Previous ruling blogged here. Rocky alleged that Red Wing markets its boots in a manner falsely suggesting they’re made in the US, when some of its lines are made wholly or partly overseas. Red Wing moved, unsuccessfully, to exclude Rocky’s survey and related expert testimony.
The survey was designed to figure out whether consumers believed that the Irish Setter, Red Wing, and Worx brands are made in the US and to determine consumer choice between two boots identical in all respects except country of origin. Awesomely, the survey was conducted at four knife and gun shows in Pennsylvania and Tennessee, limited to people over 18 who’d bought hunting or work boots or were likely to do so in the next 12 months.
There were three questionnaires. The first one showed an Irish Setter boot with labels and tags stating “made in the USA with imported materials.” The respondents were then asked, “Where do you think this product is made?” and followups. The second showed the same boot, but with references to “with imported materials” removed. The third showed a page from one of Red Wing’s catalogs and asked them “Where do you think shoes from this company are made?” and followups. For all three, respondents were asked to assume that there were two brands of boots identical except that one was made in the US and the other was made in another country. They were asked which they’d purchase (with no-preference and don’t-know options).
Because almost all surveys can be criticized (ed: almost?), flaws generally go to a survey’s weight rather than its admissibility, as long as its probative value isn’t outweighed by the prejudice, waste of time and confusion it would cause at trial.
Red Wing argued that the universe was overinclusive because it included past purchasers, not just potential purchasers. In trademark cases, a consumer confusion survey should consist only of potential purchasers, not past purchasers. But, though there’s not much precedent, the same rule may not always apply in a false advertising case. (Why? The court doesn’t discuss the rationale.) Cross-examination can address alleged overinclusiveness, which goes to weight rather than admissibility.
Red Wing also argued that the survey was underinclusive for covering only a narrow segment of the relevant market, gun and knife show attendees in two states, and that surveys should have been conducted in states with larger, more diverse populations. The court thought this was second-guessing. There was no evidence that gun and knife show patrons don’t represent likely purchasers, or that Pennsylvania and Tennessee were unrepresentative, and again this goes to weight.
Further, Red Wing argued that the survey should have been conducted in a retail setting such as a shopping mall, using a pair of boots that respondents could have tried on, with other footwear on hand for comparison, and sales staff present to answer questions. (In other words, Red Wing wanted a shoe store, not a survey.) Rocky argued that testing for country of origin preference didn’t require the same degree of replication of market conditions as testing for consumer confusion in an infringement case. Moreover, for the first two questionnaires, the boot was on a table and respondents could pick it up and examine it for as long as they wanted.
In what I consider an unnecessary holding, the court agreed that “the standards for replicating market conditions for a consumer confusion survey in a trademark infringement case cannot be applied wholesale to this false advertising/designation of origin case in which the survey was designed to measure consumer perceptions about country of origin.” What is certainly true is that, “[g]iven the purposes for which it was designed, the … survey satisfactorily replicates marketplace conditions.” The here-it-is, pick-it-up-if-you-want setup has been accepted in a lot of trademark cases. And certainly, depending on the nature of the infringement claim, surveys can do better if they offer an array of products, but that’s often because the plaintiff’s mark isn’t so strong that the respondents can be expected to think of it just by being presented with the defendant’s similar mark. In addition, sometimes the products can be expected to be presented side-by-side in the actual marketplace, which can increase or decrease likely confusion (think store-brand generics with similar trade dress to national brands), in which case the survey should reflect that. It’s absolutely the case that the presence of other brands is less directly relevant to whether the message on these boots is false or misleading (though if other brands bear country of origin labeling, that may affect what consumers expect from boots), but I’m not sure that one should say generally that the trademark and false advertising survey standards differ. Rather, the survey should replicate relevant market conditions; the presence of other brands is not necessarily relevant, depending on the claim at issue.
Red Wing also argued that the first two questionnaires were irrelevant because 85-90% of Irish Setter boots are clearly labeled “made in China.” Rocky conceded that the results couldn’t be extrapolated to those boots, but the survey was relevant to the 10-15% that are labeled similarly. Red Wing further claimed that the second survey used a “defaced” product that people would never encounter in the marketplace, but it’s still relevant because the difference between the first and the second measures the difference between boots labeled “made in USA with imported materials” and “made in USA.”
Red Wing then argued that the third questionnaire was irrelevant because it contained no designation of manufacturing origin and only tested respondents’ assumptions about Red Wing and Worx. But that was the purpose: to place before respondents the language “Red Wing Shoe Company, Inc.,” “Red Wing Shoes Since 1905,” and “WORX by Red Wing Shoes,” testing whether they believed that the brands were made in the US. This was relevant to the case.
As for the materiality question, Red Wing argued that it assumed what it sought to prove: that country of origin is material. Given the “no preference” and “no opinion” options, the court disagreed. There was a little bit of order bias, because the US option was first, but the court didn’t consider that fatal.
Red Wing contended that the questions in the first two questionnaires were leading and suggestive because the boots had tags saying “made in USA,” and thus the questions amounted only to a reading test. The court disagreed that the questions were leading, though they were suggestive given the labels and tags. Further, the question in the third questionnaire was leading because it asked respondents about “this company,” not permitting respondents to distinguish among Red Wing the company, Red Wing the brand, and the WORX brand. But this was for a jury to weigh. (What’s wrong with a suggestive question if it’s not leading? It’s true that, by focusing respondents’ attention on particular messages, you probably get a higher uptake in surveys than in the “real world,” but that’s true of all messages, and marketers ask focusing questions all the time to help advertisers make very significant decisions; the fact that most people don’t pay any attention to ads isn’t a compelling reason to allow falsity/misleadingness to persist in those ads to deceive those who do look.)
Red Wing then argued that the entire survey was ambiguous because it failed to define “made.” The FTC has determined that “made in USA” means that all or virtually all of a product is made in the US, but the court hadn’t yet ruled on whether it would instruct the jury on this standard. “[T]he FTC standard is not so much a legal definition as it is a measurement of what the FTC determined consumers believed ‘made in USA’ meant when the FTC conducted its research more than a decade ago,” and the FTC standard isn’t binding on the court or the jury. This is simply evidence of the FTC’s opinion, which the jury can give any weight it deems appropriate. (This seems to me underweighted—especially if there’s no better evidence, this is exactly the type of issue on which a court should defer to the FTC’s expertise; in addition, by enforcing this standard, the FTC has done more than measure: it has intervened in the market, shaping what consumers can expect.) Regardless, the court rejected Red Wing’s argument, pointing out that Red Wing would have fought any definition in the survey just as hard, since its argument is that “made in the USA” is subject to varying interpretations. The survey didn’t ask respondents what “Made in USA” meant to them, but that doesn’t make the survey unreliable or inadmissible.