Google+
Close

The Agenda

NRO’s domestic-policy blog, by Reihan Salam.

Does Cognitive Science Take Political Sides? Five Push-Backs



Text  



Like David Brooks, I have unabashedly leapt on the brain-science bandwagon. I’m fascinated by every new cognitive science and social psychology study that shows up in my Google Reader, and I now think our new scientific methods of inquiry into human nature are improvements on the old humanistic methods and that their insights have very important and complex implications for other fields, including public policy.

Which is why it’s saddening to see the fascinating portrait of the human mind that these sciences are painting get ransacked and sliced up, little pieces of it distorted, simplified, and politically appropriated.

The standout example is the ubiquitous claim that the case for free markets has been demolished by the discovery of cognitive biases. ‘Free-market economics rest,’ the argument goes, ‘on the assumption that people are perfectly rational and selfish; but now we have uncovered gaps in our rationality that government must fill in.’ The idea is familiar enough to Agenda readers that I needn’t call out names. But it’s so widespread and evidently so persuasive to many people that we should tackle it. Here are five kinds of preliminary objections that come to mind:

(1)   Logical: The argument, ‘Free-marketers assume people are rational; people are not rational; therefore, oppose free markets’ is one big non-sequitur. Its logical form is equivalent to ‘Alice justifies position C with claim B; B is false; therefore C is bad, and its opposite good.’ Advocating more regulation requires an active, substantive and logical argument, not just a refutation of one basis for the politically opposed view. 

(2)   Straw-man: The argument claims too much of free-market economists. The assumptions of rationality used in Microeconomics 101 are assumptions in the mathematical, rather than colloquial, sense — abstractions used to build preliminary models. Advanced economics doesn’t actually assume perfect human rationality any more than advanced physics maintains the assumption that air is perfectly frictionless.

(3)   If you’re so smart…: Last December saw an excellent “Free to Choose” symposium which challenged the growing popularity and policy conclusions of behavioral economics. As the name suggests, the participants generally had libertarian bents. But many of their challenges are pure, positive social science. Geoffrey Manne’s argument is the most accessible; he points out that one well-regarded guru’s insights into behavioral economics haven’t paid off in the markets.

 

I fear this will come off as mocking or ad hominem, but the predictive challenge is a valid one for social scientists — if it were true that behavioral economics equips us to understand, predict, and compensate for the systematic flaws of markets, then the best behavioral economists should be extravagantly wealthy by now. They’re not, and this unremarkable financial ability should warn against trusting them to apply the conclusions they have derived from meshing economics with brain sciences in their current stages of development. The withering question laypeople used to ask economists was, ‘If you’re so smart, how come you ain’t rich?’ The question for today’s would-be behavioral-economic regulators is, ‘If you’re so unbiased…?’

(4)   Evolutionary generalization: In very broad generalization, it seems that cognitive bias could actually bolster the case for reliance on decentralized market processes as against the state. Competitive markets permit an evolutionary process in which bad (‘biased’) ideas fail, and the discovery of good (‘unbiased’) ideas, prices, etc., is rewarded. Government and central-regulatory monopolies cut off competitions which would select against biased beliefs. In a market, we can choose firms with more proven rationality; under government, we’re stuck with the biases of the regulators we got. So governments should predictably exhibit average levels of bias, while private firms that have survived the processes of competitive elimination would exhibit less.

 

There are obvious problems with this generalization. For example, a free market rewards the discovery of more effective (‘instrumentally rational’) ways to get kids hooked on cigarettes, but that’s not the kind of rational discovery we want to nurture. And then sometimes a case for more regulation can rest on belief in human rationality. (E.g. the case for greater capital requirements is based on the assumption that rational, selfish people will take fewer stupid risks if their own money is at stake.) These contrary examples point to the need for more…

(5)   …specifics: Cognitive bias plays a role in many incidents of both market and government failure. For example, a recent study from Bent Flyvbjerg (which Virginia Prostel wrote about here) shows that governments consistently and predictably overestimate benefits and underestimate the costs — including time and incidental damages — of infrastructure projects. The explanation is that we are susceptible to ‘optimism bias.’ We all, including contractors, invariably overrate our own facility in achieving our goals, and selectively ignore and underweight information about the difficulties we might encounter. Our representatives and their delegates want valuable projects for low costs. The two facts combined predict that the government will be most attracted to the most optimistically biased contract proposals. This is what Flybvjerg calls “Survival of the Unfittest.” This is a predictable government failure. And it’s easy to imagine many more examples of cognitive biases causing government failure. 

So the claim that advances in the brain sciences leave no escape from the conclusion that we require less-free markets is not such a simple truth. And while the insights of the brain sciences also don’t unambiguously support anarcho-capitalism, either, my feeling is that behavioral economists have generally been more interested in the market-failure than the government-failure implications of cognitive science, in a way that has biased the field to the Left. An impartial understanding of the implications of the brain sciences would require us to devote just as much effort to understanding their implications for government failure. It would be worthwhile for free-market economists to shift their efforts from (as in item #3) pecking at the weaknesses of behavioral economics in its current state, to exploring its implications for government failure (as in item #5). 



Text  


Sign up for free NRO e-mails today:

Subscribe to National Review