An economic model is a false description of the world, and its falseness is the point.
It removes everything except the under examination, so that the can be seen.
Criticising a model for being unrealistic is therefore criticising it for existing.
The legitimate question is different and much harder to answer.
Does the preserve the feature that matters for the question being asked?
A model of trade is illuminating for one question and misleading for another.
Judging that requires knowing the setting, which is why models cannot be evaluated in the abstract.
The profession's difficulty is not the use of models but the transition from a model to a policy.
A result derived under assumptions is frequently quoted without them, and the assumptions are where the disagreement lives.
Textbook treatments compound this by presenting one model per topic as though it were the model.
Graduate training corrects it and undergraduate training, which reaches far more people, does not.
Most policy advisers, journalists and civil servants hold the undergraduate version.
Empirical work has grown enormously and has changed the balance of the field.
A generation of economists now spends more time on data quality than on , which is an improvement with its own .
A well-identified effect with no theory behind it does not travel to a different setting.
Theory is what tells you which features of a new context would break the result.
The honest position is that models are of thought rather than descriptions of anything.
Stated that way, the discipline makes a modest claim that it can defend, which may be why the claim is so rarely stated.