Negotiating Equity Terms with Universities and Tech Transfer Offices
Photo by Pixabay
This article was first posted as part of Kevin Christopher’s Nucleate Activator series on Medium.
Have universities always been so aggressive with equity terms?
The issue of university and PI equity regularly arises in my legal practice, and over the years I’ve developed a handful of strategies for dealing with aggressive equity asks which I’ll share below. First, it’s important to emphasize that grabbiness is a very recent trend. Ten years ago very few universities sought equity in student-led startups, and if they did it was a very modest amount that students usually welcomed as a sign of support from and ongoing ecosystem relations with the university. Now, many universities are demanding 20% PLUS royalties PLUS milestones PLUS various license and patenting fees from their startups. On top of that, PIs are increasingly demanding some form of equity relationship if a core technology was developed in their lab. A powerhouse PI on your advisory board is certainly not a bad thing, but it’s important to be conservative about how much equity is promised in that arrangement.
What’s the big deal with 20%, 30%, or even 40% of your cap table beholden to the university ecosystem? After all, you may love your school and proudly promote your school ties to investors and partners. The problem is that MOST startups don’t tinker in the lab for a few years in between yacht parties prior to an IPO. Typically, a founder’s life is a slog, and you need to be properly incentivized to keep pushing. Your university dilution is just the start, as you can expect to be diluted by institutional investors in multiple turns. After a few years of bare knuckle fighting, as your founders’ stock continues to vest, you’ll be anything but enamored with your cap table compadres that you never hear from except: (a) in the form of a tech transfer office, to demand diligence updates on your patent license; or, (b) in the form of a PI, to let you know of a brilliant advancement on your tech that will be part of a new company led by a next-gen grad student, who coincidentally has asked the PI to be a co-founder of that company as well. Being a founder is exciting, but you need to have a sufficiently elongated long-term view to avoid signing on to a bad deal.
You should have a basic understanding of the law that underlies the tech transfer system.
Before I get into suggestions of how to negotiate palatable equity terms, let me very briefly cover the Bayh-Dole Act, 35 U.S.C. § 200 et seq (text). Here are the important points:
A contractor (e.g., university/lab/hospital) electing rights in a subject invention agrees to file a patent application prior to the expiration of the 1-year period referred to in section 102(b), and shall thereafter file corresponding patent applications in other countries in which it wishes to retain title within reasonable times, and that the Federal Government may receive title to any subject inventions in the United States or other countries in which the contractor has not filed patent applications on the subject invention within such times.
The term “subject invention” means any invention of the contractor conceived or first actually reduced to practice in the performance of work under a funding agreement.
The term “invention” means any invention or discovery which is or may be patentable or otherwise protectable under Title 35 of the United States Code, or any novel variety of plant which is or may be protected under the Plant Variety Protection Act (7 U.S.C. 2321 et seq.).
The contractor agrees to submit on request periodic reports no more frequently than annually on the utilization of a subject invention or on efforts at obtaining such utilization that are being made by the contractor or its licensees or assignees. Such reports shall include information regarding the status of development, date of first commerical sale or use, gross royalties received by the contractor, and such other data and information as the agency may reasonably specify.
Essentially, if any part of federal funding touches an invention in a lab, even $1 of federal funding compared to $99 of private or state funding, the resulting invention is subject to this law. The law gives title to that invention to the federal government, but a university/hospital/lab (“contractor”)can elect title by agreeing to file a patent application on the invention. The contractor has to provide reports upon request of commercialization efforts, but frankly that just doesn’t happen. An invention under this law is a patentable invention, and if you are dealing with a non-patentable invention, like many AI and software technologies, then rights should fall under data rights clauses of grants rather than tech transfer office ownership. Also, an invention is something that is conceived or reduced to practice, with major technical implications. (I’ll cover these topics and more in following posts. See also my overview posted by the Berkeley Postdoc Entrepreneur Program.)
That’s a lot of lead in to what you came to read about, now let’s work through the heart of this post.
Tip 1: Get a letter from a VC saying he/she is really excited about the tech, but unwilling to invest in this cap table structure, now or in the future.
This is a very effective strategy, particularly outside of the rich coastal investor hubs. I’ve seen scenarios with equity asks as high as 53% (20% to university, 33% to PI) to what is effectively dead weight on the cap table. It’s hard for two co-founders to bypass other opportunities and invest their souls into a startup splitting only 47% of founders’ stock (eventually diluted by angels, advisors, employee/contractor pools, and institutional investors). Many savvy investors wouldn’t want to back a founder who jumps ship for med school or another startup with a better equity incentive, so the pain points here are real. A credible VC endorsement can help you work the system a bit. A university is not required under Bayh-Dole or any other law to get a good equity return to the office; if its rep tries to tell you so he/she is lying.
The equity trend is pure opportunism, and 5% of an investable company is better than 20% of something that flames out after going through a brand name accelerator (that also takes its pound of flesh). The letter is a lever, but you’ll very likely have some equity relationship, however small. Ask for a clear explanation of how the university will work for you. My startup Quantiscope has been fortunate to benefit from the MIT Startup Exchange Program, where they matchmake industry pilots and afford us stage positions to pitch during tech conferences. This is the type of benefit you should be getting from an equity holder that only wins when you win.
