Integration and Causality in International Freight Markets: Modeling with Error Correction and Directed Acyclic Graphs

Using directed acyclic graphs (DAGs) and error correction models, we study the dynamics of freight prices that comprise the Baltic Panamax Index (BPI), the index on which freight futures trading was based. The DAGs are used to make statements about the contemporaneous correlations between prices and allow us to address the construction of the data-determined orthogonalization on contemporaneous innovation covariance, which is crucial in providing inference in innovation accounting techniques. Our results provide a source of information on price discovery and suggest that the index is not appropriately composed and weighted, which may help explain the failure of the Baltic International Freight Futures Exchange (BIFFEX) contract.