The first is a famous one, from Milton Friedman, via Justin Fox’s The Myth of the Rational Market:
[T]he relevant question to ask about the “assumptions of a theory is not whether they are descriptively “realistic,” for they never are, but whether they are sufficiently good approximations for the purpose at hand. And this question can be answered only be seeing whether the theory works, which means whether it yields sufficiently accurate predictions.
As Fox writes:
To head off the obvious objection that it was ridiculous to think regular folks reason according to complex statistical rules, Friedman and Savage argued that billiards players couldn’t write down the physics formulas that underlay their shot selections but nonetheless acted as if they did.
Here’s quite a different take that I just came up on recently, from J.W. Mason of the Roosevelt Institute:
It seems to me that Deirdre McCloskey was right: Economics is not the study of the economy. Economics is just what economists do. Economic theory is essentially a closed formal system; it’s a historical accident that there is some overlap between its technical vocabulary and the language used to describe concrete economic phenomena. Economics the discipline is to the economy, the sphere of social reality, as chess theory is to medieval history: The statement, say, that “queens are most effective when supported by strong bishops” might be reasonable in both domains, but studying its application in the one case will not help at all in applying it in in the other.
Finally, a third view, from Dani Rodrik’s excellent book Economics Rules:
I wrote this book to try to explain why economics sometimes gets it right and sometimes doesn’t. “Models” — the abstract, typically mathematical frameworks that economists use to make sense of the world — form the heart of the book. Models are both economics’ strength and its Achilles’ heel; they are also what makes economics a science — not a science like quantum physics or molecular biology, but a science nonetheless.
Rather than a single, specific model, economics encompasses a collection of models. The discipline advances by expanding its library of models and by improving the mapping between these models and the real world. The diversity of models in economics is the necessary counterpart to the flexibility of the social world. Different social settings require different models. Economists are unlikely ever to uncover universal, general-purpose models.
But, in part because economists take the natural sciences as their example, they have a tendency to misuse their models. They are prone to mistake a model for the model, relevant and applicable under all conditions. Economists must overcome this temptation. They have to select their models carefully as circumstances change, or as they turn their gaze from one setting to another. They need to learn how to shift among different models more fluidly.