Legal Applications of Marketing Theory, part 5 (me on puffery)

Rebecca Tushnet, Harvard Law School, On Puffery

Puffery is a concept that purports to be about things consumers ignore and don’t rely on. It is in fact a concept about things courts ignore and won’t rule on.  At the moment, marketing and other empirical work has essentially nothing to say about puffery in the courts; puffery consists of precisely the elements of advertising for which courts neither require nor allow empirical evidence of consumer reaction.   That doesn’t make the doctrine wrong, but it does mean that explanations for the doctrine should not be founded in unsupported, mostly unsupportable judicial claims about how consumers think and what claims they disregard.  Instead, puffery should be about what kinds of claims are too difficult to evaluate for their truth in judicial settings.  That’s an epistemological determination that judges are actually well qualified to make.

Let me back up a bit: In modern advertising law in general, only factual misstatements are actionable.  Puffery is an overlay onto the fact/opinion divide: it allows courts to reject liability for what might look like factual, verifiable claims (such as the cheapest prices in the universe, or even the cheapest prices in West Virginia ) because they are too exaggerated or vague to be believed by reasonable consumers.  The Fifth Circuit wrote, for example, that “non-actionable ‘puffery’ comes in at least two possible forms: (1) an exaggerated, blustering, and boasting statement upon which no reasonable buyer would be justified in relying; or (2) a general claim of superiority over comparable products that is so vague that it can be understood as nothing more than a mere expression of opinion.”  The FTC has reasoned similarly. 

Courts think, without evidence, that consumers don’t rely on puffery.   This conclusion is also normative, and it has distributional consequences—it is about how consumers should behave, not about what advertisers should say. As Learned Hand wrote, “There are some kinds of talk which no sensible man takes seriously, and if he does he suffers from his own credulity.”  The influential treatise Prosser & Keeton on the Law of Torts says that an advertiser has a privilege “to lie his head off, so long as he says nothing specific.” 

Ivan Preston argues that current puffery doctrine is a mistaken evolution from nineteenth-century cases involving individual buyers and sellers that held that buyers couldn’t sue for fraud based on statements that they could easily have verified or disproved themselves.  When buyers were unable to verify the claims, however, the law provided them redress. But as the puffery doctrine developed, he argued, it turned into a rule that consumers treated certain claims as meaningless and therefore rejected them at the outset.  He argues that this new rule was not only unconnected to its historical foundation in fraud law, but also was inappropriate for modern mass advertising where the complexity of factual claims combined with their sheer volume mean that consumers can’t actually investigate most of the factual claims they receive.

One Additional piece that has to be understood before the full scope of the problem is understandable: advertising law, like trademark law, is probabalistic.  If 25% of consumers (net of control) are confused or deceived, almost any court would grant relief.  Deception, that is, doesn’t need to be universal to be actionable by a competitor, or by the FTC.  Deception doesn’t even need to be the most likely outcome for a given target consumer as long as a substantial number of consumers are likely to be deceived. 

Resulting problems in the law of puffery: First, it is a problem for the conventional justification of the doctrine that puffery actually works, in the sense of getting consumers to buy things.  Even the FTC has accepted that puffery works: C&H Sugar was ordered in 1977 not to call its brand “superior” to or otherwise different from other granulated sugars without substantiation.  In 1995, it successfully argued that it shouldn’t be barred from using ads such as “I love C&H the best” or “C&H tastes best,” which harmed it because its competition was free to make similar unsubstantiated claims. The FTC granted the modification, because competing ad campaigns were able to “take advantage of C&H’s inability to counter claims that … constitute puffery. . . .”  But, of course, if C&H needed puffery to compete, then puffery was affecting consumer behavior.

If puffery didn’t work, we should probably expect it to be rare.  Courts occasionally make the point that advertisers both want to affect consumers and are likely have greater-than-average insight into what might affect consumers in the sale of their particular thing.  It seems like the setup to a joke: why do claims appear in ads?  The obvious answer seems to be: to get to the sale.  One thing we might do when advertising claims are challenged in court, then, would be to presume that a factual claim matters to consumers.  The key questions we should have would be about whether a significant group of consumers receives a factual message specific enough to be falsified.  [We could still recognize that there are parts of ads that aren’t claims as such—for example, elements of ads that functioned to attract attention, thence to deliver a factual claim.]  I would suggest that the concept of unbelievability adds nothing further to the question of falsifiability—it’s possible that a sufficient exaggeration means that no falsifiable factual message has been conveyed, but we really don’t know that at the wholesale level, without looking at the specific exaggeration and the market.

I want to work through two examples from the FTC to show what I mean when I suggest that the current conception of puffery as meaning something about actual consumer perception is not working very well.

