The Slowdown of the US Economy and Asia

A. The Subprime Crisis in the US and Its Impact on Asia

A decline in the US housing market ignited hedge fund failures in the summer of 2007, with the US credit market seizing up as more losses of unknown magnitude were expected. The financial distress of the US economy immediately spread to financial markets all over the world with the Asian financial markets no exception. In this sense, the Asian financial markets are not decoupled from the US financial market, meaning that a shock to the US financial market greatly influences the Asian financial markets.

There has been a debate about whether the subprime crisis will be confined to the financial sector or eventually lead to a recession in the real economy. More researchers are now supporting the view that the subprime crisis is driving the US economy into a recession. For example, Eichengreen (2008) argued that “the U.S economy is undoubtedly experiencing a sharp growth slowdown.” The World Economic Outlook (WEO) Update (2008) also predicts that the slowdown in global growth is expected to continue through the second half of 2008.

How large will the impact of the US real economy's slowdown be on Asian economies? While it is clear that the financial markets in Asia are heavily influenced by the turmoil in the US financial market, how to assess the real economy's recession in the US will affect Asian economies is a separate issue. In this paper, I focus on whether Asia's real economy is decoupled from the US real economy or not.

How do one economy's fluctuations affect another economy's? This question is answerable by investigating two linkages between economies, those of trade and finance. I will review the implications of the two linkages and introduce a recent empirical study on the evidence.

B. Trade Linkage and Its Implications for Decoupling¹

One important linkage through which a shock on one economy can be transmitted to another is trade linkage. Since the early 1990s, the volume of world trade has increased twice as fast as that of world gross domestic product (GDP). As this pace of trade integration continues, its impact is expected to also increase.

While a number of researchers agree that trade linkage must play a crucial role in transmitting disturbances from one country to another, at least theoretically, there is no consensus on whether increased trade would lead to a greater or smaller degree of comovements across countries. On one hand there is an important theoretical reasoning that deeper trade linkages result in less synchronization of business cycles. For example, Kenen (1969), Eichengreen (1992), and Krugman (1993) all argued that if countries are more specialized in industries with comparative advantage, as long as a shock in a particular industry is less likely to be transmitted to different industries, more trade integration leads to fewer synchronized fluctuations.

On the other hand, however, there is evidence that recent trade increases are mainly driven by intra-industry trade rather than inter-industry trade as production fragmentation and outsourcing becomes the major source of trade explosion. According to a recent study by Jones (2006), trade in parts and components in the last decade of the twentieth century grew by an average of 9.1% a year, even faster than the rate of growth for overall trade.

If this is the case, then as Frankel and Rose (1998) argued, business cycles would become more positively correlated as trade integration progressed. In particular, this is more so if business fluctuations are dominated by industry-specific technological shocks. Additionally, there is a well-known and important argument that supports the positive transmission of a shock from one economy to another. If a shock drives one country to a boom that increases demand from foreign countries as well as domestically, the effects may spill over to trading partners through an increased volume of imports.

In sum, the theoretical implications of trade integration on how a shock in one country would be transmitted to another are ambiguous. Hence, an empirical investigation is in order. Canova and Dellas (1993), in one of the earliest attempts made in this area, found that while the choices of the de-trending method matter, in general there was some evidence that more trade integration leads to positive transmission of disturbances across countries. More recently, Frankel and Rose (1998) found that, based on a study of 21 industrialized countries, the more countries traded with each other, the more highly correlated their business cycles were. Following a similar method, Choe (2001) also confirmed, based on research done on 10 East Asian countries, economic fluctuations are more synchronized as trade interdependence deepens in the region. Recently Shin and Wang (2003) more directly tested the driving force of positive impact of trade integration on business cycle synchronization and found that intra-industry trade, rather than trade by itself, plays a crucial role. Calderón, Chong, and Stein (2007) extended the analyses to include both industrialized and developing countries and found that the impact of trade on the co-movement is higher among the former than the latter. They also found that the response of output correlation to trade linkages is especially higher when intra-industry trade is more pronounced.

