Although some recent studies have analysed issues relating to credit in African manufacturing, they have not directly tested for the effect of credit on firm growth. The use of bank credit can affect firm growth in two opposite ways. The effect may be positive if credit allows a firm to address its liquidity constraint and increase profitability. However, if macroeconomic shocks such as increases in interest rates make firm debts unsustainable as experienced in Kenya in the 1990s, indebted firms may shrink or even collapse. Hence, empirical testing is necessary to determine which effect dominates in a specific case. Using microeconomic data on the Kenyan manufacturing sector, the study finds that conditional on survival, the firms that use credit grow faster than those not using it. This result is robust to alternative estimation procedures, controlling for both endogeneity of the credit variable and selection bias. Convergence in firm size is significant in all the models except the GMM estimation that controls for several forms of endogeneity. The significance of convergence contradicts Gibrat's law of proportionate effects while supporting Jovanovic's learning hypothesis.