Remarks of Keith R. Fentonmiller, Attorney, Federal Trade Commission (FTC), Division of Advertising Practices

At the CMOR Government Affairs Workshop in Washington, D.C. on September 6, 2005

I’d like to thank Brian Dautch and the Council for Marketing and Opinion Research for inviting me to this Government Affairs Workshop. 

As many of you know, the FTC enforces federal consumer protection laws that prohibit fraud, deception, and unfair business practices.  Among other things, we seek to prevent material misrepresentations in advertising. We’re interested in insuring that advertising does not make false claims or fail to disclose facts and thereby mislead consumers.  In short, I’m in the disclosure business.  So, I think it only appropriate that I begin my presentation with a disclosure: The following views are my own and do not necessarily represent the views of the Federal Trade Commission or any individual Commissioner.

Like your industry, the FTC has long recognized the importance of survey research.  In fact, market research informs my day to day work.  When I’m sporting my false advertising enforcement hat – Did you know that there’s a false advertising enforcement hat?  Well, there is.  It’s kind of like the magical sorting hat from the Harry Potter stories. 

Anyway, when I’m sporting the false advertising sorting hat, the question I often face is what claims an advertisement makes?  Sometimes, the claims are obvious on their face – for example, “Just strap on this bracelet and lose 50 pounds in a month without diet or exercise!” or “Cure your cancer and heart disease by taking this simple pill!”  And the falsity of these claims is just as clear.  The sorting hat has a very easy time with these types of claims.

But sometimes the claims are not so obvious.  Sometimes, ads contain implied claims.  That is, when one looks at the ad as a whole – the ad copy, the product name, imagery – one  can take away claims that are not stated on the face of the ad. 

Take the case that the FTC brought against Stouffer Foods.  Stouffer had advertised its Lean Cuisine frozen entrees as skimping on sodium and containing less than one gram of sodium.  The express claim – that each Lean Cuisine entree has only one gram of sodium – was true.  But the words “skimp” and “less than,” combined with other claims about low calorie and fat content, conveyed the implied claim that Lean Cuisine entrees are low in sodium.  The ads failed to disclose, however, that sodium intake is measured in milligrams, not grams.  One gram contains one thousand milligrams.  Someone consuming 1 gram of sodium would be consuming nearly half of the recommended daily allowance.  That is a material misrepresentation for people watching their sodium intake.

So, how did the FTC determine that the ad actually made the implied representation of low sodium?  Often, the FTC may on its own reasoned analysis to determine what “reasonably clear” implied claims are conveyed by examining the “overall net impression of an ad.”   In other words, we can rely on the sorting hat.  But if the implied claims may not be determined with confidence from the face of the ad, extrinsic evidence must be examined, including consumer surveys and expert testimony.   In the Stouffer’s case, the FTC relied on copy testing to prove that consumers took away a low sodium claim from the ads.
Consumer research also comes into play at the remedy phase of our enforcement efforts.  As part of a consent order, an advertiser might be required to provide a particular disclosure in order to prevent a claim from being deceptive.  On rare occasions, we might insist on corrective advertising,  requiring the advertiser to include an affirmative statement to correct a false belief that was created or reinforced by the challenged advertising.
In crafting these informational remedies, consumer research may assist in not only the wording of these disclosures, but also their location and prominence in the ad.  One recent example of agency sponsored research in this area includes the Bureau of Economics’ working paper on Consumer Perceptions on Qualified Health Claims in Advertising.  The paper explores whether advertising disclosures can communicate – clearly and nondeceptively – limitations in the level of scientific support for claims about the relationship between the consumption of dietary supplements and the prevention of diseases.  Also, last year, the Bureau of Economics surveyed over 500 recent mortgage customers to examine the understandability of mortgage broker compensation disclosures proposed by HUD.  We found that the proposed disclosures were likely to confuse consumers.The FTC also relies on consumer research in the policy context.  In the course of monitoring the self-regulatory systems that restrict the marketing of violent entertainment media to children, the FTC has commissioned surveys of parents and children to assess their knowledge and use of the rating systems.   The agency also has conducted a survey of  rent-to-own customers  to determine whether abusive collection practices are widespread in the industry. 

In short, the FTC has found consumer research to be a critical tool in shaping our enforcement efforts, assessing the efficacy of different types of informational remedies, and in uncovering unfair or deceptive business practices.  More generally, I believe that the FTC recognizes the potential economic value of legitimate, industry sponsored consumer research that is conducted in an appropriate manner.  Companies have a powerful fiscal incentive to gather information from consumers about their needs and preferences for goods and services.  Armed with this information, companies can more economically target their products to a receptive consumer base and even develop new products to fill gaps in the marketplace.  Bona fide consumer research can make for more efficient markets.

