Academic journal article Contemporary Economic Policy

United States-China Trade at the Commodity Level and the Yuan-Dollar Exchange Rate

Academic journal article Contemporary Economic Policy

United States-China Trade at the Commodity Level and the Yuan-Dollar Exchange Rate

Article excerpt


Over the last few decades, the United States has been running trade deficits. In the 1980s and 1990s, Japan was said to be a major contributor. Today, the emphasis has shifted to emerging economies, especially China. In 1978 when China began her reform process and introduced elements of a free market system to her economy, the United States had a trade surplus of almost 600 million dollars. In 2002 (the last year for which the data are available), that trend has reversed itself and the United States suffered a trade deficit in the magnitude of 120 billion dollars. On the other hand, over the 1978-2002 period, Chinese currency has depreciated from 2.2 yuan per dollar to 8.4 yuan per dollar. Has depreciation of the yuan played any significant role in the trade between the two countries? (1)

One of the traditional ways of assessing the impact of currency depreciation on the trade balance is to rely upon price elasticities of export and import demands, hence the Marshall-Lerner condition. If the sum of export and import demand elasticities adds up to more than unity, the Marshall-Lerner condition is satisfied and depreciation or devaluation is said to be effective in improving the trade balance. Examples of studies that have relied upon this approach are Houthakker and Magee (1969), Khan (1974), Goldstein and Khan (1976), Wilson and Takacs (1979), Haynes and Stone (1983a, 1983b), Warner and Krienin (1983), Bahmani-Oskooee (1986, 1998), and Bahmani-Oskooee and Niroomand (1998). These studies have provided a mixed conclusion with regard to the validity of the Marshall- Lerner condition which could be due to the fact that they have all used aggregate trade data. When aggregate data are used, a significant price elasticity of one industry could be more than offset by an insignificant elasticity of another industry, yielding an insignificant elasticity at the aggregate level.

In. this article, rather than using aggregate trade data we concentrate on another avenue of research that concentrates on using trade data at the commodity level so that we can identify industries that are sensitive to the exchange rate changes. For this purpose, we consider trade between the United States and China and 88 commodity groupings (two-digit and three-digit industries). Section II introduces the models and estimation techniques. Section II] reports the empirical results, and Section IV concludes. Data definition and sources are cited in the Appendix.


A country's trade balance pertaining to commodity j is said to be the difference between her earnings due to export of that commodity and payments due to import of that commodity, as outlined by Equation(1)

(1) T[B.sub.J]=[[P.sub.J.sup.X][X.sub.J]] - [[P.sub.J.sup.M][M.sub.J]]

where T[B.sub.J] is the trade balance pertaining to the trade in commodity j; [P.sub.J.sup.X]([P.sub.J.sup.M]) is the export price (import price) of good j: and [X.sub.j]([M.sub.j]) is the export volume (import volume) of good j. As mentioned before, in assessing the impact of currency depreciation or devaluation on the trade balance, earlier studies emphasized estimating own-price elasticity of exports and imports by estimating export and import demand models. The purpose was to determine the condition under which [P.subj.sup.X][X.sub.j] or inpayments will increase and [P.sub.j.sup.M]M or out-payments will decrease. In the context of Model (1), this implies that we must estimate own-price elasticity of export of commodity and import of commodity j by estimating models in which export volume of commodity j and import volume of commodity j are linked to their respective prices in addition to other determinants. However, export and import prices are not available at commodity level to deflate value data, making the traditional approach of estimating elasticities impossible. An alternative approach would be to follow Haynes et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.