Export or Domestic-led Growth in Asia?

In recent years, some developing Asian countries claim to have started shifting emphasis from export-led to domestic demand-led growth policies with a view to achieving a more balanced growth strategy. This paper evaluates empirically how far this shift has gone. The evaluation—based on an analysis of five countries—finds no evidence that the period 1993–2003 has been marked by such a shift. It also finds that periods of expansionary domestic demand and deteriorating net exports signaled an ensuing crisis. This should serve in the future as an early warning system.


I. INTRODUCTION
Since the East Asian financial crisis erupted in 1997, countries in the Asian and Pacific region have been immersed in a search exercise to identify what policies led to the crisis and recession, and what alternative set of policies would lead them back to a path of sustained and higher growth rates (Felipe 2003).The majority view has been that the crisis was the consequence of a fundamental flaw in precrisis financial policies, which led to currency overvaluation, overborrowing, and overlending for the domestic economy; and speculative bubbles that eventually burst (for an overview see Jomo 1998, Seguino 2000, Lim 2004).
As part of the "package of solutions" to reinvigorate these economies, a number of policymakers in the region proposed shifting (some of them more openly, e.g., Thailand, and some others less so, e.g., Malaysia) from the exportled growth model to a "new development paradigm" based on domestic demandled growth.This way, it is argued, the Asian countries hit by the crisis are making efforts at diversifying their economic base away from over-reliance on external trade, the basis of the export-led growth model.Since 2001, a number of news articles have analyzed and followed this alleged shift (see, for example, the articles in The Economist [2005] titled "Heading back" (p.9) and "Thaksin's way" (pp.22-4).Thailand's Prime Minister Thaksin Shinawatra, for example, announced upon taking the helm of government in January 2001 that he was determined to move the country away from mass manufacturing for exports into domestic demand-led growth through a series of policies.The country's policymakers are making big efforts toward shifting economic policy to reduce the country's overdependence on external demand and foreign capital.The high growth rates achieved by Thailand in recent years seem to vindicate the new approach.However, Mr. Thaksin's approach is not, strictly speaking, just a transformation from export-led growth into domestic demand-led growth, if by the latter it is meant a series of policies to boost domestic demand (this will be defined in Section IV).
His policies are based on what has been referred to as a "dual track" strategy (Lian 2004) of relying on external demand (first track) and simultaneously developing domestic demand and supporting domestic enterprises (second track).Though it is true that his policies emphasize private consumption, they try to boost the demand of domestically produced goods and services (see Box 1).
Since coming to power in 2001, Mr. Thaksin's objective has been to alter Thailand's production structure with a view to reducing the country's dependence upon exports.The key is to create demand among households and businesses without creating another bubble (i.e., to avoid a household-led spending boom fueled by borrowing like in the United States).Moreover, Mr. Thaksin's strategies aim at boosting domestic demand and strengthening local enterprises and at developing indigenously owned production capacity. 1alaysia is also making an effort at diversifying its economic base; and Republic of Korea (Korea) is reported to have gone into a debt-led consumption binge right after the Asian crisis, which has led to the current mini crisis of credit card defaults and weak consumption demand that cause low growth.
It is therefore important to analyze whether the empirical evidence indicates that indeed a shift from export-led growth to domestic demand-led growth is indeed taking place across Asia, and what are the consequences of this shift.In particular, does the data appear to confirm this move toward a domestic demand-led growth strategy?This study more precisely attempts to answer the following questions: (i) Does the evidence indicate that countries are switching from exportled growth to domestic-demand driven growth?(ii) Did the export-led strategies partly contribute to the East Asian crisis?(iii) What lessons can be drawn from the different country experiences?Box 1.What is Thaksinomics In August 2004, the Thai government published a white paper entitled "Facing the Challenge: Economic Policy and Strategy", explaining clearly the economic agenda that Mr. Thaksin has been trying to implement since he came into office on January 2001.The message is that his policies try to balance past excessive dependence on external demand, urban-based mass manufacturing, and unproductive asset-building on one hand; with structural development in domestic demand, traditional sectors (e.g., agriculture, SMEs, and rural households) and entrepreneurs, and improvement in the pricing power of Thai goods and services on the other.Thus, Mr. Thaksin intends to revive domestic demand (by boosting private consumption and by developing the traditional sectors), in addition to exports.This is what has been referred to as a dual track strategy, as opposed to the single-track model followed by many countries in the region, namely, producing for exports.Mr. Thaksin's dual track strategy is five-pronged: (i) Revitalize growth at the grassroots level.The key policy initiatives are embodied in the following programs: one tambon, one product; SME and entrepreneur promotion; farmers' debt suspension; village and urban community revolving fund; The People's Bank of the Government Savings Bank; SME loans; venture capital; and asset capitalization.(ii) Jumpstart key sectors.The paper contains ideas for the key sectors of the economy.For example, for agriculture, it is argued that it is crucial to identify new demand for Thai agricultural products both domestically and abroad.For manufacturing, the government has created a new Entrepreneurs Promotion Board set up to create 50,000 new SME businesses.On tourism, his policy sets out to promote Thailand aggressively and to capture the upper middle classes of Chinese, Indians, and affluent Europeans.Regarding real estate, the government has disregarded the standard prescriptions of "fire sales" and driving asset prices to their true bottom.Instead, it has promoted asset reflation.Finally, on the finance sector, Thaksin's government has put in place a financial sector master plan to create a more efficient and competitive financial system.
(iii) Enhance economic efficiency and long-term competitiveness.The government has identified a series of industries to promote: automotive, tourism, software, food, fashion, health care services, hospitality, rubber, and furniture.(iv) Provide a stable and supportive macroeconomic environment to facilitate growth while maintaining overall policy discipline.The government has raised tax revenue, consolidated spending, balanced the budget, and retired public foreign debt.(v) Promote the external sector by expanding markets, and foster financial stability through regional and global cooperation.Under the dual track strategy, the external sector is as important as the domestic.Thus, exports remain a cornerstone of the strategy In order to address these three questions, the paper analyzes growth from the point of view of the aggregate demand components.The approach is based on the analysis of the information provided by the basic demand-side macroeconomic accounting identity, according to which output equals the sum of consumption, investment (i.e., domestic demand), and net exports.
Section II offers a summary and discussion of the export-led growth strategy as well as a summary of some recent critiques of this strategy.These criticisms have led (at least in the view of some authors) to the theoretical rationale for the alleged need to shift to a domestic demand-led growth approach.Section III defines the two types of growth strategies for purposes of the subsequent discussion.
The empirical work is carried out in the form of three complementary analyses (Sections V-VII).Since the objective of our study is limited to an expost, factual, and descriptive analysis of whether a shift to domestic demand-led growth is taking place, the methodology used is very simple.We look at output (gross domestic product [GDP]) from the demand side.This way, the latter is made up of the domestic demand components-consumption and investmentand net exports (exports less imports); and we do so from the point of view of an accounting identity, i.e., there is no attempt at modeling in the sense of understanding ex-ante, causal, or behavioral relationships.Section IV presents the results of a growth accounting exercise performed on the aggregate demand components of a selected group of Asian countries, namely, People's Republic of China (PRC), India, Republic of Korea (henceforth Korea), Philippines, and Thailand.Growth accounting apportions overall GDP growth to the contribution of each component of demand.Thus, overall growth of output is the sum of the growth rate of each component multiplied by its share in GDP.For example, the contribution of the growth of private consumption to overall GDP growth is calculated as the product of the growth rate of consumption times the share of consumption in GDP.Expressed as a percentage of the overall growth rate, it is the ratio of this product to the growth rate of GDP.The exercise provides a longrun view of these five countries in terms of the contribution of growth in domestic demand components and net exports to overall growth.
Section V broadens the analysis by looking at the expansionary versus nonexpansionary (or even contractionary) stances or positions of the private sector, government or fiscal sector, and external trade sector, over the last 20 years, in terms of aggregate demand injections versus leakages of the three sectors.Over the last three periods there have been substantial changes in demand-side parameters, such as import coefficients, tax efforts, and savings rates, along with jumps in flows such as annual exports, investments, government spending, etc.The analysis in this section looks at how output has responded to these shifts, using a simple decomposition of demand "injections" (private investment, government spending, and exports) versus "leakages" (private savings, taxes, and imports).The analysis helps identify whether the component of demand in question has an expansionary or nonexpansionary contribution to aggregate demand (naturally, ex-post, total injections must be equal to total leakages).
It must be pointed out that while the growth accounting exercise provides a long-run picture over 30 years (with periods grouped into three 10-year intervals) in terms of the growth contribution of each demand component to overall growth, the stances provide an annual graphical picture over 20 years of the different phases of growth of the five countries by identifying expansionary and nonexpansionary factors (private, government, and external sectors) in effective demand.
Section VI completes the empirical analysis with a comparison of the shares of aggregate demand components for a large number of Asian and Pacific countries, classified according to three income groups, with the shares of a group of small open European economies.Since it is impossible to carry out the growth accounting and stances analyses for all Asian and Pacific countries, the analysis of the demand shares provides an overall picture.The last section provides a summary and conclusions.
Our study leads to the conclusion that the more successful phase of development of the selected countries has been associated with significant investment increases and capital accumulation as well as with significant export growth that brought about trade surpluses or reductions in trade deficits.For the countries badly hit by the Asian crisis in 1997-1998, the instabilities were preceded by unbalanced growth in demand components, with highly expansionary domestic demand, and increasing trade deficits.This was the result of currency overvaluations, overborrowing, and overlending in the domestic private sector, and rise of speculative bubbles that most economists agree triggered loss of confidence, massive currency depreciation, and capital flight during the crisis.The harsh adjustments during the crisis resulted in the collapse of domestic demand (especially investments) as net exports recovered sharply.Thus, it was not the export-led strategy that contributed to the crisis.Conversely, it was the promotion of debt-financed domestic demand growth at the expense of net exports that precipitated it.
The analysis suggests that the best periods seem to be those when both domestic demand and net exports exhibit significant and continuous growth or improvements, as in the case of the PRC and India today, or in post-crisis Thailand.This was also the case of the post-Plaza Accord period of the second half of the 1980s in Korea and Thailand, when the reputation of the East Asian miracle reached its peak.Periods when domestic demand was highly expansionary occurred at the same time that net exports signaled an ensuing crisis, as the experiences of Korea, Philippines, and Thailand have shown.In this sense, the analysis in Section V provides an "early warning system." The comparisons between the upper medium and low-income Asian and Pacific countries show that, during the last period (1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003), the highperforming Asian countries outperformed the European countries in terms of growth in both exports and net exports.The Asian and Pacific middle-level and low-income countries have, on average, improved their trade deficits during the last period.However, the low-income countries still have very high trade deficits that need to be reduced (or, alternatively, the gap between aggregate domestic demand and domestic production has to be reduced).But there is no evidence that countries in the region have been exhibiting recently growing domestic demand shares at the expense of net exports.
Inasmuch as the study suggests that healthy growth for developing countries should be the result of growth in both domestic demand and net exports, the last section includes a general discussion about how the international trade system should be more responsive to the needs of poorer countries with a view to allowing them to benefit from international trade.The paper proposes that, to provide developing countries with the proper environment to achieve improvements in their net exports, the international trade system should provide the developing countries with mechanisms to reduce their large trade deficits.This requires: (i) a more open international trade system: richer and trade-surplus countries can contribute by opening up their agricultural, industrial, and service markets to the developing world; and (ii) poorer and deficit-ridden countries to use price and nonprice mechanisms to improve their productivity and competitiveness in the world market.
Some final words on methodology are important: (i) the demand side growth accounting and the stances exercises are not, strictly speaking, an economic model in itself (or based on a model), so no causal inferences should be drawn.The former is simply a device to split and apportion, ex-post, the growth of output from the demand side.The latter provides also an ex-post classification of how the private, government, and trade sectors contribute to expansions or contractions in output; (ii) the analysis does not take into account any supply-side consideration (e.g., the relationship between exports and technology upgrading, often brought up in the discussions of the benefits of export-led growth); (iii) although the analysis in the paper is an exercise in positive economics, it leads naturally to the normative observation that the problem being considered should not be an either-or choice between domestic demand and export-led growth, but a need to actually give both domestic demand growth and net export growth due importance and proper balance.This is especially crucial since developing countries need precious foreign exchange for their economic development, which net export earnings provide; and (iv) it is virtually impossible to clearly discern a structural change from export-led growth into domestic demand-led growth with the latest 3-year data where domestic demand-led growth was supposed to have taken place.If this is happening, it will take years, perhaps a decade, for the data to show.Hence, our analyses cover 20 or 30 years and analyze episodes of the two strategies mentioned.

