Illusions of Bayh-Dole: patent blockages and incentives

In 1979, when S. 414 was introduced by Senators Bayh and Dole–later much of the language of S. 414 would form the core of the Bayh-Dole Act–Senator Bayh made the following claim:

Some 30,000 government-owned patents are piled up awaiting takers. To that extent, the national economy is not being enriched and utilization is forestalled. It is a baffling situation until one realizes that the blockage occurs largely in the government’s patent policy.

Senator Bayh’s argument here is problematic. Most of those patents, it turns out, were for inventions made by defense contractors who had declined to take ownership of the inventions. Furthermore, the patents on these inventions by the government did not need “takers.” The federal government generally did not need to grant licenses–it was indifferent to the license as a means to generate revenue, and it was not prepared to sue as infringers American citizens and companies who were using these inventions anyway. The reality was, the federal government did not “forestall” the national economy by taking up all these inventions and patenting them.

For that matter, the 30,000 patents (28,000, really) represented only a tiny portion of the total number of patents issued. Take 1978, for instance. The USPTO issued about 66,000 utility patents. The federal government received 1,225 of them–just under 2% of the total. But this 2% we are to believe, “forestalls” the national economy.

Now here’s the thing. As best I can tell from searching the USPTO data base, universities and related institutes and foundation now hold some 51,000 US utility patents that cite federal funding, and over 120,00 patents overall in the Bayh-Dole era (1981 to the present). They don’t publish their licensing figures often, and almost never their commercialization rates, but it would appear from recent Stanford and University of California numbers, commercialization rates are below 5% at the best performing programs, and perhaps closer to UC’s estimate of 0.5% of inventions. While it is difficult to match up inventions, patents on those inventions, practical application, and commercial products, we are on pretty good ground concluding that universities are not doing any better than what Senator Bayh accused the federal government of doing. Senator Bayh reports the government’s licensing rate as about 4%. If the government’s 30,000 patents in 1979 were forestalling the national economy, then the 51,000 patents on inventions made with government support held by universities and other nonprofits must be doing much the same thing.

Of course, if Senator Bayh’s argument is dismissable as silly political bluffery, then sure, perhaps the universities and nonprofits are offering a grand benefit–but then there is also no basis for Bayh-Dole and we should be rid of it (and we should).

Senator Bayh attributes this “blockage” to government patent policy. But it could not have been true, given that most inventions owned by the government were made available to all–including their inventors, the institutions and companies that hosted the federally supported research, and most anyone else. The patents held by the government did not block the economy. But patents held by universities and other nonprofits–organizations ready and willing to sue for infringement, demanding “market rates” for their patents–most certainly do block practice. Furthermore, holding patents back in the hope of a future paying exclusive licensee means that whole areas of new technology must be set aside by industry until the patents have expired–or industry must work to undermine or circumvent those inventions to avoid patents to which most companies will never receive a license. That’s the necessary effect of having a fixation on exclusive licensing rather than non-exclusive licensing of patents. If the past government patent policy offered the faint scent of “blocking” the national economy, what a stench Bayh-Dole has been as government patent policy.

We aren’t quite done. Senator Bayh argues that without patents, private entities won’t use government-supported inventions:

The view prevails that if rights to the discovery were released to private developers on an exclusive basis unreasonable private enrichment could occur. There is scant evidence in support of these apprehensions, but the doctrine is riveted into government’s thinking. The effect is that the market incentive to develop government-financed discoveries is circumscribed and inventions are isolated from normal risk-taking and pursuit.

The concern in the Kennedy patent policy was not unreasonable enrichment but rather interference in government activities including choice of contractors; the inappropriateness of the government encouraging for-profit markets that exploit public suffering or hazards; the inherent inequity of granting a monopoly to a company selling product required to comply with government regulations. Under the Kennedy patent policy, the emphasis was on attracting private risk capital to bring an invention to the point of practical application faster than would otherwise be possible, and permitting a reasonable recovery of that risk capital for the effort. Again, think toll bridges, but don’t allow private exploitation where the bridge affects public health.

