In the aftermath of financial crises, governments can use economic policy to minimize the risk of future recurrence. Yet not all do so. To explain this divergence in responses I develop a theory of economic policy choice after financial crises. I argue that past financial crises provide information to future governments about the political costs of financial crises. This subsequently informs the need to use economic policy to insure against such crises. Focusing on the accumulation of foreign exchange reserves after currency crises, I find that when past currency crises led to political changes future governments accumulate higher levels of reserves to prevent another crisis from occurring. This effect is stronger when political change occurred in situations where governments would not expect to be held accountable, and when reserve sales were shown to be effective in preventing political change. The theory and empirical results provide an answer as to why countries experiencing a similar form of financial crisis can, nevertheless, vary in their attempts to prevent future recurrence.