Like comparable past episodes, the financial crisis of 2008 can be attributed to strong prior growth in credit, which fuelled a sharp rise in financial and real-estate asset prices. Nevertheless, it is a specific event. In particular, it was preceded by a massive underestimation of risk and an explosion of loans to insolvent households. Both mechanisms were facilitated by an excess of global saving, fostered and maintained by imperfect-information issues and various regulatory inadequacies. Classic factors such as market imperfection explain the rapid spread of the crisis to other parts of the financial system and other countries. The complexity of structured mortgage products may have amplified these effects. As in previous episodes, the response to the crisis occurred in two stages. An initial, fragmentary approach comprised intervention by monetary authorities, the injection of liquidity, and ad hoc rescues of individual institutions. As these measures failed to restore market confidence, major rescue plans were enacted in autumn 2008. As often in the past, the proposed solutions combined three main actions: guaranteeing bank liabilities, recapitalizing financial institutions, and separating toxic assets from healthy assets. The gross budgetary costs of crises are usually heavy, but must be weighed against the potentially high costs of inaction: a market collapse and its consequences are risks that need to be factored into the equation. The effect of financial crises on the long-term output (GDP) level can be lastingly negative, and is generally estimated at 1.5-2.4%. However, the estimated effect of severe financial crises is much greater.