In this paper we …nd empirical evidence of bank lending channel for Colombia, using a balanced panel data of about four thousand non-…nancial …rms.We …nd that increases in the interest rate, proxiing for the monetary policy instrument, lead to a reduction in the proportion of bank loans, out of total debt, of the …rms.This bank lending channel am-pli…es the e¤ect of the traditional interest rate channel, which leads to a reduction in total debt and spending when monetary policy tightens.Our evidence suggests that …rm size matters in the transmission of monetary policy through the bank lending channel: smaller …rms have a higher probability of being credit rationed after a tightening of monetary policy than (otherwise identical) larger …rms.