Currency crash risk in the carry trade

This thesis provides a systematic study of currency crash risk and funding liquidity risk in carry trade strategy in the foreign exchange (FX) market. Carry trade, which involves longing currencies with high interest rate and shorting currencies with low interest rate, is a popular currency trading strategy in the FX market for obtaining annualized excess return as high as 12%. This thesis studies exchange rates of 9 currencies over 13 years from a microstructure perspective. We identify a global skewness factor and use it to measure the currency crash risk. Applying a portfolio approach in cross-sectional asset pricing, we find that global skewness factor explains more than 80% of carry trade excess returns. On the other hand, funding liquidity is effective in predicting the future currency crash risk. Funding liquidity explains more than 70% of carry trade excess returns. We also use the coefficient of price impact from customer order flows to measure the liquidity, which reveals heterogeneous information content possessed by different types of customers. We find that the order flow implied liquidity risk factor can explain a fraction of carry trade excess returns but with small risk premium on quarterly basis. We provide empirical evidence to show that the excess return and crash risk in carry trade is endogenous; i.e., the crash risk premium is inherent in carry trade process. As the natural condition widely affects all investors, we argue that funding constraints are effective in explaining the excess returns of carry trade. When capital moves smoothly in a liquid condition and investor have sufficient funding supply, carry trade is prosperous in the FX market. When investors hit their funding constraints, market-wide liquidity drop, which force the carry trade positions diminishing. The exchange rates respond as that the low interest rate currencies appreciate and high interest rate currencies depreciate, which exacerbates currency crash risk and induces large loss to carry traders. Our cross-sectional analysis provides empirical evidence to show that funding constraints helps to explain the forward premium puzzle and push the exchange rate shift back to the direction the UIP expects