The Mystery of the Excess Trade (Balances)

Bilateral trade deficits are a perennial policy issue. Former Deputy Assistant U.S. Trade Representative for Japan and China, Merit Janow (1994 p. 55), notes that during the first George Bush administration, "High deficits coupled with the continuing allegations from U.S. business interests about the closed nature of the Japanese market were resulting in serious domestic political pressures for improved access to the Japanese market." Recently Robert C. Feenstra et al. (1998 p. 1) made similar comments vis-'a-vis China: "Some analysts have interpreted the large U.S.-China bilateral trade deficit as prima facie evidence of unacceptably high levels of protectionism in China, and have advocated stringent entry conditions for China's admission into WTO." Given the policy salience of bilateral trade deficits, it is peculiar that no one has ever examined them empirically for a broad set of countries. One reason for the scant study is that economists are naturally (and sensibly) loath to accept the terms of the policy debate, which considers bilateral trade deficits ipso facto harmful. A second reason is that economists believe there may be very natural explanations for bilateral imbalances. One such explanation finds its origins in macroeconomic identities that equate current-account deficits to an excess of investment over saving. From this, it may be argued that bilateral imbalances will arise naturally in trade between countries in aggregate surplus and those in aggregate deficit. Indeed, this is the principal explanation that the profession has given policymakers, and it forms the foundation of many U.S. bilateral trade initiatives such as the Structural Impediments Initiative and the Framework talks. Janow (1994 p. 55) observes that "there was (and is) little disagreement among economists that the causes of large aggregate and bilateral deficits are largely attributable to macroeconomic factors" [italics added]. A second account may rely on what may be termed "triangular trade," in which cross-country differences in the patterns of demand and supply mean that a country will run bilateral deficits with those countries that are unusually important suppliers of the goods for which the deficit country happens to be an unusually strong demander. In this paper, we use the canonical "gravity model" of bilateral trade to form predictions about bilateral trade balances. We develop two key variants of the model, in which bilateral trade imbalances arise due to aggregate macroeconomic imbalances or due to "triangular t ade," and we implement these empirically for a broad set of countries. Our results paint a dismal picture. The central explanations that economists provide to explain bilateral balances perform miserably. There are two key failures. First, actual bilateral trade imbalances are much larger than those predicted; there is a "mystery of excess trade balances." Second, even after we allow for both macroeconomic imbalances and idiosyncrasies in the structure and levels of demand and production, the models perform poorly in explaining bilateral trade balances. These failures of economists' standard explanations of bilateral trade imbalances require that we move beyond the simple gravity framework to consider alternative explanations: homogeneous goods, highly specialized intermediates, and the role of policy.