Accumulating Patents at Universities
Most US universities maintain three core businesses that earn most of their revenue:
- Selling diplomas
- Competing for federal and industry research sponsorships
- Trying to crack open the checkbooks of wealthy alumni
Since the 1980s, universities have ventured into a new line of business: patenting inventions from university research labs and brokering these patents to businesses and start-ups. Thirty years later, university patent holdings have swelled into the tens of thousands and larger research universities spend millions of dollars each year on filing for new patents. Yet, on average, over three-quarters of university patents are never licensed to companies for commercial use. Since US universities own 5% of our nation’s patents, and a growing number of patents in cutting-edge fields such as nanotechnology and biotech, even on human genes, people get worried that needlessly “locking up” basic university research will stifle innovation and create a patent anti-commons.
In the terminology of Henry Chesbrough, in the open innovation ecosystem, universities have branched beyond their traditional role of innovation explorers (they generate knowledge) to become innovation merchants (they license their knowledge to other organizations). While not everyone agrees it’s good for society for university’s to own and broker patents, from the perspective of business strategy it makes sense if a university files for a patent at the request of company that plans to license the invention.
What’s is harder to understand, however, is what motivates universities to continue to file for patents when there’s no licensee in sight. My question is are not intended to point the finger at the capabilities of university tech transfer staff who adeptly juggle large volumes of very complicated patent paperwork. Nor do I mean to question the effectiveness of any given university’s technology transfer strategy. I am genuinely curious.
Why university accumulate patents is a controversial, complicated and poorly understood topic. The debate around the topic brings to mind the seven blind men and the elephant. If you ask an optimist or a politician why universities continue to invest in unlicensed patents, they’ll tell you that it’s to incent companies to invest in developing a product, to motivate the faculty, and to make sure the university invention gets a fair shot at finding a home in the commercial marketplace. If you ask a pessimistic or someone who’s had a bad experience, they’ll tell you that universities file patents willy-nilly based on the political pull of the inventing faculty member. Or that staff don’t know what they’re doing. Or worse, that universities have become genteel patent trolls, guarding piles of un-used patents and suing companies and researchers that made money on an unlicensed university invention.
Some patent accumulation, perhaps, could be explained by the motivations above. However, the real reason universities end up with largely unlicensed patent portfolios is this: university patent accumulation is an unintended consequence of the inefficiencies inherent in our current university technology transfer model. The current tech transfer model creates a situation in which universities accumulate patents because they patent more inventions than they can license. Then, after a patent is issued, if there’s still no licensee in sight, a university’s technology transfer office is reluctant to let outside parties try their hand at finding a licensee in exchange for commission.
Two indications point to the fact that university patent accumulation reflects underlying inefficiencies in our current tech transfer model: one, data indicates that each year, university file for utility patents independent of that year’s licensing activity. Two, when comparing patent and licensing activity across universities of similar size and resources, the numbers are all over the map.
In theory, if a university only filed a utility patent application at the request of a company wishing to license the invention, there would be no patent accumulation. Although some universities report that this indeed, is their patent strategy, in fact, AUTM data from the top 50 US research university says otherwise. Most research universities file new utility patent applications without the presence of a signed license. If it getting patents was indeed a license-driven process, the chart below that depicts the number of utility patent applications vs. new licenses executive should look more like a line. (Other factors may be in play here: part of this seeming random filing could be that a single license covers several patents. Also, since a typical license takes 6-8 months to negotiate, some license completion may be spilling into over into the next year).
The only consistently demonstrated patent strategy across US universities was filing a provisional “placeholder” patent on new inventions to buy time to find a licensee and to figure out the potential commercial value and use of the invention. Across the board, most universities file provisional patents on roughly three-quarters of their new invention disclosures.
Perhaps since most utility patent applications are not motivated by company request, nor directly related to a university’s three core businesses (teaching, research and alumni donations) AUTM data indicates that patenting activity between university varies wildly, even between universities of similar size, funding and technology transfer activity. (For a clearer picture, I pulled out the California and Texas systems since their research funding and number of new disclosures is significantly higher.) University patent-related activity is all over the map. For example, at leading US research universities in 2009
- A few universities filed zero utility patent applications; one filed more than 200
- Some universities filed 50 to 60 new provisional patent applications, while another university with similar resources filed more than 300 new provisional patent applications
- The percentage of provisional patents filed per utility patents filed was completely inconsistent across universities. For example, one university filed 145 provisional patents and not a single utility patent application. The same year, another filed 221 utility patent applications and only 134 provisional patent applications.
