Online advertising: consumer perspectives

Berkeley/Santa Clara conference

Panel on Consumer Perspectives on Online Advertising

Moderator, Chris Hoofnagle

The NYT recently reported that advertisers won’t target ads to you if you search on AIDS, but might if you search for information on warts.

Norman I. Silber, Hofstra University School of Law

Rather than traditional consumer deception/unfairness concerns, we’re increasingly focused on privacy. We should also discuss consumer vulnerability to affective appeals rather than rational, and the impact of information source masking on consumers. We may want to regulate certain foreseeable practices.

The history of advertising is one in which consumers have been subjected to ever more affective appeals and unverifiable but scientific-sounding claims. Advertisers have tried subliminal appeals. Online advertising has to be put in a context, which includes product placement.

What continuity is there between older practices and behavioral advertising? The FTC staff: behavioral ads have benefits, but the practice is largely invisible. Targeting basketball ticket ads based on geography to be for the “Cal-Berkeley” game or the “USC” game, depending on location, seems fine. But if the advertiser knows that the target has a big plasma screen, and modifies the ad to assert that games are often “blacked out,” is there a problem? What if that’s a true statement in general, but this game isn’t blacked out? What if the advertiser knows the target is wealthy and sets a higher price for her than for others? What if the advertiser knows that the fan has Paypal, Visa, and Mastercard, and the advertiser accepts all three, but only gives her the Mastercard option because that’s the least expensive for the advertiser?

What Women Want: Mel Gibson gets Helen Hunt to fall in love with him by reading her mind and (inauthentically) repeating her thoughts. This is the worry.

John Horrigan, Pew Internet and American Life Project

Gathering info online is a way to minimize risks endemic to any purpose, but many online users see the internet as a risky place to do business. Dec. 2007 survey of internet users: the internet is convenient, saves time, offers bargains. But there’s significant worry over security, and frustration over finding information and the complexity of online shopping. The lower a consumer’s income, the more she worries about sending credit card information online and the fewer benefits she sees to online shopping. The story is similar but not as dramatic with respect to age (the older, the more worries).

New survey, forthcoming: where do online resources fit when buying music? When music buyers are asked where they hear about music they purchase, they say (1) radio, TV, or movie; (2) family or friends; and (3) various online tools. Moreover, 36% say that online information had a major (9%) or minor (27%) role. So even in music, where the internet has been so important, the internet comes in third.

Cellphone purchases: For researching, people used (1) expert/salesperson; (2) store; (3) internet, mostly provider’s site, and also rating sites. Role in purchase: 73% said major (27%) or minor (46%) role. A cellphone is a commitment good, so consumers are drilling down more.

Real estate: real estate searchers are equally likely to cite the internet, real estate agents, and newspaper ads, and to find them all influential.

Online information is only part of the info pathways in consumer choice, and not always influential. And people who worry about online info security are less likely to use the internet when making buying decisions—even searching for information. Security of credit card worries depresses internet use in general. Since switching is so easy for shoppers, online advertisers have clear incentives to promose a trustworthy and secure e-shopping environment. The internet is not as useful for lower-income people, who are arguably the most likely to benefit from time-saving and other efficiencies.

Joel Winston, Federal Trade Commission

Disclaimer: These are his own views, not official FTC positions. Consumers have a general feeling of lack of control: info is being collected and used in ways they don’t understand/control—everything from identity theft to government surveillance to advertiser uses. Every time they read about a data breach, they worry about their own information. About 71% are generally aware that online activities may be tracked, but the process is opaque to them, especially about how the info might be used or disclosed.

Most think that they should at least have the opportunity to opt out, and in fact that they should have to opt in.

The FTC has encouraged self-regulation as the right approach for a dynamic marketplace. Principles: disclosure; reasonable security and limited data retention. What about using tracking data for secondary purposes? If it’s disclosed to other parties, it may be of great concern to consumers, or if it’s personally identifying information. Some participants in town halls suggested do-not-track as an analogue to do-not-call, but the logistics are in doubt.

Mark Cooper, Consumer Federation of America

Online ads are Dr. Jekyll/Mr. Hyde. People want caller ID for themselves, but they don’t want to show their info to others. There’s a privacy/efficiency tradeoff. His question: how much info do I have to give advertisers to substantially improve on the 20^th century? His answer: not much, and certainly not as much as they want, as illustrated by the morning tutorials.

Everyone hates TV ads: consumers; advertisers (don’t know if it works; it’s expensive); media critics (shows can’t create audiences without advertising, so advertisers heavily influence what gets on the air: “Why is this show so dumb?”). Newspapers are clearly better, even though ads are a bigger percentage—easy to skip, easy to store for later reference, and tends to be more useful—what’s in the supermarket at what price, or what are car dealerships charging, instead of image awareness.

One-way push advertising was useless; internet allows a two-way conversation: a telephone instead of a megaphone. But do consumers know they’re involved in a conversation? Advertisers chase eyeballs rather than trying to create an audience.

Online ads can more easily be closed than TV ads—advertisers hate the TiVo. So online ads can be less intrusive in experience, but more intrusive in terms of data gathering and targeting. Ubiquity-wise, online ads are also better than TV and no worse than newspapers. Very local advertisers can get into the market, and can even sell nationally through websites when they couldn’t afford national TV ads. There are significant possibilities for improving efficiency.

The Mr. Hyde part: Behavioral tracking and targeting is inherently deceptive. If we can restrict it to contextual ads, we can get the benefits without the costs of tracking. But contextual ads sell out immediately: context is key. The more you try to find categories of people, the more you go to the 20^th century TV model of trying to find a demographic. Contextual ads can provide most of the benefits without the privacy harms.

What’s available today doesn’t approach informed consent. The data suggest that 85% of sites give privacy statements, but 99% are incomprehensible to consumers.

In the regime we have now, inertia and ignorance work to the benefit of the advertiser. Most people can’t deal with the tech. If you shift to opt-in, the advertisers will tell you, no one will do it. An easy, uniform opt-out (do not track) would split the difference.

Online advertising: tutorials

The Law & Business of Online Advertising: Berkeley/Santa Clara

Tutorial on Online Advertising Technology

Kim Howell, Microsoft privacy expert

Among other things, Howell emphasized that 3d-party ad networks enable smaller publishers to monetize their sites (e.g., bloggers) and smaller businesses to participate in ad campaigns across multiple platforms. Intermediaries need to report to publishers and advertisers how many ads they served to figure out how much to pay, which means they record impressions and users’ IP addresses. If the advertisers only want to reach net-new users, then they need to get more information: thus they add a cookie to the user’s browser.

Financial/auto website inventory sells out very quickly. The advertiser may want to target users with specific interest, but use a general-interest site. The ad network server can check the cookies and see who’s been to financial websites and show them American Express ads. It extends the inventory of “people with financial interests” to nonfinancial publishers. This is important for general-interest websites, Yahoo!Mail, etc.

What if the advertiser wants to show different ads to men and women? The ad network server can have IP address, cookie, ads served, time/date of visit, and registration—if the user has registered with a site for various reasons, such as getting content. The user provides gender, age, income, etc. The publisher passes on that information to the ad server. The ad server can correlate the user account with the cookies and serve targeted ads. (You can find what servers are being called by choosing “view source” and searching for http.)

What if the advertiser wants to show an ad to a known buyer? E.g., people who abandoned a shopping cart or people who’ve made a past purchase. The advertiser can place a web beacon on a page it controls; the advertiser, e.g., Wal-Mart, can choose what information about the user will be sent to the ad server. So when the user is on another site, the ad server can check the cookie ID and see the user is a Wal-Mart customer and show her a Wal-Mart ad. The advertiser’s privacy policy will (or should) explain how the data is being used.

Data that could be in an ad profile: IP address, unique cookie ID, all publisher sites visited where ad server serves ads, all advertiser sites visited where there’s a web beacon, any publisher/advertiser info passed to the ad server, and geographic info derived from reverse IP lookup.

