I was thinking a bit more about a point I made in an earlier post about the inherent logical absurdity of an (economic) efficiency versus “equity” (distributional ethics) tradeoff given how “economic efficiency” and “utility” are defined in economic theory (The Equity Efficiency Tradeoff and Fake Distributional Indifference, July 15, 2020), and it occurred to me this false tradeoff may come from conflation or confusion with a different tradeoff one sometimes hears economists and others discussing, the supposed tradeoff between total output and distributional ethics. The usual argument presented for such a tradeoff is that if one redistributes resources from how they would be distributed in a perfectly competitive labor and capital market or some particular expression of a perfectly competitive labor and capital market, then unfavorable incentive effects may reduce total output. In the vernacular, one may get a more ethical pie, but a smaller pie. The usual counterarguments in the context of the empirical issue of what happens to this metaphorical pie when we shift resources around are that some resources can be redistributed with minimal effect on incentives, and spreading resources around, especially to people who need it and would spend it as opposed to save it and invest it, will generate more robust consumer demand leading to higher economic growth and potentially a bigger as well as more ethical pie. What happens to the pie and why are interesting empirical questions, but that’s not what I want to discuss here. For my purposes, let’s just assume there is such a tradeoff, and if we were to address some particular notion of distributional ethics total output might suffer. What I’m interested in is the apparent contradiction between the normative scheme of neoclassical welfare economics based on “utility” as defined in economic theory, and the normative scheme expressed or implied by this notion we then have a separate normative judgment to make between total output and distributional ethics, as though total output were an independent ethical or normative consideration outside distributional ethics.
We know in the normative scheme expressed in standard neoclassical welfare economics, based on “utility” as defined in economic theory, there’s nothing special about any particular Pareto optimal (economic efficient) outcome relative to any other. They’re all the same in the context of that theory, although not generally, of course, as depending on one’s views on distributional ethics one will always be ethically superior to another. That’s how we end up with the Pareto frontier, the set of all possible Pareto optimal outcomes based on all possible distributions of economic power. That’s why neoclassical economic theory, properly interpreted, can be said to eschew potentially controversial distributional ethics. That’s what distributional indifference in neoclassical welfare economics is all about. In particular, under that normative scheme, it doesn’t matter if total output is the same across all Pareto optimal outcomes, and as far as I know there is no argument in economic theory that says total output will necessarily be the same across all such outcomes. Indeed, if one contemplates the more extreme potential Pareto optimal outcomes with economic power concentrated in the hands of one person or a small group of people it seems implausible to suggest total output could possibly be the same as under some other Pareto optimal outcomes.
However, in our alternative normative scheme, based on total output, there may be something special about certain Pareto optimal outcomes in realistic settings at least because of the presumed effect of the changes in incentives associated with redistribution. Indeed, even moving between Pareto optimal outcomes with equal total outputs becomes ostensibly inadvisable in a normative or ethical sense because of the at least temporary disruption in total output supposedly generated by the disruption of incentives to get there. So if one is at a Pareto optimal outcome then maintaining the status quo takes on normative significance. One is no longer truly indifferent between that outcome and other Pareto optimal outcomes. One certainly wouldn’t be indifferent to redistribution for the sole benefit of one person if total output tanked as a consequence. One would have a built-in argument to maximize total output unless someone else presented a stronger ethical argument to move to a different outcome, which of course would be impossible in the context of economic theory given the definition of utility and the explicit avoidance of distributional issues within that theory. So in the context of distributional disputes economists become players, the great champions of total output as the overriding ethical concern.
The two normative schemes make strange bedfellows. If one likes pie, one may note one normative argument says rather sensibly when it comes to pies bigger is better, but if one then suggests as a matter of empirical fact slicing the pie in certain ways shrinks it in the future, one finds oneself at odds with the principle of distributional indifference so central to the other normative argument. One is ostensibly making a bigger pie to potentially share, but logically one never could share it, because if one ever did one would no longer be making a bigger pie to potentially share.
One surmises some confusion between real ethical utilitarianism and the fake utilitarianism of neoclassical welfare economics may be involved. Total social welfare may be important under the former, but not the latter. The latter is confined to matters of individual preferences. In the former having a big total output to share buys one a lot of stuff one might suppose is related to total welfare, in the latter the whole world might tank, but if even one person thereby improves his or her situation one should be indifferent. They’re different normative arguments.
To put it another way, there’s no such thing in economic theory as a total amout of “utility” we can distribute to different people. There is of money or economic power or total output, but not “utility” as defined in economic theory. Or to put it yet another way, if one tries to express the tradeoff in terms of utility alone as an “efficiency equity tradeoff,” it doesn’t really make sense, as a I discussed in the post I referenced at the start. One needs to shift gears from talking about “utility” to talking about total output to get the tradeoff.
Don’t get me wrong here. They’re both perfectly legitimate normative arguments one might want to consider. I’m not arguing that. Indeed, worrying about total output may very well be more sensible that confining oneself to thinking about “utility” as defined in economics. But they’re different. Neoclassical welfare economics says one, not the other. Don’t get them twisted. If one intends to speak on behalf of economic theory one should represent it correctly. If one wants to step outside economic theory and express one’s own personal values or normative judgments, no matter how commonplace one may suppose them to be, then one should make sure everyone understands that’s what one is doing and doesn’t suppose one is speaking in one’s role as an economist or about economic theory.
How many economists express normative or value judgments they at least imply are based on neoclassical welfare economics that actually contradict that theory? How many economists use both normative schemes we’ve discussed here or switch between them willy-nilly depending on the rhetorical needs of the moment? Is there any mystery where purveyors of bad economics get their ideas? Neoclassical welfare economics has been around a long, long time and learning that theory is, I believe, widely associated with what it conventionally means to be an economist. Funny there’s this type of confusion going on just below the surface. Funny even if there’s a perfectly sound explanation I simply can’t think of at the moment because, believe it or not, I’ve studied economics, and I occasionally bother to think about such matters. Well, funny if anyone is sincerely trying to address it. I suppose entirely expected otherwise.
Neoclassical economic theory is a confusing hash of science, math, and ethical philosophy that often generates or at least facilitates confusion relating to the methods and results appropriate to each. Many economists and purveyors of bad economics are bad ethical philosophers. Even assuming no active intent to deceive, they don’t seem to care too much about their own normative or ethical arguments or expressing them carefully and accurately and addressing potential inconsistencies and errors. Their ethical thinking tends to have a rather markedly casual and anything but rigorous character. The ethical or normative aspects of evaluating economic systems and outcomes should be removed from economic theory and the purview of economists and put with democratic decision making bodies and the people. We should one day make economics a real science and economists real scientists.
