Over at The Hill, James Edwards has an op/ed piece, “Don’t sacrifice patents for politics,” that worries changes to Bayh-Dole march-in procedures. Almost everything about the piece is predicated on the “usual narrative”–which is almost but not entirely false. Let’s have a look.
Before the Bayh-Dole Act, federal research funding was your tax dollars only partially at work. Federal funds went into research — advancing our understanding and producing discoveries — but the government kept the practical application of this new knowledge bottled up.
Before the Bayh-Dole Act, there was the IPA program. It allowed universities and nonprofits to acquire title to inventions made with federal support and was used by the two federal agencies that provided most of the funding to universities–HEW and NSF. It had a 5% commercialization rate. By contrast, HEW’s own commercialization rate was 23%. Bayh-Dole’s rate is kept secret, but major universities report commercialization rates on the order of 0.5%. The government did not “bottle up” new knowledge–just the opposite. The government published discoveries and dedicated the rights to the public. The government made research results broadly available for use. Mr. Edwards has it all ass-backwards.
The problem was that big pharma in particular did not like the loss of possible monopolies. It wanted a federal policy that allowed the creation of private monopolies based on publicly funded research. When it couldn’t get its way with the HEW, universities did its dirty work for it and created the illusion that all research was being “bottled up” and that the only remedy was a “uniform”–i.e., arbitrary–federal patent policy across all federal agencies, research programs, purposes, and contractors.
This 1980 bipartisan law, along with legislative amendment in 1984 and President Reagan’s executive orders expanding Bayh-Dole’s availability, enabled federally funded basic research discoveries to be put to practical use.
Nonsense. Bayh-Dole did not enable “discoveries to be put to practical use.” Bayh-Dole enabled the creation of private monopolies without public accountability–especially so after the 1984 amendments stripped the law of many of its public protections. It’s just ideology that private monopolies are the only or best or first way that anything discovered in research can come to be used. Bayh-Dole permits the exclusion of practical use of discoveries in favor of speculation on the future value of patent rights. It does this arbitrarily–across the board, for any discovery in any federally supported research program, no matter what. No one reports practical use. No one cares about practical use. What they care about is maximizing income while deflecting criticism. The key term is “commercialization”–which is not a stated objective of Bayh-Dole. What Bayh-Dole is used for is to withhold discoveries from practical use in favor of speculation on commercial product development.
Its cornerstone is guaranteeing universities, nonprofits and small businesses secure, reliable rights to the intellectual property created under federal funding.
Well, if one is going to include Reagan’s expansion of “Bayh-Dole’s availability,” then why not include all contractors, large and small, nonprofit or for-profit? And Edwards is now way beyond rights in patentable inventions, making Bayh-Dole appear to concern “intellectual property.” This is a gross expansion of the apparent scope of Bayh-Dole. Anyone who knows better knows it is deceptive nonsense, and anyone who wouldn’t know is led down a rotten path.
But beyond all this, Bayh-Dole does not guarantee “secure, reliable” rights for universities and others. All Bayh-Dole does–arbitrarily–is allow contractors to keep rights that they acquire. And Bayh-Dole made it appear that universities should become predators on the patent rights of their faculty researchers. That is, Bayh-Dole prevents the federal government from intervening in the institutional exploitation of patent rights on research conducted for public benefit and with public support.
Government-run R&D didn’t work out so well prior to Bayh-Dole. Taxpayer dollars funded basic research, but there was no uniform federal policy about ownership or licensing of IP rights, and the U.S. government often retained them.
Right. Way more nonsense. Let’s not talk about the digital computer and the internet, not talk about nuclear power and space travel. Let’s leave alone air bags (in cars), gene splicing, oh, the heck with it. There’s too many things to mention. Government-run R&D was working remarkably well before Bayh-Dole. There were fusses in some industries, but most companies didn’t care about the patent rights. Here’s a useful table from the Harbridge House report from the late 1960s–well before Bayh-Dole:
Most companies did not care about patents as a vehicle to justify commercial investment. Of the six reported attitudes, only one cares. Even there, there were two patterns–get the patent rights before contracting with the government, or isolate the government work so it doesn’t have anything to do with the rest of the company’s business.
