Monday, November 14, 2011

Literally false images get another airing

Aviva Sports, Inc. v. Fingerhut Direct Marketing, Inc., 2011 WL 5395975 (D. Minn.)

Another day, another opinion in this case. Aviva makes fixed-air inflatable water slides and pools (this means they’re inflated and then plugged). Manley also makes fixed-air products, as well as constant-air products which require continuous inflation through the use of an air compressor; these are generally larger and more expensive. Aviva created a prototype of a constant-air slide, but Target allegedly told it not to pursue the product because Manley already “owned” the market. Target also stopped selling Aviva’s inflatable pools and slides. Shoremaster acquired Aviva, resulting in significant internal changes, including a number of lost employees. Aviva’s sales declined over the next few years.

Aviva sued Manley and four retailers for alleged patent infringement (stayed pending reexamination, which resulted in a rejection of some claims, which is on appeal) as well as violations of the Lanham Act and Minnesota UDTPA. Aviva alleged that ads and/or packages for 95 Manley products contain false representations, superimposing scaled-down images of children onto images of Manley products to make them look bigger. See, e.g., reviews at Sears or this review with helpful contrast between the Manley picture and the reviewer’s picture. Manley product imagereviewer's product image showing smaller productIn addition, Manley allegedly used larger, custom-made products for its photo shoots, rather than the actual products being advertised.

On Manley’s motion for summary judgment, the court treated the UDTPA and Lanham Act claims the same. Manley argued that Aviva lacked evidence on at least one element for each of the 95 products, and proof as to one ad isn’t proof as to any other. The court agreed that each ad needed to be proved false, but thought that evidence about one ad could also work as evidence about other ads when the ads and the products are similar. “To find otherwise would completely ignore the role of circumstantial evidence.”

However, Aviva had no standing to contest Manley’s advertising for its 32 constant-air products because it didn’t show that its products competed with constant-air products. Aviva argued that its failure to enter the constant-air market was Manley’s fault, but Aviva didn’t provide evidence of what its reputation was or could have been without the false advertising, nor that Manley’s alleged category ownership was the result of false advertising.

Turning to the fixed-air products: visual images may be literally false. The court here found the question of literal falsity best left to the jury. Aviva presented the testimony of two individuals involved in the design and creation of Manley’s packaging and ads. One testified that the products used for photo shoots were custom-made items that were roughly twenty percent larger than the actual product being sold. Moreover, she was instructed to miniaturize the children in the images to make the products look larger for "[a]lmost every package." Another senior packaging designer declared that "all of the design team, myself included were specifically directed to misrepresent the size of every Banzai product in its product packaging and advertising, as well as most other Manley Toys product lines." He also stated that "the actual products photographed for the Banzai advertisements and product packaging are custom made, and are 25% larger than those actually sold in the packaging…. [T]he children depicted in the Banzai advertising are then photoshopped into the photographs, and reduced in size to make the Banzai product appear even larger." Manley disputed whether these people worked directly for Manley, but since intent isn’t required, that didn’t matter.

Manley argued that reasonable jurors couldn’t find false advertising before 2007, when these people weren’t yet involved in Manley’s ads. Moreover, Manley argued, neither one testified as to the specific 95 products at issue. The court was unpersuaded: a jury could infer that the images of the 95 products at issue were altered enough to be literally false, and that such practices existed prior to 2007, especially if the jury finds the ads similar across time.

Aviva also offered an expert witness who physically examined 26 Manley products and analyzed the images of 36 additional products. He concluded that "many of the Manley Toys product packages misrepresent the size of the products contained in the box by shrinking the children, relative to the products, who are shown in the promotional images." Further, "the magnitude of the differential scaling between the children and the products is such that young children (ages 5-10 years old) are often portrayed to be the size of toddlers or babies."

Expert testimony was not necessarily required for the remaining products, when common sense would suffice. The jury could reasonably conclude that “at the very least, those product advertisements are literally false.” Moreover, it could compare the images of other products to products Aviva did analyze to assess their similarity and thus the falsity of the ads for the other products.

Manley argued that evidence that the images alone were false was insufficient for a jury to conclude that the ads taken as a whole were false. The court disagreed. Context matters, but a jury could reasonably conclude “that if the images misrepresent the size of the product, then the advertisement as a whole is literally false.” (Indeed, I’m hard pressed to see how a jury could conclude otherwise. Perhaps having lost earlier motions, Aviva was leery of moving for summary judgment on liability. But so far I’m not seeing Manley’s case for non-falsity.)

