Is the Patent System an Even Playing Field? The Effect of Patent Attorney Firms
Gaétan de Rassenfosse, Paul H. Jensen, T'mir Julius, Alfons Palangkaraya, Elizabeth Webster
Journal of Industrial Economics (2023), 71(1): 124-142
Research Question
Does the quality of patent attorney firms significantly influence patent examination outcomes, independent of invention quality?
Why Do We Care
The efficacy of the patent system is crucial for innovation since it allocates monopoly rights in exchange for disclosures that should enhance economic development. However, if certain attorneys have better access to patent offices than others, it could distort the level playing field intended by the patent system, allowing less deserving patents to be granted while excluding higher-quality inventions. Understanding the role of attorneys could inform policy to improve patent quality and ensure fairness in the innovation ecosystem.
What Did We Do
We analyzed patent examination decisions across five major patent offices (EPO, JPO, KIPO, CNIPA, and USPTO) using a dataset of 106,453 patent applications filed between 2000 and 2006. We employed high-dimensional fixed-effects models to control for unobserved variation in invention quality and measure attorney quality. Our approach considered multilateral equivalents where patents from the same invention family were filed in multiple jurisdictions.
What Did We Find
We find that attorney quality significantly influences patent grant outcomes, with a one-standard-deviation increase in attorney quality leading to a 6 percentage-point increase in grant probability, and a logit estimate indicating an 11 percentage-point increase. This effect is particularly pronounced at the USPTO, where attorney quality outweighs invention quality in importance. We also find that attorney quality has a stronger impact in rapidly evolving technology fields such as ICT and software compared to highly codified fields like chemicals and pharmaceuticals.
Primary Data
- Our dataset includes 106,453 patent applications from five major patent offices, representing 279,220 applications all examined between 2000–2006.
- Our data are sourced from Patent Cooperation Treaty and national patent office databases including EPO, USPTO, JPO, KIPO, and CNIPA.
Assumptions
- We assume the influence of attorney quality on patent outcomes is independent of invention quality; thus, we control for invention quality using fixed effects modeling.
- We assume there are no biases in the patent examination process linked to external factors not accounted for by the model.