The existing empirical literature on banking crises has not produced agreement on their causes. Using a sample of 75 emerging markets in 1975-1997, we attempt to determine what we know about banking crises by establishing which previous results are robust. Among the robust causes of emerging-market banking crises are rapid domestic credit growth, large bank liabilities relative to reserves, and deposit-rate decontrol. On the other hand, there is no compelling evidence of any particular relationship between exchange rate regimes and crises. Finally, the evidence that deposit insurance or a weak institutional environment heighten crisis risk appears to be fragile.