KEY POINTS

1. An exchange rate crisis is a large and sudden depreciation that brings to an end a fixed exchange rate regime.
2. Such crises are common. The typical fixed exchange rate lasts only a few years. History shows that crises can affect all types of countries—advanced, emerging, and developing.
3. Crises have economic costs that tend to be very large in emerging markets and developing countries. Political costs are also large.
4. To avoid a crisis, the central bank in a country with a fixed exchange rate regime must have the ability to peg the exchange rate. In practice, this means the central bank needs foreign currency reserves, which can be bought or sold in the forex market at the fixed rate.
5. In a simple model of a central bank, the money supply consists of domestic credit and foreign reserves. Money demand is exogenous and is determined by interest rates and output levels that we assume are beyond the control of the authorities when the exchange rate is pegged. In this model, reserves are simply money demand minus domestic credit.
6. If money demand rises (falls), holding domestic credit fixed, reserves rise (fall) by the same amount.
7. If domestic credit rises (falls), holding money demand fixed, reserves fall (rise) by the same amount and the money supply is unchanged. The combined result is called sterilization.
8. When the central bank gives assistance to the financial sector, it expands domestic credit. If it is a bailout, money demand is unchanged, and reserves drain. If it is a loan to satisfy depositors’ demand for cash, then reserves stay constant.
9. A first-generation crisis occurs when domestic credit grows at a constant rate forever, usually due to the monetization of a chronic fiscal deficit. Eventually, reserves drain and the money supply grows at the same rate, causing inflation and depreciation. Myopic investors do not anticipate the drain, and when reserves run out, they see a sudden jump (depreciation) in the exchange rate. Investors with foresight will try to sell domestic currency before that jump happens and by doing so will cause a speculative attack and a sudden drain of reserves.
10. A second-generation attack occurs when the authorities’ commitment to the peg is contingent. If the domestic economy is suffering too high a cost from pegging, the authorities will consider floating and using expansionary monetary policy to boost output by allowing the currency to depreciate, thus breaking the peg. If investors anticipate that the government will break the peg, they will demand a currency premium, making interest even higher under the peg and raising the costs of pegging still further. In this setup, at some intermediate costs, the authorities will maintain the peg as long as investors find the peg credible, but they will allow their currency to depreciate if investors find the peg not credible. This creates multiple equilibria and self-fulfilling crises.