More Logical, Experimental Economics

In  Dan Ariely's excellent blog, He again positions the non rationality of the consumer.  Or even of groups of consumers.   I have been involved in a number of projects where we attempted to model the consumer.  It turns out this is tough, they do not easily bend to engineering modeling.  The suggestion he ends up with here is to construct small experiments, like we did at our innovation centers, to test what works, and what the unintended consequences are of change.  See the credit card interest example as one of those unintended consequences.  Yet models are also not reality, and have to be carefully tested themselves.   Agent based models are also something we experimented with and are worth examining.  Consumers cannot be modeled by an equation, but as groups statistically.   Ariely ends with this obvious and important point:

" ... I realize that this is not an elegant solution because conducting rigorous experiments in public policy, in business, or even in our personal lives is not simple, nor will it provide simple answers to all of our problems. But given the complexity of life and the speed at which our world is changing, I don’t see any other way to truly learn the best ways to improve our human lot. ... "