Exploring the opaqueness of the patent system : evidence from a natural experiment

One of the objectives of patent systems is to disclose information which other agents can build on in further inventions and in their decision-making. While some observers take it as given that real-world patent systems serve this objective, we argue in this article that patent systems are highly opaque and likely to be of limited value as a source of information. We use data from a natural experiment to explore this issue. Requests for accelerated examination used to be publicly observable at the European Patent Oce (EPO). Starting in December 2001, the EPO started to treat these requests as confidential information. Using data on acceleration requests which were historically known only to the applicant and the EPO, and later provided to us, we test whether the change in the information regime impacted the actions of applicants and their rivals. We develop a theoretical model of acceleration requests and patent opposition to identify the extent to which the patent system is opaque. We confirm empirically that opposition and acceleration rates of highvalue patents change significantly in most technological areas once acceleration requests become unobservable. We interpret these results as evidence that the system is highly opaque in many fields.