Seems it’s been a while since I did a post devoted to what I called “fake distributional indifference,” a rhetorical ploy or technique interesting to me because it takes so many different forms and serves as one of the main points of connection between neoclassical welfare economics proper and bad folk economics of the “government should not interfere with the free market” variety. I’ve noticed discussions in this area often seem to get just that bit more confused when so-called “social welfare functions” enter the room, so maybe I’ll address fake distributional indifference in that context today.
As you may recall, “social welfare functions” are the intellectual Frankenstein’s monster of neoclassical welfare economics whereby certain ethical or normative values exogenous to neoclassical welfare economics and its definition of “utility” can be imported and expressed as “weights” attached to that “utility.” Much like time travel in the third installment of Austin Powers, it’s probably best if one doesn’t think about it too much. What exactly are we meant to be weighting, anyway? “Utility” as most commonly defined in neoclassical economics doesn’t even exist, per se, but is really just a funny way to talk about the preference rankings of individuals. But never mind. As seems frequently the case with overly formalized neoclassical economics, the question of whether the concepts are sensible and consistent with one another is less important than whether some particular mathematical manipulation or other gets one where one wants to go.
Given enough information on the personal characteristics and behaviors potentially relevant to distributional ethics, social welfare functions theoretically allow one to express any form of distributional ethics based on people, including concerns imported from traditional ethical utilitarianism relating to human welfare but also from other potential ethical frameworks focusing more on merit and rights and fairness and so on, albeit by confusingly slapping weights on the “utility” used in economic theory that are, of course, seemingly arbitrary within the normative framework of that theory itself and, indeed, not entirely consistent with the ethical or normative values expressed in that theory itself. However, that doesn’t mean social welfare functions are necessarily capable of addressing all ethical issues associated with resolving interpersonal conflicts of needs and desires on the basis of economic power in markets. Trying to capture other potentially relevant ethical issues using social welfare functions, issues not based on the characteristics or behaviors of the people involved but something else such as, for example, issues relating to the extent of the market, that is, issues related to when we should resolve interpersonal conflicts on the basis of economic power in markets and when to use some other mechanism, seems a rather more difficult project. I’m not saying it’s impossible, but it certainly seems quite awkward. It would presumably involve some weighting scheme that not only covers personal characteristics and behaviors but characteristics of the situation or context in question. A little indicator of whether we’re talking about vaccines or automobiles and so on. Personally, I’ve never seen a social welfare function like that, but economists can be quite ingenious when they choose to be, so maybe one day. And, of course, using social welfare functions to address the ethical issues submerged by false factual premises like perfect rationality or information, even though they concern the characteristics or behaviors of people, can present certain difficulties in the context of models requiring those premises to derive their results.
Although the theoretical availability of social welfare functions is commonly considered a significant game changer for the normative significance of neoclassical welfare economics by many economists, with some economists of both the real and armchair varieties going so far as to suggest it renders neoclassical welfare economics ethically or normatively neutral, it really doesn’t change the basic conceptual or intellectual structure of neoclassical welfare economics at all from what I’ve been calling an ethical half-theory. Some relevant ethical issues associated with resolving interpersonal conflicts on the basis of economic power in markets remain endogenous to the theory and some remain exogenous, some relevant ethical issues can be derived from within that theory and some cannot, some ethical reasoning is in, some out. In particular, the ethical or philosophical rationale for any particular social welfare function obviously remains exogenous to economic theory. The only thing social welfare functions change is that rather than applying exogenous ethical principles to the conclusions of the ethical half-theory of neoclassical welfare economics, one can import at least some of the exogenous ethical content into economic theory and express it in a funny way using social welfare function to make a muddled pseudo-economic conclusion, which still likely requires exogenous ethical principles applied after the fact to address relevant ethical issues not expressed in the social welfare function.
So what does fake distributional indifference look like in the context of social welfare functions? Same as always, with a slight change in terminology. All real markets and market outcomes come bundled with characteristics relevant to controversial ethical issues related to resolving interpersonal conflicts on the basis of economic power in markets that are explicitly set aside or implicitly ignored or avoided in neoclassical welfare economics. In the case of those ethical issues expressed in social welfare functions, one might say all real market outcomes come bundled with considerations potentially relevant to some social welfare function. As a result, one cannot express distributional indifference, or indifference with respect to other controversial ethical issues that are or should be in the same boat as distributional ethics, while supporting any real market outcome. If one supports any real market outcome, one is necessarily expressing at least an implicit opinion about the ethical considerations bundled with that outcome, including those relevant to some explicit or implicit social welfare function. Not caring or talking about those ethical issues or implicit social welfare functions doesn’t render supporting particular real market outcomes neutral. One is still expressing support for one ethical position over another, albeit possibly for hidden reasons or perhaps even for no reason at all.
Note that this remains the case even if one is thinking about classes of market outcomes because, of course, we never encounter a class of market outcomes in real life. The class is a conceptual or intellectual construct. We encounter real, specific, defined markets and market outcomes or, if you like, particular instances of a class. Preferring any real instance of a perfectly competitive market outcome to a non-perfectly competitive market outcome that theoretically may be more highly ranked under any potential social welfare function is not being neutral with respect to distributional issues. It’s explicitly or implicitly supporting one social welfare function over another, or more broadly, one set of ethical beliefs relating to evaluating market systems and outcomes over another. It goes beyond what one can say based on neoclassical welfare economics alone.
Let’s just go over some quick takes on fake distribution indifference in general to make sure we’re all getting what’s going on here.
Can one say based on neoclassical welfare economics that any real instance of a perfectly competitive market is ethically or normatively superior to any real instance of a non-perfectly competitive market? No. That’s an artifact of fake distributional indifference.
Can one say based on neoclassical welfare economics that if one is at a real instance of a perfectly competitive market outcome one should stay there because there is no legitimate reason to change in economic theory? No. That’s an artifact of fake distributional indifference.
Can one say based on neoclassical welfare economics one should not interfere with a real instance of a perfectly competitive market? No. That’s an artifact of fake distributional indifference.
In the presence of distributional indifference, there’s nothing ethically or normatively special about any real instance of even a perfectly competitive market relative to any market outcome that differs from it along the distributional dimension, or to be consistent, along the dimension of any controversial ethical issue that should be treated similarly to the distributional dimension, and certainly nothing special about any real market that fails to qualify as a perfectly competitive market. To assess or evaluate a real market system or outcome, one must address all the relevant ethical issues associated with it, those expressed in neoclassical welfare economics and those set aside, those expressed in social welfare functions and those not so expressed. One can, of course, support any market system or outcome one likes on any ethical or normative basis one likes, but one should do so honestly and not pretend one’s ethical or normative judgments are based in neoclassical welfare economics in situations where they clearly are not and, as a matter of logic, cannot be. Addressing the relevant ethical issues involved in evaluating market systems and outcomes is a job for everyone involved operating through democratic government, not philosophically inept or underhanded economists.
