Thursday, October 22, 2009

Arnold & Porter Webinar on False Advertising

False Advertising Disputes: Benefits, Challenges, and Risks of Making and Defending Against False Advertising Claims, hosted by Arnold & Porter LLP

Mary Engle, FTC Associate Dir. for Advertising Practices

FTC’s priorities: national advertising. Health claims, functional foods, green marketing, endorsement/testimonials. Health claims are becoming more prevalent in food ads, drawing increased scrutiny. Settled claims with Kellogg over Mini Wheats: be careful when transforming study results into ad claims. Airborne: consent judgment over, inter alia, misleading implied claims made through visual depictions. Also Rite Aid and CVS which sold store brand Airborne knockoffs.

Kellogg: “Clinically shown to improve kids’ attentiveness by nearly 20%”—compared to kids who had only water for breakfast. Only half the kids showed any improvement in attentiveness at all; only one in 7 improved by 18% or more, and 1 in 9 20% or more. Study was presented misleadingly! Subsequent version of ad included verbal disclosure that this was compared to no food, but the FTC didn’t think that was enough.

Airborne: Challenged the claims that Airborne prevented colds: clearly implied from depictions of germs spreading, reference to Airborne as a defense. Earlier ad versions had made express cold prevention claims, but even after the ads were toned down they implied the same thing. Settlement: up to $30 million in refunds, including private class action settlement. Commissioner Rosch’s dissenting statement: didn’t support allowing Airborne to run out existing packages for several months, and believed that FTC should have addressed assertions of Airborne’s “immune-boosting” qualities and should have required corrective advertising to dispel lingering misimpressions.

Improvita: Airborne knockoff for Rite Aid and other retailers, still pending. FTC settled with Rite Aid for Germ Defense ($500,000 in consumer redress, stores posting refund notices now) and CVS for AirShield ($2.8 million in consumer redress). These complaints did challenge immunity-boosting claims and didn’t have run-out provisions for existing packaging.

Functional foods: boost the immune system, assist with brain function, protect the heart, etc. FTC is concerned with non-FDA-approved claims—need strong science to back them up. Standard FTC injunctive provision prohibits claims unless based on competent and reliable scientific evidence, but that standard is not sufficiently precise. We will be crafting new injunctive language in future orders. More precision will increase ease of enforcement and harmonize with laws and regs administered by sister agencies like FDA. Address situations where a single piece of research, though conducted according to established protocols, gets results inconsistent with scientific consensus.

Green marketing: conducted consumer research to see how to update the Green Guides. Expect to issue revised Guides for comment in 2010. Clothing/sheet sellers deceptively advertised products as made of natural bmboo fiber, when they were made of manmade rayon from bamboo, which is an environmentally destructive process that mostly occurs in China because US regs are too tight. Challenged claims the products were manufactured using an “environmentally friendly” process and were biodegradable; also included Textile Fiber Act counts: you have to say that the end product is rayon; can’t just say it’s bamboo.

Kmart 2009 consent order: challenging biodegradability claims for paper plates. In modern landfills, paper doesn’t biodegrade.

FTC endorsement/testimonial guides: Least-noticed principal change—requiring disclosure when advertiser paid for a study touted in the ad.

Typicality: testimonials need to do more than say “results not typical,” because that didn’t work. An endorsement on an essential/key attribute will be interpreted as representing that the endorser’s experience is representative of what consumers will generally achieve with the product in actual, albeit variable, conditions of use. So, unless there’s substantiation for that, the ad should clearly and conspicuously disclose the generally expected performance in the depicted circumstance; no longer sufficient to disclose limited applicability of the endorser’s experience, because that wasn’t working. The net impression of the ad controls.

Social media marketing: pretending to be a customer and giving oneself a good review violates the FTC Act. And the guides require disclosure of material connections between seller and endorser, where the audience wouldn’t reasonably expect the connection and it would affect the weight or credibility of the endorsement. Examples: seller is compensating endorser; endorser is employee/buisness associate of seller; endorser is related to seller. When does a consumer become an endorser? When, viewed objectively, the consumer is being sponsored by the marketer. Is speaker acting solely independently, or is the speaker acting on behalf of the advertiser or its agent such that the speaker’s statement is an endorsement that’s part of an overall marketing campaign? If an advertiser sends me free samples as part of a marketing campaign and expects me to talk it up on my blog, that can be an endorsement. Linedrawing may be tough, but that’s always true—advertorials pose the same issues.

