I also have a degree in economics and think if you average out the two opinions upthread you get closer to the truth :) (No but seriously, rational expectation

I also have a degree in economics and think if you average out the two opinions upthread you get closer to the truth :) (No but seriously, rational expectations works pretty well in the limited sense of showing that economic models making other assumptions about how human behaviour expect humans to persistently make systematic errors that there's usually no justification for. It's of no value in understanding human behaviour and rather limited value in predictive modelling, especially when it emerges as the assumption that expectations are always correct in their priors, as opposed to the assumption that they eventually adapt to making random rather than systematic error. Similarly the New Keynesian paradigm is pretty good at showing that even in idealised market conditions it only takes one bit of friction to create a situation where economic stimulus results in less economic pain than letting the economy sort itself out, but if you're actually trying to predict the size of the stimulus required by estimating menu costs you're doing it wrong. Since the New Keynesians managed to adapt their models to the Lucas critique whilst having quite different policy preferences to Lucas himself, it's also pretty clear that the allegiance to models isn't blind and economists are perfectly capable of formulating different proposals for better and for worse despite rather than because of modelling orthodoxy)