Valuation of Scandinavian patent rights across industries, nationalities, and time: Analysis, estimates, and applications

Patents are observable results of research and development and, consequently, patent data are a potentially valuable resource for the study of technological change. Patents are also an important component of most countries' industrial policies. But patent counts are problematic indicators of either the value of the patent right or the value of invention because of the large variation in and skewness of these distributions. If patentholders renew patents when the expected value of the patent right exceeds the renewal cost, renewal data provide information about the value of patent rights and, potentially, about the value of patented inventions. In this thesis, this information is developed using unique patent renewal data that are disaggregated by the nationality of the patentholder, technical field of the patent, and application date.;In the first section, equality tests are employed to test whether renewal behavior differs across industry of origin and nationality. Inequality tests are then employed to order groups according to the value of their patents. The results indicate that industry of origin is important in explaining renewal behavior. Once one conditions on industry, the null hypothesis that nationality differences do not explain renewal behavior cannot be rejected. The ordering of industries resembles the results from recent surveys; pharmaceutical and other chemical patents are most valuable, followed by mechanical, electronic, and heavy industry patents. When the renewal data are examined for changes over time, evidence of an inverse relationship between patent quality and quantity is found, confirming an earlier result by Schankerman and Pakes (1986).;The second section matches the patent data to industry-level research and development expenditure data to examine whether renewal information can be used to construct patent-based indices of technological change with less noise than simple patent counts. In addition, the relationship between patent and research and development expenditure data is explored at the industry the relationship between patent and research and development expenditure data is explored at the industry level.;Finally, the appendix presents information from a survey that elicited information about the availability of renewal and other patent information from many national patent offices.