than those for trade growth. On this count, Anderson(2005a)finds that while the standard deviation of trade growth has been very sin lar to the standard deviation of GDP growth for a group of seven Asian economies, the standard deviation of GDP growth has been third as high as that for trade growth in China. In fact the relationship between gDP growth and trade growth in China looks much closer to that in the United States than it does to Chinas In my work with Nicholas Lardy( Goldstein and Lardy 2004)on Chinas growth prospects, we have underlined that it is investment and consumption that have the dominant weights (40 percent or more each)in China's GDP and that it is their behavior that typically drives GDP growth in China--not net exports. In 2004, the weight of net exports in Chinas gDP was just under 3 percent. Net exports can thus have a major influence(positive or negative)on Chinas growth ormance only in those years in which the percentage change in net exports is very large(say 75-100 percent or more). It turns or that 2005 was just such a year. But this cannot go on indefinitely as the rest of the world would surely resist a serial doubling of Chinas net export surplus Second, discussions of the impact of exchange rate revaluation should not proceed as if the exchange rate were the sole instrument of macroeconomic policy. In 2003 and much of 2004, an RMB re- valuation would have simultaneously moved the economy closer to both internal and external balance. In the classic terminology of James Meade(1951), China was then in a"non-dilemma"situation for exchange rate revaluation because the domestic economy was strongly overheated at the same time that China was running a siz able global balance-of-payments surplus. But such non-dilemma situ ations do not last forever. In fact. Chinas domestic final demand growth slowed in 2005-the near 10 percent growth in real GDP notwithstanding. This may have created a dilemma for the authorities a large, one-step revaluation would dramatically shrink the current account surplus but it could also slow economic growth more than Preliminary data for the first quarter of 2006 suggest that domestic demand growth is strengthening in China. But even if this strength ening were to prove short-lived, this does not mean that a significant evaluation of the rMb would then have to be abandoned. One -Chinas goods trade surplus, as measured by Chinese Customs, was roughly three times arger in 2005 than in 2004than those for trade growth. On this count, Anderson (2005a) finds that while the standard deviation of trade growth has been very simi￾lar to the standard deviation of GDP growth for a group of seven Asian economies, the standard deviation of GDP growth has been only about a third as high as that for trade growth in China. In fact, the relationship between GDP growth and trade growth in China looks much closer to that in the United States than it does to China’s Asian neighbors. In my work with Nicholas Lardy (Goldstein and Lardy 2004) on China’s growth prospects, we have underlined that it is investment and consumption that have the dominant weights (40 percent or more each) in China’s GDP and that it is their behavior that typically drives GDP growth in China—not net exports. In 2004, the weight of net exports in China’s GDP was just under 3 percent. Net exports can thus have a major influence (positive or negative) on China’s growth performance only in those years in which the percentage change in net exports is very large (say 75–100 percent or more). It turns out that 2005 was just such a year.12 But this cannot go on indefinitely— as the rest of the world would surely resist a serial doubling of China’s net export surplus. Second, discussions of the impact of exchange rate revaluation should not proceed as if the exchange rate were the sole instrument of macroeconomic policy. In 2003 and much of 2004, an RMB re￾valuation would have simultaneously moved the economy closer to both internal and external balance. In the classic terminology of James Meade (1951), China was then in a “non-dilemma” situation for exchange rate revaluation because the domestic economy was strongly overheated at the same time that China was running a siz￾able global balance-of-payments surplus. But such non-dilemma situ￾ations do not last forever. In fact, China’s domestic final demand growth slowed in 2005—the near 10 percent growth in real GDP notwithstanding. This may have created a dilemma for the authorities: a large, one-step revaluation would dramatically shrink the current account surplus but it could also slow economic growth more than desired. Preliminary data for the first quarter of 2006 suggest that domestic demand growth is strengthening in China. But even if this strength￾ening were to prove short-lived, this does not mean that a significant revaluation of the RMB would then have to be abandoned. One 12China’s goods trade surplus, as measured by Chinese Customs, was roughly three times larger in 2005 than in 2004. RENMINBI CONTROVERSIES 257