II. THE EXPORT-LED GROWTH STRATEGY
The export-led strategy consists in encouraging and supporting the production for exports.The rationale, going back to the classical authors, is that trade is the engine of growth, in the sense that it can contribute to a more efficient allocation of resources within countries as well as transmit growth across countries and regions.Exports, and export policies in particular, are regarded as crucial growth stimulators.Exporting is an efficient means of introducing new technologies both to the exporting firms in particular and to the rest of the economy, and exports are a channel for learning and technological advancement.Moreover, the growth of exports plays a major part in the growth process by stimulating demand and encouraging savings and capital accumulation, and, because exports increase the supply potential of the economy by raising the capacity to import.
Indeed, as a development strategy, the classical belief was that development could be transmitted through trade.Early economists justified the promotion of exports and trade with the traditional argument of comparative advantage.Accordingly, opening up a country's markets to the international market allows a country more efficient production and allocation of resources as the country can concentrate on the production of goods in which it has a comparative advantage based on its factor endowments.Thus, world trade markets allow producers and consumers of the participating countries to benefit from lower prices, better quality products, more diverse supply of goods, and higher growth.The export-led growth model seemed initially to have been vindicated with the success of the East Asian "miracle" countries, which achieved extraordinarily high growth between the 1970s and the mid-1990s, supposedly through export promotion.Since the eruption of the Asian crisis, however, there have been increasing doubts from some sectors as to the feasibility of export-led growth for many developing countries (Felipe 2003).
Recent decades have brought about other important justifications for export promotion.Some of these are: Participating in trade, especially export production and promotion, exposes a country to the latest and most advanced production and marketing techniques, and a "learning by doing" process that brings about dynamic innovation and technological diffusion into the economy.It also drives a country to higher production and economies of scale, which lead to increasing returns (see Felipe 2003).(ii) Many development economists use the "two-gap or three-gap" models of Chenery (1969), Bacha (1990), or Taylor (1993) to justify the need to earn foreign exchange via exports.According to these models, the investment−savings gap and the foreign exchange gap are major obstacles to the growth and development of many developing countries.
As countries need precious foreign exchange for their development needs (capital goods, industrial raw materials, oil and food), export earnings are more efficient means to finance these needs than foreign debt since the latter is vulnerable to adverse exogenous shocks and currency risks that may lead to debt defaults.(iii) A similar argument (see McCombie and Thirlwall 1994) claims that large balance of payment deficits, spurred by large import propensities or elasticities, may be a hindrance to growth for many developing countries.Thus, moderate trade deficits, or trade surpluses, are more desired.This, of course, implies that export growth should be in pace, or should outpace, import growth.(iv) Felipe (2003) also argues that export-led strategies allow an expansion of aggregate demand without much inflationary pressure and without the danger of a wage-price spiral, compared to strong domestic demand injections.This is partly due to the resulting real appreciation of the currency as a result of large export earnings, which tame inflation and allow real wages to rise.
It is important to mention that while the growth and development literatures do consider the export-led growth strategy, the so-called domestic demand-led growth strategy is not a term defined and used (hence we have to define it, in particular for purposes of empirical implementation; see Section III).2Therefore, it is not straightforward to place the "debate" between export and domestic demand-led strategies in a theoretical context.The term and debate or controversy between the two has appeared mostly in the media, as noted in the Introduction.
In recent years, however, a series of economists have hypothesized that the East Asian crisis had very different roots and that after several decades of being presented as the optimal growth strategy, the export-led growth model that the East Asian countries followed, ultimately gave in and even harmed the growth prospects of developing countries.These economists have put together a critique of the export-led growth model and proposed a shift toward domestic demand-led growth.Palley (2002), for example, has argued that the emphasis on export-led growth of most East Asia countries had a series of negative effects.First, it prevented the development of domestic market growth.Second, it put developing countries in a "race to the bottom" among themselves.Third, it put workers in developing countries in conflict with workers in developed countries.Fourth, there is a relationship between export-led growth and financial instability by creating overinvestment booms.Fifth, due to the emphasis placed on global goods and commodity markets, this model has aggravated the long-trend deterioration in developing-country terms of trade.Finally and most importantly, export-led growth has reinforced the dependency of developing countries on the developed world, thus becoming vulnerable to slowdowns in the latter's markets (e.g., as in the slowdown of the semiconductor world market in 1996-1997 right before the Asian crisis).Export-oriented economies are dependent on foreign (mostly Western) demand.The problem is that recessions in Europe, Japan, or US translate into slow growth in the developing world.Summing up, Palley (2002) argues that the export-led growth model followed by East Asian countries for several decades is not an optimal strategy any longer.Blecker (2002 and2003) has also contended that the adoption of a development strategy that relied on high rates of growth of manufactured exports is the root cause of the problems that led to the crisis, for such a strategy led to growing excess capacity, intensified competitive pressures, and disappointing growth performance.In a similar vein, Kaplinsky (2000) and Ertuk (2001/02) have suggested the possibility of immiserizing growth as a result of the creation of excess capacity in export-oriented manufacturing industries.During the 1990s too many developing countries entered the more advanced product categories thus creating excess capacity and fostering falling prices.Blecker (2002 and2003) has argued that the reliance on export growth suffers from a "fallacy of composition."The reason is that if too many countries try simultaneously to rely on export-led growth policies to stimulate growth under a given set of global demand conditions, the market for developing countries' exports is limited by the capacity of the industrialized nations.If demand in the developed countries stagnates, it translates into overinvestment and excess capacity in the developing countries.As East and Southeast Asian countries plunged into the financial crisis, the first policy option considered by all of them in order to resume growth was the export-led strategy.However, the problem with this strategy is that the problem of fallacy of composition was compounded, since during the last decade the PRC has been added into the equation.Export-led growth operates through a hierarchical process with less developed newcomers replacing more maturing export economies as their wages grow.The PRC poses an entirely different problem for it has a fairly large supply of labor so that it can keep wages very low and, seemingly, for a long time.
Blecker summarizes his views as follows: "the current emphasis on exportled growth in developing countries is not a viable basis on which all countries can grow together under present structural conditions and macroeconomic policies" (Blecker 2003).Palley (2002) has gone further and contends that the export-led growth model followed by many developing countries during the last few decades was part of the so-called "Washington consensus" emphasis on trade liberalization. 3As a solution, Palley proposes a new development paradigm based on domestic demand-led growth.

