Abstract
The explosion in the patenting rate in the U.S. during the last half of the 1990s is often attributed partly to an apparent decline in examination standards. We estimate a simultaneous equation model accounting for the fact that a decline in examination standards would itself induce an increase in dubious applications. We have a multi-dimensional panel, with data on the application and grant rates and country of origin and destination. We find that a 'loosening' of the grants standard by 1 percent increases applications by 8 percent in the full sample and by 3 percent in the Non-US sample. After accounting for the endogenous application response, the application elasticity of grants is around 0.124 for the full sample and 0.145 for the Non-US one. Countries whose patent applications are more likely to be successful in the US are more likely to be successful in other countries as well. These findings confirm that inventors respond to increased likelihood of success at the patent office by filing more applications, but also confirm earlier findings that the surge in patenting in the US in the last two decades appears to be driven to a significant extent by an increase in the underlying invention rate. L'explosion du taux de délivrance de brevets aux Etats-Unis durant la seconde moitié des années 90 est souvent attribuée en partie à un relâchement des niveaux d'exigence requis à l'examen des inventions. Nous estimons un modèle d'équations simultanées rendant compte du fait qu'un abaissement du niveau d'exigence devrait induire une augmentation des demandes douteuses. Nous utilisons un panel multi-dimensionnel, contenant des données sur la nature des demandes de brevet, les taux de délivrance, les pays d'origine et de destination. Nous trouvons qu'une diminution du niveau d'exigence pour la délivrance des brevets de 1% augmente le nombre de demandes de 8% dans l'échantillon total et de 3% dans l'échantillon non-US. En tenant compte de l'endogénéité.