Greed Is Good Two: Methodological Individualism and Reductive Utility - Revised

Normally I don’t like to double up like this, but I was having a fun conversation with a fellow economist the other day about whether standard neoclassical economic theory is or is not based on the notion people are selfish, which you may recall was the topic of my last post.  Setting aside the actual conversation for just a moment, I must say I find it suggestive of bad intent on the part of academic economists that people trained as economists are still having discussions like this, given that neoclassical welfare economics has been around in its current form for about one hundred years or so now.  Makes one wonder what they and their fellows in academic philosophy have been doing all this time.  But never mind, let’s try to talk it out a bit more right now, shall we?

My interlocutor argued that because economic theory analyzes human behavior in terms of individual utility it implies people behave to satisfy their own desires and not for the good of others.  He called this characteristic of economic theory “methodological individualism” and described it as “reducing” utility to the individual, which was terminology I must confess I don’t normally use myself.  He suggested if we broke this reductive conception of utility by assuming the utility functions of different people are not independent but interrelated in some way, then the mathematics wouldn’t work, and the theory would crumble in his own evocative language.  In contrast, I’ve argued economic theory is not really based on the notion people are selfish but it’s a common misperception that it does.  We each talked past one another for a bit and left it at that.  However, after thinking about it a little more I think I may now understand what he was getting at.  Let’s discuss.

First, let me say I have no doubt at all my colleague was correct about the methodological individualism, the reductive conception of utility, and that the mathematical model of the economy associated with neoclassical welfare economics will not work as intended unless the actors in that model have independent utility functions.  However, I’m rather less sure those features of economic theory rule out actors in economic theory thinking about and acting for the good of others as a part of the individual desires they’re trying to satisfy.  

Clearly, given the unobservable nature of utility as defined in economic theory, one actor could never have another actor’s utility literally show up in the first actor’s own utility function.  How would it get there?  However, what’s less clear to me is whether observable phenomena potentially relevant to the assessment of another actor’s utility under some definition of utility relevant to some interpretation of utilitarian ethics, or indeed any other observable characteristic relevant to any other of the variety of ethical beliefs we find in the real world, could not be incorporated into the utility function of a given actor in economic theory.  That is to say, it’s not clear to me the utility function of a given actor in the economic model associated with economic theory could not include utility that actor derived from fulfilling his or her own individual preferences about behaving ethically toward other actors based on whatever observable phenomenon that first actor was assumed able to perceive in that model, and that includes notably fulfilling those individual preferences by working to create a society expressing those preferences.

Nor do I think it really precludes the appearance in one actor’s utility function of utility for that actor generated by other people acting in ways that fulfill that first actor’s preferences related to what those other people should be doing, what I’ve previously called second party preferences.  (Not to be confused with the situation addressed by the de gustibus non est disputandum principle, which refers to one person second guessing someone else’s choices on the basis of that other person’s own utility.)  In terms of the conceptual analysis of the ethical component of economic theory I’ve presented elsewhere, what happens when one person derives utility from preferences relating to what another person does is that every situation involves or might involve a conflict of desires that renders economic theory irrelevant because of our inability to make interpersonal utility comparisons using the definition of utility used in economic theory.  Well, unless one could set up a market for that sort of thing, then I suppose it would just be one more externality one could potentially address on the market under any particular distribution of economic power, so it would fall into the same category as all other such interpersonal conflicts resolved using economic power on the market.  But is there or could there ever be such a market?  Seems complications may ensue because of such factors as the difficulty of setting up contracts of that sort, enforcement, numbers of people potentially involved, etc.  And how do we handle the situation analytically if a market cannot be generated?  Interesting.  However, I seem to be straying rather far from the point of this post, so let’s just leave these issues for some other day, shall we?  

So what’s the point?  Just that I’m not convinced methodological individualism and the reductive conception of utility really implies economic theory treats people as acting like self-centered amoral jerks.  That issue, of course, is conceptually distinct from the issue of whether neoclassical welfare economics expresses the ethical proposition people should act like self-centered amoral jerks, that is, that greed is good.  However, my contention that economic theory does not necessarily treat people as acting like self-centered amoral jerks certainly seems to undercut one such way such an ethical proposition could enter neoclassical welfare economics, via a logical or factual mechanism that serves as a necessary element to the ethical argument. 

