Bloggers Beware: The Ideological Implications of the FTC’s Native Advertising Regulations for Bloggers

Monday, October 26th, 2015

They say that the best marketing tool is word of mouth. Despite the allure of clever product placement, viral videos, and creative billboard designs, the best advertisements are still personal. We tend to trust the opinions and experiences of our friends, family, neighbors, and acquaintances. A Nielson study shows that 92 percent of consumers worldwide trust “earned media” (i.e., recommendations from friends and family and word of mouth) over “paid media” (i.e., traditional advertisements). In this digital age, word of mouth may look a little different, but research clearly indicates that earned media is a powerful way to win new customers. While we are not discussing and recommending different products, brands, and services in person as often, we constantly share what we’re doing, what we’re buying, and what we’re “pinning” online. In turn, we trust the activity of our peers that we see on social media and we use those opinions to shape our consuming habits. These modern realities are part of the reason why native advertising is such a big problem from the perspective of the Federal Trade Commission (FTC), and why blogger activity should be of particular concern to the FTC as well.


What is Native Advertising?

Broadly speaking, native advertising, also known as “sponsored content,” is advertising that blends into its environment. The classic example occurs in print media, where the layout of one page in a magazine looks very similar to the other pages in the magazine, but is actually a paid advertisement. In the virtual world, news websites will often include affiliate links at the bottom of an article to other news stories that are actually advertisements—they are either articles written solely for the purpose of advertisement, or they are links to other news websites where the other websites paid to advertise on this one. Sometimes, websites will contain articles written in a style typical for the website, but the whole article will be sponsored by some brand or company, and will ultimately be an advertisement for said ad or company. One big industry player that frequently advertises in this style is Buzzfeed.

While native advertising is a completely legitimate advertising tool, the FTC is concerned about consumer welfare. The FTC believes that consumers must be informed and should not be deceived. Simply put, consumers need to know if they are looking at editorial content or an advertisement. The FTC’s Endorsement Guides states that:

If an endorser is acting on behalf of an advertiser, what she or he is saying is usually going to be commercial speech – and commercial speech violates the FTC Act if it’s deceptive. The FTC conducts investigations and brings cases involving endorsements under Section 5 of the FTC Act, which generally prohibits deceptive advertising.

In practice, this means that publishers, bloggers, and other content-creators need to make sure to provide appropriate disclosures on their sponsored content. For instance, a disclaimer in small text or in a shade that does not stand out from the background would not be enough. At the core of the FTC’s regulations on native advertising and sponsored content is a deep concern about consumer trust. Therefore, the test for violating the FTC Guidelines is based on how deceptive an advertisement is, measured by asking consumers questions about said advertisement in order to determine how misleading it is. An advertisement is considered deceptive if it misleads “a significant percentage of consumers,” which Mary Engle, the FTC’s associate director of advertising practices, states is generally 15 percent of consumers, but sometimes can be as few as 10 percent.


Clear Implications for Bloggers

Blogs can be a really powerful marketing medium because bloggers with a high readership can be tastemakers. This influence is especially true of lifestyle bloggers. Lifestyle bloggers curate a certain lifestyle, pretty and packaged, and present it to their audience as something they should shoot for. The content they create can have a huge reach. For instance, BlogHer, a popular online forum for women bloggers, reports over 40 million hits a month. Since many blogs make money via sponsored posts, these hits could translate into many consumers seeing and being strongly influenced by sponsored blog posts.

The FTC has warned that there may be legal ramifications for publishers that participate in creating the content of an ad. Participating in the creation of an ad blurs the line between merely displaying a paid advertisement or promotion, and creating original content for the purposes of advertisement. Such ads could open up the publisher to legal liability. When a blogger agrees to do a sponsored post, it usually falls into this bucket. The blogger is generally charged with creating original content which showcases and talks about a certain product in exchange for compensation. Therefore, it is critical that bloggers disclose clearly that they are being paid to promote something, since the consumer could easily misinterpret it as editorial content.

Moreover, the FTC has warned that online content creators need to be especially careful with regard to social media campaigns. Recently, for instance, Cole Haan ran a “Wandering Sole” social media contest via Pinterest. The contest encouraged Pinterest users to take pictures of their favorite Cole Haan shoes and their favorite places and tag them with “#WanderingSole” for a chance to win $1,000. Though the FTC did not take enforcement action against Cole Haan, the FTC stated that the advertising campaign was misleading to the average Pinterest user, who would not necessarily know why there were Cole Haan posts, and that the hashtag “#WanderingSole” was not sufficient disclosure. This controversy about appropriate disclosures in social media points to a central concern about misleading users of social media. In essence, the FTC is concerned that increased social media use for advertising purposes could easily be misinterpreted as individuals’ endorsements of products or brands without monetary incentives. Social media campaigns are dangerous territory because an average consumer might misinterpret social media presence as a legitimate “word of mouth” recommendation. This individual nature of social media use is analogous to the capacity and presentation of a typical blogger. It plays into the same concept of consumer trust that the FTC is worried about.


A New Frontier

The FTC has stated that it is generally not monitoring bloggers. After all, bloggers are usually individuals and not big corporations, and we generally assign more responsibility and expect more care in terms of following trade and advertisement regulations from big corporations. Under the current regulatory framework, it does not make sense for the FTC to prey on individual bloggers who make mistakes by insufficiently disclosing sponsorship relationships.

However, the FTC needs to seriously consider the implications of bloggers pushing sponsored content. As mentioned earlier, word of mouth is a powerful advertising tool. Blog readers tend to feel as though they “know” the blogger. Successful blogs, especially lifestyle blogs, publish quite regularly and share personal details and personal tastes. Considering the FTC’s concerns about manipulation of consumers via native advertising tactics, bloggers and their sponsored posts could pose some serious problems. If bloggers do not properly disclose, this form of advertising can be extremely misleading to the reader. From the reader’s perspective, he might trust the blogger’s opinion much like he would a friend or coworker. Thus, though the number of sponsored posts probably pale in comparison to the number of advertisements consumers encounter through general native advertising, sponsored blog posts are likely to have a greater influence on consumer behavior. If blog readers are not properly informed that the content they are reading is sponsored in some way, each dishonest and manipulative blog post can very much shape the consumer choices of blog readers.

