Abstract. This paper critically examines the theoretical literature on banking instability that has followed Diamond and Dybvig (1983). It explores the extent to which it (a) explains banking instability within a theoretical context in which financial intermediaries improve on unintermediated markets, and (b) justifies government involvement in the financial intermediation industry. It suggests that the literature has yet to provide a satisfactory theoretical basis for banking instability as such since the intermediaries which arise from it are peculiar mutual funds that bear little resemblance to real-world banks. In addition, the paper challenges the widespread belief that this literature provides a sound foundation for government involvement in the industry. It suggests that arguments for government intervention are open to objection on various grounds, the most important one being that they are inconsistent with the existence of properly motivated financial intermediation in these models.