Sixth Trademark Scholars Roundtable: The Territorial and Product Market Dimensions of Trademark Law, UT Austin
Session 1: The Territorial/Geographic Market Dimension
Graeme Dinwoodie: territoriality is becoming more important in reconciling local and global markets.
Lionel Bently: Geographic localism used to be possible; a common story is evolutionary—the most common word is “increasing.” Increasing connectivity, move across geographic borders. But what we lack is a deep history. He’s not convinced that the change narrative is right. 19th century cases featured a lot of transnational marketing and movement of goods. Many of 1830s and 1840s brands were being marketed in America and UK. Not clear whether this was licensing, but should make us skeptical about claims to novelty that underpin much of this work.
Late 19th/early 20th century English cases on passing off. One case from 1909 where House of Lords first recognized “goodwill”—involved Spaulding, an American company making sports goods. First case that recognizes that descriptive words can have secondary meaning was a case involving camelhair belting (for use on machines), used in tropical countries where traditional materials didn’t work—which is to say, sold as part of a system of international production and distribution of goods. UK registration system was response to pressures from traders from elsewhere who feared they wouldn’t get protection under the passing off regime.
We know that there was a lot of international marketing, and the cases are informed by that. We should be skeptical of claims to novelty, and about what it is that we identify as new. Wants to think of the role of TM in British colonialism—systems that allow protection of goodwill generated in Britain in colonial markets, to allow for expansion of British business easily into those markets.
More conceptual Qs: is it desirable to protect spillover reputation that exists from the marketing of goods elsewhere? Some situations, most of us would probably agree that it is desirable to bar another from using the relevant designation. Cybersquatting type situation, where someone has just registered a well-known mark in Country B in order to hold up any potential marketing/extort money from them. No social benefit from permitting that.
Then there’s the trader who uses the mark in Country B—potential consumer deception. Well-known mark doctrine undermines the function of the registration system. Maybe because there’s no need to register; maybe because it defeats nonuse limitations on registered marks. Undermining registration might not be too bad a thing, though. Also, you increase the costs for other traders who want to find out what marks are available, which the registration system is supposed to reduce.
What is the standard? How do you find out whether a mark has a reputation? Might be easier to show a mark is well known. Tendency to think that well known is a higher standard than having a reputation. But if the level of known-ness is lower, there are problems, and the notion of known-ness in Europe is much lower than Americans seem to think.
Empirical questions: Does protecting a well-known mark facilitate expansion into other markets or not? Does it allow companies to stay out for longer because they have protection anyway? These seem to be testable Qs. Skeptic: well-known mark protection is sought by already successful businesses and is a product of rent-seeking more than real economic efficiency.
What are the effects on local investment? Absent cybersquatting/blocking, if someone innocently adopts McDonald’s in a country like South Africa or Jamaica (real examples) and invests in the mark, and in due course gets sued. We don’t know! You can say the local producer didn’t need to adopt the mark, but at least in cases of innocence that’s not fully satisfying.
RT: Thanks also for the opportunity. I’m afraid I won’t live up to it, because I just have a series of questions. (Bently’s first, historical question: “What it is that we identify as new?” raises for me the possibility of increased direct consumer access, instead of firms bringing goods across borders. Rise of mass media? But this may well be overstated. For his project, compare Christine Haight Farley on the inter-American treaties where American companies were apparently doing the same thing.)
A forehand/backhand: TM is about consumer protection, until TM is about managing relations among businesses. Or is it vice versa? Principles, then exceptions that may be so broad as to actually constitute the background rule (Dawn Donut being one possible example). Another example: exhaustion, where legitimate goods can be sold across borders … unless there are material differences. Even if the consumer is getting them from Australia herself by buying them on eBay, apparently, the variation in details, including details in the warranty, still means that there will be confusion as a matter of law. So we have exhaustion in theory but in practice a retailer can prevent it.
Dorpan v. Hotel Melia, from last year: area of use isn’t exactly reputation, certainly not zone of likely expansion. It’s confusion. Confusion might not be everything, as Mark and others have told us, but it’s the only thing. What this demonstrates to me: Lack of theoretical or even practical basis for an understanding of territoriality as something distinct from the trademark’s strength. And this is true for registered/unregistered both, I think.
Should registered marks be different? Nationwide priority means hypothetical confusion inquiry. We act as if the mark had the full scope of presence everywhere, even if it doesn’t really. This hypothetical confusion inquiry is usually done in opposition proceedings. I’ve been thinking a lot about using the registration to act as if the rights were defined by the registration, rather than by use. I think of this as a European model, though clearly the Europeans can’t entirely resist the appeal of looking to the actual use either. Implications: not just nationwide priority, but possible effects on strength (this I think is wrong, though you’re starting to see cases where courts say it, as well as courts that say that incontestability means strength), and also on similarity (where the registrant has a standard character mark but usually uses a particular design, and the defendant’s design is different but the words are similar/the same). If we have nationwide priority, why don’t we have these other features of a registration system?
Well, maybe because of the common law reasoning that resists these hypotheticals. The cybersquatting like behavior Bently talks about wouldn’t work in a use-based system, since there’s no bona fide intent to use. Dawn Donut: Based on the idea that non-nationwide businesses aren’t really that likely to become nationwide—only a few of them really do manage to make it to all 50 states–so it’s on balance better for society to let other businesses continue to use the currently nonconfusing marks even though if the senior user does expand there would then be confusion. It is a risk assessment.
Final thought: Carol Rose on crystals and mud in propertylaw. She says courts will always seesaw between them because each has features that operate as necessary correctives to the others. Should we think about TM, including territoriality, this way?
Dinwoodie: Territoriality issues often collapse to the distinction between use and registration. If you really bought registration, you’d have rights to chunks of goods (at least for 5 years in Europe) as defined by the registration.
Munich economists studied changes in 1860s when TM regime changed—inconclusive, struggling to isolate variables. Benelux—in 1973, Benelux abolished national rights and went to unitary right.
At some point, a difference in degree justifies difference in kind—it is fair to say that territorial issues are “increasing.” More interactions of national marks in foreign contexts, particularly online.
Courts do address the cost/benefit issues in the well-known marks cases, such as Grupo Gigante; even the Bancorp case where the dissent has it right. They’re thinking them through in context of defining the territorial dimension of rights. Even European courts may limit relief to countries in which a mark is known and in which therefore there is confusion, even though the right is by hypothesis EU-wide.
McKenna: Every year, we fight about whether TMs are about search costs. International level: becomes more clear that the question of what TMs are for is so much more diffuse. Right to expand in another market is a political choice about structuring commercial relationships, not a consumer issue. Compromises at the international level have little to do with the normative Qs Americans tend to think TM is about, and the layering makes analysis more difficult.
Dinwoodie: Grupo Gigante is phrased for the “poor immigrants coming over the border,” but the well-known mark doctrine comes from treaties during a period when there weren’t such immigrants and it really was a right to expand.
Bently: the people who really want the well-known mark doctrine are lawyers who fear consequences of not doing their jobs (that is, getting registrations) in a particular country. (This assumes a registration based system!)
McKenna: registration system is driven by a sense of geographic expansion of trade across the US; this isn’t new but goes to the 1800s in the US, and also true in the international level—but the expansion is all being carried out by means of registration. What differs now: we’re talking about spillovers beyond registration—scope of relief, infringement, etc. A lot of the features of the registration system are features of int’l agreements driven by old waves of product diffusion; what’s different now?
Searching for use v. searching for repute: McKenna isn’t sure we search for use in any setting. We say we do, but we’re always searching for repute. We don’t have a great idea of what constitutes use, and so we look for things like analogous use. There are some exceptions in priority disputes, but the overwhelming majority of cases really look for recognition, not use. Maybe the level of repute we’re talking about is higher versus what’s required to find basic “use,” but use isn’t doing the work.
Robert Burrell: points out that Australian courts will routinely reject consumer surveys.
Dinwoodie: true of British etc. courts while US courts fetishize them; surveys by definition look at distinctiveness to consumers, whereas nonsurvey proxies are more likely to look at what the business is doing. The evidentiary rule has effects on (or perhaps reflects) the attitude.
Robert Bone: “use” from the 19th century was about what was required to appropriate/control the mark, from a property standpoint. 20th century: move from property to goodwill, pushing us to reputation, but we still have the old use rules. We’re seeing this old use idea being pushed and crumbling, but TM holds on to the past.
Sheff: how do we determine whether a mark is well known? Surprisingly difficult to determine fame, which his current empirical project. Looking at third-party assessments of brand value. Problem is that best evidence is in the hands of the person claiming big rights; they know what consumers think about their brands better than anyone else. How do we force that info out of the private user? A registration system is a useful tool for getting one piece of information: I’m staking out a market. But it’s a very small piece if we’re interested in reputation among consumers in a jurisdiction. So, we need incentives to disclose.
Dinwoodie: in other countries with pure registration based systems (not much examination as in US) there are a lot of “squatted” marks on the register and there’s cost to getting them back.
Bone: chilling effect on local investment must not be from mark (because there’s always another mark to use), but from fear of big guys coming in.
Burrell: it’s gotten harder to do parallel imports in Australia—in US too?
McKenna: the material difference doctrine has been around for a while, but courts are increasingly willing to treat differences in warranty as material differences—or even removal of product codes. Why is that material? Well, Sheff notes that the TM owner claims it interferes with its monitoring of its customers. RT: but how could that be material to consumers?
Burrell: maybe in international cases courts require real confusion, not all the proxies we use in national cases—so proving a reputation turns out to be really quite hard.
Dinwoodie: that works in Australia because it’s not such a mess as the US.
Burrell: if you’re doing a well-known marks case in Australia, the action is technically the same but the foreign company must prove a reputation, not just prove that it’s been selling and advertising for a number of years.
Heymann: Q of whether the marks mean the same thing across jurisdictions. Does this mark have a meaning is different—do you recognize it v. what does it mean? Allowing producer to control meanings across jurisdictions—or not. McDonald’s can have different types of McDonald’s depending on local cultures.
Mid-point Discussants: Marshall Leaffer
There’s good literature on the desire of French firms to promote their goods in French-speaking Africa—could feed into the imperialism discussion.
Interaction of reputation, goodwill, and use: Convinced that de facto we’re looking more at reputation than goodwill. WIPO says that reputation should be the centerpiece. Leah Chan Grinvald argues that there should have to be a high level of secondary meaning first. McDonald’s South Africa case seems to comport with WIPO standards—need a substantial number of persons to recognize the mark, which means not a negligible number—a very liberal standard, attenuating the notion of extraterritoriality. Bulova and Vanity Fair are deprecated, after Grupo Gigante; the q of the nationality of the individual in question is less important. Even McBee v. Delica was a close case; if he’d shown real some sales in the US, he should’ve won.
Formalities of registration and renewal: it’s really hard to keep up with them in all countries. Well-known marks = important safety valve for the costly system of today.
Jessica Litman: Many more companies are international today. First, consumer understanding. Second, some degree of territoriality is how the different states/contracting parties acknowledge each other’s sovereignty and power over their own markets and citizens. Increase in multinational corporations affects both: likelihood that consumers think that Marlboro in Japan is the same Marlboro in the US. Acme Donuts in NY and California: increasingly consumers believe that two companies are related not because consumers or goods are traveling more but because they’re aware that many companies aren’t local. 50 years ago if you were traveling to a different city and saw a descriptive mark you’d assume that this was just a different local company. Multinationals also change political economy. Lobbying their various sovereigns to respect their marks cross-border.
Mike Grynberg: How strong/robust do we think these expectations are, versus the costs of weakening territoriality?
Burrell: shouldn’t assume that multinationals desire to trade everywhere; sometimes they have no interest in being in the market. McDonald’s has said outright that it won’t enter Bosnia because it’s too much of a dump. Well-known marks doctrine still applies.
Dinwoodie: they might come in if Bosnia improves.
McGeveran: they’d say that badly run McDonald’s would harm their reputation among non-Bosnian tourists even if Bosnians aren’t confused.
