The federal government has five methods to deal with the exploitation of federally supported inventions. Let’s list them and make a few points, ending with a discussion on the value of the government’s license to “practice and have practiced” subject inventions.
- Bayh-Dole’s policy and objective limiting the scope of patent property rights
- The standard patent rights clause
- Exceptional circumstances modifying the patent rights clause
- March-in procedures for nonuse, public needs, regulatory needs, U.S. mfg
- Government “practice and have practiced” license
Let’s work through these.
Bayh-Dole’s policy
Bayh-Dole starts with a statement of policy and objectives at 35 USC 200. This is more than simply a statement of legislative purpose, as evidenced by the presence of the term “policy.” The policy set forth replaces the executive branch policy that governed federal agency contracting requirements for nonprofits and small businesses. But Bayh-Dole is part of federal patent law, and the policy also changes the scope of the patent property right for inventions made with federal support. Bayh-Dole defines a new category of patentable invention–the “subject” invention. Patent law states that patents have the attributes of personal property, “subject to the provision of this title” (35 USC 261). And for subject inventions, the provisions of the title include Bayh-Dole, which is part of patent law.
The property rights in patents on subject inventions, then, are shaped not just by ordinary patent law, but also by Bayh-Dole as part of patent law, independently of the parts of Bayh-Dole that establish executive branch contracting practices. If 35 USC 200 were merely a statement of objectives, we’d be done. Prefatory surplusage. But it is also a statement of policy, and when that policy states that the patent system is to be used to “promote the utilization of inventions arising in federally funded research”–subject inventions–then there’s a limit to how the patent system can be used for subject inventions. The expression of that limit in specific bits of the standard patent rights clause and other restrictions on federal agencies is not the controlling element. The policy statements stand regardless of whether there’s specific guidance for the patent rights clause, which in its way is one mechanism by which the policy requirements of 35 USC 200 can be met. The question raised by 35 USC 200 is, Can the patent system then be used to prevent the utilization of subject inventions?
Granted, no one bothers with 35 USC 200 as a serious part of federal patent law, so the government is not about to step in and challenge universities trolling industry with patent rights they have sat on for sometimes a decade or more without developing anything, or licensing to a soon-moribund company whose investors then play the trolls to “monetize” the patents to recover their investment. There’s an argument here, but without any money to pay attorneys to make it. So tear water tea, and not much more.
The standard patent rights clause
This is the clause that Bayh-Dole authorizes the executive branch (now, NIST, delegated from the Department of Commerce) to construct for use by federal agencies in their funding agreements with universities, other nonprofits, and small businesses. Actually, there are four standard patent rights clauses–
- 37 CFR 401.14(a) [the one most people refer to, directed at nonprofits];
- 37 CFR 401.14(b) [an extension of (a) dealing with certain weapons systems;
- 37 CFR 401.14(a) less (k) [for small businesses]; and
- 37 CFR 401.9 [for inventors, to be treated as small businesses but with fewer restrictions].
[NIST has now conflated (a) and (b) into a single clause with a conditional for the nuclear weapons and propulsion systems, so the clause is now just 401.14]. (The standard patent rights clause transfers obligations from federal agencies, who are required to use the clause unless they can show reasons not to, to universities and other nonprofits, who accept the requirements of the clause as a condition of accepting the federal funds, as part of the overall funding agreement. The standard patent rights clause does more than arbitrate the conditions under which the federal government gives up its right (as a matter of contracting) to require delivery of ownership of patentable inventions made within the scope of the funding agreement. The clause also stipulates how the patent property must be managed–a patent application must be filed within a specific period, and patent applications must be prosecuted and patents maintained; licensing of patents must (for nonprofits) give preference to small businesses; exclusive licenses to use or sell in the U.S. must include the requirement that products involved in such use or sale be “substantially manufactured” in the U.S. Income from subject inventions (that is, from patents on subject inventions) is also constrained–the only expenses that may be recovered from that income are ones “incidental” to the management of subject inventions. Any remaining money must be used for “scientific research or education” (and it’s not clear whether “scientific” modifies both “research” and “education”).
It’s clear that patents on subject inventions–POSIs–are not ordinary patents. The patent property right in POSIs is not an ordinary patent property right. The text that controls the POSI property right is the statement of policy at 35 USC 200. Specific instances of that control are then expressed in the standard patent rights clause.
If the government wanted to address the behaviors of POSI owners, it could start by enforcing the government’s side of the standard patent rights clause. Require full reporting of use. Enforce the prohibition on nonprofit assignment. Look out after small business licensing. Require U.S. manufacture and reject feeble claims it’s not feasible. Audit the use of nonprofit royalties–not just at universities but at the nonprofit patent brokers that front for them.
