hub in Asia, reflecting its position as the world's factory.

Finally, technological advantages can also enable a country to become a trade hub. Japan is an example. With its superior technology advantage as reflected in its huge number of patent holdings,2 Japan has maintained its position as the major trade center in Asia. Even after transferring its labor-intensive manufacturing industries to the FIAE and China, Japan still holds trade advantages to some extent, such as the supply of manufacturing equipment and key parts and components. In fact, most of the rest of East Asian economies rely on importing Japan's upstream key parts and components, as well as equipment for their downstream production.

2.2  Trends in the evolution of China's and Japan's markets

The global market has gradually expanded since 1979 with China's opening. As China integrated more closely into the world economy, the focus of the global trade market has inclined more toward East Asia. Among the neighboring countries in Asia, Taiwan benefits the most from China's opening because of similarities in language and culture. As shown in Figure 1, the ratio of Chinese trade to Taiwan has risen rapidly since 2005, outpacing the ratios of Japan and the United States to Taiwan trade. China has replaced Japan and the United States to become the most important trade partner of Taiwan. Focusing on Taiwan's exports, China's export ratio exceeded those of the United States and Japan even earlier, in 2003 and 2004, respectively. In 2004, Taiwan's exports to China reached almost 20 percent of its total exports, and China became Taiwan's most important export destination country (Figure 2).

Figure 1.

Taiwan's trade with major trading partners, 2000–16

Figure 1.

Taiwan's trade with major trading partners, 2000–16

Figure 2.

Taiwan's total exports, by main destination countries

Figure 2.

Taiwan's total exports, by main destination countries

The increasing dependence on China's market in the past few decades is not only peculiar to Taiwan. In fact, this holds true for most of China's neighboring countries in East Asia. In China's first opening period, many MNCs in the neighboring countries first moved their labor-intensive manufacturing firms to China in order to enjoy the initially available cheap labor there. This was accompanied by increasing exports of parts and components to China from the midstream and upstream production processes. At the same time, the re-exporting of downstream finished goods to U.S. and European markets from China also increased. The neighboring home countries that sent investments to China and China itself evolved closely in the production process of labor-intensive products.

After the early 2000s, the midstream and upstream production processes and some technology-intensive industries also gradually moved to China, via foreign direct investment, as China's economy continued to grow. As shown in Figure 3, China's trade ratio with the respective major global economies has been increasing dramatically in the 21st century, especially for Taiwan, Korea, and Japan in East Asia. The trade ratios of these three economies are 5 percentage points higher than those with Europe and the United States.

Figure 3.

Ratio of each economy's trade with China to its total trade, 1993–2016

Figure 3.

Ratio of each economy's trade with China to its total trade, 1993–2016

A similar trend of increasing import dependency on China can also be seen for the main economies, as illustrated in Figure 4. As expected, Japan had the highest import ratio and the United States the second highest, after the 2008–09 financial crisis. To summarize, the trade dependency for imports and exports of most major economies has continued to increase in the 21st century. China's East Asian neighbors (Taiwan, Korea, and Japan) have a much bigger trade dependence on China compared with other main economies.3

Figure 4.

Ratio of each economy's imports from China to its total imports, 1993–2016

Figure 4.

Ratio of each economy's imports from China to its total imports, 1993–2016

In contrast with neighboring countries’ rising dependency on China as shown in Figure 3, Figure 5 shows that trade dependence on Japan has been declining. For example, in 1994, Taiwan's trade share with Japan was around 20 percent in 1994 as shown in Figure 5, far above the corresponding 1.1 percent share of its trade with China, shown in Figure 3. However, Figure 1 shows that Japan's share in Taiwan's trade continued to decline while China's share continued to rise, especially after 2001. By 2005, Japan's share in Taiwan's trade had dropped to 16.25 percent as shown in Figure 5, which is lower than the 17 percent of China's share in Taiwan's trade as shown in Figure 3. The same trend for the other main economies can also be observed by comparing Figure 5 with Figure 3. The disparity between the shares of China and Japan in the total trade of the respective economies became even wider after 2005.