Dinwoodie: that’s a very producer oriented argument, not really about consumer protection.
RT: (Note on harm stories being quite often wrong.) Now I’m confused about what goodwill means (ok, I’ve always been confused). I thought it meant something like the selling power of the mark, which is to say the power to bring sales to the TM owner. So you can have reputation without selling power?
Some discussion with Litman over whether consumers would really think Acme was the same in a different state, given other contextual elements. It may be that we shouldn’t treat Proctor & Gamble as a descriptive term and should abandon the old rule about proper names, but just because there are some crossnational brands doesn’t mean they all are, and I think consumers can still tell the difference.
Litman: Gave her students the Cracker Barrel case, and they were largely shocked that the two companies were separate entities.
RT: yes, where a brand presents itself as a national brand and you can find Cracker Barrel cheese in multiple supermarkets, I see that. But there are also plenty of businesses that are obviously local.
Bone: is there harm from any difference? The quality of Cracker Barrel cheese and Cracker Barrel restaurants isn’t really related. If they both look like national brands, you can presume you’ll be ok.
Dinwoodie: convergence among products is related to convergence in territory (Cracker Barrel being a great example). Change in business organization affects both.
One of the tradeoffs of trade liberalization is that new companies can come in and fight for market share on the established one’s turf. So while you can call the TM rent-seeking, it looks less so in the context of the broader trade deal.
Common-law system in theory starts with reality and has developed hypotheses; the registration system starts with hypothesis and occasionally consults with reality. The Q is whether they end up in a different place.
Leaffer: Goodwill: what is the mark worth is another definition, but it all depends on other things such as reputation. Reputation is a less loaded term; skeptical of arguments that well-known marks doctrine should be limited to cases in which the mark has goodwill.
Materiality in exhaustion cases, used to destroy exhaustion—does seem to be going in the opposite direction of the copyright cases.
McKenna: in terms of the harm story, are there points of differentiation between the marks? Relationship between scope of mark and territoriality. Someone need not open McDonald’s with the same logo/colors in Bosnia. Empirical evidence suggests that consumers are good at differentiation when there’s a reason for them to do that. “Big Mac Noodles” have successfully opened in some East Asian countries.
We keep talking about reputation/goodwill, but in the context of functionality he’s considered non-reputation-related advantage, and the Q is “reputation for what?” Courts are pretty slippery and use the concept as a makeweight, as in Au-tomotive Gold. The court thinks that the use of the mark on a keychain necessarily calls the reputation to mind, ignoring the question of whether the reputation is for those goods or services. We have the same problem with confusion, talking about it as all one thing, and we do that with “reputation” too.
Laura Heymann: agrees with the point about logos. Marks don’t exist in tombstone, black and white form; almost always visually presented, even though cases sometimes mention the radio. Always has a context, which affects how consumers think about the relationship between the two.
Thinking through harm story about Bosnian McDonald’s: if consumers are confused, do they judge McDonald’s business judgments as opposed to the ordinary meaning of the quality of the goods and services? That’s a different kind of reputation.
Dinwoodie: “what’s the harm?” is a distinctively American question to ask. One could ask a different question (e.g., unfair advantage/unjust enrichment).
RT: (1) a registration system has a ready answer to the “marks don’t exist in tombstone format” point, if the word mark is registered—so then we should dig into the question of why allow that kind of registration, and relatedly why we allow nationwide priority. (That is, McDonald’s doesn’t use the block letters; it uses its own distinctive font. If we nonetheless agree that McDonald’s is “using” the standard character mark, which I think we must (otherwise it should be cancelled for nonuse!), then someone who opens a McDonald’s in Bosnia with a different font is also using the standard character mark, whether or not that causes confusion). (2) On Leaffer’s exhaustion comments: There’s a connection also between exhaustion and the point Dinwoodie makes about having to compete against entrants from outside. Exhaustion means that in theory the owner has to compete with its foreign products too. The larger trade scheme needs to be considered, and the Europeans seem to be ahead of us here.
