Bayh-Dole’s nonprofit royalty and license income provisions, thus, appear to have to do with navigating tax issues for nonprofit organizations arising from patent management activities. The effort in Bayh-Dole is to connect patent licensing income and the typical nonprofit exempt purposes of universities, including research and education. If the patent licensing is just “for the money” then the activity may appear not to be related to exempt purposes and end up being taxed or the exempt status of the university might be called into question.
It’s rather odd, though, that federal law should require federal agencies to include such provisions in federal contracts with nonprofits. Why should it matter to the federal agency in contracting for inventions in federally supported research whether a nonprofit exposes its licensing income to unrelated business income taxes? Shouldn’t the issue rather be something having to do with the purpose of the research–and therefore the purpose of any patents on subject inventions–and therefore with the purpose of the money arising from exploiting those patents? That would appear to be the burden of 2 CFR 200.316–the property trust relationship, in which universities are required to be the trustees on behalf of the beneficiaries of the research. So why not require that any income after incidental expenses in the administration of subject inventions go to the support of the beneficiaries of the research? But that’s not what we have.
For another perspective on this odd clause in Bayh-Dole, let’s look at the testimony by Thomas F. Jones, the vice president for research at MIT in 1978. Jones was testifying with regard to a proposal that the IPA program should be made government-wide. Before Bayh-Dole, it was the IPA program’s expansion that was the effort to work around executive branch policy to make inventions in federal research broadly available. Perhaps you can see now the problem that was developing. The IPA program was used to end-run non-exclusive dissemination of inventions. In place of non-exclusive practice was substituted the idea of “commercialization”: that no one will develop an invention for public use without an exclusive position through which to recover the costs of development. Along with commercialization comes the depiction of competing users as “free riders” exploiting the risk capital of another to gain unfair benefits for themselves.
Lost in the commercialization rhetoric, of course, is any concept of how innovation comes about from research–that historically most innovative developments have been the result of activity outside commercial markets–“networked, non-market” stuff, to use Steven Johnson’s term. Use does not necessarily or usually depend on the production of a commercial product. And commercial development of product does not necessarily depend on a monopoly position–though efficiencies of all sorts might come about because there’s a standard, or interoperability. A standard, in its way, is a kind of monopoly, but it is one that people opt into to gain the advantages of common use, not one that people are excluded from in order to maximize income.
For the folks seeking to end-run federal policy on public access to research inventions, commercialization talk created the necessary political cover. The U.S. was falling behind Japan and Europe, and what was needed was new products from all that research. For that, it was important that research inventions were patented, to prevent the Japanese from exploiting the research to create products faster than U.S. companies could do. Utterly lost was the logic–if Japanese companies, say, could develop commercial products based on federally funded research without a monopoly position, why couldn’t U.S. companies also do that? And if Japanese companies were providing commercial products more quickly and at lower cost to the American public than could U.S. companies, then exactly what did the federal government hope to accomplish by encouraging universities to tie up all significant research discoveries in the patent system? It would appear that “commercialization” stands for “delay and pay more” motivated by an odd form of nationalistic angst.
The answer was that many U.S. companies could and did develop new products without a monopoly position. It appears that with the emergence of each new technology, an effort has been made to “corner” the market. This happened with electricity, light bulb, telephone, airplane, and television. In the 1950s and 1960s, attention turned to “wonder drugs”–compounds that could prevent disease or cure disease, or at least make acute conditions chronic (and that was the sweet spot for profits). The pharmaceutical industry was caught up in efforts to corner markets for any new compounds that might show up. When the federal government ramped up its support for medicinal chemistry on the premise that results would be available to all, the pharmaceutical companies fought back and boycotted the screening of compounds discovered in federally supported research. Monopoly or nothing.
Forces in the “deep state” of the NIH worked for years to circumvent government policy and deliver monopolies on compounds to the pharma industry. The IPA program, especially as implemented from 1968 to 1978, was the primary workaround. Under the IPA program, a university entered into a master agreement with the NIH. That agreement allowed the university to patent any invention made with NIH funding, provided that the university then followed various rules with regard to licensing. The rules made it appear that non-exclusive licensing was preferred and exclusive licensing was a last resort, and only for limited times and only on reasonable terms. But it ended up that the nonprofits–mostly university research foundations working for universities–licensed almost entirely exclusively. Lesson: no matter how many publicly laudable options are present in a policy, if there’s even one self-interested pathway allowed, no matter how obscure or put behind multiple conditions, the self-interested pathway is the one that will operate–and may be the one that was intended all along, but for the need for “protections” to get the policy established as law.
