The policy implications of estimated macroeconometric systems depend on the
formulations of their equations, the methodology of empirical model select
ion and evaluation, the techniques of policy analysis, and their forecast p
erformance. Drawing on recent results in the theory of forecasting, we ques
tion the role of rational expectations: criticize a common approach to test
ing economic theories, show that impulse-response methods of evaluating pol
icy are seriously flawed, and question the mechanistic derivation of foreca
sts from econometric systems. In their place, we propose that expectations
should be treated as instrumental to agents' decisions, discuss a powerful
new approach to the empirical modelling of econometric relationships, offer
viable alternatives to studying policy implications, and note modification
s to forecasting devices that can enhance their robustness to unanticipated
structural breaks.