Question: is there any way around cookies? What if a user deletes/disables? Answer: there are new technologies in the offing, but the major technologies rely on them. Cookie management creates “orphan” information that exists, but is unusable.

Q: Are there ways to discourage cookie deletion? A: Other than education, not really. They can say the website won’t work as well without cookies, which is true. But other than that, no.

Q: How is this connected to personally identifying information? A: The publisher/advertiser can pass info to the ad server—the ad server, e.g. Doubleclick, doesn’t have any direct relationship with customers. It’s against National Advertising Initiative (NAI) principles to use web beacons to pass personally identifiable information. So legitimate/large players don’t do this.

What is the customer’s choice? (1) To clear/block cookies; you can even create custom block lists. Those two options conflict with each other, though. If you set your options to accept only first-party cookies, most sites should work for you. (2) To set opt-out cookies with NAI members. If you choose (2), the ad server knows you’ve opted out of targeted ads and won’t show you one. When you opt out, the data is still collected in order for publishers and advertisers to know which ads were served where.

Q: If an ad network were subpoenaed, could it turn over your information? A: Assuming they’ve retained the data, yes, though they should have a retention policy keyed to business purpose.

Comment from audience: if you have a dynamic IP address, as is common in the US, your IP address won’t be tied to you absent that cookie. A: If you want to track users, going to the ISP directly would be a lot more valuable.

Tutorial on the Business of Online Advertising

Hal Varian, Google

There are only a few hundred major players in the brand advertising space: big advertisers, big publishers.

Advertiser/agency contacts publisher, negotiating a price starting from a rate card—nobody pays retail. Aspects of negotiation: cost per thousand impressions $1-20; $10 CPM=1 cent per impression. Advertising requires scale, which is why there are so few major players.

The process: Advertiser puts creative on ad side server; publisher programs publication side server according to conditions (# of repeats, targeted info); ad is fetched and displayed according to terms; payment is made. Cost of serving is 2-5 cents per CPM, paid to ad-side server. Any pub server can talk to any ad server and vice versa. Doubleclick, Aquantive, Real Media etc. just provide serving technology, not creatives/ad pricing; some large publishers and agencies have their own pub servers.

Ad networks: about 300; an aggregator that reduces the need for separate contracts, separate sales forces. Network, like Google AdSense, sells ads to publishers on behalf of a number of advertisers. There are quality issues, though—the ad might not be as good. Valuable for small publishers and remnant inventory of large publishers when they haven’t sold out all the ad space.

Ad exchanges: a marketplace where buyers/sellers can meet. It’s a competitive market with commissions, like a financial market, but participants have less control over where and how ads appear.

Why cookies? Because http is “stateless”—it has no memory. The web would fall apart without a way to maintain memory. A common use of cookies in ads: frequency capping.

Website cookies are sometimes also called 1^st-party cookies, used by publisher to sustain conversation with user. 3d party cookies are different—ad-side or pub-side.

Contextual ads, e.g., AdSense: advertiser uploads text creative, which could be the same as for a search ad. Publisher puts a bit of javascript on page. Users are shown ads contextually related to page content. CPC pricing: every time user clicks on ad, revenue is shared at a fixed rate between the ad network operator and the publisher; CPC price is set by competitive auction. Variations: image ads; CPM ads. Major plus: reduced transactions costs. It’s popular with small publishers and used as filler by large ones.

Search ads: On Google, advertisers can decide how to partition ads—whether they only want ads to show on Google, or also on search engines with which Google has agreements; geo targeting; day splitting; other options.

There are millions of keywords which can’t be priced separately. Instead, there’s an auction: each ad submits creative, keywords, bid. Ads are ranked by bid x click quality, as estimated by Google (the “secret sauce”). The best-scoring ads get the most prominent positions. The reserve price (the price you pay if you’re the only bidder/the only ad on the page) depends on ad quality—the higher the quality, the lower the reserve. Really good ads get promoted to the top of the page. Really bad ads get disabled and not shown. You’ve got to keep users as well as advertisers happy to keep users from developing “ad blindness.”

Now watch Varian set up some competition rhetoric: Publishers want max revenue. Effectively, all forms of ads are perfect substitutes. In practice, they often have a banner ad, some image ads, and some contextual ads. There’s no exclusion; on a large publisher, there can be 3 or more companies serving different types of ads, e.g. on the NYT page, plus what they sell directly. The switching costs are relatively low: all you have to do is change the html code on the page.

Another intermediary: SEOs and SEMs, which help manage search engine campaigns; he thinks that in the long run they’ll be incorporated into ad agencies.

Economic forces: pricing power. Prices are set by advertisers in competitive auction for search and contextual ads; set by competitive negotiation for display ads and ad serving.

Switching costs: For publishers, low: change code. For advertisers, they already use multiple channels, and switching is easy, or even not required: 95% of Google’s big advertisers also advertise on Yahoo! Users read a variety of content, and find it easy to switch search. There’s no exclusivity anywhere.

Supply-side economies of scale, or, what about entry? There are several players with appropriate scale, large data centers. What’s important is not tech capability but know-how.

Demand-side economies of scale/network effects. Varian’s never heard a convincing story about search engine use.

Q: Third-party trademarks?

A: His understanding is that the acceptability of this varies by country. Nobody’s quite sure what the policy is. As an economist, he likes competition! Google has a policy against misleading ads, but more choice is good.

No absolute privilege to accuse another of copyright infringement

Carmichael Lodge No. 2103, Benevolent and Protective Order of Elks v. Leonard, 2008 WL 1346641 (E.D. Cal.)

Carmichael Elks is a fraternal organization; two of its members started a series of travel guides to Elks lodges in several states. In 1988, Carmichael Elks registered the first two volumes, claiming assignment by the members. In 1998, Leonard allegedly updated and reformatted the guides and granted a royalty-free license to Carmichael Elks to publish and sell them. (Hmm—given these facts, it doesn’t seem that Leonard’s the one who can grant the license.)

A dispute arose, as they do, and Leonard told Carmichael Elks that he was terminating their license. He also applied for and received registrations for his guides. Soon thereafter, Carmichael Elks registered the third volume of the original guides.

Carmichael Elks sent a letter to other Elks members accusing Leonard of stealing the guides, stating its intent to enforce its rights “with the full force of law,” touting its nonprofit purposes in comparison to Leonard’s pursuit of personal profit, and asking them not to buy Leonard’s guides.

Leonard counterclaimed for, among other things, false advertising, unfair competition, and libel per se. The court rejected Carmichael Elks’ defense of privilege related to a legal proceeding. (The court was skeptical of the breadth of privilege under California law, since it “far extends beyond the purposes which it is said to serve,” but stated that it was “of course” bound by California law.) In California, privilege covers statements made before filing a lawsuit, including demand letters and other prelitigation communications by lawyers, as long as there’s a reasonable belief that litigation is likely to be commenced. But the statement must have a functional connection to a contemplated lawsuit. So one case found that a statement to the press about an employee’s dismissal was not covered by the privilege, even though the dismissal was the subject of NLRB proceedings. “Public mudslinging” is something courts can stop.

The letter here reflected a good faith belief that litigation would ensue, and its subject matter was the same as the lawsuit. But by sending the letter not just to Leonard, but other Elks Lodges, it went beyond a functional connection to the litigation. Carmichael Elks argued that it was attempting to achieve the same results as litigation—stop Leonard’s sales of infringing works. But the court rejected that—Carmichael Elks’ entitlement to an injunction is up to the court.

In addition, the policies behind the law (“promoting truthful testimony, encouraging zealous advocacy, streamlining litigation”) wouldn’t be furthered by granting the privilege. Demand letters can avoid the need for litigation, but this wasn’t a demand letter or other attempt to resolve the dispute. Prelitigation investigation is tied to judicial proceedings, but the letter wasn’t investigative. As a result, the court denied Carmichael Elks’ motion to dismiss the non-copyright counterclaims.

Playing chicken: "raised without antibiotics" draws lawsuit

Sanderson Farms, Inc. v. Tyson Foods, Inc., -- F. Supp. 2d --, 2008 WL 1733607 (D.Md.)

Tyson advertises its chicken as “Raised Without Antibiotics” (unqualified RWA claim) or “Raised Without Antibiotics that impact antibiotic resistance in humans” (qualified RWA claim). Its label has been approved by the Department of Agriculture. The court ruled that this was not insulation from a Lanham Act false advertising claim.

In its chicken feed, Tyson uses molecules called ionophores, which kill microorganisms in chicken by disrupting transmembrane ion concentration gradients. Plaintiffs allged that ionophores are antibiotics. The Department of Agriculture originally approved the unqualified RWA label, then revoked that approval on the grounds that ionophores are antibiotics. In late 2007, USDA issued a labeling guidance for ionophores affirming “longstanding” policy that ionophores are antibiotics because they meet the American Vetinary Medical Association’s definition of antibiotics, and the FDA agrees. Thus, birds to whom ionophores have been given can’t bear a RWA claim.

Plaintiffs alleged that the unqualified RWA claim is literally false and that the qualified RWA claim doesn’t cure the literal falsity of “Raised Without Antibiotics.” Further, these are material claims, and make an implied health and safety superiority claim over plaintiffs’ chicken. Plaintiffs’ shopping-mall survey of about 600 consumers involved two cells shown an unqualified RWA claim, one shown a qualified RWA claim, and one showed a control claim. The survey allegedly showed that 59-63% of respondents perceived a false message of safety superiority regardless of qualification. Consumers reason (1) Tyson’s chicken has no antibiotics, and (2) therefore Tyson’s chicken doesn’t impact antibiotic resistance in humans. (I.e., they reason that the qualified RWA claim is on the order of “fights the germs that cause bad breath,” in that it provides a reason why one would purchase chicken raised without antibiotics.) Thus, the qualification is ineffective.

Tyson argued that the RWA claims were approved by USDA, which has the authority to regulate poultry labels, thus precluding any Lanham Act claim. There’s no private cause of action for violation of the Poultry Products Inspection Act (PPIA), and that, defendant argued, is what plaintiffs were trying to create. A line of cases holds that statements that “comport substantively” with FDA-approved claims can’t be attacked via the Lanham Act. More broadly, “federal courts should not unduly entangle themselves in regulatory agency decisions where the agency has special expertise in the subject matter and where, more importantly, doing so would usurp the authority specifically delegated by Congress to that agency.”

Tyson was entirely out of luck with respect to the unqualified RWA claim; there was “absolutely no tension” between the Lanham Act and the USDA. Given that the USDA revoked Tyson’s authorization to use the unqualified RWA claim, Tyson “cannot rely on the USDA’s former (and briefly held) position to defend itself against allegations that it continues to run false and misleading advertisements ….”

What about the qualified RWA claim? Tyson has current approval for this language on labels. But no case on point resolves the question whether USDA approval is insulation from false non-label advertising claims. Distinguishing the FDA cases, the court reasoned that the USDA has no jurisdiction over ads, whereas the FDA does (though it has a coexistence agreement with the FTC for OTC drugs). Indeed, USDA acknowledges that the FTC has responsibility for poultry ads. Thus, the idea that drug ads are ok if they “comport substantively” with FDA-approved wording is inapplicable here.

The scope of the labeling provisions was also at issue. USDA defines labels broadly, to include all materials that accompany a product, including point-of-purchase displays. But some labeling is also “commercial advertising and promotion” under the Lanham Act. (Comment: I would think, indeed, that virtually all labeling is advertising, but apparently courts are beginning to be confused about this; if a product name can be false advertising, as in BreathAsure, then I can’t see what else on the label would be excluded.)

The court held: “Point-of-purchase materials that merely restate the language approved for the label cannot fairly be characterized as advertising. [But] point-of-purchase materials that may well be considered labeling to [USDA] often contain images and promotional slogans in conjunction with the language approved for the label. This sort of labeling is merely advertising by another name. … [USDA’s definition of labeling] will not insulate what is plainly an advertisement intended to induce consumers to purchase Defendant's product.” Thus, the complaint covers any labeling that, “despite including language approved by the USDA, contains additional images and promotional slogans that effectively turn the labeling into an advertisement.”

Of note: the court also drew support for its conclusion from a case from the Better Business Bureau’s National Advertising Division (NAD). NAD reviewed a USDA-approved “no preservatives” label and independently determined that the words were misleading to consumers when used in ads.

There’s no private cause of action under the USDA’s governing statutes, but at the same time, the USDA’s existence shouldn’t eviscerate the Lanham Act. Thus, courts have drawn lines between (1) precluded claims that involve application and interpretation of an agency’s statute and (2) allowed claims against false advertising. Plaintiffs couldn’t use the Lanham Act as a disguised effort to enforce the PPIA. Thus, if their claim had targeted Tyson’s labels, it would fail because it would directly challenge the USDA’s primary jurisdiction to determine whether a label is false or misleading.

But claims about non-label ads are not within the USDA’s jurisdiction. USDA conducts a “highly technical and scientific review” of proposed label language, but it doesn’t address whether language is misleading to consumers when combined with images and slogans. “Undoubtedly, language that is technically and scientifically accurate on a label can be manipulated in an advertisement to create a message that is false and misleading to the consumer.”

Thus, the Lanham Act’s prohibition on literally true but misleading claims can be invoked in this circumstance. To prevail, plaintiffs would need to show that the qualified RWA claim “means something different to the consumer public when viewed as part of Defendant's advertisements than the language did to the experts and scientists at the USDA during the label-approving process.” And the complaint alleges just that. (This seems to be more about lack of expertise/focus than lack of jurisdiction. Scientists are not marketing experts.)

Interview: John Villafranco of Kelley Drye & Warren

I am pleased to present the next interview in my occasional series on advertising lawyers. Today's subject is John Villafranco.

Here's his introduction: I am a partner at Kelley Drye & Warren and provide litigation and counseling services to corporations involved in advertising and marketing. My practice focuses on Lanham Act litigation, consumer class action defense, representation of clients in advertising substantiation proceedings and investigations conducted by the Federal Trade Commission and state attorneys general, and representation of challengers and advertisers before the National Advertising Division. I am co-Editor-in-Chief of the ABA's Consumer Protection Law Developments treatise, which will be published in the Fall of 2008, and I am immediate past Chair of the ABA Section on Antitrust Law Consumer Protection Committee.

Q: How did you get into advertising litigation?

I started in 1991 as a first year associate in the antitrust section at Collier Shannon Rill & Scott which, at the time, was perhaps the most prolific antitrust practice in the United States. I liked but did not love antitrust law, and I made myself available to work on advertising cases that were being handled by Judith Oldham (a gifted attorney and my eventual mother-in-law). When Bill MacLeod joined the firm after leaving his position as Director of the FTC’s Bureau of Consumer Protection, our advertising case load increased. I never worked on another antitrust case – it has been advertising all the time – and it has been a lot of fun.

Q: Tell us about some favorite advertising cases.

There have been many, but my favorite may be the Joe Camel case, which we litigated against the FTC in 1998 before Administrative Law Judge James Timony. The case was hotly contested with 17 FTC lawyers filing notices of appearance. Complaint Counsel relied on its unfairness authority under Section 5 of the FTC Act, alleging that the cartoon camel was a “substantial contributing factor” to smoking initiation among teens. I was a senior associate and it was my first meaningful opportunity to devise case strategy, present oral argument, and otherwise manage a case. And this one had a little bit of everything – pre-trial publicity, surprising documents (for both sides), and novel theories. After a few weeks of trial and immediately after we finished the successful cross examination of the FTC’s causation experts (they never could, in my mind, establish that the cartoon camel caused substantial injury), Complaint Counsel surprised everyone by telling the judge that they were withdrawing the complaint. They asserted that the issue was moot as a result of the multi-state settlement agreement that had just been negotiated between the states and the tobacco companies. There was some truth to that, but we did not see it that way. They brought the complaint, litigated it, then withdrew it right when the case was breaking our way. No FTC order, no injunctive relief. We celebrated as if it was a clean win.