But let’s get at the major untruth–that there was no uniform federal policy about ownership and licensing. That’s simply wrong. The Kennedy patent policy, kept largely intact through Nixon’s term, was uniform and specific with regard to both ownership and licensing.
Nutshell of the Kennedy patent policy: The government should take title unless a contractor has the demonstrated capacity to use the invention and an established commercial position, or it is in the public interest for a contractor to hold title, especially if it commits to developing an invention that otherwise would not be supported by public funds. The government should make all inventions it acquires broadly available to the public through dedication or licensing. Contractors who take title should have three years of exclusivity from the date of patent issuance, and the government should be able to intervene in any such patent monopoly if the public interest would be better served otherwise.
Some folks might not like that policy–most companies did not mind it. But it was most certainly uniform. What it lacked was arbitrariness. That’s what “uniform” means when one disregards context and purpose.
there was no uniform federal policy about ownership or licensing of IP rights, and the U.S. government often retained them.
We leave you to contemplate the “them.” The policy concerned inventions, not “IP” rights in general. And the proper verb isn’t “retained.” The government negotiated the acquisition of patent rights (other than in the AEC and NASA, where legislation limited the scope of private property rights in certain inventions involving atomic or nuclear energy and space technologies). When the government obtained patents, it deployed them without a profit motive. This rankled some folks, university-affiliated patent attorneys among them. Look at all those opportunities to exploit private monopolies going to waste.
As a result, hardly anybody did anything with the discoveries.
No, the discoveries were published and used without formal requirements. There’s little truth to the claim of “hardly anybody.” Well, let’s go so far as to say that hardly anybody uses any discovery made in university research, regardless of whether there are any patents involved. But what Edwards here wants us to believe is that lots of people were doing things with other discoveries, just not the federal ones. While there are instances of federal agencies refusing to license patents exclusively, Edwards will have to do some work to show that “hardly anyone” was using discoveries that the government claimed title to (and may or may not then patent) and then released for public use. Actually, he can’t show it. He’s just repeating bullshit that sounds good for the usual narrative.
Federally funded research before the Bayh-Dole era led to the U.S. government owning 28,000 patents from research it funded.
Almost all of which were made generally available, and many of which were in defense systems where the government allowed contractors to acquire title and most did not–these were mostly inventions that contractors didn’t want to own. Even then, the “commercialization” rate was cited as 4% to 5%–and much higher for HEW inventions. That is, the same as the commercialization rate for inventions owned by universities and nonprofits under the IPA program. The government did without trying as well as the universities did throwing focused effort at the work. The difference is that 95% of the university IPA portfolio, because those patents were not licensed, was unavailable to the public.
But commercial entities used only 5 percent of those inventions, including discoveries with the potential to produce tremendous economic benefit, because the government retained and controlled the property rights. This included owning the patent on an inventor’s discovery if his or her inspiration and perspiration were supported by a federal research grant.
There is not a record of the use of inventions made without a license. The 5% has to do with those patents formally licensed. The actual use rates were assumed to be much higher. It’s only the political rhetoric to make it appear that federally owned inventions weren’t being used that’s at work here. Treating political rhetoric as fact is, well, a sign of insanity.
There’s no evidence whatsoever that any of the patents held by the government had such potential. But Edwards slips from the 28,000 patents to the discoveries that must underlie the patents. Sure, the discoveries might have potential. But that is not the issue for patent policy. The issue for patent policy is whether the potential in a discovery benefits from a limited monopoly (perhaps) and if so, whether that limited monopoly should come with a profit motive (as, in exploiting suffering for profit, say) and if with a profit motive, whether the federal government should exercise greater than ordinary care with how that profit motive is exploited.
Should the goal of federal support for basic research be the creation of private monopolies operated with a profit motive without federal supervision or public accountability? If you answer yes! yes! yes!, then you have got Bayh-Dole.
Now for some crocodile tears:
This included owning the patent on an inventor’s discovery if his or her inspiration and perspiration were supported by a federal research grant.
