This thesis comprises three essays in international finance, with a focus on the foreign exchange (FX) market. The first chapter examines the pricing ability of the short- and long-run components of global FX volatility in the cross-section of carry trade returns. I find a negative and statistically significant factor risk price for the long-run component. Low interest rate currencies covary positively with innovations in the long-run component of volatility, while currencies with high interest rates covary negatively. Interestingly, I do not find significant evidence in favor of the short-run volatility component being a cross-sectional priced risk factor. I also show empirically that the long-run component is the dominant part of global FX volatility and is dynamically related to US-specific macroeconomic fundamentals such as industrial production and money balances. The second chapter investigates the time-series predictability of bilateral exchange rates from unconditional and conditional linear factor models using dollar, carry, and global FX volatility risk factors. I find evidence that all versions of the models largely fail to outperform the random walk with drift benchmark in out-of-sample forecasting of monthly exchange rate returns for individual currencies. This holds for currencies sorted into portfolios conditional on forward discounts. I also show that information embedded in dollar, carry, and global FX volatility risk factors do not generate systematic economic value to investors. In fact, linear factor models with currency-based risk factors do not outperform the random walk with drift benchmark in out-of-sample economic value analysis.