It is with this shared appreciation for the high value of consumer research that I address you today.  And it is my goal, as someone who enforces laws that touch on issues of free speech, to enlighten you and, perhaps, assuage some of your concerns about the regulation of your industry from a commercial speech perspective.

It’s my understanding from Brian that members of your industry are concerned about regulators treating survey research as a form of commercial speech.  Your main concern seems to be that regulators may try to use the CAN-SPAM Act and the Do Not Call provisions of the FTC’s Telemarketing Sales Rule in ways that will infringe on your ability to conduct survey research.  From my perspective, this concern has two components.  First, the concern that market research is entitled to a lesser degree of constitutional protection because it is somehow commercial in nature.  And, second, because of this purported commercial nature, it arguably falls within the scope of CAN-SPAM and Do Not Call regulations.
Let me preface this discussion by giving you the short answer:  your fear is unlikely to become reality if you continue to conduct bona fide survey research – in other words, research not likely to be viewed as promoting a product or service.

I’ll tackle the statutory question first, because if survey research does not even meet the statutory definitions of regulated speech, the question of whether survey research is commercial speech under constitutional law principles is irrelevant.

As you all know, the National Do Not Call Registry is a list of phone numbers from consumers who have indicated their preference to limit the telemarketing calls they receive.  To implement Do Not Call, the FTC amended the Telemarketing Sales Rule, which covers any plan, program or campaign to sell goods or services through interstate phone calls. 

Is market research such a plan, program, or campaign?   In my view, the answer is no, well, usually no.  How’s that for a lawyerly answer?
First, the “no” part.  The Rule makes clear that the do not call provisions do not apply to calls from telephone surveyors.  The Telemarketing Sales Rule Compliance guides, available online at, also makes this clear.   Likewise, the FCC’s corresponding rule on Do Not Call excludes surveys and market research from coverage.
Now, the “usually no” part.  The Rule notes that such calls may be prohibited if they are a pretext to advertise a product or a service.  In other words, Do Not Call will apply when the survey reasonably could be construed as proposing a commercial transaction.  So, usually, a telephone survey will not be covered by Do Not Call.

What about CAN-SPAM?  The CAN-SPAM Act covers “commercial electronic mail messages” – that is, e-mail whose primary purpose is advertising or promoting a commercial product or service.  Among other things, CAN-SPAM prohibits commercial e-mails with false or misleading information on the “from” and “subject” lines, and prohibits sending a commercial e-mail message without disclosing that the message is an advertisement or solicitation. The key question then is whether an e-mail promoting the participation in a survey constitutes a commercial e-mail – that is, whether the “primary purpose” is to advertise or promote a commercial product or service. 

This past January, the FTC issued its final rule establishing criteria for determining the “primary purpose” of e-mail messages.  The Rule mentions three types of content. 

First, content that involves the commercial advertisement or promotion of a commercial product or service.  Second, transactional or relationship content.  And third all “other” content. 

I’m not going to go into an in-depth application of the “primary purpose” rule.  Suffice it to say that survey research does not appear to involve transactional or relationship content, which typically involves a pre-existing buyer-seller or employer-employee relationship between the sender and the recipient.  So, whether an e-mail soliciting participation in survey research constitutes a commercial email – which CAN-SPAM regulates –  or “other” content – which CAN-SPAM does not regulate –  depends on whether a recipient reasonably interpreting the subject line of the e-mail or the body of the message would likely conclude that the message contains the commercial advertisement or promotion of a commercial product or service.

Unfortunately, the Rule does not define the phrase “commercial advertisement or promotion of a commercial product or service.”  The FTC’s Final Rule implementing CAN-SPAM appears to acknowledge this fact, at least implicitly, and suggests that the relevant definition is the definition of commercial speech under constitutional law.  So, if a reasonable consumer would reasonably interpret the subject line or body of an e-mail as constituting commercial speech, then CAN-SPAM regulates the e-mail.

So, now it’s just a simple matter of consulting that seminal Supreme Court case that defines commercial speech... And once the Supreme Court decides that case, you’ll have your answer.  Until then, you’re at the complete mercy of government regulators.  Thank you and good night. 

I’m kidding, of course.  The Supreme Court has decided numerous commercial speech cases over the years; some of those cases also could be considered seminal.  Unfortunately, the Court has failed to articulate a clear and consistent definition of commercial speech.

Since 1975, the Court has defined commercial speech as narrowly as speech which proposes a commercial transaction and as broadly as “expression related solely to the economic interests of the speaker and its audience” and then again as narrowly as speech which proposes a commercial transaction.  The last definitive statement from the Supreme Court is that the narrow definition is the definition of commercial speech.

That definition – speech which proposes a commercial transaction –  is of very limited utility.  After all, it uses the word we’re trying to define – commercial – in the definition itself.  If any of us tried to pass off that definition in English class, we’d get an “F” for sure.  Instead, says the Supreme Court, it’s the law of the land.  Like that line in the Mel Brooks movie, “It’s good to be the King.” 

