Things start here:
The Harbridge House report
According to the Harbridge House report on federal patent policy, from the 1930s until the 1950s, the pharmaceutical industry was the primary source of funding for university investigators in the area of medicinal chemistry–looking for compounds that might become prescription drugs. The pharmaceutical industry operating model relied on patents, and drug companies offered free “screening” services to university faculty in exchange for exclusive rights in any compound that turned out to have biological importance.
An invention in this area generally was not a single compound but a class of compounds–hundreds, perhaps thousands. The standard form of claiming a chemical invention involved the use of a “Markush” claim, first used in 1925, allowing broad groups of chemicals to be combined in many arrangements, each combination covered by the claim. Thus, once one had identified one potential compound of interest, a patent could be secured on hundreds (or more) variations on that compound. Those variations, then, might be subject to screening efforts to identify those with significant biological activity. The number of combinations, however, means that it is possible one would never get through screening all of them using a systematic, brute-force method.
The Harbridge House report cites one drug company that of 6,500 compounds obtained from university research (out of 40,000 total), 2 had resulted in products, 2 more were in development, and 2 more were of some interest. Thus, researchers could produce many compounds of potential interest, but only a very few might be of any importance. We might think, then, of such research as one of securing mining claims–finding some evidence of possible biological action (such as collecting a sample of a compound in use for another purpose, such as Brazilian tribes that used pit viper venom on arrow tips–later to become Captopril, a medicine to treat high blood pressure)–and then creating a patent claim that extends to as many conceivable variations on that compound as possible. The patent “protects” not only a future potential compound, which may be one in thousands, but also the exclusive right to look for that compound among all the ones that have been claimed (and thus are not available to anyone else).
When the Public Health Service ramped up research funding to universities in the area of medicinal chemistry, it was moving in on territory that the pharmaceutical industry already largely controlled. The interplay between industry expectations–monopoly in exchange for early screening and synthesizing of compounds–and the PHS expectation–open publication and available to all–created a huge problem. The PHS published new requirements for inventions in 1955, insisting that the government should have ownership of any inventive compound made with federal support unless there was some compelling reason otherwise., meaning that it would be up to the federal government to decide whether there would be any ownership claim by means of patenting, and if so what scope those claims would have in terms of classes of compounds. Until the government decided, there was “uncertainty” for everyone else who might have an interest.
In 1962, the PHS required investigators and universities to identify any parties to whom they owed commitments that might interfere with the PHS patent policy, and obtain from those parties a patent agreement that placed PHS requirements ahead of their own.
One might see the prospect for disagreement. If an inventive compound really was a class of compounds, then sending a compound to a company for screening could result in identifying new variations of the compound that were even better–but these new variations would also be compromised by the federal support, and thus unavailable for a commercial monopoly–or, rather, potentially unavailable for a commercial monopoly–there would be uncertainty with regard to the outcome of the PHS’s determination of rights. And as shown with the WARF 5-FU situation, the PHS was ready to undermine monopoly commercial positions based on patents to work that it had supported. No public subvention should be turned so readily into a subsidy for any single monopolizing commercial interest.
The pharma companies refused to sign the PHS patent agreement and therefore refused to provide screening services for potentially interesting compounds identified with federal support. University investigators sought alternatives–there were two federal screening organizations, various commercial screening companies, and some university labs that could provide services. According to the Harbridge House report and a parallel GAO report on the NIH’s problems with medicinal chemistry research, the alternatives were generally viewed as inadequate. Thus, the argument went, many compounds were not screened. We are talking about hundreds of thousands of potential compounds scattered out from a few hundred identified compounds–it is easy to see that many such compounds would never be tested anyway. And beyond these, millions upon millions of possible compounds yet “undiscovered.”
It is easy to see how one could construct an argument that the PHS invention regulations were disrupting available screening services offered by the pharma industry over the issue of patent monopolies. Furthermore, if commercial services provide the most ready path to the preparation of a new drug candidate, and the stream of new federal regulations put out of business all those companies unable to afford the expense of navigating these regulations, then refusing to allow monopolies on newly discovered compounds prevents these compounds from being screened and developed for public use. The PHS regulations fail the argument for urgency, fail the argument for expeditious development, fail the argument for collaboration between academics and industry. If the assault on human disease is a matter of war, then the public should expect to sacrifice to achieve victory, and the fundamental sacrifice is that they pay more for twenty years than they would otherwise, and in return they get the safest, most effective medicines known to humanity.
Everyone, it appeared, was working to circumvent PHS patent policy. Here’s the Harbridge House account of it from their 1968 report describing in detail the problems with the PHS (now HEW) patent policy:
The rules may sound nice, but the rules also weren’t responsive to the situation. As a result, the NIH resumed the IPA program, but now designed to circumvent PHS rules but not make a show of doing so. When the IPA program got caught out, then the same folks moved on to produce Bayh-Dole, billing it as a “uniform” policy for all inventions, government-wide rather than as a special program to address problems in medicinal chemistry.