III. DEFINITION OF DOMESTIC DEMAND-LED AND EXPORT-LED STRATEGIES
The analysis is performed in terms of the macroeconomic accounting identity: 3 The term "Washington consensus" was coined by Williamson (1990).In its original formulation, the idea encompassed fiscal discipline, reorientation of public expenditures, tax reform, interest rate liberalization, unified and competitive exchange rates, trade liberalization, openness to foreign direct investment, privatization, deregulation, and securing property rights. 4Palley certainly acknowledges that developing countries need to export.What he argues is that "the global trading system must be made the servant of domestic development, and domestic development must not be forgone for the sake of international competitive advantage" (Palley 2002, 4).For him, domestic demand growth rests on four pillars: (i) improved income distribution, (ii) good governance, (iii) financial stability, and (iv) a fairly priced supply of development finance.And the policies needed to put these pillars in place are (i) labor and democratic rights; (ii) financial reform; and (iii) a combination of debt relief, increased foreign aid, and increased development assistance through the expansion of special drawing rights.
where GDP stands for gross domestic product, C p is private consumption, C g is government consumption, I is gross domestic investments or gross domestic capital formation (GDCF), X and M are exports and imports of goods and services, respectively.We will refer to an export-led development growth strategy as one that results in: (i) high export growth, accompanied by high GDP and income growth; and (ii) improvement in net export growth, i.e., higher export growth than import growth.
Conversely, we will say that growth is strictly speaking domestic demandled if domestic demand is growing, accompanied by GDP and income growth.
The share of each component in output is defined as: A convenient way of categorizing the different possibilities for the two strategies is as follows.The first three terms on the right-hand side of identity (1) or consumption of the private and government sectors plus investments are the domestic demand components, while (X-M), or net exports, is the other component of aggregate demand.Thus the following cases can arise: (i) Domestic demand is growing and net exports are deteriorating (becoming a smaller positive number or larger negative number).If GDP growth is positive, then growth must be domestic demand-led.This is the only case where one can, strictly speaking, refer to domestic demand-led growth.(ii) Domestic demand and net exports are growing.Thus, growth is due to both domestic demand and net exports.Which one is contributing more to growth is simply an empirical issue.If domestic demand is growing faster, we will say that growth is demand-led, but weakly speaking.(iii) Domestic demand is deteriorating and net exports are increasing.If growth is positive (which is often not the case since domestic demand is usually a much larger component of GDP), growth must be net export-led.If growth is negative, the recession is due to a decline in domestic demand.(iv) Both domestic demand and net exports are decreasing.Obviously, we have an economic recession and negative growth rates are due to declines in both domestic demand and net exports.
It must be pointed out that as we separate GDP into the domestic demand and net export components, the share of domestic demand will be much larger than the net export share, usually comprising more than 90 percent of GDP when net exports are positive.(When net exports are negative, the share of domestic demand will be more than 100 percent.)This is because much of the export earnings will go to import purchases, and since net exports track the difference between these two trade variables, the magnitude becomes quite small compared to domestic demand.This is true even in the most successful export-led growth cases where export growth is double-digit.5