Addendum

I have more recent thoughts on this issue of methodological individualism, for example, see the post Methodological Individualism from May 19, 2021, or the related addendum to my post Does Neoclassical Economics Really Propose Greed is Good from February 12, 2020. I also have more recent thoughts on the issue of “greed is good,” for example, see the post on Actors and Roles in Neoclassical Welfare Economics from February 24, 2021.

Does Neoclassical Economic Theory Really Propose Greed Is Good?

Neoclassical microeconomic models often include the assumption the actors in those models obtain utility only from the consumption of specific manufactured physical goods as well as the assumption their desire to consume those goods is insatiable.  Taken together one may fairly describe the assumptions as suggesting the actors in those models behave like greedy, self-obsessed, materialistic, amoral, jerks obsessed with fulfilling their base material desires for the goods in question with no evident concern for higher ethical values or really any higher values of any kind.  However, when one sets aside those particular and essentially arbitrarily defined models and instead looks at the rudimentary ethical argument presented in economic theory meant to identify ostensibly optimal social structures, that is, the system of factual postulates, value propositions, and the logic that ties them together to produce a conclusion, there seems to be no such restriction on acceptable sources of utility.  For example, under this more general theoretical perspective, if someone freely chooses to follow some course of action A rather than B, revealed preference in conjunction with the de gustibus non est disputandum principle suggests we can infer that person derives more utility from A than B, no matter his or her motivation and in particular no matter if his or her motivation involves ethics, a concern for others, etc.  In other words, the ethical or value arguments presented in neoclassical economic theory seem to neither require nor imply people are greedy, materialistic, amoral, jerks.  

So what exactly is the status of the greedy jerk assumption in economic theory?  Well, clearly in a scientific context we can view it as a non-essential simplifying assumption one can pull out of the bag or put back in the bag depending on how well one’s model is performing on the basis of predicting and explaining human behavior and whether one’s model contains actors with characteristics and behaviors that any other actor in that model might reasonably have ethical beliefs about.  In this context, we’re not necessarily saying people are greedy jerks or should be greedy jerks, we’re just saying maybe we can model their behavior to some degree of accuracy by presenting them as acting that way for purposes of our predictive or explanatory model.  But what about in terms of the factual postulates, value propositions, and logic of the rudimentary ethical argument meant to establish optimal social arrangements and market structures?  Is the greedy jerk assumption necessary in that context?  Does the logical argument about how we can maximize total utility across society by setting up certain market structures stop working if we allow that people may derive utility from following ethical principles and other higher level motivations and not just from selfishly interacting with physical goods?  Or are we doing something funny with utility that contradicts how we’ve defined utility in economic theory?  Does revealed preference not apply in that context?  Does the de gustibus non est disputandum principle not apply in that context?  Do we really need to get in there and assess people’s motivations before we can determine if they’re really maximizing their own utility when they make consumption decisions?  Do we need people to act like that to get to our socially optimal results?  If they don’t, should economists play ethical philosopher and try to persuade them to act that way for the good of society?  If people just don’t get it or won’t play along and instead insist upon acting ethically should economists support constraining their choices and forcing them to maximize their utility properly so we can get to a socially optimal result?  Maybe hold them down and shove pie down their pie holes?  And since being or behaving like greedy jerks would seem to preclude holding any particular ethical belief relating to distributions, what are we to make in this case of the famous indifference within economic theory to distributional concerns?  We’re saying we’re indifferent to distributional concerns but as a matter of fact no one has any or perhaps as a matter of ethics no one should have any?

