I use the introduction of deposit insurance in eight U.S. states in the early twentieth-century to study the effects of deposit insurance on the banking system. Using a triple difference approach exploiting regulatory differences between national and state banks and between states, I find that insured banks experienced higher deposit growth and decreased funding costs. I also observe a replacement of demand deposits by riskier time deposits. However, I find no aggregate effects on failure rates or risk-taking. Using hand-collected micro-level data, I show that small and large banks reacted differently and that banks facing funding problems especially benefited.