Im-patent to innovate

By Z.G. | LONDON

“IF I have seen further”, Isaac Newton once demurred, “it is by standing on the shoulders of Giants”. The process of invention has long been a cumulative one, in which incremental advances are made on previous innovations. However, a new NBER working paper* by Alberto Galasso of the Rotman School of Management and Mark Schankerman from the London School of Economics has found that the patent system is curtailing this tradition of progressive innovation.

The authors quantify the impact of the patent system by measuring what happens when a patent is invalidated by the US Court of Appeals. They find that the removal of a patent significantly increases the number of future innovations derived from the original idea. This effect is attributed to inefficiencies in the bargaining process. If a downstream inventor is unable to negotiate access to foundational patents, then they will be unable to monetise any subsequent innovations and will have little incentive to build on the original idea. However, when a patent is invalidated this restriction on new innovation is lifted and the invention is free to be expanded upon.

Critically, the authors report that this effect varies widely between fields. For patents in ‘complex’ fields, where products rely on numerous patents, including electronics, IT and biotechnology, subsequent citations more than tripled after a patent was invalidated. However, for ‘non-complex’ fields, in which the average product only embodies a few patents, the effect was insignificant.

What explains the result? When a new invention requires access to a large number of patents the risk of the negotiations failing is high. Thus when a patent is over-turned the benefits - from avoiding a messy negotiation - are comparatively large. In contrast, in ‘non-complex’ fields negotiating access to a small number of patents is relatively easy, and so the cost of securing access to the required patents is relatively small.

Furthermore, the authors compare the relative sizes of the respective innovators to show that this effect is strongest when patents are owned by large firms that are used to block smaller innovators. This is likely because smaller inventors have a hard time bargaining for access rights with large companies, which see little gain to licensing their patents to bit players.

The patent system is designed to encourage innovation by granting property rights to the inventor. However, finding the right strength for these property rights is a delicate balance. Too weak and there may be insufficient monetary incentive to innovate, too strong and they may impede the development of future innovations.

The authors shy away from making any definitive suggestions for patent reform, in part because the paper only attempts to measure one side of the cost-benefit ledger: the gains from overturning a patent, as opposed to the positive incentives it generates to innovate in the first place. However, it does suggest that the potential costs of the current system are larger than had previously been thought. It also shows that constraints on innovation are concentrated in specific markets. As such it may be more efficient to target reforms at these fields specifically, as opposed to trying to reform the patent system en masse. That way the areas of the patent system which are broken can be fixed, without reducing the incentive to innovate in fields where the patent system is already working well.

* Patents and Cumulative Innovation: Causal Evidence from the Courts. Alberto Galasso and Mark Schankerman. NBER Working Paper June 2014.