Jeremy Sheff: Mary LaFrance is working on a paper on the implications of Kirtsaeng for TM—predicts shift to TM methods of exercising control/preventing exhaustion.
Mark Janis: what is the body of law on well-known marks outside the US?
Bone: Persons has this rule about “bad faith” adoption by a remote junior user. But the theory is that it’s a remote area and there’s no local meaning—what could bad faith possibly mean? Maybe it’s an evidentiary test—an idea that the D might have reason to believe that there is secondary meaning in the area. Or maybe it’s a proxy for potential market entry into the area, though that would be weird. Then you have a Dawn Donuts problem: there’s no entry yet, so what are we accomplishing by stopping the junior use? Also, as to secondary meaning/goodwill, if that’s the only evidence thereof that’s really slight.
Litman: real consumers encounter TMs in context. A strict application of territoriality treats copying logo, trade dress, business concept exactly the same as copying only the word mark. If we want to treat them differently, and her impulse is to do so, the Q is if there’s a way to do that, say by limiting famous mark doctrine to identity of logo/other visual elements.
Leaffer: there’s a paucity of cases on well-known marks. The South African McDonald’s case is therefore our fallback.
Dinwoodie: you can have passing off in many countries; the countries in which you find these cases tend to have registration-only systems and no passing off backup.
McGeveran: we should be distinguishing between products and services. The nature of what’s moving across territorial boundaries can be goods or it can be information/reputation, and those are distinct. Increase in both of those movements. Consumer’s ability to go online can defeat decision of goods-maker not to enter a country (unless exhaustion doesn’t apply!), whereas the Bosnian McDonald’s is different.
McKenna: why would we want TM rights to preserve rights to entry? We might think that owners are more likely to be good at expanding to/exploiting a new geographic market than a new product market.
Registration: maybe we have a rule for registration purposes that block letters encompass everything, but once we get to infringement we ignore that. We’ve been talking mostly about word marks: but what are we talking about with a registered trade dress? We don’t really have a long history of interpreting what registration of a product configuration means, mostly because we don’t care what the registration says in an infringement case. But as we start thinking more about stuff outside the registration system, that will force us to think harder about the relationship between the registration system and infringement actions. We wouldn’t necessarily want the McDonald’s registration for Ba Da Ba Ba Ba, which is registered in standard character form, to cover all alternatives. You might want to treat it like a patent: you get what you claim, and people who use more than that don’t infringe the patent.
Dinwoodie: Compare Louboutin. Registered mark valid, but not used.
When you export US registration for international registration purposes, what do you get? Problems are usually thought to come from the description of goods and services/classes, but also this discussion shows that there can be issues from the description of the mark itself.
Janis: now we’re talking about a regime of claim interpretation, which has its own costs.
Maybe the McDonald’s issue with the different logo should really be a 10bis claim, unfair competition not TM. There, you focus on what the D has done, not the scope of the P’s right.
Dinwoodie: one reason we’ve moved away from “use” is the rise of the service economy, where advertising is more important, and once you do that for services, why not for goods?
EU is now struggling with need for Dawn Donut type rules since rights are now so geographically broad. The internet doesn’t make local use irrelevant, it makes it more relevant.
Bently: to what degree is this related to nationbuilding and perceptions of nationbuilding—does Dawn Donut reflect a view of the nation that no longer seems accurate to us? (Comment: If anything, we’re more divided!) In Europe, the TM stuff is clearly part of a process of Europe-building, but placed registration in tension with some of the functions of TM, so we’ve started to carve out a series of exceptions from the unitary mark.
RT: w/r/t nationbuilding, contrast the nationalism of our export of copyright—one might suggest that now that the US doesn’t make anything but movies, it hasn’t felt much need to export TM, or even incorporate others’ rules. By contrast we’ve felt great need to export our copyright regimes, minus our limitations. Exporting culture and law together.
Dinwoodie: European politicians are resistant to any derogation from one EU policy; judges however worry about granting relief that covers 29 countries without reason for that breadth.
Also, © harmonization might be behind TM harmonization for various reasons and TM thus less in need of pressure from the perspective of the harmonizers.