Joakim Westerlund

Most empirical evidence suggests that the efficient futures market hypothesis, henceforth referred to as EFMH, stating that spot and futures prices should cointegrate with a unit slope on futures prices, does not hold, a finding at odds with many theoretical models. This paper argues that these results can be attributed in part to the low power of univariate tests, and that the use of panel data can generate more powerful tests. The current paper can be seen as a step in this direction. In particular, a newly developed factor analytical approach is employed, which is very general and, in addition, free of the otherwise so common incidental parameters bias in the presence of fixed effects. The approach is applied to a large panel covering 17 commodities between March 1991 and
August 2012. The evidence suggests that the EFMH cannot be rejected once the panel evidence has been taken into account.