Trademark Scholars' Roundtable part 2

Session 2: Designing Registration Systems and Institutional Questions

Robert Bone: two models. Registration as supplement to background rights, or substitute for background rights.  US is the first; second is more like other countries. 

From a normative perspective, putting aside peculiarities of federalism, strongest argument for preserving the background is the common law process for defining rights. But that implies that there’s no necessary content to the substantive benefits conferred by registration.  You can figure out what substantive benefits you want.  Some benefits do seem to demand registration, if only to provide notice (geographic reach).  But incontestability, for example, could exist without registration, as long as the mark had been in exclusive continuous use for 5 years.  Need a normative story about the advantages of preserving the underlying common law system.

Registration as substitute. Screening function.  Might also be used to expand rights, through a property notion. Not clear you need to displace the underlying system.  Choice between PTO and courts. If you thought the PTO would be better at liability, it might be appropriate to give it rulemaking power. They might not do a better job than courts, but could they do worse?  Give it Chevron deference and let it decide whether initial interest confusion is actionable.  If we see the PTO as a potential alternative, that opens up options (though that’s only an extension of the discussion of registration).

Robert Burrell: Institutional design. How would a more dominant agency affect TM law and policy? IP Australia exerts extraordinary control over the legislative agenda, which couldn’t happen here with the PTO; majority in Parliament gives it control. How does that play out in practice? Can become captured by large brand owners, who formally have a seat at the table; by the time of public consultation, IP Australia has already made its decision.  Will never give up its existing functions—Australia established a national register for security interests in personal property. But registering on the TM or patent register would still give advantages; the agency defends its turf no matter what. Also inevitably sees reform through the lens of what troubles the agency. Reforms of inter partes provisions have been made many times over recent years, but infringement has been untouched; not because infringement is untroubling, but because IP Australia deals with inter partes proceedings.  Within a few months of identifying that they didn’t know how to count public holidays for deadlines, the problem was fixed.  Well, at least there’s an agency that cares about boring technical things.

Has a huge surplus; needs to spend it in order not to be seen as a profit center for gov’t. So now we have a propaganda industry—sponsored an exhibit at a children’s museum to teach them about the importance of IP rights.  Allows them to design a logo for their smoothies.  Also pushing academic work; can fund academic projects.  The projects it funds will not be deeply critical of the office. 

When we think about institutional questions, it’s not just regulatory capture. That exists, but other considerations as well.

Jessica Litman: the current system doesn’t work very well. Incentives for overclaiming and ridiculous claims because examination is so haphazard. No real sense of whether geographic misdescription will be noticed, so why not give it a try? We could improve incentives by improving quality of examination.  If examiners were paying more attention, lawyers wouldn’t apply so haphazardly. A US problem is what Beebe calls opportunistic applications: ITUs for ridiculously invalid claims over popular words and phrases.  95 applications for “Desert Storm” from everything from toilet paper to children’s toys. 163 ITUs for some form of “Obama.” Are Americans just crazily acquisitive? How do we discourage inappropriate claiming? 

Another idea: take jurisdiction of registration reviews away from the Fed. Cir.; need for expertise is not as persuasive (if you think it is for patents). 

Could also require more specimens of use—small, but encouraging taking the job more seriously.

Sheff: 1/3 of applications are published without office action. 2/3 with an office action: ½ of those are abandoned, 12% of the 2/3s will proceed to final rejection and be abandoned. So about 50% abandonment rate. 10% appeal from final refusal, and more than 2/3 of those will be abandoned ultimately. At each gate in the process, you can get the applicant to go away just by saying no. Maybe that’s in examiners’ minds when deciding whether to publish.

Burrell: in brief defense of the PTO, there is no question that US TM examiners are much more careful than any other group he’s encountered: Australia, OHIM (another category entirely), India, New Zealand, etc.  Don’t assume that because small/medium size enterprises don’t respond to office actions they may not understand that it wasn’t a serious attempt in the first place.

McKenna: Deborah Gerhardt shows that applicants with lawyers do substantially better, and that’s particularly true with office actions.

Kur: not just about the way rights are acquired, but about having an agency that considers itself the center of the world as far as TMs are concerned.

We certainly don’t have a good way of getting rid of dreck applications, unless manifestly descriptive, which is often not the case for a popular phrase applied to various goods/services. Could be bad faith if there’s no intent to use it at all, but that’s hard to prove.

Bone: why be so concerned about smaller/medium entities?  Can understand worrying about small authors in copyright, maybe even small inventors. But TM?  Students come up to him with various enterprises they want to register TMs for; he says “get a lawyer.”  If they don’t, well, there are lots of costs that small businesses face.  (I think Dawn Donut reflects this insight: most businesses are not going to expand and don’t really need the full panoply of federal TM rights.) 

Why not require fees for abandonment, equal to the costs incurred by the examination?  Internalize the external costs from an unsuccessful application?  (Or refund part of the now-higher fee after success?)  Penalties require merits determinations, whereas his approach would be an ordinary fee-shift.  Ordinary fees may sort of do that now but then why do we have such a high filing rate?

Dogan: could make initial filing fees higher and maintenance fees lower.

Bone: that could chill applications.

Dogan: ex post makes it administratively difficult to collect.

Litman: post a bond.

McKenna: consider funding sources generally. PTO thinks of applicants as clients; incentives are to issue.  Could structure a fee system that gave greater incentives to reject, not just issue. PTO is trained to say that the statute says the TM shall be registered unless there’s some barrier; the default is registration, and that presumption must be overcome. The natural outcome of an application is a registration.

Kur: that comes from the Paris Convention.

Litman: but they overcome that impulse on a fairly regular basis.

McKenna: registration’s meaning doesn’t have to displace scope of rights/meaning of right, as with 1905 registration system. Not binary between registration or common law.

Dinwoodie: Agrees with McKenna: what is it you get with registration? If it’s metes and bounds definition of property, that’s different from starting point for court’s analysis—if the latter, court may be able to expand the right subsequently.  Presumptions; double identity (same goods/same mark in the EU); other possibilities.

Consumer protection orientation may justify registration for small/medium enterprises in a different way than encouraging product differentiation (which I find an interesting way to divide the world—I don’t see TMs as encouraging product differentiation).

Goldman: multiple flavors of registration—ITU, supplemental register, principal register.  PTO suggested IP clinic for TM registrations—but Goldman wants to know who the clients are and why we should be manufacturing TM rights (or at least registrations) at a discounted price. 

McKenna: if you think those rights exist anyway and that you can move them into the registration system rather than multiplying them, there might be something to be gained by moving them in and serving a notice function. If you think it will strengthen or duplicate them, then it’s a different question.

Goldman: believes it’s the latter; we’ll be telling these businesses “you should be in the IP game.”

McGeveran: small landscaping service with a cute name. Consumers need protection from passing off. Registration relies on landscaper to do its job; if it doesn’t do it, then consumers aren’t protected. If you change the nature of the rights by registration, it’s different.

Bone: consumer protection is hostage to business decisions throughout the process. Transfers of business; naked licensing exists and consumers are confused. Can’t get away.

Dogan: but against a tradition of respecting consumer expectations through use-based system, would be a major change to move to a registration-required system.

Dinwoodie: §43(a) protection will exist anyway, but those rights should be more limited—no dilution, for example.

McGeveran: how to design the narrow alternative and keep it narrow. 

Dinwoodie: take away some of the phrases in §43?  They don’t even appear in §32.

RT: the problem we have is that §43 is so broad that it covers everything one might want from §32. Then courts think that registration must mean something, which is sensible in and of itself, and so they make rights even broader … which then can’t be confined to §32 because of the broad language of §43.  Dinwoodie suggests that we should make people prove more under §43 than they have to prove to win under §32, including materiality, but what else is left to make them prove given the breadth of §43?  (Still experimenting with this idea: try double identity (plus confusion) as a requirement under §43, while allowing different goods/services when there’s a registration? There’s a bit of weirdness in that because you’ve specified your goods/services in a registration, but in theory you also have to prove rights based on use in §43 so that limitation is implicitly in there.)

Leaffer: crush of applications has been important to change in quality. ITUs and Madrid.  We have the information to make quality registrations, but we don’t have the resources.

Some discussion of whether there are fields with higher than average/lower than average oppositions and abandonments. Kur: Some fields have special issues, like pharmaceuticals where there is a separate process of approving names at the medical level.

Dinwoodie: back end (core/periphery of rights) affects how much effort you want to put in on the front end.

Janis: sounds like claim construction for utility patents, which is not a compliment.

Grynberg: to get a shift to the front end, build in an idea that courts will start deferring. Some courts give Skidmore-like deference to the TMEP, but could go further to Chevron rulemaking/adjudication.  One possibility is that PTO’s problems then go up on steroids as it defines the scope of a registration. But PTO could also have a carveout authority—a backwards ratchet.  DMCA exemptions as possible model?  (Only if you don’t have to go back every 3 years.) PTO has a whackamole quality: a few get through/inconsistency. With notice and comment and rulemaking, organizations can bring focused attention to bear.

Bone: public choice dynamics in copyright; exceptions for incontestability aren’t written in similarly industry-specific terms, but why do we have the ones we have?

McKenna: idea was that we don’t want to have to reprove every time the right to the mark. You don’t need incontestability for that; presumptive evidence of right to mark, but that was the intent anyway—secondary meaning wouldn’t have to be shown every time, even though burden shifting would also have gotten the job done.  Functionality was created by courts, but put in after Silstar.  (Bone: but why didn’t firms exert enough pressure to stop that amendment?)

Dinwoodie: a lot of exceptions are things that might change over time. Something might be nonfunctional at registration and become functional, similar to genericity—dynamic.  Because the main opposition to incontestability was FTC/competition policy; the exceptions therefore track competition.

Litman: TM bar is both plaintiffs’ and defendants’: INTA’s predecessor/ABA committee had some concern about false positive registrations that became incontestable but had other defects that weren’t related to secondary meaning, but didn’t worry too much about losing secondary meaning.

Burrell: Australia used to have incontestability until 1995; Red Bull is about to push through a significant change. Incontestability can protect TM owners against retrospective changes to the law made by common law reasoning. If the judiciary changes the standard and invalidates a bunch of marks, incontestability is a protection against those changes.

Sheff: registration can add an in terrorem effect to enforcement activities, especially when small/medium entities are on the defense side as they often are.