In a rational profit-maximizing world, banks should maintain a credit policy of lending if and only if borrowers have positive net present value projects. Why then are changes in credit policy seemingly correlated with changes in the condition of those demanding credit? This paper argues that rational bank managers with short horizons will set credit policies that influence and are influenced by other banks and demand side conditions. This leads to a theory of low frequency business cycles driven by bank credit policies. Evidence from the banking crisis in New England in the early 1990s is consistent with the assumptions and predictions of the theory.