Anti-Kickback Statute Remedies: Short-Term Gain for Long-Term Pain?

Monday, October 26th, 2015

The United States brought a suit against Novartis Pharmaceuticals Corporation (“Novartis”) under the False Claims Act claiming that Novartis orchestrated a kickback scheme which caused the government to pay out tens of millions in improper Medicare and Medicaid reimbursements. The government claimed that Novartis paid kickbacks, disguised as performance rebates or discounts, to pharmacies that switched patients from competitors and generics to the Novartis drug Myfortic. The Anti-Kickback Statute, 42 U.S.C. §1320a-7b(b), prohibits offering, paying, soliciting, or receiving remuneration to induce someone to purchase or recommend a drug covered by Medicare or Medicaid. The Novartis complaint states that a company could violate the Anti-Kickback Statute by offering financial benefits to a pharmacy in exchange for physicians recommending that their patients switch from one prescription drug to another.

Many False Claims Act suits settle. In the present Novartis litigation, for example, pharmacies like CVS Caremark, Walgreen, and Accredo have settled. Perhaps the harsh penalties imposed by the Anti-Kickback Statute make it more likely that defendants will settle rather than proceed to trial. A conviction under the Anti-Kickback Statute results in mandatory exclusion from participation in federal health care programs. Considering that, between Medicare and Medicaid, the federal government spends hundreds of billions of dollars on prescription drugs, such a sanction would result in a substantial loss of revenue for a health care company. While pharmacies can’t easily take on the risk of a conviction under the Anti-Kickback Statute, so far it seems like Novartis is willing to. Novartis may be willing to take the gamble because the government, while asking for damages and fines up to $3.35 billion, has not yet called for exclusion as a remedy in the case.

While exclusion, the ultimate penalty, would certainly cause crushing losses for Novartis and deter future pharmaceutical companies from instituting kickback schemes, exclusion could also prevent some patients from getting certain medicines. As the Wall Street Journal stated, “The feds need to be cautious about excluding a drug maker that provides numerous needed medications, because doing so could unnecessarily harm many patients.” The government’s discretion is particularly needed where, as here, the relevant drugs are still within the patent-protected time frame.

Patents are a notable exception to antitrust laws. Patents generally provide drug companies with a 20-year monopoly. Due to the extremely high cost of researching and testing new drugs, such a monopoly protects and promotes innovation of new drugs. Not every drug that goes through clinical trials reaches the market. In order to recoup losses sustained from the failed drugs, pharmaceutical companies need the temporary monopoly to make money on their successful drugs. Once generic versions of a drug are approved by the Food and Drug Administration and enter the market, most patients would rather pay less than use the brand-name version. Generic drug companies don’t have to go through as much research and testing as the original drug creators do, thus making production significantly cheaper. This process in turn provides cheaper prescription drugs to the public. However, the industry must balance the revenue required to create new drugs with the desire to provide affordable prescription drugs to those who need it.

The temporary monopoly system created by patents has promoted innovation in drugs. However, the temporary monopoly can be abused and can prevent patients from getting the medicine they need. Drug companies can charge what they want when they have the only treatment on the market. For example, a health start-up increased the price of Daraprim, a parasitic infection drug, from $13.50 to $750 a tablet. This price increase made it too expensive for some hospitals to keep Daraprim in stock and thus will surely delay treatment for their patients. Daraprim, approved by the FDA in 1953, isn’t even a brand-new, patent-protected drug.

By eliminating a player from a market that is already small, the mandatory exclusion provision of the Anti-Kickback Statute exacerbates the monopoly of certain pharmaceutical companies and drives up drug prices even more. On the one hand, the convicted company once competed against other drug companies, who can now charge a higher price with less competition. On the other hand, the convicted company owned the patent for a drug which only it sold. Either patients can no longer get the drug, or the convicted company would sell the patent, in which case the purchasing company still retains the monopoly and is still free to charge any price.

In the Novartis case, the government claims to have received half a million dollars’ worth of reimbursements from Medicare and Medicaid for Myfortic. However, is that amount really as high as the government makes it out to be, especially compared to the higher drug prices (and thus higher reimbursements) a monopoly-like market would entail? Mandatory exclusion forces the pharmaceutical market to be more monopoly-like, thus increasing the price the drug market can support. This knowledge can prevent large drug companies like Novartis from settling, because they believe the government has more to lose. Conversely, the pharmacies, which do not have such monopoly power, can’t afford to be shut out of the market and therefore can’t risk conviction under the Anti-Kickback statute. Therefore, as in the Novartis case, the pharmacies have settled while the pharmaceutical company has stayed in the fight. In conclusion, while the government does need a harsh penalty to prevent pharmaceutical companies from providing kickbacks to promote their products, exclusion hurts the pharmaceutical market by making it more like a monopoly, thus increasing drug prices and decreasing access.

The SEC’s Appointment Problem & Its Likely Solution

Wednesday, October 21st, 2015

With the passage of Section 929P(a) of the Dodd-Frank Act in 2010, the SEC saw the largest expansion of its administrative enforcement power to date. Prior to that, SEC administrative proceedings were limited to obtaining an order enjoining violations of the Exchange Act. Section 929P(a) amended this, authorizing the SEC to seek civil monetary penalties from “any person” in an administrative hearing.

While the SEC still brings a majority of its cases in federal courts—roughly 63 percent this fiscal year through June—a growing proportion of cases have been tried in its internal administrative tribunals, a trend the agency intends to continue. These internal proceedings are presided over by Administrative Law Judges (“ALJs”), whom the SEC describes as “independent judicial officers … [imbued with the power to] issue subpoenas, conduct prehearing conferences, issue defaults, and rule on motions and the admissibility of evidence.”

Expansion of internal administrative proceedings, however, has hit a snag; myriad challenges have been brought contesting the constitutionality of the ALJs’ appointments. Former Standard & Poor executive, Barbara Duka, for example, has brought a challenge in the U.S. District Court for the Southern District of New York, alleging “that the ALJs’ appointments violate the Appointments Clause because the ALJs, as ‘inferior officers’ under Article II, may only so preside on due and proper appointment by a constitutional Officer, here, the [SEC] Commission[,]” which has not occurred. In other words, the allegations suggest that the ALJs are inferior officers, rather than employees, and thus must be appointed under Article II. Ms. Duka’s challenge is likely to result in a favorable ruling, as Judge Richard Berman issued a preliminary injunction, as Ms. Duka had “demonstrated irreparable harm along with a substantial likelihood of success on the merits of her claim ….”

