Use Bayh-Dole 35 USC 202(a)(ii) to deal with (future) drug prices

The Washington Monthly has a new article out by Alicia Mundy on high drug prices and Bayh-Dole. Mundy reports on efforts to use march-in procedures under Bayh-Dole to force lower the price of drugs that were developed with federal support. There’s substance to the argument. James Love has a well documented petition to the DHHS that makes a strong case, too.

Perhaps. But it will take more than an executive order that changes federal agency policy for march-in, since the basis for march-in is hardwired into 35 USC 203. Furthermore, the awful (and university patent broker influenced) implementing regulations for march-in (37 CFR 401.6) are so bad that even if an agency tried to march-in–with a legitimate cause–an offending contractor (or licensee) could stall the process for years.

An executive order might require federal agencies to use march-in whenever there appeared to be reason (such as for all the subject inventions that have never been licensed–free the 99%, at least!). But it is difficult to see how march-in criteria, as they are written, can reach to high drug prices where there is widespread availability. The criterion that folks want to use is this:

action is necessary to alleviate health or safety needs which are not reasonably satisfied by the contractor, assignee, or their licensees;

Read carefully: the action must be (a) necessary (not discretionary); (b) must alleviate needs (not wants–cf. Mick’s comment); that are (c) not reasonably satisfied by the private patent owner and crew.

Is a need “reasonably satisfied” if something is available–Xtandi, say–for $120K a year when it could be sold profitably for $5K? There’s a lot riding on “reasonably” here, especially if drug companies argue that it costs hundreds of millions of dollars (billon!) to do the clinical and formulation work to develop a new drug (and get to bill the costs of all the failed compounds against the winners as well). In the case of Xtandi, it appears the total cost of development was under $500m. In the first three years of sales, the drug has paid off its development cost many times over. Nice. 

The backing argument is that without the huge possible upside (or two decades of price gouging based on a monopoly on something life-critical), speculative investors won’t back companies developing new drugs and people won’t get any drugs whatsoever. If people–er, um, pension funds and endowments and private equity funds–are not presented with a huge upside earnings potential, then they will pull out of investing in U.S. pharma stocks, the stock prices for these companies will collapse, and the pharma industry in the U.S. at least will be ruined.

That argument may indeed be spot on. Should you care? Should someone make it a law that you don’t have any choice?

The question, though, is whether we, as a society, value a speculative industry based on turning acute suffering into chronic suffering so much we are willing to subsidize that industry with public money–whether on the research side or coming from state pension funds and public university endowments–or even from Bayh-Dole.

Mundy introduces Bayh-Dole with the usual narrative:

A complex piece of legislation that took four years to write and pass, Bayh-Dole was designed by its sponsors, Indiana Democratic Senator Birch Bayh and Republican Robert Dole of Kansas, to encourage the commercialization of federally sponsored research.

Not bad. We might note that Bayh-Dole disagreed with the model of commercialization that had been embedded in federal patent policy for decades–that commercialization arises from broad access based on practical use. We might also be more specific–to encourage institutional (not inventor, not corporate) ownership of patents on inventions made with federal support. Or snarky–to route the results of research conducted on the premise that it will contribute to the public welfare through patent brokers to create private monopolies as a betting pool for oligarchs (which in the U.S. look like pension funds and endowments and foundations).

Now, “usual narrative” stuff:

At the time, much of that research was sitting on shelves in university and federal labs

There is no evidence for this. The research was published, publicly available, and if there were patents, they were held for the public by the government–or were made available to contractors with equitable commercial positions or who committed to develop an invention to the point of practical application. The research was available. Not sitting on shelves. Not behind a paywall. If it was sitting, it was because no one had a use for it, or no one had a use for it that required the development of a commercial product, or that no one thought that one had to develop a commercial product first, before any other use. And who was willing to make the argument that the federal government had an obligation to help create private monopolies that prevented use so that speculators could charge monopoly prices to folks who already could and wanted to use the invention? Ah, yeah, that.