Nonexclusive factors: did the advertiser compensate the speaker, provide product for free? What are the terms of any agreement between them? What’s the length of the relationship? Did the speaker previously receive free products? What’s the value of free products received? Overall, is this part of a marketing campaign? Only then are disclosure requirements triggered.

David Mallen, Associate Director, NAD

NAD as alternative to litigation, though can’t get you a preliminary injunction. What’s new? Aggressiveness in comparative advertising, resulting in an uptick in NAD challenges. Implied claims/puffery. Green marketing. Social media.

Often a battle of the tests—when S.C. Johnson’s Glade comparative performance claims were challenged, for example. NAD tries to offer guidance on substantiation.

Sensory testing: Kraft Foods, #4915—claim was that Tombstone pizza was preferred over Red Baron; mouseprint said it was a pepperoni comparison. Two issues: if the test is only pepperoni, can you make such a broad claim? Parties make a variety of pizzas, and pepperoni wasn’t the majority sold. Second, how do we conduct comparative sensory tests? Basic issues: is there a standard industry test; is the methodology sound; did the study test the actual products at issue (often a problem with dietary supplements where the test was only on one ingredient and extrapolation is difficult); can a correlation be drawn between the results and the challenged claims; are the parameters of the test consumer relevant?

Puffery: don’t suggest health benefits without adequate substantiation.

Green marketing: most claims are faith-based advertising: consumers have no choice but to trust; they can’t verify. If they can’t trust those claims, it will undermine faith in ads as a whole. Confusion over substantiation, meaning of claims to consumers. One case: advertiser relied on its supplier’s certificate of biodegradability. Not good enough. “Better for you” claim overall on the basis of one attribute—that’s not a good idea. Clorox Liquid Laundry: “more sensible for the environment” because a plant-based surfactant; that was true, but that doesn’t mean that overall the product had substantiated a claim that it was more sensible for the environment—it’s still a complex chemical with difficulty breaking down, and there was no lifecycle analysis that could support a broad green claim.

Social media: big challenge for NAD. Used to be easy to figure out what advertising is. Dyson, Inc. #4619: a YouTube clip of a product demo, performing really well and showing the competition performing badly. NAD made itself clear that it would consider these types of things “advertising.” Will try to harmonize with FTC guidelines and provide consistent message to industry.

Blogger disclosure: weknowdiets.com—“We have compiled the most comprehensive database of information for people who are looking for a trimmer body and healthier lifestyle.” Gives MiracleBurn an “editor’s choice” award one week after another—this is actually owned and operated by MiracleBurn. A clear violation of the old testimonial guidelines as well as the new ones. NAD plans on following this issue.

Michael Mazis, prof. of marketing, AU

Key issues in false advertising surveys: universe; open v. close-ended questions; “noise”; filter questions; control questions v. control ads.

How do attorneys get into trouble? Clients are often emotionally upset by competitors’ ads. Surveys are blunt instruments. People don’t retain all that’s in an ad; take away one or two messages. A subtle message can’t be assessed well by a survey. Tendency for attorney/researcher to push the envelope and include leading questions.

Memory v. stimulus-based surveys. Memory: Eveready confusion (look at this product: who makes it? Requires consumers to have brand in memory), secondary meaning, and dilution surveys. Universe is crucial! If you define it too broadly, people won’t have the product in memory. Stimulus-based: show the ad; the universe is important, but not as important, because memory isn’t as crucial, so universe definition will often go to the weight of the survey.

Open ended questions: preferred by courts; less likely to be biased, but may produce general responses. Closed-ended: greater potential for bias, but likely to produce more specific findings. Most surveys have both types. Ideally, you get consistency: people play back a claim in the open ended questions and again in the close ended questions (comment: the Mylanta Night Time Strength case was like that). What if you get little or no playback in the open-ended questions, but strong playback in the close-ended ones? People may suspect bias, and you’ll need an explanation for the inconsistency.

