The main objective when a country implements capital controls is to prevent large fluctuations in the exchange rate and asset price volatility. The direct mechanism through which these policies work is simple: a tax on foreign borrowing reduces flows, which prevents the price from changing considerably. Since foreign borrowing involves transactions in the foreign exchange market, the price of the asset can also be thought of as the exchange rate. However, the empirical literature has not come to a consensus on the effectiveness of capital controls on managing the exchange rate. Therefore, could there be other channels, different from their direct effect on flows, through which capital controls have the undesired effect of increasing fluctuations in the exchange rate? In particular, can capital controls increase the sensitivity of prices to sudden changes in capital flows?

The dissertation answers this question using two approaches. First, I embed into a market microstructure model a mechanism through which capital controls reduce the ability of the market to sustain large amounts of foreign capital without a substantial change in their price. This characteristic is called market depth. The deeper the market, the price reacts less to adjust for an excess supply or demand of the asset. Second, I verify the existence of the theoretical mechanism in the data by analyzing the case of Mexico and Brazil. I focus on these countries because they are two similar foreign investment destinations, but with the main difference that Brazil has implemented capital controls in the past and Mexico has not.

In the first chapter I present a survey of the theoretical and empirical literature of capital controls and capital flows. In the second chapter I present the theoretical model that analyzes the effect of capital controls on market depth. The third chapter proposes a new measure on capital account restrictiveness. This measure is, to the best of my knowledge, the first index of capital controls that has quarterly periodicity and that is an intensive index. Finally, the last chapter analyzes two econometric models that explain the effect of capital controls on the levels and composition of capital flows, and on the probability of extreme events of flows.

If policymakers choose to implement capital controls for their short-term effect on the exchange rate, my results show that there are permanent effects on price sensitivity that could outweigh their immediate benefits. Moreover, the new measure proposed in this work can be used to find new evidence on the effect of capital controls on capital flows.