I saw a nice expression of bad economics in the conservative style the other day, “Economists shouldn’t pick winners and losers, they should just give policy advice based on efficiency.” You see it, right? No? Even now? Fine, let’s talk through it.
First, we have the typical ambiguity about “efficiency.” Are we talking about “economic efficiency” defined with respect to “utility” specifically? Or general technical efficiency defined with respect to some unspecified socially agreed goal X? To the extent we’re talking about efficiency with respect to attaining socially agreed upon goal X, that seems a technical matter that wouldn’t involve economists addressing ethical issues relating to X, such as resolving interpersonal conflicts of preferences, economic power, etc. Granted, there may be ethical issues, tradeoffs, considerations involved in choosing, pursuing goal X, but in this case the goal is not being generated by economists themselves, so they’re not choosing the winners and losers associated with goal X. The situation is completely different, of course, if we’re talking about so-called “economic efficiency” defined specifically with respect to “utility,” which is a goal economists themselves provide, based on explicitly normative neoclassical welfare economics.
To the extent we’re talking about “economic efficiency” defined with respect to “utility,” a famous feature of the sort of “utility” defined in neoclassical welfare economics is that it does not support interpersonal comparisons of “utility.” Therefore, comparing two situations that differ in the sense of requiring the resolution of some interpersonal tradeoff is not something one can do within the framework of neoclassical welfare economics, even if one situation is “economically efficient” and one not. Suggesting economists break distributional indifference by arguing for any instance of an “economically efficient” outcome in that case, ignoring interpersonal tradeoffs, is not proper neoclassical welfare economics, it’s bad economics in the conservative style. Basically, one would be implicitly taking sides on the ethics of resolving the interpersonal conflict, thus rendering the original advice oxymoronic. One cannot, in realistic contexts, pursue only “economic efficiency” without at the same time choosing winners and losers. It’s something that really only makes sense in the arbitrarily, restricted world of theoretical economics, what I call The Fairy Land Of Economic Theory, in which any “economically efficient” outcome is literally equivalent to any other because some relevant ethics are exogenous. If one were in the attenuated, artificial Fairy Land of Economic Theory in which certain ethical issues are set aside, or maybe a hypothetical world in which everyone agrees those ethics, restricting oneself to “economic efficiency” would be uncontroversial, but not in reality.
What are the options for following real neoclassical welfare economics in realistic contexts? 1) Express indifference, pass the ethical issues to others. 2) Explain the theory expresses indifference but adduce explicit ethical arguments of one’s own to give policy advice. And, of course, 3) explain what “economic efficiency” really means, why it’s an attractive goal, if there are no other ethical issues to consider relating to resolving interpersonal conflicts of preferences, and really even individual preference rankings (ignorance, etc.). One can express any ethical ideas relating to the resolution of interpersonal conflicts of preferences one likes without pretending one is talking about neoclassical welfare economics, without resorting to bad economics in the conservative style. Just be honest about it. The problem with bad economics in the conservative style is not that it promotes conservative ideas relating to the ethics of the definition, distribution, use of economic power, but that it does so in an underhanded way. It’s bad economics and creates confusion and conflict.
