Kenneth Arrow on Behavioral Economics, etc.

A few weeks ago I blogged about Conor Clarke's insightful interview with Paul Samuelson. Well, Mr. Clarke is back at it, and he recently conducted an interview with Nobel prize-winning economist Kenneth Arrow (part one here, part two here, and part three here). Arrow is best known for his contributions to general equilibrium analysis, welfare economics, and Arrow's Impossibility Theorem. The interview addresses subjects such as healthcare reform, behavioral economics, and climate change. Here are a few things I found interesting (interview questions are bolded):

What do you think of behavioral economics?

Yes, there is a tendency to go to these psychological arguments. But my problem with these arguments is not that they are wrong -- though sometimes there are wrong -- but that they are not helpful.

Why?

They don't predict anything!

I agree that behavioral economics is only useful if it increases the accuracy of predictions made by economists (see here and here). Nevertheless, I wouldn't go so far as to say that behavioral economics doesn't "predict anything." Behavioral economics does a good job of predicting how people make retirement decisions (hint: they follow the status quo, so default rules are exceptionally important determinants of behavior).

Back to the interview - Arrow on the uncertainty surrounding climate change:

The treatment of uncertainty [with climate change] strikes me as one of the most confusing things about climate change. . . . But with climate change there is an awful lot at stake! And the time horizon is long, and you have the risk of these incredibly high cost but low probability events...

Well, the least that can occur still looks pretty bad, and I don't want to overdo the uncertainty question. I think that, even if you take a very conservative point of view, things are bad. 150 years from now, the average person will be a lot poorer then they would be otherwise be.

On health care costs:

The basic reason why health costs increased is that health care is a good thing! Because today there is a lot more you can do! Consider all these expenses that are diagnostic. Cat scans, X-rays, MRIs and now the proton-powered whatever-it-is. Something that is the size of a football field, cost $50 million, and has all sorts of diagnostic powers. A lot of these technologies clearly reveal things that would not be revealed otherwise. There's no question about it. Diagnostics have improved. Technology has improved. You know, sending things through your blood stream to help in operations, instead of cutting you open. It's incredible. But these things are costly. But for older people longevity is increasing by a month each year. Now, whether that creates other problems with retirement and social security is another question. But, nevertheless, preserving life is a good thing.

On the erosion of professional standards:

And if one explanation is the professional standards, why do you think the professional standards have changed?

Sometimes I think it's because of the Chicago School.

I'm not sure if the last one was a joke, but blaming the Chicago School for the decline in professional standards seems a bit over the top. Still, I would definitely recommend reading the entire interview.

Don't Text and Drive!

According to a Virginia Tech Transportation Institute study, texting while driving is 23 times more dangerous than driving while not distracted. By comparison, drunk driving is "only" seven times riskier. Unsurprisingly, the study's authors encourage regulators to ban texting while driving:

Texting should be banned in moving vehicles for all drivers.

I couldn't agree more (though I have been guilty on occasion). Talk about a massive negative externality!

Posner v. Thaler - Behavioral Economics

Richard Posner, one of my academic heroes (see here and here), recently criticized the Consumer Financial Protection Agency Act of 2009 in the Wall Street Journal. Although I don't know the details of this act, I was somewhat surprised that in his editorial, Posner went further and criticized behavioral economics and Richard Thaler, one of behavioral economics' strongest proponents:

The plan of the new agency reveals the influence of “behavioral economics,” which teaches that people, even when fully informed, often screw up because of various cognitive limitations. A leading behavioral economist, Richard Thaler of the University of Chicago Booth School of Business, wrote “Nudge: Improving
Decisions About Health, Wealth, and Happiness” last year with Cass Sunstein, who is President Barack Obama’s nominee for “regulatory czar.”

Mr. Thaler, whose views are taken seriously by the Obama administration, calls himself a “libertarian paternalist.” But that is an oxymoron. He is a paternalist with a velvet glove—as the agency will be. Through the use of carrot and stick, the agency will steer consumers to those financial products that it thinks best for them, whatever they naïvely think.

. . .

Behavioral economists are right to point to the limitations of human cognition. But if they have the same cognitive limitations as consumers, should they be designing systems of consumer protection?

Two things struck me about this argument. First, libertarian paternalism, at least as it is defined by Thaler and Nudge co-author Cass Sunstein, is not an oxymoron. They argue this point forcefully in an article called "Libertarian Paternalism Is Not an Oxymoron." This article is one of the forces that converted me to the importance of behavioral economics. Among other things, the article argues that regulators will inevitably influence behavior, since default rules (such as voluntary or mandatory enrollment in retirement plans) affect decisions.

Second, although it may have been tongue-in-cheek, I am surprised by Posner's claim that regulators, because they have cognitive limitations, cannot encourage people to make better decisions. This is a poor argument. Regulators have time to make considered decisions, many consumers do not. Regulators are specialists, most consumers aren't. I was pleased to read that Thaler, in a response to Posner, makes very similar points:

The premise of behavioral economics is that humans are not perfect decision-making machines. We are busy and distracted. We have fields that we know well, but are amateurs in most other domains. If our car breaks down, we go to a trained mechanic. Even the best mechanics will make some mistakes (they are human), but for most of us they still have a better chance of getting our cars to work than doing it ourselves.

Feasibility Analysis: Ambiguous and Diffused Costs

Many environmentalists are uncomfortable with cost-benefit analysis (CBA). CBA undeniably has a number of technical problems. I have argued, for example, that analysts should, in some cases, account for altruistic preferences; others have argued that analysts should ignore immoral preferences and existence values. One main reason many people oppose CBA, however, is that it often justifies allowing pollution and natural resource consumption, as long as the benefits of the risk-producing or consumption activity outweigh the costs.

Thus, a number of policy makers have proposed alternative decision procedures. Feasibility analysis is one that has recently gained traction. Feasibility analysis instructs policy makers to reduce specific pollutants within an industry to the minimum feasible levels, given technological and economic constraints.

University of Chicago law professors Eric Posner and Jonathan Masur recently provided a critique of feasibility analysis, criticizing its vagueness and its potential to cause disbursed harm to consumers:

The problem is there is considerably ambiguity throughout the process. . . . Feasibility analysis, as noted, tends to focus heavily on concentrated harms like plant closings. It pays little to no attention to the harms regulation may impose on consumers. From a welfarist standpoint, this blindspot in feasibility analysis is difficult to justify. That's an example of where feasibility analysis leads to overregulation compared to the socially optimal result, but Posner and Masur stress that feasibility analysis can lead to underregulation as well, particularly of smaller and/or less profitable companies or industries, which would face proportionally higher costs at complying with socially optimal regulation.

Posner and Masur further contend that, because of the ambiguity in feasibility analysis (for example, are automobiles and airplanes in the same industry?) agencies often resort to an informal CBA:

[A]gencies undertaking feasibility analysis seem to smuggle in some cost-benefit analysis on the side . . .

This result is unsurprising to me, since, on an intuitive level, CBA seems to correspond with how people make actual decisions. Before deciding to write this post, I did a quick determination of whether the benefits outweigh the costs. So if bureaucrats don't receive clear directions on how to choose among policy options, it seems likely that they will resort to CBA.