First, the FTC Endorsement Guides.  If you think that puffery is about subjectivity and variation among consumers’ understanding and that reasonable consumers don’t rely on puffery, the FTC’s approach to endorsements shouldn’t make sense to you. The FTC takes the position that an endorser has to disclose connections to an advertiser when they wouldn’t be obvious from context and when knowledge of the connection would be relevant to the consumer in weighing the endorsement.  So far, so good. But the FTC—quite rightly, I think—requires disclosure even when the endorser is otherwise just offering her opinion: these clothes are so cool! This hair color looks fabulous on me! Failure to disclose the connection is deceptive where the audience is likely to believe that the speech is uncompensated opinion.  If the underlying claim is pure immaterial puffery of the kind on which consumers are irrebuttably presumed not to rely—and the underlying claim in a social media endorsement is often indeed exactly that kind of claim—how can it possibly be important to consumers to know that the endorser is being compensated?

The answer is that consumers, in general, want opinions to be in some sense authentic, and they care about whether a speaker is getting paid; her influence will be less if she discloses that payment.  The fact of the financial connection is itself verifiable, so its absence can be misleading, even if all that gets said in an endorsement is stuff that consumers weren’t supposed to be relying on as a matter of law. The endorsement guidelines thus inherently, if covertly, recognize that puffery does work.  Because puffery works, it is important to regulate undisclosed endorsements, whether or not they make other factual claims.  The lack of disclosure gives us a factual hook of sufficient specificity that the courts and the FTC can handle: determining whether there was in fact an undisclosed relationship.

Second, the FTC Green Guides. In puffery discussions, courts often say that vagueness matters: some words or statements are too vague to have one specific meaning.  That was the rationale, for example, in a case involving the claim “America’s Favorite Pasta.”  There are different possible meanings of vagueness: a statement might be too vague even for an individual to get a specific message in response to the statement, but the statement might also have a lot of varying interpretations among heterogeneous consumers, at least some of which have specific definitions in mind. The AFP court endorsed the latter view, saying that favorite might mean most-purchased, but it might also mean that people liked it best but couldn’t often afford it.  Yet if the thought is that different consumers will fill out words like “favorite” with different meanings, then we could if we put the empirical work in actually figure out what those meanings are and whether they’re shared across a substantial number of relevant cases.

Which brings me to the Green Guides: The FTC’s general rule for advertisers is that they have to substantiate factual claims that are conveyed to a substantial number of relevant consumers.  FTC, relying on its own research into the meaning of general environmental benefit claims (“green” and “eco-friendly”), found that substantial numbers of consumers understood a variety of things from those claims:

61 %: made from recycled materials;

59 %: recyclable;

54 %: made with renewable materials;

53 %: biodegradable;

48 %: made with renewable energy;

45 %: non-toxic;

40 %: compostable

27 %: no negative environmental impact.

As a result, the FTC said in its Green Guides that

(b) Unqualified general environmental benefit claims … likely convey that the product, package, or service has specific and far-reaching environmental benefits and may convey that the item or service has no negative environmental impact. Because it is highly unlikely that marketers can substantiate all reasonable interpretations of these claims, marketers should not make unqualified general environmental benefit claims…..

I think this is a correct treatment of vagueness that has multiple plausible falsifiable meanings. If a substantial number of consumers receives a sufficiently specific and false meaning, we should be concerned.

Thinking of puffery as being about consumer comprehension instead of administrability leads courts into mistakes. Example: In Date v. Sony Electronics Inc., 2009 WL 435289 (E.D. Mich. 2009), Sony advertised its television as offering “Full HDTV,” and “1080p” (the best available technology). The TVs, however, could not display a 1080p signal. Instead, at best they could display an upconverted 1080i (interlaced) signal from a 1080p device. The upconversion process results in undesirable artifacts like feathering that make the viewing experience worse.

Sony argued that its claims were puffery, based on a prior similar case. In Johnson v. Mitsubishi Digital Electronics America, Inc., 578 F. Supp. 2d 1229 (C.D. Cal. 2008), the court concluded that, although Mitsubishi designated its television set as a 1080p television set, the phrase 1080p “does not convey a specific claim that is recognizable to the targeted customer.” The Johnson court thought 1080p only had meaning for engineering professionals, and that all that the plaintiff wanted was a top-of-the-line set (top of the line is classic puffery). Because he didn’t understand what 1080p meant, the claim was puffery to him.

The Date court pointed out that Sony put the term on its specification sheet addressed to consumers, suggesting that it wasn’t puffery. But more evident, I would argue, was that 1080p had a specific meaning, and consumers didn’t need to know its technical requirements in order to be moved to act by it any more than they need to know how their statins work or why the drugs are called statins.  If a consumer receives a message that she thinks is factual, credible and material, even if she can’t be particularly specific about the details, then she can be harmed if that message is false.  And it was really easy to prove that 1080p was false as applied to the Sony TV.  But focusing on consumer understanding leads to errors like the Mitsubishi court’s.

