N. Emrah Aydınonat — Using and Abusing Models in Economics: A Review of Rodrik’s Economics Rules

Dani Rodrik’s new book Economics Rules: The Rights and Wrongs of the Dismal Sciencewill be available soon. Here is an early quick review of the book.

If you do not have much time, here is the main message of this review: Economics Rulesis an excellent book; a must read for economists, philosophers of economics, and policy makers, and of course for economics students. And if you consider yourself as a critical economist, a heterodox economist, or a social scientist who is critical of the way in which economists work, behave, etc., the book has a lot to offer to you too (probably a lot to disagree about, but also a new perspective on economic models). While Dani Rodrik defends economics against popular lines of criticism; he also presents his own critique.

If you have more time, please read on.

Good review. Lots of references. It's more an article than a post.

Here are some of my thoughts.

A key problem in interpreting models is mistaking a special case for the general case. There are a lot of useful economic models, but none of them describe the general case.

But I think that Rodrik is out on a limb in thinking that the objectors have it wrong in criticizing economic models for not being representational enough. The kerfuffle is about economists confusing special cases with the general case and representing their models describing either special cases or thought experiments as a general case model on the assumption that a "real" science provides a general description of the world. 

Objectors are correct in pointing out that this isn't borne out by evidence, meaning that to the degree models are successful, they are only so regarding special cases, and even at that the models are highly stylized owing to restrictive assumptions to make them tractable.

Objectors also complain that the assumptions of many models are too restrictive to be representational. Therefore, it is unreasonable to expect let alone claim that the kind of model is representative of the world.

For example, no economist actually believes that general equilibrium is strictly true, only that under certain restrictive assumptions economies tend toward general equilibrium "in the long run," which is left unspecified.

The issues arise  when the model is used as a basis for formulating policy as if the assumptions actually hold universally so that the model is representation of reality, when they don't even hold in an actual special case. The model is just a thought experiment used as a heuristic.

My view as an outsider is that the quest for a general theory of economics, models on, say, Newton's laws of motion, is a chimera, a mere mirage, that will never be reached because the variable are too numerous (covering many disciplines in addition to economics) and too changing (time and context dependent). 

All that economists can hope to do is to describe parts of the elephant as in the story of the seven blind men. Since the whole is more than the sum of its parts, the total of these models won't do it either.

Using and Abusing Models in Economics: A Review of Rodrik’s Economics Rules
N. Emrah Aydınonat | Associate Professor, Bahcesehir University, Istanbul, Turkey

ht Tyler Cowen

Also

Information Transfer Economics
What's wrong with Dani Rodrik's view of economics
Jason Smith