The Bayh-Dole secrecy provisions ensure that there will not be public accountability that might challenge the illusion and the practices that take place behind its appearances. Thus, there is no use data for federally supported inventions. There is no licensing data specific to POSIs–patents on subject inventions. There is no way to assess the status of FOIL technologies–fragmented ownership, institutionally licensed inventions. There is no way to match royalties to inventions, nor to show how invention use and royalty payments might be skew. And there is no way to show that royalty payments are being used as required by the Bayh-Dole standard patent rights clause–university administrators may pocket the income and do what they want.
A recent case in California involved public records requests under the California Public Records Act (CPRA). A request was made for records that were created by local government officials using private email accounts. They argued that what they did on private email accounts was not a public record. The California Supreme Court disagreed:
It is no answer to say, as did the Court of Appeal, that we must presume public officials conduct official business in the public’s best interest. The Constitution neither creates nor requires such an optimistic presumption. Indeed, the rationale behind the Act is that it is for the public to make that determination, based on information to which it is entitled under the law.
Open access to government records is essential to verify that government official s are acting responsibly and held accountable to the public they serve. ( CBS, Inc. v. Block (1986) 42 Cal.3d 646, 651.) “Such access permits checks against the arbitrary exercise of official power and secrecy in the political process.” (Ibid.) The whole purpose of CPRA is to ensure transparency in government activities.
If public officials could evade the law simply by clicking into a different email account, or communicating through a personal device, sensitive information could routinely evade public scrutiny.
I’ve added some paragraphing and boldface. This argument from the California Supreme Court is relevant to our discussion of secrecy in Bayh-Dole as well. The purpose of public reporting is precisely to hold in check government abuse of power, by allowing the public to make determinations of what is in its interest, based on information that the public is entitled to under the law.
In the case of Bayh-Dole, we have a law that aims to circumvent public disclosure law in the form of FOIA. Those manipulating Bayh-Dole have aimed to make the law appear to prevent the public, as a matter of law, from being entitled to any information about the use of subject inventions.
If those behaviors and outcomes could withstand public scrutiny, there would be no need for the shrill defenses put up by these organizations and individuals. If those behaviors and outcomes were remarkable, there would be no need for statutory secrecy. Something is deeply amiss.
Within the original Bayh-Dole Act, the one passed in 1980, the reporting requirement was left outside the secrecy requirements for invention reports and patent applications at 35 USC 205 and instead placed within the requirements for a standard patent rights clause to be used in funding agreements at 35 USC 202(c)(5). This is a huge difference in treatment. As a piece of a federal funding agreement, the report requirement permits federal agencies to receive information from contractors who have chosen to step in and use the patent system to promote the use of federally supported inventions for the public benefit. Everything in the network of relationships that a contractor may then make–with inventors, with patent brokers, with companies, and even within their own organizations involving the use of license income–then requires a conveyance of that public purpose, however it might be expressed. Bayh-Dole gives contractors flexibility in expressing that public purpose (which may well have been a defect in the practice of federal agencies in the 1970s), but the reporting provision to be placed in all federal funding agreements had the purpose of allowing the public to see just what flexibility was being used and with what outcomes.
In the original reporting requirement, Bayh-Dole made a distinction between the information that the public should receive and information that was properly privileged and confidential. Any of the information might be regarded as “sensitive” in some way. If a university was not licensing any of its POSIs, that information might be embarrassing and a university might not ordinarily make embarrassing information public. If a university at some point actually has licensed some POSIs, are the licenses almost always exclusive? And if not to a big company then how often to a small company with the purpose of selling the small company to a big company? (And in that latter case, the exclusive license to the small company gets sold and assigned to the big company–as if the license had been to the big company all along). Any of this information might be considered “sensitive” to university licensing programs–“privileged and confidential.”
The reporting requirement in the original Bayh-Dole, however, was not concerned with removing all sensitive information from public scrutiny. Instead, the original 35 USC 202(c)(5) permitted federal agencies to distinguish between sensitive information that should be publicly reported so everyone could assess how contractor flexibility influenced outcomes and sensitive information that might reveal the otherwise confidential operating information of licensees, such as release dates and budgets. In the original clause, all that is indicated is that agencies must contract to have the right to reports on the use of federally supported information, and the efforts at “obtaining utilization.” Bayh-Dole does not require agencies to enforce this provision and actually request reports. It just requires that the agencies to put the provision in federal funding agreements. (If Bayh-Dole wanted to make the use of this provision required, it would have required each agency to make a report–to the President, to Congress, to the public–regarding the status of each federally supported invention in contractor hands; but Bayh-Dole’s original reporting apparatus was swiftly eliminated, permitting three decades of unsubstantiated announcements that Bayh-Dole was an unqualified success–that alone is enough to spark a relatively deep suspicion that Bayh-Dole is rather an unqualified scam).
Let’s push this point further. Bayh-Dole requires a reporting clause in all funding agreements. Bayh-Dole uses language to permit federal agencies to differentiate between contractor sensitive information that should be reported publicly and contractor sensitive information that is privileged and confidential and thus may be excluded from FOIA disclosure under the (b)(4) exemption. That’s the purpose of the use of “may” in the original version: “may be treated as,” not “shall be treated as.” There’s a huge difference behind such an apparently trivial change in usage.
Imagine this reporting clause as itself a kind of public records law regarding sensitive information. For patent work, there are three classes of such information. There’s the reports of inventions and patent applications–that stuff is addressed in 35 USC 205. Then there’s the status of a contractor’s efforts to promote use. That’s one part of the sensitive information in 35 USC 202(c)(5). Then there’s the information received from licensees regarding their performance. That information is normally sensitive within the context of a license agreement between patent owner and licensee. But that information, too, may be reportable, and may be treated as “privileged and confidential” information–privileged because it was intended by the licensee only for the use by the licensor; confidential because it is not information that licensee-companies normally release to the public. The “may” in the original Bayh-Dole reporting provision allows federal agencies to distinguish between these last two categories of sensitive information. The change to “shall” removes federal agency discretion in determining which information should be public and which information should remain a government secret.
Now look at the implementing regulation for the reporting provision, at 37 CFR 401.14(h):
Such reports shall include information regarding the status of development, date of first commercial sale or use, gross royalties received by the contractor, and such other data and information as the agency may reasonably specify.
Here the regulations call out three data elements: status, date of first commercial sale or use, and gross royalties. There’s a good argument that these data elements are intended to be public. The status of development involves whether the contractor has licensed a POSI, and if not what efforts the contractor is making to do so. There may be sensitive information involved, but the fundamental question of what the contractor is doing should not be. The federal agency must determine what information should be exempt from FOIA.
Example: a university reports that it has not licensed a POSI but that an offer to license is posted at the university’s web site. None of this information is confidential–it is posted at the university’s web site. The fact that the university reports this information to the government, then, too is nothing that is excludable from FOIA, since this information is not of the kind ordinarily kept secret by the university. But in the changed Bayh-Dole, “may” becomes “shall” and the information in this report is required to be excluded from FOIA under (b)(4). That is, what’s excluded is the public fact reported by the university–something that FOIA does not permit to be excluded under (b)(4). The information fails the basic test–it is not confidential or privileged information. But Bayh-Dole requires agencies to treat it as such, and the implementing regulations go further and require federal agencies to promise to keep the information secret, even as the university makes the information public. The report to the government is secret, so we cannot see what the university has reported and verify whether the report is accurate, nor can we then see across the reporting for each subject invention held by the university what the university is doing with patents to promote the use of each invention. This scheme violates FOIA even while it cites FOIA. Bayh-Dole requires federal agencies to contract to keep from public disclosure information that is not confidential or privileged–as if that information were confidential and privileged.
Thus, there is basic information on a university’s efforts to “obtain utilization” that certainly is public. We might consider the case, then, where the university has done nothing–the invention is not used (or the university is so indifferent it doesn’t even know), there has been no marketing of the invention (or if there has, that marketing has lapsed and perhaps is out of date), and there’s no licensing activity. Now none of this information is necessarily public. The test might be if someone called the university’s licensing office and asked whether a given invention was available for licensing. Under the standard patent rights clause, the university is not obligated to answer! But if it does answer, then perhaps the information is not confidential. But a standard answer is to require the requester to sign a non-disclosure agreement, so perhaps a university might try to hide, even, its lack of doing anything with a subject invention.
If a federal agency requires a report that involves this information, is the information merely “sensitive” or is it “privileged and confidential”? Who can say, other than a judge at the end of seven years of multi-million dollar litigation? That’s a defect in the law, since it does not have to be this way. I’ll argue (perhaps you might, too) that this information, no matter how sensitive, should be required to be made public. Call it an essential characteristic of the public covenant that runs with a subject invention when a contractor chooses to take title to an invention made with federal support–the contractor must report that it has ownership of the invention, that it has filed patent applications, that it has or has not made efforts to “obtain utilization.” And this information–without even revealing the details of the invention (“a nonconfidential summary”) or the patent work–is never exempt from FOIA, no matter how embarrassing it might be to a university, no matter what it might reveal about university patent practice or compliance with the patent rights clause.
