The Global Economic Crisis and Trade and Growth Prospects in East Asia

The global economic crisis has revealed that growth in East Asia is highly vulnerable to slowdown in exports to major advanced economies. This paper shows that in precrisis years, at least one third of growth in the People’s Republic of China was due to exports, and this proportion was even higher in most other East Asian economies. In the coming years the region may not be able to go back to rapid export-led growth because of sluggish growth and high unemployment in advanced economies, and the need to reduce trade imbalances in order to secure global stability and avoid frictions in the trading system. It will be necessary not only to rebalance domestic and external components of demand, but also to embark on industrial restructuring necessitated by a shift from export-led growth to growth led by domestic demand and intraregional trade.


I. Introduction
After several years of impressive growth, the world economy encountered an equally impressive downturn starting in the third quarter of 2008, triggered by financial fragility and imbalances generated by speculative lending and investment and debt-driven spending in major advanced economies (AEs), notably the United States (US). Initially, there was widespread optimism that growth in the developing and emerging economies (DEEs) of East Asia 1 would be decoupled from the difficulties that pervaded AEs, and the region would continue to surge ahead as an autonomous growth pole. Sound balance-of-payments positions and self-insurance provided by large international reserves accumulated from current account surpluses and/or private capital inflows were expected to protect them against the kind of financial shocks that had devastated the region during 1997-1998. In the event, however, the region could not avoid a significant drop in growth, in large part because of a sharp contraction in exports. Growth fell even in countries such as the People's Republic of China (PRC) despite massive countercyclical fiscal packages and aggressive monetary easing in response to fallouts from the crisis, while many others incurred output losses for the first time since the 1997 Asian financial crisis.
Like the earlier episodes of instability, this global crisis too has revealed certain structural weaknesses and vulnerabilities among various DEEs in Asia. As a result of the growth strategies pursued, economic activity has come to depend heavily on exports to major AEs, and hence became highly vulnerable to their interruption. There is now an increased agreement that the dependence of growth in the region on foreign markets needs to be reduced, particularly since the world economy is unlikely to go back to the conditions prevailing before the outbreak of the crisis, characterized by rapid expansion of exports to AEs and plenty of cheap money and footloose capital. This paper examines, in the light of the recent experience, the policy challenges faced and options available in sustaining rapid growth in East Asia over the medium term. Section II discusses the extent of dependence of growth in East Asia on exports, notably to the US and the European Union (EU). This is followed by a discussion of medium-term prospects of the latter economies and their implications for growth in East Asia. The final section will examine policy responses needed in the region to a sustained slowdown in exports to AEs, both in terms of rebalancing domestic and external sources of growth and industrial restructuring necessitated by a shift from export-led growth to growth led by domestic demand and intraregional trade.

II. Export Dependence, Trade Shocks, and Growth in Asia
Trade has been by far the most important channel of transmission of deflationary impulses to DEEs from the global crisis. 2 After growing by close to 10% per annum during the years before the crisis, world trade volume started to fall sharply in the last quarter of 2008 and throughout the first half of 2009. Despite the subsequent recovery, it registered a decline of close to 13% for the year as a whole (WTO 2010).
In most East Asian DEEs in the years preceding the crisis, exports were the most dynamic component of aggregate demand, growing faster than domestic investment and consumption; and even at double-digit rates in the PRC, the Republic of Korea, and Singapore in real terms. With the downturn of the global economy, this rapid growth was followed by a sharp contraction in the last quarter of 2008, with exports falling by doubledigit rates in most countries until the last quarter of 2009.
The impact of export contraction on economic activity has varied according to the importance of exports in the income generation process in comparison with the domestic components of aggregate demand. To account for the contribution of different components of demand to income and growth, one needs to go beyond the conventional growth accounting based on ex post national income identity. In this identity, the trade balance or net exports (that is, exports minus imports) describe the ex post contribution of trade to income, but does not provide a correct measure of dependence of income on exports because all imports are deducted from exports, and imports used for domestic consumption and investment are not accounted for. It thus underestimates the contribution of exports and overestimates the contribution of domestic demand to gross domestic product (GDP).
