The Fall of the Economists' Empire — Robert Skidelsky

The problem is not so much with the modeling, actually. People are free to construct any models that please for whatever reason. The problem is with the conclusions that are drawn from the model when they exceed the limitations of the of the assumptions.

This is not a problem with modeling but with logic. Drawing conclusions that exceed the scope and scale of the premises in a context other than the model is flat out illogical, and any inferences drawn on this basis are unsound, that is, do not follow from the premises of the argument and the reasoning about them.

Models don't automatically transfer to the world that they purport to model. This is a fundamental of scientific method. Semantic interpretation is required, along with methodological rigor, and hypothesis testing is necessary.

Project Syndicate
The Fall of the Economists' Empire
Robert Skidelsky | Professor Emeritus of Political Economy at Warwick University, fellow of the British Academy in history and economics, member of the British House of Lords, and author of a three-volume biography of John Maynard Keynes
ht Lars Syll