A great exchange from the two bloggers at Defective Equilibrium, the New Zealand version of Marginal Revolution:
This doesn't mean we have to argue that markets are always and everywhere a non-starter. It is merely an acknowledgment that the model proferred by neo-classical economics is not the final draft of economic explanation, but that it is akin to the Newtonian Physics awaiting its Einsteinian upheaval. For me this is tremendously exciting, as it opens up the study of economics and allows its foundations to be put under the experimental zoom lens. If people are predictably irrational, then it is something we need to be aware of as empirical scientists and integrate into our models of market behaviour.
Uday used a useful analogy of Newtonian vs. Eisteinian physics. Newtonian physics gave us a good approximation of how the world worked for hundreds of years - but as Einstein showed, it turned out to be fundamentally misconceived. Einstein's insights gave us a deeper, more accurate understanding of how things actually work.
But have we since abandoned Newtonian physics? Of course not. In the vast majority of observable situations (that is, except for research physicists) Newton's laws still appear to hold. Engineers need not worry that people moving around in their buildings will increase in mass as they move and thus crash through the floor, for example.
While earning a master's degree in mechanical engineering, I don't think I ever learned about Einstein's theories beyond the cursory mention that Newton's stuff is really just an approximation for how things really work, but it's so accurate for most real world applications that it's all we're going to teach you. Ten years from now, perhaps this is how neoclassical economics will be thought of.
Image from here.