John Allen Paulos in his Who’s Counting column at ABC News:
…Lena Edlund of Columbia University and Evelyn Korn of Eberhard-Karls-Universitat Tubingen, have published an intriguing paper, “A Theory of Prostitution,” in the Journal of Political Economy.
Making simplistic but more or less plausible assumptions and applying the tools of economic model-making, they searched for the answer to a puzzle: Why is it that prostitution is so relatively well-paid?
Before getting to why this is, they document that in diverse cultures and over many centuries, prostitutes have indeed made much more, sometimes several multiples more, than comparably (un)skilled women would make in more prosaic occupations. From medieval France and imperial Japan to present-day Los Angeles and Buddhist Thailand, this income differential has persisted, although its size depends on various factors.
More here.