I was talking to a helpful fellow the other day about “positive economics” and how I might be unaware that economic concepts like “optimality” and “economic efficiency” were not always really used in a normative sense in the context of neoclassical economic theory because they have technical and formal definitions that make them really part of positive economics. It was an interesting point of view that made me wonder whether it might be worthwhile to just take a step back and say a few words about what we mean by positive and normative in the context of economics.
Positive economics is about what is true empirically, factually, scientifically, or about what is true formally, logically, mathematically. In contrast, normative economics is about what one believes one or others should so, evaluating and assessing economic systems and outcomes, defining goals and objectives, giving policy recommendations. Neoclassical welfare economics is, at least under normal conditions, normative economics. It purports to evaluate or assess economic systems and outcomes. It purports to show the advantages of market systems. It carries implicit or explicit recommendations about what people should do as far as economic arrangements that go beyond the question of what is. It can and should be distinguished from neoclassical economics more broadly in the sense the theoretical framework or structure of neoclassical economics more broadly can be used to construct positive, scientific economic theories or models meant only to predict or explain empirical phenomena. One can take seriously someone who claims to do positive neoclassical economics; however, one cannot take seriously someone who claims to do positive neoclassical welfare economics.
The potentially confusing bit is that even a normative or ethical theory like neoclassical welfare economics will always have positive bits. Ethical arguments typically involve value premises, factual premises, and logic or math to tie them together. The latter two maybe considered positive elements or inputs to the normative argument. Neoclassical welfare economics doesn’t contain any real factual or scientific premises I can think of right now anyway. The famous behavioral assumptions are tautological and other potentially factual elements are generally considered false but useful simplifications. However, neoclassical welfare economics does, of course, use logic and math. If one confines oneself to the mechanics of the normative argument presented in neoclassical welfare economic, the logic and math, one may suppose one is doing positive neoclassical welfare economics. However, that is incorrect in nearly all cases.
If neoclassical welfare economics is being used in a normative way, what one is actually doing when one confines one’s attention to the positive bits is not really positive neoclassical welfare economics but bad normative neoclassical welfare economics. One is basically doing normative economics but failing to identify and evaluate some or all of the necessary normative or value inputs for consistency, philosophical plausibility, and ethical controversy, and hence ignoring the crucial issue of whether the logic and math combined with whatever value inputs one has chosen to specify leads to the normative or value outputs or conclusions one is purporting to establish. One is restricting oneself to evaluating the form of the argument rather than the content of the argument, but when it comes time to actually use a normative theory in a realistic or practical setting the content matters.
Technically, of course, one could do a purely positive form of neoclassical welfare economics as a sort of parlor game with no significance for the real world and by strenuously refuting anyone who tried to misuse in a normative way to, for example, to evaluate or assess economic systems and outcomes, to recommend market resolutions to resolving interpersonal conflicts of needs, wants, and desires. However, if we’re going to keep it real, no economist really does that. Not only do economists fail to confront those who use neoclassical welfare economics as a normative theory, they typically end up doing so themselves, if not explicitly then implicitly, which is probably pretty much inevitable given the normatively charged language and concepts involved. If economists want to continue working with neoclassical welfare economics, they should acknowledge it as normative economics, take the identification and assessment of the normative or ethical or value inputs and the issue of whether those inputs support the normative or ethical or value conclusions seriously. They should pause their mathematical research and brush up on their ethical philosophy.
