While we work on repeal of Bayh-Dole, a dismal failure of a law, and work out how to repair the damage–the lost non-commercial development community that exists only in a few areas such as open source software and ought to be robust in medicine but is not; the disconnect between university-hosted science and professional and industrial use of that science; the disinterest by federal agencies in funding development of work deemed important enough to fund as research but not important enough to fund as discovery–what can be done?
Here’s one thing: enforce Bayh-Dole.
Almost nothing in Bayh-Dole having to do with public interest is enforced. Not the law, not the default patent rights clause, not the government license. Stuff is ignored, waived, not acted upon, kept secret.
Enforcement,compared to repeal, is fast. Enforcement does not require Congress to act. Enforcement can be done by executive order and by action by the Department of Justice. Enforcement can be done through audits, with demands for reporting under oath.
What should be enforced?
Start with federal agencies in their compliance with Bayh-Dole and move then to federal agencies enforcing on contractors the standard patent rights clause Bayh-Dole requires them to use as a default when they can’t bear to use a better clause.
Executive orders can require federal agencies to develop alternative standard patent rights clauses to address situations in which it is clear that the default position on inventions should not be contractors may keep whatever inventions they can manage to acquire. That means opening up research tools and unique research resources. That means allowing anyone to make and use an invention. If there’s going to be an assertion of exclusive rights, keep that for sales, and raise the standard for what has to happen before someone can hold a patent right to prevent development directed toward products that might be sold.
Executive orders also can require federal agencies to act on the government license that comes with every subject invention for which a contractor elects to keep title. That license is broad–to make, use, and sell, and have others make, use, and sell, for any government purpose. That license extends to states (which cannot be sued for infringement anyway). So direct federal agencies, anywhere that the government is paying for the use of inventions made in work that it has supported, to determine whether they could provide that same public benefit by making and using the invention themselves or commissioning others to make, use, and sell product based on the invention on behalf of the government.
Think of “reasonable terms” this way. If the federal government, acting on its license, can make, use, and sell a particular invention for less than what a company holding a patent on that invention has priced products made under that patent, then the federal government should do its own making, using, and selling–or authorize others to do so. In the case of drugs under patent, authorize generic manufacturers. That’s a basic comparison. A reasonable price is what it would cost the government were it to act on its license. In some cases, all that entails is the government asking a company to do the manufacturing for it. Consider Xtandi–sold for something like $80/pill. A generics company offered to produce Xtandi for $3/pill, including its reasonable profit. The federal government could take that offer tomorrow, since Xtandi is based on a Bayh-Dole invention. Screw it. Do it.
While folks are on the government license, enforce the scope of that license. That means enforcing the fundamental architecture of Bayh-Dole. Bayh-Dole applies whenever a federal contractor acquires ownership of an invention made in work receiving federal support. The “work” can be any project, anything with a scope of work, a purpose, a goal. The federal government need only provide funding for some portion of that work. At the start, for research, or in the middle for applied research and prototypes, or towards the end for development or testing. Federal funding anywhere in the “work” makes any invention in that work, when acquired by a federal contractor, a Bayh-Dole invention, even if the “inventing” itself doesn’t involve the use of federal money. You follow? What’s the purpose? To discover compounds that can be developed into drugs. Then what’s the work? Anything from research to discovery to screening to formulation to delivery to testing? Anything invented anywhere in that string of activities, if acquired by a federal contractor, becomes a Bayh-Dole invention.
The government, then, has a license to any such invention. Most of these inventions, however, go undisclosed by federal contractors because they use a different standard, not in Bayh-Dole, to decide if an invention is a subject invention. They claim that only inventions directly funded by federal money come within the scope of Bayh-Dole. But that’s not true. For which see the discussion on scope at 37 CFR 401.1. But also see Bayh-Dole’s definitions, including those of funding agreement, contractor, invention, and subject invention at 35 USC 201.
It gets better. A federal contractor may add additional contractors to a funding agreement. That’s the upshot of the definition of funding agreement at 35 USC 201(b). A contractor is added by “any assignment, substitution of parties, or subcontract of any type.” If a federal contractor assigns an invention, authorizes another party to act in its stead, or subcontracts work, those new parties accepting any such deal become parties to the funding agreement and by the definition of contractor at 35 USC 201(c) are also contractors. Thus, if a university assigns a subject invention to a company for “development” into a commercial product, then that company becomes a federal contractor and any patentable inventions that company makes, whether patented or not, once the company acquires them, become Bayh-Dole inventions. The company must disclose them or risk losing title to them, and the federal government has a license to those inventions to make, use, and sell, and authorize others to do so on behalf of the United States.
Bayh-Dole is also express that when a nonprofit assigns a subject invention, any company receiving that assignment must comply with the nonprofit’s patent rights clause. Under Bayh-Dole, nonprofits have special obligations. See 35 USC 202(c)(7). One of these is that they must share royalties with inventors. So when a company acquires a nonprofit subject invention by assignment, the company also must share royalties with the inventors–regardless of what the nonprofit shares of its take. Another is that the nonprofit must use all income earned with respect to a subject invention for scientific research or education, and may deduct only expenses incidental to the administration of subject inventions. For a university, this is all fine, since most whatever it makes after sharing royalties arguably goes to either research or education. But when a university assigns a subject invention to a company, the company has this constraint as well–anything the company generates by way of income from the invention goes to the specified uses, with only deductions for specified expenses.