“The line between ‘mere’ employees and inferior officers is anything but bright.” The Supreme Court in Freytag, the seminal Appointment Clause case, articulated a definitional standard for inferior officers, predicated on “significant authority pursuant to the laws of the United States.” The Freytag court concluded that Tax Court judges meet this “significant authority” standard, as their duties include taking testimony, conducting trials, ruling on the admissibility of evidence, and enforcing compliance with discovery orders.

At no point has the SEC seriously suggested that the ALJs are appointed by the Commission itself, as would be required under Article II. Rather, the SEC has maintained that ALJs are not inferior officers but employees, and thus not governed by the Appointments Clause. The SEC’s argument is predicated on an interpretation of Freytag that views “the authority to issue final rulings” as a necessary component of the “significant authority” standard—an authority SEC ALJs do not possess. This interpretation of Freytag has, thus far, not been persuasive. In Duka v. SEC, Judge Berman stated, “the SEC ALJs are ‘inferior officers’ because they exercise ‘significant authority pursuant to the laws of the United States.’”  Furthermore, in the Freytag opinion itself, the Supreme Court noted that the authority to issue final decisions was not a necessary component of inferior officers as “this argument ignores the significance of the duties and discretion that special trial judges possess.”

Whether a “quick fix” is possible has not been articulated by the SEC. Judge May queried whether such a solution existed during a judicial conference, to which the government responded that “the Commission should not act precipitously to modify its ALJ scheme.” In the event that ALJs are determined to be inferior officers, which appears likely, the SEC would need to deviate from the status quo. Although the Commission has ducked the question when posed directly, there does not appear to be any great hurdle standing in the way of meeting the Appointments Clause. Indeed, the FTC, facing a similar challenge to an ALJ proceeding, ratified the appointment of the in-house judge “to ward off any possible claim that [the] administrative proceeding violates the Appointment Clause.” Nothing suggests that the SEC is incapable of a similar prophylactic measure.

Fortunately for the SEC, it seems improbable courts would willingly apply such a ruling retroactively given the administrative impracticalities of vacating all past proceedings. Rather, the ruling is likely to be applied prospectively—according past proceedings de facto validity—and stayed to allow the SEC to implement remedial measures, i.e., ratifying the ALJs’ appointments. Far from a novel solution, this would be akin to the approach taken by the Supreme Court after bankruptcy courts were ruled unconstitutional in Northern Pipeline Construction Co. v. Marathon Pipe Line Co. The court noted that, “Th[e] limited stay w[ould] afford Congress an opportunity to reconstitute the bankruptcy courts … without impairing the interim administration of the bankruptcy laws.” Other examples of prospective rulings intended to cure the impracticality of vacating past actions include: Buckley v. Valeo, Chicot Cty. Drainage Dist. v. Baxter State Bank, and Ins. Corp. of Ir. v. Compagnie des Bauxites de Guinee.

Although it appears likely that the appointment of ALJs will be deemed to violate the Appointments Clause, the practical impact would be limited given a prospective ruling.  Assuming the SEC adheres to the appointment requirement moving forward by ratifying its ALJs, the administrative proceedings can continue unimpeded.  Indeed, the SEC may even undertake ratification without announcement to save face, although it may be well advised not to do so given the public exposure of this issue.



Wendt Revisited: How Fictional Character Endorsement Could Undermine Copyright Protections

Wednesday, October 21st, 2015

There has been a recent phenomenon of actors, when endorsing companies and products, eerily recreating their television characters in commercials. For example, in a recent series of Lincoln commercials, Matthew McConaughey appeared in a close recreation of his character from the show True Detective. The advertisements, like the show, featured McConaughey driving along empty roads, pessimistically musing on deep subjects. Some commentators like Conan O’Brien took the opportunity to poke fun at the similarity by dubbing True Detective lines over the Lincoln commercials. In another example, Kevin Spacey recently released a series of commercials for E*Trade in which Spacey appears to dress and speak in a manner similar to that of his character Frank Underwood from House of Cards. This phenomenon threatens to bring back to the surface the tension between federal copyright law and state-granted rights of publicity. One body of law suggests that individuals have control over their likenesses, even when the representations simply remind the consumer of them. The second body of law gives protections to the creators of fictional characters through copyright and other protections. Though the Ninth Circuit has said that the right to publicity can trump copyright protection of fictional characters, the debate may be reopened by the proliferation of these commercials.

Protection of Fictional Characters

Copyright law does not provide any independent or separate protection for fictional characters. Though a character may be protected in some ways by normal copyright law, when characters have extended uses (e.g. toys or trading cards) or are created independent of a specific medium, they are often left insufficiently protected. Though trademark protection might be available for some fictional characters, for most, trademark would be an odd and inappropriate match. Thus, fictional characters occupy a precarious position in copyright law: sometimes protected while other times left to be copied by others. Additionally, in film and television, it may be even more difficult to enforce copyright infringements where the owner of the copyright may be indebted to the actor who helped to create the character.

Rights to Publicity

Rights to publicity granted by states, most notably California, allow individuals to control the way in which others use their likeness. This right allows individuals a fairly intangible ownership of the attributes of their public persona, which they have developed. In the perhaps most famous and liberal application of the right to publicity, Vanna White, host of the game show Wheel of Fortune, sued Samsung under California law. Samsung had been running an advertisement in which a robot styled like Ms. White stood next to a game board reminiscent of that used in Wheel of Fortune. Certainly, this would remind most viewers of the show and Ms. White. However, the advertisement did not directly mention Ms. White, nor did it use, without consent, any copyrighted material. The Ninth Circuit found that the advertisement violated Ms. White’s right to publicity, as it utilized the persona she had worked to develop without permission or compensation.  In a more narrow interpretation, the District Court for the Southern District of New York found that a commercial which featured an M&M candy dressed like the Time Square “Naked Cowboy” did not violate the “Naked Cowboy’s” right to publicity because the commercial, though an allusion to the “Naked Cowboy,” was not, under New York law, direct enough in its portrayal.

Wendt Revisited

In respect to fictional characters in film and television, copyright protections and rights to publicity have often been on a collision course where studios, the usual owners of the copyright, and actors feud over control of the characters. Wendt, a 1997 case, remains the most prominent authority on this issue. Paramount Studios had licensed its copyright for the show Cheers to Host International, which then placed recreations of the famous Cheers bar with robot bartenders as advertisements in airports. The two actors who played the bartenders in the show sued, claiming their rights to publicity had been violated. The Ninth Circuit found that the rights of the actors came before the copyright protections asserted by Host and Paramount. In his dissent from the decision to not rehear the case en banc, Judge Alex Kozinski pointed to the dangerous implications of such logic. He argued that these rights might make it impossible for a copyright owner to adequately use the copyright. For example, if a television show decided to create a spinoff or recast a character, or if a movie decided to make a sequel, an actor, with a newfound right to publicity in her character, could demand to be compensated or even keep the studio from using the character in the new product.

The argument presented by the plaintiffs in Wendt and adopted by the court conflates the rights of the characters with the rights of the individual actors involved. The actor, asserting her personal control of her likeness, is able to control a character, which she portrayed as a subset of her likeness. However, the character, which many people helped create (e.g. writers, directors, other actors) and the traits of the actual individual may very well be conceptually distinct.


These new advertisements, in which actors resemble their fictional characters, threaten to bring Wendt back into the forefront. Moreover, they validate some of the concerns of the Wendt dissent. In these future cases the parties may be reversed, with the actor and advertiser defending a claim of copyright infringement from a copyright owner, but the stakes will remain the same. If the actor has virtually unlimited right to sell and publicize the fictional character the actor previously portrayed, then copyright protection may be ineffective. Studios and other copyright owners may be held hostage by actors’ rights to publicity

What is a “Threatening” Facebook Post?

Wednesday, October 21st, 2015

On June 1, 2015, the Supreme Court ruled in Elonis v. United States that an online post cannot be considered “threatening” without intent from the author to convey a threat, vacating Anthony Elonis’ conviction.  Elonis, in the midst of a divorce, had posted under a pseudonym on Facebook rap lyrics indicating that he wanted to kill his wife and the President of the United States, among other people, leading to an indictment on five counts of threat under 18 USC § 875(c).  Under the statute, “[w]hoever transmits in interstate or foreign commerce any communication containing any threat to kidnap any person or any threat to injure the person of another, shall be fined under this title or imprisoned not more than five years, or both.”  The jury convicted him on four of the five counts, sentencing him to forty-four months in prison.

Elonis requested a jury instruction that “the government must prove that he intended to communicate a true threat” in order to convict him—a request that was denied.  Elonis’ appeal to the Third Circuit was unsuccessful.  The Third Circuit upheld his conviction under the reasonable person test—whether a reasonable observer would consider the post to constitute a threat, regardless of whether the author intended for the message to act as a threat.  The Supreme Court overruled the reasonable person test, stating that the true test for conviction under 18 USC § 875(c) is whether the author had a subjective intent to convey threat through the message.  Writing for the majority, Chief Justice John Roberts noted that “[h]aving liability turn on whether a ‘reasonable person’ regards the communication as a threat—regardless of what the defendant thinks—’reduces culpability on the all-important element of the crime to negligence.’”  The Court has “long been reluctant to infer that a negligence standard was intended in criminal statutes.'” While the reasonable person standard is appropriate for tort cases, criminal conviction requires proof of subjective intent.

Is it enough that Elonis knew that his post could be conceived as a threat?  The majority opinion did not make this clear—a point that Justices Alito and Thomas criticized in their opinions.  Alluding to the decision as a cause of confusion among the lower courts, Alito wrote, “The Court holds that the jury instructions in this case were defective because they required only negligence in conveying a threat.  But the court refuses to explain what type of intent was necessary.” So while negligence is not enough, the requisite mental state required for a conviction remains to be seen.

Lenz v. Universal and the Suggestion of Robots to Assess Copyright Infringement

Saturday, October 10th, 2015

In February 2007, Stephanie Lenz uploaded a homemade video to YouTube of her young children dancing, with Prince’s “Let’s Go Crazy” playing on a home stereo in the kitchen. About four months later, Universal Music sent YouTube a takedown notification that identified Lenz’s video as an unauthorized use of Prince’s song. Represented by the Electronic Frontier Foundation, Lenz claimed that Universal wrongly went after the video, in violation of Section 512 of the Digital Millennium Copyright Act (“DMCA”).

On September 14, 2015 the United States Court of Appeals for the Ninth Circuit ruled in favor of scores of YouTube users like Lenz, holding that copyright holders must consider fair use before trying to remove online content.

Copyright Holders Must Consider Fair Use

Section 512(c) of the DMCA requires a copyright holder to include a statement that it believes in good faith that a particular use is unauthorized. In the first part of its ruling, the Court held that copyright holders must consider fair use before sending a takedown notice.

Codified in the Copyright Act of 1976 (17 U.S.C. § 107) (“Section 107”), the fair use doctrine, allows us to use portions of a copyrighted work for a variety of purposes, including comment, teaching, and scholarship. Whether a particular use is a fair use—and therefore not copyright-infringing—depends on four factors:

  1. Purpose and character of the use, including whether such use is of commercial nature or is for nonprofit educational purpose
  2. Nature of the copyrighted work
  3. Amount/substantiality of the portion used in relation to the copyrighted work as a whole
  4. Effect of the use upon the potential market for or value of the copyrighted work

This is a fact-intensive inquiry for a court overseeing a copyright dispute. The Ninth Circuit held that fair use is not just an affirmative defense, but rather is expressly authorized.

Universal Media’s procedures to determine copyright infringement did not explicitly address fair use: An employee would watch a video and determine infringement based on whether there was significant use of a song (especially if the song was recognizable) and if it composed a significant portion of or was the focus of the video. Thus, the Ninth Circuit stated that a jury should determine whether the procedures were enough to form the required subjective good faith belief.

Can Computers Fairly Assess Fair Use?

