Information about Asia and the Pacific Asia y el Pacífico
Journal Issue

Chapter 1: China’s Changing Patterns of Trade

Koshy Mathai, Geoff Gottlieb, Gee Hee Hong, Sung Eun Jung, Jochen M. Schmittmann, and Jiangyan Yu
Published Date:
September 2016
  • ShareShare
Information about Asia and the Pacific Asia y el Pacífico
Show Summary Details

This chapter takes a close look at China’s trade data to document how patterns of trade are changing. It starts with the background behind China’s rise as an export powerhouse and provides an overview of the structure of Chinese trade. It then examines the question of whether China is moving up the value chain, and whether any such moves are also being accompanied by an exit from labor-intensive light manufacturing, as was seen in previously dominant export economies. Finally, it examines rebalancing, and in particular the implications for the composition and size of China’s imports. Findings in these areas suggest opportunities and challenges for other countries, including in the Mekong region, and these issues are taken up in subsequent chapters.


China has followed an export-led growth strategy since its “reform and opening up” started in the late 1970s. As part of the decision to increase the role of market mechanisms, one of the first reforms implemented was to open up trade with the outside world. This decision was soon followed by the 1979 law on Sino-Foreign Equity Joint Ventures, officially welcoming the foreign direct investment (FDI) needed to create a manufacturing sector in what was then a heavily agricultural economy.5 In 1980, the first four special economic zones (SEZs) were created—in Zhuhai, Xiamen, Shenzhen, and Shantou—to help kick-start FDI, and then 14 coastal cities and three regions (the Yangtze, Min Jiang, and Pearl River Valleys) were designated “open areas” for foreign investment, with limited red tape and generous tax incentives. Further trade reforms, including significant tariff reductions that reduced the cost of critical imported inputs, continued throughout the 1990s and 2000s.

China also benefited from a series of favorable external developments. First, the East Asian exporters that had dominated global manufacturing in the 1970s and 1980s were faced with sharply rising wages and land prices, reflecting their relatively limited factor endowments. Indeed, the initial surge in manufacturing investment in China occurred as ethnic Chinese entrepreneurs from Hong Kong SAR and Taiwan Province of China sought to take advantage of lower production costs on the mainland (Gereffi 1999). Second, the prevailing exporters’ currencies appreciated significantly after the Plaza Accord in 1985, while the renminbi depreciated. Third, the prevailing manufacturing Asian powers generally faced quotas in the West that constrained their access to key consumer markets; China’s rise as an exporter, by contrast, coincided with, and was reinforced by, a global move toward trade liberalization. While China also faced quotas, it nonetheless managed to increase export earnings by upgrading quality and diversifying into non-quota items, and it benefited from World Trade Organization (WTO) accession in 2001 as well as the 2005 expiration of the Multifiber Agreement.

Figure 1.1.Nominal Effective Exchange Rates

(August 1985 = 100)

Sources: Haver Analytics; and IMF staff estimates.

Figure 1.2.Average Import Tariff


Sources: UN Comtrade; and IMF staff estimates.

China’s initial focus in the 1980s and early 1990s was on the labor-intensive light manufacturing of simple consumer goods. China specialized in goods with short production chains and low unit-value inputs—goods like apparel, footwear, furniture, and toys were dominant, just as they had previously been for Taiwan Province of China, Korea, and Japan (Riad and others 2011). By the early 1990s, labor-intensive light manufacturing accounted for more than 40 percent of China’s exports, consistent with the country’s factor endowments. Even though such industries were characterized by relatively low wages and unskilled work, these early forays into manufacturing had what were then surprising levels of performance in terms of job creation and foreign earnings (Scott 2006).

While market share in such segments continued to grow, China increasingly moved toward the assembly of electronic goods and machinery. At first, this was still labor-intensive manufacturing, but these new sectors involved sophisticated and costly inputs that China could not produce domestically. The country thus began to integrate into global supply chains (see Box 1.1). This so-called “processing trade” initially looked like a “triangle” of production. First, Western firms would export high-tech parts and components to Japan, Korea, and Taiwan Province of China, where skilled workers would convert them into sophisticated, technology-intensive intermediate and capital goods. Second, these would be exported to China, where producers would also import raw materials, accessories, and even packaging materials, all in bond, into SEZs. Third, simple, labor-intensive processing or assembly of these inputs would take place, and finished products would be re-exported, typically to advanced economies (Gereffi 1999). Over time, this simple model became more complex, both because production chains grew longer and more specialized and because the producing economies’ comparative advantages evolved.

