Tuesday, October 06, 2009

Flash forward: cellphone lawsuit continues

MetroPCS Wireless, Inc. v. Virgin Mobile USA, L.P., 2009 WL 3075205 (N.D. Tex.)

Eric Goldman pointed me at this interesting case, which shows how trademark has metastasized even in the hands of a court not particularly willing to give the trademark owner the benefit of the doubt.

MetroPCS reconfigures cell phones from other providers to operate on its wireless network, a process called reflashing. Virgin Mobile doesn’t like MetroPCS doing this to its phones, so after some threats they sued each other.

Customers play a flat fee for unlimited use of MetroPCS’s network, without a longterm contract. Virgin competes for the pay-as-you-go market, which is marketed to consumers with low income or poor credit who can’t afford longterm contracts. Virgin customers buy minutes and only pay for what they use. They use Virgin handsets that bear the VIRGIN MOBILE trademark on the face and show the logo on the electronic display. There’s also proprietary software that allows the handset to connect to the Virgin service and to download content like ringtones and graphics. Virgin sells its handsets below cost, and argues that this is only economically feasible if it can recover its losses by selling airtime. So, if a customer buys a handset and then reflashes it for service on a competitor’s network, Virgin suffers. It makes handsets with software and security measures to “control access” to its software and prevent alteration. And the handsets come with terms of service prohibiting alteration of the hardware or software and barring use of the handset on any other wireless network.

MetroPCS, on its part, reflashes handsets only when a handset owner requests and buys service with MetroPCS, part of which is signed agreement to terms specifying that the handset will use a network other than that of the provider whose marks appear on the handset; that the coverage may be different; that they’ll get only voice and text messaging; that they don’t have a contract with another service provider; that the owner isn’t part of a scheme to get bulk quantities of handsets; and that the owner won’t use the original provider’s trademarks in selling, distributing, or advertising his/her handset.

Virgin alleged direct and contributory trademark infringement, dilution, and tortious interference with contract.

The court first considered whether MetroPCS’s conduct could constitute “use in commerce.” This first required a discussion of the law of reconditioned/altered goods. Basic principles: if the product is, after extensive repairs/alterations, essentially a new product, the original trademark must be removed. But if it’s not a new product, it’s okay to retain the original mark so long as it’s clearly and distinctly sold as repaired or reconditioned. And courts are particularly reluctant to use trademark law as a barrier when the product’s owner requests a repair or modification: some cases hold that’s not even use in commerce, where use is defined as trading on the trademark’s goodwill—repair and return to the original owner isn’t a use in commerce. But if the owner-requested repairs are so great as to make the substance of the transaction a sale, that’s “use” because the seller is effectively selling a new product under the original trademark to a purchaser “who believes he is paying for (and/or using) a repaired good that is of the same make as that which is represented by the trademark.”

So: how much does MetroPCS alter a Verizon handset? Is it just a change in software, or a new product with a different core function: connecting to MetroPCS’s network instead of Virgin’s? Virgin argued that MetroPCS’s recognition that handsets could be damaged or made inoperable by reflashing undercut MetroPCS’s argument that its changes weren’t that significant. The court found that summary judgment had to be denied on this point. There was evidence that the industry treats handsets as distinct from the services on which they operate, and some carriers unlock the handsets they sell under certain circumstances. But the industry default is to lock a handset to a network, especially for prepaid handsets. (Not entirely clear to me that industry practice translates to consumer knowledge any more than the use of DRM in music conformed with consumer expectations, but presumably this can still be litigated.)

MetroPCS also argued from common sense: an electronic device is distinct from the service it’s configured to use, like a TV set is distinct from the cable service it receives and would not be a new product if reconfigured to receive satellite signals. The court disagreed because, due to restrictions imposed by the current market, handsets aren’t analogous to TVs. Over 90% of handset retail is controlled by four carriers, which limit access to handsets that tend to be locked to particular networks. The carriers make it difficult to take your handset with you when you switch services. So reflashing might create a new product.

Assuming that it does, there is still the issue of likely confusion. (Let’s pause for a moment to appreciate that the court noticed that there were actually two issues here! Sadly, there are courts that would have inferred confusion from the presence of a new product.) Along with the traditional multifactor test, the court also considered three factors relevant to sales of an altered product: the extent and nature of changes made to the product, the clarity and distinctiveness of the labeling on the reconditioned product, and the degree to which any inferior qualities associated with the reconditioned product would likely be identified by the typical purchaser with the manufacturer. But Virgin didn’t contend that point of sale confusion was likely, given the conspicuous and unequivocal terms MetroPCS customers confront, so the real theories here are post-sale confusion and initial interest confusion. Complicated enough? Here is a good case for Bill McGeveran’s argument that trademark doctrine has become so complex and reticulated that it’s almost impossible to dodge trial even with a strong defense, which has trademark-expansive and speech- and competition-contracting results even when the doctrine theoretically allows a lot more freedom.

Virgin argued that consumers observing others’ use of the handsets, using the handsets themselves, or buying reflashed handsets in the secondary market,

“are likely to be confused as to the association of the handsets and service with Virgin Mobile and to mistakenly believe that the decreased functionality of the handsets is attribut[able] to Virgin Mobile.” The court found a genuine dispute of fact on post-sale confusion and thus didn’t consider initial interest confusion.

The usual top two factors—mark strength and similarity of marks—were of “little value” under the circumstances, since MetroPCS keeps Virgin’s mark on an altered Virgin product. The court noted that MetroPCS doesn’t mark the handsets with anything showing they’ve been reflashed. On similarity of retail outlets and purchasers and similarity of ad media used, Virgin’s evidence would permit a reasonable jury to find likely confusion, because the parties both target lower-income, bad-credit consumers and use similar media to market themselves.

The court found that a reasonable jury couldn’t find that MetroPCS intended to derive benefit from Virgin’s reputation by retaining Virgin’s mark on reflashed handsets. (Stated this way, MetroPCS is apparently entitled to an instruction in its favor at trial, which is an interesting result.)

To show actual confusion, Virgin submitted online sales listings by owners of reflashed phones, which purportedly showed that other companies were associated with MetroPCS, e.g. offering a “Motorola MetroPCS” or “Verizon Metro PCS” phone, and posited that these showed actual and potential post-sale confusion. The court held that a reasonable jury couldn’t find that these listings showed actual confusion about the relationship between MetroPCS and the original handset maker; at most they showed potential post-sale confusion.

The court then evaluated product similarity and the extent of change to the plaintiff’s product together, and found a genuine dispute of fact on this, as noted above. Relatedly, it looked at the clarity and distinctiveness of the labeling on the altered product, and noted that MetroPCS doesn’t label reflashed handsets. MetroPCS argued that such labeling would be pointless, given the written pledges its customers already sign not to use the original providers’ trademarks in selling or advertising the handsets and that the nature of the market “virtually guarantees” disclosure of the service a handset is configured to use, because otherwise a buyer wouldn’t know whom to contact for service. But the court concluded that not all customers got that pledge, and that disclosure doesn’t necessarily mean lack of confusion about affiliation. (Query whether a label “reflashed” would help. What, other than not reflashing, does Virgin think MetroPCS ought to do?)

Next: the degree of care exercised by potential purchasers. MetroPCS claimed that handset consumers necessarily take care in the secondary market, because they need to pick the right device for the network. But, the court concluded, that doesn’t mean a lack of confusion about the relationship between the service provider and the trademark owner (that is, whether Virgin approves of the ability to use a Virgin handset on a different network). This, in a nutshell, is why we need a materiality requirement in trademark to match that in false advertising: who the heck cares? If people are “confused,” it’s because they never consider this question and, if asked, simply have to guess what Virgin’s relationship is to MetroPCS.

On to the degree to which the typical purchaser would attribute to Virgin Mobile any inferior qualities from reflashing. The court found a question of fact for the jury on this. In conclusion: summary judgment in favor of MetroPCS denied.

Next came dilution. The court pointed out that dilution was founded in concern for use of a famous mark on unrelated goods or services, and questioned whether dilution applies at all between competitors. Still, MetroPCS didn’t raise the issue, so the court assumed that dilution could occur between competitors. In a previous case, the Fifth Circuit rejected a dilution claim by a vacuum cleaner manufacturer against an independent vacuum cleaner sales and repair store because trademark doesn’t let markholders control the aftermarket in their products, even if that means that independent repair shops use lower-quality used parts when repairing the vacuums. A trademark owner shouldn’t be allowed to “cry dilution” every time a resold or repaired product reflects poorly on the mark it bears. “We refuse to encourage anti-dilution law to metastasize in this manner.” (Applause!)

