comparison charts might infringe if lacking a disclaimer

Penn Engineering & Mfg. Corp. v. Peninsula Components, Inc., 2022 WL 3647817, No. 19-513 (E.D. Pa. Aug. 24, 2022)

This seems like a silly result to me, shifting the burden to comparative advertisers, but it's often much harder to get summary judgment in a trademark case than in comparable cases.

The parties compete in the industrial fastener market. PennEngineering claims a PEM family of marks and sued Peninsula for trademark infringement, counterfeiting, false advertising, and unfair competition.

The court denied Peninsula’s motion for summary judgment except for pure keyword buys, counterfeiting (based on resales of PEM goods), and claims based on marks Peninsula didn’t use. That left for trial claims based on some online ads that used PEM in text and on cross-reference charts used for comparison, as well as trade dress claims based on a “square-in-square” clinching nut.

keyword ads, no use in text

Bizarrely calling keyword buys “keyword conquesting,” PennEngineering sought summary judgment based on alleged sales diversion. The court pointed out that “diverting customers is a key aspect of competition. … Here, there is no dispute that the links are clearly labeled as belonging to Peninsula and there is no likelihood of confusion where the use of trademarks as trigger words is hidden from the consumer.” 

However, allegedly according to vendor mistake, some ads did show PEM in display text. The court found that the requisite intent is intent to confuse, not intent to use, and that a factfinder would have to decide whether these ads were confusing. “[A] reasonable jury could conclude that there is a likelihood of initial interest confusion from the visible use of ‘PEM’ in Peninsula’s advertisements because the advertisement is not clearly labeled. On the other hand, a reasonable jury weighing the Lapp factors could instead conclude that that there was no intent to confuse (rather, a vendor mistake) and that the average Internet user’s ability to understand the difference between ads and direct search results diminishes the likelihood of confusion in these limited instances.”

keyword ads with mark in text

Because nominative fair use is the defendant’s burden in the Third Circuit, the court let claims based on cross-reference charts continue, even though it seems hard to understand why these comparisons would be confusing about source. Peninsula’s charts “identify Peninsula products that are available as substitutes for [PennEngineering’s] products.”

cross-reference chart

Although the customers are sophisticated, the fasteners cost only pennies per unit; the parties also disputed whether “a handful of emails sent to Peninsula—which PennEngineering interprets as attempting to reach PennEngineering instead—constitute more than de minimis evidence of actual confusion.” So there might have been confusion, though the court didn’t explain why there’d be confusion because of the comparison charts.

The court said that it could be necessary to use the plaintiff’s mark for comparison purposes even if fasteners are “staple items” (commodities). [Notably, plaintiff apparently has two-thirds of the relevant market.] “[I]t is clear that a party may use comparative advertising for any items, including ‘staple’ items.” Although the court rejected the claim that a cross-reference chart comparing types of Bob’s Burgers to McDonald’s burgers would constitute trademark infringement, because “there is no likelihood of customers confusing Bob’s Burgers and McDonald’s from looking at such a chart,” the charts here were apparently different.

The strongest argument I can find for this result is plaintiff’s argument that “the absence of disclaimers on the cross-reference chart ‘suggests a manufacturer-distributor relationship, or that [Peninsula] is a division of PennEngineering or a division of a common parent.’” Why would the relevant consumers think that? What makes Bob's Burgers different? The court thought that this constituted a factual dispute over whether the chart portrayed the true relationship between the parties.

The court also declined to grant summary judgment on whether the claimed double square trade dress, which was registered, was functional—the registration meant the burden was on defendant, and while its evidence created a factual dispute, the registration also weighed in favor of nonfunctionality.

the design

Peninsula, like PennEngineering, makes floating fasteners with square-in-square designs.

A floating fastener uses an outer retainer that allows an interior nut some “float” to accommodate minor misalignment. “Double squares” refers to a fastener with an interior square that rotates with a screw and is stopped by an exterior square, within a certain amount of “float” between the outer square’s corners for misaligned parts, as shown below:

PennEngineering’s VP of Product Development testified about the way the smaller square provides some float while the larger square prevents rotation, which is “necessary for the functionality of the part.” PennEngineering’s expert also opined that the interaction between the two square shapes “is a functional feature of” the product. And Peninsula cited a prior trademark prosecution by PennEngineering for admissions that the product design is functional.