And then came the fast food wars. In 2001, Truett Cathy, the founder of Chick-fil-A sought a preliminary injunction in Chick-fil-A, Inc. v. Burger King Corp., attempting to prevent Burger King from describing its first chicken sandwich as “the original chicken sandwich.” The claim was self-comparative, in that Burger King had just launched its BK broiler chicken sandwich and needed to call the existing chicken sandwich something. Mr. Cathy did not see it that way. He asserted that he was the true inventor of the chicken sandwich -- the very first human being to put chicken between two slices of bread -- and therefore Burger King could not claim that they had “the original.” A Library of Congress search yielded recipe books from the 1800s that included chicken sandwiches and the motion was denied.

Three years later, in Steak ‘n Shake Co. v. Burger King Corp., Burger King found itself defending motions for a TRO and preliminary injunction seeking to enjoin Burger King's launch of its new “steak burger” sandwiches. Steak ‘n Shake alleged violations of federal and state trademark and anti-dilution statutes. They claimed (you guessed it) that they invented the steak burger and that consumers associated “steak burger” with Steak ‘n Shake. Same result – motion denied.

Another favorite was BellSouth Telecommunications, Inc. v. Hawk Communications, LLC, a 2004 Lanham Act case that we brought in the Northern District of Georgia. In that case, BellSouth – the largest ISP in the region – sought a preliminary injunction to stop Hawk from advertising that its narrowband internet service allowed consumers to “Dial Up at DSL Speed.” Hawk advertised heavily all over the southeastern United States, including on billboards right outside the BellSouth headquarters, and it was driving BellSouth crazy. We had previously prevailed on all issues before the National Advertising Division, but Hawk chose to ignore the NAD decision. In granting our motion for preliminary injunction, Judge Shoob credited the NAD decision (a precursor to Judge Carter’s decision in the November 2007 case Russian Standard Vodka (USA), Inc. v. Allied Domecq Spirits & Wine USA, Inc., which I recently heard described as a “love letter to the NAD”), and extended the doctrine of necessary implication into the Eleventh Circuit.

Q: What are the emerging trends in advertising law?

When I started in 1991, advertising was neatly categorized under the headings of print, television, and radio. Today, it is all over the place: viral and buzz marketing, product placement, mobile marketing, consumer-generated content, and who knows what’s next. This is one aspect of the practice that I love – it’s on the move.

Q: Over the past several years, an increasing number of courts have both expressly recognized that prudential standing limitations apply to false advertising actions under the Lanham Act and have applied those limitations to bar certain plaintiffs from pursuing claims. Last year, the Eleventh Circuit’s decision in Phoenix of Broward, Inc. v. McDonald’s Corp. granted McDonald’s’ motion to dismiss for lack of prudential standing, and the Eleventh Circuit affirmed (joining the Third and Fifth Circuits in holding that prudential standing limitations applied to false advertising claims under the Lanham Act). Do you expect this trend to continue in other circuits?

I do. The courts are already overburdened; tossing cases where a judge can reasonably conclude that the plaintiff is not the litigant best suited to assert a claim is an easy way to lighten the load. I expect we will continue to see the courts express skepticism about expansive standing under the Lanham Act, especially where the plaintiffs are corporate customers or suppliers. Having said that, we are not at a point where a precise showing of injury is required, nor will we ever be. As I learned in the Joe Camel case, showing that an advertisement actually caused injury is a very difficult thing to do.

Q: What is your advice for a law student who wants to practice in the field? Given that most schools don’t teach advertising law, what are the best courses to take?

The Federal Trade Commission’s Bureau of Consumer Protection would be the ideal place to start a career in advertising law. If you intend to start in private practice, then I would suggest that you focus your job search on the leading advertising law firms. There are only a handful with sizeable practices that will provide a steady diet of advertising work: Kelley Drye (of course), Mannat Phelps, Reed Smith, Davis & Gilbert, Frankfurt & Kurnit, Venable, Loeb & Loeb. Interested students will want to get a summer associate position with one of these firms. As for recommended course work, I would not worry too much about it – this is an area that is better learned in practice than in the class room -- although classes that focus on unfair competition and trade regulation would not hurt. And of course, if you are at Georgetown, any class taught by Professor Oldham (my father-in-law) will ensure that you are ready for the rigors of the practice. You just can’t get enough Lord Mansfield.

Iowa Trademark Scholars' Roundtable, Session 3

Session 3: Trademark Use, Unfair Competition Law and Other Causes of Action

Introduction: Robert Burrell

Australia’s had a TM use requirement since 1905, and a modern doctrine of TM use for at least 50 years. The temptation is to take that experience and try to apply it to the US. We need to be careful about how we use comparative material; IP scholars too often take an isolated provision from someone else’s law and compare to an equivalent without asking whether the provision does the same job in both jurisdictions. Confusion doesn’t necessarily play the same role in other jurisdictions as it does in the US; it’s not the same touchstone of liability. That’s partly because registration has played such a different role outside the US. Common law: registration gives a narrow but very strong right; the plaintiff simply didn’t have to demonstrate confusion. So in those jurisdictions there’s a temptation to try to mitigate that nearly strict liability.

In Australia, the scope of registration has expanded, but confusion is still not a key trigger of liability, and courts are trying to retrofit the infringement test to build in a confusion requirement. TM use provides one way of accommodating confusion.

Is TM use anachronistic? Not at all. It plays a key role in confining TM protection. It’s not an attempt to reimpose technical requirements like affixation—it doesn’t work like that in Australia at all. The Australian experience bears out many of Barrett’s points about the competition-preserving elements of TM use. The Fender case—an attempt to prevent resale of second-hand guitars: the courts used TM use to protect that market. Johnson & Johnson case: a registration for caplets. Courts used the TM use doctrine to protect defendants who were using the invented term in a descriptive way.

Mothercare-Othercare case: an attempt by a clothing store, Mothercare, to prevent publication of a book called Mother Care/Other Care, a feminist argument that non-parental care was a fine thing. The store argued that its customers would be horrified by that argument. Free speech concepts were vindicated by application of the TM use doctrine despite Mothercare’s registration of its mark for, among other things, books.

Also courts can use TM use to positively shape consumer reactions: courts accept short-term consumer confusion for other interests.

Confusion weighs heavily against a finding of no TM use, however. The doctrine’s attractiveness is its flexibility. It’s most helpful for fashioning very clear exceptions to liability: second-hand markets; use on T-shirts; use on book covers. Parliament often responds by codifying the decision—comparative advertising, second-hand markets, and others. Is there anything about TM use that makes it uniquely suited to this role? Possibly not; but it works in practice.

Historically TM use has performed many diverse functions. Barrett’s list of factors might move us away from that flexibility. He’s also wary that TM use can serve as a gatekeeper outside the realm of very clear exceptions. In 15 cases he can think of where TM use was critical, they were split 8-7 between defendants and plaintiffs.

There are areas where prediction of Australian law is very different. 3-D marks are one example—how do you apply TM use? Use on the internet is another example. The TM use requirement might also have unexpected consequences. In Australia, TM use has contributed to a downward pressure on the distinctiveness test: issuing the registration won’t prevent other traders from using it other than as a badge of origin, so there’s less need to protect other traders by denying registrations. Now Australia accepts ridiculous things as distinctive. We can imagine similar pressure on the functionality doctrine.

We can also imagine TM use feeding back into nonuse cases. If use on a T-shirt isn’t use as a mark, then it’s not use when the TM owner does it either.

Australia has yet to confront Google/keyword buy cases. There’s also no historical contributory infringement doctrine in the way the US understands it.

Historically, deception has been considered a property of the mark, not of the conditions of the marketplace; courts are beginning to broaden into a more general confusion inquiry. TM use builds some of that in.

Eric Goldman: (1) TM law needs better channeling/gatekeeper doctrines. In his mind, the paradigmatic TM infringement is counterfeiting. The further we get from that, the more problematic the doctrine becomes. Principally we use defenses to channel today, and they aren’t enough. Functionality works pretty well in channeling claims to patent; genericide might also be pretty good. Descriptive/nominative fair use, however, are not particularly good or unified, and the burden is on the defendant. TM use promises a channeling function; it’s not an ideal solution.

(2) Looking at TM law as consumer and producer protection. From a consumer standpoint, once we get away from counterfeiting, other law does a lot better at protecting consumers than TM. Contract law, false advertising law, consumer protection law, etc. We need to retire the “order Coke, get Pepsi” example. We already know the retailer can’t do that—it breaches a warranty to the consumer and might be fraud. What does TM law bring to that party? TM law doesn’t need to weigh in, though we might want the producer to have a cause of action.

Producer policing: We don’t even really think of TM law sitting on top of retailers in the way we’ve been dealing with TM law sitting on search engines and other intermediaries. Why haven’t we had a conference on: spitting out a coupon for X when you buy Y; stocking house brands next to national; up-selling (if you like X, try Y); loss leaders—discounting one product to sell others; etc. These practices are ignored by legal doctrines and plaintiffs, who tend not to sue retailers even when it’s clear that someone’s liable. Possibly because so many other doctrines control retailer behavior, like antitrust and false advertising. What can we learn from that? If TM law doesn’t need to govern retailers’ profiting from 3d-party TMs, maybe we can cope with similar practices in new areas.

(3) TM law is not industrial-strength—it’s not good enough to do what we’re asking of it. The statutory drafting is weak—it’s been 60 years and it’s not working any more.

Discussion of Goldman’s remarks focused on the issue that, if we want to segregate behaviors out of TM, we still have the contested question of which behaviors we want to regulate. False advertising; gripe sites; etc. could all be regulated under different laws (or things like gripe sites could be affirmatively approved by different doctrines)—but we need to know what we think is harmful.

Bohannan: Thinking about issues as false advertising changes the rhetorical draw of property, which has important influences on how cases come out. If the paradigm is false advertising, a gripe site seems far from the central case; if the paradigm is trademark as property, then a gripe site seems like unfair use of the mark.

Margreth Barrett: In older cases, someone who made a false representation about source without using the mark could be sued for passing off. That’s been incorporated into the Lanham Act, but we don’t use it much. Focusing on passing off or false advertising would move us away from obsessing about business goodwill and property interests, building in a materiality requirement. In a gripe site case, you’d have to show a separate identifiable misrepresentation other than use of a mark.

Litman: Are you satisfied by how TM law deals with counterfeiting currently?

Goldman: Conceptually, it just fits in so neatly to the utilitarian rationale for TM. The Lanham Act version is serviceable.

Litman: INTA sat down to revise TM law in 1980, which became the 1988 amendments; it had all sorts of Christmas tree ornaments on it. If we were to draft a TM law of 2010, it would look a lot like INTA’s 1980s proposals, and maybe that’s okay; it would probably give a lot of powers to TM owners, as the counterfeiting provisions do now. If we need a revision, we need to think about cabining the scope of reform!

Goldman: Just from a Coaseian perspective, the statutory language is so unclear.

Lemley: But the lesson of legislation is that different language is not necessarily better and introduces its own linguistic ambiguities.

Goldman: It’s true that we need to know what we want before we start on an improvement jag. The statutory language we have now is not helpful.

McKenna: Why not try to address everything in TM and hope for robust preemption? Write TM law to allow keyword ads, and not allow another theory to supplement TM.

Goldman: Agreed, preemption and channeling go together.

Lemley: The move to say “that’s for patent law to deal with” or “that’s for Congress” is usually addressed to a court: it means, “please don’t try to fix the problem [if there is one], and we hope that nobody else will either.” Because we don’t have particularly good standards for figuring out what “unfair competition” is, punting to unfair competition is unhelpful.

Dogan: Our proposal allows search engines can be regulated under the law of false advertising, or even secondary liability for enabling passing off/TM infringement.

Lemley: He’d go further: outside TM law, there’s good reason to have false advertising law, but no reason to have a free-ranging cause of action for unfair competition.

McKenna: Lemley wants us to decide that certain cases are within the domain of TM and are allowed, such that resort to other causes of action is not allowed. Like copyright/patent preemption: the conduct allowed by TM law should not be prohibited as such by other law.

Litman: Congress thought that the Lanham Act would preempt state law, but the courts were in an anti-preemptive period when the issue reached them.

Barrett: We’ve heard ominously little from Dinwoodie & Janis! What’s realistic in terms of changing doctrine? There used to be a specific requirement that goods be similar for infringement, which created problems of its own, and of course TM owners were unsatisfied with that limitation. Materiality might be a possibility to slip into cases, but it might not be enough.

Ten or 15 years ago, courts would just have looked at validity of the mark and confusion. Now, they’re more often also asking about use of a mark. So courts’ awareness has been raised by new types of claims. Now we need to get them to focus on what that use language means. McKenna is convincing in arguing that “use as a brand” is ultimately dependent on consumer perceptions, and so we’re not off the slippery slope. She doesn’t want to find use just because consumer surveys show that people think that use X or Y, like comparative advertising or banner ads, requires permission. (My reaction, of course, is to say that we should use lots of presumptions, even irrebuttable presumptions, to get off the slope.)

Barrett would require some close perceptual connection between the mark and the defendant’s content—not necessarily visual, but something like that. And a distinct commercial impression created by the defendant’s use. She wants a standard that can be applied on a motion to dismiss or a motion for summary judgment.

She also was intrigued by Burrell’s suggestion of a hydraulic effect of tightened trademark use on other standards, like distinctiveness—making it harder for TM owners to win cases on TM use could encourage greater initial grants of rights. (This reminds me of the inverse relationship between protecting trade dress as combinations of features—which are more likely to be nonfunctional than features alone—and finding infringement—which is harder when a defendant copies only some of those features. Courts have of late been coming down hard on plaintiffs on both aspects, so there need not be a tradeoff, but I definitely see the logic Burrell articulates.)

Austin: Dinwoodie & Janis still leave him with the concern: against what are we trying to protect people?

Dinwoodie: It’s better to put the harms at issue on the table, and we can only talk about them in specific: the harms/benefits of comparative advertising. That’s why it’s more useful to focus on the particular practices on the back end, rather than sorting practices out based on some concept of “use.”

Dogan: But certain behavior hasn’t historically been thought of as branding behavior, and we can point out to the courts why that’s so, and why we require branding.

Janis: He’s concerned about the subtleties of alternatives, like materiality: we’d have to make difficult decisions about what kind of evidence we need—is materiality a requirement of specific consumer perceptions, or is it a proxy for something else? He hasn’t seen a specific enough alternative. (Comment: This strikes me as requiring perfection from alternatives when his proposal is at least as messy.)

Lemley: We could have a doctrine for pictures of buildings in public places, and one for domain names that aren’t part of the core URL, and one for search engine uses, and one for taking advantage of preexisting confusion, and one for every other use that we think is ok, but (1) there will be lots of loss in the interim, and (2) the people who will get the benefit of carveouts, especially if they’re statutory, will be the big players like Google, not the people who want to build FedEx furniture. A broad TM use doctrine gives coverage to things that historically were never sued over.

Austin: Another way to say it: what’s wrong with blunt instruments? This isn’t surgery. They may be over or underinclusive. If they’re overdetermining, what’s the loss? That’s Lemley’s point: the chilling effect without TM use, even with the availability of surgically accurate instruments down the pike in defenses, outweighs the benefits of targeting. Most of the time the harms Dinwoodie & Janis talk about don’t seem that bad: so you hear about another pizza store when you’re searching for one in particular.

