The way a country manages its currency can affect its volume of trade, capital flows and income. A country, especially one with high degree of trade openness, needs to find the most suitable exchange rate arrangement to reduce the volatility of its currency value and output. This thesis examines the optimal choice of exchange rate arrangements from various aspects.The first essay examines whether or not the Northeast Asian economies, that is, China, Japan, Korea, and Taiwan can form a currency union, where a single currency and a uniform monetary policy are adopted, or an exchange rate union, where all the currencies are pegged to an internal or external currency or currency basket. It also attempts to find the optimal currency or currency basket for the four Asian economies if an exchange rate union is feasible. Structural VAR models with identification assumptions are applied to analyze the correlations of supply, exchange rate, monetary, and demand shocks. The paper finds that the shocks of these four economies are not symmetric, implying that the Northeast Asian economies are not ready yet to form a common currency union. However, it is found that these economies can form an exchange rate union with a major currency basket including the U.S. dollar, the Japanese yen and the Euro as currency anchor. The paper also examines the option of pegging to a basket of regional currencies, similar to the Asian Currency Unit (ACU).The second essay uses panel data on emerging and developing countries to study the interrelationships between balance-sheet currency mismatches and the choice of exchange rate regimes. The evidence shows that the lack of exchange rate flexibility reinforces currency mismatches and increases dollar liabilities, but the estimated economic effects are small.The third essay uses a three-country model to analyze the choice of the optimal weights of a currency basket for emerging market economies. This model assumes imperfect capital mobility and allows the domestic country to impose a reserve requirement on capital inflows. The optimal weights are derived by minimizing the loss from the volatility of output and trade. The result shows that the optimal weights are affected by variance of the cross exchange rate between the two major currencies, the covariance between inflations in the two large countries and the cross exchange rate, the relative weights assigned to trade and output, and price and exchange rate elasticities of trade, demand and supplies.
