There is much interest in statistical analysis of patenting trends by academic institutions (where they are an applicant), but there are problems in gathering the data.
Their names vary -- university, college, institute, school, and so on, besides foreign language equivalents. Many terms could mean something else, as in institute. Also, in the UK, there are at least two major players which do not use any of these terms -- Isis Innovation (for Oxford University) and Imperial Innovations (for Imperial College). There is also the complicating fact that some institutions license their innovation to companies such as BTG, sometimes before any patent application is actually published, or between the application's publication and the grant.
These factors make it hard to get true figures.
There is a very useful survey on pages 16-28 of the PCT Yearly Review 2014 of how universities and public sector organizations use the PCT system. The PCT, or Patent Cooperation Treaty, offers a way to simplify, cheapen and delay expenditure when an applicant wishes to secure protection in foreign countries. Instead of applying to many countries for foreign protection within the permitted 12 month period after the initial filing, a single application is made to WIPO in Geneva and a single document is published 18 months after the initial filing (coded as WO among patent documents). A search report listing relevant prior art is either included with the publication (in which case it is coded A1) or follows later (in which case the A3 search report supplements the already published A2).
Only after the A1 or A2 is published do requests come to the applicants from member states of the PCT listed on the application asking if the applicant wants to go ahead with requesting an examination to decide if a patent can be granted. If relevant a translation into an acceptable language is requested. Each country or regional system then decides individually whether or not to accept the application. All this means that in cases where the applicant wants to withdraw because the invention does not seem to be new, or there is a lack of finance, or the market is not perceived to be adequate, the applicant can withdraw and save the cost of translations and other costs.
The survey states that nearly 7.5% of all received applications through the PCT in 2013 were from academic institutions. While traditionally these were largely from America or Europe, Asia was catching up fast.
The survey goes on to say that only 4% of German or Italian academic institutions patent in their own name,12% in France, 20% in The Netherlands, 32% in the UK, and 53% in Spain. The source quoted is from a 2011 paper in Research Policy, 40(1), 148-164, The European university landscape. An abstract and an opportunity to purchase the paper is at the ScienceDirect website.
The survey tends to distinguish between university and public research organizations (PROs) in its figures and pie graphs, giving numerous facts and analysis. The survey says that its use of "filed" really means published -- which is confusing !
Allowing for the factors mentioned above, in 2013 universities published through the PCT published 9,804 applications, and PROs 4.411. Both were increasing more rapidly than the private sector, especially universities. They are becoming more aware of the potential value of their inventions.
On pages 22-24 there is discussion of co-applicants, where more than one applicant is named on an application. Typically this is when some sort of collaboration occurs, either in the actual research or in financing (or both). During 2011-13 13.7% of published PCT applications had co-applicants. This was 16% for universities and 19% for PROs.
Page 25 gives the top five university applicants from each continent and the continent's figures as a whole for 2005-13. Of the total of 28,153, 11,823 were from North America. 9,065 were from Asia. 6,421 were from Europe, including Isis Innovations, and Imperial Innovations, which are known as surrogates for Oxford University and Imperial College. A similar table for PROs is on page 27.
There is also analysis by technical field. It confirmed my impression that academic patenting is biased towards biology and chemistry.
All in all, a very useful survey. All of this comes down to technology transfer as any exploitation of the value of the patents is done by licensing out, selling or spinning off companies. There is much research into this topic, such as the 2013 book by the OECD, Commercalising public research: new trends and strategies.