Here’s the stark reality. Most research doesn’t get “used” by commercial concerns, doesn’t become “products”–even when the research is done by commercial concerns. There’s just no reason why it should, why that is the purpose of research or those supporting research. Research shapes what is possible–especially exposes what could be possible. Into that space, new technology may move. There may be utterly non-commercial reasons for the move. There may be utterly non-profit “incentives” for the move. Research develops capabilities, open ups possibilities, challenges consensus claims, explores the implications of claims, and yeah tries to solve problems, but even there, there may be ten solutions and twenty more that are possible. In the case of Xtandi, there were over 170 compounds to choose from–and probably plenty more than that. Well, if Xtandi is compound 169′ and not compound 37, then gosh, looks like compound 37 is “sitting on the shelf” even though the “invention” is getting used. And so are plenty of other compounds. And a monopoly patent held by UCLA and licensed exclusively to Medivation (now acquired by a subsidiary of Pfizer) ensures they likely won’t be used for two decades. Well now.

because companies could not get secure enough title to the discoveries to be willing to invest the extra dollars necessary to turn them into salable products.

“Enough title” is garble. Under the IPA program, nonprofit contractors were required to “secure” title to subject inventions they chose to patent. Under DoD research programs, contractors could routinely request and receive permission to acquire title. Under the Kennedy patent policy–eventually destroyed by Bayh-Dole–federal agencies were to allow contractors to acquire title when they had an established capability and commercial position, or when the public welfare would be better served with contractor ownership. For any of these contractors, federal policy already provided for disposition of title. Many of the patents owned by the federal government were ones that federal contractors–especially commercial defense contractors–did not want. The patent served no useful commercial purpose on top of the invention, on top of the opportunity, on top of the technology, on top of the goal.

The federal government then could use a patent to record the technological advance (in the patent literature, making further patenting push to a more demanding requirement for novelty and non-obviousness) and to ensure that rights were broadly available to anyone who wanted to use (while addressing issues of importation–a little protectionism still left in patent laws) and dealing with matters of foreign patents (more touchy and thoughtful then than now–the federal government considered making our research available to the rest of the world as a service, rather than seeking to tie that research up in private monopolies exploiting the access in other countries).

because companies could not get secure enough title

And this bit of the usual narrative is pure nonsense. It’s not Mundy’s fault–it’s nonsense in the narrative. Mundy just reports it. For university drug research, the IPA program provided for “secure enough title.” There, Bayh-Dole didn’t change much of anything (well, it did, but not in the usual narrative).

There was an argument about “title uncertainty”–but it was an entirely different argument. That argument was whether patent brokers could keep title that ought to have gone to the federal government to support public availability. The patent brokers wanted to step in and intercept that title on the way to the government to create private monopolies which they could then sell access to. The patent brokers did not like federal agencies taking their time to consider what was in the public interest, especially when the burden was on the broker to make a positive case that a private monopoly was better than a public, open-access monopoly. The uncertainty had nothing to do with companies not getting “secure enough title.”

We are talking about federal research supporting scientists and engineers at universities and other nonprofits. Under the IPA program–directed at just these organizations–companies would never get title. All they could get would be a license, with an institution or patent broker on the other side of the deal. When asked directly about this at a Congressional hearing–why not just let the companies own the patents? why make things pass through a patent broker? The university patent brokers had no good answer. It was just political necessity, they said, because Congress was not ready to let companies own title to patents made with federal support. And that was only true in the technical sense that Congress did not have to make ownership go one way or another–things were managed by executive patent policy, and so Congress did not have any particular reason to decree arbitrarily what would be in the public interest in any given research contracting situation, especially when the contract was managed by a nonprofit.

