His Ph.D. dissertation research was in the field of mathematical economic theory.

The title of the research paper was, "Price Determination in an Overlapping Generations Example Economy with Producible and Nonproducible Capital". This paper attacked conventional Monetarist analysis by applying a generalization of the Capital Asset Pricing Model to argue that Monetarist analysis is illogical. In his dissertation, he also introduced the methodological idea, "what might be true is sometimes more important than what is true", to argue that scientific theory must be logical as well as consistent with observed facts.