On the other hand, the conventional exports/GDP ratio overestimates the income (value added) generated by exports because it ignores the foreign (import) content of exports. Since exports use, directly or indirectly, imported intermediate goods, parts and components, any contraction in exports would bring, pari passu, a contraction in imports used for the production of exportables, thereby tempering the impact of export declines on income and growth. Thus, in order to correctly assess the impact of export shocks on income and growth, it is necessary to identify direct and indirect import contents of exports, using input-output linkages.
Furthermore, a contraction in exports reduces income not only directly but also indirectly, through its effects on domestic consumption and investment. The Keynesian multiplier establishes an indirect link between exports and income through consumption. There are also strong knock-on effects of exports on investment, particularly where an important part of manufacturing is export-oriented.
Estimates for import contents of different components of aggregate demand and spillovers from exports to domestic demand are not always readily available for DEEs. However, it is known that a common feature of East Asian DEEs closely linked to international production networks is that not only their export/GDP ratios but also the import content of their exports is high.
Evidence suggests that import content of exports from the PRC is between 40% and 50%; that is, domestic value added generated by exports is less than 60% of their gross value. In value added terms the share of exports in GDP is no more than 20%. A very large proportion of foreign content of exports consists of imported parts and components directly used in sectors producing exportables. Almost two thirds of domestic value added contained in exports are generated in industries supplying inputs to sectors producing exportables. Thus, a contraction in exports would cut value added and income not so much in export sectors as in those supplying inputs to them. The import content of consumption in the PRC is very low compared to its main trading partners, barely reaching 10%. Around 60% of Chinese imports are used, directly and indirectly, for exports; less than 15% for consumption; and some 20%-25% for investment.
Despite high import content, one third of growth of income in the PRC in the years before the outbreak of the global crisis is estimated to have been due to exports because of their phenomenal growth of some 25% per annum in real terms. This figure goes up to 40% if spillovers to domestic consumption are accounted for, and further to 50% if knock-on effects on domestic investment are also added. The sharp contraction of exports in 2009 is estimated to have dragged down GDP by more than 3 percentage points, without allowing for spillovers to domestic demand. 3 This meant a sharp swing of more than 6 percentage points in the contribution of exports to GDP compared to precrisis years. Despite massive government intervention, this is only partly offset by faster growth of domestic demand so that GDP growth in 2009 remained some 2 percentage points below the 2003-2007 average.
Available evidence suggests that the import content of exports is lower in Indonesia, the Republic of Korea, the Philippines, and Thailand than in the PRC. With the exception of Indonesia, these countries also have higher export/GDP ratios. Consequently, in valueadded terms the share of exports in GDP is higher in these countries than that in the PRC, reaching 30%-40%. This means that they are more susceptible to export shocks. Accordingly, it is estimated that in these countries the impact of export contraction during 2008-2009 on GDP varied between 4 and 6 percentage points, without accounting for spillovers to domestic demand. Malaysian and Singaporean exports have higher import content than the PRC exports, but these countries also have much higher export/GDP ratios. Consequently in value added terms their exports account for a higher share of GDP not only than the PRC but also the first group of countries. Accordingly, export shocks in 2008-2009 had a stronger impact on GDP growth in Malaysia and Singapore, possibly reaching double-digit figures.
There appears to have been considerable variations in the impact of trade and financial shocks on private domestic demand, notably investment. The impact has been particularly strong in, respectively, the Republic of Korea; Taipei,China; Malaysia; Thailand; and the Philippines where private investment, notably in durable equipment, slowed down or contracted significantly. This, together with a slowdown in foreign direct investment (FDI), is the main reason for sharp drops in the share of domestic investment in GDP during 2008-2009 even though in most of these countries government investment outlay rose as part of the countercyclical policy response to the crisis (ADB 2010).
All in all, it can be estimated that, on average, contraction of exports during 2008-2009 reduced GDP in East Asian DEEs by 5-6 percentage points. When spillovers to domestic demand are accounted for, this figure is likely to be much higher. Indeed, according to an estimate by the United Nations Economic and Social Commission for Asia and the Pacific (2010) for East Asian DEEs and Japan, the impact of the 2009 shortfall in exports on GDP reached 7.8 percentage points, accounting for both direct and indirect effects.
