Abstract: The correlation between Chinese and international commodity prices may be nonlinear because of China’s minimum agricultural commodity purchase price policy and temporary storage policy. In order to research this nonlinear dynamic correlation mechanism, we construct a nonlinear Granger causality test model and a nonlinear autoregressive distribution lag model including Chinese and international agricultural commodity (soybean, corn, rice, and wheat) price variables. Our empirical results reveal that a unidirectional causal relation exists between international and Chinese prices for soybeans and corn; specifically, international prices of soybeans and corn Granger-cause Chinese prices of soybeans and corn. Moreover, the pass-through effects between Chinese and international commodity prices are asymmetric; Chinese agricultural commodity prices respond more strongly to positive shocks than negative shocks of international agricultural commodity prices.

Keywords: international agricultural commodity price, domestic agricultural commodity price, asymmetry, nonlinear Granger causality, NARDL model