This paper will investigate the causality of bank lending channel by applying the non-causality test proposed by Toda and Yamamoto (1995). We apply the quarterly data from 1947Q1 to 2009Q4 in the United States and divide the sample into six sub-samples to conduct our empirical studies. Based on evidence of this thesis, we verify the bank lending channel does exist in the early years of the United States. We also examine the U.K. and the Japan data, but we find that there is no strong evidence that the bank lending channel exists in these two countries.