If a jury found literal falsity, Aviva would be entitled to a presumption of consumer deception. As to materiality, Aviva presented two consumer depositions testifying that the product images affected their purchasing decisions. Manley argued that the testimony of one consumer per product was insufficient. The court found this argument meritless, under the circumstances. (This may be a litigation deathmatch, but if I were Manley’s attorneys I’d be thinking hard about settlement. What is a jury going to think about these pictures?)

Aviva [I think the court misidentified the source] submitted an expert report from Hal Poret, who conducted a consumer survey. One cell was shown the actual image and description of a Manley product as it appears in the ads and packaging. The other was shown an allegedly “accurate” image, and the same description. Both cells were shown images and descriptions of three other products, including an Aviva product. Participants then ranked the products and explained their reasons. In the group shown the ad, 69% would have selected the Manley product as their first choice and 10% would have selected the Aviva product as their first choice. In the group shown the allegedly "accurate" image, 25% would have selected the Manley product as their first choice and 28% would have selected the Aviva product as their first choice. The court found this testimony relevant and reliable circumstantial evidence of materiality with respect to the other Manley products.

As to injury and causation, Aviva’s burden depended on the type of relief it was seeking—injunctive relief, defendant’s profits, and plaintiff’s damages (also costs). Injunctive relief requires proof of likely injury and a causal link between the injury and defendant’s conduct, but not proof of specific damage. Because the Lanham Act protects consumers as well as competitors, courts shouldn’t be reluctant to grant injunctions to allow competitors to act as vicarious avengers of consumer rights. Poret’s study was evidence of likely injury causally related to the allegedly false advertising; that was enough for an injunction.

For money damages, plaintiffs have to prove actual damages causally linked to the false advertising, and damages must be awarded as compensation, not penalty. The difficulty of proving exact damages, combined with the necessity of showing record support for damages, means that the plaintiff must prove the fact of damage with certainty but not the amount. Aviva argued that the erosion of its revenues when Manley entered the inflatable market, plus Manley’s revenue growth, provided evidence of damages, as did the loss of Target as a client while Target continued to sell Manley products. However, in a noncomparative advertising case, evidence of plaintiff's decreased sales, defendant's increased sales, and plaintiff's loss of sales to the defendant may not sufficiently prove causation without a link to the specific false claims challenged.

Aviva also compared its increased sales of other products in similar markets to declining sales of inflatable pool and water products. The court found that this comparison did not prove that the difference was caused by Manley’s false advertising. Other factors such as competition from other sellers and Manley's greater physical presence on retailers' shelves or attractive licensing agreements may have made the difference. A jury would have to make too great an analytical leap to find that Aviva’s losses were caused by false advertising.

Why couldn’t the jury be aided by Poret’s study? It examined only one Manley product and one Aviva product, and was admittedly "not designed to quantify the number of sales gained or lost due to use of the Enlarged Image on Banzai products or to quantify damages due to use of the Enlarged Image." (Okay, but that’s quantification, not the basic fact of harm causation, which I thought was the point.) Still, the court found that it didn’t meet the heightened evidentiary requirements for recovering monetary damages. Comment: Manley succeeded here where it didn’t elsewhere, breaking down each item of proof and arguing that, on its own, that item was insufficient.

What about disgorgement of profits? This has to be compensatory, but doesn’t require a finding of actual damages. Courts in the Eighth Circuit are divided about whether willfulness is required for disgorgement, and the court here found that the 1999 amendment to the Lanham Act made clear that willfulness is not required, though it is a factor to be considered. For false advertising, a plaintiff must establish defendant’s sales of the products at issues, and the defendant bears the burden of showing deductions, including sales not due to false advertising, just as with infringement. Nothing in the Lanham Act suggests that the rules should be different for trademark infringement and false advertising.

The court found that Aviva could make out a case for disgorgement. It had sufficient evidence of likely damage, and of Manley’s sales and profits from the products at issue.

Manley also moved to exclude the testimony of Aviva’s experts, and Aviva did the same with Manley’s rebuttal experts. This led to more detail about Poret’s study: 60% of the respondents who saw the actual ad who chose Manley as their first choice cited size as a reason for their choice. In the altered image group, only 45% gave size as a reason for choosing the Manley product first. Among those who chose the Aviva product as their first choice, size was a reason for 47% in the actual ad group, whereas 81% cited size in the altered image group.