Exmaple: AbForce, FTC v. Telebrands from 2004. Claim: electrical contraction of stomach muscles; reminded consumers of commercials for other ab belts; no express claims about weight or fat loss; showed toned models. The claims were all implied. Open ended questions: 22% said product will give you flat abs, lose weight/fat, gets you in shape; closed-ended said it would result in well-defined abs (65%), lost inches on the waist (58%), lost weight (43%), alternative to exercise (39%), and so on. Is the AbForce ad deceptive? Possible objections: (1) there was yea-saying to closed-ended questions. (2) Just relying on preexisting beliefs about other ab belts—the earlier beliefs are deceptive, not this ad. (3) Nondeceptive elements of the ad might be biasing the results.

Alternative explanations are noise: extraneous factors that make it impossible to state that the ad caused the observed results (here, deception). Controls can help. Filter questions: eliminate respondents who are guessing or yea-saying. So: “Did the ad, say show or imply that Ab Force improves users’ appearance, fitness, or health?” Anyone who said yes went into the close-ended questions, but anyone who said no didn’t. Filter questions cause respondents to be dropped from the survey. Overfiltering is a defense strategy: the fewer respondents answer a closed-ended question, the less likely you are to find deception. Quasi-filter: “Did or didn’t the ad say, show or imply … or don’t you know?” Necessary but not sufficient. “Don’t know” gets an average 25% takeup—a more conservative way of doing the research.

Control questions: items not in the ad yet plausibly associated with the product. Ab Force: lowers blood pressure (6%), relieves stomach ucler pain (5%), relieves nausea (4%)—subtract these results to get net deception. Good, but not enough.

Best practice: control ads. Control for preexisting beliefs; control for nondeceptive elements of ad. Goal: show respondent a nondeceptive ad as close as possible to the allegedly deceptive ad. Modified ad: purge misleading elements and correct any misimpression.

Easier to do with a print ad, which can be digitized and modified; harder with a TV ad. Depending on the ad, there may be little remaining of the ad.

Vermont sued RJ Reynolds for Eclipse advertising: a cigarette that “may present less risk of cancer, chronic bronchitis and possibly emphysema.” RJ Reynolds relied on the word “may.” How many people played back the concept of “may”? Hardly anyone even saw “may.” Words like “cancer,” especially for smokers, are the ones they pay attention to.

Two controls: one ad was “a cigarette that presents less risk of cancer, chronic bronchitis and possibly emphsyema.” No difference in reception. Then a really meaningless ad with little claim left; didn’t convey much.

Disclaimers: most common in creating controls for TV ads. Correcting alleged deception with an additional statement. Potential problems: what should the statement be? Is the message effective? Ongoing case: FreeCreditReport.com – case between Experian and LifeLock. What does “free” mean? It’s free if you sign up for a paid program. Control ad with disclaimer: to get your free credit report you must sign up for a trial membership. If you don’t cancel within the 7-day trial period, you’ll be billed $14.95/month. Difficulty with proving a negative: did the ad fail to communicate conditions associated with the free credit report. Pretty significant differences in open-ended and close-ended responses both; people did notice the disclaimer.

If you can’t fix the ad, use a different ad for the same product, as done in the Kraft v. FTC milk content case. Issues: must have an available nonmisleading ad; which ad to use?

Finally, could use a different brand’s ad—used when no control is available. Doan’s: advertised for decades that it contained a special ingredient that other brands didn’t have, but it was just an aspirin variant. Problems comparing, but best they could do.

There is no cookbook for studies; they each have to be designed. Control questions/filters are useful, often necessary, but you really need a control ad to control for prior brand beliefs. Trade-offs exist in selecting control ads. There’s no substitute for a knowledgeable, experienced, and independent expert. A bad expert will come back to haunt you; courts aren’t stupid.

Randy Miller, Arnold & Porter

Lanham Act is useful when speed is critical. Courts are more willing these days to order advertising stopped, because of a body of caselaw allowing injunctive relief. Listerine “as effective as floss”; DirecTV; Splenda; Tysons Food “no antibiotics” case—all granted injunctions requiring ads to be pulled. Sometimes, as a result, the C&D approach works. Always call competitor first; you won’t get a TRO unless you’ve gone to your competitor and asked it to stop.

Cheat and retreat: run a disparaging ad and then stop and claim mootness: the Gatorade/Powerade “incomplete sports drink” dispute. Voluntary cessation won’t always help, though; the Tyson’s court said that if you’ve already stopped you should have no problem with my injunction not to restart.

No comments: