Following up on our recent discussion of the problems of considering utility theory as a foundation for economic analysis (which in turn was a reprise of this post from last September), somebody named Mark pointed me to a 2007 article by Luigino Bruni and Robert Sugden, “The road not taken: How psychology was removed from economics, and how it might be brought back,” which begins:
This article explores parallels between the debate prompted by Pareto’s reformulation of choice theory at the beginning of the twentieth century and current controversies about the status of behavioural economics. Before Pareto’s reformulation, neoclassical economics was based on theoretical and experimental psychology, as behavioural economics now is. Current discovered preference defences of rational-choice theory echo arguments made by Pareto. Both treat economics as a separate science of rational choice, independent of psychology. Both confront two fundamental problems: to find a defensible definition of the domain of economics, and to justify the assumption that preferences are consistent and stable.
Bruni and Sugden explain it all very clearly. To put it another way, basing economics on a science of rational choice, independent of psychology, makes about as much sense as basing chemistry on a science of chemical bonding, independent of physics.
At the same time, I don’t think economics is a subset of psychology, any more than psychology is a subset of biology, or chemistry a subset of physics. Supply and demand, prices, inflation, unemployment, and all the rest: these can be usefully studied without direct relevance to psychology research. At some point, the psychology does have to come in—just as the physics comes in to chemistry—but a lot of econ can be done at a more direct level, which is fine.
The political angle
What I want to write about here, though, is the connection between political views and attitudes toward economic foundations.
Many of the foundationalists and supporters of utility theory are politically conservative while its opponents tend to be on the left. Center-left economists such as Paul Krugman and Mark Thoma, along with economists further to the left such as John Quiggin and Peter Dorman, are skeptical of utility-based microfoundations (and, in the case of Dorman, this skepticism is long-standing, predating current economic debates by decades). On the other side, right-leaning economists such as Gary Becker rely heavily on utility theory and typically are negative about the use in economics of psychological measurements such as in happiness research. (See also this summary by Noah Smith.) And it goes back earlier than that: John von Neumann, axiomatizer of utility theory, was conservative too (and stood out from among most of his peers in this way).
This correlation between political attitude and intellectual orientation makes sense: rational choice is related to the idea that people are making the best choices for themselves, which in turn is related to the conservative ideas of minimal taxation and redistribution. This is not a strict logical relation—it could well be that redistribution could improve the outcomes from individuals’ rational decisions—but I see the connection. The liberal argues that the government should do X (for example, provide free preschool), the conservative retorts that parents who don’t send their kids to preschool are making an informed choice and the government shouldn’t interfere. Etc.
This all seems natural—but I could also imagine the correlation going the other way. A key theme of conservative thinking, now and in the past, is skepticism of systematizers, utopians, and “isms” in general. It would make a lot of sense to me if a liberal such as Paul Krugman (following the example of his liberal and technocratic forebear, Paul Samuelson) to lean heavily on a rational-actor model (perhaps tweaked a bit via prospect theory etc), and conversely for a conservative to distrust the clean simplicity of utility theory and, as a conservative, feel more comfortable with the real-world ambiguities of decision making as revealed by psychology research. Similarly, I could well imagine a pragmatic conservative resisting the rationalists’ backstop claim that, even if people are not individually rational, in aggregate they act as if they are behaving according to a utility function.
Recall my favorite quote of G. K. Chesterton, from the very end of his celebrated book on George Bernard Shaw:
I know it is all very strange. From the height of eight hundred years ago, or of eight hundred years hence, our age must look incredibly odd. We call the twelfth century ascetic. We call our own time hedonist and full of praise and pleasure. But in the ascetic age the love of life was evident and enormous, so that it had to be restrained. In a hedonist age pleasure has always sunk low, so that it had to be encouraged. How high the sea of human happiness rose in the Middle Ages, we now only know by the colossal walls that that they built to keep it in bounds. How low human happiness sank in the twentieth century our children will only know by these extraordinary modern books, which tell people that it is a duty to be cheerful and that life is not so bad after all. Humanity never produces optimists till it has ceased to produce happy men. It is strange to be obliged to impose a holiday like a fast, and to drive men to a banquet with spears. But this shall be written of our time: that when the spirit who denies beseiged the last citadel, blaspheming life itself, there were some, there was one especially, whose voice was heard and whose spear was never broken.
Chesterton was a Catholic conservative of the early 1900s, Shaw was a socialist, and both were famous for expressing their ideas in paradox.
Shaw, the leftist, associated progress with material happiness, while Chesterton, the rightist, said things were better in the Middle Ages. Nowadays, the debates usually go in the other directions, with people on the left being less positive about material progress and people on the right being more likely to be techno-optimists (setting aside politically-loaded topics such as the theory of evolution).
On to statistics
The other funny thing to me is that, in statistics, it seems to be considered conservative to be skeptical of models, to go for approaches that make minimal assumptions. And the hard-line econometricians hate models and want to go nonparametric all the way—even though, from my perspective, they give out the same vibe as the hard-line economic theorists who defend utility theory.