The Supreme Court had the opportunity to clarify its definition of commercial speech two years ago, but declined.  Some of you may be familiar with Nike v. Kasky, which found its way to the California Supreme Court.  Starting in 1996, Nike was publicly accused of mistreating and underpaying workers at foreign facilities.  Nike responded by sending out press releases, writing letters to the editors of various newspapers, and mailing letters to university presidents and athletic directors.  In 1998, a California resident sued Nike for false advertising, asserting that, in order to maintain or increase its sales, Nike made a number of false statements concerning the working conditions under which Nike products are manufactured. 

The California Supreme Court focused on three factors in holding that Nike’s speech was commercial speech.  First, Nike is a commercial speaker because it sells athletic shoes and apparel. 

Second, Nike intended to communicate to potential consumers, namely, athletic directors and university presidents; even the press releases and letters to the editors were designed to reach the general public, who also purchase Nike products. 

Third, the content of Nike’s statements was commercial because the statements concerned its business operations and were made in order to maintain its sales and profits.  As a result, the court held that if the statements were false, they were actionable under California consumer protection law.

The U.S. Supreme Court granted review of the California Supreme Court’s decision and took extensive briefing in the case.  The Court, however, dismissed the appeal on technical grounds.

Some day the Court may take another stab at defining commercial speech.  But even under the California Supreme Court’s expansive definition in the Nike case, I think that consumer research seeking information on attitudes toward a commercial product would not qualify as commercial speech.  Although a market research firm might be considered a commercial speaker if it is obtaining information on behalf of a commercial client, the survey, if bona fide, would not necessarily communicate any message about the client’s product or business.  Rather, the communications would primarily involve the gathering of information from consumers.

No doubt, some aspects of a survey may involve providing information to the consumers.  For example, “If I told you that Brand X contains this ingredient, would that affect your purchase decision?”  The question tells the respondent that the product contains the ingredient, but the context of the statement is neutral and does not suggest that the consumer should buy or not buy the product because it contains the ingredient. 

By contrast, take the statement, “If I told you that Brand X contains an ingredient that both softens the skin and renders it impervious to bullets, would you prefer it over Brand Y, which does not contain that ingredient?”  This statement is not neutral and suggests to the respondent that she should buy Brand X, particularly if she happens to be a gangster with dry skin.  Thus, depending upon the amount and type of information conveyed about a product, a survey might qualify as commercial speech under the California Supreme Court’s view if the survey reasonably could be construed as an attempt to induce consumers to engage in a commercial transaction.

To re-cap: In my view, legitimate survey research, the goal of which is to obtain consumers’ opinions about non-commercial issues, such as their views about political candidates, is not commercial speech.  In fact, such research would appear to lead to the discovery of truth and foster better self-government, ends that the First Amendment is designed to protect to the highest degree.  Legitimate survey research also is likely fully protected speech even if the goal is to obtain consumer opinions concerning a class of products or services or even a particular, brand-name product or service.  The only caveat here is if the methodology of the survey or the content of the questions could lead a reasonable person to believe that the surveyor is seeking to persuade him or her to purchase the particular product or service.  But assuming that is not the case, such a survey does not fit any of the existing tests for commercial speech.  It does not propose a commercial transaction, nor does it primarily involve the communication of information about a product for the purpose of inducing a commercial transaction. 

Again, the reason I’ve been discussing the definition of commercial speech is because we are trying to figure out whether an e-mail requesting participation in a legitimate consumer survey involves the advertisement or promotion of a commercial product or service for purposes of CAN SPAM.  Since it generally will not, there is little reason to believe that run of the mill consumer research will be covered by the CAN-SPAM Act.

Likewise, the Do Not Call rule applies only to any plan, program or campaign to sell goods or services through interstate phone calls.  In other words, the Rule applies to speech which proposes a commercial transaction – commercial speech.  Thus, in general, telephonic survey research would not meet the definition.  This conclusion is buttressed by the Rule itself, which, as I previously mentioned, states that it does not apply to calls from telephone surveyors.

In closing, I would leave your industry with this piece of advice: If market research or polling is bona fide, in other words, is not an attempt to induce the sale of a product or service or to build brand awareness, it likely will not be considered commercial speech.  After crafting your surveys and polls, put on your sorting hat and ask yourselves, “Could a reasonable consumer construe this as an attempt to sell me something?”  If the answer is no, the surveys are probably not commercial speech, and any regulation that purports to cover such speech is subject to the most exacting constitutional scrutiny.  And lastly, if you’re going to conduct face-to-face consumer research, please, please, do not forget to take off the sorting hat.  Otherwise, you’ll look ridiculous and completely skew your results.

Thank you very much and I look forward to any questions from the audience.