Box 1.1.Global Value Chains in International Trade

International trade has become increasingly dominated by global value chains (GVCs). Thanks to improved logistics and technology, production has been broken into component parts, allowing economies to focus on the stage(s) at which they are most competitive. Goods like electronics lend themselves to supply chain trade as they are lightweight, have robust components, and feature an assembly process that can easily be broken into a series of standardized components. This has facilitated the entry of new countries into global manufacturing and trade. Advanced economies generally participate in GVCs in high-value-added activities—both upstream, in research and development (R&D) and the production of sophisticated components, and far downstream, in the branding, marketing, and distribution of final products. In between are lower-value-added activities such as processing and assembly, where emerging and developing economies have increasingly played a role—this pattern is captured in the well-known value-added “smile” schematic. GVCs also promote technology transfer and productivity gains, and over time countries usually move to higher-value-added activities.

Box Figure 1.1.Value-Added “Smile” Chart

Source: IMF staff estimates.

Box Figure 1.2.World Trade to GDP


Sources: Haver Analytics; and IMF staff estimates.

The rise of GVCs has deepened the interdependency of trading nations and had dramatic implications for the global economic landscape. The original “triangle” of production has become substantially more complex. Many countries have moved up the value chain, and Korea and Taiwan Province of China now produce many of the first-stage parts and components that previously came from Europe, Japan, and the United States. Moreover, production has become further fragmented, with unfinished goods typically crossing national borders multiple times before reaching their final destinations. This helps explain the fact that global trade grew faster than GDP over much of the past two decades (Baldwin and Lopez-Gonzales 2013). In addition, because GVCs facilitate the entry into manufacturing, by breaking up complex goods into simple production steps, it also explains the massive growth in trade by developing countries: more than half of developing-country value-added exports involve GVCs, and their share of global trade in components has quadrupled since the mid-1980s. These developments underscore the importance of looking beyond gross trade data and examining the value-added contributed by each country, as is possible in the OECD’s Trade in Value Added dataset.

As processing trade grew increasingly important, China emerged as the primary assembly point of the Asian supply network. By the early 2000s, processing trade rose to 60 percent of China’s total trade (though it has since declined), and even outside the processing trade, manufacturers increasingly relied on imported inputs, such that the share of foreign value added (FVA) in China’s total exports was as high as 50 percent—“Made in China” was only halfway true. Nonetheless, even in value-added terms, China has become the world’s largest exporter. FDI inflows from Japan and ASEAN surged, and bilateral trade balances evolved such that China went into substantial surplus with the United States, Europe, and Japan while it went into deficit with Asian neighbors that were upstream in the Asian supply chain (notably Malaysia, Taiwan Province of China, Korea, Philippines, and Thailand) as well as with commodity producers (Australia, Brazil, Saudi Arabia, Russia, Angola, Iran, and Indonesia). China now accounts for about two-thirds of Asia’s imports of intermediate goods, 25 percent of capital goods exports from Japan and Korea, and nearly 50 percent of the region’s exports of intermediate goods.

Figure 1.3.Manufacturing Domestic Value Added

(Percent of global value added)

Sources: OECD Trade in Value Addded (TiVA) database; and IMF staff estimates.

Figure 1.4.China’s Trade Balance by Type of Trade Partner

(Billions of U.S. dollars)

Sources: UN Comtrade; and IMF staff estimates.

1/ EU-27, Hong Kong SAR, and United States.

2/ Korea, Malaysia, Philippines, Taiwan Province of China, and Thailand.

3/ Angola, Australia, Brazil, Chile, and Saudi Arabia.

Somewhat curiously for a still low-income, labor-abundant country, electronics—and, to a lesser extent, machinery—became China’s principal exports. The bundle of goods exported by China increasingly resembled the bundles exported by richer countries. In particular—and perhaps as a legacy of central planning—China produced and exported a higher share of capital-intensive products, such as machinery and transport equipment, than countries with similar levels of per capita income (Rodrik, Hausmann, and Hwang 2006; Rodrik 2006; Schott 2008). The apparent sophistication of the Chinese export mix led some to predict a surge in GDP growth as income “caught up” to the country’s export dynamism. Though income did in fact grow rapidly, the magnitude of the final goods overstated China’s contribution, which was initially labor-intensive and low in sophistication. Still, participation in supply chains, even at this labor-intensive stage of production, helped China develop a level of manufacturing competence that set the stage for later developments—the first step was simply bringing production inside the borders (Gaulier, Lemoine, and Kesenci 2007; Baldwin and Lopez-Gonzales 2013).

China’s export success and participation in supply chains brought increasing attention to value-added, rather than gross, trade data. As noted by Ravenhill (2014), the economics profession was late in appreciating the economic impact of global value chains, holding its first major symposium on the subject only in 2001. Over the subsequent years, a few key stylized facts emerged from the studies in this area. First, gross data on bilateral trade offered a misleading picture of relative competitiveness (Johnson and Noguera 2012); for example, according to the OECD Trade in Value Added (TiVA) database, as of 2011, the U.S. bilateral trade deficit with China was 35 percent smaller in value-added terms than in gross terms. Second, adjusted for value added, global trading patterns were once again well explained by factor endowments, as per standard Heckscher-Ohlin trade theory. And third, given the increasingly large import component of exports, growth had become less sensitive to export dynamics and to exchange rate movements than would be expected in a non-supply-chain world (IMF 2014, 2015).6,7

Given the nature and extent of China’s trade, the country’s rapid growth has had important spillovers to the rest of the world. China’s imports have fueled growth in Asian neighbors, as well as in commodity exporters worldwide, and there have also been conscious attempts to strengthen economic ties, as with the Free Trade Agreement between ASEAN and China (IMF 2011; IMF 2012). Arora and Vamvakidis (2010) show empirically that the spillover effects of China’s growth have increased significantly in recent decades. While China’s dramatic gains did increase pressure on other Asian economies to seek areas of comparative advantage (Ahearne and others 2006), for the most part, China did not increase its exports at the expense of other Asian economies—what Asian economies lost in market share in the United States and Europe, they more than gained in exports of higher-value-added intermediate goods to China (Kim, Kim, and Lee 2006). Put another way, complementarities thus far have been more important than competition for Asia on the whole. Either way, China’s recent difficulties and its attempts to rebalance to slower, consumption-led growth are being watched closely across the world.

China’s trade structure


As suggested above, the composition of Chinese exports has evolved over time. Four basic categories—clothing and plastic toys, electronics, industrial machinery, and manufactured metal goods—have consistently accounted for 60–70 percent of total exports. During the 1990s, the share of electronics and machinery in the total grew sharply, while the share of garments and footwear has fallen. (As discussed below, however, in absolute terms and as a share of world exports, China remains a major garment and footwear exporter.) There have been sharp changes within electronics as well, with China moving from simple TVs, radios, and white goods, such as refrigerators and washing machines, in the early 1990s to computers in the early 2000s, and mobile phones and valves and tubes in more recent years. One constant, however, is that China has remained primarily an exporter of final goods rather than intermediates; final goods account for roughly 60 percent of total exports, much the same as in the 1990s. This does not, however, imply that China is still involved only in assembly—rather, as discussed below, China in many cases now produces intermediate goods and assembles them into final goods exports.

Figure 1.5.Key Chinese Exports

(Percent of total)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.6.Key Electronics Exports

(Percent of electronics exports)

Sources: UN Comtrade; and IMF staff estimates.

In terms of export partners, advanced economies still dominate but the broad trend is toward an increasing role for emerging markets. The United States and Japan used to receive 45 percent of Chinese exports, but this is now down to about 30 percent—20 percent for the United States and 10 percent for Japan—and another 15 percent of Chinese exports go to Korea, Germany, and the Netherlands. Emerging markets are less important but have increased their share materially since 2008—this is particularly true of Vietnam, Malaysia, Indonesia, Thailand, and Mexico. The rise in emerging market demand for Chinese exports is most pronounced in final goods—phones and computers, in particular—in line with rising incomes in those economies.8

Figure 1.7.Main China Export Partners

(Percent of total)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.8.Major Change in Export Partners

(Percentage point change in share of Chinese exports, 2013-08)

Sources: UN Comtrade; and IMF staff estimates.


Imports are less concentrated than exports and their composition has evolved even more significantly. Five categories—electronics, energy, industrial machinery, raw ore/metal scrap, and vehicles—currently account for about 60 percent of total imports. Machinery dominated China’s import needs as it initially industrialized. As processing trade grew, electronics imports started growing rapidly, though recently, energy and commodity imports have grown even faster, reflecting both the surge in domestic (investment) demand and the rise of global commodity prices as markets adjusted to the scale of China’s appetite for raw materials. Within electronics, China mostly imports relatively sophisticated intermediate goods such as circuits, resistors, and semiconductors. Unlike with exports, imports are dominated by intermediate goods, which account for 70 percent of the total. As with exports, however, this composition has been stable since the 1990s—China continues to produce most of its own final goods and does not yet have substantial imports of sophisticated final goods from abroad.

Figure 1.9.Key Chinese Imports

(Percent of total)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.10.Key Electronics Imports

(Percent of electronics exports)

Sources: UN Comtrade; and IMF staff estimates.

There have also been major changes in the origin of China’s imports over the past two decades. Japan has seen the greatest change, with its share falling from almost 25 percent in the early 1990s to under 10 percent today. By contrast, Korea’s share grew, and by 2013, that country had become the biggest source of imports into China. Commodity exporters, such as Australia (iron ore), South Africa (metals), Iraq (oil), and Switzerland (gold), also grew in importance, especially since 2008. “China” itself emerged as a major location of imports in the early 2000s, but this includes the way goods that are exported to Hong Kong for light processing and then re-imported into China are recorded.