I’m not sure quite how to deal with this.
First angle: the government didn’t force anyone to do work for it. If someone wanted a contract, then they lived with the result. Contract research firms such as Arthur D. Little argued that they routinely gave up patent rights to whomever they did research for, the federal government included. One knew this result going in, inventors and managers and finance VPs and shareholders. The business model was to do the research, not to hold the rights out against the sponsor or against anyone else the sponsor might work with, or against an industry or the public.
Second angle: with federal ownership of title, an inventor had free access to the invention whether the inventor continued working for the contractor or not. But with Bayh-Dole, inventors lose that access–the university excludes all use while seeking an exclusive licensee–a monopolist speculator or a commercial developer (universities don’t differentiate).
Third angle: the invention has to be made within the scope of the contract (or is related to the work, for the AEC and NASA). There were mechanisms for reporting such scope of claims, but the standard was not “inspiration and perspiration” but rather whether any claim in a patent was supported by federal funding. Edwards repeats the “if even $1” b.s. that the Supreme Court mocked in Stanford v Roche. No one learns.
Fourth angle: boo-hoo-hoo for the poor inventors. But the universities took the inventors’ rights anyway–and under the IPA program they were required by federal contract to do so. So why are we here holding out sympathy now for inventors who (i) chose to seek federal support; (ii) work in universities that take ownership if they can; (iii) and who gain access to their inventions through federal dedication of their work (if they are so lucky); and (iv) likely have a greater chance of seeing their names on a patent than if they work through their university and its patent brokers?
Fifth angle. If title is so important, then why not let inventors own their patent rights and let them choose how they think those rights should be deployed? Why not let federal agencies work directly with inventors rather than force ownership to patent brokers? Even Bayh-Dole provides for this pathway. But universities make a huge effort to prevent it from ever operating.
Remember, that in industry one primary use of the patent system is to document the technical expertise of one’s technical team. No one expects to get rich off the patents–the point is to show priority and skill.
Before this law, federal agencies had 26 different sets of rules controlling commercial use of federally owned IP.
Nope. The 26 laws were established by Congress to advance specific missions. As one report concluded in 1969 with regard to another report: 
And:
The Kennedy patent policy recommended flexibility within a uniform set of defaults. The 26 different laws that agencies operated under reflected just a few of the hundreds of programs operated by the federal government. For universities in particular, such a diversity was nothing new: universities receive research funds from scores of different sources (now, hundreds), each with its own expectations with regard to patents. There are ways to deal with it. Reducing 26 special federal program requirements to one arbitrary system barely touches the diversity of contracting requirements that universities must still deal with. And given that most of the rules involving patents had to do with agencies acquiring title, the issue was not at all whether there must be one arbitrary federal policy on patents, but rather what practices, in a given context, would best serve the public interest. Bayh-Dole reduced this question to what practices would best suit patent brokers. Nice.
Grantees had to negotiate a waiver to take title to any discoveries. They obtained title on a case-by-case basis.
Except not in the IPA program to which over 70 universities and non-profits were enrolled. For those, title was upfront when they decided to patent. The waiver programs had to do with agencies that did not implement an IPA system (only HEW and NSF used IPAs). The issue was not the procedure of requesting a waiver–the issue was whether a waiver was in the public interest. Put another way: whether the creation of a private monopoly to exploit an invention was in the public interest. For universities, who received the bulk of their money from the HEW and NSF, this was a non-issue because of the IPA program and the relative ease otherwise of requesting a waiver. For agencies such as DOE (where the Purdue Research Foundation finally kicked up a storm that got Senator Bayh’s attention), it was indeed case-by-case. And that could be slow, and uncertain, and denied. “Waiver” then hides the issue. What is the public interest in a private contractor creating a private monopoly on research conducted in the public interest? What is the point of that, in particular, when the contractor is itself a university–a public service organization–and without any particular experience or resources or mission to develop any invention for public use? Why should the monopoly become so important?
The government only allowed nonexclusive licenses to its patents.
