Chapter

III. Implications of Asia’s Regional Supply Chain for Rebalancing Growth

Author(s):
International Monetary Fund. Asia and Pacific Dept
Published Date:
April 2011
Share
  • ShareShare
Show Summary Details

A. Introduction

Persistent global imbalances are raising concerns about the long-term sustainability of strong growth in Asia and across the world. Although current account imbalances in key countries declined in 2010, global imbalances are set to widen again over the medium term, as highlighted in the April 2011 World Economic Outlook. A sustained strengthening of private domestic demand will be needed for Asia to contribute to the resolution of global demand imbalances. Greater flexibility of real effective exchange rates would be an important part of the package of measures that would help to reduce large external surpluses.

The role of exchange rates and shifts in demand in global rebalancing cannot be fully understood without taking into account the increasing role of vertical integration in Asia’s trade.1 Much of the debate over rebalancing growth has been focused on the U.S.-China trade imbalance, and on the role of bilateral exchange rate realignments in resolving this imbalance. However, gross exports can overstate China’s exposure to U.S. demand as they also reflect U.S. demand for intermediate goods of other Asian economies, whose exports to China are used as inputs for Chinese exports to the United States. Moreover, the importance of imported intermediate goods in the production of exports implies that the price competitiveness of exporters depends not only on their own currency but also on the currencies of their suppliers. A depreciation of supplier currencies, for example, would mitigate the loss of competitiveness from a real exchange rate appreciation in the home country. Separately, understanding Asian supply networks can also help assess the impact of disruptions to production in one country, as in Japan after the March 2011 earthquake and tsunami, on the rest of the region.

Against this background, this chapter focuses on two main questions:

  • How are Asian economies linked through vertical integration, and to what extent do they compete in third markets?

  • How do vertical linkages, on the one hand, and competition on third markets, on the other, affect the role that exchange rates and shifts in global demand can play in reducing external imbalances?

The analysis proceeds in three steps. Section B of the chapter describes the most salient features of Asia’s emergence as the world’s leading exporter. Section C discusses the nature of bilateral trade links in the region and of deeper vertical integration. Section D focuses on the potential implications of vertical trade integration in Asia for the effect of exchange rates and shifts in global demand on export competitiveness and external surpluses.

The main findings can be summarized as follows. Over the last decade, Asian economies have become part of a greater supply network, increasingly centered on China.2 Not only has China relied intensively on imported inputs from the region, in particular from Japan and Korea, but the rest of Asia has also increasingly relied on inputs from China. Nevertheless, sustained disruptions to production in Japan could also affect regional supply chains. As a result of growing vertical integration, the competitiveness of Asian economies depends not only on movements of their own currency, but also of the currencies of their suppliers. In other words, the price of exports to the final market (for example, the United States) will rise by more if the exchange rates of both the final exporter and its suppliers appreciate (vis-à-vis the currency of the final market) than if the exchange rate of only the final exporter appreciates. Moreover, in the presence of vertical integration, China’s share of Asia’s surplus vis-à-vis the United States is smaller than suggested by balance-of-payments data, which do not account for Asia’s intermediate exports that are embedded in China’s exports to the United States. Hence, it is misleading to focus on bilateral imbalances, as effective rebalancing will involve all major Asian economies.

B. Asian Exporters: Between Partnership and Competition

Asia is the world’s leading source of exports, but the most striking trend is the rapid growth in intraregional trade.3 While global trade and Asia’s trade with economies outside the region have doubled since 2000, intra-Asian trade has tripled, and regional trade involving emerging Asia, in particular, has increased even faster. As a result, Asian economies accounted for 35 percent of world exports in 2009, compared with 25 percent 10 years earlier, with the share of intraregional exports rising to 55 percent from 45 percent over the same period (Figure 3.1).

Figure 3.1.Asia: Export Trends

Source: United Nations, Comtrade database.

Intermediate goods exports have accounted for about 70 percent of the annual export growth in Asia over the last decade—more than double the contribution of capital and consumer goods together.4 This has been particularly the case for the ASEAN-5, the NIEs, and Japan (Figure 3.2). Exports of intermediate goods have been particularly strong to other Asian economies, whereas consumer goods and intermediate goods have contributed roughly equally to the increase of exports outside Asia. As a result, intraregional exports are more intensive in intermediate goods than are other exports. The average share of intermediate goods exports between Asian trading partners has increased to nearly 80 percent in 2009 from about 60 percent a decade earlier (Figure 3.3).

Figure 3.2.Selected Asia: Contributions to Export Growth

(In percent of total export growth; 1998–2009 average)

Source: United Nations, Comtrade database.

