More than two decades ago, I took a course in intermediate microeconomics. The textbook was R. Robert Russell and Maurice Wilkinson's Microeconomics: A Synthesis of Modern and Neoclassical Theory. "Modern", in this case, refers to the use of set theory terminology, linear programming, and proofs like those in an introductory real analysis class. In contrast, "Neoclassical" refers to the use of continuously differentiable functions. In any case, the substance of the theory - which is only one possible theory - is unaffected. (I would not have been clear on this at the time.)
One day, our professor was teaching us about oligopoly and the theory of the kinked demand curve. And, in response to a question, the professor said something like, "This is a theory I actually believe". Yet, in the rest of the classes, when he was teaching us to manipulate utility functions or production functions or to take Lagrangians or whatever, he never expressed an opinion of the empirical applicability of what he was teaching us.
I also recall that our professor made a special effort to teach us input-output analysis one week. This topic was not in the textbook, if I recall correctly. But Leontief was coming to give a lecture (not to our class, but in a big lecture hall, that is, CC308). And our professor wanted us prepared. As it was, Leontief's lecture did not concern the details of input-output analysis, but the complaint that most of then contemporary economics was unconcerned with empirical results. Most economists did not even cast their theory in a form where it could be connected up to empirical data that one might collect.References
- Wassily Leontief (1982). Academic Economics, Science, New Series, V. 217, N0. 4555 (9 July): pp. 104-107.
- Wassily Leonteif (1983). Academic Economics Continued, Science, New Series, V. 219, No. 4587 (25 February): 904.
- R. Robert Russell and Maurice Wilkinson (1979). Microeconomics: A Synthesis of Modern and Neoclassical Theory, John Wiley & Sons.