The more successful exporters of manufactures with high export/GDP ratios have been hit particularly hard by trade shocks even where the import content of exports is relatively high. These include not only countries that enjoyed large current account surpluses such as the PRC, Malaysia, Singapore, and Thailand, but also the Republic of Korea where the current account was broadly in balance in the run-up to the crisis but which had an important part of its manufacturing industry export-oriented. By contrast, countries that are not so closely integrated into the global trading system and international production networks in manufacturing, such as Indonesia and the Philippines, suffered relatively less from contraction in exports.

III. Prospects for Growth in Advanced Economies and East Asian Exports
A return of the global economy to the kind of rapid and broad-based expansion enjoyed during the early years of the decade until 2008 is no doubt desired by all, but the main question is how to achieve this without the accompanying financial fragilities and trade imbalances that led to the most severe postwar global economic crisis (Akyüz 2010a). There is a wide agreement that a return to "business as usual" with the US continuing to consume beyond its means and absorbing the PRC's exports by issuing growing amounts of dollar liabilities is not a sustainable option-it is a recipe for heightened international monetary and financial instability, and disorderly and deflationary adjustment to global economic imbalances.
The prospects for global stability and growth are thus believed to depend crucially on rebalancing the US and the PRC-the largest deficit and surplus countries, respectively. In view of the central place occupied by the dollar in the international reserves system, it is recognized that international monetary and financial stability crucially depends on spending discipline by the US in line with its income, allowing for a fundamental and sustained balance-of-payments adjustment. However, in order to maintain growth, the US should not simply reduce the pace of domestic absorption but also shift to exportled growth. An orderly US adjustment would also require, inter alia, a shift by the PRC from export-led to consumption-led growth, and the realignment of the exchange rate of the renminbi against the dollar. In this way, prospects for global stability are expected to improve without sacrificing growth. 4 A significant adjustment by US consumers is already under way, brought about by massive losses of personal wealth caused by the subprime crisis. Personal savings have moved into positive territory and, on recent trends, may reach 10% of disposable income in coming years. 5 By contrast, stimulus and bail-out packages have resulted in a significant increase in fiscal deficits, which now exceed 10% of GDP, pushing the public debt ratio toward 100%. Clearly, with consumer spending staying behind income growth, attempts to reduce public deficits and debt through tax increases and spending cuts would be highly deflationary. This explains why the US administration is reluctant to undertake fiscal retrenchment before recovery is strongly in place and growth is restored.
Given consumer retrenchment, a growth-driven US fiscal adjustment would require strong export growth. Indeed, a shift from consumption-led to export-led growth is the main objective of the National Export Initiative launched by President Obama in his State of the Union address, targeting to double exports in 5 years. However, even with rapid export expansion, US growth is likely to be sluggish due to a decline in potential growth brought about by the crisis. 6 Actual growth may fall even further if bond markets force a swift fiscal adjustment. In other words, the US may have to pay the cost of decades-long bubble-bust cycles and aggressive monetary and fiscal policy easing in response to consequent crises, by going through slow and unstable growth for some years to come. But, whether or not growth slows down over the medium term, net demand stimulus from the US to East Asian DEEs is likely to be significantly lower than in the years preceding the financial crisis. This means that domestic demand in surplus countries, including the PRC, needs to expand much faster than in the past in order to maintain a relatively rapid global growth.
Germany and Japan are also among the major surplus countries that need to adjust and add to global demand. Although these countries often escape attention because their bilateral trade surpluses with the US are much smaller than that of the PRC, they have also been running large amounts of current account surpluses-$250 billion in Germany and $210 billion in Japan before the onset of the crisis, compared to $370 billion in the PRC. They have both been siphoning off global demand without adding much to global growth. During 2002-2007, exports grew 25 times faster than domestic demand in Germany and 8.5 times in Japan, while this figure was less than three times for PRC. In both Germany and Japan, the contribution of exports to growth was much higher than that in the PRC during the years preceding the crisis. This lack of dynamism in domestic demand has been due to the falling share of consumption in GDP along with stagnant or falling real wages, slow employment growth, and a downward trend in the share of wages in GDP (Akyüz 2010a).
