Sixth layer: Government rights
Seventh layer: Enforcement of patent rights clauses
We must also consider the flip side of contractor non-compliance with Bayh-Dole’s standard patent rights clause–federal agency enforcement of the patent rights clauses, waiver of these clauses, and federal agency indifference. Federal agency enforcement, waiver, and indifference then form the sixth and seventh layers of Bayh-Dole.
Bayh-Dole the statute is largely complied with, because that’s a matter of federal agency compliance. But Bayh-Dole is a law that establishes the terms of federal contracts. Bayh-Dole has its effect through those contracts. If the patent rights clauses are not enforced, then Bayh-Dole fails in its purpose, even though federal agencies comply with the law by including the patent rights clauses required by Bayh-Dole if they don’t bother to enforce those patent rights clauses. Bayh-Dole does not require federal agencies to enforce the patent rights clauses. Talk about a gaping assumption made by the law. Once the agencies have properly placed the patent rights clauses in their funding agreements, their obligation under Bayh-Dole is apparently over. They don’t have to request reports on utilization, they don’t have to take ownership for failure to disclose or file patent applications, they don’t have to march-in to protect the public, they don’t have to require U.S. manufacture for exclusive licenses in the U.S. They don’t have to act on the rights they have under the government license to practice (make, use, and sell) and have practiced.
We may distinguish four separate areas in which federal agencies can avoid the bother of enforcing the patent rights clauses: reporting, audit, waivers, and indifference.
Bayh-Dole requires a provision that gives federal agencies the right in any funding agreement to request reports on the use of subject inventions. But Bayh-Dole does not require federal agencies to request the reports, does not stipulate what must be in the reports (unlike Stevenson-Wydler, by contrast), and purports to exclude all reports that are made from public disclosure. Thus, simply by not asking for reports regarding the use of inventions, federal agencies can ignore a large swath of the substance of Bayh-Dole–the practical application of subject inventions, the use of the patent system to promote the use of inventions arising in federally supported research or development, free competition and enterprise, and the like. If a federal agency is not aware of nonuse or unreasonable use, then it has no basis to consider using march-in procedures to protect the public.
Similarly, even when a federal agency receives reports, it has no independent means to verify the accuracy or completeness of the reports–or even their truthfulness. Without regular audits of university practice under the standard patent rights clause, federal agencies have little means for establishing whether a contractor is compliant or not. Thus, without audits specific to the patent rights clause, conducted by auditors who have a full working knowledge of the clause, there will be little basis for federal agencies to question university practices and require changes to that practice to bring universities into compliance. As the GAO report on university practices under Bayh-Dole put it,
The agencies we visited relied on the universities to ensure that all federally funded inventions were meeting the requirements of the law and regulations.
Nothing appears to have changed in the intervening two decades.
The agencies’ monitoring activities consisted largely of collecting and recording the information the universities provided. The agencies generally did not have data for some areas, such as whether the universities were giving priority to small businesses in licensing or how they ensured substantial domestic manufacture under exclusive licenses.
Bayh-Dole operates on a “don’t ask, don’t tell” model. This is what the GAO report calls “decentralized.” Federal agencies are good at collecting paperwork, and universities are happy to comply. But as for the substance of the law–ownership, delegation, patent monopolies, and money–the federal agencies simply don’t have the data because they don’t have to ask for it, don’t have to reveal it when they do get it, and don’t have to act on it even then.
But there’s more to Bayh-Dole. Within Bayh-Dole’s provisions are a series of waivers that federal agencies may use to ignore non-compliance. If a contractor fails to disclose a subject invention or fails to give notice to elect to retain title, a federal agency does not have to take ownership of the invention. The agency has the right to “request” ownership, and has that right for 60 days after the “discovery” of the failure. If the federal agency takes no action, the obligation to report or elect to retain title is simply waived.
Similarly, if a contractor fails to file a patent application or chooses not to maintain or defend an issued patent, a federal agency has no obligation to request assignment of the invention. That is, if licensing patents exclusively as the only way to attract private capital for development is Bayh-Dole’s reason for existence, then federal agencies sure don’t have much by way of regulatory requirement to protect that opportunity. And even if the government does take title to an invention, unless the contractor has failed to disclose the invention, the contractor retains a non-exclusive license that includes the right to sublicense everyone the contractor had obligations to prior to the start of the federal grant. So where then is the monopoly patent right so necessary to attract to risk capital?
Federal agencies are given the right to request reports of invention use from contractors, but that same provision (h) does not require federal agencies to request the reports. Agencies do not have to request reports, and don’t have to request specific information on anything when they do request reports. Reporting, it would appear, is not an essential element of Bayh-Dole, either. Since Bayh-Dole attempts to exclude reports from FOIA (there’s a question whether Bayh-Dole has legally done so–it appears that Bayh-Dole’s requirement does not comply with FOIA), we don’t know, actually, what agencies have requested reporting, what those reports contain, and what federal agencies have done, if anything, to follow up. If the GAO report from 1998 is any guide, federal agencies have done nothing more with the reports than file them.
