Friday, February 06, 2009

The Boundaries of IP at William & Mary

Intro and first panel later, when I've written up my notes.

The Boundaries of Patent Law

Usual disclaimers, squared: I’m not a patent person. I will get stuff wrong.

Margo A. Bagley, University of Virginia School of Law, Invention Creation Activity as a Boundary in Patent Law

Perfume, book/movie set in 1800s, about a man born with no smell, which caused people to be uncomfortable around him. So he decided to create the perfect scent, and was successful. Unfortunately, he did it by (spoiler alert) killing young women. Suppose he wants a patent on his new and nonobvious process/result. Should the illegality and immorality of his actions impact patentability?

We usually don’t look at inventor behavior. But your patent can be unenforceable if you (post-filing) withhold info from the patent office with intent to deceive; or prosecution laches; or patent misuse post-issuance. What about pre-filing? The existing doctrines rely on the idea of unclean hands.

Real situations where illegal activity is taken into consideration: a number of countries have made noncompliance with genetic resource access laws a reason for denying patentability to any invention. India, Brazil, Costa Rica, Andean community, and China (#3 in patent applications worldwide and growing fast). New amendments: if the acquisition and exploitation of genetic resources used to complete an invention was contrary to law, no patent; and if your patent depends on genetic resources, you have to indicate the source. Done to give effect to Convention on Biological Diversity: fair and equitable sharing of benefits with sovereigns, on mutually agreed terms and with informed consent.

Proposals in multilateral fora like WIPO and WTO to include mandatory disclosure of origin, and even the possibility of post-grant sanctions if requirements weren’t met.

Other instances: Germany: using human embryonic stem cells created in Germany; US: falsifying data to obtain a federal grant. What kind of causation should count? If the researcher speeds on the way to the lab and makes a discovery, that probably shouldn’t preclude patentability.

Other than illegality, there’s also immorality in invention creation. Normal inquiries are about use of invention or nature of invention—gambling machines, abortion pills, torture devices; transgenic animals, human/animal chimeras. What about invention creation activity? Generally ignored. But the patent office did reject a patent based on ground up skeletal remains acquired by digging up graves within a week after burial—that’s immoral activity.

Recent example: the European Patent Office had to apply European Patent Convention art. 53, which says that patents shall not be granted if they’re contrary to ordre public or morality, provided that exploitation of the invention shall not be deemed to be contrary merely because it’s prohibited by law or regulation in some or all European states. Challenges of application: balancing tests; public acceptability/abhorrence; no clear standard. Amendment: included a list of immoral inventions: processes for cloning humans, modifying the germ line identity of human beings; uses of human embryos for industrial or commercial purposes. None of these dealt with invention creation, but patentability of those things. But a patent application for a cell culture was rejected because the only way to make the invention was to destroy human embryos, on the ground that that was commercial use. Euro. Patent Office took a broad view of what counts as commercial use, and what counts as an invention. Other patent offices have rejected similar types of inventions.

The research is legal, but has been deemed immoral. Whose morality will control? So for example, Switzerland moved to honor the interests of plants in themselves; researchers concluded that decapitating flowers on the roadside would be immoral.

In the US, we don’t read moral into useful, though the Weldon Amendment prevents the PTO from issuing patents on human organisms by use of appropriations power.

Unethical activity in patent creation: violation of duty of informed consent in Moore v. Regents of the University of California; use of Nazi data.

Possible remedies: claim invalidity; unenforceability; ownership transfer. Invention creation activity can and should be a boundary in patent law.

Mark A. Lemley, Stanford Law School, Distinguishing Lost Profits from Reasonable Royalties

Boundary issues in doctrine: how we distinguish between remedies for patent owners. Two basic types of remedies: either lost profits or in no event less than a reasonable royalty, designed to compensate patent owners, not to punish infringers or require disgorgement of profits. Patent infringement is almost never a culpable offense; you can infringe a patent without knowing and with your own inventions. Outside pharma, only 2-3% of cases involve allegations of copying. So we don’t want plaintiffs to get windfalls.

If you lost profits, you get those back; if you didn’t, you still get something, like what the parties would have negotiated in a counterfactual where they negotiated.

As a general matter, you want lost profits. Reasonable royalty is thought of as a floor. A patent owner with a monopoly can sell at a monopoly price; an infringer will drive the price down, and their total surplus will be lower. There is therefore no price that would compensate the owner as a license fee; owner’s lost profits will be greater than infringer’s profits. But a reasonable royalty by definition assumes that the parties could agree, and the infringer would be unwilling to pay more than profit, so it is theoretically lower.