- Spending on patent-related expenses ranged from less than a million dollars at one university, to over $16 million at another
The fact that universities file for patents without a clear commercial license in sight is not necessarily a bad thing, despite legitimate concerns about the impact of patent accumulation on our innovation ecosystem. Nor is it necessarily a bad sign that across universities file for patents in such widely varying numbers and with no apparent underlying strategy. As patent brokers, universities have the right to take an individualistic approach that reflects their own unique internal logic and is driven by local need and local mission. Extreme local variance could indicate an underlying strategy unique to that university’s areas of research expertise. If patent filings reflect the number of high-potential new inventions that came in that year and are based on sound market research and an intelligent case-by-case basis, then cross-university variance may simply indicate diligence and adherence to a locally-defined patent strategy.
At the end of the day, a university’s relationship with its patent portfolio is a complicated one. In comparison, a company’s patent strategy reflects and is a tool to support its larger business strategy (see the fantastic Kauffman-funded report by Ted Sichelman and Stuart Graham). It’s not so straightforward with university patent strategy. The current technology transfer method rests on a model in which universities attempt to be innovation merchants, despite the fact their core business is not brokering IP and the patents they broker have no relation to their core revenue streams.
- A hefty patent portfolio does not attract tuition-paying students and does not play a part in faculty advancement.
- Federal funding isn’t allocated according to universities that own a particular type or certain number of patent.
- Companies bearing research sponsorships are typically drawn by faculty expertise and university research capabilities.
- Alumni donate money because they get their name on a bench.
In other words, successfully licensing university inventions is an incredibly difficult task! Patent accumulation is not the fault of under-performing tech transfer offices, nor the malevolent master plan of greedy university administrators or faculty. Instead, it’s much more subtle and complicated than that. Universities simply don’t have the same incentives for filing and hanging onto a patent that companies and startups do. Nor do they have the same resources and business models that permit IP brokering companies to succeed (although even successful IP brokers accumulate patents and regularly go out of business).
Why do universities own significantly more patents than they can license? Well, for several reasons:
- Being a patent merchant is darn hard work and even the best IP brokers end up with more inventory than they can sell
- Patents that staff can’t find licensees for are set aside and no longer marketed. Most universities are reluctant to permit third party agents to take over their commercialization process
- Picking out patents that may have future commercial appeal is nearly impossible given the fact that by design, most university research is early stage and covers a huge range of territory
- Releasing patents into the public domain or regional IP pools is scary and involves its own legal and political complications;given the fact that patents can costs tens of thousands of dollars, letting patents loose is difficult to fiscally justify
- Licensing unlicensed patents older than 2-3 years with non-exclusive, royalty-free, no cost, no terms, “go in peace” license may not please some companies, but it may lure others out of the woodwork; again, hard to justify, may invoke resistance from inventors; politically risky
- Staff are given too many patents to handle; finding the right companies to invest in one’s raw IP is time-consuming and sophisticated work
- Universities don’t get to choose their patents or develop their portfolio in a particular direction, say becoming a “nanotech specialist; patents get dropped off at the door
- Universities don’t need their patents to build up their core business; university technology transfer is a sideline at most universities, not a core function
- It’s not unusual for patents to be pursued at the request of a faculty member, even though there’s no commercial interest in the technology
- The number of patents issued remains a core “performance” metric at many university technology transfer office
We should turn the debate away from the supposed motives (both for good, and for evil) underlying university patent accumulation. That debate barks up the wrong tree and doesn’t provide a solid foundation for stakeholders to rationally think about what to do next. Here’s what comes to my mind:
- Does patent accumulation harm anyone? If so, whom and how?
- Is getting and hanging onto unlicensed patents worth the university’s time and money?
- After 30 years of the current model, are universities effective innovation merchants?
- If you think change is needed, what, specifically would help?
- Would the introduction of commercial free agents into the university tech transfer process help break up the back log of unused patents?
Melba Kurman writes and speaks about innovative tech transfer from university research labs to the commercial marketplace. Melba is the president of Triple Helix Innovation, a consulting firm dedicated to improving innovation partnerships between companies and universities.