Figure 1.11.Top Chinese Import Partners

(Percent of total, China reported)

Sources: UN Comtrade; and IMF staff estimates.

1/ Imports from “China” imply goods exported to Hong Kong SAR that are lightly processed.

Figure 1.12.Change in Share of China’s Imports

(Percentage points, 2013–08)

Sources: UN Comtrade; and IMF staff estimates.

Moving up the value chain

This section presents a number of pieces of evidence that together demonstrate that China has moved into higher-value-added activities over time.

China’s domestic value-added (DVA) ratio has risen over time and now exceeds those of Korea and Taiwan Province of China. Any investigation of whether a country is moving up the value chain must surely start with its DVA ratio, or the share of the value of its final exports that is produced domestically. This can be obtained from the TiVA dataset, but not for every year, and with a substantial lag—the latest observation is for 2011.9 China’s DVA ratio has risen over time, though the rate of increase has flattened out after 2008. Perhaps surprisingly, the ratio now exceeds those of Korea and Taiwan Province of China, though China still lags Japan, the United States, and Germany.

Figure 1.13.Domestic Value-Added Ratio

(Percent of gross exports)

Sources: OECD TiVA; and IMF staff estimates.

Figure 1.14.Domestic Value-Added Ratio, 2011

(Percent of gross exports, by contribution)

Sources: OECD TiVA; and IMF staff estimates.

The “aggregate” DVA ratio can mask significant developments within each sector and should be complemented with other evidence. Figure 1.13 reveals that, in each year, “knowledge-intensive” production has had lower domestic value-added content than “capital-intensive” and “labor-intensive” production10—this is as expected, because sectors like electronics and machinery typically rely on complex imported inputs, often from more advanced countries, while sectors like garments, toys, and furniture either have inputs that have a less important share of the final output value or are relatively simple (for example, a lightly processed raw material) and can be produced locally. And this is why the aggregate DVA ratio can be misleading—China’s shift “across sectors”—for example, from simple, but high-value-added activities like light manufacturing to more complex, but nonetheless lower-value-added electronics assembly—may be masking moves “within sectors”—for example, from simple T-shirts to complex jackets made of high-performing materials.

In absolute terms, domestic value added has grown sharply, and especially so in “knowledge intensive” production. An increase in value added is, by definition, GDP growth, and the sharp increase in total value added shown in Figure 1.15 was a key driver behind China’s rapid GDP growth during the period. The figure also shows that growth was especially pronounced in “knowledge-intensive” sectors, such as electronics, chemicals, machinery and equipment, electrical and optical equipment, and transport equipment. Indeed, these sectors now account for close to two-thirds of China’s total value added. This more rapid growth in knowledge intensive value added is prima facie evidence that China is “moving up the value chain,” even if the aggregate DVA ratio is relatively flat.

Figure 1.15.Chinese Manufacturing Domestic Value Added

(Billions of U.S. dollars)

Sources: OECD TiVA; and IMF staff estimates.

It is also striking that since 2010, processing trade11 has sharply declined in importance. As noted above, processing trade accounted for more than half of total Chinese exports in the early 2000s, but this segment of trade has stagnated over the past five years while other exports have continued to grow. The share of processing trade has now fallen to about one-third of the total. Since processing exports, by definition, are more reliant on foreign value-added than other exports are, the relative decline of this sector strongly suggests that China is moving up the value chain by replacing its reliance on sophisticated imported inputs with domestic production.

Figure 1.16.Processing Exports

(Percent of total exports)

Sources: CEIC Data Company Ltd.; and IMF staff calculations.

Consistent with such developments, import intensity in key export sectors is falling. Without highly technical knowledge of both manufacturing processes and industrial classification codes, it is not easy to identify, from a country’s gross trade data, which imports relate to inputs used for the production of particular exports. Nonetheless, the granular classification of traded products provided by the Standard International Trade Classification, referred to as SITC, allows this matching exercise for some of China’s key export sectors (computers and TVs, radios, phones, and capital goods more broadly) because it clearly specifies the codes for both the final good and the intermediate parts that relate to it. As shown in Figures 1.17 and 1.18, such ratios suggest that the value of imported parts required to produce these exports has declined over time. Barring secular declines in the relative prices of such parts—which we have no reason to believe have occurred—this evidence suggests that China has started moving upstream, producing some of the inputs it needs onshore, and thus moving up the value chain. In fact, these data likely understate the degree of onshoring, given that some of the imported inputs are used for final use domestically in China. This process appears to be continuing for computers, but for the other sectors, import intensity started to flatten out around the time of the global financial crisis.

Figure 1.17.Import Intensity of Select Sectors

(Ratio of imported parts to exports of final goods)

Sources: UN Comtrade; and IMF staff estimates.

1/ In SITC codes, imports of 759 divided by the sum of exports of 751 & 752.

