Economics is a unique tool-set for deciphering the causes of human behavior. It assumes that people are rational, incentive-driven, and utility-maximizing. The framework is simple and the logic is pure. Behavioral Economics, on the other hand, is the branch of economics that brakes the with homo economicus assumption. It tests models of rationality in the real world. I will argue that, while behavioral economics points to some unrealistic assumptions made by classical economists*, it does not resolve the issue in any meaningful way. First I will give an account of a classical economics framework. Secondly, I will outline the behavioral response, its conclusions, and its flaws. Next, I will summarize the previous two points with an example. Lastly, I will list some other critiques of behavioral economics. I will argue that the critique of behavioral economics regarding classic economics is best resolved by more classic economics.
Let's start by talking about classical economics. There is a certain economic approach to economic modeling. When building a classic economic model one has to first define some assumptions. These assumptions might seem (and mostly are) wildly unrealistic. My Corporate Finance Professor said to us that assumptions are inherently unrealistic, that’s why they are called assumptions. Modigliani and Miller’s Nobel-prize-winning Capital Structure Theory starts with a list of 13 assumptions, one more unrealistic than the other (No taxes, symmetric information, no transaction costs…). Once these assumptions are defined one has a framework on which to build conclusions in a logical and mathematical way. One thus creates an alternate world with assumptions and proves that in this alternate world things would be this way or that way. One can then brake assumptions, changing the alternate world, and come to different conclusions. These conclusions are always true. They are a logical deduction from the assumptions made at the beginning. There are reasonable steps and proofs for everyone to follow.
Now come along the new cool kids from out of town: the behavioral economists. They grew up in Psychology-Ville, they have long hair, smoke, and play electric guitar. Tversky and Kahnemann smash prospect theory on the table. Thaler whips out his Nudge. Arieli shows off his Predictable Irrationality. The cheerleaders immediately leave their nerd classical economics boyfriends for the hot new dudes. Nobel Prizes left and right: Economics finally found its way into the real world. But what's so cool about the new kids on the block? They showed that most of economic theory doesn’t hold in the real world. They go out and test the classical economic models in the field or the lab and observe that most people think that they are better than average, can be primed or anchored, and are influenced by available rather than true information… It seems that nothing that the classical economical models predicted is true. Not only do the cool new behavioral kids steel the hot cheerleaders from the glasses-wearing classic economic nerds, they now rename classical theory with “Biases” and “Heuristics”. “Hindsight Bias”, “Planning Fallacy”, “Cognitive Dissonance”, “The Better-Than-Average Effect” replace classical theory.
But as with the cool new kids in school, the mystery fades and one is left with the question: What was it that made them so cool? Sure they showed that classical theory does not apply to the real world… But, was that not clear from the start? As pointed out above: Classical economics builds and the alternate world with its assumptions. Within this framework, its conclusions are very true. Noone said we wanted to explain the real world. Behavioral Economists pointed out that theories with unrealistic assumptions are unrealistic. That’s not too impressive. What’s even less impressive is the conclusion that behavioral economists draw from this. They see a deviation from rational behavior, call it a bias, fallacy, or heuristic and move on. What did that tell me about the world? Not much…
Let me give you an example:
Classical economists assume that people are pretty good and accurate at information acquisition; Sure, its costly and asymmetric but the equilibrium of an information economy will show accurate results (Hayek is great on that).
Along come behavioral economists and ask people to rate their driving skills from 1 to 10. 90% of people rate themselves above 5. This is a revolution in the behavioral world. If the majority of people think that they are better than the majority there is something wrong. Information is not accurate and people are overconfident. Let’s call this the “Better-Than-Average Effect” and move on.
But classical economists don’t give up so fast. They dissect the problem with theory. What if this question is too broad. Aren't there many aspects of driving that one can arbitrarily choose to answer this question? I myself am a great driver of highways but completely crap at parallel parking. I’d give myself a 7 in highway driving and a 2 at parallel parking. If someone asks me to rate my driving I will opt for the better aspect of driving and ignore the parallel parking for now. So, classical economists come along and ask people to rate themselves on different aspects of driving - low and behold the “better-than-average effect” decreases. People don’t overestimate their skills as much as with the broad question asked by behavioral economists.
Behavioral economists have a different starting point. They begin by observing humans in their natural habitat and then coming up with a theory post hoc. I would argue that these theories can never explain human behavior. They start with an infinite amount of variables (the real world) and try to arbitrarily boil it down to human irrationality. The more scientific approach is to start with a limited amount of variables (assumptions), predict human behavior on a theoretical level, test it in the real world, observe that the theory doesn’t hold, go back to the drawing board and revise your variables and assumption, test again, revise again… until you have a model with a finite amount of variables that can sufficiently explain an aspect of human behavior. This is our only chance of finding the true causes of things.
Other criticisms of behavioral economics have relied on the disapproval of experimental economics as a method or the inherent contradiction of Parental Libertarianism (Thaler). I would defend behavioral economics on both points. I think experimental economics is a great way of testing given certain constraints. Classical economists (Gary Becker f.e.) say that Parental Libertarianism, the notion of using behavioral insights for the greater good, is an inherent conflict. I disagree. Here I am on the side of behavioral economists saying that we are influenced by all kinds of factors regardless of whether someone planned them or not; So why not use that fact and arrange certain aspects of society to improve the lives of people, based on solid data.
In conclusion, my problem with behavioral economics is not a philosophical one, it is a methodological one. I think that pointing out human irrationality by observing the complicated real world is circular and meaningless. Behavioral economists made their point clear that classical economic models fail to explain the real world. The only solution to this problem is more classical, more assumptions, more variables, more theory, in essence, more classical economics.
* Here Classical Economics refers not to free market economics, but to economics that works with assumptions, most notably the homo economicus assumption.
Comments