IV. DEMAND-SIDE GROWTH ACCOUNTING EXERCISE
In this section we perform a growth accounting analysis on the components of demand.As indicated above, the objective of this exercise is to apportion overall growth between domestic demand and net exports.Technical details are shown in Box 2.
The five countries chosen provide a relatively wide spectrum of experiences and results: PRC, India, Korea, Philippines, and Thailand.The first two are the oft touted Asian success stories in the most recent decade due to their opening up to international trade, and the latter three countries were countries affected by the Asian crisis in 1997-1998.Table 1 gives the shares of the expenditure components of GDP at constant prices for the five countries.Table 2 shows the average annual growth rates of GDP and of demand components over the intervals 1973-1983, 1983-1993, and 1993-2003.Table 3 provides the growth rates of the expenditure components weighted by their shares in GDP.This gives, in growth rate terms, the contribution of each component to the growth rate of GDP.Finally, Table 4 displays, in percentage share, the contribution of each aggregate demand component in overall GDP growth.

Box 2. Demand-side Growth Accounting
Real output from the demand side is given by the National Income and Product Accounts as where GDP stands for gross domestic product, C p is private consumption, C g is government consumption, I is gross domestic investments or gross domestic capital formation, and X and M are exports and imports of goods and services, respectively.
In growth rate terms: where the symbol ^ denotes growth rate of the variable.
The above simply states that the growth rate of GDP is the sum of the products of the shares in GDP times the growth rates of private consumption, government consumption, gross domestic investments and exports, less the product of the share of imports and its growth rate.
Real values were derived for 1973, 1983, 1993, and 2002  (3) For a continuously increasing positive x the above method will yield a higher annual average growth rate than taking the actual annual growth rates of x in years 1974, 1975 up to 1983, and then averaging them. 6he method employed here also uses the GDP estimate without taking into consideration the statistical discrepancy between the value added GDP estimate and the expenditure GDP estimate.That is, the GDP in the denominators of the shares in equation (2) uses equation (1) exactly without including the statistical discrepancy.This allows the expenditure shares to sum up to exactly 100 percent, and for equation (2) to sum up exactly to the GDP growth rate.

A. People's Republic of China
The tables show that the PRC registered high domestic demand growth in the first two periods, 1973-1993, while its net export position deteriorated and was negative. 7This happened even as the growth of exports posted annual averages of more than 20 percent (since imports increased more than exports).The last period, 1993-2003, however, saw not only continuing large growth in domestic demand components, but also a strong shift from negative net exports (or trade deficits) to high positive net export (or trade surplus) positions, as export growth accelerated and import growth decelerated.Thus the PRC's growth experience during the last period points to high growth in both the domestic demand components and in the net export component.Domestic demand contributed around 90 percent to the double-digit GDP growth of the PRC in 1993-2003, while net exports contributed around 10 percent (Table 4).It is also important to point out that, in all three periods, investment growth outpaced consumption growth (Table 2), so that the last period saw a larger contribution of investment than consumption to GDP growth, an increase in the share of capital formation (to more than 40 percent), and a continuing decline of the share of private consumption.It must be emphasized that in the last period the share of net exports to GDP grew substantially, reflecting the PRC's transition from a negative contributor to growth to a high positive contributor (see Fan and Felipe 2005).

B. India
India registered positive average annual GDP growth during the three periods, but lower than the PRC.The first two periods  were marked by growth in domestic demand as net exports deteriorated.During the last period, when India opened up to the international market, the country exhibited even higher growth, with higher growth in the domestic demand components, but now the trade deficits improved so that net exports contributed slightly to overall GDP growth.During the third period, the growth rates of exports and imports more than doubled, with exports outpacing imports, leading to the decline in the trade deficits (net exports became a smaller negative number).Like in the PRC, investment increased more than consumption in the last period, with the consequence that the share of capital formation increased, while that of private consumption fell.But the high share of consumption still made this component of demand the largest contributor to growth in the last period.Finally, the last period saw an increase in the share of net exports to GDP (actually a decline of its negative share to GDP) and a slight decline in the share of domestic demand to GDP (see Fan and Felipe 2005).

C. Republic of Korea
1973-1983 was the high-growth period for Korea, when it started being touted as an East Asian tiger.During this period, the domestic demand components of GDP grew very fast.Export growth exceeded 20 percent during this period and surpassed import growth so that the country registered net export growth.At the same time, there was strong domestic demand growth.The trade surplus position reversed during 1983-1993 as Korea began exhibiting trade deficits in the early 1990s, even if exports continued growing at a very high rate.The very high GDP growth during this second period, therefore, was due to high growth of domestic demand, with net exports deteriorating and turning negative toward the 1990s.Trade deficits continued until the Asian crisis.The third period, 1993-2003, reversed the trade deficits, and the country returned to positive net exports starting in 1998, at the height of the Asian crisis.Because of the significant contraction of the economy in 1998, the growth rate of the last period was lower than those registered during the last two periods, though still respectable.The last period saw a slower growth of consumption and investment than in the previous periods, with investment actually losing share of GDP (reflecting the investment collapse of 1998).Net exports contributed to GDP growth in this last period, and increased its share in GDP, while the share of domestic demand fell.

D. Philippines
The Philippines exhibited respectable growth during 1973-1983, with domestic demand growing significantly.Trade deficits (negative net exports) worsened in this first period.1983-1993 was a difficult period for the Philippines, marked by the economic collapse of 1984 and 1991.Average annual growth was low during 1983-1993, which saw a decline in investment and low growth in consumption.Trade deficits also worsened, contributing to the low growth.The period, therefore, was characterized by stagnation, with net exports not improving by the end of the period (1993).The last period (1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003) saw an improvement in growth rates, but net exports continued to be negative and did not improve in absolute terms, though they did improve as a percentage of GDP.The Philippines, therefore, is the only case among the five countries analyzed where all three periods, including the last one, were marked by growth in domestic demand and deterioration in net exports, although there was an improvement in terms of the share of net exports to GDP (to a smaller negative number).