Talk about a limitation on price! The company either prices low because it does not care to contribute income to scientific research or education, or it prices high and hands whatever it makes over to those purposes. No dividends for investors. No recovery of past costs. No fancy dancy headquarters construction.
And here’s the ruse universities use to avoid all this. They grant what they label “exclusive patent licenses.” Those licenses insist that the university retains title to the licensed patent. But those exclusive patent licenses routinely convey all substantial rights in the subject inventions. Courts have repeatedly held that any instrument that conveys all substantial rights in an invention conveys ownership in the invention–it’s an assignment regardless of what it is labeled. A telltale sign of assignment is that the “exclusive licensee” has the right to enforce the patent. Only the owner of an invention can enforce the patent on that invention. And that right is included, one way or another, in nearly all university exclusive licenses. They exclusively license patent rights, and they also assign the underlying claimed invention. Bayh-Dole’s nonprofit requirements apply. Enforce those requirements. Do it by executive order if federal agencies balk.
Don’t let NIST fuss around trying to change Bayh-Dole’s implementing regulations to make it appear that the federal government does not have a broad license or that inventions made by companies who accept assignment of subject inventions are not also subject inventions. And audit every nonprofit exclusive patent license involving a patent citing government support to determine whether it also assigns a subject invention. If it does, audit the university and the assignee to ensure that all income less allowable deductions has gone to scientific research or education. That’s a matter of enforcement. It can happen now. There’s not even a procedure in Bayh-Dole by which a contractor can delay the audit or demand for compliance with some drawn out appeals process. If a contractor (nonprofit or assignee from a nonprofit) has violated the law and breached the nonprofit patent rights clause, and refuses to make good on it, then sue them for the amounts they owe, in effect, the public. And while the government is at it, if the contractor doesn’t make good, take title to all those undisclosed inventions owned by the company and which became subject inventions when the company accepted assignment of the subject invention.
We can also turn to what Bayh-Dole calls the most important provision of the law, the requirement for US manufacturing, 35 USC 204. The provision applies to any subject invention, but is limited to exclusive licenses in the U.S. to use or to sell. For any such license, product using or based on using a subject invention must be substantially manufactured in the United States. Thus, as companies rush to avoid having to dedicate the income earned from subject inventions to scientific research or education, they end up having to manufacture in the United States unless they downgrade all the way to non-exclusive.
This is important. The most visible premise of Bayh-Dole was that the United States was falling behind other countries in technology development. Building a manufacturing base on new research discoveries was just the thing to restore U.S. leadership and create jobs for Americans. Sounds nice, in a nationalistic sort of way, but why not? For biomedical stuff, this means that what we invent here (in work receiving federal support) ought to stay here, be “made in the United States by United States industry and labor,” to quote 35 USC 200, which lays out Bayh-Dole’s policy and objective. Bluntly, for subject inventions, the manufacturing supply chains should be U.S. based. Not based in foreign countries. Nothing personal–it’s a U.S. first sort of thing, for inventions made in work with U.S. funding.
It’s funny–funny-strange, not funny-funny–that in all the ballyhoo about how successful Bayh-Dole has been for the biotech industry, there’s not a peep about how the (apparently) many (no more than 200) drugs have all been manufactured in the U.S. Clearly, they haven’t. And it’s not because they haven’t been licensed exclusively in the U.S. It’s assured that many of them have been licensed exclusively in the U.S. But what is this? Bayh-Dole 35 USC 204, the most important bit of the law, allows federal agencies to waive the U.S. manufacturing requirement.
The NIH, leading the way with waivers, has gone so far as to set up a web site to expedite waivers. All a contractor has to do is argue that it tried to find a U.S. manufacturer and couldn’t, or didn’t bother to try because it was obvious there wouldn’t be one that would take the deal the contractor had on offer. It’s funny-sad! Here the law sets up to promote U.S. manufacture for new discoveries and then creates a backdoor by which contractors can take the manufacturing offshore on the thinnest of reasons–and the NIH enables it. An executive order can put an end to the practice. If a contractor cannot find an exclusive licensee willing to source product made in the U.S., then it should offer non-exclusive licenses. If the point is to encourage U.S. manufacturing, then the last thing is for a federal agency–NIH, here–to allow contractors to license exclusively for foreign manufacturing, effectively cutting out future U.S. manufacturing rather than helping to build new capability where before it was lacking. So, end the NIH waiver program and focus on returning biomedical supply chains to the U.S. for inventions made in federally supported work. If contractors don’t like it, tough–that’s the law, and the core stated purpose of the law–so they can either work harder to build up U.S. manufacturing for their discoveries or license non-exclusively and let U.S. manufacturers have access whenever they get around to wanting to make this new stuff.
There–some examples of how enforcing Bayh-Dole and its patent rights clauses would go a long way to addressing the lack of innovation, the abuse of monopoly patent positions to price gouge, and the failure of Bayh-Dole to develop U.S. manufacturing when it ought to.
There’s much more. We haven’t even got to the abuse of federally supported inventors that Bayh-Dole enforcement would end. But even these things would be a good start. Doesn’t need repeal. Doesn’t need amendment. Doesn’t need new regulatory codification. Just enforce the law that there, and use executive orders to prevent federal agencies from working to undermine the law, ignore the standard patent rights clause requirements, and refuse to act on the outright government license that comes whenever a contractor chooses to keep ownership of a subject invention.