Forecast Failure, Expectations Formation, and the Lucas Critique.

Since forecast failure is due to unanticipated large shifts in deterministic factors,'sensible' agents should adopt 'robust forecasting rules'. Unless the model coincides with the generating mechanism, one cannot even prove that causal variables will dominate non-causal in forecasting. In such a non-stationary world, 'rational expectations' do not have an epistemologically-sound basis: agents cannot know how all relevant information enters the joint data density at every point in time. Thus, although econometric models 'break down' intermittently when deterministic shifts occur, that is not due to the Lucas critique and need not impugen their value for policy analyses