There may well be information involved in the status of a subject invention that is indeed “privileged” or “confidential.” A university might report problems in patent prosecution that are delaying interest by potential licensees–that information, exchanged between university patent counsel and the PTO, is privileged and excluded by (b)(4) from FOIA disclosure. A university might receive feedback from a potential licensee who has received an evaluation license (with non-disclosure requirements) that they can’t replicate a feature of the invention. That information might well be confidential, as companies aren’t in the habit of releasing information about new technology they are evaluating under a non-disclosure agreement. A federal agency operating under the “may” must determine what information is what information is exempt from FOIA disclosure. That same agency operating under “shall” is required to promise to treat all information as exempt from FOIA–even the fact that a university has offered evaluation licenses or that the university is involved in an extended patent prosecution.
We can consider the two other data elements called out by the standard patent rights clause in a similar fashion. The date of first commercial use or sale would not appear to be one that requires secrecy, especially if that date is combined with the requirement in the definition of “practical application” that the use must be “established” and that the benefits of that use must be “available to the public” and “on reasonable terms.” For there to be “utilization,” there must be, as far as the apparatus of Bayh-Dole is concerned, practical application. Before practical application, all we have are “efforts at obtaining utilization.” Those efforts have both a public element and a confidential or privileged element. Those confidential or privileged elements involve the patenting process and information exchanged with potential licensees or with licensees working to develop products or adapt an invention for their own use. Thus, one may report for public consumption that an invention has been licensed for evaluation, and that fact is and should be non-confidential, while the name(s) of the companies involved may be confidential because those companies don’t normally release such information and might suffer competitive harm if the government released this information.
But at the point of practical application, certain information necessarily must become public. That’s the only way to “establish” that there is use. But more–if benefits are available to the public, then the public must necessarily know of these benefits. We are not facing a situation in which the primary mode of providing benefits to the public is that the public does not know that such benefits are available, or that the public is benefiting from the use of the invention but with care to ensure that the public will never know that they are. That’s nutso stuff. Further, if the public benefits from the use of a subject invention, then for practical application to be met, the benefits must be available on reasonable terms. That is, the terms, too, must be public. We are not to read Bayh-Dole as requiring that the terms are reasonable but the public cannot know these terms, or that even if they public does not know these terms, the federal government must keep what it knows about these terms–what the public knows–as a government secret. That, too, is nutso stuff. Or, more technically, such practices fail to comply with the express provisions of FOIA.
Under a “may” provision, a federal agency must distinguish the sensitive information that is public or must be made public (if only by public request) from the sensitive information that is privileged or otherwise ordinarily not provided to the public by the non-federal entities that possess that information. Under a “shall” provision, no such determination is permitted. All information received is to be treated as secret, as exempt from FOIA. The effect is to efface the operation of Bayh-Dole, to prevent anyone from obtaining information from the federal government that should be public, that often is public anyway, and that might confirm or challenge claims that Bayh-Dole has been successful in allowing universities (in particular) to use the patent system to promote the use of subject inventions.
Consider, too, “gross royalties.” It’s not at all clear what “gross royalties” might mean. In a license arrangement, a licensee might be required to report “gross sales” as part of showing a royalty calculation (to adjusted gross sales, to net sales, to whatever). But that’s not “gross royalties.” Further, a university may receive “royalties” under a patent license. A “royalty” is, essentially, any consideration for the grant of the license. Thus, a royalty may be financial or it might involve non-financial consideration, such as a license to related technology or special performance, such as offering developed product to underserved populations at a deep discount or for free–serving a public interest of importance to the licensor may be a substantial form of consideration. Bayh-Dole, however, distinguishes royalty and “income” (35 USC 202(c)(7)):
(B) a requirement that the contractor share royalties with the inventor;
(C) except with respect to a funding agreement for the operation of a Government-owned-contractor-operated facility, a requirement that the balance of any royalties or income earned by the contractor with respect to subject inventions, after payment of expenses (including payments to inventors) incidental to the administration of subject inventions, be utilized for the support of scientific research or education;
Royalties must be shared, but there may also be “income earned” and that income earned is, clearly, distinct from “royalties.” “Income earned” would appear to be broader and could encompass money received from selling products based on a subject invention rather than licensing others to do the selling. Thus, a university might sell a research tool based on a subject invention and thereby earn income, but that income would not be a royalty because there is no license involved, and hence there’s no consideration for a license, and hence nothing called a royalty. Interestingly enough, “income earned” then does not even depend on the presence of a patent. A university could earn income from the subject invention without having a patent–just a patent application, even perhaps one that’s finally rejected on whatever basis. Then who ever is paying–buying or paying on the prospect that a patent will issue, but no patent ever does–is not paying a royalty. It’s just “income earned” with respect to a subject invention, not with respect to the patent on the subject invention, the POSI. One might go so far as to view sponsored research funding, to the extent that sponsored research can be construed to be provided “with respect to” a subject invention (such as to develop the invention or explore variations or properties of the invention), to be within the scope of “income earned” but not shared with inventors because not a royalty.