The preference for United States industry (i) is set up in Bayh-Dole as the most important provision in the patent rights clause. Section 204 starts out with a “Notwithstanding any other provision of this chapter”–that is, Section 204 takes precedence over all other elements of Bayh-Dole, just as Bayh-Dole takes precedence over nearly all other laws regarding inventions. Senator Bayh introduced his bill with the premise that Bayh-Dole was needed to spark American technology development and manufacturing. Bayh-Dole’s statement of policy and objective calls for the patent system to be used to promote inventions made with United States industry and labor. We might expect Bayh-Dole to insist on American manufacture, or at least that Americans have the opportunity to manufacture. But that’s not the case. The Section 204 requirement narrow (it applies only to exclusive licenses to use or to sell in the U.S.) and weak (it requires products to be “manufactured substantially” in the U.S. without any guidance on what “substantially” means, and can be easily designed around by relatively uncreative use of assignments and licensing). Furthermore, all that is required is a letter of assurance from a licensee on the matter. Contractors are not required to enforce the requirement.
Bayh-Dole permits federal agencies to waive the requirement based on either a contractor tried and failed to find a U.S. manufacturer or didn’t bother to try at all. There must be enough requests for waivers of section (i) of the patent rights clause that the NIH has set up a web site devoted to processing waiver requests. Despite all the ballyhoo about the importance of U.S. manufacturing, it turns out that there’s almost nothing about section (i) that anyone has to enforce. Contractors and agencies have to make a show of the paperwork, but don’t have to actually do anything substantive.
Consider as well march-in. Again, Bayh-Dole devotes a section (203) to march-in. March-in is taken up in the standard patent rights clause at section (j). Again, however, federal agencies do not have to march-in. Nor can any member of the public–even inventors–force an agency to initiate a march-in proceeding. In addition, the implementing regulations add a substantial apparatus to the march-in procedure that makes march-in unworkable–formal determinations, opportunities for contractor testimony, opportunities for appeals and litigation. Howard Bremer bragged after the implementing regulations for march-in were issued that he worked to sure they would not operate. And Bremer did an excellent job, since march-in has never been successfully used. Thus, although march-in is recited as a sure protection of public interest, despite rampant nonuse of subject inventions–universities have acquired over 50,000 patents citing federal support, most of them unlicensed and many of those licensed undeveloped–no federal agency has bothered to march-in. Similarly, federal agencies have declined to march-in based on unreasonable use, such as exploitation of public health issues by using patent monopolies that nonprofits have failed to break up to promote free competition and achieve public benefits available on reasonable terms.
If we look at the nonprofit-specific requirements, we see that federal agencies can waive the restriction on assigning inventions to organizations that don’t have a primary function in managing inventions, though they cannot waive the requirement that the nonprofit patent rights clause goes with the assignment. But as we have seen, most of the patent rights clause that matters is waivable anyway. Similarly, federal agencies can waive the small business preference at (k)(4). The Secretary of Commerce is reduced to requesting changes to a university’s licensing program–there’s not even a provision for granting small businesses access to inventions when a nonprofit contractor fails to comply with either attracting small businesses or giving preference to small businesses in licensing.
Finally, we may look at the government license in subject inventions to practice and have practiced. This is a broad grant of rights–“to practice” was defined in executive branch patent policy for two decades to mean “to make, use, and sell.” Further, the “Government” was defined to include state and even municipal governments. Nothing in Bayh-Dole counters the idea that any other scope was intended. For all that, there’s nothing in Bayh-Dole that requires any federal agency to act on its broad license to practice and have practiced.
Thus, there’s no requirement in Bayh-Dole that federal agency must grant a right under its government license for companies to sell product to meet federal obligations, even when federal and state governments spend billions on prescription drugs invented with federal support that are priced ten times higher than necessary to make a “reasonable” profit on their sale. The government license to practice and have practice is entirely unrelated to march-in. If the government routinely exercised its right to practice and have practiced–including authorizing companies to make and sell product on behalf of the government–then companies holding exclusive licensees (as assignees) to subject inventions would have to bring their prices down if they wanted to compete in the government marketplace as well as in the private marketplace. But Bayh-Dole does not require federal agencies to act, and in over thirty years, there’s little evidence (is there any?) that federal agencies have exercised the rights that Bayh-Dole provides for them.
Thus, in addition to substantial university non-compliance with the standard patent rights clause, federal agencies can waive almost all substantive elements of the patent rights clause or simply choose to fail to act, and even to choose to fail to come to know.