Study shows patent licenses don’t lead to tech transfer

It’s been another bad week for America’s benighted patent system: a Texas jury ordered Samsung to pay a patent troll $16 million for using Bluetooth — even though the “inventor” admitted Bluetooth had been on the market for years before he patented it. The system is a mockery.

Still, its defenders say, the system is working because it facilitates “tech transfer” and the passing of critical knowledge from inventors to companies. But now even that justification is collapsing in light of a new study that suggests patent licensing often does nothing at all to promote research or new products.

The study, titled “Does Patent Licensing Promote Innovation?,” is by Mark Lemley of Stanford and Robin Feldman of UC Hastings, and is based on a survey of 188 people whose job involves negotiating patent licenses at major companies and elsewhere.

The results, to put it mildly, are depressing. The screenshot below relates to patent trolls, and shows how often the survey respondents say that deals with the trolls result in technology transfer (top graph) and in personnel transfer (bottom graph):

Lemley/Feldman

In other words, those who paid patent trolls (which are responsible for the lion’s share of legal activity around patents these days) for a license say they received virtually zero benefit from doing so.

This is significant because the patent troll industry — which consists of a coterie of law firms, investors, judges, insurance fixers and so on — likes to claim that patents induce inventors to share knowledge that they would otherwise keep to themselves. But as the survey suggests, the licenses do no more than force companies to pay a tax for old technology, or for products they’re making anyways. Here is how two respondents, quoted in the survey, expressed it:

“Virtually every license my company has taken has been to ensure freedom of action for products or services we already offer. We have never received any value from a patent license other than to avoid litigation.”

“[Trolls] do not have any of the details worked out and they do not put any capital at risk developing any product, service or market. NPE’s simply exact a tax . . .”

The paper also notes that the same phenomenon, in which licensees pay money for nothing, is also pervasive when universities are the ones wielding the patents. Instead, as with the trolls, university patent deals rarely lead to meaningful tech transfer or innovation. The findings could have important implications at a time when more universities, including MIT and Boston University, are using decades-old patents to demand money from Apple and other big companies.

While the study doesn’t offer specific prescriptions for reform, it does repeatedly raise the fact that many of the (apparently useless) patents at issue are being asserted at the end of their life-cycle. This points to a solution in the form of shorter patent terms or, as Professor Brian Love has suggested, the President could order the Patent Office to change its fee structure so as to discourage trolls from hoarding patents on obsolete technologies.

The paper notes that its sample size is small, and that more study is needed, but the survey could still gain attention in light of the provenance of the authors.

Feldman has been instrumental in revealing the massive trolling operation, involving hundreds of shell companies, carried out by Intellectual Ventures, the standard bearer for the patent trolling business. Lemley, meanwhile, is one the country’s most prominent antitrust and intellectual property scholars, whose work has shown how patents and innovation are not one and the same, and that many of the conventional justifications for awarding patent monopolies are false.

The study also comes at a time when Congress is trying for a third time to pass a bipartisan bill to reform the patent system.