I’ve argued before and in print that the rudimentary ethical argument neoclassical economic theory uses to establish ostensibly optimal market arrangements doesn’t really require the assumption that people are greedy, materialistic, amoral, jerks and indeed that assuming it does require that assumption amounts to an error of interpretation that serves to shut down conversation about distributional issues.  Indeed, aside from the obvious fact that people do clearly have ethical beliefs and clearly do act upon them, the proposition no one has any ethical beliefs would seem to put economists, with all their talk of maximizing total utility and social optimality and so on, in the awkward position of being the exception to their own rule.  Very special people who do in fact have ethical beliefs, ostensibly utilitarian ethical beliefs, and are interested in acting accordingly.  If they’re not very special, if they’re the same as everyone else, then they shouldn’t be contemplating social optimality and should be thinking instead about what works out best for themselves in a distributive sense, perhaps how economists can establish a monopoly of everything or pull the wool over people’s eyes or something along those lines.  I’ve also argued the ethical proposition that people ought to behave like greedy, materialistic, amoral, jerks is not a logically necessary part of the ethical argument in economic theory about what we need to do to reach a social optimum.  At least, if it were, then there could be no sense in which economists or economic theory could be indifferent with respect to distributional issues.  We would say be saying that as a matter of ethics no one should have any distributional beliefs, including supporting any particular system of property rights or the existence of markets themselves, and if they do they should stop it right away.  So confusing when one tries to do science and ethics at the same time, isn’t it?  But the conclusion I think must be that for the purposes of the ethical argument put forward in neoclassical welfare economics about what we need to do to create socially optimal market structures, people acting like greedy, materialistic, amoral, jerks is not a necessary condition.

Addendum 

I eventually decided the best way to describe what’s going on here is to think of neoclassical welfare economics as an ethical half-theory in which some relevant normative or ethical issues are purposefully excluded, and to draw a distinction between what I begin calling at some point the Fairy Land of Economic Theory on the one hand and real life on the other, in which those issues cannot be so excluded. The restriction on preferences relating to distributional ethics applies to the subjects of economic theory in the Fairy Land as a precondition to deriving results predicated on Pareto improvements defined with respect to “utility,” which won’t work in the presence of distributional beliefs relation to relative “utility,” but it doesn’t translate to restrictions in real life. Rather, applying an ethical half-theory to real life require exogenous ethical content. Neoclassical welfare economics on its own is inapplicable to real life. For example, see the post on Methodological Individualism from May 19, 2021.

The Contributions of Neoclassical Economic Theory

The good part of neoclassical welfare economics, the valuable contribution to knowledge, is the attempt to see how far we can get analyzing the desirability of different market structures using certain factual premises regarding human behavior while avoiding controversial ethical issues especially distributional issues (or most of them anyway, we get a few potentially controversial ethical issues as soon as we decide to resolve interpersonal conflicts of needs and desires using market mechanisms and at least one potentially controversial distributional issue as soon as we decide a system of property rights to set up a market in the first place, as any anarchist worth his or her salt will be happy to explain).  In this context we can show some market structures are better than others, if there are indeed no controversial ethical or distributional issues, that is. The conclusion, if the factual premises are true, is that if one has a choice of market structures, all of which are consistent with one’s ethical views and in particular one’s distributional concerns, then a perfectly competitive market structure looks pretty good.  It’s unfortunate economists have never been able to leave it at that but instead try always to confuse the issue relating to the factual premises of the argument and also to apply economic theory in situations that clearly involve (other) potentially controversial ethical issues especially distributional issues where it generates rather more heat than light and obscures the real issues rather than illuminating them.  Economists really have no business showing us their equations and graphs when the real issues are conflicting values and ethics.  Is it hubris?  Is it egotism?  Is it chicanery?  Is it stupidity?  Is it rhetorical cleverness?  Are they just trying to foist their own distributional values on unsuspecting people while shielding those views from criticism and debate?  Are they hired guns for people who make out well under certain distributional arrangements and don’t want to argue for those arrangements honestly and openly?  So hard to tell, isn’t it?  But one thing is for sure.  When bad economists misuse neoclassical economic theory in ways and in contexts for which it was never designed they can do a good deal more harm to society than good.

Addendum

The factual premises used to remove the ethical controversy from the inputs to neoclassical welfare economics, assumptions like perfect information and perfect rationality, are never true except in what I call the Fairy Land of Economic Theory. To the extent one has evaluated the relevant normative inputs under those factual premises, one’s normative evaluation of the theory will be inapplicable to the real world.