"The Costs and Bene fits of United States Patents" is the title of a paper by a four-man team consisting of John Turner, James Bessen, Peter Neuhäusler and Jonathan Williams. Labelled both as Boston Univ. School of Law, Public Law Research Paper No. 13-24 and Boston Univ. School of Law, Law and Economics Research Paper No. 13-24, it is conveniently available via SSRN here. According to the abstract:
"We use a detailed data set to estimate the costs and benefits of United States patents. To estimate costs, we combine data from Derwent Litalert with a proprietary dataset of non-practicing entity (NPE) lawsuits collected by Patent Freedom, and use an event study approach to estimate losses suffered by alleged infringers during 1984-2009. To estimate benefits, we combine patent data from the USPTO and EPO with financial data from CRSP and COMPUSTAT, and use market-value regressions to estimate the value of patent rents for publicly-traded US firms during 1979-2002.
We find that costs exceed benefits overall and that the gap between costs and benefits has grown across time. Surges in the number of NPE lawsuits, lawsuits filed over Computers/ Communications patents, and lawsuits brought against non-manufacturing, software and telecommunications firms contribute to the increase in the gap. Growth in costs outstrips growth in lawsuits, in part, because events in these fast-growing categories have higher-than-average per-event dollar costs".This blogger is no economist, but he can't help wondering: what about all those patents that are never litigated, as well as those which are happily and uncontentiously licensed -- and what about all those patents which are not asserted by NPEs?
Thanks go to Gerry Gavigan for forwarding this link