What actually is the behavioral approach? I have suggested that behavioral can be understood as an opening up of the strict rational choice model.
But in this lecture, the term is discussed in the context of the evolution from policy evaluation to program design. Though note that the reason why we engage in designs is because the existing programs when evaluated turn out to fall short. And why are the existing programs failing? Because of incentive issues. And how do we study incentives? Rational choice analysis. So then we move on to design programs that are supposed to structure incentives. But when we construct designs, people don't actually behave the way we designers thought they would. And we get this disjoint between design and behavior time after time. So what do we do? We design the system better to take into account underlying behaviors and structure the system to "nudge" folks to do what it is we designers want them to do.
This is a very clear cut lecture and reflects a thoughtful and interesting mind. I enjoyed this discussion, but I believe all of those who think about behavioralism should read Vernon Smith's Nobel Prize lecture on the difference between ecological and constructivist rationality (and Gerd Gigerenzer's Rationality for Mortals) and then read backwards to an older essay by James Buchanan, "Natural and Artifactual Man." Both essays in their own way, push the argument that choice is individual and contextual and not abstract. Our decisions are based on our subjective assessment of the trade-offs we face in pursuing our subjective desires. And yes, we have problems of self-control and self-delusions, but we also respond to incentives even in these areas. And, moreover, there are no doubt issues associated with stated intentions and actions pursued, and there are issues of a disjoint between stated intentions and outcomes realized.
Good economics has to look at not only immediate actions, but also the indirect results of actions pursued. The lecture has a great slide which contrasts Spock with Homer Simpson, and then flashes a picture of Brad Pitt when he was wearing a rather unattractive beard. This was to reflect that human choosers are neither perfectly rational nor completely foolish, but instead capable but fallible human choosers. I think this is right. But it is also the model of man one reads in Adam Smith and David Hume, and also one reads in Ludwig von Mises and F. A. Hayek. In fact, I think one would benefit greatly from reading the first 100 pages of Human Action, not with a focus on the defense of deductive logic and its role in praxeology, but instead with a focus on the nature of choice and what is required for a human chooser to the center of analysis. Along these lines, I think the best recent work is Richard Wagner's Mind, Society and Human Action.
Mario Rizzo is currently working on the question of rationality as well as a critique of variants of the behavioral economics that underlies much of the discussion which in this lecture is viewed as "design principles".