Manley argued that the study was not relevant because Poret didn’t test consumer perception of any Manley ad taken as a whole. The court disagreed. “The information accompanying the images included descriptions of the products, dimensions, and price--the same information a consumer would see when viewing the product on a website (minus the consumer comments, ratings, and consumer-posted images).” Manley argued that Poret should have showed participants entire packages or full internet ads, but Poret explained that this would have been infeasible (there was no way to create a good control group with a non-manipulated image) and unreliable (because actual internet ads include customer comments, which distort responses to the ads). Manley didn’t identify any critical omissions, and anyway expert testimony on the image alone would be relevant to a jury’s determination of literal falsity.

Manley also argued that Poret’s testimony was irrelevant because he used information from the internet, when Manley’s products are primarily sold in stores. “Poret's study was designed only to demonstrate the effect that the product image had on consumer decisions. Whether the consumers saw the image in the store or on a website does not affect the relevance Poret's study or conclusions.” At most, this went to weight and credibility, not admissibility.

Manley then challenged the survey’s reliability, first arguing that Poret was wrong to take the altered image as an accurate depiction of Manley’s product, because he didn’t create the image himself or know how it was created. All he did was “eyeball” it in comparison to unspecified photos on the internet. The court found that Poret’s assumption that the “accurate image” was at least more accurate than the Manley image was not fundamentally unsupported. Aviva’s experts explained how they created that image, using known anthropometric data and basic mathematical formulas. Poret’s eyeball test was also some evidence that his assumption was warranted. “Perhaps most tellingly, Manley has not presented any evidence that the ‘Accurate Image’ was anything other than accurate, at least with respect to the size of the product.” (Again I think that the smart money is that Manley should consider writing a check. And consulting its advertising injury coverage.)

Manley then identified six additional variables between the two images: the actual size of the image, the number of children using the product, the number of children actively engaged with the product, the background and landscaping, the coloring of the product, and the image of a girl whose feet were apparently chopped off. Poret didn’t think these differences were important, and Manley argued that he couldn’t know that for sure. The court found these criticisms relevant to weight, not reliability.

Next, the court rejected Manley’s argument that Poret failed to consider other factors, such as product features, colors, and prices, affected consumers. To the contrary, Poret acknowledged that these features mattered. “He did not opine that size was the only factor responsible for consumers' purchasing decisions; he merely concluded that size was the only factor responsible for the difference in purchasing decisions between the Control Group and the Test Group.” By keeping the descriptions constant, he focused on the effect of apparent product size. The Lanham Act doesn’t require the falsehood to be the only factor consumers care about. “The fact that there was a difference between the product choices among the two groups, when the only variable changed was the image of Manley's product, is reliable evidence that consumer decisions were materially affected by the image, regardless of the other factors.”

Manley then argued that Poret’s testimony was unreliable because he didn’t compute an error rate. Poret explained that, for a non-probability study, an error rate is inappropriate. Manley’s rebuttal expert testified that it could have been computed, but not that it should have been, given the type of study. Moreover, the methodology was reliable: “Manley identified no problems with Poret's sampling plan, shopping mall selections, respondent selection, interviewing process, check-in procedures, or interviewing period.” This was a randomized double-blind study of 303 participants, asking appropriate questions.

Aviva also offered the testimony of Dr. Michael Houston, an expert in consumer behavior and marketing whose research has focused on the impact of visual components on consumers. He opined on the deceptiveness of Manley’s ads and the resulting improvement in Manley’s market position compared to Aviva, as well as on the effect of dissatisfaction with Manley’s products on trust in the entire product category. The court agreed that there were “serious concerns” about the factual basis for Houston’s opinions, because he kept no record of the products he viewed, the websites he visited, or the number of consumer comments he read and was unable to reconstruct them in deposition. “This lack of factual detail and documentation is concerning, to say the least.”

Houston drew broad conclusions based only on the two consumer depositions and various online complaints, but he didn’t know what the people who didn’t complain thought or what fraction of sales the complainers represented. By Manley’s calculations, the gripers constituted less than 0.008% of the volume of sales of the ninety-five products. (I should note that the black-letter law is that complaining is difficult and most people don’t bother, so even a few complaints can be evidence of a deeper problem. Perhaps the internet is changing this some, but most people still can’t be bothered.)

“Houston had no factual basis for forming conclusions related to non-complaining consumers.” He couldn’t know that, as he testified, many other consumers would have reacted the same way, or how many would have done so. (Why isn’t his general expertise on the effects of visuals relevant here? I should note that the deposition portions quoted sound like he was pretty vague in his responses, but this seems like a preparation problem rather than something inherent in this kind of testimony.) He also failed to account for other factors relevant to consumers’ purchasing decisions, such as price and availability (that is, Manley’s better distribution system). He also didn’t know enough about the parties, and other competitors, to testify that Manley’s market position let it harm Aviva. Moreover, in terms of consumers being driven away from the product class, he testified that dissatisfaction with quality was more likely to drive consumers away and that consumers were also dissatisfied with the quality of Manley’s product, not just with the size.

In the end, “[w]hile sometimes ‘observations and highly specialized expertise may suffice to form the basis of an expert opinion,’” Houston failed to "explain how that experience leads to the conclusion reached, why that experience is a sufficient basis for the opinion, and how that experience is reliably applied to the facts." After his specific conclusions were excluded, all that was left was “essentially an academic lecture about the importance of visual information in marketing--that ‘[a] picture is an effective way to convey product information.’” This was a matter well within the knowledge or experience of lay people and thus superfluous. (As I argue in a forthcoming piece, ordinary people are actually quite unaware of just how much pictures influence them. Again, this testimony might have been badly configured, but I can see how expert testimony in this vein would have offered the jury something it didn’t know.) His testimony was excluded.

Aviva also offered the testimony of Dr. David Krauss, an expert in human perception and the science of measuring sizes and ratios in the human body, "to provide a scientific analysis of various Manley Toys products and packaging to determine whether the images that appear on product packaging misrepresent the actual product sizes." Krauss physically examined 26 Manley products and analyzed product packages for 36 more. For the 26, he measured the dimensions of the products as they appeared on the packaging and then their actual dimensions. He then used that scale factor to figure out whether the children were appropriately scaled. He did that by measuring the visible body dimensions of the children on the packages and comparing the extrapolated size of the child to the average size of a child of the estimated age. To deal with variability, he used the lowest 5th percentile (so that 95% of the relevant population would be larger), and then added further wiggle room to reduce the effect of measurement errors. He did similar things with the 36 products for which he only had pictures, using other similar products to get dimensions.

He concluded that 74% of the packages misrepresented product size by shrinking the children relative to the products. Young children (5-10 years) were often shown toddler- or baby-sized. Manley argued that Krauss wasn’t qualified because he’d never used this approach to determine whether ad images were distorted. The court was unimpressed. Krauss used similar methods in all his work, and had specialized knowledge in applying these principles and methods to image analysis. Specialized knowledge in advertising or marketing was not required.

Manley also challenged the reliability of Krauss’s investigation because his methods had never been used before in a false advertising case and because his cutoff for a misrepresented image was arbitrary and not generally accepted in the scientific community. Moreover, Manley argued that he made too many judgment calls, had too small a sample size, and didn’t use a double-blind procedure, among other things. Nope, said the court: “Krauss's study basically involved measuring images of children and products on a box, measuring the actual products, and then, using basic arithmetic, determining what the actual size of the children would be if placed next to the actual products. Krauss then determined whether that size was below the 5th percentile for a child of that approximate age.” This specific application might have been novel, but the underlying principles were generally accepted and widely used. “The novel application of accepted, published, reviewed, and tested techniques to a new set of facts does not make the analysis unreliable.”

Manley argued that Krauss’s cutoff for identifying misrepresentation wasn’t linked to consumer deception, but that missed the point. The cutoff wasn’t designed to determine any effect on consumer behavior. It was merely used “to conclude with some degree of statistical certainty that the images of children were reduced in size compared to the products.”

The court excluded the testimony of Aviva’s expert Frances McCloskey on Aviva’s damages, for the failure to make the causal connections discussed above. However, her testimony on Manley’s profits on fixed-air products survived. Disputes over whether certain products were actually different from the allegedly falsely advertised products were factual issues for the jury to resolve.

Aviva also challenged Manley’s rebuttal experts challenging Poret’s survey, Krauss’s report, and McCloskey’s conclusions. Aviva argued that, because the rebuttal experts only criticized Aviva’s experts and performed no analysis of their own, they identified no reliable principles or methods; they merely identified flaws without establishing whether any of those flaws would have provided different results. However, district courts have held that rebuttal expert witnesses may criticize other experts' theories and calculations without offering alternatives. The court found that these rebuttal experts’ testimony was not unduly speculative and would help the jury weigh the evidence presented; it was too early to decide whether their testimony would be cumulative to cross-examination at trial.

No comments: