Yes, The Assumptions Of Neoclassical Economics Are Unrealistic; And Yes, You Are Allowed To Criticize Them
According to popular rhetoric, there are ways to criticize neoclassical economics, and then there are ways not to criticize neoclassical economics. A big No-No, the experts say, is the argument that the profession is flawed because its assumptions are too unrealistic.
Economists recognize that their assumptions are unrealistic. Most, I hope, can see that a Walrasian general equilibrium is not something that actually exists in the real world. Nevertheless, they still think it is a useful simplification that can yield valuable insights, a perfectly reasonable position to be sure. After all, this is how science is supposed to work: simplifying assumptions need to be made in order to make models tractable. Physicists do this all the time. For example, they often need to assume “zero friction” in order to zoom in on other, perhaps more interesting, properties of motion. Anyone criticizing neoclassical economics for using simplifying assumptions is therefore usually seen as ignorant for not understanding the scientific method.
But allow me to argue the opposite in this post: namely, that one can – and indeed one should – criticize the simplifying assumptions that neoclassical economists make. Why? Well, because economics is different than physics of course!
The subjects studied in physics – e.g., atoms, planetary masses, etc. – do not change after physicists study them. The same is not true in economics. When economists model social phenomena, they’re not just describing “what is,” as they so often proclaim. On the contrary, by modeling social phenomena, economists actually end up changing what is.
This performative link is made because although economists recognize that their simplifying assumptions shouldn’t be interpreted literally, the general public does not often make the same recognition. For example, when economists arrive at a conclusion like the first fundamental theorem of welfare economics (FFTWE), which states that any Walrasian equilibrium is Pareto-efficient, they recognize that Pareto efficiency is an interesting theoretical result but not one that can be observed in reality. But people outside of the economics paradigm don’t usually proceed with the same level of caution. In particular, policymakers often see conclusions like the FFTWE as evidence that not only are Pareto improvements desirable, but also that the key to achieving them is to structure markets in such a way that mimics a Walrasian setting. Hence, we must cut taxes, deregulate, and encourage people to act more self-interestedly, the politicians argue.
Of course, after the neoliberal policies are enacted, the economist inevitably comes back defensively, saying that the FFTWE was misconstrued; that we shouldn’t blame the model, but rather that we should blame those who jumped to erroneous conclusions. The way I see it, though, that’s not a sufficient response to get the economist off the hook. Indeed, economists can’t just stand back after their models are published and claim innocence; they have a responsibility to make sure that their simplifying assumptions aren’t interpreted in the wrong way, so that politicians don’t use them to sway policy.
In my opinion, it seems that there’s a cost-benefit calculation that needs to be done here. On the one hand, there are benefits from making simple and tractable models. Although Walrasian equilibriums don’t exist, thinking about them still helps to clarify some things – perhaps making them a good “scratchpad,” as Paul Krugman would say. On the other hand, there are (potentially huge) costs if policymakers misinterpret economic models.
My guess is that most economists would be unhappy with this sort of cost-benefit analysis, as it would likely significantly restrict their modeling efforts. Fortunately, there is a way forward: economists could drop the questionable declaration that their work is positive and admit that their profession is a fundamentally normative endeavor.
In other words, economists could stop offering allegedly value-free statements, such as: “These are the conditions that will lead to Pareto efficiency.” They could, on the other hand, make it clear that Pareto efficiency is something that they find normatively attractive, and that they have a step-by-step blueprint for getting us there. If they make this admission, then there are no modeling restrictions; economists will be able to model whatever they want, whenever they want. And they won’t need to worry about their results being (mis)interpreted by policymakers, because that’s precisely the point of any normative endeavor.
But – and this is a big BUT – once economists step into the normative playing field their models are fair game for political philosophers to scrutinize. For example, if an economist normatively defends the FFTWE, the political philosopher can respond by saying, “That’s an interesting result, but are you aware that you’re measuring welfare by looking at revealed preferences? And are you aware of the problems of measuring welfare by looking at revealed preferences? Have you read Martha Nussbaum’s criticisms of subjective welfarism? What happens if preferences are adaptive? Is it really possible to “purify” preferences à la Dan Hausman?”
At this point, most economists will likely have no idea what the political philosopher is even talking about. In fact, many may see the political philosopher’s comments as merely sloppy thinking and hand-waving theorizing, without acknowledging the obvious reality that the world is too complex to be modeled by simply observing buying-and-selling decisions in the marketplace.
The political philosophers have been desperately awaiting the day when economists admit that they’re doing normative philosophy and not positive science – when they stop running from normative questions, and stop citing Milton Friedman ranting about leaves growing on a tree or something. If economists are unwilling to make this admission, then holding them responsible when their models are misinterpreted by the general public is fair game. And if their simplifying assumptions are the cause of the misinterpretation, then it is perfectly acceptable to criticize them.