The Third Circuit very clearly doesn’t require that a feature be “essential” to be functional.  Nor did the existence of other designs disprove functionality. “Further, PennEngineering does not show that these alternative shapes would allow the same amount of float in both directions.”

Nonetheless, Peninsula didn’t meet its burden of proving functionality, because it didn’t introduce its own expert report or explain exactly how the design “affects the cost or quality of the article,” or how “the ‘exclusive use of [the feature] would put competitors at a significant non-reputation-related disadvantage.’ ” PennEngineering’s expert opined that the method of manufacturing the shape “entails the same cost and produces the same quality for any shape, although ‘some shapes may be marginally easier to form than others.’” That was enough to avoid summary judgment.

Counterfeiting through reselling legitimate PennEngineering products: Protected by first sale. There’s an exception when goods are “materially different,” which [supposedly, though not always in practice] requires consideration of “whether the allegedly infringing products are likely to injure the goodwill developed by the trademark owner in the trademarked goods.”

PennEngineering argued that resold products were “materially different” because they weren’t covered by PennEngineering’s warranty or customer service. But this wasn’t a material difference where no reasonable person would ever claim on a warranty for a two-cent fastener (the warranty did not cover consequential damages), and where its own VP of quality was unaware of any warranty claims from reseller sales and even stated that, if such a claim did occur, PennEngineering would likely honor its warranty the first time for a given customer. PennEngineering also apparently was unpersuasive in analogizing to an unauthorized used car dealer selling a Ford truck.

Such a truck would not have the original Ford warranty and dealer services but would have the Ford logo and all original parts without physical alteration. According to PennEngineering’s theory of counterfeiting, the truck is “not a Ford.” Yet, a material differences claim requires showing “differences that are likely to damage the goodwill developed by the trademark owner.” Unlike a fake Gucci bag that falls apart and injures an unsavvy, label-anxious customer’s goodwill toward Gucci, a used Ford truck is still a Ford. PennEngineering’s “ceci n’est pas une Ford” theory of counterfeiting falls flat.

False advertising: PennEngineering argued that Peninsula’s use of cross-reference charts and copied performance data constituted false advertising, because PennEngineering had test results supporting its performance data and Peninsula’s comparisons were based only on its expectations that the results would be the same.

The court declined to grant summary judgment on literal falsity. In the Third Circuit, “although the plaintiff normally has the burden to demonstrate that the defendant’s advertising claim is false, a court may find that a completely unsubstantiated advertising claim by the defendant is per se false without additional evidence from the plaintiff to that effect.”

So, were these claims completely unsubstantiated? “Peninsula admitted that it reverse-engineers the parts and then assumes that the performance ‘should’ be the same as the PennEngineering products.” By contrast, “PennEngineering describes a time-consuming and expensive process to test each of its products for performance ratings (e.g., how much weight each fastener can secure).” Peninsula “publishes performance data on thousands of products that show identical performance but … only independently tested around 51 of its products,” and PennEngineering argued that it “sometimes disregards its own test results to publish PennEngineering’s data instead.”

PennEngineering’s expert opined that the way in which the fasteners are manufactured affects the final performance properties. Peninsula’s Director of Product Development admitted that Peninsula does not know how PennEngineering manufactures each part and does not specify manufacturing processes for its manufacturing vendors.

However, PennEngineering did not actually show that any particular performance rating published by Peninsula was incorrect.

For summary judgment purposes, Peninsula’s testimony that the goal of the reverse-engineering program was to follow the “form and function” to achieve the same performance, plus its expert test results showing that the performance characteristics of a subset of the products tested were the same, would allow a reasonable jury to conclude that Peninsula’s performance data has a “semblance of support,” which would defeat the per se falsity theory. PennEngineering also failed to show actual deception, as required because it was seeking both injunctive relief and damages.

Likewise, the court denied summary judgment on the argument that the charts would mislead consumers based on an expert’s opinion (apparently not backed by a survey) that the cross-reference charts were likely to mislead customers “into believing that the corresponding [Peninsula] products are equal in quality and all other respects.”