Think about it. If a nonprofit wanted to take title to inventions made with federal support ahead of the government, the public interest argument would be that the nonprofit would do so without a profit motive–would do a better job than passive open availability to encourage folks to use. You know, teach the invention (which the federal government might not be prepared to do) or find companies that would be happy to find out about the invention and move to use it (again, something that university-based folks might have more incentive and capability to do, being closer as it were to the research and the community that might benefit). That is: the university public mission might have greater capability and information to move a discovery to public use. In this, a patent in the hands of a nonprofit would serve the same function as a patent in the hands of the federal government–to create a public commons, to encourage use. There was no profit motive–just a need to find support for the patenting and transfer work–and that support need not come from licensing patents any more than support for research or teaching or public service come from licensing patents.

So the IPA program proposed non-exclusive, royalty-free licensing as the default. Or at worst, non-exclusive licensing with no more than a “reasonable” royalty. Anything else would be rare, thoroughly justified, and subject to public accountability and review. If an exclusive deal wasn’t accomplishing its purpose or wasn’t justified, then the monopoly part of the license could be terminated by the federal agency if the nonprofit wasn’t diligent or willing to do so itself. The only difference between nonprofit ownership of patents on inventions made with public money for a public purpose and federal ownership of those same patents was the ability of the nonprofit to do better than the federal government in doing what the federal government was trying to achieve–public availability, public use, public benefit.

But when the federal government reviewed the IPA program, they found that nearly all the inventions were being licensed exclusively. That is, the universities weren’t using their patent monopolies to create public availability of the invention–they were using their patent monopolies to create speculative positions in patent rights. These positions, the university patent brokers claimed, were necessary to attract “the investment required to create commercial products.” The university patent brokers did not bother to complete their statement, and include the whole truth–“that would have otherwise been used without commercial products or would have been developed by multiple companies each pursuing its own competitive use of the invention.”

Not only that. While the licensing rate for privately supported inventions under university patent broker management was reported to be anywhere between 25% and 50%, the rate for IPA patents was more like 5%. WARF, a leading advocate for the IPA program and then for Bayh-Dole, didn’t have a single IPA-based invention become a commercial product between 1968 and 1975. The IPA program was, on its own evidence, operating at 20% the rate of privately held patents, and no better than the federal government was doing without trying. But there was a huge difference. Every invention covered by a federally owned patent that wasn’t licensed for “commercial” development was otherwise available to everyone. Every invention covered by a university-owned patent that wasn’t licensed for “commercial” development wasn’t available to anyone.

Companies did not need “secure enough” title to federally owned inventions. They just used them. That was all that was needed. Companies never got title to university-owned inventions made with federal support. They did not get to decide whether a patent would be acquired or something would go to the public domain. They did not get to decide how a patent might be used, whether to document technical capability or to cross-license or to defend against aggressive competitors or to contribute to a standard or to get a seat at the table in discussing industry roadmaps or interoperability. All of those possibilities were sacrificed to a demand from the university patent broker that the only legitimate use of the patent was to create a private monopoly by which to make a commercial product based on the invention. That is, to sell product that pays a royalty to the patent broker. That is, the premise of university ownership was to preclude broad access and use in favor of royalty payments from a favored “business partner.”

Who could possibly want to be on the other side of such deals? Not most companies. Not most industries. Mostly speculators. Ones that wanted to create monopolies to deny access, deny use, deny competition–and in that environment create commercial product, or to make a gesture toward creating a threat to industry practice so that they get bought out by someone in that industry. If there’s going to be a product, it will come at monopoly pricing, not at competitive pricing. For this, speculators indeed want “secure enough title.” That means: they don’t want a license that promotes competition. They don’t want monopoly that’s uncertain. They don’t want a monopoly with march-in rights that would unwind it just when the monopoly was working at maximum success–vast demand, high prices, low costs. That’s the policy debate–whether patent brokers should have the freedom from public accountability to take up all inventions made with federal support in order to place some few (under 0.5% as it turns out) with speculative monopolies based on huge interest and possible public benefit without any demonstration that a monopoly was the only instrument by which public use and public benefit might be obtained.