McKenna: The bluntness depends on whether it’s conceptual or functional assessment. Whether a design indicates source is actually a very fine-grained assessment.

Iowa Trademark Scholars' Roundtable, Session 2: dilution

Session 2: Trademark Use and Dilution.

Beebe: What is the relationship between tarnishment and search costs? Despite his sympathy with Lemley & Dogan, the search costs explanation for dilution is actually so similar to confusion that it takes dilution apart, but tarnishment appears different.

One way to approach use in dilution is to ask what is our theory of the TDRA? New §43(c)(1) requires “use of a mark or trade name in commerce” likely to dilute “the famous mark,” thus apparently imposing a “use as a source-identifier” requirement. Beebe is even willing to say that the plaintiff’s burden in dilution is to show that defendant’s mark is perceived as a mark (in which case even Haute Diggity Dog might have been ok), but then again the language is similar to the old FTDA, and “designation of source” was deliberately dropped from the law.

The TDRA might even be a model for writing a use requirement into the Lanham Act more generally. It also gives us potential redundancy in the fair use provision by protecting any fair use other than as a designation of source.

Dogan: Tarnishment is a more harmful version of blurring.

Lemley: Tarnishment has a potentially different mental effect that can come more readily from a single use than blurring—the salacious/problematic use is so relevant that it by itself influences my view of the mark.

My thoughts:

For those who do not believe in a separate trademark use requirement and think that proof of confusion is proof of use, like McKenna, it would logically follow that proof of dilution is proof of use for purposes of dilution. But this creates serious problems for the US concept of dilution; too many things cause dilution that we still want to allow, so we need a bunch of exclusions. If we defined dilution as commercial free riding and disfavored comparative advertising, parody, and so on the way the Europeans do, we’d have a lot fewer problems of definition, though I obviously think that would make dilution an even worse idea. In fact, I have taken the position that our concept of dilution, combined with our exclusions, render current dilution law an unconstitutionally irrational regulation of commercial speech.

Relatedly, if your definition of dilution is interference with source significance. Trademark owners quite vocally claim that non-source-identifying uses can be dilutive, specifically tarnishing and genericizing uses. We may disagree, but I have to respectfully part ways with Mark McKenna’s argument that use as a mark or as an indicator of source is inherent in the concept of dilution (as he more persuasively argues that it is also inherent in confusion). Interference with source-identifying function can be caused by acts that are in themselves not source-identifying. Only a purely Schechterian dilution law, which nobody seems to have (or want) would be tied to the junior user’s source-identifying use.

I increasingly see trademark use as a heuristic—a way of implementing two intertwined propositions, one normative and one descriptive. Normatively, certain practices are desireable, usually because they either promote competition or free speech; we thus should exclude them from the scope of trademark law, whether confusion or dilution. Descriptively, such practices are also exceedingly unlikely to cause the harms targeted by trademark law, whether confusion or (perhaps less persuasively) dilution.

Seen in this way, a use requirement joins a number of other useful heuristics in trademark law, most prominently nominative fair use and the rule in Rogers v. Grimaldi, but also rules like the secondary meaning requirement in Wal-Mart for product design. Justice Scalia is quite up-front about intertwining descriptive and normative—he acknowledges the possibility of inherently distinctive product design in occasional cases, but declares that the game isn’t worth the candle. It’s not surprising that the language of use as a mark and consumer perceptions of use as an identifier of source show up in these kinds of cases too.

Here’s a radical claim: the multifactor confusion test has been a mistake. It leads to continual expansion of the law. Notably, the major innovations (and the classic limiting doctrines) take the form of substitutes or bypasses for the confusion test. Along with Rogers and Wal-Mart, there’s functionality; TM use; nominative fair use; abandonment; genericity; and the proposed materiality requirement that McKenna, Barrett, and I like. Kozinski holds in Mattel v. Walking Mountain that transformativeness in copyright fair use is a matter of law, no matter what your consumer survey says; why not make most of these doctrines, including confusion, into a matter of law?

This move would be even more important in dilution because we have no idea what dilution is; rather than searching for harm, simply ban identical or nearly identical uses of very famous marks (subject to exceptions, like use in noncommercial speech).

This might enable us to avoid confusion/dilution conundrums because we would no longer be asking about the harm, if any, from a particular act, but about the type of behavior. Competition might be better off if “exclusive right to use” really meant “exclusive right to use” within the scope of registration (again, subject to exceptions/defenses like descriptive fair use). We wouldn’t ask about confusion as part of plaintiff’s case, though lack of confusion might be a defense. Note that this is not a restrictionist position—TM owners get a lot out of not having to show likely confusion within their field of registration—but it might be better for restrictionists overall if that enabled lots of uses by noncompetitors.

Bone: He blames the Polaroid case! Maybe we have to worry about noncompetitive goods causing harm in certain cases; the Second Circuit gave up, though, and made potential sources of harm factors in every case.

The exclusions in dilution define a set of activities that don’t create the harms we care about. But we might say there’s no such set. Here’s a story about the harm of the Tiffany strip joint: the meanings of “Tiffany” jewelry are part of the product, so tarnishing uses change the product for the users. It’s more likely, however, that a branding use will affect the firm—Tiffany in a porn movie won’t have as much effect on Tiffany as a store as Tiffany in a source-identifying context. The similarity in marketplace contexts—that they’re both source-identifying—causes the dilution.

There’s a network/node information quality story to tell about blurring.

Bone: On the enforcement cost side, a TM use requirement for blurring is good. Free riding cuts across everything, creating a huge risk that judges will use blurring against anyone. You want to nip that in the bud with a TM use requirement.

Tarnishment may be different; categorical defenses might suffice to protect parody, etc.

Goldman: If you buy these harms, the TM use requirement seems counterproductive; a blurring use could be noncommercial.

Bone: The harms flow from source-identifying uses.

Litman: If dilution did work that way, naming my dog Tiffany would be different than naming my store Tiffany. If our minds don’t work that way, Bone’s story doesn’t work.

Comment: This seems like a just-so story to me. I don’t think there’s any evidence that mental categories map in any way to “use as a mark.” In other words, “car” is no closer to “jewelry” than it is to “dog,” so the name of any one of them is no more or less likely to have an effect on the others.

Lemley: It might be that all these uses dilute, but we worry more about restricting the non-TM uses for other (speech) reasons than about restricting TM uses. We’re also quite sure that a dilution law without a TM use requirement is unconstitutional, but a limited law might be constitutional. (I’m quite sure it’s not!)

Beebe to Dinwoodie: Does §43(c) strengthen your argument that use isn’t a separate requirement in the rest of the Lanham Act?

Dinwoodie: People will make that argument, but Congress wasn’t thinking about the issue.

Dogan: Referential uses tend to change meaning of marks but still reinforce their singularity as source identifiers. Source identification is a very strong limiting principle.

Bone: The other account of blurring is added associations, which can come from references. If you reject that account of blurring, source identification makes more sense.

Dinwoodie: The Europeans are limiting dilution to the “essential function” of a mark—its marketing function—a move that forces courts to ask what they’re meant to be protecting. A German court found that toy replicas of cars were allowed because they didn’t interfere with the essential function of the mark for cars.

Iowa Trademark Scholars' Roundtable, part 2

Eric Goldman: Is TM use the right solution or just the best option on the table?

Lemley: What are the alternatives? All advantages are comparative: bring me a doctrine that works better. All existing combinations of little doctrines are insufficient, subject to manipulation in new technological environments. A keyword ad exception would be too small.

Bone: Say more about search costs—desktop clutter is also a search cost.

Lemley: Fair point. Lemley wants effective operation of a marketplace. TM’s normal way to do that is to respond to a particular information failure—but sometimes decreasing search costs isn’t worth other costs, which explains the functionality doctrine.

Bone: This move disturbs me. Saying that decreasing confusion isn’t worth it is different than saying TM doesn’t care about certain kinds of confusion. That’s an empirical debate about the costs of reducing search costs. Alternatively, we could argue that some search costs wouldn’t count.