2/ In SITC codes, imports of 7649 divided by the sum of exports of other categories within 76.

Figure 1.18.Import Intensity of Capital Goods

(Capital goods parts imports/capital goods exports)

Sources: UN Comtrade; and IMF staff estimates.

Note: in BEC codes, imports of 42 divided by exports of 41.

China also increasingly appears to be producing sophisticated, upstream parts and components onshore. As China integrated into global supply chains, its imports of sophisticated inputs from Korea and Japan grew sharply. The top five products (at the three-digit level of industrial classification), including such items as electrical circuits, LCD screens, and valves/tubes, are shown in Figure 1.19. These goods are all intermediate goods (specifically “parts and accessories,” under the Broad Economic Classification [BEC] used in UN Comtrade data), and they are also classified as “difficult-to-imitate research-intensive goods,” using the definition in Hufbauer and Chilas 1974 and Yilmaz 2002. Over the course of the past decade, China has been gaining export market share in the sophisticated products that it earlier exclusively imported. The exact pattern depends on the good in question, but with LCDs, for example, China moved upstream by onshoring more sophisticated production that previously took place in Japan, then Taiwan Province of China, and more recently Korea, as shown in Figure 1.20.12 Another example—a certain type of transistor—is shown in Figure 1.21. In short, China has moved up the value chain.13

Figure 1.19.Top Five Electronics Exports to China

(Share of total exports to China, 2014)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.20.LCD Screens

(Export market share)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.21.Transistors (<1 watt)

(Export market share)

Sources: UN Comtrade; and IMF staff calculations.

Finally, China’s move up the value chain is reflected in the evolution of bilateral trade balances. Advanced North Asia for many years ran a large collective trade surplus with China. While China exported low-technology goods to Japan, Korea, and Taiwan Province of China, these flows were outweighed by China’s import of relatively high-technology goods from these countries—many of these were parts intended for assembly and then export to the rest of the world. The trade surplus decreased sharply after the global financial crisis, and most of the decline has occurred in relatively high-technology sectors. This may suggest that China is now producing these parts onshore. At the same time, China continues to export low-tech goods—as discussed below, it is not exiting these sectors in any dramatic manner.

Figure 1.22.Advanced North Asia’s Trade Balance with China

(Percent of GDP)

Sources: UN Comtrade; and IMF staff calculations.

All of these pieces of evidence combine to paint a clear picture of China’s moving up the value chain in several different ways. As China has gained more manufacturing experience and as its labor productivity has risen, it has gradually moved into the production of more sophisticated goods. The processing trade has become less important, and even within regular trade, China has become less reliant on imported inputs in producing these goods. And reflecting these changes, China’s value-added ratio has risen and its bilateral trade deficits with Asian neighbors have decreased materially. These developments suggest that China is increasingly competing with advanced economies like Japan, Korea, and Taiwan Province of China.

Existing labor-intensive production

The rapid growth of China’s export market share naturally raises questions about whether it can be sustained. Even as China is gaining share in an increasing number of sophisticated areas, as highlighted above, there are growing questions about its competitiveness in the lower-end, labor-intensive sectors where it began. One issue is that the excess-labor dividend may be in its twilight. The working-age population is expected to start shrinking this year before contracting deeply over the coming decades.14 Such demographic issues along with other factors such as falling demand for factory jobs by Chinese youth have put sharp upward pressure on China’s wages, which have risen considerably faster than productivity, and faster than wages in competing countries (albeit from a very low level).

As China has moved up the value chain, the relative importance of labor-intensive production has declined. Figure 1.23 shows that the share of labor-intensive goods in China’s total gross exports started declining as early as the early 1990s. At the same time, both the “difficult to imitate” and “easy to imitate” varieties of research-intensive production have increased in relative importance. Since the global financial crisis, there has been some reversal in these trends but it remains too soon to determine whether this trend persists as global output returns to potential.

Figure 1.23.China Export Breakdown by Factor Intensity

(Percent of Chinese exports)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.24.China World Market Share by Factor Intensity


Sources: UN Comtrade; and IMF staff estimates.

Somewhat surprisingly, China still appears to be highly competitive in labor-intensive goods overall. As shown in Figure 1.24, China’s world market share in labor-intensive production is higher than its share in any other sector and has been increasing steadily over the years, with only recent evidence of some plateauing. Figure 1.25 complements these messages by showing that China is by far the most important global exporter of labor-intensive goods, and again, its share is just beginning to flatten out.

Figure 1.25.Global Export Market Share in Labor Intensive Goods

(Percent of world exports in labor-intensive goods)

Sources: UN Comtrade; and IMF staff calculations.

