Estimating the import demand function in the autoregressive distributed lag framework: The case of China

This paper uses the concept of cointegration for empirically analyzing the long-run relationship of China's import demand function. The analysis employs the annual data for the sample period from 1978 to 2009. The purpose of this study is to investigate and explain China's import demand functions and provide a more in-depth analysis of China's import behavior. The autoregressive distributed lag (ARDL) and dynamic ordinary least square (DOLS) techniques were used for estimating the long-run coefficients of price and income elasticities. The empirical results from ARDL bound testing approach and Johansen's method of cointegration provide strong evidence of the existence of a long-run stable relationship among the variables included both in the traditional model and the disaggregated expenditure model of import demand. In addition, the disaggregated import demand model estimated in this paper provides a complete description of the determinants of China's imports, and offers empirical results that are significantly different from those obtained in existing studies (Tang, 2003). This is an important finding for resolving the issue of trade imbalance from the perspective of China's policy formulation. We are grateful to Kazuhiro Ohtani, Akio Namba, and Mark Wynne for helpful comments and suggestions