Lemley: He advocates a rule that we don’t consider cost/benefit in each specific case.

Jessica Litman: She takes a positivist approach: the word choices in the statute were made intentionally, moving from affixation to a very similar definition of use in commerce. Barrett’s point is that Congress knew what it wanted. Litman’s view of TM use is far even from Lemley & Dogan. There’s always a potential harm: a hard-core movie where a character washes his clothes with Tide can do real damage to Tide because people could really believe it was paid product placement. That’s not actionable, not because it’s nominative fair use or free speech but because that fails to state a claim.

Dogan: How do you deal with the advertising problem (this interpretation means that use in advertising isn’t “use in commerce” with respect to marks for goods) or the waiter who substitutes Coke for Pepsi without ever saying the terms?

Litman: She accepts the physical affixation-esque requirement. If you don’t fit within §32, she’s happy to send you to §43, and if you don’t fit within §43 she’s happy to send you to Congress.

McKenna: But the definition of infringement refers to use in advertising. So it was explicitly contemplated.

Litman: That’s §43. She’s not willing to throw the whole affixation baby out. (Sometimes I really love hardcore stances like this.)

Dinwoodie: One could take the “deeming” provision of the definition of use in commerce to be nonexhaustive—such uses are deemed use in commerce but others might be too.

Lemley: Then we have to throw out all our law about rights acquisition too.

Dinwoodie: No, Congress wanted to eliminate token use.

McKenna: But then all sorts of things that we don’t (necessarily) think are source-identifying automatically count, like putting a device on a cap. The move to §43(a) doesn’t make sense to him—if the point is to identify things that are ok, how does moving to §43 help? Also, §43 still requires “use.”

My interjection: the Lanham Act only defines use in commerce with respect to what counts as use as a mark, not what counts as false advertising.

Barrett: The concept needs more work, but she thinks fraudulent marketing could be distinguished from TM infringement—the Restatement (3d) of Unfair Competition covers not just false advertising but passing off through acts other than use as a mark. That’s ok with her because it avoids propertizing words and symbols.

My comment: At least you’d get away from the multifactor confusion test, where the big 3 (strength, similarity, relation of goods) will favor the plaintiff in many cases where it shouldn’t.

Barrett: Yes, there’s even a materiality requirement.

Heymann: It’s not uncommon to exclude categories of relationships or harms as a matter of law—done with duty and proximate causation in tort (Palsgraf).

Bone: That’s a function of the underlying value: sometimes it’s nonutilitarian. Is there a nonutilitarian value in TM?

Goldman: The statute doesn’t work, fundamentally, for how we think about TM. We aren’t going to reach consensus on statutory interpretation. If we as scholars can’t agree on principles, it’s hopeless.

Dinwoodie: Congress delegated a lot to the courts. What do we do when Congress also used very specific words to inform the common-law development of doctrine? On dilution, the Supreme Court looked very closely at the statutory language; likewise, the original Lanham Act was focused on registration and Congress’s language is very important for registration. Outside that realm, Dinwoodie is more comfortable playing with the language and allocating discretion to courts.

Lemley: All sorts of judicially created doctrines have no explicit statutory basis, even in copyright. If you read the Lanham Act language as covering the waterfront on use in commerce, that leaves little room for maneuver. If Dinwoodie is right that the language is only for rights acquisition and not infringement, though, you could say (1) there’s no TM use requirement or (2) courts are free to create a common law TM use requirement.

Barrett: Congress codified what it understood to be the common law, with exceptions like incontestability, and expected courts to continue to develop the common law.

Austin: He is pessimistic about consensus about what counts as an externality/what should be done. There are two different conceptions of the role of the state: (1) paternalistic, using TM to protect consumers; (2) requiring calibrated and robust reasons for regulation, with the default being no regulation. (Comment: note however that property concepts can be used to flip (2), since property protection is usually seen as a neutral background rule rather than state intervention.)

Dinwoodie: Consumer search costs aren’t the only thing TM is for.

Lemley: He finds it ironic that the claim “search costs theory is insufficiently rich to bring in culture, free speech, etc.” is being used to create a world free of those things because we allow TM to expand with no barrier. (My thought: Lemley doesn’t want to say, at least not here, that search costs does account for these things, but First Amendment absolutists would say that marketplace of ideas concepts can encompass these all. The Gay Olympics case comes to mind.)

Dinwoodie: There are other ways to limit TM! Multiple values exist and need to be engaged.

Bone: But really, these uses we’re talking about increase search costs—but who cares? They don’t increase search costs in the ways TM cares about.

McKenna: The argument is that individual assessments of search costs are outweighed by dynamic effects on institutions—e.g., search engines.

Dogan: Search costs minimization is the justification for TM rights, but the good is improving information in the market to increase competitiveness. Search costs aren’t the end result. Goldman points out that intermediaries have incentives to collate information well and TM may interfere with that process.

Bone: Free speech doesn’t always improve information for consumers. There’s a balance between confusion and harms to First Amendment values. We don’t think consumers will adapt to the law in every case and we don’t think “Barbie Girl” improves consumer information. The hard cases are where there is some confusion and increase in search costs.

Lemley: We accept a First Amendment set of rules/values which limit TM in cases even of TM use: Haute Diggity Dog involved a TM use, but other doctrines are relevant. Within the domain of uses that aren’t branding/advertising uses, should we have a rule or a standard?

McKenna: Part of what leads to “confusion uber alles” is the search costs story. That view makes it too easy to tell stories about how any use causes increased search costs.

Barrett: How about an objective multifactor test?

McKenna: Objective multifactor? (Said in tone indicating that the phrase is a contradiction in terms.)

Barrett: Offers factors for assessing TM use: (1) Can the consumer perceive the defendant’s use of the mark? (2) Does the defendant closely associate the mark with goods or services it’s offering for sale or distribution? (3) Is the use likely to make a separate commercial impression on consumers? (4) Is the defendant’s use in noncommercial speech? (5) Is there some other important non-TM purpose for the defendant’s use? (6) Does the defendant’s use further a business that’s overall pro-competitive, or a service that gives useful information to consumers? (7) If it’s new technology, is there any useful analogy to more familiar practices?

Dogan: What do you get out of the rest of it if (1) and (2) are answered yes?

Barrett: (1) is dispositive, (2) is a minimum threshold.

McKenna: If you say that bringing Pepsi in response to a request for Coke is use under this, I don’t see how Google gets out.

Dogan: She’s troubled by the lack of constraint imposed by a multifactor test, as Beebe has documented.

Christine Bohannon: In copyright you can always claim a lost licensing fee to fight fair use—there’s a similar circularity. We need some baseline. Looking at weak cases is a good start. Barrett’s factors seem like reasonable questions for assessing likely confusion.

Lemley: What worries him about a multifactor test is rules v. standards, and the in terrorem effect of threats to sue; he wants a rule. That said, how do we define branding use? Need not set out with specificity in the statute; could give examples of branding and nonbranding, then figure out how to distinguish at the margin. (Trade dress, presumably, would be some tertium quid.) It’s wrong to require absolute precision from TM use when the alternative is the horribly flawed “confusion.”

Bone: Courts will develop the idea of branding and develop fact-sensitive factors because that’s what courts do unless you make clear they can’t. Forget about the Pepsi/Coke passing off example; don’t try to make TM use do too much.

Lemley: But passing off is in the Lanham Act without an intent requirement, so all cases will get shuffled there if you funnel non-affixation uses into passing off. You can make the same claims under §43(a)—an implied representation of endorsement. (Comment: that would happen unless courts adopt Dastar-esque preemption and say “what you really have here is a failed TM claim precluded by the TM use doctrine, so you can’t bring a passing off claim.”) Even polestar cases of non-use would now need this filter if courts adopt Barrett’s test, and that scares him.

