Can the Baltic Dry Index predict foreign exchange rates?

The Baltic Dry Index (BDI) is commonly perceived as a leading indicator of economic activities. In this paper, we explore whether the BDI has predictive ability for exchange rates of fourteen major currencies against US dollar. Results of panel regression demonstrate that the BDI provides statistically significant long-run predictability of currency returns. An increase in the BDI is associated with a depreciation of currency. The three sub-indices of the BDI also show significant predictive power with similar patterns. The in-sample and out-of-sample results of individual time-series regressions illustrate that the BDI shows significant predictive ability in majority of the cases and displays an inverted U-shaped predictive pattern. Our results imply that by capturing information about economic fundamentals the BDI is a useful predictor for exchange rates.