Interest rate differentials and the dynamic asymmetry of exchange rates

We propose a unified econometric strategy to revisit the informational content of interest rates differentials (IRD) for predicting exchange rates. The novelty of our approach consists in allowing for a time-varying asymmetry component in the conditional distribution of the depreciation rate, therefore explicitly modeling the link between interest rates and the likelihood of a depreciation. To assess the economic significance of IRD as a predictor, we derive a directional forecasting procedure from our model and apply this technique to daily exchange rates of the Euro and the Swiss Franc. We document in-sample and out-of-sample performances significantly superior to benchmark models, both in terms of sign forecasts and trading profits. Overall, we find the dynamic asymmetry component to be driven by interest rate differentials, but also by general uncertainty and past unexpected shocks. These findings empirically confirm currency crash theories for recent time periods, suggesting that the larger the difference between interest rates, the more likely the high yield currency appreciates but also exhibit larger depreciation risks.