Bayh-Dole–six parts real, one part faux

Bayh-Dole is a law in six real parts and one faux part. There’s

  1. a policy part (35 USC 200, 201, 206, 210-212);
  2. requirements for contractor owned inventions (35 USC 202(a, b), 203, 204, 205);
  3. requirements for federally owned inventions (35 USC 207-09);
  4. requirements for standard patent rights clauses (35 USC 202(c));
  5. standard patent rights clauses (37 CFR 401.14(a, b), 37 CFR 401.9)
  6. implementing regulations (37 CFR 401)

Despite all this apparatus, there is also a seventh part to Bayh-Dole, the faux part. It’s the account of Bayh-Dole that’s fake and deceptive, the part that never gets implemented, complied with, never used. Faux Bayh-Dole dominates practice and dominates as well not only public accounts of Bayh-Dole but also academic articles–even ones critical of Bayh-Dole. There is significant cognitive dissonance involved in watching an academic argue that a faux practice under Bayh-Dole is the fault of Bayh-Dole and suggest some (generally silly) policy remedy.

While universities and federal agencies make a show of compliance with the superficial bureaucratic parts of Bayh-Dole, the substantive parts of the law and implementing regulations are ignored, misrepresented, kept secret, and unenforced. So there’s great fanfare regarding the particulars of when a university must disclose; what form the disclosure must take (including what form the disclosure may be submitted on); time frames for reporting to the federal government, electing to retain title (or not), and filing patent applications; and including a government rights statement in each application. None of this stuff has a wit to do with practical application, innovation, or public benefit–it’s all bureaucracy and paper fussiness.

But where the government obtains a broad license to practice and have practiced all subject inventions, the government rarely–if ever–practices or has practiced those inventions.

As for the public and inventor protections–there are few and the ones there are ignored. The march-in provisions have never been successfully used. The preference for small business was gutted in 1984. The preference for US manufacturing is easily circumvented. The requirements on nonprofit assignment of subject inventions are ignored. The restrictions on nonprofit use of royalties and income relating to subject inventions are ignored. And everything that might be reported is made a government secret. Most of what is required of universities and nonprofits is walked back in the law, or in the implementing regulations. Agencies can waive preference for US industry, waive assignment restrictions, waive reporting requirements; agencies can ignore march-in conditions, can ignore reporting defaults–and they do, generally. Within the standard patent rights clause, the written agreement requirement is simply ignored by universities and federal agencies alike.

Does all this suggest a law that you expect to operate as written?

And that’s the problem–Bayh-Dole is a law that runs cover for a set of practices that gut nonprofit research independence, stifle the public domain, and exploit the public interest. The primary bit is to feed patent monopolies on publicly funded research inventions to pharmaceutical companies via nonprofit fronts. By using nonprofits, the idea goes, the whole affair will appear virtuous. If it is virtuous for nonprofits to feed patent monopolies to pharma, then it must also be virtuous for the federal government to do the same thing, provided that the details are kept secret. And if it’s virtuous to feed patent monopolies to pharma, then it also must be virtuous to feed those patent monopolies to venture capitalists to attempt to resell to pharma. And if all of this is virtuous, then surely it is virtuous to engage venture capitalists to attempt to sell the patent rights to anyone fool enough to want those rights. And if all those purposes are virtuous, then of course it must be virtuous for states to fund efforts to develop “technology” to the point where venture capitalists are willing to speculate on reselling it to pharma or to fools. And if that’s the case–that states, for “economic development” reasons find value in pushing this scheme along, then it is virtuous to hold everything behind a proprietary paywall, just in case someone wants to pay to speculate, and it is virtuous to strip inventors and investigators alike of ownership and control over their self-directed work.

The down side of virtue-signalling is that it can cover for practices that are anything but virtuous.

Bayh-Dole is a dismal failure. First, because it has never been implemented. Second, because it creates virtue-cover for a set of practices that are themselves a dismal failure. Third, because it has led to those failed practices becoming the dominant form of research transaction, driving out viable alternatives.

There are no data showing Bayh-Dole is successful. Folks like Joe Allen are political operators, using political bluffing as fact. And there are people who want to live the life of illusion, and if deception is necessary, then it must all be good because it is in the public interest for research to be bottled up behind a patent paywall, and it is in the public interest that public health should be exploited for profit by corporations provided with patent monopolies for research subsidized with public money. These would be mere buffoons if they weren’t so maliciously effective in mucking up research enterprise.

But nothing changes. We may conclude that, based on Bayh-Dole and its faux implementation, next to nothing in the $35 billion or so the federal government spends on university-hosted research is worth bothering with. It just doesn’t matter. It’s pork. It’s a honey pot to keep clever research people out of industry and government here and abroad. But it’s not about invention and discovery, or promoting use, or promoting free competition and enterprise, or developing American industry. It’s a welfare scheme for pharma and for the patent brokers and nonprofit administrators willing to compromise their institutions to have even the tiniest share of the monopoly pricing that might be had and to encourage speculation in this share, using their institutions’ reputation as the bait.

Explaining the details of Bayh-Dole is like pointing out bugs in Windows 95. No one uses Bayh-Dole, so they actually don’t care how it operates.

If Bayh-Dole were enforced, the changes would shut down most university and federal licensing operations instantly. Rats would scatter from the barn. A whole lot of “guidance” would get removed from university and AAU and APLU and COGR web sites. AUTM, however, no doubt would keep its guidance up forever, claiming that Bayh-Dole has been wrongly interpreted to mean what the words of the law say. You know, like the Supreme Court’s ruling in Stanford v Roche.

In this context, repealing Bayh-Dole would simply leave in place the faux version of the law, now required by university patent policies altered to “comply” with Bayh-Dole. Instead, faux Bayh-Dole must be repealed.

The path to repeal faux Bayh-Dole starts with three steps:

Require public disclosure of basic subject invention use information that demonstrates that each subject invention has met the standard of practical application as defined by Bayh-Dole. Sunlight the rats!

Audit university practice for substantive compliance with Bayh-Dole’s policy statements on the property rights of owners of patents on subject inventions. Practical application, free competition and enterprise, American industry and labor, small company preferences, assignment compliance, use of funds compliance. More sunlight on the rats!

Require compliance with the (f)(2) written agreement requirement in the standard patent rights clause at 37 CFR 401.14(a), and make clear that this agreement displaces any university claim on subject inventions as a condition of participation in federally supported research. Protect inventors from rodent-infested institutions!

That’s the start.


This entry was posted in Bayh-Dole, Bozonet, Policy. Bookmark the permalink.