Modeling the Duration of Patent Examination at the European Patent Office

TLDR

The study analyzes the duration of the patent examination process at the European Patent Office. Using data on economic and technical relevance, EPO capacity, workload, and novel citation measures, the authors estimate competing‑risk models to characterize the processes leading to withdrawal, refusal, or grant. Highly cited applications are granted faster but withdrawn more slowly; overall duration rises with complexity, originality, reference count, and EPO workload, while controversial claims slow grants and hasten withdrawals, whereas well‑documented applications are approved more quickly and withdrawn more slowly.

Abstract

We analyze the duration of the patent examination process at the European Patent Office (EPO). Our data contain information related to the patent’s economic and technical relevance, EPO capacity and workload as well as novel citation measures which are derived from the EPO’s search reports. In our multivariate analysis we estimate competing risk specifications in order to characterize differences in the processes leading to a withdrawal of the application by the applicant, a refusal of the patent grant by the examiner or an actual patent grant. Highly cited applications are approved faster by the EPO than less important ones, but they are also withdrawn less quickly by the applicant. The process duration increases for all outcomes with the application’s complexity, originality, number of references (backward citations) in the search report and with the EPO’s workload at the filing date. Endogenous applicant behavior becomes apparent in other results: more controversial claims lead to slower grants, but faster withdrawals, while relatively well-documented applications (identified by a high share of applicant references appearing in the search report) are approved faster and take longer to be withdrawn.