Virgin argued that the vacuum cleaner case was inapplicable because MetroPCS isn’t an innocent second-hand store, but a deliberate creator of inferior MetroPCS products bearing the Virgin mark. The court reasoned that the vacuum cleaner case hadn’t involved a dispute about whether the repaired products were still the same product; in such cases, referring to the used/repaired item’s trademark is the only feasible way to identify the product.

So, in the genuine aftermarket, it’s natural that consumers associate a repaired/rebuilt/used product with its mark and base their opinion of a product on its performance after years of use and repairs. Those things don’t go to the point of anti-dilution law, which is to protect against increased consumer search costs. “Because aftermarket sellers of used genuine trademarked products accurately reference the trademark, and because consumer opinion of a product is often informed by the product’s performance after months or years of use and periodic repairs, the aftermarket seller’s use of the trademark does not increase consumer search costs or reduce the signaling power of the trademark.” (Cue me disagreeing strongly about the search costs rationale in general, and being baffled in specific: all the court has established is that dilution is inherent in a market for durable goods, not that it doesn’t happen, especially as compared to a world in which the trademark owner could suppress the secondary market.)

Anyway, the same factual issue of whether there was a new product precluded summary judgment on dilution by tarnishment.

Next, Virgin argued contributory infringement. MetroPCS argued that Virgin had no evidence that its customers engage in direct infringement, and that its requirement that customers pledge not to infringe absolved it. To win, “Virgin Mobile must adduce evidence that would permit a reasonable jury to find that someone committed an act of direct infringement and that MetroPCS either intentionally induced that person to commit the act or continued to supply reflashed handsets to that person when it knew or had reason to know that he was engaging in trademark infringement.”

Assuming that Virgin could show direct infringement, it still failed to produce sufficient evidence of intentional inducement or knowledge plus continued supply. Virgin argued that MetroPCS didn’t consistently require the pledge from everyone, and that MetroPCS should have known about direct infringement because of “open and notorious” activity on the internet. Even though there was a genuine issue of MetroPCS’s consistency, that fact wasn’t material, because there was still no evidence of intentional inducement or of knowledge of particular infringing consumers to whom MetroPCS continued to supply handsets. At most, MetroPCS continued to sell to the general public when it had reason to know that some past customers had resold their reflashed handsets, and that’s just not good enough to establish contributory infringement in trademark.

Then, applying standard copyright preemption doctrine, the court determined that Virgin’s contract claims were not preempted by the DMCA’s exemption for making phones compatible with new networks. The contract claims alleged an extra element beyond “rights equivalent to any of the exclusive rights” granted by copyright. Arguably this is the wrong analysis, because the issue isn’t copyright preemption but paracopyright preemption (especially since the contracts here purport to operate against the world—like property rights, not traditional contract rights; note the interaction with the tortious interference argument below). Moreover, the exemption at issue here was allowed by the Librarian of Congress on the explicit reasoning that interconnection is, as a policy matter, not something with which copyright law or anticircumvention law should interfere. Still, I wouldn’t be surprised to get the same result out of a court that paid more attention to the difference between copyright and anticircumvention rights.

On Virgin’s tortious interference with contract claim, Virgin had to show that its contracts prohibited reflashing. MetroPCS argued that there was no evidence that the handsets it reflashed were from people in privity with Virgin, rather than from people who’d already bought the handsets on the secondary market. Virgin argued that its contract terms traveled with the handsets, so sellers couldn’t transfer to buyers more rights than they had, but the court rejected this attempt to make contract terms run with the handset. Virgin sold the handsets; it didn’t license them. The use restrictions in the terms of sale were consideration for the sale, but Virgin retained no interest in the handsets. (Note how different the result would likely have been in a pure software transaction—Virgin cited software cases, which the court rejected because they were founded on copyright.)

Virgin therefore couldn’t win summary judgment on its intentional interference with contract claim, because it couldn’t establish the existence of a contract between Virgin and any particular MetroPCS customer, though it will get its chance to prove intentional interference at trial.

No comments: