Euclid, Friedman, and the Marginal Revenue Product Curve

I always found economics a hideously frustrating subject: We're asked to understand the world in light of models that portray stuff like 'pure competition' or 'monopoly', and make all sorts of absolutely unsustainable assumptions thereabout. Is there ever an industry completely free of collusive behaviour? Is there ever a product for which one can't possibly find a replacement?

Of course not. And, oddly enough, I found the explanation in an essay collection dated by several decades. Milton Friedman's Essays in Positive Economics has the feel of an outdated text meant primarily for experts (the three-page footnotes are certainly rather frightening), but getting past some of the arcane terminology reveals his startling instinct for clarity: Those economic theorems are much like those we'd find in a geometry or physics textbook, in that they assume absolutely ideal and unsustainable conditions to create a set of basic principles from which more advanced ideas can be derived. We wouldn't have developed a Pythagorean Theorem if Pythagoras and his contemporaries spent all their time quibbling about how a 'line' could never be 'straight' in the real world.*

All very obvious, right? Unfortunately not. For anyone who studies economics, there's an irresistable tendency to shoehorn real-world situations into the relevent economic models. Thus, Microsoft's high market share in the 'software industry' (a classification so broad we might as well talk about Ford being involved in the 'objects that use circles' industry) makes it a 'monopoly' which, perforce, will use Monopoly Pricing Strategies economists devised to illustrate how a monopoly would price its products if it did exist -- they didn't think they had to note that it can't.

Meanwhile, when a not-perfectly-competitive, but surely not oligopolistic, market doesn't behave as the economic literature suggests it might, it's scolded as a 'market failure'. How dare that market pretend it isn't precisely what we hypothesize it is? I can only imagine a world in which geometers behaved the same way: "You're pretty tall. I think you're a line segment... hey! You don't exist in those dimensions, so put that isosceles triangle down!" Meanwhile, economists are still using graphs and theorums to conclusively prove that the future is predictable and the past is just a fluke. Talk about side-splitter theorems.

*Multiculturalist Editor's Note: Yes, I know, the Chinese developed that theorem half a millenium before Pythagoras, but you get the idea.

Edit: Welcome, Carnival of the Capitalists readers!