There has been an ongoing debate over whether the very notion of a mainstream, neoclassical economics is still valid. As many commentators have pointed out, various departures have been embraced by portions of the profession which, by sociological standards, have to be considered mainstream: these include game theory and interactive expectations with multiple equilibria, behavioral departures from homo oeconomicus, importation of biological and psychological measures of well-being, and so on. So where are we now?
First, note that each item on this list constitutes a relaxation of a default assumption that other economists, not directly engaged in that departure, typically rely on in their own work. Those who are not themselves behavioral economists generally invoke “U-max” unthinkingly, those who are not personally constructing game-theoretic or similar models invoke convexity assumptions to ensure stable, mono-equilibria, etc. The violation of the core axioms coexists with the default status of those same axioms. I think in this paragraph I am simply describing what currently takes place, not theorizing or explaining it.
Second, here’s a hypothesis. Neoclassical economics has evolved to serve an ideological function which is promoted through incentives, the selection of new adepts, and a conceptual hegemony: the purpose of economics is to solve economic problems with minimum, and ideally no, recourse to politics. Thus welfare economics in particular plays a central role, since it is the basis for proposing economic solutions that don’t depend on a political deliberation or selection process. What the theoretical departures that don’t migrate to the core have in common is their incompatibility with welfarism. The exception proves the rule here. Consider the case of the new institutional economics largely centered on transaction cost theory. This has arguably entered the core of the discipline despite having once taken the form of a departure. Is it coincidental that practitioners in this field have no difficulty modifying welfarism to account for institutional frictions?
Reinforcing these ideological pressures are the network externalities associated with core doctrines: like computer programs they gain in value as more users employ them. (I’m indebted to a manuscript I’m currently reviewing for this insight.) That said, the example of NIE convinces me that pure lock-in is not the only consideration in the inertia at the heart of economic doctrine; ideology is also important.
UPDATE: Google is not indexing this post! I wonder if it's because of the header? Poor Imre Lakatos must be turning over in his grave.
UPDATE^2: Imre can relax. Google, after a long hiatus, hoovered me up.