E. Thailand
Thailand registered very high growth in the first two periods, 1973-1983 and 1983-1993, with both investment and consumption growing very fast.This was accompanied by deteriorating net exports in the two periods. 8The deterioration of net exports during 1983-1993 was accompanied by spectacular growth rates in both exports and imports.The last period saw a significant fall in the GDP growth rate, as a consequence of the Asian crisis, which hit Thailand in 1997, and resulted in steep GDP and investment declines.Because of this, investment fell during the third period while consumption grew slowly and net exports turned from negative to largely positive.Thailand's GDP growth in 1993-2003 stemmed largely from improvements in net exports, which contributed 71 percent of the country's overall growth.Thus, Thailand's post-Asian crisis improvement in net exports was the main contributor to growth during the last period, rather than domestic demand.
Table 5 summarizes the results of the growth accounting exercise.The overall picture that emerges from the analysis of the selected countries indicates that during the first two periods, especially 1983-1993, domestic demand was the main driver of growth, as net exports deteriorated.The last period, on the other hand, was accompanied by significant improvements in the net exports position of the selected group of countries (with the exception of the Philippines).This is true for countries experiencing continuous growth (PRC and India) and for the countries hit by the Asian crisis (Korea and Thailand).The PRC and India registered high domestic demand growth in the last period, simultaneously with net export growth (and very high export growth).Korea and Thailand saw net exports drastically turning from negative to highly positive and contributing significantly to growth, as the domestic demand components grew more slowly.
It must be noted that in the East Asian tigers such as PRC, Korea, and Thailand, export growth actually decelerated in the last period compared to the previous one, but export growth was still double-digit.On the other hand, the growth rate of imports decelerated more with the consequence that all three countries saw improvements in their net export positions.
Export growth accelerated very strongly in India during the last period, much more than imports, leading to the reduction of the country's trade deficit.The Philippines had the slowest growth in exports in the last period, and it is the only country with deteriorating net exports.
The net export share to GDP improved in all five countries.Even countries with negative net exports (or trade deficits) improved their positions.India was able to reduce its trade deficit in terms of magnitude.Trade deficits increased in magnitude in the Philippines, but declined in terms of the share in GDP.Tables 1  and 4 show that the share of domestic demand and its contribution to growth have decreased during the last period.Conversely, the share of net exports and its contribution to growth have increased.
The conclusion is that there is no evidence that the net export position of the selected countries deteriorated during the last period.And as a consequence, there is no evidence that growth during the last period was domestic demand-led and at the expense of the net export position.

V. DECOMPOSITION ANALYSIS OF STANCES IN THE PRIVATE, GOVERNMENT, AND TRADE SECTORS: AN EARLY WARNING SYSTEM
In this section we analyze the stances of the private sector, the government (fiscal) sector, and trade sector for the five selected countries.The technical details of the aggregate demand decomposition analysis are provided in the Appendix.The private sector stance or direct "own" multiplier on output is given by (I p / s p ) where I p denotes gross private investment and s p is the savings rate out of GDP.If (I p / s p ) is larger than GDP, then private investment is larger than private savings (or, alternatively, private disposable income is smaller than private spending, composed of private consumption and private investments).Under these circumstances, the private sector is exhibiting an "expansionary stance" on aggregate demand, i.e., demand injections are larger than demand leakages.The government or fiscal stance is (G / t) where G is government spending and t is the tax effort out of GDP.If (G / t) is larger than GDP, then government spending is larger than tax revenues, and the government will exhibit an expansionary stance on aggregate demand, i.e., it exerts positive net injections on aggregate demand.Finally, the external sector stance is (X / m), where X denotes exports of goods and services and m is the propensity to import out of GDP.If (X / m) is larger than GDP, exports exceed imports, and the trade or external sector is exhibiting an expansionary stance on aggregate demand, i.e., export injections exceed import leakages.The period covered in this analysis is 1983 to 2003, using real values in the national income accounts for aggregate demand components.
As is shown in the Appendix, the sum of net injections of the three sectors is zero.This implies that a surplus in any due sector must be compensated with a deficit in at least due of the other two sectors.While small deficits can be sustained for considerable periods of time, large deficits will demand corrections, more drastic the larger the deficit.It is in this sense that this decomposition analysis provides an early warming system.
The results are presented in Figures 1-5, which plot the stances of the three sectors vis-à-vis GDP.
The PRC's slow transformation into a market economy and its participation in world trade has brought almost uninterrupted high growth to the country from the late 1970s until the present.The very high private saving rates (above 35 percent) have allowed the private sector "stance" to be nonexpansionary throughout most of the period 1983 to 2003, while maintaining a very high share, as well as growth, of GDCF (discussed in the previous section).Since the early 1990s, the fiscal stance has been expansionary.The external stance became expansionary starting 1990 and has remained positive until the present.The major expansionary stances in the last period came from the government and external sectors (Figure 1).1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 GDP Private sector stance

Fiscal stance External stance
In recent years, the Government of the PRC has reduced the expansionary fiscal stance in an attempt to avoid overheating of the economy.This explains why the external sector has emerged as the leading expansionary sector in recent years.
As with the PRC, relatively high private saving rates in India, a lowincome country, have allowed a nonexpansionary private sector stance and supported a GDCF of around 20-25 percent of GDP during most of 1983-2003.Figure 2 shows the consistent nonexpansionary stance of the private sector.This sector's stance falling below GDP seems to be widening in recent years as the private saving rate is close to 30 percent of GDP.1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 GDP Private sector stance