Turns out to be very hard to show lost profits: owner has to show it had the production capacity to make the sales, that there were no available substitutes, and a number of other things. If you don’t get lost profits, you’re relegated to reasonable royalty even if you’re a direct competitor. This is not a realistic royalty, as discussed above. So courts think that the number they come up with is too low. So increasingly, as lost-profits-entitled plaintiffs get kicked into reasonable-royalty territory, courts find ways to jack up “reasonable royalties.” Sometimes courts just say expressly that they need a kicker, and the Federal Circuit approved “discretionary increases.”

Courts also have rejected the idea of apportionment, when a technology is only one part of a larger picture. Imagine a hybrid auto; invention of the engine’s a big part of the value. But there are other parts of the car that also contribute to value, like the wheels and the windshield. But the Federal Circuit excludes those from the assessment of the reasonable royalty—just figure out what the product is and calculate a percentage. That could lead to the right result, if you’re lucky, but it doesn’t always.

Lost profits world has “entire market value” rule. If I’m the inventor of the hybrid engine, and people only buy the car because of the engine, then all the infringer’s sales are at the cost of the inventor, because the inventor would have made the sale without the infringement even if the infringer has a better windshield wiper. This makes sense in lost profits, but no sense in the reasonable royalty context, yet courts have imported it. So the patent owner, who may not be practicing, comes in and says I’m entitled to the entire value of the hybrid; then the patent owner on the wheels use comes in and says so am I; you quickly get well over 100% of value.

And it gets worse. Sometimes goods are sold in packages: if I sold you a car, I would have sold you a service package and tires etc.—convoyed sales. Again, makes sense with lost profits, but now we’ve imported it into reasonable royalties, giving owners percentages of sales of completely noninfringing products. WTF?

Patent owners have now decided they’d rather have the supposed floor than the supposed ceiling. In the Monsanto seed-saving cases, where Monsanto sues farmers for replanting seeds in violation of the license, we know exactly how much Monsanto is injured: $6.50 per bag of soybeans, which is the price to everyone in the world. Monsanto doesn’t want lost profits; the reasonable royalty in this brave new world hovers between $35-50 per bag. No wonder they want reasonable royalties! That can’t be the right result.

Suggestion: need a boundary. Market participant-owners should get lost profits, and we should do our best guess about what those are, even if it’s imperfect. Non-market participant-owners have a different injury: the amount they could have licensed the invention for, and reasonable royalty should compensate them for the value of their contribution, but not for the value of the other components of the product. Easing proof of lost profits, we’d stop importing those concepts into reasonable royalties.

Michael J. Meurer, Boston University School of Law, Patent Examination Priorities

The time patent examiners spend examining applications: 18 hours on average. Want to consider: error costs; strategic responses by patent applicants (don’t want reforms to be evaded); ability to monitor how examiners implement whatever the rules are; the extent to which patent examiners can help define clearer boundaries of patents; effort cost—maybe some things are better addressed after issuance/at trial; benefit of information provided by patents to the public.

Errors can be false positives or false negatives; add in errors on scope, which can be too narrow or too broad. Where should examiners concentrate effort to reduce errors? Scope.

There’s a “piece rate” which keeps the assembly line moving, connected to a customer service culture. This minimizes false negatives but encourages bad applications. The current system has failed, but he’s still looking for improvements. (1) Auditing: a second pair of eyes, an innovation pushed by people unhappy with the quality of business method patents; applicants therefore evaded it by drafting to get out of the class of business method patents. Auditing could be made unevadable. (2) Reform training and culture to pay more attention to the scope of claims that issue.

What’s the PTO good at compared to the courts? Some kinds of prior art are hidden from the examiner, making the cost infinitely high. PTO’s probably also bad at policy judgments, like 102(b) public use or maybe even 103 nonobviousness. Some examiners will be good at enablement, others won’t. We should try to increase technical expertise.

Some tasks are harder when claim scope is unclear. You don’t need claim boundaries to figure out whether there’s utility, and not so much to figure out enablement. Where you really want to nail down boundaries is in 102(a)/prior art—need to know what the claim is to know whether the prior art reads on the invention.

Do rates of licensing and litigation depend on the mix of examiner error? Error that induces more applications increases search cost an thereby encourages litigation and harms “good licensing.” Breadth errors are greater threats than validity errors.

John F. Duffy, The George Washington University Law School, Rules and Standards on the Forefront of Patentability

There are two standard ways of thinking about articulating legal principles: rules and standards. (Or, as Carol Rose might say, crystals and mud. You can tell I’m teaching Property this semester, eh?)

In re Bilski: Duffy advocated a standard, and the court wanted a rule. His motivation to write this paper: rules fail in this area of law. But innovation is unpredictable (it wouldn’t be innovation otherwise) so the idea that rules can guide innovation will probably fail in the long term. But there may be value to failure.