The fact that more trade integration reinforces transmission of shocks across trading partners has an important implication for East Asia, especially in understanding the impact of the recent slowdown of the US economy on East Asian economies. Traditionally, the US market has been an important outlet for exports from East Asian countries. However, the importance of the US market has become substantially lowered. Instead, trade integration among East Asian countries has been greatly enhanced.

Following the methodology used in Park and Shin (2008), Table 1-1.A shows how the trade intensity measure has evolved over time in Europe and East Asia.² They divided the world economy into three blocs—the US, East Asia, and European Union (EU)—and defined the trade intensity measure between an individual country i and any of the three blocs (i,b) by normalizing exports (imports) of country i with bloc b by total exports (imports) of country i.:

where x_ibt denotes total nominal exports (US$ value) from country i to bloc b (b=US, EU, and East Asia) during period t; m_ibt denotes total nominal imports (US$ value) from bloc b to country i during period t; and X_it and M_it denote total global exports and imports of country i during period t.

Table 1-1.A [ PDF 15.5KB | 1 page ] shows the export intensity results for East Asia. The whole sample (1990:1– 2006:IV) is divided into three subsamples: period I (1990:1–1996:IV), period II (1999:I– 2002:IV), and period III (2003:I–2006:IV). As expected, for every individual East Asian country, the trade share with the East Asia bloc (the last column) is the highest. This is especially the case for Hong Kong, China; Indonesia; Malaysia; Singapore; and Taipei,China—in period 3, the share is either over or close to 0.6. In contrast, it is lower than 0.5 for four countries including People's Republic of China (PRC), Japan, and Republic of Korea (hereafter Korea), the area's three largest economies. When we calculate the unweighted and weighted average export shares of the East Asia bloc, they are 53.7% (unweighted) and 49.7% (weighted) in period 3.

On the other hand, the trade share with the US decreased from period 2 to period 3 in every country but the PRC. The average export share of the US is 17.2% (un-weighted) and 18.7% (weighted) in period 3, which is much smaller than that of the East Asia bloc. The two countries for which the export share of the US is over 20% are the PRC (22.7%) and Japan (23.3%).

The extent of regional trade in East Asia is even more remarkable if we compare the achievement of East Asia with that of Europe. Table 1-1.B [ PDF 15.6KB | 1 page ] shows the export trade shares for European countries. For every individual EU country, again the export share with other countries in the EU bloc is the highest, which shows strong regional trade integration in Europe. However, for a number of countries, it actually decreased over time. The average export share of the EU is higher in period 1 (weighted: 55.7%; or un-weighted: 60.1%) than in period 3 (weighted: 53.5%; or un-weighted: 53.9%). This finding suggests that the decreasing trend continued even after the euro was introduced in 1999, which is quite surprising. Further, the weighted average share of the EU in period 3 is comparable to the intra-regional export share in East Asia. Given that an East Asia-wide free-trade agreement (FTA) or common currency is yet to be established, it is quite notable that East Asia countries have achieved this degree of intra-regional trade integration.

In general, EU countries' dependence on the US market is lower than East Asia countries' but has been increasing. The average export share of the US is 6.4% (un-weighted) and 7.3% (weighted) in period 1, and 8.2% (un-weighted) and 8.7% (weighted) in period 3. Overall, however, compared to East Asia countries, EU's exports are more diversified over the world.

Table 1-2.A [ PDF 15.8KB | 1 page ] shows the import shares for East Asia. Generally the findings are similar: for every country, the import share of the East Asia bloc is largest. It also increases over time for most of the countries. For every country, the US' import share also decreases, at least from period 2 to period 3. It is interesting to note that the PRC's reliance on the US is also decreasing. On average, the US' import share decreased from 16.1% (un-weighted) and 17.1% (weighted) in period 1 to 11.0% (un-weighted) and 10.4% (weighted). Table 1-2.A also shows the import shares for EU. Again, the results are similar as those of the export share.

The evidence so far indicates that the intra-regional trade integration in both East Asia and Europe is quite high. In this sense, it is likely that business cycle fluctuations are more regionally synchronized. We also found that the dependence of both regions on the US market is considerably lower than their dependence on regionally close economies. While EU countries' dependence on the US is generally lower than East Asia countries', one interesting characteristic for East Asia countries is that its dependence on the US has a clear negative trend. This implies that the influence of the US economy on East Asian countries resulting from trade integration is getting weaker.