Figure 3.3.Asia: Exports of Intermediate Goods

Source: United Nations, Comtrade database.

Asian economies increasingly have formed a supply network, with China taking the role of an assembly hub for final goods exports, notably consumer goods. In this “triangular trading pattern” (Gaulier, Lemoine, and Ünal-Kensenci, 2005), direct export competition has been accompanied by trade partnerships. For Asia, excluding China, the share of intermediate goods exports to China in total exports has doubled over the last decade, whereas the share of direct consumer goods exports to the United States and euro area has steadily declined (Figure 3.4). A corollary of this development is the ongoing shift of market share in leading export markets toward China. For instance, while Asia’s share in U.S. consumer goods imports has remained stable over the last two decades at about 40 percent, China’s share has risen nearly threefold, from 9 percent to more than 25 percent, whereas shares of other Asian exporters, particularly Japan and the ASEAN-4 among emerging Asian economies, have declined (Figure 3.5).

Figure 3.4.Asia (excl. China): Exports to G-2 and China

(In percent of total exports)

Source: United Nations, Comtrade database.

Figure 3.5.Share in U.S. Imports: Consumer Goods

(In percent)

Source: United Nations, Comtrade database.

As a result, competition in intermediate goods markets has become more important for Asian economies’ overall exports than competition at the final goods level. To assess the intensity of competition in a simple way, we use an export similarity index (Appendix 3.1). The closer that index is to 100, the more similar is the export structure between two Asian economies and thus the higher is the degree of potential competition between them. The results suggest, as one might expect, that the degree of export competition is generally higher for countries at comparable levels of development. This holds across all major product categories, that is, consumer, capital, and intermediate goods exports (Figure 3.6 on capital goods, for example).

Figure 3.6.Selected Asia Export Similarity Index

(Top five country pairs for capital goods)

Sources: United Nations, Comtrade database; and IMF staff calculations.

Moreover, competition appears to have increased over time, broadly in line with patterns of economic convergence across Asia. For example, in the case of capital goods, only five country pairs had an index of more than 80 in 1998–2002, compared with nine pairs in 2005–09. According to this metric, competition in capital goods exports was initially relatively stronger between Japan and Korea and between China and the ASEAN-5, but has since become stronger also between both Japan and China, and the NIEs and China.5

C. Evolving Trade Networks: Old and New

A closer look at vertical integration of exports between country pairs underscores the profound role of China in regional trade.

  • Based on direct and indirect trade flows for intermediate goods, and on data from Asian input- output tables that allow one to trace the final sources of supply and demand, we estimate that China ultimately accounts for about 50 percent of all trade flows in imported inputs in Asia, more than double its share in 1995.6 As such, China has become, for many of its Asian trading partners, the single most important destination of intermediate goods exports. The predominant intermediate goods suppliers to China have consistently been Japan, Korea, and Taiwan Province of China, accounting for almost 80 percent of China’s imported inputs (Figure 3.7; Appendix Table 3A.1).7

  • But China’s role as a supplier to other Asian economies has also rapidly grown. China’s share in direct and indirect intermediate goods exports within Asia has doubled to nearly 30 percent in 2009, from 15 percent a decade earlier.

  • Korea has the closest link to China. It directly exports nearly 40 percent of its intermediate goods exports to China, four times its exports to Japan, its second most important export destination of intermediate goods. Adding indirect exports—intermediate goods that Korea exports to other countries in the region, such as the ASEAN-4, and eventually find their way into Chinese production—Korea exports about 60 percent of its intermediate goods to China.

  • China’s share has also increased in the intermediate goods exports of the ASEAN-4 economies. At the beginning of the 2000s, Japan was, by a wide margin, the most important destination of intermediate goods exports, accounting for an average 20 percent of direct exports from these countries. A decade later, Japan’s share has fallen to an average 15 percent. By contrast, China’s share more than tripled over the same period from about 4 percent to 14 percent.

  • Considering that, in addition to intermediate goods, capital goods are also used as inputs, the supply-chain links from Japan and Korea to China are likely significantly stronger. China’s ascent as a leading exporter has been fuelled by a high investment rate (April 2010 Asia and Pacific Regional Economic Outlook).8 For Japan and Korea, capital goods exports to China now account for 20 percent to 25 percent of their total capital goods exports (a fourfold increase from a decade earlier), making China by far their single most important capital goods export destination in Asia, and comparable to the United States or the European Union as export markets (Figure 3.8).9

Figure 3.7.Selected Asia: Direct and Indirect Intermediate Goods Trade

(In percent of total regional flows)

Sources: Institute of Development Economics, Japan External Trade Organization, Asian Input Output Tables, 2000; United Nations, Comtrade database; and IMF staff calculations.

