[Added a comment at the end.] PhRMA published recently a white paper arguing that Bayh-Dole’s march-in provisions should not be used to mitigate monopoly patent pricing. Here’s the basic argument from the executive summary. It’s a few sentences, but it is good to capture some context:
There have been several recent petitions to the National Institutes of Health (NIH) to use march-in rights in an effort to directly reduce the prices of innovative medicines.
So true.
These misguided efforts threaten to undermine the success achieved under Bayh-Dole over the past 36 years in both fostering early basic research and ensuring the use and translation of those early findings into new medical innovations.
Actually, no. Bayh-Dole doesn’t have anything whatsoever to do with “fostering early basic research.” And while it does have to do with patents on “early findings,” it does not do much at all to “ensure” use or translation of findings into “innovations.” Bayh-Dole prevents federal agencies from acquiring inventions made at universities with federal funding–if the inventors assign their rights to the university that hosted their inventions, an invention management organization, or other approved assignee. How patents on such inventions come into use is not Bayh-Dole’s concern. In fact, the march-in provisions were deliberately weakened to make it difficult for the federal government to have much to say how anyone used (or didn’t use) their patents on subject inventions.
The limited march-in right established by the authors of Bayh-Dole reflected an understanding of the inherently costly, risky, and uncertain nature of drug development and the need to provide clear, consistent, and predictable ground rules for government licensing to encourage public and private sector collaborations to harness promising scientific and technological research into advances for patients and consumers.
A mouthful of bureaucratese. Let me help. Latker (NIH) and Bremer (SUPA to become AUTM) drafted Bayh-Dole to create a way to end-run HEW policy decision. HEW wanted to make the results of basic health research available to everyone–something that the pharma industry objected to and therefore boycotted the development of NIH discoveries. Bayh-Dole interposed Congressional policy to override executive branch executive orders and their interpretation by federal agencies, replacing a flexible program with federal agency oversight with a uniform or much less flexible program generally lacking federal agency oversight.
The scheme under Bayh-Dole involves diverting patentable inventions to university control, to keep them from moving to the federal patent commons or to the public domain where they would be available for use and development by anyone. In return, pharma companies have the opportunity to license rights exclusively (from universities and their brokers, without faculty investigators or inventors or federal agencies having any say about it) for the inventions they choose, paying about 1% royalty and keeping 99%.
For this scheme to be most advantageous to both the universities and the companies, it is necessary to preserve monopoly pricing and enforcement of patent rights. Greater value means greater income from sales with less competition, and that means greater royalty income for university administrators along for the ride. University inventors are welcomed financially into the scheme with a requirement that universities also share some (unspecified) portion of the university’s royalty income with the inventors. There is no provision for university inventors to object to administrators’ choice of licensing scheme, or terms, or choice of commercial partners. Universities often also insist that inventors *must* accept royalty payments (common arguments: it is a requirement of federal law; you will have a tax liability for any money you are owed but don’t accept). Monopoly for money is just the kind of thing patent administrators (who happened to be patent attorneys) would cook up to maximize business. Capture everything, create a pool, let companies pick and choose. License a few things for a big wow. Advertise the monetary successes to attract more participates. Make everything else look like efforts in the public interest. All done.
The intent of limited march-in authority was to ensure that grantees were in fact making efforts to commercialize the licensed technology and bring applications to market to the benefit of patients and society.
Actually, there is no “intent” stated for the march-in rights. Just as with the standard patent rights clause, march-in procedures are postponed to the implementing regulations, where Bremer and Latker succeeded in making a mess of them. The law suggests oversight, but the implementing procedures say it’s not exactly possible. Under the IPA system, a university had three years from the date of patent issue to demonstrate that it deserved to continue to administrate a patent on a subject invention–show development to the point of practical application, or show a non-exclusive licensing program, or make a substantive case for it. Under Bayh-Dole, none of that–just show that there’s an expectation that effective steps will be taken, and if someone objects, then tie them up in knots for years in procedural maneuvers and appeals.
The whole point of commercialization in a corporation is to turn a profit. PhRMA argues that somehow their companies’ goal is to benefit “patients and society.” That knowingly misses the point. The question that matters is what price (to the community) the profit (to the companies). The vision that PhRMA puts forward is one in which companies scramble to find new drug candidates, taking great risks and running up huge expenses (these, to be seen as a measure of their commitment to benefiting society, apparently), and when they do find something, it is crucial that they gain a monopoly position on it, so they recover their expenditures and also have an incentive to start the search and development process all over again. The prospect of monopoly pricing and a beat-back of competition using the same compounds and techniques is what drives this overall speculative betting pool. Universities play the role of the house. Companies win big or lose bog according to their luck and ability to count cards. The public gets what it gets and the government pays for the lion’s share of it–first in doing the research, and then again in paying for the drugs as they are sold to folks on government medical insurance programs such as Medicare.
A few sentences on:
The threat of march-in as an approach to regulate drug prices would create substantial uncertainty for private sector technology development partners and dramatically alter the framework that has contributed to the growth and sustainability of the robust U.S. R&D ecosystem.
That is, march-in would undo the monopoly betting pool and make it less interesting for companies seeking maximum profits to draw upon federally funded subject inventions. That much, surely, is true. But “dramatically altering the framework” doesn’t mean things get worse for patients or society or even for universities and their money-grubbing administrators. And it doesn’t mean that things get worse for companies willing to get involved in developing new therapies for medical use.
Some history. The Institutional Patent Agreement system was built to comply with the Kennedy Statement of government patent policy. Here’s the operative provision for universities:
Universities did not have, generally, established commercial positions that warranted the government to give up its own claim to acquire inventions made with government funding. So they came under this section (c), which left it up to each agency to create policy as the need required. A contractor needed to commit to reporting and to bringing an invention to the point of practical application, so long as the public interest is served.
Within that policy framework, the IPA required that a university acquiring “principal rights” in a subject invention must make the invention available under a license with “adequate safeguards against unreasonable royalty and repressive practices.” The IPA stipulated that “Royalties shall not, in any event, be in excess of normal trade practices.” Now this is all interesting but it does not get at price, only royalties. That is, the consideration paid through any chain of assignees and licensees should be in line with “normal trade practices.” The burden of the IPA is to prevent universities and their invention management organizations from getting greedy and overcharging for permissions to use federally funded inventions.
But there’s nothing in the IPA on the prices that licensees can charge the public. The price protection is built into the nature of licensing. If licenses are non-exclusive, then competition will manage prices, and collusion on pricing is a matter for anti-trust law–as the Kennedy statement makes clear. (And Bayh-Dole repeats this mantra at 35 USC 211). For exclusive licenses, the limits on the time to achieve practical application–three years from patent issue–and on the term of any exclusive license–sooner of three years from first commercial sale or eight years from the date of the license–means that soon after the patent issues or there’s a product on the market, competition will step in to manage price.
Bayh-Dole does away with all this apparatus–nothing on price, of course, but also nothing on competition to manage price. (Well, there is something on competition in 35 USC 200–more about that). The Bayh-Dole march-in procedures replace the IPA’s by switching the burden from the university to show why it should continue to have the primary responsibility to administrate patents on a subject invention to the agency to show why it should intervene. The new standard is that the university and its licensee can show that there is an expectation that effective steps will be taken to achieve practical application. No time frame. No actual demonstration of practical application. There’s not much at all to work from if someone wants to use Bayh-Dole march-in procedures to get at monopoly pricing practices. On this point, PhRMA is spot on.
Let’s say that we feel the need to control monopoly pricing of drugs that are based on subject inventions (another discussion, another day!). What are the points of attack to the defenses that university monopolists built into Bayh-Dole in cahoots with the drug industry? Did folks in Congress believe that university administrators and especially their patent brokers would provide that public oversight if only they had title to inventions and were left to do as they pleased? Universities through their policies and public standing certainly could look out for the public interest. It’s not clear, however, how small businesses figured in the argument, except that as the potential victims of big companies they were worthy of government largesse and were mostly harmless.
