Predictive power and unbiasedness of implied forward charter rates

This paper examines the efficiency and predictive power of implied forward shipping charter rates. In particular, we examine whether implied forward 6-month time-charter rates, which are derived through the difference between time-charters with different maturities based on the term structure model, are efficient and unbiased predictors of actual future time-charter rates. Using a dataset for the period January 1989 to June 2003, results of different statistical tests, including the cointegration approach, suggest that implied forward rates are in fact unbiased predictors of future time-charter rates in the dry bulk freight market. In addition, it is found that implied forward rates yield superior forecasts compared to alternative univariate and multivariate time series models. However, while the unbiasedness hypothesis is found to hold, on average, we find that chartering strategies based on simple trend-following trading rules in this cyclical market are able to generate economic profits even out-of-sample. This highlights how standard tests for unbiasedness do not always capture cyclical predictable components in the market behaviour.