Figure 3.8.Japan and Korea: Export of Capital Goods

(By destination; in percent of total exports)

Source: United Nations, Comtrade database.

Asian trade networks outside China have become relatively less significant over time. This trend has been particularly evident for Japan, which was the central hub in the 1990s, accounting for about 40 percent of imported inputs in other regional economies, compared with about 25 percent in 2008 (Figures 3.7 and 3.9). However, unlike China, Japan has traditionally taken the role of net exporter of parts and components rather than net importer.

Figure 3.9.Selected Asia: Change in Intermediate Goods Trade

(Change in percent share in regional flows)

Sources: Institute of Development Economics, Japan External Trade Organization, Asian Input Output Tables, 2000; United Nations, Comtrade database; and IMF staff calculations.

  • As a result, direct and indirect input linkages between the two leading industrial economies in the region, Japan and Korea, have also become relatively less important. Although their mutual supply relationships accounted for about 10 percent of regional import flows in intermediate goods in 1995, a decade later the Japan—Korea import links represent less than 5 percent.

  • Intermediate goods trade among the NIEs and the ASEAN-4 has generally been less significant than the links described above as a share of regional flows, with direct and indirect bilateral flows of intermediate goods typically accounting for less than 1 percent of regional flows. However, notable exceptions include the outsourcing links between Malaysia and Singapore, and Taiwan Province of China and Korea, which each accounts for about 1½–2 percent of intra-Asian intermediate trade flows.

Although Asian supply chains have become increasingly centered on China, sustained disruptions to production in Japan as a result of the March 2011 earthquake and tsunami could affect them. For all major Asian economies, Japan remains the second most important source of intermediate inputs after China, and it remains the single most important supplier to China, where it accounts for 36 percent of all imported inputs sourced from Asia. Moreover, its supply of electronic components and capital goods may not be easily substitutable as the outsourcing hierarchies reflected in bilateral surplus and deficit positions in intermediate goods trade have remained relatively stable. Japan has consistently run intermediate goods surpluses with all its Asian trading partners, except for a small deficit with the ASEAN-4 economies (Figure 3.10).10

Figure 3.10.Japan: Trade Balance in Intermediate Goods

(In billions of U.S. dollars)

Source: United Nations, Comtrade database.

D. How Does Vertical Integration Affect External Competitiveness and Rebalancing?

The increased degree of vertical trade integration in Asia affects the impact that movements in exchange rates have on the competitiveness of Asian economies. In the presence of vertical integration, the cost of imported intermediate inputs can account for a significant share of the exporter’s total cost. On average over 2005–08, the share of imported value added in exports ranged from about 10 percent in Japan to about 40 percent in the smaller open economies, such as Malaysia, having increased sharply since the previous decade.11 The cost of the imported intermediate inputs (in terms of the final export market currency) is not affected by the exchange rate of the exporter, but by the exchange rate of its suppliers (relative to the export market currency). Hence, if only the exchange rate of the exporter appreciates, while those of its suppliers remain constant (in terms of the export market currency), the impact on the exporter’s total costs in export market currency terms is lower than if the currencies of the suppliers also appreciate.12

Conventional real effective exchange rates only partially reflect changes in export competitiveness. Real effective exchange rates compare the prices of domestic and foreign goods at the same stage of production, and therefore do not explicitly account for the impact of movements in the exchange rates of “upstream” supplier economies on the costs in “downstream” export markets. Based on Thorbecke (2010) and Thorbecke and Smith (2008), we augment real effective exchange rate calculations to take account of vertical linkages and estimate an “integrated effective exchange rate (IEER)” based on the breakdown of an export good’s total costs (expressed in the export market currency) into the value added produced directly by the exporter and the value added produced by its supplier economies.13 The main difference of the IEER from the conventional real effective exchange rate (REER) for a given country is that in addition to that country’s own exchange rate movements, it also factors in its supplier economies’ exchange rate movements vis-à-vis its final export markets.14 The main results are as follows (Figure 3.11):

  • In the case of China, the integrated effective exchange rate has appreciated more slowly in recent years than the conventional effective exchange rate. This reflects the fact that currencies of important China-supplier economies, in particular Korea, have depreciated vis-à-vis China’ major export markets.

  • Likewise, although Korea’s real effective exchange rate has depreciated by about 25 percent since the global financial crisis, its integrated effective exchange rate has depreciated much less (by only about 15 percent), as the currencies of key suppliers to Korea’s economy (in particular Japan and China) have appreciated over the same period.

  • By contrast, changes in Japan’s integrated effective exchange rate have been more closely tracked by movements in the real effective exchange rate, and in particular, its appreciation since 2007. This is explained by the relatively high local content of Japanese exports and because the currency of the most important supplier, China, also generally appreciated.