Bayh argues that patents represent “market incentive” to develop inventions. Perhaps this is so, but the patent system’s Constitutional purpose is to promote the “progress” of the useful arts, and the fundamental exchange is publication of the invention, for which a right to exclude others is granted for a limited time. When the government issues a patent to itself, it has met the expectation of publication and so does not need to exclude others. In the patent system, it is the publication that serves the public benefit. The incentive of exclusive patent rights is offered so that there is publication, not as an incentive to develop the invention. In fact, in US patent law, there is no working requirement. An inventor has no obligation to “develop” any invention, nor to sell product, nor to license the invention for the use by others. In US patent law, an inventor may sit on a patent until others stumble on the same thing and then sue those others for infringement–patent trolling, as it is called. As far as I can tell, it is rarely the case in industry that a new product is chosen for development because there is a patent monopoly available on some bit of it. Yes, among speculating investors, a patent might carry some strange sign of monopoly. But Senator Bayh does not argue that his bill is designed to offer public subsidies to speculators, or that speculation is the thing that is lacking in the national economy.

Again, it is a strange sort of argument, but one that has been repeated. The folksy way this is put is “what is available to all is used by none.” It’s nonsense, but it must sound good, given how often it shows up. In this light, consider Senator Dole’s remarks on the same day, with regard to S. 414. Senator Dole worries that the United States is losing out to foreign competition.

The deterrent factor that the Government places in the path of American business–the refusal to grant patent rights–is removed in the case of foreign firms whose government policies, by establishing a strong collaboration with the private sector, guarantee the development of inventions to which equal access is insured through our Freedom of Information Act. As a result, American research is made available to foreign and domestic firms alike.

This is a bizarre argument, other than its place in an exercise of political bluffery. The patent system publishes the details of inventions for all. It does not matter whether the federal government owns that patent or the federal government insists that a university must own the patent–foreign firms gain access to the details of the invention in either case. FOIA is simply a non-issue. But Senator Dole’s argument here does provide an explanation for Bayh-Dole’s emphasis on secrecy, and in particular 35 USC 205, to prevent those foreign firms from gaining access to inventions before a patent issues. Of course, university-based inventors have routinely published their work well before patent applications are ever filed, resulting in the loss of foreign patent rights anyway. Unless Bayh-Dole took steps to prevent such publication, Senator Dole’s worry about sharing research inventions with foreign companies is useless. His worry has nothing to do with patent policy.

But think about the logic going on here. Foreign companies are benefiting from the mere publication of inventions made with federal support. These companies, apparently, are prepared to develop these inventions without the need for patent monopolies. Yet American companies will not develop these same inventions unless they are given the “incentive” of patent monopolies. How can this be? What is it about non-US companies that allows them to develop inventions into commercial products and thus threaten US leadership in technological innovation at the same time that US companies refuse to develop any new products unless they hold monopoly patent rights? It makes no sense because there is no sense to the argument. It’s another political bluff.

Under the Kennedy patent policy, the sensitive area had to do with how the federal government enforced patent rights in foreign jurisdictions. What should the federal government’s policy be with regard to suing infringers in other countries under foreign patents obtained by the federal government? Here’s how the Kennedy patent policy handled it:

Where the government may acquire the principal rights and does not elect to secure a patent in a foreign country, the contractor may file and retain principal or exclusive foreign rights subject to retention by the government of at least a royalty free license for governmental purposes and on behalf of any foreign government pursuant to any existing or future treaty or agreement with the United States.

That is, the federal government reserved a license on behalf of the foreign government involved, subject to an agreement with that government. Contractors obtaining ownership of inventions made with federal support could not sue foreign governments for infringement unless the federal government permitted it.

Bayh-Dole does away with such niceties. At the same time, it does little to provide American companies with any advantage. Think of it this way: foreign companies may practice subject inventions in any jurisdiction where an American university or nonprofit or small business has not obtained a foreign patent. They are in these jurisdictions free to practice, relative to the federally supported invention. But American companies are all subject to a claim of infringement and only one company, the exclusive licensee, is granted the right to use or to sell products–after paying a royalty to the patent owner. While foreign companies may be denied access to the US market for a time, they have the rest of the world in which to compete at an advantage. So much for US leadership in technological innovation.