In Germany, reliance on exports for growth through wage restraints (the so-called "competitive disinflation") and at the expense of consumer demand has been compromising growth and stability in other eurozone countries that are unable to restrain wages to the same extent, but locked into a common currency. It has played a major role in growing external imbalances and debt accumulation in several eurozone countries including Greece, Portugal, and Spain where current account deficits as a percentage of GDP have been hovering around double-digit figures, and debt ratios have been rising rapidly. The region is highly vulnerable to a financial turmoil due to the combination of large and what looks like unsustainable sovereign debt in these countries, and a high degree of exposure of European banks to sovereign default.
Avoiding such an outcome would call for strong growth, but this has been severely constrained by continued beggar-my-neighbor policies of Germany. The deflationary policy bias would be aggravated by premature fiscal consolidation that the countries in the region are determined to pursue under the pressure of bond and currency markets, posing the risk of a double-dip. According to the IMF (2010a and 2010b) potential growth in the euro area, currently at zero, will only reach 1.5% over the medium term and actual growth would remain below the rates attained during precrisis years. With increased concerns over sovereign insolvency and premature fiscal retrenchment, the figure may even be lower, if positive at all. Under these conditions, even if the spectre of a doubledip recession could be averted, Europe is unlikely to provide a rapidly expanding market for East Asian exporters for several years to come.
Japanese growth prospects look as bleak as Europe, expected to be the worst among the G-7 by the Organisation for Economic Co-operation and Development (Fujioka 2010). Sluggish markets in the EU and the US would cut down Japan's growth not only directly but also indirectly since Japan provides over 15% of PRC's total intermediate imports used in production for exports. Indeed, a very large proportion of negative export shocks to the PRC during 2008-2009 were passed on to Japan alongside the Republic of Korea and Taipei,China, being the main suppliers of parts and components to the PRC. Therefore, Japan is unlikely to provide much independent growth stimulus to lesser developed countries in the region. With consumer retrenchment and external adjustment in the US and sluggish growth or stagnation in the EU and Japan, it will be very difficult for East Asian DEEs to return to business as usual and continue to enjoy rapid growth based on expansion of exports to AEs.

A. The PRC: Shift to Consumption-Led Growth
For countries closely linked to the East Asian production network, policies and performance of the PRC, as well as major AEs, hold the key for medium-term growth prospects, given the sheer size of the PRC and their close trade linkages. It is now broadly agreed that the PRC cannot go back to precrisis export growth rates in excess of 25% per annum, more than three times the projected medium term growth in world trade volume (IMF 2010b). An aggressive export push in the markets of AEs or DEEs is likely to meet strong resistance, creating conflicts in the trading system. If, on the other hand, the rate of expansion of the PRC's exports comes down to a more acceptable level, say 10%, then, without a fundamental change in the pace and pattern of domestic demand, its growth may barely reach 7%. Growth may drop a lot more if the creditdriven investment bubble bursts, exposing bad loans and excess debt in over-investing local governments, giving rise to difficulties in the banking system and, eventually, to a financial crisis.
An option is to expand rapidly in the markets of poorer economies where imports are constrained by foreign exchange availability, by simultaneously providing them the necessary financing through lending and/or direct investment. However, there are limits to what these economies could absorb while securing sustainable external debt and asset positions. For instance, total merchandise imports of least developed countries are less than 10% of the PRC's exports, while for sub-Saharan Africa as a whole, the ratio is less than one-to-five. Besides, the PRC's expansion in these markets could threaten domestic industry in certain sectors such as clothing, a main manufactured export item of several low-income countries, including the larger ones such as Bangladesh and Viet Nam. While such an option could help both the poorer countries and the PRC in expanding trade, it should only be seen as a step in the transition from export-led to domestic-demand-led growth, rather than as a way of postponing the necessary adjustment.
Another option would be to lower the foreign content of exports by upgrading and through import substitution of high-tech parts and components so as to enhance their contribution to growth. Such a transformation has been under way, but even if accelerated considerably, it will take a long time to have its effects felt on domestic content of exports. Besides, it would imply continued growth of the PRC's trade surplus, thereby aggravating global imbalances.
The solution should be sought primarily in raising domestic consumption much faster than has been the case so far. Since the beginning of the decade until the global financial crisis, investment in the PRC went ahead of consumption, and the demand gap was filled by rapidly growing exports. The share of private consumption in GDP fell from over 50% in the 1990s to around 36% on the eve of the crisis, while that of investment rose to 45%. Consumption as a share of GDP remained stable during the 2008-2009 downturn while the investment rate has been pushed up to 50% by the recent fiscal stimulus package. With a sustained slowdown in the pace of exports, a return to a growth path of some 9%−10% will require reversing the downward trend in the share of private consumption and the upward trend in the share of investment in GDP.