But while there is not yet evidence of a marked shift out of labor-intensive production in aggregate, the beginnings of decline are seen in some key sectors. As shown in Figure 1.26, light manufacturing sectors all appear to be at an inflection point—global market shares seem to have plateaued in many of these categories and are even dropping for furniture. There is also some plateauing in final electronic goods, but the evidence of such a transition is far less pronounced: there is some flattening in simple white goods and computers but continued sharp growth in the market share of telephones.15

Figure 1.26.Export Market Share: Simple Consumer Goods


Sources: UN Comtrade; and IMF staff estimates.

Figure 1.27.Export Market Share: Consumer Electronics


Sources: UN Comtrade; and IMF staff estimates.

The durability of China’s success in labor-intensive production is striking, but also difficult to explain. As shown in Introduction, earlier exporters’ success in sectors such as garments was relatively short-lived; declines in market share set in after less than a decade, and occurred quite sharply, with countries like Korea almost completely exiting garments within a space of 15 years. Such a transition may yet be seen in China—firms in other countries may out-compete Chinese producers, or Chinese producers may invest abroad (outward FDI) and move their manufacturing to foreign subsidiaries. These processes are already happening to some degree, particularly in certain sectors, but overall it is remarkable that China has remained competitive in labor-intensive production for so long.

China’s inland provinces may be a possible reason for the sustained performance of the Chinese labor-intensive sectors, but this is not obvious in the data. Part of the explanation of China’s continuing strength in labor-intensive exports stems from the size of China’s labor force and the ability to draw cheap labor to coastal production areas. In addition, some argue that production moves inland as wages on the coast lose competitiveness (the popular example being Foxconn). This story, however, is not clear in the data. While the inland production of basic manufactures has risen and the share of FDI going inland has reached 50 percent, the share of total exports produced by inland provinces has not risen; production for export has remained mostly on the coast, where wages are still the highest (Figures 1.2831). Moreover—and perhaps somewhat counterintuitively—inland provinces have seen greater increases in the production of high-tech goods than in light manufacturing. Inland production may indirectly have helped to sustain coastal exports, but it may also be that the extreme efficiencies, network effects, and other factors associated with exporting from China’s coastal provinces have caused the “geese” to stop flying.

Figure 1.28.Foreign Exports by Location of Producer

(Trillions of U.S. dollars)

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

Figure 1.29.Industrial Production of Inland Provinces

(Percent of national total)

Sources: Chinese Statistical Yearbooks; and IMF staff estimates.

Figure 1.30.FDI by Region

(Percent of total)

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

Figure 1.31.Inland Wages as Share of Coastal Wages


Sources: Chinese Statistical Yearbooks; and IMF staff estimates.

In sum, after two decades of sharply rising wages, there is initial evidence of some plateauing of China’s dominance in labor-intensive goods. But this evidence is not yet a clear trend and is occurring at levels of market share far above where previous manufacturing powers lost competitiveness. The thrust of this information is suggestive of advantages to Chinese manufacturing that go beyond purely labor-related competitiveness, such as network effects or economies of scale. Building on the findings in the previous section, if firms in China are moving the production of imported inputs onshore, they may find that they can keep assembly and control the entire production process. This is a reversal of the move captured in Hummels, Rapoport, and Yi (1998) toward “vertical specialization”—importing inputs and exporting outputs—toward the “horizontal specialization” in which countries trade in goods produced largely in one country.


China’s export-led model and its resulting current account surpluses generated substantial global attention. In addition to a number of underlying advantages such as abundant labor, the Chinese authorities made several explicit policy choices including heavy intervention, capital controls, financial repression, and low returns to labor. From a domestic perspective it was a very successful strategy that led to dramatic increases in per capita income and total factor productivity without obvious evidence of overheating (Blanchard and Giavazzi 2005). In the initial years, growth was highly imbalanced across the provinces and skill levels, but this improved over time as infrastructure reached deeper into the Chinese interior. The model was nonetheless subject to two critiques—first, domestically, there was inadequate development of health and other services, excessive precautionary savings as a result of weak safety nets, and a misallocation of investment; and second, other countries saw themselves facing a sharp—and, in their view, unfair—loss of competitiveness, particularly in labor-intensive industries.

China has made important progress in reducing external imbalances since the global financial crisis. The weakening of external demand during the crisis was offset in large part by government stimulus that drove the share of investment in GDP from 40 to almost 47 percent, thereby preventing a sharp fall in growth. Without a corresponding increase in savings, this additional investment reduced the current account surplus from 10 percent of GDP in 2008 to just 2 percent today. Moreover, the growth contribution from net exports fell from 0.5 percentage points in 2001–07 to –0.8 percentage points in 2008–14. In this context, reserve accumulation, a key proxy for the broader imbalance, ceased, and foreign exchange sales began in 2014–15.

Figure 1.32.Investment and Saving Balances

(Percent of GDP)

Sources: Haver Analytics; and IMF staff estimates.

Figure 1.33.Private Consumption and Investment

(Industrial countries and emerging markets; average, 2008–14)

Sources: IMF, World Economic Outlook; and IMF staff estimates.

The progress in external rebalancing, however, has come at the cost of a large increase in domestic imbalances. The source of excessively large current account surpluses before the crisis was not overly low investment, but rather excessive savings. Thus, responding with an investment-led stimulus resulted in the current excesses in capacity, particularly in the manufacturing and real estate sectors. Moreover, the easy credit conditions that helped finance the investment boom have given rise to financial sector vulnerabilities.

In more recent years, China has had some initial success in addressing these imbalances and moving toward a more sustainable growth model. The authorities are attempting to move toward consumption- and services-led growth and away from investment and manufacturing. Services have been on a steady upward march since 2011 and overtook manufacturing as the largest share of GDP in 2012. Meanwhile, consumption, which fell from 62 percent of GDP in the late 1990s to 49 percent at the start of the global financial crisis, has plateaued and started to increase marginally. Nonetheless, as shown in Figure 1.33, China remains a major outlier on both consumption and investment. Progress could speed up if the authorities advance on key structural reforms outlined during the Third Plenum in 2013.

Figure 1.34.Rebalancing—Supply Side

(Percent of GDP)

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

Figure 1.35.Rebalancing—Demand Side

(Percent of GDP)

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

Rebalancing will likely have important implications for China’s trading partners. Input-output tables suggest that consumption is currently only half as import-intensive as investment (Figure 1.36), and rebalancing toward consumption could thus imply lower overall imports, at least during a transition period; over the longer run, however, one would expect the import intensity of consumption to rise as China’s relative prices adjust. In addition, there will be a material shift in the composition of China’s trade partners as a result of rebalancing, with significant effects on those countries that export investment-related goods, but not consumption goods, to China. And all of this is on top of the trend, described above, toward onshoring the production of upstream inputs, which will also directly lower imports.

Figure 1.36.Import Intensity in GDP Components

(Percent of GDP, 2011)

Figure 1.37.China’s Primary Goods Imports

(Percent of total China imports)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.38.Real Commodity Imports Growth

(Year-on-year percent change of 12-month rolling sum)

Sources: CEIC Data Company Ltd.; and IMF staff estimates.

Rebalancing has already had an important impact on investment-related goods, including commodities. Since 2005, real import demand for commodities has outpaced that for imports more broadly, and China now accounts for about 15 percent of global commodity imports. Petroleum, iron ore, and copper are particularly important to China, with other metals and coal also significant. In the most recent data, while petroleum demand has remained robust, real demand for iron ore has weakened with the slowdown in infrastructure and real estate investment, as has demand for copper (though this is relative to 2012–13). The clearest weakness is in coal, which is down by almost 50 percent. Potentially even more important than the quantity effects are the impacts on prices, as the elasticity of commodity prices to Chinese commodity demand has historically been far above unity. China has also contributed to weakening global commodity markets by increasing supply. Finally, quantities and prices of imported machinery and other investment inputs have also declined.16

Figure 1.39.China—Metals Consumption

(Percent of world total)

Sources: World Bureau of Metal Statistics; and IMF staff calculations.

Figure 1.40.China—Oil Consumption

(Percent of world total)

Sources: BP Energy Statistics; and IMF staff calculations.

Meanwhile, the slow increase in China’s consumption so far has yielded only a modest increase in consumption imports. Indeed, an increase in imports may not materialize in a substantial way if Chinese consumers’ needs are met from domestic production.17 Up to now China’s share of world consumption imports has increased, particularly since 2008, but that share is still very small, in part because of weakening consumption growth rates in the recent period. Moreover, consumption goods make up a far smaller share of China’s imports than is the case in other countries. China’s main goods imports have been passenger cars, and in terms of services, outward tourism has increased sharply.

Figure 1.41.Consumption Imports

(Percent of respective total imports)

Sources: UN Comtrade; and IMF staff estimates.

Figure 1.42.China Consumption Imports

Sources: UN Comtrade; CEIC Data Company Ltd.; and IMF staff estimates.

Going forward, the expectation is that Chinese external and internal rebalancing will continue. The process may not be monotonic, as progress on domestic rebalancing may temporarily reverse external rebalancing on account of the less import-intensive nature of consumption. But over time, the aggregate impact of reforms—including greater exchange rate flexibility, financial sector liberalization, reduced capital controls, and a smaller role for the state—should reduce the current account surplus and increase the sustainability of domestic demand. Trading partners will need to adjust appropriately as changes in the contours of Chinese demand are likely to be long-lasting.


