Anti-Comparative Advantage: A Puzzle in U.S.-China Bilateral Trade

From 1992 to 2011, the total trade volume between the U.S. and China increased by 25 times, and China's share in U.S. total imports increased from 5% to 20%. However, the U.S.'s share in China's total imports dropped from 11% to 8% in the same period. In the major categories of U.S. exports to China, Waste & Scrap increased from 744 million dollars in 2000 to 7, 562 million dollars in 2008, rising 916% times and becoming the No.1 product that the U.S exports to China. It is important to understand what explains these structural changes, and to ask whether the principle of comparative advantage determines the structure of U.S.-China bilateral trade. Interestingly, we find an "Anti-Comparative Advantage" puzzle: the U.S. exports less to China in sectors where it has greater comparative advantage, while China exports more to the U.S. in its sectors with greater comparative advantage. To further study this issue, we extend Eaton-Kortum model of bilateral trade to multiple sectors and test it empirically using US and China trade data. We find that after controlling for the importer's demand, trade costs and factor intensities, etc., comparative advantage cannot explain U.S.-China bilateral trade flows. The puzzle survives various robustness checks.