The rule of thumb for when an intellectual monopoly should be granted is if, absent monopoly power, a given innovation would not come into existence. This is generally true in fields where the cost of innovation is high, but the cost of imitation is low.
This is a broad standard that makes it likely there will be considerable inter-industry differences with respect to the utility of patents, necessitating empirical research to determine which fields need (or don’t need) access to patents. A new study published by the National Bureau of Economic Research conducts a review of the empirical evidence related to patenting and innovation and confirms such heterogeneity.
The main finding of the review is strong evidence for the necessity of patenting for pharmaceutical innovation. Using both survey data from those in industry and empirical findings, the author confirms patents are important to the pharmaceutical industry.
A 1973 study cited found that if patents were replaced with a system of compulsory licensing, there would be only an 8% decline in R&D funding overall, but a 64% reduction in pharmaceutical research. A 1986 survey further confirmed this general finding, with respondents indicating that 60% of drug inventions would not be in existence absent patenting.
Part of this has to do with the nature of innovation in pharmaceuticals vs. other industries. Pharmaceuticals tend to be discrete products, while other industries rely on a number of different patented components. When innovation is a “one-shot” deal, the case for patenting is strong, but if innovation is incremental by nature in an industry, the case becomes weaker.
What about other industries? In industry survey data, the traditional justification to prevent copying wasn’t the only motivator. Blocking rival firms from patenting and preventing litigation also played important roles.
But the more relevant findings have to do with the motivations for firms to innovate. A 2000 survey found that lead time, secrecy, and offering complementary services or products were more important for innovation than patenting.
Finally is the question of how disclosure relates to innovation. The author describes disclosure as the “grand bargain” for a patent, and many firms cite the need to reveal the nature of their innovation as a reason not to patent.
By using data from academic researchers, the review finds that patenting itself is not a great hindrance on innovation. (In some cases, researchers weren’t even aware of patenting.)
Instead, control of (and ability to exclude others from using) the tangible, patented materials was a greater hindrance to research. This is because “compositions of matter,” as physical entities, “require affirmative consent (and sometimes negotiation, formal contracts and payment) for access.” Research methods, on the other hand, do not face this logistical hurdle.