Bayh-Dole mandated that the labs and universities could patent their discoveries and sell the royalty rights to private-sector firms.

“Mandated” is wrong (the law is directed at federal agencies–the object of mandate should be agencies, not a relative clause). “Their” is wrong (the inventions were not the universities’ property). And “sell” is wrong (license). And “royalty rights” is garbled (patent rights). And the word “exclusively” is left off. And “without public accountability” is left off. Here’s a clearer statement, not in the usual narrative:

Bayh-Dole prevented federal agencies from intervening when a nonprofit patents inventions taken from inventors and research projects supported with federal funds and exclusively licenses those inventions to private-sector firms, seeking to share in the speculative value of the monopoly so created.


Bayh-Dole greatly reduced federal intervention and public accountability when a nonprofit exclusively licenses inventions made with federal support to private-sector firms.

Universities could license under the IPA program, among others. What university patent brokers wanted was the freedom to license exclusively without public oversight, without the threat of march-in, without limitations on the term of exclusivity, without worrying about public purposes. The only “public purpose” was making money for universities by demanding commercial products be developed. For most inventions–99% or more–there was no basis for commercial products. That’s not because the inventions were worthless–or even not “commercially valuable”–but because there was no need for only a commercial version of the product for their use. University patent brokers routinely deny company requests for a royalty-free non-exclusive license–even for inventions arising in research sponsored by the company!

The patent brokers could not imagine any other public purpose than making money. The profit motive was the heart of the patent laws. Exclusive licenses were the best way to attract people with a profit motive–and what a better profit motive than the monopoly speculator, who could use the monopoly to extract the maximum profit, and could therefore attract other investors who wanted a piece of the action? This is what Bayh-Dole was set up to promote. One can invest in production–put up money to retool a factory to produce a product–or one can invest in speculation–put up money to bet on the future value of patent rights. Sure, there might also be a product, but there might also be a company (that does other things, or gets bought up before it can do anything), or an opportunity to troll industry (if other companies implement a version of the invention anyway).

Bayh-Dole eliminated the IPA requirement that a university have a viable technology licensing program, policy, and practice. Bayh-Dole eliminated federal agency review of university licensing programs, but for a speck of possible review of small business preference licensing practice. Bayh-Dole made reports of invention use secret. There are no public data on the success or havoc of Bayh-Dole. Bayh-Dole eliminated most of the IPA march-in provisions and made narrow and arbitrary the basis for a federal agency to review the public interest served by university licensing practices. Bayh-Dole even mismatched the statement of purposes for the law with the march-in available to protect the public interest. This was by design. Bayh-Dole makes march-in so difficult and slow that it will never be used within the time frame that it is needed.

The Washington Monthly article distinguishes two kinds of government rights that might be used to deal with high drug prices: march-in and royalty-free rights for government purposes. Neither is a stroke of a president’s pen away from solving the high drug price problem. March-in is designed and drafted to fail. Royalty-free government rights are restricted to “practice” of an invention–not sale.

Folks will have to hit the law where it is weakest. One way to do that is to hit the “exceptional circumstances” clause in 35 USC 202(a)(ii):

Each nonprofit organization or small business firm may, within a reasonable time after disclosure. . . , 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.

There’s where an executive order can hit home–at least for future drug development, say, five or ten years from now. Aim it only at nonprofits. Leave the small businesses out of it for now. There is still apparatus to navigate–the law gives the Secretary of Commerce the responsibility to decide whether an agency has made a proper determination (35 USC 202(b)(1)):

If the Secretary of Commerce believes that any individual determination or pattern of determinations is contrary to the policies and objectives of this chapter or otherwise not in conformance with this chapter, the Secretary shall so advise the head of the agency concerned and the Administrator of the Office of Federal Procurement Policy, and recommend corrective actions.

Notice how the whole law turns on the belief of the Secretary of Commerce. The Secretary has no obligation to document a decision–just review and believe. An executive order may properly direct the Secretary of Commerce’s beliefs, for instance, that competition is necessary to the exploitation of inventions made with federal support.