Given what’s happened since, one might say that the drafters of Bayh-Dole exploited the reputations of universities to work their scheme, and then set out to sell the scheme to university administrators as better than working with Research Corporation. With no oversight, however, and no accountability, things rapidly got out of hand toward chasing the almighty dollar–not that anyone other than university administrators and patent attorneys have consistently made money but for a handful of big payouts. That’s the nature of a public scheme created by patent attorneys. Perhaps every profession dreams of its own good:
The wery hunter, slepinge in his bed,
To wode ayein his minde goth anoon;
The juge dremeth how his plees ben sped;
The carter dremeth how his cartes goon;
The riche, of gold; the knight fight with his foon;
The seke met he drinketh of the tonne;
The lover met he hath his lady wonne.
Ac the manne of lawe dremeth bayh-dole to passen.
So march-in, by design, cannot get at monopoly pricing. Even if creating competition through non-exclusive licensing might, no one has a position in the Bayh-Dole scheme to choose competition over the prospect of monopoly rents.
Attacks on such pricing based on trying to lower the royalties negotiated in university licenses fare no better. Those royalty rates are often 1% or below. Pushing those rates lower will not lower the price charged–it will just marginally increase the profit position for the monopoly licensee.
There have been a number of articles aiming to fix Bayh-Dole’s march-in procedures or to find ways around them. One calls for the “recalibration” of Bayh-Dole. There’s a proposal out there (paywalled, summary here) that the government could use its royalty-free license to drop the price it pays on drugs based on subject inventions, but again, that’s likely to be only 1% or less, and the effect is to shift money from the universities to the government, for which the universities will howl like stuck pigs and the howl will be worse for the agency than the tiny financial gain.
Consider, then, two other forms of attack that do not require changing Bayh-Dole at all–just go after non-compliant practices elsewhere in the law–practices that create inappropriate monopolies in the first place.
Attack 1. Hit assignment. Create competition. 35 USC 202(c)(7)(A).
(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)
Here’s the thing. An assignment is determined by the instrument not by its label. Universities routinely label as exclusive licenses what courts have found to be assignments. An assignment, as determined by the courts, transfers substantially all rights in the patent. The court in Prima Tek II put it this way:
Section 261 recognizes, and courts have long held, that an exclusive, territorial license is equivalent to an assignment and may therefore confer standing upon the licensee to sue for patent infringement.
A court reviewing an exclusive license granted by the University of Washington concluded it was an assignment. The license had the following provisions:
(1) to “make, have made and sell” products world-wide covered by the University’s patent rights;
(2) to grant one or more sub-licenses to make, have made, use and sell licensed product(s) covered by UW’s patent rights;
(3) to bring suit in its own name, or if required by law, jointly with UW, for infringement of the licensed patent rights including the right to sue for past infringement of the licensed rights;
(4) to enjoin infringement and to collect damages, profits and awards of whatever nature recoverable for infringement;
(5) to settle any claim or suit for infringement of the patent by granting the infringing party a sublicense for the patent.
The reservation of rights to the university to practice the invention for research purposes were not sufficient to overcome the effect of the instrument–it was an assignment.
Now consider that in monopoly licensing, universities frequently use just this strategy. The UC Davis template patent license, for instance, covers each of the five points–jurisdiction-wide rights (2.1), right to sublicense (3.1), right to sue in its own name and enjoin infringement and settle any claims (17). It reserves minimal university rights to practice (2.2), including rights for other universities and non-profits. That’s hardly enough to interfere with the monopoly commercial rights of an exclusive licensee. The UC Davis exclusive license is basically an assignment.
But if it is an assignment, then for subject inventions it comes within 35 USC 202(c)(7)(A)–the license cannot be made without federal agency approval, and if made, it must carry with it the same provisions that apply to the university. That is, the patent carries a public covenant if assigned, and that public covenant requires the royalties earned by the assignee to be (i) shared with the inventors and (ii) any balance after costs of administrating the invention must be dedicated to “scientific research or education.” So, no profits passed to shareholders as dividends.