Figure 5.

Ratio of each economy's trade with Japan to its total trade, 1993–2016

Figure 5.

Ratio of each economy's trade with Japan to its total trade, 1993–2016

The import dependency on Japan for those economies also declined over the same period, as shown in Figure 6. However, the dependent ratios are more diverse. Taiwan, South Korea, and China are the top three, in terms of the share of imports from Japan relative to their total imports (Figure 6). Taiwan has always held the first position, indicating Taiwan's high dependence on imports from Japan. ASEAN ranks second, followed by South Korea, China, the United States, and the EU. Obviously, the technological-advantage-induced trade hub status of Japan is reflected in this result. To some extent, this is consistent with Choi's (2015) findings that only exports in intermediate goods promote exports in East Asia and not the FTAs. Most of the manufactured goods trade in East Asia is actually trade in parts and components, with many of the final goods exported to the EU and the United States. The different positions of Japan and China in GVCs is the result of their different hub advantages, apart from the advantage of their respective large domestic markets. China has the assembly base advantage, while Japan has the technological advantage.

Figure 6.

Ratio of imports from Japan to total imports for major economies, 1993–2016

Figure 6.

Ratio of imports from Japan to total imports for major economies, 1993–2016

3.  Revisiting the China-Japan twin-hub trade pattern

To illustrate the long-run evolution of the trade hubs in East Asia, we develop a region-specific hub-ness measure (HM), based on the bilateral hub-spoke HM index by Baldwin (2004). Accordingly, the bilateral hub-spoke index HMoh, denoting the “hub-ness” importance of a hub country to country o, is defined as
HMoh=Xoh/jXoj×1-Xoh/iXih=Sohx.(1-Sohm),
(1)
where Xij is the exports from country i to country j, Sohx is the export share of country o to country h, and Sohm is the ratio of imports from country o to the total imports of country h.

Obviously, the value of HMoh lies between 0 and 1. A higher value of HMoh implies higher export dependency of country o on the hub h, and/or country o is less important to country h for its import market.

Using bilateral trade data from the United Nations (UN), we compute year by year the HMoh for each pair of countries. By linking country o and h for those with HMoh > 0.5, as calculated by Baldwin (2004), a pattern with bicycle-like wheels is obtained, with China and Japan as separate hubs, surrounded by separate “spokes” countries. To show the trend, we further extend the HMoh index to region-specific hub-ness measure, denoted as HMhR, by the following formula:
HMhR=oRHMoh.XohXohiRxihiRxih.
(2)
Clearly, it is the weighted average of HMoh for the hub h over all the “spoke” countries o, in the region R. The weight is a ratio of the exports of the export country to the hub country over the exports of all the “spoke” countries to the hub country. A higher HM hub-ness index implies h country's hub-ness for the region is higher.

3.1  Evolution of trade hubs in East Asia

The results of the long-run evolution of the trade hub in East Asia, based on the computed HMhR for East Asian countries (namely, China, Japan, Taiwan, South Korea, Singapore, Malaysia, Thailand, the Philippines, Indonesia, and Vietnam) are depicted in Figure 7. The far higher value of the HM for Japan and China as shown in the figure indicates the existence of the China-Japan twin-hub pattern in the years before 1997.

Figure 7.

Regional HM indexes in East Asia, 1993–2014

Figure 7.

Regional HM indexes in East Asia, 1993–2014

Our findings, which add to the literature, indicate that China's regional HM surpassed Japan in 1997 and has continued to increase since then. By contrast, Japan's regional HM index has been declining over the same period. This demonstrates a clear trend of a rising China and a declining Japan in the evolution of regional trade hub-ness in East Asia over the past few decades. After the 2008–09 global financial crisis, Japan's HM declined steadily to less than 8, while China continued to surge forward to become a unique trade hub in East Asia.

