This paper applies a recursive approach to evaluate the credit – output relationship empirically using Peruvian quarterly data for the period between 1992 and 2009. Given the nature of the series, econometric analysis is based on the estimation of a vector error correction model (VECM) which establishes the existence of a dynamic relationship between the growth rate of output and the "credit impulse", the latter defined as the change in the growth rate of credit. The results show that the credit impulse in domestic currency (soles) contains relevant information to predict the growth rate of output in the short run.