What’s the effect of this attack? First, it takes all the fun out of a monopoly license. A licensee, to avoid assignment, has to leave substantial rights with the university licensor–not just a territorial carve out, but rights that include activity in that territory. Two forms of rights are relevant:
- The right to make, have made, and use the invention.Any commercial vendor then has to compete with local use. A commercial version of the invention has to be better or less expensive than an organization implementing the invention themselves, directly. A lab medicine department, for instance, may readily implement disease assays from the literature without relying on, or waiting for, the development of a commercial version. The commercial version, when it shows up, has to be better or cheaper than what they already have. Thus, there will be market pressure on price.
- The right to sell product licensed under the patent.If there is competition without collusion, then there is pressure again on price. There are two ways to go about this. First, the university licenses non-exclusively, but requires licensees to develop product or lose their license–thus, each non-exclusive licensee must front the costs rather than wait to ride the development efforts of the first developer. A variation is that the university delays licensing after the first non-exclusive license for a period–say, three years–to give the first licensee an advantage. A second variation is to charge somewhat more for later non-exclusive licenses, reflecting the added value of the first licensee’s development, and reducing the royalty paid by the first licensee proportionately, so that there is a financial advantage for early adopters and developers over those taking licenses later in the development process or after there’s a commercial product available.Second, the university requires the assignee to sublicense non-exclusively on FRAND terms at some point after there’s a product on the market–with enough delay that the assignee recovers its investment in development and then pricing adjusts to deal with competition–much as was anticipated by the IPA approach.
Thus, this first attack forces the monopoly licensee to give up profits and devote income to further science or education, or accept a true exclusive license that’s not an assignment and allow competition in use, development, or sales to shape price, rather than to rely on monopoly rights to set prices.
Attack 2. Hit scope of property rights. Limit the ability to enforce. 35 USC 200:
It is the policy and objective of the Congress to use the patent system to promote the utilization of inventions arising from federally supported research or development . . .
This section of Bayh-Dole is part of federal patent law. It is labeled “policy” and that label means something. It stands in parallel to the government patent policy it supplants in the executive branch. It is, therefore, the framework in which any patent owner of a subject invention must work. The personal property right represented by a patent is expressly subject to federal patent law (35 USC 261):
Subject to the provisions of this title, patents shall have the attributes of personal property.
The provisions that matter for subject inventions–also defined within patent law by Bayh-Dole–starts with this statement of Congressional policy. That statement does not just rationalize the existence of Bayh-Dole–that could be done well enough in the legislative record, along with the desire to restore national innovation competitiveness and give university patent attorneys more work and the like. But the statements made in 35 USC 200 are different in effect–they are part of patent law, they shape the property right available to an owner of a patent on a subject invention. They create a covenant, as you will, that runs with the patent, that conditions what any owner of the patent can do. Patents on subject inventions are not ordinary patents, constrained by federal contracting requirements. They are abynormal patents, with their own set of property rights–and those rights differ significantly from ordinary patent property rights.
35 USC states that the patent system, if it is used, must be used to promote the utilization of subject inventions. That is, the patent system is not available to prevent the use of subject inventions unless preventing some uses is a means of promoting other uses.
Bayh-Dole is clear, too, what it means by “utilization” (35 USC 201(f)):
(f) The term “practical application” means to manufacture in the case of a composition or product, to practice in the case of a process or method, or to operate in the case of a machine or system; and, in each case, under such conditions as to establish that the invention is being utilized and that its benefits are to the extent permitted by law or Government regulations available to the public on reasonable terms.
We have been through this before recently. Five elements: (1) use (2) established, i.e., publicly disclosed (3) with benefits (4) available to the public (5) on reasonable terms. All five elements have to be present for the definition to be met. All five elements have to be present to meet the Congressional policy that limits the use of the patent system to the promotion of utilization–that is, to the promotion of practical application.
Thus, although federal agencies have no authority to march-in on price, the Department of Justice does have standing to take up the issue of what constitutes “reasonable terms”–unlike the march-in procedures criteria limited to “reasonable availability.”
What guidance is there on reasonable terms? There is, of all things, the IPA, drafted by the same folks who brought us Bayh-Dole. There we find a similar definition of subject invention (but now not a matter of federal patent law) and practical application (with only slight differences in wording). It is clear that an effort was made to bring forward all sorts of material from the IPA into Bayh-Dole. In the IPA, “reasonable terms” means, operationally, competitive terms–non-exclusive licensing, licensing exclusively only with justification and then only for limited times, and a potential loss of exclusive control if development to the point of practical application has not been achieved in three years from the date of patent issuance.
There is no indication that anything changed from the IPA to Bayh-Dole between, say, 1978 when the IPA was still in use and 1981 when Bayh-Dole replaced it. Thus, while reasonable pricing is not directly reviewable by march-in (because efforts might be being made to lower the price, and that’s enough to defeat the express review standard), there is a direct attack available through 35 USC 200. A reasonable term–as to price–is one established in a competitive market–either a market in which the competition is local use vs commercial product (make your own cider or buy it from us) or the competition is competing commercial products, differentiated by price, quality, availability, features, service but not by monopoly rights.
Even if there is no such market, as a result of a monopoly position (see Attack 1), a reasonable price can be estimated based on what happens when a product with monopoly standing goes “off patent.” What are the generic prices then, when there is indeed competition? What are the prices for goods when local use but not sales is permitted? The reasonable price must be competitive with the cost of making something for one’s self or one’s organization. These are readily definable standards that can be used to argue that higher prices, monopoly prices, fail the Congressional policy and therefore jeopardize the license (it falls outside the property rights established for patents on subject inventions) and the patent itself (it is being misused to secure property rights that the owner is not entitled to). In essence an unreasonable price–a non-competitive price, even in the absence of competition–is patent misuse, an illegal tie.
The “safeguards” on the public interest that were so evident in the IPA approach have been broken up. The obvious part, the march-in procedures that everyone points to, were gutted so that they appear to be safeguards but really have nothing to do. The other part, the statement of Congressional policy, was hidden in plain sight, but displaced to the head of the law, where it might be mistaken for fluffy text needed by Congress to explain why they have done such a thing as Bayh-Dole. But 35 USC 200 is anything but surplusage. It has substantial import that has lain dormant all these years. Perhaps it is time to attack from the desert rather than from the sea.
[Here is Niels Reimers, at the time the director of Stanford’s relatively new Office of Technology Licensing, writing in 1978 to the U.S. Attorney General with regard to proposed new legislation regarding federal disposition of inventions made in work receiving federal funding:
Ironically, what is needed are cases where government contractor.s have been able to make “monopolistic profits” from sales in the commercial sector as a result of patent protection derived from government research. Such cases would encourage others that there is indeed something useful in results of government research. . . . The problem at this juncture is not excessive profits from commercializing patented government research results, but minimal commercialization and profits.
In other words, monopoly pricing should be encouraged by the federal government, to compensate for the apparent lack of uptake of research results from federally supported projects. Reimers argues that some abuse of the public must be encouraged–especially in things like public health, which has always been the battle Bayh-Dole and before it the IPA program aimed to fight–to somehow jumpstart investor interest in making products based on federally funded research. This thought is very strange. The problem is not merely in the abstract excessive profits. The problem is unreasonable pricing. Bayh-Dole has no problem at all with excessive profits, provided that those profits are earned in the context of “free competition and enterprise” and using US labor and industry. One might think, rather, that how federally funded research is selected and conducted might be the problem, not that the results are great but not so great that either federal agencies or companies are willing to adopt them for making products. But no one taking millions of dollars from federal agencies for research, and raking off a third of that to pay administrative salaries, is about to argue that the federal research funding allocation system sucks, even if it does suck, and it does suck.]