Comment by jcutrell | Hacker News Reader

Comment by jcutrell | Hacker News Reader

Parent Full thread

I think a "good" behavioral economist would not explain exactly why, but instead provide theories with as much data to describe why those theories exist.

A trial shows that people often exhibit behavior A. Why they exhibit behavior A may be complicated.

In your example, there's a sense of authority and conforming to a norm that can be used to describe those behaviors. But there's also the cognitive barrier (which we call lazy) that requires someone to exert mental effort to change away from a default.

It's not that one is wrong and the other is right. All of these may be right. But B.E.'s best contribution to society isn't to explain so much of _why_ we act in these ways - but instead to accept that despite our desires or perspectives of rationality, we do act in these ways.

Understanding the "why" may help us to make better decisions, certainly; however, being too prescriptive about "why" may lead us down the wrong path.

The way that behavioral economics is very often presented is not with the epistemic humility you describe here however. This is in part the fault of media popularizers and entities that use behavioral economics to justify and push their own agendas but in my experience the well known names in research are also culpable, both in their own popularizations and applications of their research and in their academic work.

There is an admirable strain in classical economics of epistemic humility: assuming that when human agents act in the world in ways that don't appear to match economic theory, it is at least as likely that the agents are acting in their own best interests and the economic theory, model or data are wrong than the reverse.

Behavioral economics tends to follow a less admirable strain of economic thought that assumes when agents don't follow the predictions of our over simplified model, it is the agents who are wrong and not the model. This is merely misguided in it's benign form but becomes pernicious when it is inevitably extended to "fixing" the agent's behavior through some form of coercion.

My criticism of BE is when it makes too strong a claim to agents acting irrationally rather than questioning where its own assumptions or models may be incomplete.

I find it somewhat useful if overly simplistic as an aid to analyzing and improving my own decisions. I dislike its use as another tool to justify telling people what to do. I also find its lack of introspection into its own flaws and limitations as a field disappointing.

Reply on news.ycombinator.com