This article examines the causality among economic growth, export expansion and export composition in China since economic reform. The cointegration tests show that cointegrating vectors do not exist either between economic growth and export composition or between export expansion and export composition. Meanwhile, the null hypothesis that economic growth and export expansion are not cointegrated is rejected. The error correction model shows that there is a bi-directional causality between export expansion and economic growth. Meanwhile, there appears no causal relationship between export composition and the other variables.