Use the government license in Bayh-Dole

According to news reports (here’s where I first read about it–follow the link there to the article in the Baltimore Sun), the state of Maryland is attempting to deal with high drug prices through legislation that gives the state the right to sue companies that raise drug prices without a good cause. Perhaps that will work. If I’m at a drug company, then, I start with an outlandish list price (so I never have to raise it) and discount as necessary to get to my optimal market price.

For drugs based on compounds invented with federal support, such as in research at universities supported by federal grants, there’s another way. The Bayh-Dole Act, which controls patent rights on inventions made in federally supported research, requires federal agencies to receive a license “to practice and have practiced” each invention for and on behalf “of the United States.” 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 .  .  .  .

This language has been used in executive branch patent policy since 1963. “Practice and have practiced” is defined to mean “to make, use, and sell; and to have made, used, and sold.” The scope includes states and municipal governments. See Here:

Governmental purpose–means the right of the Government of the United States (including any agency thereof, state, or domestic municipal government) to practice and have practiced (made or have made, used or have used, sold or have sold) throughout the world by or on behalf of the Government of the United States.

In Bayh-Dole, this language is varied, but the scope is still clear. “United States” is, well, the states. For drugs based on compounds invented with federal support, states have a license to “have made and have sold” those drugs for any state purpose.

It would be interesting to see a state do more than object to increases in drug prices. For instance, Xtandi, a prostate cancer drug invented with federal support, sells in the US for about $70 a pill, but a generic drug company says they could produce the drug for $3 a pill. Why shouldn’t a state just just license a generic manufacturer to produce the drug for everyone that the state otherwise has to cover as a matter of health insurance? Maybe that’s just state employees, but that’s a start. For each 8-month treatment with a generic version of Xtandi, a state would save over $50,000. 20 patients = $1m in savings.

This approach has nothing to do with “march-in” procedures. Folks keep banging against that window. March-in procedures under Bayh-Dole were designed by patent brokers not to operate. They bragged about it afterwards. March-in depends primarily on availability, not price (it depends on what federal agencies decide “reasonable terms” means), and even then on a determination by a federal agency that march-in conditions have been met, and even then on a determination that march-in should be used as a remedy. Once a march-in proceeding has started, the patent owner has plenty of opportunities to delay, appeal, and litigate to prevent the march-in. And, oh, there’s no mechanism by which the public can request a march-in. It’s entirely up to each federal agency whether to gather information at all, let alone act. And the NIH has shown no interest in acting.

Worse, march-in is set up to appear to be a “taking” of private property for government purposes. It isn’t this, actually, but that’s beside the point. As a taking, the claim is, when a federal agency marches in, it may have to compensate the patent owner for the value of the taking. Whoa–you see what happens then? The government may have to pay the patent owner the value of what the government will otherwise save by compelling the patent owner to grant more licenses, creating competition, lowering the price of the drug, and the like. While health insurers may save money–making shareholders in such companies happy–the government might not save much at all. Another booby trap placed in Bayh-Dole by the patent broker operatives. Since Bayh-Dole sets the terms for patent rights in federal research contracts, a failure by an owner of a patent on a subject invention is not a breach of law but a breach of federal contract. March-in is a remedy for that breach. But Bayh-Dole doesn’t come out and make this clear, either.

Under the prior executive branch patent policy, which Bayh-Dole apparently (it’s not clear–at least the law is consistently muddy) superseded, federal agencies were directed to contract so that a qualifying contractor obtained only “principal rights” to a patent on an invention made with federal support, and the government did not have to “march-in” based on some failing of the contractor with regard to making an invention available for public benefit. If the government created a regulation that required the public to use the invention, then the government could require non-exclusive licensing. Didn’t matter what the patent owner had done, was doing, or was hoping to do. The patent owner never had such rights that when the government took action, the patent owner’s private rights were diminished. The patent owner simply had room to operate pending government actions that might limit that room. Bayh-Dole, of course, replaced this way of viewing patents on inventions made with government support with the idea that the contractors could own and if the government “marched-in” it was something of an eminent domain action rather than calling in the option that one always had or acting to remedy a breach of the patent rights clause in the federal funding agreement. Bayh-Dole is set up to prevent march-in. March-in is a dead end, by design and in practice both.

But the government license is entirely separate and has nothing to do with march-in, availability, reasonable terms. Nada. The government license is an outright grant made by each owner of a patent on a “subject invention”–an invention made with federal support under a research contract. The government can exercise its license in any way it pleases, for any purpose within the scope of the license–“for or on behalf of the United States.” That is, if the federal government or a state or a municipal government has authority to do something (say, pay for health benefits), then the license extends to that doing, and to anyone who provides services to that doing, to “practice and have practiced.”

Under Bayh-Dole, federal agencies receive licenses on behalf of the United States–so each state, in addition to the federal government, enjoys the benefit of the license outright. It is an asset of the state. To not use that asset is a waste of state resources. Rather than fuss with march-in, folks should be going directly after the government license.

If one wants to get a sense of the mental (and logical) geography, here’s how it works. Federal funding divides up a market into a private-side and a government-side. On the private side, the owner of a patent on a subject invention can exploit the patent to create products to use and sell, subject to some limitations imposed by Bayh-Dole and its standard patent rights clause. The patent owner can sell products or license others to make and use the invention, with all sorts of combinations possible. The patent owner can even authorize sales to the federal government, and the federal government can be happy to purchase products this way, especially if these are the only products available (and a monopoly is one possible outcome of a patent, yes?). All this is private-side market.

But there’s another market, and that’s the federal-side market. On the federal side, federal agencies (and states, and municipal governments) have the freedom (under the license to practice and have practiced) any subject invention for any government purpose. They can make and use the invention and they can commission companies to make and use the invention and sell products based on the invention for, or on behalf of, any government. The government-side market arose to deal with procurement needs of federal agencies where they were leading the development of technology–for instance, military applications, atomic energy, and space technologies–where there was no private-side market. Government-side markets also addressed issues in competitive bidding for research services, where awarding a contract to one company should not give that company patent positions in research results that would prevent other companies from bidding to do follow-on or related work for the government. And government-side markets included anything that had to do with regulatory matters and treaty obligations. If there was a government purpose involved, then privately held patents based on inventions funded by the government were not going to stand in the way of government actions.

The government market for inventions is distinct from private-side patent exploitation opportunities. If a patent owner wants to “call forth risk capital” to develop an invention made with federal support for the private market, it may do so. If the government sees an advantage in purchasing product based on the invention in the private-side market, it may do so. But if the government sees an advantage in making and using the invention directly and authorizing others to make, use, and sell product based on the invention for the government, it may also do that.

Private patent owners of inventions made with federal support have no right to exclude any practice in the government-side market. Put another way, owners of patents on subject inventions agree, as a condition of obtaining the right to be owners, that they will not sue any US government for infringement (nor sue in the Court of Federal Claims for compensation) nor sue any company for infringement that is authorized to practice by any US government. When the government practices or has practiced an invention under its government license, it in no way changes the rights that a patent owner on a subject invention has. In fact, things go the other way. If the government does not exercise the license it has to practice and have practiced, then it converts the patent owners patent into a still more valuable asset than was bargained for.

Imagine: If I pay for you to buy a car on the condition that if I need it you’ll share it with me, and I never, ever bother to need it–even to the point that I pay you to drive stuff around for me rather than doing it myself–then you are getting more in the deal than we bargained for–I pay for the car, and you get the car entirely to yourself. In the case of patents on inventions made with federal support, we end up with two cars–one for the private-side market and one for the government-side market. Both can be driven independently. If the government does not drive its car, then the value of the private-side car goes up anywhere that the government-side market is significant.

Thus, if the government does not practice its government-side license, it is donating the value of that license to the owner of the patent practicing in the private-side license. It’s a government give-away, on top of giving away the right to manage the patent in the first place. That’s a sweet deal for the patent owner, but a lousy deal for government and for the public.

Not every invention made with federal support can be readily practiced by the government. Some inventions get combined with other inventions that aren’t federally supported, and the resulting product is covered by other patents that the government does not have a license to. But other inventions can be practiced because the other technology necessary to develop useful products is available–in the public domain, in alternative forms, as a standard. And it is important to note that the government license is not simply to replicate whatever product is on the market; rather, the license is as broad as the claims in the patent, and thus includes all the technology the patent owner has reserved, not simply the bits the patent owner has exploited to create a given product. Thus, there may be other methods, compounds, devices, circuits, and the like that do similar things and which the patent owner has not developed at all for the private-side market and which may work perfectly well for government-side market needs.

Thus, a starting point for dealing with the cost of drugs in the private-side market is to develop the government-side market for these drugs when they have been invented with federal research support. If the government is going to fund research leading to the invention of therapeutic compounds, then the government ought to also develop the resources to exploit those compounds in the government-side market. While the government does not have to pay to take a compound through clinical trials or figure out dosage levels or delivery packages or provide training to physicians, the government does have an outright license to practice and have practiced the invention for the government-side market. It is as if the government is a licensed generic drug manufacturer from the get-go. The private side patent owner knows this, has granted the right. The competition is built into the relationship–but it is only competition in the government-side market.

The government’s way to deal with problems in the private-side market is by means of march-in, which is basically not at all, never. It’s the government license that matters, and it’s time that the government stop wasting its asset and stop donating the value of its license to private patent owners working monopoly deals at monopoly prices. States can complain, but they have at their disposal a broad license and they should being using that license.



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