The main reason for underconsumption in the PRC is not excessive household savings. They are no doubt high, but not always higher than those in other DEEs. In the past few years they have remained around 20% of GDP, broadly the same as household savings in Malaysia in the 1980s and in India in recent years. As a proportion of household disposable income, they are in the order of 28% compared to 32% in India. However, at more than 50% of GDP, the PRC national savings rate exceeds that of India by a large margin because of significantly higher corporate savings or profit retentions-over 20% of GDP compared to 10% in India. The PRC's corporate profits and savings are also much higher than those in late industrializers in Asia. While household savings as a proportion of disposable income have been rising in recent years, their share in national savings has been declining because of sharply rising corporate savings. 7 The disparity between consumption and investment and the consequent dependence of the PRC on foreign markets is largely the outcome of the imbalance between wages and profits. Wages in the PRC constitute a very large proportion of household income because government transfers and investment income, including dividends, are very small. Despite registering impressive increases, wages have lagged behind productivity growth and their share in value added has been declining. The downward trend in the share of wages in GDP is almost perfectly mirrored by the share of private consumption in GDP. 8 A return to trend growth in PRC thus crucially depends on a sizeable increase in the share of household income in GDP and a corresponding decline in corporate profits and investment, which are boosted by tax incentives and the practice of nonpayment of dividends to the government by state-owned enterprises. This calls for a higher share of wages in value added and significantly greater government transfers to households, particularly in rural areas where incomes remain depressed. Greater public spending on social infrastructure in health, housing, and education would not only improve social welfare but also serve to reduce relatively high precautionary household savings. These expenditures and income transfers can be financed with dividend payments by stateowned enterprises.
The policy response by the PRC to export shocks from the crisis was not designed to address the problem of underconsumption in any fundamental way. Rather than giving a major boost to household income and consumption, public intervention has focused on investment. Less than 20% of the fiscal package of some 16% of GDP has been allocated to social spending, with the rest going mainly to infrastructure investment in roads, railways, ports, and airports. It has pushed the investment rate to 50% of GDP, aggravating the problem of excess capacity that had pervaded several sectors and increased the dependence of growth on exports. Policies designed to revive real estate demand and an unprecedented growth of mortgage lending to households created a bubble in the property market, with real estate investment growing by close to 40%. While private consumption held up thanks to several incentives, particularly for car purchases, it did not provide much impetus to offset the sharp decline in exports. The increase in investment is estimated to have contributed between 80% and 90% of growth in 2009 (Wolfe and Ziemba 2009aand 2009b, Ho-Fung 2009.
The PRC will be in a weaker position to give a similar positive response to fallouts from a possible second dip in global economic activity that may result from debt difficulties and premature fiscal tightening in Europe. Indeed, the longer the adjustment to underconsumption is delayed, the greater the vulnerability of the PRC to instability of economic activity in its main markets in AEs. Instability in the latter economies is likely to persist since the crisis response has not eliminated its root causes-namely, excessive indebtedness and rapid liquidity expansion.

B. The PRC's East Asian Suppliers: Stimulating Domestic Demand
As noted, the dependence of growth on exports is greater in East Asian DEEs closely participating in the Sinocentric production network than in the PRC. These economies have direct exposure to a sustained slowdown in exports to the EU and the US, which together account for more than a quarter of total exports of some of these countries. They also have a significant indirect exposure through the PRC. Although the PRC has become the largest export market for an increasing number of East Asian DEEs, an important part of the PRC's imports from them is used for inputs into exports of consumer manufactures to the EU and the US. For every $100 worth of processing exports of the PRC to the US and the EU, about $35-$40 accrue to East Asian DEEs.
This means that while the PRC is a major importer from East Asian DEEs, it is not a major market for them since an important part of the PRC's imports is destined for exports rather than used internally. Domestic consumption in the PRC generates proportionately much less demand for imports from East Asian DEEs than its exports to the EU and the US. Consequently, a shift by the PRC from export-led to consumption-led growth could result in a significant slowdown of its manufactured imports from East Asian DEEs. Thus, at its current pattern of domestic spending, the PRC market is not a good substitute for the EU and the US markets for East Asian DEEs. The PRC cannot replace the US even if it maintained GDP growth of some 10% based on domestic consumption rather than exports, as its GDP is about one third that of the US; the share of households in GDP is much smaller; the population saves a much higher proportion of disposable income; and the import content of household consumption is much lower than that of the US. Briefly, a PRC-US rebalancing can make it quite difficult for East Asian DEEs to sustain rapid export-led growth.
Outside the PRC a main reason for excessive reliance on exports is underinvestment.
In several economies including Indonesia; Malaysia; the Philippines; Singapore; and Taipei,China, investment rates have been hovering around 20% of GDP in recent years, less than half the rate in the PRC. In none of these economies have investment rates recovered back to the levels attained before the 1997 crisis. 9 Even recognizing that the pre-1997 investment boom was an unsustainable bubble driven by massive capital inflows, recent investment rates are too low to generate a rapid growth of either productive capacity or effective demand.
Exceptionally high investment in the PRC and low rates of investment in the rest of East Asia are related. Generous incentives provided by the PRC to export-oriented FDI play an important part in attracting large amounts of investment from the region. There is a need to redistribute aggregate investment within East Asia, from the PRC toward the rest. This would be greatly helped if the PRC were to start focusing on domestic markets and dismantling incentives to export-oriented FDI.
Private consumption has also been weak in most East Asian countries. In the Republic of Korea; Malaysia; Taipei,China; and Thailand its share in GDP barely reaches 55%-much below the rates in more affluent countries such as the US (over 70%) and the EU (some 60%). Singapore is another underconsumption economy in the region where the share of private consumption in GDP has been declining since the beginning of the decade-it is now below 40% while national savings are as high as 53% of GDP, very much like the PRC. In several of these cases underconsumption has its origin, in part, in low and/ or declining shares of wages in income. 10 They thus face the dual task of raising both consumption and investment while allowing wages to grow faster.

C. Intraregional Trade and Industrial Restructuring in East Asia
The crisis has uncovered a high degree of vulnerability of East Asian DEEs to trade shocks, raising the question if export-led growth is no longer viable. This echoes the dilemma that Arthur Lewis pointed out in his Nobel Lecture three decades ago (Lewis 1980) that dependence of growth in DEEs on AEs through trade would make it difficult to catch up and close the income gap. The solution proposed by Lewis was to develop an internal market in DEEs and south-south trade.
East Asia has ample space in these respects. Traditionally the balance-of-payments constraint is seen as the main reason for the dependence of DEEs on exports. But East Asian DEEs export not simply to earn foreign exchange for imports needed for capital accumulation and utilization of productive capacity, but to find markets for goods for which there is little or no domestic demand because of imbalances between investment and consumption and between profits and wages. The solution is not to turn inward away from world markets, but to stop relying on cheap labor and cheap currency, and to start allowing wages and private consumption to grow in tandem with productivity, and to underpin the expansion of productive capacity by providing growing internal and regional markets in final goods.
While absolutely necessary, rebalancing domestic and external components of demand would not be sufficient to sustain growth based on the expansion of domestic markets. A shift from export-led to consumption-led growth would also require significant industrial restructuring. An important part of the PRC's exports are specific to foreign markets with little domestic demand. Similarly, much of its physical and human infrastructure linked to manufacturing is geared to exports rather than domestic consumption. A relatively sizeable excess capacity is known to exist in both industry and infrastructure and this can be laid bare with a permanent slowdown in exports to the EU and the US.
For many of the PRC's regional suppliers, the task is even more onerous. The parts and components destined for the PRC as inputs into exports to the EU and the US has even lesser domestic use as they are specific to the regional production network and the associated division of labor. For the same reason they cannot easily be diverted from the PRC to third markets.
A slowdown in exports to AEs can be expected to produce significant consolidation in the Sinocentric East Asian production networks. It could reduce the number of suppliers of parts and components as increased competition drives out high-cost producers. It could also restrict the scope for division of labor and the number of segments in value chains through import substitution in parts and components in countries assembling them into final goods. A significant restructuring and consolidation along these lines could restrict growth of exports and income in the PRC's East Asian suppliers even if the PRC continues to grow on the basis of expansion of domestic consumption. 11 Establishing industries for final goods for domestic consumption and exports using these parts and components would not always be a viable option when countries lack the kind of productive endowment needed in all other segments of production.
