AUTM a few weeks ago pointed favorably to a description of the Bayh-Dole Act posted by the University of Pittsburgh. Let’s have a look, then.
The post is titled “What It Means for Technology Commercialization.” While “It” is ominous in these Steven King days, let’s say “It” is the clownish Bayh-Dole Act and not something yet more horrible. We will take Pittsburgh’s AUTM-endorsed account of Bayh-Dole section by section.
The Bayh-Dole Act is a federal law enacted in 1980. This legislation, cosponsored by Senators Birch Bayh and Robert Dole, enables universities, nonprofit research institutions, and small businesses to own, patent, and commercialize inventions developed under federally funded research programs within their organizations.
So far, pretty good. Bayh-Dole was enacted in 1980–into effect in mid 1981. And Bayh and Dole did co-sponsor the bill in the Senate. And Bayh-Dole does, indirectly, “enable” universities and others to own inventions made with federal research funds. It’s an odd use of “enable,” however, and one has to know just enough fake history to have the courage to use “enables.” The fake history is that prior to Bayh-Dole the federal government in its contracting claimed ownership of all patentable inventions made with federal support FAKE. If a university got assignment of an invention, then the government made the university assign the invention to the government anyway IT HAPPENED A FEW TIMES. The government stockpiled patents and refused to license them to anyone FAKE, ensuring that the public would not benefit from any of the great research universities were otherwise doing FAKE. Bayh-Dole came along and reversed all this FAKE, by “enabling” universities to own inventions made with federal support.
The actual history is that some agencies of the federal government from the 1950s on allowed universities to own inventions made with federal support–but most universities refrained from doing so. When an invention made by faculty was to be owned, all but a handful of universities referred inventors to invention management agencies, especially Research Corporation, often via a university affiliated “research foundation.”
Some agencies, such as the Department of Defense, allowed inventions to be owned by contractors; other agencies–the NIH and later the NSF–used a master agreement, an “Institutional Patent Agreement,” with selected universities. Thus, many of the universities receiving federal funds for research could own inventions made with federal support–provided they first obtained that ownership from their inventors. Notably, under the IPA program, the NIH reviewed the university’s technology transfer program before allowing a university to join the program. Once a university joined, the master agreement required the university to have a patent agreement with each research employee under which the employee promised to assign to the university each invention made with NIH support that the university had chosen to file a patent application on. Just to be clear–the university could require assignment of the NIH-funded invention only after it had made the decision to spend the money to apply for a patent. Disclose, decide, own; not disclose, own, fuss around deciding; and certainly not the current rage of own; disclose; fuss around deciding. Pillage, then burn. Not burn, then pillage. You get the idea now.
What Bayh-Dole actually does–directly–is to require each federal agency to use a standard patent rights clause in all research funding agreements with universities unless the agency can justify an exception. Bayh-Dole applies to federal agencies and to a certain class of inventions. Bayh-Dole does not apply–directly–to universities or inventors.
The formal name for the act is the “Patent and Trademark Act Amendments of 1980,” and it created a uniform patent policy among the federal agencies that fund research.
The bold is in the original. Giving the formal name for bill in which Bayh-Dole is included is rather arcane. The Patent and Trademark Act makes various changes to federal patent law. Bayh-Dole is just one part of the bill. Why not add PL 96-517? Or 94 Stat. 3015? Or the enabling regulations, which matter way more to universities, 37 CFR 401, and especially the standard patent rights clause at 37 CFR 401.14(a)? It’s as if whoever is writing this piece really has no clue about Bayh-Dole–which, as we will see, is the case. This is just the opening warning.
The idea of a “uniform” policy makes it sound like there had been no uniform policy. But there was a uniform policy, first established by President Kennedy in 1963 and continued with only slight modifications by the Nixon and Carter administrations. Here’s the Kennedy statement:
The prudent administration of government research and development calls for a government-wide policy on the disposition of inventions made under government contracts reflecting common principles and objectives, to the extent consistent with the missions of the respective agencies. The policy must recognize the need for flexibility to accommodate special situations.
Bayh-Dole superseded this uniform policy with a different uniform policy. The Kennedy policy emphasized agency flexibility. Bayh-Dole emphasizes arbitrary requirements, and embeds those requirements in federal patent law, not in executive branch policy. The problem with flexibility is that some federal agencies did not require assignment of inventions to the government; others used a master agreement; others required assignment to the government; and some–including NASA and what’s now the Department of Energy–operated with laws that dictated ownership of inventions.
For federal funding, invention ownership was not any different from the range of possibilities one encountered with private funding for university research. We might say that federal agencies matched their invention ownership positions to their public missions. Some agencies did a better job with this matching than others, but the problem was not that there was no “uniform” approach. There was. And the best advice of the day, the Harbridge House report, made clear that no single approach to inventions would meet the needs of federal agencies or be acceptable to industry, where companies varied across a spectrum with regard to the role of patents–some ignoring patents and others using patents in various ways, and some entirely focused on patents.
