Despite the recent growth in capital flows across countries, direct trade linkages are still more important than financial linkages in determining how shocks to the world's largest economies affect a variety of markets around the globe.
In the first half of 2002, when accounting scandals, terrorist threats, and disappointing economic growth produced a 17 percent drop in the U.S. stock market, some other economies' markets predictably followed suit. For instance, Mexico's markets lost 11 percent; Ireland's dropped by 14 percent; and Finland's plunged by 30 percent. By contrast, other markets experienced exuberant returns: Iceland and South Africa both jumped by more than 20 percent during that period, while Colombia and South Korea also registered double-digit gains. Why do sudden swings in the market of the world's largest economies appear to spread to some smaller markets but leave others unaffected?
Kristin Forbes and Menzie Chinn tackle this question in their recent study A Decomposition of Global Linkages in Financial Markets Over Time (NBER Working Paper No. 9555). Although they acknowledge that their analysis is "only a start" and that many additional factors must be considered, they conclude that "direct trade linkages are still more important than financial linkages in determining how shocks to the world's largest economies affect a variety of markets around the globe."
Forbes and Chinn start by theorizing that a country's market returns are determined by global factors (such as international interest rates or commodity prices), sectoral factors (as measured by returns for industry-specific stock indexes), cross-country factors (returns in other financial markets), and country-specific factors. They then consider the cross-country linkages among five large economies -- the United States, Britain, Japan, France, and Germany -- and 40 developing countries, disaggregating the cross-country impact into four distinct links: direct trade flows, trade competition in third markets, bank lending, and foreign direct investment (FDI). Finally, they assess how the importance of these linkages has evolved over time (1986-2000), and whether bond markets are related across countries similarly to stock markets.
Forbes and Chinn find that cross-country and sectoral factors tend to be important in determining stock market returns around the world. As could be expected, the major economies in each region prove particularly important for nearby markets. Movements in the United States have a particularly important impact in the Americas, for instance, and markets in Germany, France, and the United Kingdom are especially influential in Europe. Market relationships also follow traditional colonial patterns; for example, the performance of British markets is a large factor for nations such as Australia, Canada, and Hong Kong.
Among the cross-country factors, Forbes and Chinn find that bilateral trade flows, as measured by a country's reliance on exports to the largest economies, are the most important. One surprising finding is that, after controlling for other linkages, foreign investment flows from large economies do not appear to significantly influence stock market returns in smaller markets.
The impact of cross-country factors also has evolved over time, the authors show. Forbes and Chinn divide their study into three periods of equal length: 1986-90, 1991-5, and 1996-2000. In the first two periods, cross-country linkages tend to have low explanatory power. However, from 1996 to 2000, "bilateral linkages through trade and finance become substantially more important determinants" of how shocks are transmitted from large markets to countries around the world. Again, direct trade flows prove the most important factor, with bank lending and trade competition in third markets also playing a role. However, FDI flows remain insignificant throughout the different periods. And, on a country-by-country basis, the United States became increasingly important in the transmission of market shocks in the 1996-2000 period, while British and Japanese influence declined.
Finally, the authors conduct a similar exercise for bond markets although, because of data limitations, they limit the scope of the study to the 1994-2000 period. Once again, sectoral and cross-country factors remain significant and more important than global factors in determining market returns.
The authors stress that their study does not incorporate several potentially important bilateral linkages, such as portfolio investment, trade credit, and exposure to multinational corporations. Nevertheless, Forbes and Chinn end their paper by reaffirming their key conclusion: "Despite the recent growth in capital flows across countries, direct trade linkages are still more important than financial linkages in determining how shocks to the world's largest economies affect a variety of markets around the globe."
-- Carlos Lozada