Figure 3.11.Selected Asia: Standard and Integrated Real Effective Exchange Rates (REER)

(2000 = 100)

Sources: Institute of Development Economics, Japan External Trade Organization, Asian Input Output Tables, 2000; United Nations, Comtrade database; IMF, Information Notice System database and staff calculations.

In general, movements in the integrated exchange rate appear to be more closely related to exports than changes in the conventional real effective exchange rate (Figure 3.12). The relatively strong and statistically significant correlation between changes in the integrated exchange rate and export growth suggests that accounting for vertical trade integration is important when gauging the effect of exchange rate movements on external competitiveness.15

Figure 3.12.Asia: Real Effective Exchange Rates (REER) and Real Export Growth

Sources: IMF, Information Notice System database, and staff calculations.

Ignoring the role of imported inputs in total export costs, the conventional real effective exchange rate tends to move more than the integrated one, and generally overstates the potential impact of a country’s exchange rate movements on its exports.16 By contrast, adjusting the real effective exchange rate to reflect the degree to which Asian economies compete in third markets yields another alternative measure of effective exchange rates, one that tends to move relatively closely with the conventional one (Box 3.1). This also appears to support the view that vertical integration is a relatively more important determinant of Asian economies’ export competitiveness than horizontal competition (in goods at the same stage of production) in third markets.

Ignoring vertical integration as the primary driver of intra-Asian trade may also overstate the importance of regional domestic demand as a driver of growth. As highlighted in the April 2010 Asia and Pacific Regional Economic Outlook, vertical integration masks the importance of final sources of demand outside Asia, especially for the smaller, more open regional economies. The upshot, however, is that the loss of market shares of Asian economies to China in final goods exports to advanced economies understates their total exposure. Through vertical trade, all Asian economies have increased their indirect trade exposure to the United States and other advanced economies.

Moreover, bilateral trade surpluses based on gross exports are an imperfect indicator of global demand imbalances. The bulk of the U.S. trade deficit in consumer goods, for example, appears to roughly match China’s trade surplus in consumer goods (Figure 3.13). But we have shown that Chinese exports embed a significant amount of inputs imported from other economies in the region. A measure of bilateral trade imbalances that accounts for vertical integration should repartition the imported inputs embedded in China’s exports across Asian economies, and add them to their direct exports. On such value-added-based terms, China’s trade surplus with the United States shrinks, whereas that of most other Asian economies increases. Similar to the impact of exchange rate movements, this suggests that as the degree of vertical trade integration increases, imbalances should be assessed at the level of supply chain networks on a value-added basis instead of individual countries’ trade balances based on gross exports and imports (Figure 3.14).

Figure 3.13.Net Exports to World Markets: Consumer Goods

(Percent share in world exports minus percent share in world imports, excluding intragroup trade ; 2005–08 average)

Source: United Nations, Comtrade database.

Figure 3.14.Selected Asia: Share in Trade Surplus vis-à-vis U.S.–Value Added and Gross Trade

(In percent of total; 2005–08 average)

Sources: Institute of Developing Economies, Japan External Trade Organization, Asian Input Output Table, 2000; United Nations, Comtrade database; Haver Analytics; and IMF staff estimates.

E. Conclusion

Asia’s high degree of vertical integration implies that a durable reduction in its imbalances will require adjustment across all major Asian economies and currencies. The economies of the region have become part of a greater supply network, with China taking the role of an assembly hub for final goods exports, notably consumer goods. Hence, gross trade exposures mask the impact of shifts in global demand on Asian economies’ exports and growth. Moreover, the new trade structures defy conventional views of the impact of exchange rates on external competitiveness. A real effective exchange rate appreciation, unless it is matched by appreciations of supplier currencies, may have a smaller impact on exports and global rebalancing than conventional trade models and effective exchange rates suggest.

Box 3.1.Horizontal Competition and the Real Effective Exchange Rate

To account for the fact that product differentiation tends to lower the degree of potential competition, we also calculate a similarity-adjusted real effective exchange rate (Appendix 3.1 for details). The implication is that countries whose exports are more similar should be given greater weights when effective exchange rate measures are used to assess the impact of bilateral exchange rate movements on competitiveness. However, comparing the weights of our similarity-adjusted real effective exchange rate with those of the conventional real effective exchange rate suggests that major deviations, in particular concerning big trading partners, are typically limited. As a result, similarity-adjusted real effective exchange rates tend to move relatively closely with conventional ones. Nevertheless, there are a few noteworthy country-pair exceptions where actual competition may differ from what is implied by weights in the conventional effective exchange rates.

