Three essays on international finance
This thesis contributes to three major fields in the international finance literature: forward premium anomaly, monetary policy and exchange rate determination, and measuring monetary policy expectations from asset prices. In the first chapter, I develop a production-based asset pricing model predicting that excess returns of currency carry trade compensate investors for the commodity price risk. Commodity producers differ in their exposure to the export price risk. Exchange rate-commodity price covariance, procyclical interest rates, negative price of exchange rate volatility, and countercyclical currency risk premium arise endogenously. Empirically, risk factors implied by the model explain up to 55% of time-series variation in carry trade returns across developed countries, and generate substantial risk- and transaction costs-adjusted returns as tradable strategies. In the second chapter (jointly with Igor Pozdeev), we document a drift in exchange rates before monetary policy changes across major economies. Currencies tend to depreciate by 0.8% over ten days before policy rate cuts and appreciate by 0.5% before policy rate increases. We show that available fixed income instruments allow to forecast monetary policy decisions and thus that the drift is exploitable by investors. Buying (selling) currencies ten days in advance of predicted target rate hikes (cuts) earns on average a statistically significant excess return of over 40 basis points per ten-day period after trading costs. We further demonstrate that this return is robust to the choice of holding horizon and monetary policy forecast rule. Our results thus pose a major challenge for the risk-based explanations of the exchange rate dynamics. In the third chapter (jointly with Igor Pozdeev), we document overnight index swaps (OIS) to be unbiased predictors of future short rates in developed economies, bearing no significant risk premium for maturities up to one year. We show that the OIS underlying overnight rates accurately reflect the target rates set by central banks. We extract the implied future target rates from the OIS prices to predict the outcome of monetary policy meetings around the world. In the US, a randomly selected triplet of a target rate hike, cut and no-change is correctly classified using the OIS-implied rates in 99.9 and 98% of cases five and ten days before a FOMC announcement respectively. We report similarly high prediction accuracy for other developed countries.