Fiscal stance External stance
Imports have been increasing since the 1990s but at lower rates than in the other countries.Export growth, however, has outpaced import growth in recent years leading to smaller negative net exports and to a small nonexpansionary external stance.
The very low tax effort (below 10 percent of GDP during most of the 1983-2003 period) and high government spending has made the government the only sector with an expansionary stance.This impact of the fiscal expansionary stance on aggregate demand, though large and increasing in recent years, is moderated by the growing gap between GDP and the private sector stance, and the improvement in the external stance.
Figure 3 shows that an expansionary private sector stance and nonexpansionary external stance during 1983−1985 were reversed in the second half of the 1980s.This shift to an expansionary external stance took place at the time the optimism about the East Asian "miracle" was at its height.This high foreign exchange earning capacity of the country was an important component of the country's success.1983 1985 1987 1989 1991 1993 1995 1997 1999  The appreciation of the won, high short-term capital inflows, speculative bubbles, and the fixed exchange rate regime of the 1990s, however, brought back an expansionary private sector despite the country's very high private saving rates.This was accompanied by a reversal to a nonexpansionary (and at times contractionary) external stance between 1990 and 1997.This contributed to the loss in confidence in Korea in the period right before the Asian crisis.
As Korea got enmeshed in the Asian crisis, the deep recession and sharp currency depreciations in late 1997 and throughout 1998 effected a sharp reversal, with the private sector stance shifting sharply from expansionary to highly contractionary, and the opposite in the case of the external stance.This situation continues until now, though quite subdued compared to the situation in 1998-1999.In recent years, Korea has been experiencing difficulties in increasing its GDP growth rate due to weak consumption demand.Korea's fiscal stance has historically been nonexpansionary except in 1998-1999 as a result of the Asian crisis.Therefore, the only expansionary sector in Korea in the post-Asian crisis period is the external sector as net exports remain significantly positive.
The Philippines's economic history since the 1980s has been marked by alternative periods of growth and recessions.The sharp recession and crisis in the mid-1980s caused a sharp reversal in the private sector stance from highly expansionary in 1983 to highly contractionary.Correspondingly, the contractionary external stance in 1983 turned into expansionary in 1985-1988.Economic recovery in the late 1980s brought the private sector increasingly back to very positive territory in the 1990s, despite that 1990-1993 were years of stagnation.The most expansionary period of the private sector was 1993-1997.Accompanying the high expansionary stance of the private sector were increasingly negative net exports, which returned starting in 1989, and rapidly increased in the 1990s (reaching more than 10 percent of GDP).
The Philippines was also hit by the Asian crisis in the second half of 1997 and throughout 1998.The sharp currency depreciation, initial high interest rates, and slight recession tamed the high expansionary stance of the private sector (making it briefly contractionary in 1999 and 2000) and brought net exports to positive territory in 1999 and 2000.
The ensuing economic recovery (though weak and slow) returned the private sector to expansionary territory and the external sector to contractionary in recent years (2001 to 2003), but at much lower levels than before the Asian crisis.
High government injections and deficits in the mid-1980s were met with fiscal austerity in 1987-1992 due to debt overhang as the country joined the decade-long debt crisis that afflicted most Latin American countries during 1982-1992 (Philippine fiscal deficits remained high during 1987-1992, but this is not reflected in Figure 4 because much of the government spending was due to debt payments, and net lending and bailout of government corporations).Fiscal surpluses were attained in 1994-1997 but these were reversed in 1998 due to the Asian crisis.The Philippines faces another fiscal crisis as the tax effort has continued its decline after the Asian crisis, and as debt payments and failing government corporations (especially the National Power Corporation) are absorbing much of government spending.The fiscal stance turned to expansionary starting in 1999, but not markedly since government spending has been largely concentrated on debt payments and net lending to failing government corporations.1983 1985 1987 1989 1991 1993 1995 1997 1999  Summing up, the recent years in the Philippines have been marked by an expansionary stance in the private and fiscal sectors, and a contractionary one in the external.But the levels are much lower in this latest period than before the Asian crisis.
Like Korea, Thailand's private sector stance moved from expansionary to nonexpansionary in 1985-1987, and its net exports turned from negative to positive in 1986 (Figure 5).Net exports turned slightly negative in 1988.Thus, like Korea, Thailand's peak of being an Asian "miracle" came during the second half of the 1980s, when its net exports were either positive or small negative figures, and its private sector was not too expansionary.1983 1985 1987 1989 1991 1993 1995 1997 1999  Also like Korea, currency overvaluation, high short-term capital inflows, speculative bubbles, overlending and overborrowing, and a fixed exchange rate regime brought the private sector's stance into significantly expansionary starting in 1989, and continuously strengthened during 1990-1996.Correspondingly, its external stance turned into significantly contractionary throughout the 1990s, especially in the few years before 1997, the outbreak of the Asian crisis (which originated in Thailand).
The fiscal stance was largely nonexpansionary during 1987-1995.More so than in Korea, Thailand had very severe private sector and trade sector adjustments during the Asian crisis and its aftermath.The private sector became very highly contractionary, especially in 1999 and 2000, and remains significantly negative until now.The external stance turned into very highly expansionary, especially in 1998 and 1999, and has remained this way in recent years.
The fiscal stance turned expansionary (with fiscal deficits) during 1997-2000 because of a decline in the tax effort and social and economic spending due to the Asian crisis.These were restrained in 2002 and 2003 as tax efforts improved (unlike in the Philippines, where the tax effort continued its decline until the present).
There was a continuous increasing import propensity from the mid-1980s to the present, with a short respite in 1998 because of the crisis.But export growth has outpaced import growth in the post-Asian crisis period.Thus, like in Korea, 2002 and 2003 saw Thailand's trade sector as the only one providing significant expansionary stance to aggregate demand.It must be noted, nevertheless, that reduction in the nonexpansionary stance of the private sector during the last few years was to some extent the result of Thai Prime Minister Thaksin's policies.For the time being, the private sector stance is still nonexpansionary.However, if it becomes an overexpansionary stance, then one must be cautious that the situation does not revert to that of the precrisis period, that is, highly expansionary private sector stance leading to significant trade deficits financed by large foreign borrowings (very vulnerable to interest and exchange rate shocks).
Therefore, it must be pointed out that, despite the attempts of Mr. Thaksin at switching from export-led to domestic demand-led growth, net exports still provide a key ingredient to Thai growth, while the private sector and fiscal stances-the domestic demand sectors-have been actually nonexpansionary in the latest years.If anything, Mr. Thaksin's policies must be seen as an attempt at increasing aggregate output vis-à-vis aggregate demand (domestic absorption).If one thinks of net exports (X-M) equivalently (through the National Accounts) as the difference between aggregate output (GDP) and domestic absorption (the sum of consumption plus investment and plus government expenditures), it seems that the Thai government's five-pronged strategy (see Box 1) aims at boosting output rather than demand.
We are now in a position to provide an answer to the three questions posed at the beginning of the paper.

(i)
Is there Evidence of Demand-led Growth Replacing Export-led Growth The answer to this question is a clear No.The external sector is the one with the strongest expansionary stance in recent years in three out of the five countries studied, namely, PRC, Korea, and Thailand.For Korea and Thailand, it is the only sector providing an expansionary stance.For the PRC, the government is very consciously reducing its expansionary stance to avoid overheating the economy.Since its private sector has historically exhibited a nonexpansionary stance (due to the country's high savings), the trade sector provides a major force in the expansion of aggregate demand. 9n India, the high fiscal expansionary stance is growing, but growing nonexpansionary and offsetting pressures from the private sector and improving net exports (though still negative) are reducing this expansionary domestic demand pressure on aggregate demand.
In the Philippines, the post-Asian crisis led to a return to expansionary stances in the private and fiscal sectors, and negative net exports.But the expansionary stance of the private sector and negative net exports are substantially lower than before the Asian crisis.
In the cases of India and the Philippines, which exhibit contractionay external stances and expansionary stances in the private and/or government sectors, the former have diminished in recent years, i.e., the joint expansionary stances of the domestic private and fiscal sectors are smaller in recent years.

(ii) Did an Export-led Strategy Contribute to the Asian Crisis?
Again, the answer to this question is a clear No. Korea, Philippines, and Thailand followed a growth strategy characterized by a bias against exports during the years before the Asian crisis.This bias has been well documented and consisted of overvaluation of the currency, overlending and overborrowing for the domestic sector, and creation of speculative bubbles in the nontradeable sectors.This resulted in highly negative net export positions, and the exaggerated expansionary stance of the private domestic sector.For Korea and Thailand, this hurt the strong East Asian miracle image they had achieved in the second half of the 1980s.The Asian crisis and its aftermath has been a painful reversal of the earlier situation in these three countries.
These results directly contradict the arguments of Palley (2002) presented earlier, i.e., that the export-led growth strategy was partly to blame for the East Asian crisis and led to biases against the domestic demand sector.In fact, our simple analyses have shown that it was an overexpansionary stance in the private sector and growing trade deficits that marked the immediate period before the Asian crisis for Korea, Philippines, and Thailand.This reflected the policies of overvaluation of the currency, overlending and overborrowing in the domestic private sector, and speculative bubbles (in the nontradeable sector) that most scholars agree influenced the loss of confidence that triggered the Asian crisis.
(ii) Are there Any Lessons to be Learned from the Five Countries Experiences?
The most obvious result coming out of the analyses is that the "best" periods for our selected countries have been those when both domestic demand and net exports exhibited impressive growth.This corroborates the earlier justifications for export-led growth, especially the argument that developing countries need precious foreign exchange to finance their import needs.It must be pointed out that this corresponds to the definition of domestic demand-led growth weakly speaking (both domestic demand and net exports are increasing).The PRC has demonstrated that this kind of growth can be sustained for long periods.India just adopted this kind of strategy in the late 1990s, as its high domestic demand growth is accompanied by impressive export growth and improvements in its trade deficits.Korea and Thailand followed this strategy in the second half of the 1980s, when their reputation of East Asian Tigers was at a peak.A deviation from this strategy-accompanied by the overvaluation of the currency, the overlending and overborrowing syndrome, and the rise of speculative bubbles in the nontradeable sector-seemed to have led them toward the Asian crisis.Our analysis indicates that they actually have reverted to the earlier strategy of promoting both domestic demand and net export components of the economy during this postcrisis period.

VI. COMPARING EXPENDITURE SHARES OF OPEN EUROPEAN COUNTRIES AND THE ASIAN AND PACIFIC REGION
For comparison purposes, we now analyze the expenditure shares of a group of small, open developed economies in Western Europe (Belgium, Denmark, Netherlands, Sweden, and Switzerland), and compare them with those of the developing countries of the Asian and Pacific region.
We divide the Asian and Pacific countries into three groups: upper-income economies (UA), middle-level income countries (MA), and low-income countries (LA).The low-income countries (LA) coincide with the World Bank's latest categorization of low-income countries in its latest World Development Report.The middle-level income countries of Asia (MA) coincide with the countries that are under the World Bank's categorization of lower middle income countries.The upper income economies in Asia (UA) are those countries in the Asian and Pacific region that are above the income brackets for the lower middle income countries as categorized by the World Bank.
Figures 6 and 7 show the average shares of exports and imports, respectively, in GDP for the European countries, UA economies, MA countries, and LA countries.indicates that the European and UA economies have significantly higher shares of exports in GDP than the MA and LA countries.The UA economies have by far the highest share of exports among all countries and their export share has increased the most between 1983 and 2003.For all categories of the indicators, distorting the averages).LA countries (LIC) include Azerbaijan, Bangladesh, Bhutan, Cambodia, India, Indonesia, Kyrgyz Republic, Lao PDR, Mongolia, Nepal, Pakistan, Papua New Guinea, Tajikistan, and Viet Nam.The World Bank categorizes the countries according to 2002 gross national income (GNI) per capita using the World Bank Atlas method.LA countries (which coincide with the World Bank category of low-income countries) are countries with $735 or less; MA countries (which coincide with the World Bank's category of lower middle-income countries) are countries with GNI per capita of $736-2935, UA economies are countries in the Asian and Pacific region with more than $2,935 GNI per capita.All the selected European countries fall in the World Bank's category of high-income countries, with $9,076 GNI per capita or more.For the following countries 2002 data were used due to lack of 2003 data: Bhutan, Fiji, India, Lao PDR, Papua New Guinea, Tajikistan, Vanuatu.Sources of data were from Key Indicators (ADB various years), International Financial Statistics (IMF various years), UN Statistical Division, and World Bank Country Profiles.countries, the shares of exports and imports have grown fast between 1983 and 2003.Changes in the import share are quite different (Figure 7).The European countries had higher import shares than the MA and LA countries in 1983, but in 1993 both the MA and LA countries had exceeded the import share of the European countries.In 2003, the MA countries' import share still exceeded that of the European countries, but the share of the LA countries again fell below that of the European countries.The UA economies again have had the highest share of imports since 1983, and their import share is also growing the fastest.The above analysis shows that UA and MA countries-where most of the emerging markets in the Asian and Pacific region are categorized-have, on average, improved their net exports position between 1993 and 2003, and that their share of domestic demand has also been declining.The UA economies are even outperforming the European countries in terms of exports and net export shares.
The decreasing share of the domestic demand components in the UA and MA countries is borne out in Figures 9 and 10, which graph the average shares of consumption (private plus government) and gross domestic capital formation (or gross domestic investments), respectively, to GDP.
Figure 9 indicates that consumption shares fell in all groups of countries between 1983 and 2003.It is clear that although the MA and LA countries have higher consumption shares than the European and UA countries, the shares are decreasing more quickly over time in the first two groups of countries.The UA economies have the smallest share of consumption, even lower than that of the European countries.For gross domestic capital formation (GDCF), Figure 10 shows that all the categories of Asian countries have, since the early 1980s, higher shares of investment to GDP than the European countries.The latter have very stable gross investment shares of between 18 and 19 percent of GDP.The UA economies had the highest investment share in 1983 and 1993.But this share, as well as that of the MA countries, fell during the 1993-2003 period, with the UA economies' investment share losing almost 10 percentage points.The LA countries, on the other hand, increased their investment share between 1993 and 2003.
Thus, on average, there is no indication that strong domestic demand or consumption-led growth has been taking place in the developing countries of the Asian and Pacific region during the last period.The shares of consumption and GDCF have declined during the last period in the UA and MA countries (i.e., the share of domestic demand has declined, which means that the net exports share has improved).The LA countries' consumption share also fell, but their GDCF share increased.This is a positive indication that the lower-income countries, which are capital-scarce, are accumulating capital at a faster rate than the other groups.The outstanding performance of the UA economies in terms of exports and net exports reinforces the general perception that these countries (Hong Kong, China; Korea; Malaysia; and Taipei,China) are some of the strongest export performers in the world.
What is worrisome is the very large negative net exports still plaguing the low-income Asian and Pacific developing economies.We shall elaborate upon this issue in the last section.

VII. SUMMARY AND CONCLUSIONS
A.

Summary of Results
In brief, the main conclusions of the growth accounting and stances analyses are as follows: (i) There is no evidence that the last period was marked by domestic demandled growth at the expense of net exports.On the contrary, the countries hit by the crisis, such as Korea and Thailand, lessened domestic demand expansion and strengthened net export growth.Domestic demand and net exports have been growing in countries not hit by the crisis, such as the PRC and India.(ii) In general, the Asian and Pacific countries were able to reduce their trade deficits during the last period, so that the share of net exports increased visà-vis that of domestic demand.(iii) There is no evidence that the export-led strategy contribute to the Asian crisis.On the contrary, the export-led strategy, as we have defined it, was not implemented during the period right before the crisis.This period was marked by more overexpansion in domestic demand, and deterioration of net exports.(iv) Periods when domestic demand was highly expansionary occurred at the same time that net exports signaled an ensuing crisis.This should serve as an early warning system.The periods when the countries analyzed performed the best were those when both domestic demand and net exports exhibited impressive growth.This corresponds to what we defined above as domestic demand-led growth weakly speaking.
Below are two more conclusions of normative nature.