True, when it bothered with licenses at all. The government took patents without a profit motive. Patent practice without a profit motive does not look at all like monopoly speculation or exploitation. Non-exclusive licenses can hurt small players competing against large players. Non-exclusive licenses can limit who will invest huge amounts in developing an invention and training a new market for it. But non-exclusive licenses can also stimulate competition, can provide for multiple sources of a product, can build up expertise across an industry rather than concentrate that expertise in a single company subject to the whim of its profit objectives, and non-exclusive can permit widespread local use without demanding monopoly commercial development into a product form. There are trade-offs. Patent brokers preferred monopolies. But they haven’t made a case that such monopolies are uniformly in the public interest. In fact, the opposite appears to be true.
Lack of a solid property interest in these discoveries meant uncertainty. Private-sector developers and investors couldn’t afford that kind of risk. Thus, commercialization of these discoveries rarely materialized.
Again, simply not true. In many industries, people jumped at the chance to develop their own versions of invented things. Fertilizers, for instance. Tomato pickers. Stating as a general case what reduces to a handful of issues, most notably around the pharmaceutical industry’s participation in compound screening for medicinal chemistry, misrepresents the situation. For medicinal chemistry, pharma organized a boycott of HEW-supported compounds. For pharma, there was not so much uncertainty as certainty that HEW would grant them only a non-exclusive license for the compound. Any monopoly would have to be in the formulation. Poo on that, went pharma. Even then, according to the Harbridge report, the issue for pharma wasn’t the monopoly so much as the potential “contamination” by federally supported work of work that the companies wanted to keep separate. HEW wanted in many cases background rights as well, so they could make the practice of the invention widely available. What’s the point of gaining a title (or even a license) to just a part of an invention, while the primary rights that control the invention remain a private monopoly? That’s a tough issue–and one that doesn’t get resolved in the general case but for an arbitrary policy. In the particular case, it’s a matter of negotiation, but even then, there’s no guarantee that two parties will find a way to agree. That’s uncertainty, perhaps, but it’s an uncertainty created by one’s negotiating goals, not by any policy requirement.
As National Academy of Sciences researchers put it, before Bayh-Dole, “[T]he incentives to pursue further development and commercialization were severely attenuated and the capacity to do so severely limited. Government agencies, in particular, had no incentive and negligible capacity.”
Well, there’s no citation here. But I think this refers to an NAS report by a committee–important people in their way, but hardly researchers. They repeated the same usual narrative about Bayh-Dole, so we have some confirmation bias. But look at the nature of the rhetoric used by the NAS authors. They pin the public interest on “incentives”–as if there must be an enhanced profit motive to do something beneficial for the public. By “incentive” the NAS authors must mean “monopoly with a profit motive.” That is, the government patent policy limited the opportunities for private monopolies. But companies all the time operate without monopolies and do just fine, investing and developing and commercializing. Look at the internet. Open standards. Lots of investment and development. Competition. Folks get over it. The internet is not unique this way.
As for government agencies, the NAS report is close. They did not have “an incentive”–namely they had no need for either a monopoly or a profit motive. The argument Edwards is making here in a backhanded way is that for all discoveries the only means of public benefit comes through profit-motivated monopolies. The NAS authors point out the government did not have such motives, nor incentives nor capacity. So?
Even the Kennedy patent policy recognized that exclusive rights were sometimes needed to “call forth risk capital” to complete the development of an invention to the “point of practical application.” The problem was not uniformity in the policy. The problem was the interpretation of the policy across the various missions of the various federal agencies. Only the patent brokers apparently wanted uniformity–arbitrariness. Even the pharma companies would have been satisfied with a longer term of exclusivity, along with HEW compliance with section 1(b) of the Kennedy patent policy. That’s where the case ought to have been made, and not government-wide.
In other words, the problem Congress solved with Bayh-Dole was the government telling the private sector: Yes, you may use this discovery for commercial purposes, but you don’t own it, your competitors may also get a license to it and years from now the feds may unhand your meager hold on whatever you do with this invention with your own investments.
Not at all. Patent attorneys found an entry point to get Congress to tell patent brokers they could be the bottleneck for all inventions made with federal support. Send in the clowns. The federal government across various agencies, and especially DoD, HEW, and NSF, all had policies and programs that allowed contractors to own their inventions. For research in which the government was the primary funder and user, there was no need for any contractor to have a monopoly–all that would be excluded would be the government itself or others that the government wished to contract with for additional or competitive services. In the context of competitive procurement, there was absolutely no point in allowing contractors to create private monopolies when the government was paying the costs of development, and for its own use.
And the “unhand” bit is pure drama. The government’s intervention was never to exclude use. Thus, all that happens is that a monopoly gets weakened by competition. That, too, is a balance. But where’s the reasoning to support the idea that private monopolies around publicly supported research in the public interest are themselves uniformly in the public interest? Congress solved a problem all right–but not a problem that reflected the public interest in the results of research.
Today, having certainty and exclusivity over the IP facilitates moving such discoveries to commercial product — of vital importance in R&D-oriented, capital-intensive, technologically sophisticated sectors such as life sciences.
All this, just to get to a euphemism for big pharma. In pharma, perhaps, there’s a rationale for monopolies–and that appears to be the motivation behind this flurry of papers defending Bayh-Dole in such grotesque terms. But it is here in pharma that the debate is happening. The rest of the discussion is pure distraction. “Don’t change Bayh-Dole because all that other stuff would be damaged.” But the place where the damage is taking place, where the private monopolies are not working as desired, is precisely in pharma. If Bayh-Dole is modified, it can be modified to make pharma an “exceptional circumstance” under 35 USC 202(a)(ii):
Each nonprofit organization or small business firm may, within a reasonable time after disclosure as required by paragraph (c)(1) of this section, elect to retain title to any subject invention: Provided, however, That a funding agreement may provide otherwise…
(ii) in exceptional circumstances when it is determined by the agency that restriction or elimination of the right to retain title to any subject invention will better promote the policy and objectives of this chapter
So just add an express exceptional circumstance involving the disposition of title in compounds having medicinal uses. There, restrict the exploitation of patents by universities, nonprofits, and their exclusive licensees. The basis to do so is already in Bayh-Dole in the statement of Congressional policy. That’s where the pricing issue is addressed–through “free competition and enterprise.” There can be no free competition and enterprise if a patent right is held as a monopoly for its entire term. It is a breach of federal patent law specific to inventions made with federal support to exploit a patent as a monopoly for its entire term. All that’s needed is express recognition of this matter of law with regard to pharmaceuticals. That’s a civic decision, not a business one.
Pharmaceutical companies will complain–because their shareholders will complain, because the betting pool will be disrupted with uncertainties and with restrictions. But we are talking about allowing the speculators to control the game. It is as if sports betting has become so valuable that the games themselves exist for the betting, and if not for the betting, the games would end. At that point, the games really ought to end. They become cock fights, not entertainment. As for medicinal chemistry, it never was pitched as entertainment, and it does not exist to serve speculators. If no private investment can ever be made with a social purpose, then perhaps the pharma game needs to change. There are other ways to develop compounds to address public health. We don’t have to rely on speculators betting on the monopoly fortunes of huge companies as our only means to do so.
Bayh-Dole uses America’s property rights-based patent system “to promote the utilization of inventions arising from federally-supported research or development,” the statute reads.
And in the same clause Bayh-Dole also reads:
to use the patent system …. to ensure that inventions made by nonprofit organizations and small business firms are used in a manner to promote free competition and enterprise without unduly encumbering future research and discovery
Bayh-Dole is part of federal patent law. The “property rights” of the patent system are based in what the patent law allows. Here’s 35 USC 261:
Subject to the provisions of this title, patents shall have the attributes of personal property.
Patents are not “personal property.” They have the “attributes” of personal property. Further, the scope of that property right is “subject to the provisions” of federal patent law. The provisions of patent law that pertain to the property represented by patents on “subject inventions”–a defined term in patent law for inventions made with federal support–are expressly those of 35 USC 200. No contractor should be allowed to assign an invention to a pharmaceutical company. Indeed, Bayh-Dole prohibits universities from doing so without federal agency approval (35 USC 202(c)(7):
In the case of a nonprofit organization, (A) a prohibition upon the assignment of rights to a subject invention in the United States without the approval of the Federal agency, except where such assignment is made to an organization which has as one of its primary functions the management of inventions (provided that such assignee shall be subject to the same provisions as the contractor)
Standard university exclusive licenses frequently meet the definition of assignment recognized by the courts. Mislabeling does not matter. If the university does not reserve significant rights–right to sublicense, right to sue for infringement or settle infringement, right to practice in some fields, or limitation on term–then the transaction is an assignment, not a license. We might then note that what pharma is arguing for is that universities be permitted to continue to violate federal law and feed them monopoly access to federally supported inventions that might relieve human suffering. For that pharma should be arguing that Bayh-Dole must be changed to permit assignment of patent rights, so long as the assignment is labeled “exclusive license.” Otherwise, for any given assignment of an invention to a pharmaceutical company, that company is exposed to the compliance requirements of the nonprofit–including using all income after expenses of managing the patents for scientific research or education–none of it can go to shareholders as a dividend, none of it can go to development or other business fun. Want to hit on pharma? Then look at all those exclusive licenses. Do they meet the courts’ definition of assignment? If so, then go after the profits. Pharma will have to backpedal and give up significant rights, and from there opens the prospect for “free enterprise and competition” despite the concerns of shareholders that such a thing will disrupt their fantasy drug betting game.
Bayh-Dole has also contributed to expansion of the tax base and faster progress in America’s technological advancement. All this hinges on secure IP rights, regardless of a discovery’s connection to federal research funds at the earliest stages, well before the long, hard, uncertain, expensive commercialization phase began.
Bayh-Dole has been an economic disaster, other than as it has been an end-run around HEW (now DHHS) efforts to limit the effects of private monopolies. There, it has meant happy times for pharma companies. They get the benefit of massive federal funding, they get the benefit of massive venture capital investment (or speculation) to take early risks (and run up salaries in such competitive speculation), and then they get the benefit of monopoly development and sales. For some discoveries, there is indeed an expensive path to commercial product. No question. And for some few discoveries, a commercial product will be the only way any benefit is realized. Granted that, too. But it’s a long, hard way from that handful of things to a general case for all discoveries made with federal funds.
And it’s a stupid case, a case that can’t be won. Most research discoveries can be used and developed for use without huge private investments, and those that require private investment often can be developed by collaborative effort–by sharing the costs and coordinating on the approach, creating industry roadmaps and standards and shared tools and platforms. If there’s billions of dollars of profits waiting at the end of the effort, then there is still billions of dollars there if there isn’t monopoly pricing. Generic drug companies make profits, too. And if there’s still billions of dollars, those dollars are still plenty if spread around among competing companies–companies competing on quality, on manufacturing efficiency, on speed to market, on availability, on price.
This is not what pharma wants to contemplate, of course, though at one point some years ago now a senior VP of a pharma firm confessed to me privately that pharma needs to find a way to develop open source techniques to spread the early risk and expense of developing compounds for medical use.
The rest of Edwards’s piece is the usual arguments against “march-in”–that is, against public accountability and oversight, against any enforcement of the law or change in the law to allow it to be enforced. Read that at your leisure. The bottom line: the world as we know it will come to an end if march-in procedures are disturbed. Keep in mind Howard Bremer later bragged about how he and other university folks had to step in and make the march-in procedures so difficult that they would never operate. As for the lead-in to the defense of the march-in procedures, it’s bunk all the way down.
March-in by design is not a way to get at monopoly pricing. In that, the pharma apologists have a point. Competition is the way to get at monopoly pricing. That means attacking assignments mislabeled as exclusive licenses. That means enforcing the requirement that the patent system be used for subject inventions to promote free competition and enterprise. That means looking carefully at where Bayh-Dole could be modified to end the speculative patent bubble that drives up health care costs–and we may as well start with the exceptional circumstances clause, and stick a pin in it.