China’s trading patterns have already started to change. There is a clear move into higher-value-added activities, and this has been happening for a number of years. Somewhat surprisingly, there has been no obvious exit from labor-intensive production in aggregate, but it appears to be starting to plateau, with slight declines evident in some key sectors like garments, footwear, toys, and furniture; in other words, China may be at an inflection point with respect to labor-intensive goods. However, while inland production has picked up, this appears mostly to be to service domestic demand in China; exports are still produced almost exclusively on the coast. Finally, there is mixed evidence on rebalancing—imports of machinery and certain commodities, like coal and iron ore, are clearly declining, but others, like oil, continue to grow; at the same time, while Chinese consumption is increasing, this has so far translated into just a modest increase in imports of consumption goods (though Chinese outbound tourism has picked up more sharply).

Annex 1.1. Taxonomy of Trade Classifications

1. Technology Intensity (ISIC Rev. 3) – Classification of manufacturing industries into categories based on R&D intensities. See OECD.

High-technology industries: Aircraft and spacecraft; pharmaceuticals; office, accounting, and computing machinery; radio, TV, and communications equipment; medical, precision, and optical instruments

Medium-high-technology industries: Electrical machinery and apparatus, not elsewhere classified (n.e.c.); motor vehicles, trailers, and semi-trailers; chemicals excluding pharmaceuticals; railroad equipment and transport equipment, n.e.c.; machinery and equipment, n.e.c.

Medium-low-technology industries: Building and repairing of ships and boats; rubber and plastics products; coke, refined petroleum products, and nuclear fuel; other non-metallic mineral products; basic metals and fabricated metal products

Low-technology industries: Manufacturing, n.e.c.; recycling; wood, pulp, paper, paper products, printing, and publishing; food products, beverages, and tobacco; textiles, textile products, leather, and footwear.

2. National Accounts and Broad Economic Categories

Final Goods

Consumer Goods

112 – Food and beverages, primary, mainly for household consumption

122 – Food and beverages, processed, mainly for household consumption

522 – Transport equipment, non-industrial

51 – Passenger motor cars

61 – Consumer goods not elsewhere specified, durable

62 – Consumer goods not elsewhere specified, semi-durable

63 – Consumer goods not elsewhere specified, non-durable

Capital Goods

41 – Capital goods (except transport equipment)

521 – Transport equipment, industrial

Intermediate Goods

111 – Food and beverages, primary, mainly for industry

121 – Food and beverages, processed, mainly for industry

21 – Industrial supplies not elsewhere specified (n.e.s.), primary

22 – Industrial supplies, n.e.s., processed

31 – Fuels and lubricants, primary

321 – Fuels and lubricants, processed (motor spirit)

322 – Fuels and lubricants, processed (other than motor spirit)

42 – Parts and accessories of capital goods, (except transport)

53 – Parts and accessories of transport equipment

3. Trade by Factor Intensity

Raw-Material-Intensive Goods

SITC 0 Food and live animals

SITC 2 Crude material, inedible, except fuels (excluding 26)

SITC 3 Mineral fuels, lubricants, and related materials (excluding 35)

SITC 4 Animal and vegetable oils, fats and waxes

SITC 56 Fertilizers

Labor-Intensive Goods

SITC 26 Textile fibers

SITC 6 Manufactured goods classified chiefly by material (excluding 62, 67, 68)

SITC 8 Miscellaneous manufactured articles (excluding 88, 87)

Capital-Intensive Goods

SITC 1 Beverages and tobacco

SITC 35 Electric current

SITC 53 Dyeing, tanning, and coloring materials

SITC 55 Essential oils and resinoids and perfume materials; cleansing preparations

SITC 62 Rubber manufactures, n.e.s.

SITC 67 Iron and steel

SITC 68 Non-ferrous metals

SITC 78 Road vehicles

Easy-to-Imitate Research-Intensive Goods

SITC 51 Organic chemicals

SITC 52 Inorganic chemicals

SITC 54 Medicinal and pharmaceutical products

SITC 58 Plastics in non-primary forms

SITC 59 Chemical materials and products, n.e.s.

SITC 75 Office machines and automatic data-processing machines

SITC 76 Telecommunications and sound apparatus and equipment

Difficult-to-Imitate Research-Intensive Goods

SITC 57 Plastics in primary forms

SITC 7 Machinery and transport equipment (includes semiconductors/excludes 75, 76, 78)

SITC 87 Professional, scientific, and controlling instruments and apparatus, n.e.s.

Source: Yilmaz (2002) based on earlier work by Hufbauer and Chilas (1974).

4. OECD Factor Intensity Breakdown


Food, beverages, tobacco

Wood, paper, and publishing

Basic metals and fabricated metals

Labor Intensive

Textiles, textile products, leather, and footwear

Manufacturing n.e.c., recycling

Knowledge Intensive

Chemicals and non-metallic mineral products

Machinery and equipment

Electrical and optical equipment

Transport equipment

    Other Resources Citing This Publication