List of historical failures: Early doctrine was supposedly that you can’t patent a principle—tried to say this was a rule, though it sounds standard-like; by the end of the 19th century, commentators noted that the subject matter covered by the invention was known as the principle of the invention—so the rule failed as a rule. Another example: you couldn’t patent a new use of an old machine, recognized by SCt in 1875. Seventeen years later (one patent life later) they backtracked. Another example: you can’t patent life. 5-4 decision ultimately allowed it. Federal Circuit rules have a half-life of about 10 years, which is bad given that patents last for 20.

Why do rules fail? Because the law in the US is broad and vague, and has no obvious limits.

Rules do, in the short term, permit property rights. Rules of inclusion—safe harbors—might be a good thing. That’s probably the best defense of Bilski. But courts should try to keep them stable. Failures can also contain the seeds of appropriate new standards.

Recommendation: courts should try to avoid big rules, especially at the higher levels of appellate courts. It would be a catastrophe to grant cert in Bilski, no matter how it comes out. If it’s codified in SCt precedent, it’s much harder to change. Also, if we create rules, we must tolerate their over and underinclusiveness.

Commentator: Joshua D. Sarnoff, American University — Washington College of Law

When he hears “boundaries,” he thinks “transgressive.” The doctrines we’re dealing with are attempts to naturalize politics; these papers make them unnatural so that we look at alternatives. What people think is out of bounds—moral restrictions on the grant of property rights—happens actually all the time, so we should think about when and why we do that. Given the narrowness of the current interpretation of the experimental use exception, perhaps almost all research using patented materials is illegal and we should be denying patents left and right.

Lemley’s paper: concerns how patent owners are supposed to be treated by the patent system. Sarnoff doesn’t agree with Lemley’s view of lost profits; in the lost profit situation, we don’t try to put the patentee in the position it would be in absent infringement; we give lost profits and reasonable royalties—compensation that’s “adequate” is a legal question.

Meurer and Duffy are about how the system should be structured to better effectuate policy goals; they don’t take the system for granted. Sarnoff finds Meurer’s assumptions problematic, though: if you change the legal doctrines, it will change examination priorities. So let’s start with the doctrines and figure out what they’re doing, not necessarily start with the examiners. For example, definiteness; pre-KSR obviousness—shift from rule to standard changes the examination process.

Lemley: Believes in compensation. If you say adequacy is a legal conclusion, you accept undercompensation.

McKenna: If you think of real situations of lost profits, you’ll largely have a patent owner who should be getting more than what the infringer makes; that’s effectively an injunction, because it’s not worth it. If that’s available only to market participants, that fits really well with eBay.

Brett Fischmann: What counts as failure? If legal doctrine evolves over time, is it failure just because it changes?

Duffy: It’s not a failure if it provides short-term stability and reduces litigation. If that’s their goal, that will be a success. It’s a temporary compromise, like a truce. He thinks the need for change will be obvious within a few years, but maybe we can tolerate under/overinclusiveness for now.

Lemley: If he thought Bilski had given comprehensive guidance on anything, he’d agree.

Q: Is our goal to stop inventors from doing immoral research, or to prevent immoral research from gaining legitimacy? Relying on Nazi research won’t change anything, but it may grant wrongful legitimacy.

Bagley: We haven’t answered that question, and we should. US tends to want the invention out there, regardless of source.

Q by me: Crystals and mud? Isn’t this just an inevitable cycle?

Duffy: The cycling happens really fast here. Some things have been clear/stable for 200 years, like if the invention has been open for more than a year. What makes this area different from areas in which we see stability? (Carol Rose-inspired answer: considerations of equity—the bumbling good guy and the sharp operator?)

Samuelson: Destabilizing expectations in the middle of the game isn’t always bad; Duffy’s been willing to change the rules in other areas. Is Duffy offering a standard for patentable subject matter, in which some things are outside patentability, or is he rejecting any limits on what can be patentable? That’s not a standard, that’s just a rule that’s very broad. Patents on a storyline?

Duffy: No. He’d exclude some stuff. We may be able to refine our intuitions about what’s patentable by looking at these specific patents.

Lemley: He’s uncomfortable with that, willing to be more hardline about allowing everything in. People’s instinctive reaction to the new is that it’s bad, so you wouldn’t have gotten software patents, biologicals, etc.—you’d exclude the things that are most innovated.

Duffy: Software patents were originally heavily restricted, but they were allowed. Eventually we began to realize this stuff was like other fields of engineering: recognition of facts on the ground, and there was no ironclad rule in the way of patentability.

1 comment:

Unknown said...

It was a pleasure hearing you at the symposium today! I think the high point of your talk had to be explaining what 'slash' was to a room full of law professors and law students.