Trade integration among East Asian countries is likely to drive their business cycles, decoupling them from those of the US. A few caveats are, however, in order in interpreting in this way. First, as illustrated above, the dependence of Japan and the PRC, the two largest economies in East Asia, on the US market is still quite large. In particular, the PRC's dependence on the US increased from period 1 to period 3. Second, while not explicitly shown in Tables 1-1.A and 1-2.A, the inter-regional trade integration in East Asia is getting deeper as other East Asian countries' exports of parts and components to the PRC increase.³ To the extent that the PRC plays a role of assembling imported parts and components and exporting the final goods to the US, a shock to the US cannot affect only the PRC but also indirectly affect the rest of East Asia.

C. Financial Linkage and Its Implications for Decoupling⁴

Another important avenue through which a shock is transmitted from one country to another is financial linkage. In the literature, at least three channels have been emphasized. First, capital flows, induced by return differentials, tend to move to countries with positive shocks from those with adverse shocks. Hence, by pulling out capital from adversely affected countries, deeper financial integration aggravates the countries' economies further while, at the same time, by pouring capital to booming economies, may overheat them. Hence, financial linkage can contribute to generating asymmetric fluctuations of business cycles.

Second, however, in some other circumstances, financial linkage can lead to more synchronized business cycles. For example, if a country facing shortage of liquid assets takes out capital from another country, a shock in the former country's financial market can be transmitted to the latter country's financial market in the same direction. To the extent that the shock in the financial market affects both countries' real sectors, fluctuations of the real economies are synchronized as well. Further, Claessens, Dornbusch, and Park (2001); Calvo and Reinhart (1996); and Cashin, Kumar, and McDermott (1995) argued that capital flow can generate business cycle co-movements for the countries in the same area that experience in- and outflows of capital at the same time. For example, during the Asian crisis and the Latin American crises, a number of countries in the same area facing outflow of capital simultaneously suffered recession.

Third, for a longer term view, better risk-sharing attained through greater financial market integration induces higher specialization of production and, hence, larger asymmetric shocks across countries. In other words, better income insurance provided by risk-sharing across countries enables each country to take more risk by specializing in specific industries. Therefore, as long as business cycles are driven by industry specific shocks, more financially linked countries tend to face less synchronization of business cycles.

The above arguments show that, at least theoretically, financial integration also does not lead to an unambiguous conclusion on the business synchronization. There is one difference, though, between trade and financial integration. That is, due to transportation costs, while trade integration is likely to progress among regionally close countries, since financial assets are weightless, there is no clear reason for financial integration to track the same route. In fact, as emphasized by recent studies, among others, Kim, Lee, and Shin (2008), East Asian economies have strong ties with the global financial markets such as the US markets.

Table 2.A [ PDF 15.2KB | 1 page ] provides the geographical distribution of total portfolio investment asset holdings for East Asian and European countries in 2003. The data are obtained from the Coordinated Portfolio Investment Survey (CPIS), published by the International Monetary Fund (IMF). From the table, it is clear that the degree of intra-regional financial integration is much lower in EA than in Europe. While European countries hold on average 58% of their portfolio assets within Europe, the share of intra-EA asset holdings is about 14% on average for eight East Asian economies. Interestingly, among European countries, the UK, one of the global financial centers, holds the largest share positions in East Asia (11%) and the lowest intra- Europe share (42%).

In contrast, the share of US assets is about 40% on average for East Asian countries, which is much larger than the US share in Europe, which is 17%. Especially, Japan (36%), Korea (46%), Philippines (69%), and Thailand (64%) hold a larger share of their portfolio assets in the US. The share of European assets is about 25% on average for East Asian countries, which is also larger than their intra-regional share. Hong Kong, China (27%); Japan (35%); Malaysia (23%); and Singapore (39%) hold a larger share of their portfolio assets in Europe. These figures clearly show that a number of East Asian countries have stronger ties with the US or Europe rather than among themselves.