I once had this discussion with some Mexican technology transfer officers who were attending a training course I was part of. When they discussed patent strategy, their plan was simple: file a patent application in Mexico and another in either the US or Europe. Try to license the patent to a big foreign company. Make lots of money. Why not Mexican companies? Nope, not reliable or already government monopolies. I pointed out that the effect of obtaining patents was that, if unlicensed, the patents worked against the development of the technology in either Mexico or the US/Europe. If there was unencumbered competing technology available, the demand for a royalty-bearing exclusive license worked against the patented technology getting developed. In other words, the only way patents contributed is if the underlying inventions were immediately put into practice by anyone ready to do so. To hold out for an exclusive licensee meant death by blockage. To hold out for a market leading big corporation to take an exclusive royalty-bearing license was not only death by blockage but also death by fantasy (and blockage).

Yet we continue to have this idea that patents represent an incentive to companies, and that somehow all, or most, or even many inventions made with federal support will be developed and used only if there is the “incentive” of a patent monopoly. Even if this were true (and it’s obviously not–companies are willing to develop products even in the face of universities claiming infringement), the government’s policy in Bayh-Dole is that these companies must also expect to pay a royalty on their exclusive right. It’s a patent monopoly, but it comes with a 1% to 6% hit on earnings. Quick, if you were to choose between a patent monopoly with no hit on earnings or a patent monopoly that you must share with a university but only after a six-month or year-long negotiation for a license, which would you choose?

No, the “incentive” of the patent monopoly held for license by a nonprofit is among the worst incentives possible. The Harbridge House report (1968) found that licensed inventions did much worse than inventions that were owned by a company with experience. Yet Senators Bayh and Dole did not recite that fact. If Bayh-Dole had intended to put the patent incentive in the best possible position, it would have provided for ownership of each subject invention by an American company. Nonprofits would have been required to designate upfront for each federal grant an American company as recipient of any subject invention, preferably royalty-free. If no company was available, then the federal government could request ownership. That would have been a much better system. But then, that approach would have made explicit what Bayh-Dole is really about–using government subsidies for research to create patent monopoly positions for American biomedical companies.

Senator Bayh poo-poo’d the idea that companies might get unjustly rich through government-assisted patent monopolies. But that’s just what we have seen happen with prescription drugs. The biomedical industry publishes public broadsides announcing that through Bayh-Dole, American has become the market leader. And that leadership is measured by profits–and those profits come from monopoly pricing. The American public does not obtain the benefit of such drugs on “reasonable terms.” The public gets the benefit–what benefit there is–on unreasonable terms. $80 or more dollars a pill where the manufacturing cost plus a good profit is $3 in the case of Xtandi.

Bayh-Dole was built on good bluffing political rhetoric and ultimately was passed after Senator Long let up his opposition as a present to Senator Bayh, who had been voted out of office. The problem for the rest of us is to see the situation directly and not through political rhetoric forty years old and just as empty then as it is now. Yet organizations such as AAU, COGR and AUTM and  have repeated these same arguments for all these years, like a religion.

Bayh-Dole has never been implemented on its terms. Most universities refuse to comply with the substantive provisions of the standard patent rights clause or acknowledge that patents on subject inventions are not ordinary patents. Whatever has happened during Bayh-Dole has not come about because of Bayh-Dole, but because of a faux version of the law created by university administrators and perpetuated by their legal advisors. The one thing that has happened during Bayh-Dole is that biomedical companies have indeed gained monopoly access to the inventive results of publicly funded research, have charged monopoly prices, and have made good money for their investors. This has been a big win for these companies and speculators, at the expense of the American public, inventors, research directors, and a host of companies that would have worked with most any other research finding, but that it was held behind a patent paywall to justify monopoly licensing practices for biomedical companies.

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