B. There should be No Conflict between Growth in Exports and in Domestic Demand
Successful and sustained growth requires growth in both domestic demand and net exports.The demand-side growth accounting exercise and the decomposition analysis of stances from the private, government, and trade sectors provide some useful lessons for appraising the discussion of domestic demandled versus export-led growth.
First, growth of successful countries like the PRC, and to a lesser extent India, is based on a combination of both domestic demand componentsespecially GDCF-and exports.It is clear that developing countries should have adequate investment levels in order to grow and develop.There also has to be appropriate growth in consumption so that the population's welfare improves.These can be achieved at the same time as the country succeeds in developing and improving its export sector.In fact, in terms of technology deepening and learning by doing, growth in both sectors will be complementary and mutually reinforcing.
It is when overemphasizing one strategy at the expense of the other that the growth strategy becomes unbalanced, unstable, and problematic.Clearly, the growth strategies of Korea, Philippines, and Thailand in the 1990s (before the Asian crisis) overemphasized expansionary tendencies in the domestic private sector demand at the expense of net exports.This is reflected in the frequently discussed roots of the Asian crisis: currency overvaluation as well as overlending or overborrowing syndrome-spurred by inflows of short-term speculative capital-that brought high growth to the domestic and nontradeable sectors, and deterioration in the net export positions.
Conversely, the harsh adjustments undertaken by the Korea, Philippines, and Thailand during and after the Asian crisis saw recessions and a collapse of gross investment as the net export positions improved.There are prominent economists (Krugman 1999, Stiglitz 2002) who believe that the adjustments and policies imposed on the East Asian countries hit by the crisis were overly harsh, especially on domestic demand, and contractionary.Whatever side one takes, it is clear that the sacrificed growth and resulting decline in the growth of productive capacity in the Asian countries hit by the crisis is a harmful consequence of the strategy they followed (currency overvaluation, overlending, and overborrowing), which reversed the healthy balance and the desirable progression of both domestic demand (and capital goods sector) and the tradeable sectors achieved during the second half of the 1980s.

C. Countries with High Trade Deficits will Benefit from a More Open International Trade System and Export Promotion
Can all countries increase their exports at the same time?We finally address the question posed by Palley (2002), Blecker (2002 and2003), and those who contend that not all developing countries can achieve successful export-led growth, inasmuch as positive net exports and trade surpluses correspond to trade deficits in other countries, and as the markets of the weaker countries (mostly in the developed countries) are gobbled up by the richer, high-performance countries.
Countries with high trade deficits (concentrated in the low-income countries) will benefit from a more open international trade system and promotion of their exports through price and nonprice competitiveness.It must be pointed out that Figure 8 shows that even if the low-income Asian countries (LA) had high negative net exports in 2003, this position had not, on average, deteriorated from that of 1993, despite the high export growth of countries like the PRC, India, and other large countries that strengthened their export sectors in the 1990s.This is one encouraging sign, at least in the Asian and Pacific region.It must be added, however, that the net exports position of many countries may not have deteriorated too much due to the very large and growing trade deficits of the United States.Expected adjustments, especially through the depreciating US dollar, may correct this situation in the medium term.
There are some other encouraging signs.The fast growth and expansion of the PRC has quickly opened up a potentially large export market for other developing countries.This will benefit many Asian countries, and already has benefited Korea; Malaysia; Thailand; and Taipei,China.The task now is to expand the benefits to the middle-level and low-income countries in the Asian and Pacific region.India is another country that has been growing fast in the last period.Its opening up to the world trade market has also opened a large export market.
The breakdown of trade talks in Cancun in 2004 also points to the strong need to push for trade reforms in developed countries to allow more agricultural, industrial, and service imports from the developing world.
The conclusion is that for an export-led development strategy to cover as many countries as possible a more balanced and equitable growth in exports and imports across the world is required.This in turn necessitates that: (i) all countries, including richer and trade-surplus nations, open up their markets to poorer countries; (ii) extra efforts from the poorer and latecomer countries promote their export sector via price and nonprice competition, and develop the necessary technological, physical and human infrastructure to be competitive.The first one obviously requires the cooperation and participation of rich and trade-surplus countries so that developing countries can access the big world markets and reduce their trade deficits with the surplus countries.Trade liberalization of poor and trade-deficit countries alone (without the opening of the markets of rich and trade-surplus countries) will obviously lead to perverse results.The latter requires twin growth in the domestic demand and tradeable sectors inasmuch as much of this infrastructure building will be part of domestic demand.
A more balanced and equitable international arrangement in world trade will lead to smaller trade surpluses and smaller trade deficits across countries in the world as more developing countries will be able to share the benefits of international trade.

APPENDIX DECOMPOSITION ANALYSIS: INJECTIONS AND LEAKAGES
The methodology used follows Godley (1999).Consider the definition of GDP or Y as:

Figure 10 .
Figure 10.Average Share of GDCF to Nominal GDP: 1983, 1993, 2003 Figure 10.Average Share of GDCF to Nominal GDP: 1983, 1993, 2003 , T denotes net taxes, denotes private consumption, denotes private investment, G denotes government spending, X denotes exports, and M denotes imports.whereI is total gross domestic investments or gross domestic capital formation.
using UN Statistics Division data, which has a continuous series of expenditure component measures from 1973 to 2002 in constant 1990 prices.Data for 2003 was derived from the 2002 data above and the latest growth data from ADB's Key Indicators for Asia and the Pacific 2004 (ADB 2004) or the IMF's International Financial Statistics.For the Philippines, the UN Statistics Division has a complete continuous series from 1973 to 2003.India does not have data for 2003 as of December 2004), so its data only covers up to 2002.

Table 1 . Shares of Expenditure Components in Real GDP (1990 Prices)
a India's 2003 data not yet available as of November 2004.Sources: UN Statistics Division, Key Indicators (ADB 2004).

Table 2 . Average Growth Rates of Expenditure Components Based on Constant (1990) Prices
a India's 2003 data not yet available as of November 2004.Sources: UN Statistics Division, Key Indicators (ADB 2004).

Table 4 . Contribution of Demand Components to GDP Growth
Sources: UN Statistics Division and Key Indicators(ADB 2004).

Table 5 . Phases of Domestic Demand-led and Net Export-led Growth in Selected Asian Countries
s is the savings rate; and where M=mY where m is the propensity to import.From the above, one can obtain the Keynesian income multiplier as:>Y, itimplies that the private sector is injecting more demand into the system than incurring demand leakages.Similarly, if ( ) / G t >Y, it implies that government spending exceeds government taxation, so that the government is exerting positive net injections to aggregate demand.