Thus we end up with a tension between “commercialization,” with its emphasis on monopolies, profit-seeking, and exclusion of all other use, and “public availability” with its ideas of access and competition. Commercialization was effective in getting the IPA program in place and eventually getting Bayh-Dole passed when the IPA program was shut down–oddly, because it did not do an effective job in promoting rapid commercialization. Apparently, exclusive licenses motivate investment in patent rights but let licensees with a paid-for monopoly take their time in developing product.
Into this context comes Thomas Jones to testify on “The Desirability of Institutional Patent Agreements for Universities.” Jones testifies not just for MIT but also for the Association of American Universities, the National Association of State Universities and Land-Grant Colleges, the National Association of College and University Business Officers, American Association of State Colleges and Universities, and the American Council on Education. Quite the list. Notice the presence of the business officers in the list, and the lack of anyone representing, say, faculty or investigators or inventors–even though the management of inventions involves, first of all, the personal property of inventors.
Here’s Jones’s basic argument:
It’s hard not to question whether Jones has the calculus reversed–universities were expressly forming licensing operations for the money (as the University of California had done in 1960, and as MIT had done even earlier). Technology transfer was the means to get that money, but what was meant by technology transfer was not transferring new capability to use discoveries but rather conducting a trade in patents on those new capabilities. Even if the IPA program resulted in “effective” technology transfer, one still had to consider whether a monopoly trade in patents was better than making research technology broadly available–faster? more efficient? greater impact? lower prices? more competition? more choice? faster development and application?
Things varied by industry practice, according to the Harbridge House report. Some industries were fine without the patent monopolies. Others, like the pharma industry, lived for their monopolies and saw open systems as a threat.
Jones then works through his theme. Universities do basic research and lack the ability to develop products. Inventions are by-products of research, not the objective. To develop research inventions into products takes a lot of money that universities don’t have. Hence, inventions have to be transferred to industry for development before anyone may benefit from them.
Sounds good–except it’s off. The idea that’s presented is that the public benefits from “concepts” developed in university research when there are “commercially viable developments,” which are expensive. What’s left out is that the public may benefit in all sorts of ways that do not require commercially viable developments, or, if commercially viable developments are needed, these are not particularly expensive to implement. The idea that university research inventions must necessarily be “early stage” or “high risk, high return” is not true in the general case, though there are situations when it is of course true. It’s this stating a local, infrequent case as a general one that’s the problem. Jones makes it appear that monopoly commercialization is the only, or best, or primary, or first method by which the public might benefit from inventions made in university-hosted research.
While universities may lack the capacity to develop commercial products, many inventions made in university research can provide public benefits without being made into commercial products, or may be made into commercial products without substantial expense, or may be made into commercial products while also being used by the public, or may be made into commercial products at considerable expense through collaborations in industry so that expense is shared, or made into commercial products at considerable expense without a need for a monopoly position at the outset.
Other inventions may be used directly by any capable person or organization with minimum development, as is the case with many method patents (and with disease assays, for instance). Once the method is taught, it can be used most anywhere.
Even where development is required, the thing that matters then is the improvement. If development is obvious, then the expense is often not consequential. If development is not obvious, then the improvements will also likely be patentable inventions, and an exclusive commercial position can be had from the improvements rather than the original invention. In practice, a broad early patent aims to preclude inventions in improvements and functional equivalents, and thus to control development so that one has to obtain a patent license to practice, to do research, or to use improvements.
Thus, while it is true that some inventions require development, the general case is not true. It just sounds true. And it’s politically expedient to make a case that sounds true if one’s goal is to get the federal government to give up on the idea that subvention research ought to result in results made broadly available to the public for use, quite apart from whether there is commercial exploitation in the form of products to be sold. Industrial use of inventions may be an entirely acceptable result, without the need for product versions. One may go so far as to argue that for many inventions, the pathway to commercial product passes through a stage of widespread industrial and public use. That is, commons before product. And when product, competitive choice in product.
We might also point out that if a “concept” thought up at a university is directed only at a commercial product and that product requires a huge expense to develop, then it’s entirely likely that the concept was not “invented” in university research, but is rather the product of imagination–not something that anyone has been commissioned to investigate. Imagined concepts are not something unique to university research, but rather what people come up with when they are creative. The challenge for this sort of “concept” is that the further one is from commercial practice, the more difficult it is for this sort of thing to make its way into industrial uses or commercial products. More expense, more waste, more learning things the hard way. It may be that there’s a path to commercial success–as a profitable product–but indeed that may be expensive.