A controversial issue in the literature on banking regulation is whether there is contagion risk, or not. This paper derives a framework to test for contagion risk and applies it to a data set of monthly bank failures under the US National Banking System from 1880 till 1936. To capture the count nature of bank failure data, an autoregressive Poisson model is used. The empirical results indicate that there is contagion risk in banking. An initial failure could generate further failures without intervention by the authorities. This finding suggests that there may be a role for the central bank as lender of last resort to assist ailing banks, whose failure is expected to have a systemic impact.