Tim Richter

This study examines whether shocks from macroeconomic variables or asymmetric effects are more suitable for explaining the time-varying volatility in the dry bulk and tanker freight markets or whether both effects should be incorporated simultaneously. Using Baltic Exchange indices during the sample period from March 1999 to October 2011 on a daily basis, we separately analyse the impact of macroeconomic shocks and asymmetric effects on the conditional volatility of freight rates by using a GARCH-X model and an EGARCH model, respectively. Furthermore, we simultaneously investigate both effects by specifying an EGARCH-X model. Assuming not only a normal distribution but also a t-distribution in order to better capture the fat tails of error terms, three important conclusions emerge for modelling the conditional volatility of freight rates: (i) The assumption of a t-distribution is better suited than a normal distribution is. (ii) Macroeconomic factors should be incorporated into the conditional variance equation rather than into the conditional mean equation. In addition, the number of macroeconomic factors that exhibit explanatory power decreases under a t-distribution. (iii) While there seem to be no asymmetric effects in the dry bulk freight market, these effects are strongly pronounced in the tanker freight market. Our empirical findings have important implications for freight rate risk management.