3.2  Evolution of trade hubs by sectors

Theoretically, the reasons for being a trade hub may differ from industry to industry and may lead to different trade hub patterns. We apply the HM approach on two different industries—office machines and automatic data processing machines (SITC 75) and apparel and clothing accessories (SITC 84)—to verify this. As is well known, SITC 75 involves a higher and more diversified GVC, with China serving as the assembly base for the final stage of production. By contrast, SITC 84 is basically final products only. The results are shown in Figures 8 and 9, respectively.

Figure 8.

The HM hub-ness index for SITC 75 in East Asia, 1993–2014

Note: SITC 75: office machines and automatic data processing machines.

Figure 8.

The HM hub-ness index for SITC 75 in East Asia, 1993–2014

Note: SITC 75: office machines and automatic data processing machines.

Figure 9.

The HM hub-ness index for SITC 84 in East Asia, 1993–2014

Note: SITC 84: articles of apparel and clothing accessories.

Figure 9.

The HM hub-ness index for SITC 84 in East Asia, 1993–2014

Note: SITC 84: articles of apparel and clothing accessories.

As shown in Figure 8, before 1998, Singapore was the top trade hub for the industry SITC 75, followed by Japan hub and China hub. After 2001, China hub rose sharply with a tri-hub pattern emerging in the 1990s. However, the evolution of rising China and declining Japan and Singapore finally resulted in one unique China hub after the 1990s.

The evolution of trade hubs for SITC 84 in the textile downstream sector are reported in Figure 9. Significantly different from the SITC 75 sector, no double or triple hubs can be found. Instead, Japan's HM value uniquely stands out before 2006. However, Japan's unique trade hub status declined over time; after 2008, China caught up with Japan to result in a twin-hub pattern. In sum, for the textile downstream industry, it started with a unique Japan hub in the early 1990s but shifted into a weak Japan-China twin-hub pattern, after the financial crisis of 2008.

3.3 Impact of ASEAN integration

As discussed earlier, the size of the home market matters in the evolution of trade hubs. A country with a larger size is more likely to become a trade hub. Therefore, an effectively integrated ASEAN, as implied by the aim of the ASEAN Economic Community, may have the potential of fostering an ASEAN hub. To investigate this likelihood, we treat ASEAN members in our sample as one economy to re-compute the evolution value of the HM over the same period, for the two sectors of SITC 75 and 84. The results are reported in Figures 10 and 11, respectively.

Figure 10.

The HM hub-ness indexes for SITC 75, in ASEAN and East Asian countries

Figure 10.

The HM hub-ness indexes for SITC 75, in ASEAN and East Asian countries

Figure 11.

The HM hub-ness indexes for SITC 84, in ASEAN and East Asian countries

Figure 11.

The HM hub-ness indexes for SITC 84, in ASEAN and East Asian countries

Surprisingly, as shown in Figure 10, the ASEAN hub outweighs Japan after 2006 for the SITC 75 sector, although its HM value is still far below China's value. Therefore, to a certain degree, we can observe a tri-hub pattern with one big China hub and two small ASEAN and Japan hubs, for the period after 2006.

As for the SITC 84 of the downstream textile sector in Figure 11, the unique Japan hub pattern has changed to a weak China-Japan twin-hub pattern after 2006, while an ASEAN hub seems unlikely.

4.  Concluding remarks

In this paper we first raised four likely reasons for fostering a trade hub. Then, based on the bilateral hub-ness index developed by Baldwin (2004), we developed a region-specific trade hub measure and applied it to the East Asian region to investigate the long-run evolution of trade hub patterns in East Asia. The major findings are as follows:

First, we reconfirm the existence of the Japan-China twin-hub pattern in East Asia, as raised by Baldwin (2004). However, the twin-hub pattern appeared much earlier, in early 1990s. The twin hubs began to divide after 1997 as the gap arising from a rising China hub and a declining Japan hub, and widened after China's accession into the WTO in 2001. The home-market-size effect appeared to have contributed greatly to the evolution of the China hub.