The “policy and objectives” of the chapter are set out in 35 USC 200. There, patents on subject inventions are to be used to promote “free competition and enterprise.” How does a monopoly license for the term of a patent do such a thing? Either the term must be shorter or the license must be something less than exclusive (and certainly less than an assignment decked out to appear to be an exclusive license). The exceptional circumstances clause can be used not only to restrict the private “retention” of rights in subject inventions but also to enforce the use of those rights consistent with the “policy and objectives” at 35 USC 200. The restrictions, in such a case, make express the restrictions that are already present in 35 USC 200–interpreted for the case of, say, drug development, for the removal of doubt (and to discourage gaming the system and contempt of law).

Drug development clearly may come within the “exceptional circumstance” provision: there is broad public need, especially for conditions that are terminal or acute, federal funding is indicated to stimulate research to address these conditions without any need for a profit motive–and more so, without the interference of a profit motive, or the unethical use of monopolies to increase the cost to suffering patients once a therapeutic has been produced. Furthermore, the drug industry is steeped in monopoly exploitation of suffering. It is precisely where greater public vigilance is needed. This is where a federal agency is clearly well within the statutory authority of Bayh-Dole to restrict the right to retain title. Make that right subject to fair pricing requirements. The IPA did that by limiting the term of exclusive licenses to eight years overall or three years from the date of first commercial sale.

Reintroduce those restrictions, or others based on a reasonable return–a longer term if prices are at competitive rates not monopoly rates. (That is, don’t fragment the market but  Maybe the big pharma companies will not want to license patent rights to such patents. Maybe speculators won’t want to support the pre-clinical trials to shape up a compound for further development. But maybe there’s still plenty of profit for the federal government to work with companies willing to do the work. After all, if a company such as Medivation, which was speculators looking for an opportunity to contract out all the work to develop AR169′ into Xtandi, could do it, then so could the federal government or companies less dependent on outrageous profits and more focused on providing a reliable public service. I know, I’ve heard the same stories from VCs, that someone comes in and pitches a public purpose with all sincerity, and afterwards the VCs laugh that anyone could be that naive to think a public interest argument could have anything to do with their investment decisions. Even if there is a public service that arises from VCs seeking their own speculative self-interests–and I’m sure there is–why should the federal government or the everyday citizen construct a law to route everything through the hands of VCs or patent brokers or monopoly speculators? Why not compete with those folks? Or at least, why not support alternatives to those folks’ operating model? Why should professional gambling be the only game in town that produces therapeutic drugs? Why should therapeutic drugs be one of the places where the “big money” is to be made in society? That’s the policy issue.

Hitting on the exceptional circumstances provision will cause university patent brokers erupting in protest. Some university administrators will refuse to allow faculty to accept federal grants–as they have in the past. But the issue is not whether some people object–clearly, some will–but what is in the public interest. What is odd is that it would be university officials that would argue for high drug prices–that their own institutional financial self-interest should dictate a public policy in which the university is better off with a 1% share of a monopoly that exploits public suffering–and so should everyone else.

The absurd position, beyond the corruption of public institutions, is that the only way to produce beneficial medicines is by creating private speculative monopolies. If that’s federal policy, if that’s what Congress intended, then put that in Bayh-Dole expressly and damn the apparatus that appears to worry such a thing but has no effect whatsoever. Just propose that Bayh-Dole be changed to toss the apparatus, to make it express federal policy and law that the output of all federal research conducted at nonprofits should first and only be offered for monopoly speculation. If that’s not the case, if that law can’t be passed, then that’s not the federal policy, and that’s the purpose of the law.

Meanwhile, to deal with the monopoly prices of drugs currently on the market and supported with federal funds, enforce Bayh-Dole’s statement of Congressional patent policy at 35 USC 200.

This entry was posted in Bayh-Dole, Policy. Bookmark the permalink.