Thus, “gross royalties” is tied to the operation of licenses, not other disposition of subject inventions. We might posit that “gross” here means “in aggregate”–that is, for all licensees of a given POSI. The aggregate income would not reveal the particular payments of any given licensee unless there were only one licensee. Even if the usual university practice were to license non-exclusively, gross royalties might be a problem at first, if there’s only one licensee in a year’s reporting. Or “gross” might mean “in aggregate for all subject inventions.” That would only be a problem if a university had licensed only one POSI and thus the income could be traced to a single company. In cases such as these, under a “may” disposition, a federal agency could withhold financial information under FOIA’s (b)(4) until there was sufficient additional royalty activity that the particulars of any given licensing relationship could not be recovered from the aggregate “gross” royalties.
Universities take gross royalties to be a sign of “success.” There’s some merit in that, for ordinary patents, certainly. But for subject inventions, which run with a public covenant that places use with public benefits above profits, even monopoly priced profits, then gross royalties may be a warning sign. If the profits are outlandish, it is possible that the public benefit is not on reasonable terms.
Think of patent profit two ways. In one way, whoever needs the product must pay a monopoly price. The few who need the product (perhaps it is a therapeutic drug and their lives depend on it) will pay huge sums for it. The profits are immense. For an ordinary invention, this is wild success exploiting a monopoly position for whatever the market will bear, and where the market is one for people’s lives, the profits truly will be sweet. But those huge sums are also a sign that the terms are not reasonable. “Reason” here does not mean “whatever the market will bear.” “Reason” means “as a matter of public policy that places access ahead of market concerns, so long as those providing the product recover their costs with a reasonable return for their risk capital.” Again, “reasonable return” is not “any monopoly return short of anti-trust” or there would be absolutely no reason to stipulate “reasonable terms” in the law.
Now think of patent profit another way. A product is developed and provided to the public at very low cost. Many people acquire the product, happy to pay the asking price. The profits are immense. The monopoly position is still exploited–but now rather used to record transactions rather than to limit transactions in favor of a higher price. If others could not provide the same product at a lesser price point or with better quality or with features adapted to particular uses otherwise not well served, and the product is available to all, then the terms are “reasonable” wouldn’t you say, even though there’s a POSI monopoly and there are huge profits?
To ascertain whether a given invention has been managed in the public interest or for a predominantly private interest, it is helpful to know something about the “gross royalties.” One cannot reason directly from gross royalties to success or corruption, but gross royalties provides a starting point for a public evaluation of what Bayh-Dole is creating (or encouraging or looking the other way about).
One might extend “gross royalties” to ask universities to report the share of those royalties that go to inventors, the share to the other “costs incidental to the administration of subject inventions.” That information, especially at public universities, in aggregate, is generally reportable under state records laws. Thus, the standard patent right clause identification of “gross royalties” might be directed at the proper handling of funds by the university contractor and not be at all about threatening to reveal the underlying financial operating model of POSI licensees. Again, the point: under a “may” regime, federal agencies have a great deal of information that they may gather and report that would indicate how Bayh-Dole was operating by showing how contractors were handling each and every subject invention that they claimed. Under a “shall” regime, which was imposed in 1984, we will never know. While agencies may collect information, they are required to promise not to disclose that information, even when that information is otherwise disclosable under FOIA, and even when that information is vital to any assessment of whether Bayh-Dole is operating as intended or claimed or whether some university invention practices are more effective than others.
This is a study in ten parts. Here are the links.