The Federal Trade Commission's 2003 innovation report revealed an interesting fact: the pharmaceutical industry is largely satisfied with today's patent system while the electronics, software and Internet industries are not. This article suggests that a difference in governing law accounts for the difference in satisfaction. The federal Food and Drug Act requires pharmaceutical inventions to be proven safe and effective before they can be sold. It thus requires completion of the entire inventive process for pharmaceuticals. Our patent system, however, has no analogous requirement for the other fields. In them, applicants may stake a claim to rivals' later inventive effort after completing only the very first step - conceptualizing - of a lengthy inventive process. The result is patents on abstractions that hold up real inventors, as in Blackberry, eBay and countless other cases to come. Unfortunately, developments in biotechnology, including patenting gene segments, suggest that the biotech industry may be closing the gap in early-stage patenting. This article suggests two ways to improve our patent system and reduce the difference in industry satisfaction. The first is to abolish the doctrine of constructive reduction to practice. A statutory amendment would require inventors, before receiving a patent, to invest reasonably in making, building, testing or at least simulating something concrete, in order both to demonstrate feasibility and to attract real seed capital. A second amendment would convert Section 103's nonobviousness criterion from an abstract test of cognitive difficulty to an economically meaningful test. The new test would encourage review of concrete economic factors such as investment of risk capital, real progress in bringing an invention to market, the assumption of technological risk, i.e., risk of total failure for nonmarket reasons, and a proven need for protection from free riders. Unlike current Section 103, the new test would consider how the invention was made and would permit hindsight. It would do so on the theory that patents protect not cognitive brilliance, but investment of risk capital in the entire process of invention, in which conceptualization is only a first step. The test would also recognize that investment, concrete progress, and risk are susceptible to proof in retrospect. The article outlines how, if adopted, these changes could help rationalize our patent system economically and eliminate the difference in satisfaction between pharmaceuticals and other fields of industry, whose own participants now see patents as impeding progress as often as promoting it