Both arguments have their merits, but neither is compelling. But that’s politics. There’s nothing about the “uncertainty of title” that really matters, though the problem is pivotal. Prescription medicines could be placed beyond the scope of patent property rights–along with laws of nature, business methods, and surgery techniques. Other forms of government regulation might provide exclusive commercial positions to reward private investment, by-passing patent issues altogether. Alternatively, the federal government could have done a better job working through the entire system. Rather than merely fund “basic research” into medicinal chemistry, it could have developed its own high-throughput screening operations and from there pushed new therapeutic compounds all the way to the market–to a federal market established to improve public health. Like space. Like nuclear energy. Like atomic bombs. You see the pattern anyway, along with the ironies.
Instead, the stasis of the argument involved “certainty” of title to inventions. Companies wouldn’t participate in screening unless they were sure of their future monopoly for whatever they “discovered” in screening. (Here’s the irony for you–clinical trials are an advanced form of just such screening, and in those, drug companies insist that universities give up all rights in anything “discovered” while conducting the clinical trial–one of the few areas in which universities do not routinely claim an ownership interest in inventions made in sponsored projects–one might think, then, that if the PHS (then HEW, then DHHS) were the prime mover in an effort to find new therapeutic compounds, from the bazillion of such compounds possible, then anyone involved–academic chemist, drug company screener, manufacturer, pharmacist, physician–owed the PHS any monopoly interest in anything discovered along the way that might assist the effort. To fail to give up such monopoly rights was to assert the right to disrupt the effort for money–a kind of extortion. So screw the monopolists (this is a technical phrase to put an edge on it).
But this sort of screwy thinking fails to take into account social behaviors–even rational folks (other than chimpanzees) tend to decline deals that are otherwise in their interest to teach others a lesson in how to share opportunities. And this is what the drug companies did to the PHS–and beneath the surface, academics investigators and even the patent counsel at the NIH agreed with the circumvention. But it was the patent brokers–and especially folks at WARF–that made the public argument in favor of industry monopolies in matters of medicine.
The argument they constructed to get their way was based on the perception (i) that there were 28,000 unlicensed government patents (even though the vast majority of these were DoD inventions that contractors had declined to own and were made available to anyone, formal licensing not necessary); (ii) that the federal government had no expertise in licensing to encourage private development (even though federal licensing rates for biomedical inventions were 5x better than the research foundation rates for federally supported inventions); (iii) that “uncertainty of title” was the issue that blocked private management of federally supported inventions in the public interest (even though the issue was actually not title but private monopoly); (iv) that a “uniform” government patent policy was needed that gave subvention contractors (nonprofits and small businesses) the clear right to hold title in patents (even though there was a uniform policy already and what was proposed in its place was arbitrary); and (v) that various protections for the public interest placed in the regulations would take care of any problems that might arise (even though these protections would be designed not to operate).
It’s all very clever, very political–and how the drug companies won the battle to preserve their territory, drive out of business the small companies that lacked the resources to meet the regulatory requirements, and obtain the benefit of federal funding as a subsidy for their own drug discovery efforts. Even if we accept the idea that patent monopolies were designed precisely for the purpose of developing medicines as a lucrative commercial venture, there’s nothing that leads one to the idea that such a policy should be applied uniformly to all inventive work in all markets for all products and expect a comparable “public benefit.” Here’s the Harbridge House conclusion:
That is, no arbitrary government-wide patent policy will do. “A balancing of government objectives” is necessary. Kennedy’s executive branch patent policy says as much:
The argument behind Bayh-Dole disagrees with both of these statements and implements an arbitrary default government-wide patent policy, but calls it “uniform” and ignores the findings of the Harbridge House report, which identified two areas in which the government might productively allow contractors to hold title in inventions made with federal support. First, what’s now called a “dual use” situation in which the invention developed for the government isn’t suitable for a non-government market. In such cases, if the contractor has commercial experience, then allowing the contractor to obtain a patent position to develop the invention for a private market appears to be a good idea:
Second, we have work that’s left incomplete by the federal agency and involves an area of industry that’s patent sensitive:
This is the invention of the form that is now depicted by university licensing offices as the only sort of invention that exists. The pattern of testimony regarding inventions from 1978 to the present routinely describes inventions as “early stage” and in need of “industry investment” which will be much greater than the cost of the research that has led to any given discovery or invention. The Harbridge House report makes clear that it’s just not the case. In particular, the report found that in the Department of Agriculture and Department of Interior, the “public service” agencies, all inventions that were reviewed got to market without any private patent monopoly:
That is, the inventions as produced under contract were sufficiently useful that no one commercial firm had to allocate significant development money to complete the work to make the invention useful as a matter for sale.
The final segment is here: