Bad Science in Search of 'Bad' Patents

Abstract:     
This paper draws attention to fundamental deficiencies in studies that have been relied upon as authoritative sources on patent grant rate comparisons among national patent offices. The two prominent studies analyzed here had employed erroneous methods to compare patent grant rates, resulting in false high indications of such rates at the U.S. patent office compared to foreign patent offices. The three identified categories of analysis errors found in these studies were (i) the misapplication of conditional probability; (ii) miscounting invention applications; and (iii) failure to account for patent obsolescence and application attrition due to the widely differing delays among national patent offices. These errors over-estimate the U.S. patent grant rate by tens of percents, creating a pervasive misconception that such studies prove that examination at the U.S. patent office is the least rigorous among national patent offices. A subsequent section of this paper presents the results of a study that correctly estimates the patent grant rate for applications filed in the U.S. patent office as being in the range of 60% to 76%. Finally, it is shown that the scope and average number of claims in patent applications differ substantially among national patent offices. It is concluded that these differences render even accurate patent grant rate comparisons among national patent offices of very little probative value as indicators of examination rigor and patentability standards.

Keywords: patent, grant rate, patent application success probability, allowance rate, patent statistics, claims, national comparisons, patent quality, claim examination, patentability standards, probability, conditional probability

JEL Classifications: C11, C13, O34, O57

Accepted Paper Series