Tip 2: Offer equity in a holding company which sublicenses the IP to an operating company which takes in venture capital.
This is a little bit atypical, so you have to spell this out in order to get buy-in. I’ll help you spell it out.
The traditional licensing model is centered out a particular patent or patent family. It includes terms that tie the royalty and milestone obligations to the in-licensed patent on a claim-by-claim basis. If additional patents outside of the university IP form the basis of a marketed product, then the license typically include a calculation that minimizes obligations to the university licensor proportional to the weight of their licensed IP in the final product. Here’s an example:
For a combined product or service that contains or uses a Licensed Product or Licensed Service, and at least one other active product or service that is not a Licensed Product or Licensed Service (a “Component”) (such combined product or service, a “Combination Product”), Net Sales shall be calculated as:
A/(A+B) x [Net Sales of the Combination Product (without regard to this formula)], where
“A” is the total of Net Sales of each Licensed Product or Licensed Service contained within or used in the Combination Product when Sold separately in the applicable country during the applicable time period; and
“B” is the total of Net Sales of each other Component contained within or used in the Combination Product when Sold separately in the applicable country during the applicable time period.
If A or B cannot be determined because Net Sales for the Licensed Product or Licensed Service or the other Component sold alone are not available in a particular country, then Net Sales for purposes of determining royalty payments shall be agreed by the parties in good faith based on the relative value contributed by the Licensed Product or Licensed Service, on the one hand, and each other Component, on the other hand.
This excerpt is missing definitions that tie the Licensed Product or Licensed Service to one that incorporates the patent that a licensor holds. Essentially, for decades it was common practice that university/hospital/lab licensors considered it reasonable that an existing company that in-licensed IP might be combining multiple flavors of IP into a consumer product sundae, and that startups would eventually build their own proprietary portfolios. The equity trend goes well beyond the traditional model, and also fails to appreciate that some licensor IP may be, well, junk. Patents fail to issue for many reasons; sometimes it’s because of a PI’s own prior publications. In other cases, the initial disclosure may just set the table for downstream IP that’s filed solely in the startup’s name, and at investor expense, that actually covers the marketed product. Why should the licensor and/or PI carry a portion of your company in cases where the initial IP isn’t even relevant to the value of the company?? With these arguments in hand, you should be able to present a structure where the holding company gets a good return relative to the value of licensed IP it holds, and the university obtains a fair return from that arrangement. Your downstream investors and investors benefit from more equity in the company being available for their financial and labor commitments. Everybody wins!
Tip #3: Obtain a low-cost option, push it out as far as you can, file your own IP that improves upon / blocks the technology, and use that to force the licensor’s hand.
This doesn’t work in every case. If we’re talking about a chemical technology and a university’s initial patent application is pretty broad, this might not work. However, even in those instances you could file formulation or specific use patent applications that would block the source patents if issued. This devalues the threat of the tech transfer office and/or PI to license the source IP to a competitor. It also allows you to make a valid argument that the source IP is only a small component of the company and thus not deserving of a significant portion of company equity.
Tip #4: Identify a similar technology somewhere else, negotiate a term sheet, and use it to force the licensor’s hand.
This option is not for the faint of heart. It can create some bad juju, but it is a real path. Every university has hundreds of patents sitting on a shelf, and it is likely that you can find something in the same hemisphere that you can theoretically obtain, relatively cheaply, for at least an initial an IP position. If the founders hold the real know-how, then additional IP can always be filed downstream. You get to say you’re a startup out of X’s lab at X University, with an IP framework, without a big hit to your cap table. This is of limited applicability, and you’ll understand best whether it can work for you.
Tip #5: Argue that the technology is not patentable, and therefore is not subject to Bayh-Dole patenting and commercialization requirements.
This is a possible route for many computational startups. The first response from a university will probably be that the employment agreement you signed assigned all IP to the university, whether or not patentable. You could then posture that you will open-source the technology (which as many of you know doesn’t always spell out everything), and launch a company free and clear of the university that uniquely uses the technology with underlying expertise in its creation, and recognition of its authorship. Their reply might be that you won’t be as investor friendly without exclusive patents/IP, but there are plenty of unicorns that had no exclusive IP at the outset.
Tip #6: Ask for a non-exclusive license instead with a low, single-digit equity alternative.
The first response I usually get when suggesting this is something along the lines of investors being freaked out with only a non-exclusive license. Maybe so with less sophisticated investors. But here are some of the reasons this can work in your favor:
The licensing staff in tech transfer offices are typically held to quotas and incentivized based on numbers. If they’ve already licensed a technology, they’re unlikely to be out shopping the same tech to others.
This is especially true when the licensor holds an equity position in your company. Why would they want to do a deal with someone else that limits your competitive market and thus limits the potential value of their equity position?
Often times, a patent license is only as good as the know-how that comes along with it. If you’re a startup founder and originator of the technology, you are not going to be available to facilitate the university’s additional non-exclusive licenses to others.