At the subsectoral level, for the SITC 75 (automatic data processing industry), it started with a tri-hub pattern of Singapore-Japan-China in the 1990s. But the evolution of a rising China hub and a declining Japan and Singapore finally led to one unique China hub in the recent decade.

In the labor-intensive final products of SITC 84 (apparel and clothing accessories), it started with a unique Japan hub in the early 1990s and transformed into a weak Japan-China twin-hub pattern after the financial crisis of 2008.

Finally, by treating ASEAN as a single economy to estimate the likelihood of an ASEAN hub, we find the possibility of a weak ASEAN hub emerging in SITC 75, but it is not likely to appear in the downstream textile sector of SITC 84.

Notes

*

The authors are grateful to Siow Yue Chia, Prema-chandra Athukorala, and other participants of the Asian Economic Panel (AEP) meeting, Seoul, 29–30 March 2018, for helpful comments on an earlier draft of the paper. We also thank financial support under MOST107-2410-H-001-023 from the Ministry of Science and Technology, Taiwan. Chun-Chien Kuo is the corresponding author.

1 

Global value chains became a popular pattern in East Asia, as addressed in Athukorala (2011). Although FTAs may play some role in the formation of GVCs, market forces seem to be more important. FTA-induced preferential tariff advantages can be offset by the spaghetti bowl effect as raised in Baldwin (2006). Empirical studies by Kimura, Kuno, and Hayakawa (2006) also provide little support to FTA-induced exports.

2 

According to the number of patent holdings by country registered in the U.S. Patent and Trade Office, Japan held 33,222 patents compared with Taiwan's 3,538 in 2001; Japan held 52,409 patents compared with South Korea's 17,924 in 2015. Taiwan and South Korea were the second-ranked patent-holding countries in 2001 and 2015, respectively. That is, Japan's patents outweigh the second Asian country by about 3 to 8 times.

3 

This is consistent to Athukorala's (2011) findings that the pivotal role of China as the premier center of final assembly under GVC has strengthened Asia's economic interdependence in the region.

References

Athukorala
,
Prema-chandra
.
2011
.
Production Networks and Trade Patterns in East Asia: Regionalization or Globalization
?
Asian Economic Papers
10
(
1
):
65
95
.
Baldwin
,
Richard
.
2004
.
The Spoke Trap: Hub and Spoke Bilateralism in East Asia
.
KIEP CNAEC Research Series
04-02
.
Seoul
:
Korea Institute for International Economic Policy
.
Baldwin
,
Richard
.
2006
.
Multilateralizing Regionalism: Spaghetti Bowls as Building Blocks on the Path to Global Free Trade
.
The World Economy
29
(
11
):
1451
1518
.
Choi
,
Nakgyoon
.
2015
.
Global Value Chains and East Asian Trade in Value-Added
.
Asian Economic Papers
14
(
3
):
129
144
.
Kimura
,
Fukunari
,
Arata
Kuno
, and
Kazunobu
Hayakawa
.
2006
.
Does the Number of RTAs Matter?
Paper presented at a joint conference of KEBA, RCIE, and KIET, Globalization and Regional Economic Development, 15–16 December, Gyeong Ju, Korea
.
Krugman
,
Paul R.
1979
.
Increasing Returns, Monopolistic Competition and International Trade
.
Journal of International Economics
94
:
469
479
.
Krugman
,
Paul
.
1980
.
Scale Economies, Product Differentiation, and the Pattern of Trade
.
American Economic Review
70
:
950
959
.
Krugman
,
Paul
.
1993
.
The Hub Effect: Or, Threeness in International Trade
. In:
Theory, Policy and Dynamics in International Trade
,
edited by
W. J.
Ethier
,
E.
Helpman
, and
J. P.
Neary
,
Chap. 3
.
Cambridge, MA
:
Cambridge University Press
.
Kuo
,
Chun-Chien
.
2017
.
The Hub-Spoke Trade Patterns ASEAN: Lessons from the Electronic and Footwear Sectors
.
Paper presented at an international conference, 30 Years after DOI MOI Policy in Vietnam, 29–30 June, Taipei
.