TIIP Newsletter

This paper builds on work published by the authors in “Applications, Grants, and the Value of Patent,” Economic Letters 69 (2000), 109-114. Using European Patent Office (EPO) data on published patent applications and grants, Guellec and van Pottelsberghe make the assumption that inventions revealed in successful patent applications are, on average, more valuable than inventions revealed in rejected or abandoned applications. The assumption seems quite reasonable and represents an important new approach to the search for valid indicators of patent value. The authors took 100 random samples of unstated sizes from 391,440 applications filed with the EPO between 1985 and 1992 and focused on eleven technology sectors and eleven countries or groups of countries in the European Patent Organization.

Their study assessed the effects of different patent strategies, domestic and cross-border research and ownership collaboration, and technological complexity on the success of EPO applications. The main findings are as follows:

(1) The authors found a significantly higher rate of success (hence, greater value) in patent applications that had first been filed nationally and then transferred to the EPO under Chapter 2 of the Patent Cooperation Treaty (PCT) than in applications filed directly with the EPO. They also found a significantly higher rate of success when applicants took advantage of the longer delay afforded by Chapter 2 in comparison with Chapter 1.

(2) Prior research had found a correlation between patent value and the number of countries in which patent protection is obtained. The authors, however, found that this is not a clear-cut indicator of patent value. The highest rate of success occurred with applications that designated the U.K., France, Germany, and up to two other countries. They suggest that applicants seeking protection for a large number of countries may simply have less experience with patenting. Greater experience would probably teach that securing protection in only the largest markets may be sufficient to capture the full potential value of the invention. The relationship between invention value and the number of countries in which applicants seek protection is complex and warrants further research.

Findings pertaining to research and ownership collaboration included the following:

(1) Collaboration among researchers appears to increase invention value. Applications with at least two inventors were more likely to be successful than those with only one. This indicator proved especially true when the collaborators were from different countries. This result contrasts with the findings of Allison, et al. who found that the average number of inventors was a negative predictor of litigation, which can also serve as a measure of patent value.

(2) Joint patent ownership produced mixed results. When the joint applicants were from the same country, applications were less likely to be successful than those with a single applicant, but when joint applicants were from different countries, applications were more likely to be successful than those with a single applicant. The authors suggest that these findings could be caused by the greater likelihood that domestic joint applicants are individuals, whereas international joint applicants are more likely to be members of multinational corporations with far greater resources. To me, this suggested explanation is not compelling, and warrants further investigation.

(3) With respect to technological diversity, the authors weighed the number of International Patent Classifications (IPCs) in which patent applications were placed against the likelihood of a patent grant. Each IPC is said to define a technology area; the larger the number of IPCs a patent fits into, the reasoning goes, the more technology areas it touches upon, and, hence, the larger the number of potential infringers. If the average number of IPCs is a useful measure of patent scope, one might expect it to be correlated with the likelihood of litigation and thus private patent value. Previous research has yielded mixed results here, but in this study the authors find (using eight-digit IPC classes) that the probability of receiving a patent decreased with the number of IPCs in EPO applications.

The authors argue that technical diversity has a negative impact on the likelihood of an EPO application being successful and is hence a contra-indication of invention value. Another interpretation, however, holds that the IPCs do not identify technology areas adequately. Allison, et al. developed their own definitions of fourteen technology areas and subjectively assigned each patent to one or more of them. The average number of such areas was a highly significant predictor of litigation.

Guellec and van Pottelsberghe make a meaningful contribution to the patent value literature, not only in their findings but also—and perhaps more importantly—in their use of a new approach that compares successful and unsuccessful patent applications.


		The Value of Patents and Patenting Strategies: Countries and Technology Area Patterns by Dominique Guellec (OECD) and Bruno van Pottelsberghe de la Potterie (Solvay, Brussels), Economics of Innovation & New Technology 11 (2002), 133-148.
		--Summary by John R. Allison, The University of Texas at Austin
		This paper builds on work published by the authors in “Applications, Grants, and the Value of Patent,” Economic Letters 69 (2000), 109-114. Using European Patent Office (EPO) data on published patent applications and grants, Guellec and van Pottelsberghe make the assumption that inventions revealed in successful patent applications are, on average, more valuable than inventions revealed in rejected or abandoned applications. The assumption seems quite reasonable and represents an important new approach to the search for valid indicators of patent value. The authors took 100 random samples of unstated sizes from 391,440 applications filed with the EPO between 1985 and 1992 and focused on eleven technology sectors and eleven countries or groups of countries in the European Patent Organization. Their study assessed the effects of different patent strategies, domestic and cross-border research and ownership collaboration, and technological complexity on the success of EPO applications. The main findings are as follows: (1) The authors found a significantly higher rate of success (hence, greater value) in patent applications that had first been filed nationally and then transferred to the EPO under Chapter 2 of the Patent Cooperation Treaty (PCT) than in applications filed directly with the EPO. They also found a significantly higher rate of success when applicants took advantage of the longer delay afforded by Chapter 2 in comparison with Chapter 1. (2) Prior research had found a correlation between patent value and the number of countries in which patent protection is obtained. The authors, however, found that this is not a clear-cut indicator of patent value. The highest rate of success occurred with applications that designated the U.K., France, Germany, and up to two other countries. They suggest that applicants seeking protection for a large number of countries may simply have less experience with patenting. Greater experience would probably teach that securing protection in only the largest markets may be sufficient to capture the full potential value of the invention. The relationship between invention value and the number of countries in which applicants seek protection is complex and warrants further research. Findings pertaining to research and ownership collaboration included the following: (1) Collaboration among researchers appears to increase invention value. Applications with at least two inventors were more likely to be successful than those with only one. This indicator proved especially true when the collaborators were from different countries. This result contrasts with the findings of Allison, et al. who found that the average number of inventors was a negative predictor of litigation, which can also serve as a measure of patent value. (2) Joint patent ownership produced mixed results. When the joint applicants were from the same country, applications were less likely to be successful than those with a single applicant, but when joint applicants were from different countries, applications were more likely to be successful than those with a single applicant. The authors suggest that these findings could be caused by the greater likelihood that domestic joint applicants are individuals, whereas international joint applicants are more likely to be members of multinational corporations with far greater resources. To me, this suggested explanation is not compelling, and warrants further investigation. (3) With respect to technological diversity, the authors weighed the number of International Patent Classifications (IPCs) in which patent applications were placed against the likelihood of a patent grant. Each IPC is said to define a technology area; the larger the number of IPCs a patent fits into, the reasoning goes, the more technology areas it touches upon, and, hence, the larger the number of potential infringers. If the average number of IPCs is a useful measure of patent scope, one might expect it to be correlated with the likelihood of litigation and thus private patent value. Previous research has yielded mixed results here, but in this study the authors find (using eight-digit IPC classes) that the probability of receiving a patent decreased with the number of IPCs in EPO applications. The authors argue that technical diversity has a negative impact on the likelihood of an EPO application being successful and is hence a contra-indication of invention value. Another interpretation, however, holds that the IPCs do not identify technology areas adequately. Allison, et al. developed their own definitions of fourteen technology areas and subjectively assigned each patent to one or more of them. The average number of such areas was a highly significant predictor of litigation. Guellec and van Pottelsberghe make a meaningful contribution to the patent value literature, not only in their findings but also—and perhaps more importantly—in their use of a new approach that compares successful and unsuccessful patent applications. © 2003. Verbatim copying and distribution of this entire article for noncommerical use are permitted provided this notice is preserved.