Do analysts' forecasts of term spread differential help predict directional change in exchange rates?

We show that Blue Chip analysts' forecasts of US-Australia and US-UK exchange rates cannot accurately predict directional change in these exchange rates. The literature suggests that the cross-country term spread differential contains useful information for predicting exchange rates. We show that the difference between analysts' and random walk forecasts of the US-Australia (US-UK) term spread differential has directional predictability for the US-Australia (US-UK) exchange rate for 1997-2007 but not for 2008-2015. For the former period, the predictions generally imply symmetric loss, meaning that they are of value to a user who assigns similar loss to both incorrect upward and downward moves.