What the NIH and university patent brokers wanted, however, was some stamp of approval that permitted inventions in the area of medicinal chemistry to be patented and handed off as monopolies to pharmaceutical companies. Since this practice was not supported by HEW, of which NIH was a part, NIH patent counsel and university patent brokers had to make it appear that institutional patenting and monopoly licensing was not only the right thing to do for important discoveries in pharmaceutical science but also for all federally supported inventions. A special special case (new drugs as a special case of those special cases in which the only way remaining for the public to benefit from an invention is that if private risk capital is attracted by the offer of a limited monopoly) had to be made to appear to be the general case. Thus, “uniform” was created as a political cover to make all federal agency invention management practices adopt the monopoly model that university patent brokers and the NIH were using to circumvent HEW public purposes. “Uniform” in reference to Bayh-Dole means “arbitrary.” It also means “make dealing in patent monopolies appear uniformly virtuous.”
The consequence of an arbitrary policy is that as situations vary, the policy becomes inequitable. Under the prior IPA program, for instance, a university had to demonstrate that it has a viable technology transfer program before it is allowed to participate. Under Bayh-Dole, a university does not have to demonstrate anything about its program. It can have an awful program and still it has a chance to own inventions. What value is there in a “uniform” policy that turns a blind eye to atrocious university invention management?
Clearly, from the point of view of administrative slumlords running atrocious invention licensing programs, such “uniformity” is highly desirable. We might posit that those universities that make the biggest deal about the importance of Bayh-Dole run the worst licensing operations. If they ran great licensing operations, Bayh-Dole would not figure in their public communications.
There is almost nothing a university has to do to comply with the standard patent rights clause authorized by Bayh-Dole: delegate invention responsibilities to research personnel, educate them on the importance of timely reporting of inventions, report to the government those inventions that are reported to the university, flow down the patent rights clause in subcontracts. Remarkable as it may seem, none of these obligations are to be found in Bayh-Dole. They are all local to the standard patent rights clause. Pulled as it were from the posterior cortex of the same person from the NIH who drafted Bayh-Dole and then tried to finish the work in the implementing regulations.
Whatever additional obligations that arise happen only after a university has acquired ownership of an invention made with federal support. Then a pile of mostly useless administrative posturing breaks loose–nothing that advances technology commercialization: time lines to file patent applications, and requirements to report on use (except the agency doesn’t have to require reports and the reports are government secrets), notice of government funding and rights placed in patent applications, a license to the government, and fussiness about exclusive licensing (which agencies can waive), and about assignments (which agencies can waive), and about licensing to small businesses (which agencies can waive), requiring licensing (which to do agencies have to march naked through blackberries and so don’t), and how to manage royalty and other income. The complications of the standard patent rights clause show up only after a university acquires ownership of an invention made with federal support. And those complications have next to nothing to do with technology commercialization. Why would any sane inventor assign a federally supported invention to a university? And why would any sane university administrator think that university ownership of such inventions is a really good thing for technology commercialization? Well, apparently some university administrators think this arrangement is really keen. We might draw some inferences about their state of mind.
While Bayh-Dole establishes “uniform” rules regarding what a federal agency must require in a default patent rights clause, agencies have broad discretion at each point to waive the rules, decline to act, and generally ignore the rules. Thus, there has never been a successful march-in proceeding, despite thousands of university inventions that have never been licensed or used. The U.S. manufacturing requirement can be waived. The small business preference has no consequences. Assignments of inventions can be labeled exclusive licenses and agencies don’t care, despite rules. Universities can use royalty income pretty much however they wish, despite the rules on cost recovery and use of the balance after expenses.
Criticisms of the previous policy prompted this change. Congress perceived the need for reliable technology transfer mechanisms and for a uniform set of federal rules to make the process work.
The odd thing in this is the reference to “the process.” There is no process identified in Bayh-Dole. The implication here is that Congress endorsed in Bayh-Dole the practices that universities deploy today to own, market, and sometimes license patents. That is, Congress endorsed the practice of institutional ownership of inventions, to be licensed as monopolies exclusively to favored companies, so that the university could participate in the profit from higher prices and reduced competition provided by patent positions. That, in a nutshell, is “the process.”
The federal government’s inability to effectively commercialize technologies derived from federally funded research resulted in hundreds of valuable patents sitting around unused.
Now comes bogusness concerning the federal government’s patents. According to the Harbridge House report in 1968, some federal agencies had a nearly 100% success rate in commercializing inventions–notably the Department of Agriculture. The NIH’s own commercialization rate was 23% for biopharma inventions–comparable with the best rates reported by university-affiliated invention management firms and 5x the rate that those firms reported for federally funded inventions under the IPA program. When questioned by Congress on university lack of success in licensing, university officials admitted that they had not done all that well, but (brightening) that there was great potential for them to do much better. At the hearings on the bills that would be merged to become Bayh-Dole, university advocates recited a single report that there were 28,000 government patents that had not been licensed. What those advocates did not include was that most of those patents were defense-related and that the contractors had declined to take ownership of the inventions, even though Department of Defense contracting policy allowed them to do so.