Table 2.A also presents data on the size of total international portfolio asset holdings. The largest foreign investor is the US, whose holdings amount to about US$3.1 trillion, followed by the UK, about $1.7 trillion. In East Asia, Japan (US$1.7 trillion); Hong Kong, China (US$335 billion); and Singapore (US$144 billion) are the major investors. The other five East Asian countries hold a much smaller size of assets, on average, of about US$6 billion. In comparison most European countries hold a much larger size of assets.

In Table 2.B [ PDF 17.2KB | 1 page ], I have updated the Kim, Lee, and Shin (2008) table to the latest available year, 2006. Now we can see that the intra-regional share in East Asia increased substantially to 21.7%, which is slightly larger than the US share, 20.7%. However, the decrease in the US share is mainly due to relatively smaller investors such as Indonesia, Philippines, and Thailand. The major investors such as Hong Kong, China; Japan; and Singapore decreased their US share only slightly, by 3%, 2%, and 0.5%, respectively, indicating that the reliance on the US financial market in terms of East Asia's aggregate asset holdings is still eminent.

In contrast, the geographical distribution of total portfolio asset holdings for Europe did not change much. Their intra-regional financial integration is still strong, the average Europe share being about 58%. The UK's regional share is continuously lowest among European countries, amounting to 37%.

D. Trade and Financial Integration and Their Impacts on Business Cycle Comovements⁵

The results from the previous section indicate that East Asian countries are being significantly integrated in terms of trade but not as much in terms of finance. How will this feature, i.e., the intra-regional trade integration and extra-regional financial integration, affect the pattern of business cycles in EA? Has there been a divergence between cyclical changes in EA and the US? In this section, I will try to answer these important questions by briefly summarizing the findings in Park and Shin (2008).

In order to answer the above questions, Park and Shin (2008) investigated how trade and financial integration affects the influence of movements of the three blocs' business cycles on the business cycle of each individual East Asian and European country. More specifically they constructed the following equation for EA:

where is the cyclical components of output for East Asian country i and and are the cyclical components of output for the three blocs, the US, Europe, and East Asia. The cyclical measures are obtained by applying the Hodrick-Prescott filter. The output measures of the bloc are simply weighted average of output, where the GDP size is used as weights. To eliminate the influences of country i on its own bloc (East Asia), country i is excluded in calculating the cyclical component of the East Asia bloc, which is denoted by the superscript EA-i. Similarly they constructed another equation for Europe:

where is the cyclical components of output for European country j and its influence is eliminated in calculating the cyclical component of the European bloc.

By estimating equations (1) and (2), they defined the business cycle co-movement measures of output for an individual country vis-à-vis the US, EU, and East Asia blocs as the estimates of the three coefficients, , respectively.

In order to investigate how the degree of business cycle co-movements has evolved over time, the authors divided the whole sample into three subsamples: period I (1990:I–1996:IV for East Asian countries and 1990:I–1998:IV for EU countries), period II (1999:I–2002:IV), and period III (2003:I–2006:IV). For East Asian countries, to avoid the influence of the financial crisis, they eliminated the financial crisis period (1997:I–1998:IV).

Table 3 reports the degree of business cycle co-movements for East Asian countries (Table 3.A [ PDF 13.7KB | 1 page ]) and European countries (Table 3.B [ PDF 13.3KB | 1 page ]). We find that the business cycles of most East Asian countries are quite synchronized, especially after the Asian crisis.⁶ In fact, countries such as Korea; Malaysia; the Philippines; Singapore; and Taipei,China show very strong ties of business cycle co-movements with the East Asia bloc in period 3. In contrast, the PRC and Japan, the two largest countries accounting for 70 % of East Asia bloc's output, show much weaker and even negative co-movements with the East Asia bloc in period 3. The degree of business cycle co-movements of East Asia countries with the US is generally lower than that with the East Asia bloc. The PRC is an exceptional case where its business cycles are more synchronized with the U.S than with the East Asia bloc in all the three periods.

In Table 3B, we also find evidence, but somewhat stronger than in East Asia, that business cycles of EU countries are also synchronized. Especially business cycles of EU countries such as Austria, Finland, France, Germany, Italy and Netherlands demonstrate strong comovements with those of the EU bloc. Further, in general, the average measure of business cycle co-movements with the EU bloc is higher than that with the US bloc or that with the East Asia bloc. However, we did not find any strong evidence that the business cycle comovements are getting stronger. In fact, the average co-movement measure, whether weighted or un-weighted, is highest in period 2 and lowest in period 3.

Park and Shin (2008), then, investigated how the co-movement measures for each individual country are influenced by the progress of trade and financial integration. Their findings are summarized as follows. First, they found strong evidence that deeper trade integration reinforces output co-movements. Further, they found that, while trade integration strengthens business cycle co-movements in both East Asia and Europe, this effect is stronger in Europe. Second, they found that while financial integration also contributes to business cycle co-movements, its impact is much weaker. The influence of financial integration, if any, on business cycle co-movements is also larger in the EU.

E. Discussion

The results in Park and Shin (2008) suggest that business cycles of East Asia countries are becoming more synchronized, possibly leading the region to decouple from the US. The driving force behind this scene is the deepening regional trade integration. Recently, intraregional trade exploded in East Asia, and the extent of regional trade integration reached a comparable level to that of the EU region, a scenario which should contribute to synchronized business cycles within the region. Interestingly, financial integration of East Asian countries is more pronounced in their ties to the global financial markets such as those in the US and the UK However, since the impact of financial integration on business cycles is much weaker than that of trade integration, stronger financial integration of East Asian countries with the US does not necessarily keep East Asia countries from decoupling from the US.

Some caveats are necessary to interpreting the results of Park and Shin (2008) in the above way. First, as stated already, the two largest countries, the PRC and Japan, do not show clear tendency toward more trade integration with other East Asian countries. In particular, the intra-regional share of the PRC's exports has been decreasing and the US share has been increasing. Athukorala (2005) showed that international product fragmentation—the cross-border dispersion of component production/assembly within vertically integrated production processes—is an important source of the deepening trade integration among East Asian countries. Since the final destination of assembled goods is more likely to lie in other regions such as the US, he argues that “product fragmentation has made the East Asian growth dynamism increasingly reliant on extra-regional trade (Athukorala 2005: 1).” If this is the case, a negative shock in the US, which would reduce imports from the PRC, would indirectly affect other East Asian countries' exports to the PRC, which implies that East Asia may not be decoupling from the US.

Following Athukorala (2005), I report in Table 4 [ PDF 16.3KB | 1 page ] the share of parts and components content for East Asia. In the original table the sample period stopped in 2000 and I extended it to 2006. East Asian countries cover the PRC; Hong Kong, China; Indonesia; Japan; Korea; Malaysia; Singapore; Taipei,China; Thailand; and Philippines. The share of parts and components content in the intra-regional trade in East Asia continued to increase from 2000 to 2006. In contrast, the same share in trade between East Asian countries and the US decreased from 2000 to 2006. These results show that product fragmentation is still the major source of intra-regional trade in East Asia, while its importance in trade with other regions, such as with the US, has weakened recently.

Second, while the impact of financial integration on business cycles is generally weaker, it is the US financial markets that are in trouble. In terms of their depth as well as their size, the US financial markets have been and remain by far the largest in the world economy. A shock in the US financial markets has a great influence on the worldwide financial markets. In general, as long as the real sector is insulated from the occurrences in the financial markets, the increasing correlations of asset prices do not necessarily imply that the business cycles of the real sector are synchronized. The question is if this shock will be confined in the financial markets or will it spill over to the real sector.

Third, another feature of the impact of financial integration on business cycle co-movements is that it is highly nonlinear. That is, the impact of financial markets is asymmetric across business cycles (WEO 2007). Asset price correlations tend to increase significantly during bear markets and recessions. Hence, there is a possibility that the impact of a negative shock is much greater. Further, if the shock is so major that the US financial markets suffer from a major crisis, this financial shock is more likely to be spilled over to the US' real sector.

Download this Paper [ PDF 696.5KB| 28 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.