Reducing the dependence of growth in the region on the EU and US markets calls for a gradual move away from vertical specialization and network trade to product-based horizontal specialization and intraregional trade in final goods, as in the traditional sequential "flying geese" pattern of division of labor. 12 This would require the more advanced economies to phase out labor-intensive manufactures and allow them to move to lesser developed countries, eventually to become importers of such goods, rather than retaining such industries by keeping wages behind productivity and following a policy of cheap currency. For countries at similar levels of industrial development and technological maturity, a shift from vertical to horizontal specialization would promote the kind of intraindustry trade that underpinned rapid expansion and integration of Europe during the golden age when countries were producing and trading similar products in industries for passenger cars, consumer electronics, and other household durables.
Foreign direct investment from more advanced economies would certainly play a key role in relocating labor-intensive industries in the lesser developed countries in the region, thereby helping to deepen regional integration through both trade and investment. This was the course taken by Japan and subsequently some of the larger first-tier NIEs in the process of industrial upgrading and productivity growth. Closing the gap between wages and productivity in the PRC would not only help accelerate growth of domestic consumption and demand for final goods, but would also encourage firms in certain laborintensive industries to move them gradually to poorer countries in Southeast and South Asia through FDI. This could also contribute to a more balanced distribution of investment throughout the region.
11 For a discussion of consolidation of international supply chains and evidence during the recent crisis, see Milberg and Winkler (2010) who call the former horizontal consolidation and the latter vertical consolidation. 12 In the flying geese pattern of growth and division of labor, trade turns out to be complementary when there is diversity among countries in their stages of development and resource endowments, since different countries would be specialized in different industries. However, as endowments (skills and technology) change, competitive pressures will develop, leading to shifts in industries among countries; see Akyüz et al. (1998) and Ahearne et al. (2006).
To become a regional locomotive in the flying geese process, the PRC would also need to increase its imports of final goods both from lesser developed and more advanced countries in the region. This calls for not only an increase in the PRC's domestic consumption as a proportion of GDP, but also its import content. While the share of final goods in the PRC's imports from the region has been increasing in recent years (Athukorala 2008;Kim, Lee, and Park 2009), production sharing continues to dominate intraregional trade.
The second-tier NIEs in Southeast Asia-Indonesia, Malaysia, the Philippines, and Thailand-have been squeezed between low-wage producers in the PRC and other emerging exporters of labor-intensive manufactures and highly-skilled, innovative producers in the first-tier NIEs. Many of the sectors vacated by the more-advanced firsttier NIEs has been captured by the PRC rather than by these higher-wage second-tier NIEs (Ahearne et al. 2006). To avoid falling into a middle-income trap, these countries need to move up the industrial ladder. Most of them have ample domestic resources but, as noted, they lack animal spirit and investment needed for such a restructuring. Re-animating private investment and directing them toward high-productivity, skill-intensive sectors would serve the dual purpose of acceleration of domestic demand and industrial restructuring.
Indeed, the industrial restructuring needed to change the product mix to match domestic and/or regional demand in East Asia would need to be achieved primarily through new investment rather than by reshuffling the existing production capacity since skills, capital equipment, and organizational structures are often industry-specific and even product-specific. However, the new incentive structure associated with the rebalancing of domestic and external demand, including changes in wage-productivity configurations and exchange rates, may not be sufficient to drive investment toward areas that need to expand in order to meet a faster growing domestic market. Therefore, policies designed to accelerate domestic demand relative to exports to AEs need to be accompanied by an industrial and investment strategy for dovetailing the structure of production to the pattern of domestic and regional demand.
Industrial restructuring and shifts in intraregional division of labor would also have implications for trade policies. Trade liberalization in regional free trade agreements often encourage network trade since tariffs charged on vertically integrated goods decline by a multiple of agreed reductions (Yi 2003). However, their relative impact on network trade and trade in final goods also depends on how the provisions for "rules of origin" are formulated. A shift of emphasis from network trade to trade in final goods could thus require important adjustments in intraregional tariff structure and trade concessions.