This dissertation consists of three essays on capital controls and exchange rate regimes. The first essay, under the background of international monetary policy trilemma, empirically investigates the validity of the proposition that holding the degree of exchange rate stability constant, a decrease in capital mobility through imposition of capital controls will enhance monetary independence. Using a panel dataset covering 88 countries for the 1995-2010 period and system GMM estimation, this paper finds that 1) capital controls help improve a country’s monetary independence; 2) the effectiveness of capital controls depends on the types of assets and the direction of flows they are imposed; 3) the choice of exchange rate regime has important impact on the effectiveness of capital controls on monetary independence. The second essay examines the role of capital controls on economic growth. Conventional wisdom suggests that allowing international capital flows improves domestic investment and growth by providing extra resources through international capital market, yet the flows can be misallocated to finance speculative or low-quality domestic investments. Using a panel dataset covering 78 countries over 1995-2009, this paper finds that 1) capital control policies promote economic growth after taking into account a country’s de facto level of capital flows; 2) controls on capital inflows helps a country’s economic growth, but not controls on outflows; 3) restrictions on different asset types affect growth differently. Capital controls on equity type flows are less effective than controls on debt type flows or direct investment. The third examines the role of exchange rate flexibility on current account balances. Global imbalances have become an important issue for economists and policy makers. Greater exchange rate flexibility is often suggested as a means to achieve faster and more efficient adjustment in the current account. However, previous empirical studies show little support for this hypothesis. This essay revisits this issue with a large panel dataset and Threshold VAR model and finds that 1) the speed of the current account adjustment is higher in a regime with greater exchange rate variability; 2) some existing popular exchange rate classifications may not capture actual exchange rate variability as well as expected.