Iowa Trademark Scholars' Roundtable, part 1

Because of travel delays, I arrived after the discussion had begun. NB: If you don’t already have a familiarity with arguments over “trademark use,” this set of posts will likely be hard to understand; I recommend reading at least some of the sources in the previous post first.

Mark Lemley: McKenna’s argument is an incompleteness theorem: we can’t completely describe a formal system within that system. “Source” needs an outside definition. Does that mean we need abandon all hope? No, no more than Gödel’s incompleteness theory means we stop teaching math. Beebe’s study demonstrates the problems of courts deciding cases ad hoc: intent, which is entirely malleable, dominates; it’s so manipulable because we have no idea of “intent to do what?” This is a fundamental problem in confusion, and may also be so in use.

The tendency is for courts to adopt a use requirement but not do anything systematic or robust with it. Problems include ambush marketing/attempts by TM owners to stop competition. “Unfair competition” is redundant in their eyes. We need a legal doctrine to say that not all competition is unfair, and confusion isn’t it because there’s always a story about confusion over something. We need to avoid a factfinding requirement that will lead to lots of trials, especially because surveys can be so readily manipulated. Even Google, which is aggressive on this, doesn’t allow use of a TM in the text of an ad when the law clearly would (explicit comparative advertising). In practice, a search for Rescuecom no longer produces a targeted result because R.com threatened to sue the Geek Squad etc. and drove them off of bidding on the term.

Graeme Dinwoodie: Two questions drive this first session—what is TM use, and how would it change TM policy? (Nice framing, Graeme!) There are also institutional questions about the role of courts v. Congress, which may be much more significant. Some of this is US-specific but judges v. legislators is more universal. Dinwoodie believes that, even if we want to validate certain practices, let’s accept that confusion of different kinds might cause harms, then weigh them against the costs of regulation.

(Comment: This is why I teach TM as containing a bunch of rules about how “the game isn’t worth the candle”—such balancing seems to me dangerous to competition unless it’s done at the abstract level.)

Mark Lemley and Stacey Dogan say that non-TM uses don’t increase search costs; TM use is then a redundant requirement. It’s an enforcement cost issue that leads to “TM use” as the name of the theory. Margreth Barrett focuses more on competition and free speech issues that need to be internalized to TM law. It seems from Bentley’s discussion (which I missed) that the European notion of use is circular as well. The biggest lesson: use is very fact-intensive when linked to its purpose, even when the question involves association and not confusion.

Dogan: McKenna’s position is close to hers. Can we draw a line and say that courts can determine as a matter of law that these aren’t TM uses: (1) plaintiff has a weak case—this is the kind of behavior unlikely to cause confusion because historically it’s been allowed/unchallenged; and (2) there’s an important countervailing societal value—e.g., descriptive fair use. Add a new category: people whose relation to confusion is at best secondary/contributory.

Rules are critical to shaping norms and consumer expectations. Traffix and Wal-Mart give her some hope that this can be achieved. On the issue of harms: information overload and confusion over whether search results are organic are paid, which are harms Dinwoodie and Janis invoke, are not TM issues but unfair trade practices that can and should be regulated as such, not through TM law.

Dinwoodie: Agrees these are mostly unfair competition cases.

Barrett: We can have a false marketing cause of action under §43(a)—dishonest behavior not requiring use as a mark. This is underutilized in US law. (She is not referring to §43(a)(1)(B) but a more specific cause of action.)

McKenna: Courts ought to use values to limit TM, but all those other doctrines have specific names and rationales. We need to know what links those doctrines to TM use if TM use is primarily about protecting search engines. Why push anything over to unfair competition?

Dogan: No, don’t push: if the problem is labeling of paid search results, regulate that regardless of TM law.

McKenna: Courts have fetishized “confusion” in the abstract—we should focus on the kind of confusion TM is supposed to avert.

Dogan: Playboy in the 9^th Circuit mushed the ads and the organic results together.

Bob Bone: TM use can function as (1) boundary policing—what TM is supposed to cover in general, like copying in copyright: it defines a kind of activity regulated. In that view, nominative and descriptive fair use are evidence of a broader principle defining the domain of TM law. He doesn’t find that persuasive. Or it could be (2) an enforcement cost-limiting doctrine, not really about boundaries, but about the judgment that a rule creates fewer errors than case-by-case factfinding. He sees more emphasis on (2) these days and likes it better.

You need to know what kinds of errors are being made. Are courts messing up the inferences piled on inferences allowed by TM? E.g., intent to free ride à inference of intent to deceive à inference of likely confusion. In such cases, courts are implementing a different policy, not just making doctrinal mistakes. They may be mistakenly undervaluing free speech etc. So we need to find out what problem we’re trying to solve. Reeducating judges on doctrine isn’t sufficient. We need a rule at the front end. Maybe it needs to be done over smaller domains than TM law as a whole, though Bone thinks that single examples as Dinwoodie suggests may be too narrow.

Lemley: These aren’t competing but complementary stories. Search costs are the boundary definition, and the search costs theory explains where courts are going wrong. It’s true we can’t define TM use exactly, but our response is to tell the enforcement costs story. That helps us decide where to draw the line. If our only concern is that threats to sue deter certain behavior, there’s no reason to care unless there’s an underlying principle.

Bone: But it’s good to start with an analytic distinction before we recognize the complementarity. With copyright, we start with originality and copying and only then define fair use; we have a core domain. (I’m not sure I agree with this, and Bone seems to be uncertain himself—arguably, and as the CLS folks might tell us, each concept is defined by its opposition.)

Dogan: In the Netcom case, we saw that courts even define “copying” through common law interpretation, keeping the normative goals of the law in mind.

Graeme Austin: Dogan’s “weak case.” What makes a case weak? “Trivial” confusion. How will a surfer be harmed by mistaking a site for a Playboy site momentarily? In a weak case, other stories are available—the consumer is not confused, gets the joke, recognizes difference, understands that aftermarkets exist. Ambush marketing is a good example; research suggests that some people believe Nike sponsors everything. This ties into whether mere psychic harm to a guy who lands on a non-Playboy site is cognizable (like the harm in Victoria’s Secret).

Does that all go back to free speech and competition? Every counterstory maps to one or the other. So why wait until the defense stage to make this determination?

Dinwoodie: There may be other harms! A little or a lot. KP Permanent on remand is an example. If there’s 95% confusion, the counterstories are less plausible. We should make the harm story admissible for potential validation. And yes, that has a cost. Some defenses might develop as a matter of law.

McKenna: Weak cases can be weak in a lot of ways. A few people confused could mean lost sales, or a lot of people could be confused but not about anything important.

Laura Heymann: We’re discussing TM use as applying to certain relationships—defendant/mark or defendant/consumer. We could also speak of permissible relationships between plaintiffs and defendants around a mark. The expansion of infringement into sponsorship confusion is what’s causing the problem: the perception of whether defendant needed permission to use the mark.

My comment: In antitrust, courts developed a series of rules about what was automatically ok or not; most things are now rule of reason rather than per se illegal. That was driven by economic analysis; in TM it’s behavioral economics. We are suggesting a return to per se rules in order to allow more challenged behavior, or at least a requirement to prove what we now presume based on circumstantial factors like intent.

Dogan: consumer expectations/the iterative effect of rules on consumers make TM different from antitrust.

Christina Bohannon: The analogy in TM would be which practices are likely to cause confusion and which aren’t. But outside confsuion, we’re asking whether a use is the kind we want to allow. This relates to antitrust injury: is the harm due to competition—is it the kind of harm antitrust/TM cares about?

Lemley: The TM injury doctrine! You can’t sue because your harm is the wrong kind: a policy question.

Bohannon: See also harm due to substitution v. harm due to lost reputation in copyright’s fair use analysis. In copyright we’re comparing effects on creativity to each other—but in TM we aren’t comparing commensurables in the same way.

Dogan: Quick look v. rule of reason—takes pro- and anti-competitive arguments into account—that would be a way to deal with TM in weak cases.