  • For consumer goods exports, the weight of the euro area countries in Japan’s effective exchange rate is about 5 percentage points higher than simple trade weights in the conventional real effective exchange rate measure suggest. A similar adjustment applies to the weight of the United States in Korea’s consumer goods exports.

  • For capital goods, the weights of Japan and China in Korea’s similarity-adjusted effective exchange rate are about 2 percent and 5 percentage points higher, respectively.

  • Regarding intermediate goods, the weight of Japan in Korea’s basket and vice versa is about 3 percentage points higher, an increase of more than 20 percent over the conventional trade weights (figure). Competition, or similarity-adjusted, trade weights between China, on the one hand, and Japan and Korea, on the other, are also higher, although the required adjustments are generally smaller.

Japan and Korea: Change in Similarity-Adjusted Exchange Rate Weights

(By major intermediate goods trading partner; in percent of standard weight)

Source: IMF staff calculations.

Note: The main authors of this box are Adil Mohommad, Olaf Unteroberdoerster, and Jade Vichyanond.
Appendix 3.1.

Export similarity index

The export similarity index between two economies j1 and j2 is defined as

where s(i, j) represents the share of good i in country j’s exports.

Because the index is more precise the higher the degree of disaggregation (or the greater the number of goods), it is based on SITC-5 level trade statistics. At this level, consumer goods comprise 287 line items, 171 capital goods, and 884 intermediate goods. Note that the similarity index is invariant to economic size because it is based on shares in total exports.

Integrated effective exchange rate

The measure of the integrated effective exchange rate for a country is defined as (Thorbecke, 2010; and Thorbecke and Smith, 2008):

where j denotes the country’s export destinations, k its suppliers, DVA the share of domestic value added in gross exports, reer the real effective exchange rate, rer the real bilateral exchange, inputk the share of supplier k in the country’s intermediate goods imports (in value-added terms), and wj the weight of destination j in the country’s exports.

Our estimates of the share of domestic value added in exports and the shares of supplying economies in imported value added are taken from Mohommad, N’Diaye, and Unteroberdoerster (2011). They are based on analysis of Asian input-output (AIO) tables encompassing nine regional economies (China, Indonesia, Japan, Korea, Malaysia, the Philippines, Singapore, Taiwan Province of China, and Thailand) and the United States. We obtain AIO 2000 tables from the Japanese External Trade Organization and follow updating procedures in Pula and Peltonen (2009) using National Accounts and the United Nations, Comtrade database.

To the extent that they cover similar countries, our estimates, in particular the variation across Asian economies and over time, are generally comparable with other studies using different data sets and covering earlier time periods (such as Hummels, Ishii, and Yi, 2001 for OECD countries; Koopman and others, 2010 using GTAP data; and Powers, Wang, and Wei, 2009 using Asian input-output tables for 1990 and 2000).

Similarity-adjusted effective exchange rate

We define our measure of the similarity-adjusted real effective exchange rate as:

simj, i represents the export similarity index between country j and i, wj, i is the weight of country j in country i’s real effective exchange rate basket, and rerj, i is the bilateral real exchange rate.

As such, reersimi is country i’s real effective exchange rate where the weights of trading partners are adjusted for the degree of export similarity. The adjusted weights are normalized to one.

reersimi is calculated for capital, consumer, and intermediate goods, with a country’s overall effective rate being a weighted average of the three and the weights derived from their respective share in total exports.

Methodology for construction input flows matrix in Figure 3.7.

Based on Asian cross-border input-output tables, we derive the Leontief inverse matrix. Each element of a given column in this matrix specifies the input requirement from each row country i for a unit of output in the column country j, both flowing directly from i to j and indirectly from i to any third country k, and from k to j (the indirect relationships are not restricted to 3-step chains, but take into account chains of any length within the input-output relationship). Multiplying gross output of each column country j with the i, j-th coefficient in the inverse matrix gives total direct and indirect input flows from i to j. Summing these input flows into all j’s equals total input flows within the region. In Figure 3.7., each cell in the grid indicates the share of a given bilateral flow from i into j in the total (direct and indirect) input flows within the region.

Table 3A.1.Share in Intermediate Goods Exports1(In percent of total exports)
1998–2000AustraliaChinaHong Kong SARIndonesiaIndiaJapanKoreaMalaysiaNew ZealandPhilippinesSingaporeThailandTaiwan Province of ChinaVietnamUnited StatesEuro area
Australia7.64.13.13.125.511.62.35.71.86.62.57.40.58.79.7
China1.326.51.71.517.68.31.70.21.33.51.94.30.916.612.9
Hong Kong SAR0.962.60.71.23.82.31.30.11.43.51.43.70.311.15.8
Indonesia3.86.53.62.429.49.93.70.22.010.62.05.41.09.010.4
India1.53.315.91.97.22.61.90.20.63.22.82.10.831.524.6
Japan1.39.68.02.20.98.64.50.23.45.54.59.70.529.112.0
Korea2.017.511.22.91.511.44.70.23.95.32.16.81.720.58.3
Malaysia2.04.15.41.83.213.14.20.32.120.84.75.90.620.610.9
New Zealand30.35.52.61.71.519.58.41.71.41.51.12.80.514.07.5
Philippines0.42.26.60.50.214.03.85.40.010.74.58.70.226.915.7
Singapore2.24.26.914.62.05.63.320.60.32.75.36.10.914.810.8
Thailand1.66.58.22.51.116.33.06.80.22.713.06.61.816.912.9
Taiwan Province of China1.24.531.01.70.710.83.63.50.23.25.22.41.619.810.8
Vietnam18.717.01.53.00.416.61.85.10.16.211.14.93.83.16.6
2007–2009AustraliaChinaHong Kong SARIndonesiaIndiaJapanKoreaMalaysiaNew ZealandPhilippinesSingaporeThailandTaiwan Province of ChinaVietnamUnited StatesEuro area
Australia24.11.12.310.224.711.12.13.00.72.63.64.71.12.86.0
China1.821.12.14.410.810.12.60.21.33.22.14.02.317.017.1
Hong Kong SAR0.672.80.72.63.02.41.10.00.91.91.62.41.04.94.0
Indonesia3.612.51.67.624.89.46.00.42.09.53.33.41.55.59.0
India1.218.611.53.03.93.53.00.20.83.62.81.52.419.324.6
Japan1.124.86.82.31.312.23.40.12.13.46.29.11.415.810.0
Korea1.238.18.82.33.39.42.30.21.93.92.46.02.410.47.5
Malaysia4.013.65.63.85.013.55.70.51.915.56.53.71.89.19.8
New Zealand34.211.81.63.31.812.65.52.32.13.22.31.71.410.06.1
Philippines0.712.313.91.10.615.84.74.20.07.43.55.01.18.621.1
Singapore2.712.912.89.73.85.56.114.40.32.35.04.61.79.09.2
Thailand5.114.99.55.13.414.93.48.10.42.57.32.74.59.68.5
Taiwan Province of China1.133.218.81.41.28.04.62.70.12.34.52.32.910.16.8
Vietnam15.714.31.22.81.020.73.76.00.12.18.94.43.08.67.3
Sources: United Nations, Comtrade database; and IMF staff calculations.
Table 3A.2.Share in Capital Goods Exports1(In percent of total exports)
1998–2000AustraliaChinaHong Kong SARIndonesiaIndiaJapanKoreaMalaysiaNew ZealandPhilippinesSingaporeThailandTaiwan Province of ChinaVietnamUnited StatesEuro area
Australia3.75.03.51.65.22.02.518.31.310.21.22.70.724.117.9
China1.220.51.10.412.03.61.40.10.83.90.92.60.430.121.0
Hong Kong SAR1.135.40.30.46.61.91.40.21.43.41.23.70.327.515.1
Indonesia1.70.92.70.59.80.44.60.21.233.86.51.40.422.813.0
India1.82.73.41.85.60.93.40.21.17.82.60.82.826.338.9
Japan2.85.55.41.10.85.42.40.41.75.22.49.70.436.420.4
Korea2.76.05.21.21.79.42.20.21.34.01.25.90.833.424.8
Malaysia5.01.94.61.51.612.01.20.51.415.31.72.80.332.317.9
New Zealand48.11.14.10.30.72.61.21.10.43.00.31.20.222.413.3
Philippines2.70.53.30.40.023.63.00.50.14.80.611.60.130.918.0
Singapore2.82.25.99.81.87.42.99.80.41.62.64.50.732.215.5
Thailand7.12.03.90.60.614.81.22.90.81.610.62.00.528.822.8
Taiwan Province of China1.81.514.00.80.516.32.32.00.10.92.91.50.736.618.0
Vietnam3.50.77.92.00.137.67.81.20.10.86.81.211.11.817.5
2007–2009AustraliaChinaHong Kong SARIndonesiaIndiaJapanKoreaMalaysiaNew ZealandPhilippinesSingaporeThailandTaiwan Province of ChinaVietnamUnited StatesEuro area
Australia7.13.92.42.23.31.92.022.90.77.11.90.80.929.014.0
China1.922.41.33.27.04.21.80.20.54.11.31.81.226.223.2
Hong Kong SAR2.344.00.82.06.82.01.40.30.83.31.22.10.617.315.1
Indonesia3.47.04.21.412.52.64.50.51.527.23.91.01.213.915.1
India3.33.91.84.72.01.28.00.10.723.12.90.41.723.722.4
Japan2.819.16.52.01.88.52.50.41.44.63.87.81.421.316.3
Korea1.225.34.31.32.66.22.00.20.86.21.31.61.519.226.6
Malaysia3.78.14.92.22.56.91.70.40.89.44.31.60.936.915.7
New Zealand44.62.31.30.81.42.41.51.00.25.41.40.50.226.810.2
Philippines0.410.95.50.80.220.26.25.60.02.51.81.70.232.911.1
Singapore3.48.67.416.65.86.33.014.50.42.14.42.62.515.46.9
Thailand7.218.18.22.71.810.51.84.60.61.04.92.11.617.617.3
Taiwan Province of China1.432.89.71.11.45.31.72.00.10.52.22.01.621.916.3
Vietnam3.19.75.90.60.615.53.63.40.31.17.42.41.522.123.1
Sources: United Nations, Comtrade database; and IMF staff calculations.
Table 3A.3.Share in Consumer Goods Exports1(In percent of total exports)
1998–2000AustraliaChinaHong Kong SARIndonesiaIndiaJapanKoreaMalaysiaNew ZealandPhilippinesSingaporeThailandTaiwan Province of ChinaVietnamUnited StatesEuro area
China2.121.40.70.224.72.30.60.30.51.30.21.00.430.613.8
Hong Kong SAR2.612.10.30.19.40.90.50.30.81.60.51.80.448.120.5
Indonesia2.01.23.30.615.10.62.20.20.713.51.51.10.236.920.7
India2.11.33.10.38.70.81.80.40.72.10.60.60.442.534.6
Japan2.32.79.00.50.23.81.00.31.63.91.78.20.340.424.1
Korea1.75.84.01.10.430.20.50.30.61.20.62.01.336.813.2
Malaysia3.20.65.01.20.515.91.00.61.016.41.51.60.536.414.7
New Zealand21.72.14.41.20.215.61.63.62.12.92.03.80.822.415.7
Philippines1.01.13.20.70.115.71.30.70.11.60.61.50.262.89.4
Singapore4.04.47.611.62.016.53.012.20.52.83.12.52.216.910.7
Thailand2.61.73.52.30.421.10.92.10.30.84.51.70.741.615.8
Taiwan Province of China1.90.710.60.70.217.61.31.10.21.01.61.40.745.215.9
Vietnam1.98.13.41.10.424.93.81.80.21.83.11.08.79.530.2
2007–2009AustraliaChinaHong Kong SARIndonesiaIndiaJapanKoreaMalaysiaNew ZealandPhilippinesSingaporeThailandTaiwan Province of ChinaVietnamUnited StatesEuro area
China3.011.70.70.616.03.41.60.40.71.70.71.00.635.922.0
Hong Kong SAR3.516.80.40.58.52.00.50.30.61.90.52.51.236.824.0
Indonesia2.51.91.91.19.71.73.90.31.76.41.81.11.342.422.3
India2.31.63.30.32.70.32.20.30.92.50.70.21.641.439.5
Japan3.010.511.00.90.57.01.10.41.33.03.26.51.029.321.4
Korea3.814.35.41.41.321.70.60.81.31.51.63.62.926.713.0
Malaysia7.72.63.53.71.710.91.71.11.918.06.21.51.924.613.0
New Zealand23.47.23.23.20.310.53.73.12.72.31.93.80.917.216.5
Philippines1.41.42.51.90.414.72.93.50.42.23.21.21.446.216.6
Singapore4.011.57.08.63.010.93.010.90.52.58.32.43.114.39.9
Thailand4.33.13.92.71.518.61.34.80.62.83.71.62.530.718.0
Taiwan Province of China2.96.511.80.90.613.43.91.50.41.31.73.51.834.915.1
Vietnam2.32.71.51.10.311.12.91.80.24.81.20.72.741.225.4
Sources: United Nations, Comtrade database; and IMF staff calculations.

Note: The main authors of this chapter are Adil Mohommad, Olaf Unteroberdoerster, and Jade Vichyanond.

Vertical integration refers to the geographic dispersion of production processes across countries and international trade in intermediate goods (as opposed to final goods).

In documenting these trends, this chapter builds on work in previous IMF Regional Economic Outlooks. IMF (2007a) focuses on intraregional trade networks in Asia; IMF (2008) proposes measures for assessing indirect trade exposures that are masked by vertical integration; and Mohommad, N’Diaye, and Unteroberdoerster (2011) uses Asian input-output analysis to derive measures of export dependence based on value added.

Throughout this chapter, unless otherwise mentioned, “Asia” is defined as comprising the ASEAN-4 (Indonesia, Malaysia, Philippines, and Thailand), Australia, China, India, Japan, New Zealand, and NIEs (Hong Kong SAR, Korea, Singapore, and Taiwan Province of China).

The definition of consumer goods, capital goods, and intermediate goods is based on the broad economic categories (BEC) classification in the United Nations, Comtrade database. Intermediate goods range from raw materials, such as primary industrial commodity supplies (food, nonfood, and fuels) to processed industrial supplies, parts, and accessories.

This overall trend appears to be particularly clear for the subsegment of high-tech exports. First, the number of country pairs with a similarity index of 90 or higher has increased from four during 1998–2002 to six in 2005–09. Second, in the earlier period, the top-ranked pairs consisted of either industrial or emerging Asian economies. Third, in line with its rapid economic development, China’s similarity indices of high-tech exports with both Japan and Korea have risen sharply, and, with the latter now close to 100, are significantly above any other pair. Consistent with economic convergence, Murshed (2001) finds a rise in intra-industry trade in manufactured goods for all Asian economies. Nevertheless, although consistent with broad economic trends, the similarity index may overstate the degree of competition as it does not capture differences in quality or technological sophistication within product categories.

For a detailed description of the methodology used to update the 2000 Asian input-output tables provided by the Japan External Trade Organization (JETRO) and to calculate direct and indirect trade exposures, see Appendix 3.1 and Mohommad, N’Diaye, and Unteroberdoerster (2011). The JETRO data encompass nine regional economies, including China, Indonesia, Japan, Korea, Malaysia, the Philippines, Singapore, Taiwan Province of China, and Thailand.

Estimates of the value-added content of Chinese exports also point to China taking a greater share of the value chain (that is, bringing some of the intermediate inputs production onshore).

Consistent with the role of capital goods as imported inputs, Eichengreen, Rhee, and Tong (2007) find that Asian exports to China are positively affected by the growth of Chinese exports and “that this effect is mainly felt in markets for capital goods.” Similarly, Johansson (2006) finds that inward FDI positively affects Chinese electronics exports.

For a description of various types of global value chains and their impact on trade and investment patterns, see Milberg and Winkler (2010). Using firm-level surveys, Kawai and Wignaraja (2011) find that growing production networks have benefited from firms’ growing use of regional free-trade agreements.

The relatively stable outsourcing hierarchies at the country level are mirrored by relatively stable cross-border production networks at the firm level, which are often led by multinational companies in advanced Asia as documented by Kimura and Ando (2005); Borrus, Ernst, and Haggard (2000); and Thorbecke (2010).

As an example, assume a 3-country world in which country A exports only final goods to country B. Assume further that the domestic value added in A’s exports is 50 percent, while the other 50 percent are imported inputs from country C. Let A’s exchange rate appreciate 10 percent vis-à-vis B and C (and hence the cross-rate between B and C remains constant). Although A’s exchange rate appreciates 10 percent vis-à-vis B, the effective cost of A’s exports in B’s currency only rises 5 percent. This is because supplier C’s exchange rate vis-à-vis B is unchanged, and supplier C accounts for 50 percent of the value of A’s exports to B.

Of course, the input shares could change in response to a change in relative costs induced by exchange rate movements, but such changes typically happen only gradually over time.

As with the conventional REER, trading partner weights are applied to a given country’s own exchange rate movements, which matter for the share of domestic value added in that country’s exports. As for the share of imported value added, an effective rate is applied, taking into account the share of each supplier in the imported value added and the share of each export market in exports (and applying these weights to the real exchange rate between the suppliers and export markets). See Appendix 3.1 for details.

This result also holds if the integrated exchange rate is calculated based on estimates of the domestic and imported value added in exports from other studies (Koopman and others, 2010, based on Global Trade Analysis Project (GTAP) data for 2004; and Powers, Wang, and Wei, 2009, based on Asian input-output tables for 2005).

Consistent with the generally stronger correlation of the integrated effective exchange rate with export growth, Thorbecke and Smith (2008), using a standard export model (Goldstein and Khan, 1985) in a panel data set for Chinese exports to 33 countries, find that an appreciation of the renminbi has a much larger negative impact on exports with a high domestic value-added content than on those with a low domestic value-added content. In addition, using the integrated exchange rate approach, Ahmed (2009) finds that in the case of China, a renminbi appreciation vis-à-vis supplier economies in the region has no significant impact on processing goods exports (that is, exports with a relatively low domestic value-added content).

    Other Resources Citing This Publication