Talking about patents in law schools and policy spaces we often say that a patent ‘teaches’ something. What a patent teaches or is meant to teach the invention. The idea is that patents represent a bargain between an inventor and society; a patent is meant to be a description of an invention that would allow someone who makes that kind of thing (we call this someone ‘a person skilled in the art’) could make the invention. Some patents are just method patents – they describe a method for making something, some describe the thing itself but with the kind of detail that enables the thing to be made.
It will come as no surprise to readers, except those who just arrived from 1985 by time travel*, that there is a great deal of criticism of contemporary patents systems failing to disclose enough information or the right kind of information. Sometimes this is because the person who gets the patent has never actually made the thing: implemented the invention we would say in patent-talk.
But what if ask the same question not of individual patents but of data that the patent system holds about innovation. That is more or less what Jonathan Berger and I did. We looked at a ten and a half year set of patent data on patents in South Africa. We recently published a report on our research which you should download.
A summary of our findings is available in an article in Business Day.. The Intellectual Property Unit at the University of Cape Town also discusses the research.
I have to admit that I am both surprised and impressed that Business Day published that article because it is rather detailed and dare I say dry in places. It explains that when we looked at the data we found that only about 10% of the patents awarded in South Africa are awarded for South African inventions. The other 90% are given to foreign entities, mostly corporations. If the patent bargain is to give a state back monopoly to an inventor in exchange for an teaching an invention then for 90% of the state back monopolies being handed out either the monopoly granted by South Africa is unnecessary to encourage disclosure because the invention was patented somewhere else first and a patent in another country such as the United States was enough to induce disclosure or the South African monopoly was needed to enable disclosure but the profits made from the South African patent are taken out of South Africa. Since South Africa is a relatively small market it seems more likely that South African patents are superfluous as incentives to innovate.
South Africa grants patents that are based on foreign patents because of a different bargain, The Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS).**
South Africa does not turn away people who have already patented in other countries because they have already disclosed the invention but instead uses the patent application in that other country to put the applicant in South Africa ahead of anyone else claiming a patent for that invention (in patent talk this is known as ‘establishing priority’). One frequent response to the 90% foreign versus 10% local by pointing out that South Africa could somehow increase the percentage of local inventions.***
Our research concentrated on that 10%, asking what is patented in South Africa and who is patenting it. Read the article or better still the article and the report for the findings. One significant finding is that most high tech research which is also patented in other countries is State funded, carried out at a historically privileged university or a research institute such as the CSIR. Patenting by individuals and even by companies is mainly in low technology areas. South Africa unlike any medium or large economy does not actually examine patents, that is check whether someone has already invented the invention or whether the patent actually teaches anything. After looking at the codes for the patents, reading the titles of the patents and the abstracts where those are available there does not seem to be an emerging technology in which the South African private sector is leading innovation.
This kind of research is unglamorous, arduous, detailed and sometimes quite boring. The conclusions it yields are often sobering rather than exciting. But every discussion about patent reform must be grounded on the reality of how much innovation is actually taking place in South Africa and how much of a role patents actually play in that innovation.
*Yes, it is entirely probable, indeed likely, that one of the first things that someone who has time traveled from 1985 would do is read this blog.
** For the more pedantic readers, yes there are several treaties involved, why don’t you detail these in the comments.
*** But exactly how to change that percentage turns out to be a non-trivial question.