Some preliminary thoughts about implications: First, I want to revisit the difficulty that puffery works: Puffery may be effective in influencing purchases without being either provable or falsifiable in conventional judicial terms. Thus, a determination that a claim is pure puffery should arguably trump evidence that it actually influences consumers—but only if the reason for finding puffery is the difficulty of proof of truth, rather than vagueness or multiple possible meanings. And, as I suggested earlier, exaggeration should be rejected as a separate defense or category of puffery; the question is always what factual message consumers are likely to receive, if any.  If the claim is “we’ll save you a million dollars on car insurance,” we can ask whether a substantial number of consumers receives a message that they can expect to save a significant amount compared to other insurers, and whether that message is false.  The lawyers’ fighting would of course shift to whether the inquiry into falsity was a manageable judicial task from current disputes over puffery—but at least we’d have better definitions, and courts forced to consider heterogeneous groups of consumers might even be moved to look more rigorously for multiple possible meanings in an ad, where appropriate.

What’s the proper boundary of falsifiability, then?  I think it should have to do with the difficulty of getting reliable results from consumers, or of figuring out what the possible specific factual meanings are.

This approach also has implications for judicial treatment of images. The Second Circuit has said that  Time Warner Cable, Inc. v. DirecTV, Inc., 497 F.3d 144 (2nd Cir. 2007): “Unlike words, images cannot be vague or broad.” While one standard definition of puffery—general claims of superiority that are so vague as to be meaningless—fits images badly, the other—“an exaggerated, blustering, and boasting statement upon which no reasonable buyer would be justified in relying,” could be applied.  This to me gets it absolutely backwards!  Not all images have completely transparent meanings; especially in ads, images need to be interpreted.  So images absolutely could convey a vague or broad meaning, depending on what the ad was doing. But the Second Circuit’s abandonment of verifiability and focus on reasonable reliance led it to judge consumers—and to find them wanting.  In Time Warner Cable, Inc. v. DirecTV, Inc., 497 F.3d 144 (2nd Cir. 2007), DirecTV ran Internet ads showing unwatchable TV images contrasted to sharp and clear images. The district court agreed with Time Warner that DirecTV’s own rationale for running the ads—that consumers were highly confused about the then newly emerging HD technology and needed to be educated that both digital equipment and digital signals were required to experience HD quality—was reason to think that consumers might rely on the ads. The court of appeals found that the district court clearly erred. The ads were not even remotely realistic, and the court found it difficult to imagine that any consumer, no matter how unsophisticated, could be fooled into thinking cable’s picture quality would be that bad.  I would suggest that’s a problem of the court’s imagination; a consumer might know that “ordinarily” her cable wouldn’t be anything near that bad. But why would a consumer transitioning to HDTV have been confident about what cable would look like when she attached an analog cable feed to her new HDTV?

A final thought on the role of cost-benefit analysis: One reason we might reject liability even if a significant number of consumers receives a false factual message from a claim is that the claim provides benefits to a different group of consumers, significant in size or in some other way.  That’s not a justification that requires a puffery defense, nor does a puffery defense obviously help explain what benefits the nondeceived consumers might be getting—by definition, if the issue is puffery then they’re not getting specific truthful information, though maybe they’re getting enjoyment from a cool ad.  A clearer understanding of puffery could help us when we ask “is there anything lost to the nondeceived group if we get rid of or reformulate the part of the message that’s causing the deception?”

Q: more discussion of online reviews by individuals. Their ability to puff all they want if unaffiliated may be some of their attraction.  Also compare to gov’t: cities can puff all they want about how they’re great places to live. If we regulate puffery in advertising, won’t people be vulnerable to deception by uncaught puffers?

RT: I don’t think there’s any evidence that the general advertising regime can teach consumers an appropriate level of skepticism—credulity seems to vary by demographics.  I want people at least not to be fooled when they buy TVs.

Q: cheap talk/costly talk in economics: at what point is information ignorable?

Q: remedies: research suggests you can’t combat puffery w/facts, only w/anti-puffery: “best pen on earth” has to be fought with “worst pen on earth.”

RT: similar to Green Guides, where even adding explanation “green: made with recycled materials” conveys a bunch of other green claims.

Q: endorsement: isn’t the problem that the person who’s selling is anticipated to puff, but if someone else is puffing, it’s interpreted as their unbiased opinion to which people can pay attention?

RT: but that still means that there’s actionable information in the semantic content of the words—to say consumers don’t rely on puffery is just wrong.  It’s true that the deceptive part is the failure to disclose, but it then affects whether the puffery is “credible.”