The government could also add more standard patent rights clauses for specific situations. A patent rights clause specific to biomedical inventions would be one to consider. There, specific restrictions on the price of any resulting commercial product might be in order. There are various ways to pressure price. One is with non-exclusive licensing, so there’s competition. Another is to limit the term of exclusive licenses. A third is to split licensing into make/use (non-exclusive) and sell (exclusive, if necessary), so that sold products compete with DIY use. But one can also pressure price by requiring that in exclusive licenses, the proceeds from sales are subject to the same public interest limitations that nonprofits accept–other than recovering costs incidental to the development of subject inventions for public use, all profits must be used for “scientific research or scientific education.” Not for, say, CEO bonuses or dividends to shareholders or fancy new buildings or even development of drugs that aren’t federally sourced. That would mean that high prices based on monopoly positions generate more money expressly dedicated to supporting the subject invention to commercial development pathway. If pharma’s argument is that monopoly prices support more research, then a new standard patent rights clause says, “OK, make it so.”
Exceptional Circumstances
Bayh-Dole requires the use of a standard patent rights clause in each federal funding agreement. But Bayh-Dole also permits federal agencies to modify that clause, depending on circumstances. Here’s 35 USC 202(a):
(a) 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 . . . .
The “exceptional circumstances” are not called out as “rare” circumstances–just circumstances that are “exceptions” to whatever are the basic assumptions of the law. Here, the “exception” is simply (if that’s a word that can be used without humor in relation to Bayh-Dole) that a federal agency determines that a different approach (restricting, eliminating) a contractor’s right to retain title “will better promote the policy and objectives” of the law.
Bayh-Dole goes on to set out the conditions under which an agency can change the standard patent rights clause. An agency has to determine that condition (ii) applies, has to file within 30 days of the funding agreement a determination with the Secretary of Commerce, and the determination must “include an analysis justifying the determination.” The Secretary of Commerce then may object to the determination and “recommend corrective actions.” The Office of Federal Procurement Policy can object to agency determinations and ban them for various situations. And contractors can object to determinations, triggering a stormfest of paperwork and procedures and reviews guaranteed to delay if not derail determinations.
It is clear from this right of a federal agency to modify the standard patent rights clause when doing so will “better promote the policy and objectives” of 35 USC 200 that the standard patent rights clause is not the sole expression of that policy and those objectives. Removing a requirement from the standard patent rights clause (such as restrictions on the term of an exclusive license granted by nonprofits) does not mean that such restrictions are wrong. It is just that an agency, to implement such a restriction, must make a determination that the restriction better promotes the policy and objectives of the law–that is, that it is a circumstance that stands as an “exception” to the provisions of the standard patent rights clause. For a given class of subject inventions, such as therapeutic compound inventions or software method inventions or disease assay inventions, the “exceptional circumstance” rests with the nature of the invention and the means by which that invention comes to be used.
Again, the “exceptional circumstance” is not some rare circumstance–it is just one that lies outside the standard drafting of the patent rights clause dictated by Bayh-Dole. Think of it–before the 1984 amendment to Bayh-Dole, permitting a nonprofit licensor of a POSI to grant an exclusive license for the full term of the patent was an “exceptional circumstance.” The 1984 amendment made that exceptional circumstance the default, shifting the burden to the federal agencies to justify a deviation that would restrict the term of a nonprofit’s license.
There could be more standard patent rights clauses, if NIST wanted to create them, that incorporated circumstances that were “exceptional” to the provisions required by Bayh-Dole–it’s just a matter of the government deciding when there might be such standard “exceptional” circumstances. Limiting nonprofits election of title to subject inventions allows for all sorts of conditions throughout the apparatus of Bayh-Dole–from preventing nonprofits from acquiring title to requiring that they acquire title only by a voluntary choice of the inventor to limiting what they can do with that title as a condition of their obtaining title. There’s a whole world of policy opportunity for the federal government to improve the required or allowed practices under a patent rights clause. It’s just a matter of the government having the will–and smarts–to do it.
March-in procedures
March-in procedures have received the majority of the attention in Bayh-Dole debates over what should happen with subject inventions. The idea of march-in goes back at least to the Kennedy patent policy of 1963. There, it was not called “march-in” but rather operated as a restraint on the “principal rights” in a patent held by a contractor.
The Kennedy patent policy posits two markets for inventions made with federal support. The first market is the government market, and the Kennedy policy requires the federal government to take ownership of inventions made primarily for the government market–where the government is developing something as a product, where government has the primary knowledge or ability to use the invention, where the only or primary market is bidding to do more work for the government, and the like.
The second market is a private market, where an invention might be used, developed, and controlled for private purposes, outside the government market. In later terms, for military-related inventions, the distinction came to be called “dual use.” Develop something that is used in a government market for use in a private market. Tang. Space food sticks. Fission. The Kennedy patent policy defined what “principal rights” could be left with a contractor to develop use and product for a private market.
The march-in procedures in the Kennedy patent policy began with limitations on licensing in a private market. A contractor had three years from the issue date of a patent on a federally funded invention to establish that the invention was being worked, with benefits reasonably available to the public–or has made the invention available for license royalty-free or on “terms that are reasonable under the circumstances.” That much was the hard-wired march-in requirement.
If various conditions were met, the government could then intervene in the private market. If the contractor failed to establish use of the invention, then the government could require non-exclusive licensing that might do so. In essence, the government could force a contractor who both failed to establish use of an invention and to make the invention available non-exclusively to make the invention available non-exclusively. If the contractor took the initiative to license non-exclusively, then the contractor got to set the “reasonable” terms; otherwise, if the government intervened, the government got to set those terms. If there was money involved, that money still came to the contractor. It was still a private market, but one that did not have available the full property rights of a patent holder, including the right to exclude all uses of an invention for the life of a patent, or to charge so much for a license that no one was willing to pay the asking price.
The government could also intervene in a private market when doing so was necessary for some government purpose:
(i) to the extent that the invention is required for public use by governmental regulations, or (ii) as may be necessary to fulfill health or safety needs, or (iii) for other public purposes stipulated in the contract.
The first of these conditions does not depend on anything that the contractor is doing–not whether the contractor is working the invention or trying to. If the invention is required by regulation, then the government can make the contractor license non-exclusively. Again, any royalty would be returned to the contractor–and if regulations require public use, then there could be a huge windfall to the contractor in all those licenses granted–compared to trying to develop an invention in isolation and having only three years of exclusivity before having to justify a longer term to the government. Non-exclusive does not mean “without benefit to the licensor.” Cohen-Boyer and Axel and Hall were all non-exclusively licensed, for millions. Patent “trolls” license non-exclusively, for millions. Why license to just one player for what is usually a minority market share, turning everyone else in industry hostile, when one can license to the entire market?
The second condition–“to fulfill health or safety needs”–means that fulfillment is necessary sooner than the three years that the contractor has for exclusive rights. If the fulfillment is necessary after three years, then it’s just a matter of more non-exclusive licenses. If we are talking about after the three years of exclusivity, the burden is on the contractor to show that whatever the contractor is doing or will do will fulfill health and safety needs–benefits reasonably available to the public– better than letting others also work the invention. Much depends, then, on what is “reasonably available to the public.”
The third condition is based on what’s in the contract. This is especially important for proposals coming from faculty at universities, where public mission may be an express component of the proposal. If it’s in the proposal, then the government can intervene when the contractor’s actions don’t square with the public commitments made in the funding contract.
Bayh-Dole takes up these conditions (at 35 USC 203), largely via the Institutional Patent Agreement program that had been terminated in 1978 (due apparently to university abuse of the program and the program’s apparent disregard for top-level agency patent policy–that is, not enforcing the terms of the master agreement). However, Bayh-Dole makes some changes–subtle in wording, significant in import.
Similar to the procedure under exceptional circumstances, to “march-in” a federal agency must make a “determination” that march-in meets one of four conditions. These conditions include nonuse and failure to comply with the U.S. manufacturing requirement for exclusive licenses to “use or sell.” The other two conditions, however, are the ones that have received pressure–the action is “necessary” to
(2) alleviate health or safety needs which are not reasonably satisfied by the contractor, assignee, or their licensees;
(3) meet the requirements for public use specified by Federal regulations and such requirements are not reasonably satisfied by the contractor, assignee, or licensees
Bayh-Dole shifts the requirements for an agency’s determination to march-in to the implementing regulations, along with a process for contractor appeal of both march-in and determinations of exceptional circumstances. Those implementing regulations, in turn, were drafted with the intent (as described later by Howard Bremer) to prevent them from ever operating. And so far, the drafting has been effective. The federal government has never used march-in–not for nonuse (which is everywhere), not for health or safety needs, not for federal regulations.
In the case of federal regulations, there does not appear to be any record of how many subject inventions have actually been the subject of federal regulations for public use–perhaps there have been none. If so, the absence might indicate that nearly all federally supported research inventions have little or no impact on public policy; hence, no need for march-in. An alternative explanation is that there have been such needs, but they have been met by contractors. A third explanation is that there have been such needs, but federal agencies have chosen not to intervene–meaning that the private monopoly has greater value in public policy than meeting regulatory requirements or public health or safety needs. Again, these explanations are purely templates because there is no data to indicate which of these, or some other explanation, accounts for the lack of any agency march-in on these conditions.
Noticeably absent from Bayh-Dole’s march-in conditions is the Kennedy patent policy condition that the government can intervene for any purpose stipulated by a funding agreement. Thus, Bayh-Dole in its way repudiates any commitments–even when made express–in funding agreements with regard to public purposes that might involve inventions, such as contribution to a standard, broad access, or benefits directed at specific groups (such as, say, individuals suffering from a given disease or injury). The repudiation is not merely of federal agency purposes, but also of commitments made by the proposers of work–such as university faculty. Under Bayh-Dole, institutional management of POSIs takes precedence over the public commitments of those proposing and doing the work that leads to invention, in part because federal agencies have no authority to intervene to enforce those commitments against institutional actions that disregard them.
The government could intervene–perhaps with regard to 99% of all POSIs–but it does not. What is not objected to by public policy is public policy.
Government license
Entirely separate from march-in in Bayh-Dole is the government’s right to a license “to practice and have practiced” any subject invention in which a contractor acquires title and elects to retain that title. Here’s 35 USC 202(c)(4):
With respect to any invention in which the contractor elects rights, the Federal agency shall have a nonexclusive, nontransferrable, irrevocable, paid-up license to practice or have practiced for or on behalf of the United States any subject invention throughout the world
In executive branch patent policy, from the Kennedy policy of 1963 on, the government license includes “to make, use, and sell” the invention. (The Kennedy policy uses “practice and have practiced,” defined to mean “make, use, and sell”; the Nixon modifications to the Kennedy policy uses “make, use, and sell . . . by or on behalf of.”) Bayh-Dole continues and replaces portions of executive branch policy, but does not disturb the meaning of “to practice and have practiced.”
Similarly, in the Kennedy patent policy, and again in the IPA program, the “Government of the United States” was defined to include state and municipal governments as well as the federal government. In Bayh-Dole, a federal agency is to be granted a license that involves the “United States”–the scope is the “United States.” The basic meaning of “United States” would appear to be broader than the federal government. In everyday usage, “United States” means the states of the U.S., individually and collectively. “I come from the United States” can mean “from the federal government” or “from Oregon” or from “Eugene.” Does “United States” in Bayh-Dole include state governments? That would be a reasonable reading. Does “United States” include municipal governments? That was the expressly defined meaning of the “Government of the United States” under the Kennedy and Nixon executive branch policies (with slight variations in how they got to the definition). And “United States” would appear to signal a potentially broader scope than “Government of the United States.”
Finally, considered “for or on behalf of.” This construction is in both the Kennedy and Nixon executive branch policies, and its use continues in Bayh-Dole. “For” and “on behalf of” are two distinct statements of scope, not simply one statement of scope repeated immediately. “On behalf of” means “in the interests of” or “to aid” or “as a proxy for.” The one providing the aid is not restricted to the one being aided. “For” here means generally “suiting the purposes of,” or “done at the initiative of.” Taken together, the two expressions are rather clunky for the scope of a license. Why not just write something direct, such as:
the Federal agency shall have a nonexclusive, nontransferrable, irrevocable, paid-up license to practice any subject invention throughout the world
If this construction were used, the practice of the invention would be limited to the federal agency that provided the funding, and would not extend to the rest of the federal government. The federal agency could still sell product incorporating or based on the invention, but the agency would have to do the selling and could not commission others to do so. Now add in “have practiced”:
the Federal agency shall have a nonexclusive, nontransferrable, irrevocable, paid-up license to practice and have practiced any subject invention throughout the world
Now the scope expands–the federal agency can authorize others to practice (make, use, sell) the invention. There is no express limitation here on the reasons for the “have practicing” but patent practice, for “make,” anyway, constrains the making to others acting under the direction of the licensee, for the licensee’s purposes. Here’s an explanation from a guide to patent licensing for corporate counsel:
The “have made” right, which refers to a licensee’s right to engage a third party to manufacture a licensed product on the licensee’s behalf, is not one of the exclusionary rights enumerated in the patent statute. However, the Federal Circuit recently confirmed that, unless specifically stated otherwise in the grant clause, the right to “make, use and sell” a licensed product also includes the implied right to have those licensed products made by a third party.
But “have practiced” is broader than “have made” and is stated by Bayh-Dole expressly–it’s not something implied at all. Thus, “have used” (to produce some benefit for the “United States”) and “have sold” (as authorized by the “United States”) are wrapped into “have practiced.”
Imagine an invention that allows rapid repair of truck tires; the government could want to use the invention to repair the tires of postal service trucks, and so might do those repairs itself, or might authorize one or more companies to provide roadside service to do those repairs–“have practiced.” But can those authorized companies provide the same roadside services for non-federal trucks? Ah, that’s the separation between the federal government market and the non-federal market (and if Bayh-Dole preserves the idea that “United States” includes state and municipal governments, then these governments are, for the purposes of Bayh-Dole’s license, also part of the federal government market). An objective interpretation of adding “and have practiced” is that the license is unbounded–one can authorize anyone to “make” or “use” or “sell,” but unlike a “sublicense” that grants general rights (such as to exploit the non-federal market), the “have practiced” rights are restricted to the federal government market (which as we have seen is greater than just the activities and authorized purposes of the federal government itself).
Now consider the effect of adding “for or on behalf of” to the scope of the license.
the Federal agency shall have a nonexclusive, nontransferrable, irrevocable, paid-up license to practice or have practiced for or on behalf of the United States any subject invention throughout the world
Do you see the clunkiness now? If the construction was “for or on behalf of the Federal agency” then it would make better sense. “For” would indicate “what the agency may do acting for itself” and “on behalf of” would indicate “what the agency may authorize others to do to serve the agency’s purpose in the federal government market.” But Bayh-Dole substitutes “the United States” for “Federal agency.” The federal agency license includes “to have practiced” “on behalf of” “the United States.” That is, a proxy to the federal agency may act as a proxy to any other entity included within “the United States”–clearly more than just the federal agency, more than any other federal agency (there’s a whole list of the entities that can be on the federal side of a Bayh-Dole funding agreement–see 35 USC 201(a), which points to 5 USC 102 and 105), and including state governments and municipal governments. While the state governments and municipal governments, at least, are not within the scope of the license grant, the federal agency receiving the license appears (nothing is certain until there’s been three years or more of litigation) can authorize the practice of a subject invention by a proxy (including a company) to serve the purposes or needs of any of these entities.
The federal agency acquires a license that permits it to authorize any practice of a subject invention by anyone providing services to any entity included in “the United States.” The scope does not extend expressly to the non-federal market, but the non-federal market is potentially much smaller given the breadth of the license to “practice and have practiced” “for or on behalf of” “the United States.” Double proxies are wild.
Now how does all this work? First, the government license is not a matter of march-in. No determinations are required for the government to exercise its license. The grant of the license is outright, as a condition for a contractor to elect to retain title to a subject invention. Allowing the election of title is in its way the consideration for the grant of the license. March-in has to do with the status of subject inventions in the non-federal market. There, the federal government may identify nonuse, or unmet need, or unmet regulatory mandate, or failure to obtain U.S. manufacture, and “march-in” to require the patent owner to grant licenses to permit use or meet the need or regulatory mandate or U.S. manufacture.
March-in under Bayh-Dole has to do with federal intervention in the private side exploitation of a POSI. To the extent that the federal government also purchases (or pays for) goods and services made available in the private market, the government may benefit from the initiative and efficiencies of that market. Just buy from those that are making and using and selling product and services based on the subject invention. But if that private market is monopolized and expensive, or inefficient, or not producing the needed goods and services, then the federal government can use its own license to practice and have practiced to meet its own needs, using whatever resources drawn from either the federal market or the non-federal market–for or on behalf of–to do so. The federal government does not have concern itself with the vagaries of the non-federal market. If that market is monopolized, but the monopoly does not rise to anti-trust concerns, then it’s just a messed up market and the federal agency may readily decide not to intervene via march-in. And there are plenty of defenses installed in Bayh-Dole and the implementing regulations to make march-in almost but not entirely unworkable.
By contrast, there are no defenses installed at all in the government license to practice and have practiced. The government has the rights of this license, the license is broad, and the license is irrevocable and free. And the license is world-wide.
Consider how this license might operate for, say, a prescription drug based on a subject invention (there are all sorts of other issues, but let’s stick with this one thread first). In the private, non-federal market, a university license-assigns the patent rights to a company that then develops and sells a product, a prescription drug. Because the company has a monopoly position, it can charge a monopoly price. Thus, Xtandi, based on a subject invention and license-assigned by UCLA to Medivation, now a part of Pfizer, can sell in the U.S. for $88 a pill and a generic drug company based in Canada can sell it at a profit for $5. The private market is expensive, in part as a result of the monopoly created by UCLA’s license-assignment, which did not break up the monopoly to promote competition. But the NIH declines to march-in and change the conditions of the private market (or even to enforce the patent rights clause restrictions on such assignments).
But the federal government, through such programs as Medicare and Medicaid, also pays for Xtandi for individuals enrolled in these federal programs. Under its world-wide government license to practice and have practiced (including to have sold) for or on behalf of the United States, NIH can authorize the Canadian company to make and sell Xtandi pills for individuals enrolled in federal programs. If NIH’s own leadership balks at doing so, all it takes is an executive order from the President to unbalk it. No march-in. No tangle of regulations. No act of Congress required. The Canadian company could not make and sell Xtandi for the private market, of course. But the federal market for Xtandi is large enough as it is that the federal market would be a profitable market. The government would realize a 10x drop in costs associated with Xtandi–we are talking $10K per patient per year. The savings run into the millions. Repeat with just a few such drugs and the savings run into the billions.
Now consider what happens. There are two markets, then, in competition: the monopoly private, non-federal market and the generic (open) federal market. In the private market, with private insurers, Xtandi is $88 a pill and patients and insurers pay a huge price–a lucrative price, a happy-is-the-day price, if one is at UCLA or Pfizer or an investor in Medivation, or in love with the idea of monopolies driving society by enriching those that deserve to be rich–the folks that lift inventions from inventors and exploit patents at some risk to create monopolies. If this doesn’t happen to be your vision of desired societal drivers (at least not in the matter of health care), then the federal market looks a great deal more like the market in which to work.
If the federal government opened up a program that covered only prescription drugs based on POSIs–on patents on subject inventions–then anyone could choose to move to the government program for that portion of their health care needs. Xtandi would go from $88 a pill in the private, monopolized but happily legal market to $5 in the federal market. That may cause the private market to diminish, but that’s an effect of competition, and an effect of giving people a choice regarding their health care.
There’s really no way for anyone to “negotiate” monopoly drug prices. There’s no leverage position. The alternative is that a patient does not get the medication. There’s no compelling reason for the monopolist to lower prices other than that no one takes the price–and not taking the price is the same as refusing the medication, which may be life saving or at least life extending or pain reducing or injury mitigating. There is no effective BATNA–best alternative to a negotiated agreement–and hence there is no meaningful basis to “negotiate” monopoly drug prices. Yes, one can legislate drug price controls and one can legislate things that cannot be protected by patent, and one can legislate tighter antitrust laws. But legislation tends to build up apparatus to fix apparatus to fix apparatus until it seems that things are apparatus all the way down.
Meanwhile, the federal license under Bayh-Dole is outright, broad, and available with at most the need for an executive order to require federal agencies to use their license to practice and have practiced whenever a POSI is licensed to maintain rather than to break up a monopoly. There might be a three-step test:
(1) has a POSI been licensed-assigned?
(2) has a product been developed within eight years?
(3) is the product offered to the public on reasonable (non-monopolistic) terms?
If the answer to (1) is yes, then the owner and licensee-assignee of the POSI is put on notice that there’s a clock ticking and that the federal government anticipates the need to look after its own federal marketplace. If the answer to (2) is no, then the federal agency can choose to march-in to break the monopoly in the POSI or it can commission companies to develop product exclusively for the federal market. If the answer to (3) is yes, the the government purchases from the private market for federal market needs; otherwise, rather than fuss about negotiating lower prices or wasting time trying to get legislation through the pharma lobby in Congress, the federal government exercises its license to practice and have practiced and authorizes private organizations (and even universities and research institutes, to the extent that they have the capability) to make and sell the drug, but only to those within the scope of “the United States”–federal agencies, state governments, municipal governments.
Nothing in patent law requires a patent owner to pursue monopolies, even though a patent right grants the patent owner a right for limited times to exclude all others. When to exclude and when to include is a matter of judgment that depends on the interests of the patent owner. A patent owner might be self-interested in a greedy sort of way or self-interested in a visionary sort of way, or despite Ayn Rand’s fuss about it, altruistic in a self-interested sort of way. For universities and other nonprofits with express missions of public service in the altruistic sort of way, it comes as a form of cognitive dissonance (if not a creeping sense of corruption) to find university officials among the biggest, unreflective advocates of pursuing monopolies and championing the importance of license income over the effect of monopoly pricing on society, especially in areas such as health care.
The effect of Bayh-Dole has been to move the ownership of inventions from the control of faculty investigators and inventors–who historically have been publicly spirited (often publishing to dedicate an invention to the public domain) and highly selective of when to rely on patenting (where a patent is a needed tool to engage industry in development) and even when patenting, highly selective in the use of monopoly licensing rather than non-exclusive license–to that of institutional owners captivated by non-selective monopoly-preserving behaviors. That effect alone is cause for concern. Perhaps the survey that should be taken is asking inventors of subject inventions whether they would have preferred to publish or if patenting to non-exclusive license or if licensing exclusively for commercial developing reserving the right to license for making and using for broad research and public use outside of the need for a commercial version of the invention. Maybe we would find that 3x or 10x the number of inventors would have favored open licensing strategies to the compulsory, non-selective, monopoly-preserving strategies in use by university administrators.
Here’s one argument that will be raised against the federal government’s use of its license to practice and have practiced. If the federal government does this, then it will diminish the interest companies have in taking exclusive licenses that preserve the patent monopoly. If such a reduction takes place, so the argument goes, fewer subject inventions will be “commercialized” and Bayh-Dole’s inspired vision will be thwarted, the public will suffer, and federal research dollars will go wasted. There a plenty of variations that place qualifiers and attempt to cover things over with abstractions and the like. But there’s no reason to accept this argument. Whether companies have diminished interest in taking an exclusive license has nothing to do with whether a subject invention is developed for use with public benefits. Companies might take non-exclusive licenses, or licenses with exclusivity for only eight years. If there is a return on investment, then companies will still consider the investment. But perhaps companies with owners demanding high profits or high valuation at the public expense (in the case of life-saving therapeutics) is a lousy way to do things, and other associations–nonprofits, say–or a new class of social-purpose for-profit might step in and do the work needed, with honor and commitment, and having a fair wage for quality work is the only thing needed, without also offering a betting pool for speculators on the endeavor. I know, it’s like imagining a college without a football team to contribute to sports betting, but it is possible and even realistic and even desirable to condition speculators–the proclaimed saviors of society–to the idea that not every societal initiative to improve the lives of its members must involve some sort of speculative betting on the value of the outcome.
Certainly the federal government has no need to create a welfare program for technology speculators, and especially no need to focus such a welfare program on life-saving technology for the benefit of the profits of corporate entities. Yes, I know that public pension funds are neck deep in investments in these corporations, and that many smaller companies feed in the ecosystem of the corporate pharma and biotech giants. There is a whole universe of livelihoods in there, and they ride, like a monster parasite, on the back of people who have suffered a health problem–those people, their families, and the governments that aim to help them (and the insurance companies that aim to profit by betting that they can be helped through medical care that spreads the cost of treating otherwise financially catastrophic conditions).
Bayh-Dole creates the conditions for monopoly behaviors for POSIs and does a crappy job of establishing protections in the patent system and in the standard patent rights clauses for addressing the public downside of monopolies derived from publicly spirited faculty research and publication. No one cares about limits to patent property rights. No federal agency makes the effort to carve out exceptional circumstances from the defaults. No federal agency works to enforce the key monopoly-limiting elements of Bayh-Dole and instead they focus on the monopoly-producing elements (such as stripping inventors of ownership) and on the proper handling of paperwork (oh, look, a squirrel!). And no federal agency dare attempt a march-in.
But Bayh-Dole also presents the federal government’s own method to address monopolies on POSIs by giving the federal government the right under license to exploit subject inventions for the federal market–to practice and have practiced, for and on behalf of, “the United States.” If the federal government acted on this license just once, then the federal government would have leverage on the non-federal, private market as well. If universities chose to preserve their patent monopolies on POSIs and license-assign rights to companies, then those relationships would be subject to competition for the federal market. If the monopolist companies and their university licensors want a share of the federal market in addition to the private market that Bayh-Dole provides for them to exploit, then they will have to price roughly no higher than the cost to produce the drug plus the cost distributed over each unit to develop a federal generic version of the drug. That is, they will have to price as if there is competition, at least for selling into the federal market. That, then, is the alternative to negotiating drug prices–and the federal government already has this right under a broad license.
Failure by the federal government to exercise this right has contributed to the success of monopolies on prescription drugs. We might go so far to say that since the federal government has these licenses, and these licenses are clearly assets with great potential value, then the federal government is wasting a huge amount of money and should be reported via whistleblower regulations to force a change in federal practice (and having practicing). For that matter, the federal government could offer monopolist companies the opportunity to buy out the government’s license–in the case of Xtandi, say, for a price that would put Xtandi at $5 a pill for all federal government beneficiaries. No negotiation on drug prices, but the result is that the federal government and affected families pay 1/10th what they are paying now. Spread that across the drugs and other technologies that have been licensed as monopolies by universities and we are talking tens of billions of dollars in savings per year.
If the federal government expanded health care coverage to all Americans who wanted it just for POSI monopoly controlled prescription drugs, then the Bayh-Dole government license would permit the federal government to authorize generic versions of these drugs for the federal market. If a POSI monopoly company wanted to compete for the federal market, then it would need to price accordingly or buy out the federal license. Otherwise, its monopoly market becomes restricted (in the U.S.) to those Americans who choose not to use the federal POSI program for prescription drugs. If those Americans want to pay $88 a pill (or their insurers do) rather than $5, that’s their free choice.
Bayh-Dole is a law that has been only half-used, and more than half-badly at that. But Bayh-Dole has a chance to redeem itself, and for that to happen, either university monopoly Grinches have to grow larger hearts (unlikely, but hope springs eternal) or the federal government has to be directed (perhaps even likely these days) to exercise its license to practice and have practice.
There are all sorts of nasty bits underlying such an effort. For instance, development of an invention often requires additional inventing, and those additional inventions may not be POSIs. Thus, though the government gets a POSI license, it doesn’t get any access to the additional patents that might result. There are ways to address those concerns too–the effort adds costs to design around, but the resulting design-arounds also are not restricted to the federal market and might (happily in its way) destroy the value of the patents used to delay or run up the cost of the federal market. Where those patents are held by federal contractors (such as universities with tiny Grinch hearts), perhaps the federal government would simply decline to provide more research funding. That might get the attention of the university folks with the green skin. So there are complications, and there are ways to navigate those complications. And there will be consequences.
Exercising a federal license takes money (and speculator value) out of the private market. That’s a bubble that might benefit from some deflating. But any deflation is the result of following public policy pertaining to health care. And here we see the policy question: is it the purpose of federal policy to create speculative investment opportunities that fuel the profits of all those folks dependent on POSI monopolies or is the purpose of federal policy to produce discoveries that contribute to public health and welfare? One might say that’s a false dichotomy and somehow, magically, there can be both. Perhaps. But in the present implementation of policy, people with disease and injury exist to so that corporations can extract a greater share of their wealth (if not all of it), enabled by monopolies on POSIs unchecked by the federal government’s exercise of its own license to practice and have practiced. If the goal of public policy is to deliver live-saving/altering/improving therapies at 1/10th the cost in the private monopoly-dominated market, then isn’t that where federal policy should focus, and not on enhancing the private monopoly market through indifference to the federal government’s own license assets?
The federal government has a mandate to use the assets it has in the public interest. It’s just wrong for the federal government to have a POSI license–it is the clear intent of Congress that the government have this license–and to waste that license repeatedly. POSI monopolies get stronger and more bold. Drug prices go up. People suffer. The government pays many times more than it should. People bitch and moan about the prices and do nothing to address the government’s own culpability in creating the monopoly pricing in the first place. Bayh-Dole was designed to be an end run around just these sorts of government efforts to deliver cutting edge discoveries to the public on reasonable terms. Much of Bayh-Dole is a bungled mess that has been exploited to create a massive stockpile of federally supported inventions held behind university patent paywalls, dedicated to monopoly-preserving transactions. But Bayh-Dole also holds its own salvation in the form of the government license to practice and have practice. Leave march-in to addressing the ills of the private POSI market. Use the government license to deliver what’s needed by the federal market, and use the very idea of using the government license to bring down private-supplied POSI drugs to competitive levels by
- breaking unresponsive POSI monopolies
- encouraging universities not to make POSI monopolies in the first place
- creating opportunities for companies to serve the federal market
- expanding the federal health care program for POSI therapies
- offering POSI monopolies to buy out the relevant federal license in exchange for reasonable pricing for the federal market of POSI therapies
Historically, government markets have been a huge driver of innovation. Think of the markets that formed around the defense industries as they sought new weapons and new communications and new ways to help people heal. Think of the markets around space exploration and atomic/nuclear energy (not just bombs and reactors, but all those biomedical applications). Where a government purchases innovation for its use, all sorts of good things happen. But if the government by policy limits its purchases to private monopolies, it steals money from its own services (and from taxpayers, and from patients) to fatten up investors. That’s a cruddy public policy (unless one works in, say, university licensing, pharma, or conventional patent law). But there’s a shiny good public policy sitting there on the shelf in Bayh-Dole, crystal clear as the day Congress set it there to be used by the federal government. That’s one asset in Bayh-Dole that’s still sitting on the shelf, but needs to be used.