In the second part of the ruling, though, the Ninth Circuit suggested that computer algorithms may be a “valid and good faith middle ground for processing a plethora of content while still meeting” the DMCA’s requirement. The Court further noted that humans could provide a final check to ensure the flagged content indeed is infringing. However, the consideration “need not be searching or intensive.”  Some legal and industry experts note the lack of clarity here: If the investigation is not “intensive,” then how much is enough for “good faith” consideration? The middle ground suggestion of computer algorithms raises fairness questions: In his partial dissent, Judge Smith questioned whether a computer program could apply all four factors of Section 107.

Given that users upload upwards of 300 hours of video onto YouTube every minute, companies presumably must resort to computer algorithms and robots to trawl the Internet. On the one hand, tasks like tracking copyrighted audio in YouTube videos can be easily automated. YouTube uses an automated copyright system called Content ID, where uploaded videos are scanned against files submitted by copyright owners. Computer algorithms can, however, be over- and under-inclusive as well. Recently, Content ID flagged hundreds of gaming videos uploaded by gamers, and YouTube has defended its robots’ actions despite support from video game companies for the gamers’ clips. Professor Noah Feldman of Harvard Law contemplates that parody can easily include an entire work, while someone may bury a pirated copy within extraneous material to avoid algorithms.

There are other more absurd examples of wrongly accused copyright infringement, including interruptions in a live speech given by Michelle Obama and at the Hugo Awards before Neil Gaiman’s acceptance speech. So far, humans seem to find it much easier to determine that a television show clip shown at an awards show is a clear example of fair use.

Many companies have already automated their DMCA takeover notification system. Naturally, there now is a growing history of automated DMCA takedown notification abuse. Automattic, the company in charge of WordPress, recently described an ironic instance of a robot takedown. The company received a takedown notification stating that a particular site infringed on an academic’s copyright. Lo and behold, that site was the academic’s: He had employed an agency that used bots to scour for infringements, except this time the bots mistakenly targeted the copyright holder.

Last year, a federal district court judge in Florida ordered Warner Brothers to unseal documents regarding its notice and takedown practices in connection to a now-defunct litigation related to Hotfile, a file-storing and file-sharing service. There, Warner Brothers employed a system that looked only at titles, page names, and other superficial attributes of files in Hotfile’s systems. Judge Williams expressed skepticism that a computer automated process could meet DMCA requirements if it did not look into file contents at all.

Final Thoughts

The Ninth Circuit’s opinion may curtail takedown notifications to “those instances where [computer algorithms] might reasonably be capable of achieving the right result occasionally, rather than what we seem to have got now, where it’s more or less free to run amok.” The Lenz ruling seems to clarify the requirements for computer algorithms identifying copyrighted material—thus hopefully relieving DMCA takeover notification abuse like in Hotfile. However, because the Court did not provide further specifics or standards for a computer algorithm, further test cases will need to flesh this out.

Until computer learning becomes refined enough to address factors such as the nature of the work, the opinion does not seem as big a boon to fair use as it could be. If Universal only employed humans to review “the minimal remaining content a computer program does not cull,” then they are still relying on the algorithm to do the bulk of their work. As noted, automated notices are imperfect. Kembrew McLeod, professor at the University of Iowa and frequent commenter on areas of copyright law, believes that “fair use does not exist in practice because companies that are relying on these databases of copyrighted work can immediately shut off the public’s access” via takedown notifications. As the situation of Lenz and the Electronic Frontier Foundation indicates, standing up to a notification is more an act of principle than something any normal YouTube uploader can undertake.

Proliferation of Health Mobile Apps in the Era of Big Data

Thursday, October 8th, 2015

Explosion of Health Mobile Apps

The market for health apps is booming. According to IMS Health, a healthcare analytics provider, there are more than 165,000 mobile health apps available on the market. Research2guidance, a mobile market research firm, estimates that there are currently more than 100,000 health apps available in the iTunes and Google Play stores and predicts that the market for mobile apps will be approximately $26 billion by 2017.

The scope of health data handled by such apps has been expanding as well. For example, Apple HealthKit provides platforms for health data categories such as body measurement, fitness, sleep, and vitals. Beginning with iOS 9, HealthKit accepts new data types such as “reproductive health” that include cervical mucus quality, menstruation, ovulation test result, sexual activity, and spotting. Other more sophisticated health apps include blood glucose monitors and remote electrocardiogram (EKG) monitoring.


Gap between the HIPAA Regulation and the Current Landscape of Mobile Apps

The Health Insurance Portability and Accountability Act (HIPAA), which was enacted in 1996, ensures individuals’ health data privacy. HIPAA was written nearly 20 years ago, at a time when the health apps market did not exist. Despite some changes made in 2009 and 2013, it is questionable whether HIPAA provides adequate protection of privacy.

HIPAA privacy rules govern only the handling of protected health information (PHI) by “covered entities.” Therefore, HIPAA protects individually identifiable health information held by entities such as health plans, hospitals, and physician groups. Meanwhile, de-identified health information, which neither identifies nor provides a reasonable basis to identify an individual, is not subject under the HIPAA restriction. In 2009, Congress passed the Health Information Technology for Economic and Clinical Health Act (HITECH Act) to expand the scope of HIPAA to business associates of covered entities.  In 2013, the Department of Health and Human Services (HHS) issued the Omnibus Rule, which expanded the definition of a business associate to include any person who “creates, receives, maintains, or transmits PHI on behalf of a covered entity.” Some of the mobile app developers are classified as business associates and thus required to be compliant with HIPAA. For example, an app that helps patients make appointments with physician groups would be business associates of the covered entity. Also, some apps that store PHI are required to follow HIPAA privacy rules.

Nevertheless, many health mobile apps are used only by individual consumers, and thus there is no HIPAA implication for their developers. Moreover, many of the apps encourage users to disclose certain information. For example, many activity-tracking apps allow users to upload their activity levels and compare them to those of their friends. The rationale is to provide healthy competition and motivation to live healthier lifestyles.

Such disclosure of non-PHI health data, however, may not be as harmless as it appears. In the era of Big Data, health care companies have found value in analyzing data about their patients and potential patients. Data brokers may compile data collected from health mobile apps and draw powerful and sensitive inferences regarding individuals’ heath statuses. For example, health care companies may be able to analyze the probability of an individual having a heart attack by evaluating data on food intake and daily activity levels. Health care companies may also flag users who look up particular symptoms and conditions on apps that provide health information.


Possible Solutions

One solution to this potential infringement on personal privacy is to let the market take over. Among 165,000 health apps, 12 percent of apps accounted for 90 percent of all downloads, according to the IMS Institute’s estimation. In the fiercely competitive market, apps that do not provide adequate privacy protection may not be able to survive. Recent data breach incidents from other industries have raised consumers’ awareness about data security issues. In fact, Apple’s HealthKit, a healthcare platform for many popular apps, employs a strict privacy policy and bars developers from sharing data with third parties.

Another possible solution is to expand HIPAA under the auspices of the HHS. Currently, HIPAA coverage focuses on whether one is a covered entity or not, rather than on the content of data. Instead of limiting the HIPAA governance to covered entities and their business associates regarding PHI, Congress may expand its scope to cover health data itself. However, this solution has limits. At what point does data become “health data”? Most people probably agree that data such as blood pressure are likely to be health data. But how about activity levels and sleeping patterns? It is true that data analysts may be able to use such data to draw powerful inferences on individuals’ health statuses, but one could argue that seemingly non-health data, such as marriage status and work history, do the same.

Another solution is enforcement by the Federal Trade Commission (FTC). The FTC has regulated data security practices across industries. Recently, the FTC brought a data security enforcement case against LabMD, a HIPAA covered entity. One of the issues in the case was whether the FTC could act in a situation where the HHS has been the primary enforcement authority under HIPAA.  The FTC took a broad view of its authority and argued that HIPAA does not present a bar to undertaking enforcement under the FTC’s own standards.

Yet another solution is state enforcement. In 2013, Illinois Attorney General Lisa Madigan called on administrators of popular health-related websites to address concerns about their collection of health information online. The Attorney General asked that the health-related websites disclose how much the sites capture users’ health information and whether third parties have access to such data.



In this era of Big Data revolution and the proliferation of health mobile apps, privacy and data security will be important issues for the health care industry. We would expect that Congress, HHS, the FTC, state attorney general, and others would bridge the gap between HIPAA and the ongoing challenges provided by health mobile apps.

DOJ Antitrust Investigation: Is it Time for Airline Discipline?

Thursday, October 8th, 2015

Airline industry ticket pricing is a fun cat-and-mouse, prisoner’s dilemma-like game. That is, until the Department of Justice decides to launch a major antitrust investigation.

Numerous characteristics of the airline industry, as discussed below, render it especially susceptible to a basic four-box Cournot oligopoly problem. When all airlines restrict the number of seats flown (capacity) under profit-maximizing levels in a competitive market, they all earn more profits through higher prices than if they operated at the competitive market level. However, if some airlines break discipline by adding capacity, they earn more profits at the expense of those who are still restricting capacity. Importantly, unlike in the prisoner’s dilemma, the consequences when a competitor breaks discipline are minimal because other airlines immediately respond by adding capacity back to the competitive market level. Thus, deciding independently to restrict capacity is a low-risk move with potentially high rewards. In other words—it’s an easy game, so don’t cheat.

Under the Sherman Act, any agreement between corporations to restrain supply, thereby inflating prices, is considered cheating. The DOJ launched an investigation in July after multiple airline executives made comments at an airline trade show that could be interpreted as signaling capacity restrictions to each other. According to Bloomberg, Delta President Ed Bastian said that the airline is “continuing with the discipline that the marketplace is expecting.” When discussing the importance of avoiding so-called overcapacity, American Airlines CEO Doug Parker said, “I think everybody in the industry understands that.” Soon after the DOJ opened its investigation, numerous plaintiffs have filed private class action lawsuits based on the same price-fixing claims with Quinn Emanuel LLP representing one group.

Although the DOJ need not prove an explicit, written agreement, there must be significant evidence of an implicit agreement. Neither independent capacity decisions nor being in an oligopolistic market is a violation of the Sherman Act. Conscious parallelism occurs when competitors monitor each other’s behavior and choose to mirror it, even when such activity has the effect of raising prices.

The airlines have three potential defenses. First, the airlines can argue that they were merely talking about their own strategies and had no intention of signaling each other. Second, the airlines may argue that they have a duty to provide guidance to investors even if it borders on signaling strategy. Finally, the airlines will likely argue that because they are competing vigorously with each other, accusations of collusion are unfounded.

The airlines’ first argument is undercut by the nature of the market. Discipline only works, as former Department of Justice Antitrust Attorney Robert Connolly explains, if other airlines also maintain capacity discipline. Adding supply lowers prices across an entire industry selling such a similar service. In that light, the airlines also didn’t help themselves by mentioning the “marketplace” and the “industry” in their discipline comments rather than sticking to their corporation.

Second, Wall Street analysts have criticized the investigation, arguing it threatens their ability to get guidance from corporations. While the tension between guidance and signaling is noted in some litigation, that the comments came at a trade show with competitors present rather than on an earnings call constitutes, at the least, bad optics. When comments are ambiguous, context matters, and the airlines could have picked a context more consistent with their claim that they were only providing guidance. Moreover, because Wall Street has been calling for capacity control as a way to increase profits, it is unclear whether the underlying motivation of the criticism actually emanates from a genuine concern about guidance or a concern that the investigation will cause airlines to add capacity, which theoretically reduces profits.

Since those criticisms, the nexus to Wall Street has gotten more serious. Bloomberg has reported that the DOJ investigation is looking at whether airlines have been communicating strategy with each other through major common shareholders. One study has concluded that airfares are up to 11 percent higher for airlines that are commonly owned by the same major stockholders.

Finally, airlines argue that they are competing vigorously and consumers are doing well because of it. They point to complex airline industry metrics such as revenue per seat mile and other real-world examples like the battle for Seattle between Delta and Alaska Airlines.

The Seattle story is complex, however. First, Alaska is small but provides an important counterbalance to the big four airlines as one of only two alternatives that has a significant network and is not an ultra-low-cost carrier. Thus, aggressively attacking its primary hub is not exactly the competition the industry needs. Some even suggest Delta’s behavior may be predatory, another accusation major airlines have faced historically. Moreover, Delta acknowledges that its decision to build a hub in Seattle was motivated by Alaska’s refusal to end its partnerships with foreign airlines and exclusively feed Delta international flights out of Seattle. Delta has recently expressed disdain for foreign airline competition. Part of the reason Alaska may have wanted to diversify its partnerships is that it has been considered a long-term acquisition target for Delta. That move would further consolidate the industry. Thus, what looks like terrific competition is infected with strains of anticompetitive behavior.

The airlines’ arguments may not be enough to fend off a DOJ investigation driven by high structural antitrust susceptibility and current conditions that have escalated antitrust concerns.

Structurally, a significant inelastic air travel market among business travelers and high costs of new competitive entrants raise the risk of oligopolistic pricing. Critically, the prices consumers are willing to pay for air travel diverges substantially, with an elastic leisure market and an inelastic business market that needs to travel and has no available alternative (virtually no high-speed rail, for example). That being said, all consumers are extremely price-sensitive when picking between two competing airlines as each offers the same basic service. Together, these factors explain the basic and historical antitrust problems in the industry.

However, the executives’ remarks also came at a particularly sensitive moment in the industry. As Connolly explains, four major mergers that have led to 80% of domestic airline seats being flown on four airlines, plummeting fuel costs without reciprocal fare decreases, and cramped and sold-out flights all create an especially high antitrust risk environment. Moreover, with fewer airlines each running an immense operation, there is much less incentive to expand market share or create a more comprehensive route network for customers, two of the previous driving forces behind maintaining capacity. Finally, the airline industry has entered a new era of profitability that invites renewed scrutiny.

In this case, it is likely that the remarks can be explained as an overzealous attempt to tell Wall Street what it wants to hear rather than strategic signaling. Thus, ultimately, whether the DOJ investigation leads to a lawsuit depends largely on whether the airlines have been as undisciplined with their private statements as they have been in public.

While the investigation is ongoing, a new generation of airline executives should position themselves for the potential upcoming public relations battle by recognizing the public importance of air travel. Air travel shares many characteristics with public utilities given its pervasive impact on the economy and our quality of life. In fact, the executives should remind themselves that, for many years, airlines were regulated as public utilities. If airlines don’t want to be regulated like utilities again, including circuitously through antitrust law, they will have to do the one thing that differentiates them—compete.

Evaluating Commissioner Gallagher’s Dissent of the Rules Promulgated Under § 953(b) of the Dodd-Frank Act

Thursday, October 8th, 2015

On August 5, 2015, the Securities and Exchange Commission voted to adopt a set of rules that require public companies to disclose the ratio of CEO compensation to that of the median compensation of all employees. These rules were promulgated under order of Congress, which mandated the SEC issue them under § 953(b) of the Dodd-Frank Wall Street Reform and Consumer Protection Act (“Dodd-Frank”). Nevertheless, the final vote authorizing the rules was 3-2, with the Republican-appointed Commissioners Daniel Gallagher and Michael Piwowar dissenting.


Not only did Commissioner Gallagher appear skeptical that the median pay ratio rules serve any legitimate purpose, he also had a more targeted criticism of the finished product: namely, that the rules too expansively define “employee,” as they do not allow corporations to exclude overseas workers beyond a threshold, certain seasonal employees, or part-time employees. In light of these perceived deficiencies, Commissioner Gallagher felt that the rules serve only to “name and shame registrants into reducing CEO pay,” and he “question[ed]…whether [these] pay ratio rules are constitutional.”


Commissioner Gallagher’s concerns arise out of the compelled nature of the speech that § 953(b) demands; such concerns are not misplaced, as at least two separate provisions of Dodd-Frank have been targeted under this strategy. As a general proposition, the First Amendment protects both the “right to speak freely and the right to refrain from speaking.” However, commercial speech is typically measured against a lower standard than purely private speech. That standard was articulated in Zauderer v. Office of Disciplinary Counsel, where the Supreme Court held that the disclosure of “purely factual and uncontroversial information…reasonably related to the State’s interest in preventing the deception of consumers” is constitutional given that such disclosures are not “unduly burdensome.” Zauderer itself was markedly ambiguous as to what the outer boundaries of the government’s power are—which is to say, whether the government can compel speech for reasons other than to correct specific deception. In 2014, the D.C. Court of Appeals clarified (en banc) in the affirmative, holding that the Zauderer test can be invoked for government interests above and beyond correcting deception. The precise interests were left unnamed, but such a holding potentially protects the median CEO pay ratio rules from undergoing more searching scrutiny, as public companies do not often explicitly deceive their consumers about executive pay.


Thus, any challenge to the rules has three potential avenues for success: (1) contending that the CEO pay rule ratio formulae are not purely factual and uncontroversial; (2) contending that the rules do not reasonably relate to the sort of State interest indicated in Zauderer; or (3) contending that collecting the information necessary for the disclosures is unduly burdensome. Commissioner Gallagher unequivocally suggested that the CEO pay disclosure rules have defects that would fall into avenue (2) by terming them “name-and-shame,” and avenue (3) by observing that industry groups expect the cost of compliance to be approximately $1.3 billion a year. However, the overall dissent was clear. The Commissioner “could potentially support” a CEO pay disclosure set of rules, but only if any such proposal grants managers sufficient discretion to determine which positions qualify for the median employee compensation statistics. In other words, Commissioner Gallagher was principally concerned that the content of the CEO pay rule disclosures fails to be purely factual and uncontroversial.


If a Court were to find that these rules require a disclosure beyond mere fact and non-controversy, than the subsequent constitutional review would be conducted under a more stringent standard than the one articulated in Zauderer. However, such a finding is unlikely. While the Supreme Court has yet to offer precise color to the clause “purely factual and uncontroversial,” at a minimum it is clear that there are two somewhat separate criteria that must be met to fulfill this standard. The first is “definitional”—the compelled speech must possess or recognize some feature of a product or service, and not “advocate” for a political purpose. Courts have found that product specifications, legal classifications, and the consequences of the product’s use (including its true costs and externalities) are “factual” in nature, while graphics or symbols that are manipulated to convey a particular message are not. The second criteria is more nebulous, but in the words of one commentator, the compelled fact should not be “opinion[ated] or ideolog[ical],” which is to say that it should not infringe upon an individual’s “self-determination, autonomy or freedom of mind.”


The CEO pay rules clearly meet this standard. As a threshold matter, the rules are plainly factual: a ratio derived by comparing the values of compensation received by an officer and a median worker is an indisputable numerical truth. Moreover, there is nothing ideological about this number. Only when intermingled with an individual’s pre-existing values or politics does this ratio take on an ideological tone. The rules as currently authored convey valuable information to shareholders, who can use it to benchmark management compensation across firms. Granted, Commissioner Gallagher’s contention that the rules do not allow for sufficient managerial discretion dampen the credibility of the reported ratio. That does not make the information conveyed by the ratio non-factual. Undoubtedly the Commission considered the possibility that allowing too much variability would lead to manipulation, thereby watering down the value of the reported ratio. However, insofar as insufficient discretion is an issue, nothing prevents companies from providing the legally required ratio and subsequently offering context or a more nuanced alternative ratio in a footnote or additional disclosures. Subsequent disclosures would only empower shareholders to make informed decisions about management and labor compensation.


Commissioner Gallagher’s August 2015 dissent identified some serious potential deficiencies with the CEO pay disclosures promulgated by the SEC. If his chief complaint—that the rules do not provide sufficient individuality so as to rise to the level of unconstitutionality—is brought before a judge, it is unlikely to prevail.

“Sweeping” the GSEs’ Profits to the Treasury is Legal, but Problematic

Tuesday, October 6th, 2015

The slow return of the housing market and its two backbone financial institutions, Fannie Mae and Freddie Mac, has created not only huge profits for the U.S. government, but also a flood of lawsuits claiming unconstitutional government takings of privately owned shareholder profits. It has been seven years since Fannie Mae and Freddie Mac were placed in government conservatorship, and much has changed for the government-sponsored enterprises’ (“GSEs”) balance sheets. Back in the fall of 2008, they had a combined $5.3 trillion in outstanding debt, requiring $187.4 billion of capital injection from the United States Treasury to salvage the companies, and in effect, the U.S. housing market. Since then, the GSEs have regained profitability and paid back the Treasury $230.8 billion to date, $43.4 billion more than they received from the Treasury.

However, due to an amendment to the original conservatorship agreement between the Treasury and the Federal Housing Finance Agency (“FHFA”), the conservator of the GSEs, these payments are not considered a repayment of principal debt owed to the Treasury, but rather dividend payments. This amendment, known as the “the net-worth sweep,” allows the Treasury to “sweep” nearly all of the net income of the GSEs on a quarterly basis into the government’s coffers. With the GSEs’ profits in the tens of billions ($84 billion for Fannie and $49 billion for Freddie in 2013), this arrangement certainly works to alleviate the government’s crushing deficit.

The deal, however, is not without its losers. Large private investors of the GSEs bet big during the financial downturn on the return of the GSEs’ profitability. But as the mortgage market bounces back, their expected returns have been “swept” to the Treasury, leaving them with effectively worthless shares in the GSEs. The private investors have responded by filing a volley of lawsuits directed at the Treasury, arguing that the “net-worth sweep” constituted a violation of the Fifth Amendment’s “takings clause,” which protects against seizure of “private property for public use without just compensation.” So far, the claims have been unsuccessful. Nevertheless, a settlement with the investors and a reform in the “net-worth sweep” agreement may be prudent next moves for the government.

Leading-Up to the Sweep

The FHFA became conservator of the GSEs in 2008 under an agreement in which the Treasury would provide up to $200 billion in bailout funds in exchange for preferred stock valued at an equivalent amount, warrants to purchase as much as 79.9% of the common stock, and a mandatory 10% dividend. Over the next four years, the Treasury infused $187.5 billion into the GSEs, but the mortgage entities still found themselves cash-strapped and unable to meet their obligations, due in part to the mandatory dividend payments to the Treasury.

To remedy the situation, the FHFA and Treasury amended their agreement. As amended, the Treasury retains its preferred shares in the companies. However, instead of collecting the 10% dividend payments, the Treasury “sweeps” on a quarterly basis nearly all of the net income of the GSEs into a slush fund for general government expenses, including cutting the government’s deficit. The GSEs’ private sector preferred shareholders – including institutional investors, insurance companies and pension funds – are left with little, both in terms of dividends and the value of their equity.

Legality of the Sweep

Many of these exasperated shareholders are now suing the U.S. government for taking their property for public use without just compensation, in alleged violation of the Fifth Amendment. In a recent panel discussion at Columbia University, Bill Ackman, the largest private shareholder of the GSEs, called this “net-income sweep” arrangement “the most illegal act of scale” he has ever seen the U.S. government do. As he sees it, “If the U.S. government can step in and take 100% profits of a corporation forever, then we are in a Stalinist state and no private property is safe….”

So far, the courts have disagreed with Mr. Ackman.  Last fall, Judge Royce Lamberth of the United States District Court for the District of Columbia dismissed without argument two of the institutional investor lawsuits, holding the “net-income sweep” is within the wide-reaching authority Congress granted the Treasury and FHFA under the Housing and Economic  Recovery Act (HERA). Furthermore, HERA states, “Except as provided in this section or at the request of the Director, no court may take any action to restrain or affect the exercise of powers or functions of the Agency as a conservator or a receiver.” Lastly, Judge Lamberth held that private shareholders failed to plead a cognizable property interest in the dividends or liquidation preferences of their preferred shares in the GSEs. According to the court, the shareholders purchased the stock certificates with the “background principle” that, under HERA, the GSEs are “subject to regulatory oversight, including the specter of conservatorship or receivership under which the regulatory agency succeeds to ‘all rights’ of the GSEs and shareholders.” Therefore, since the GSE shareholders necessarily lack the right to exclude the government from their investment when the FHFA places the GSEs under conservatorship, and since the “right to exclude” is “one of the most essential sticks in the bundle of rights that are commonly characterized as property,” the private GSE shareholders had no cognizable property interest in the dividends of the stock certificates that the government took preference over. In essence, these investors should have built into their investment expectations the realization that the government has the authority to supersede their dividend rights. Therefore, it is not an unconstitutional taking for the government to exercise that authority.

This past February, the U.S. District Court for Southern Iowa followed Judge Lamberth’s lead, dismissing a similar case brought by another GSE shareholder. The precedent created by these two federal court holdings is a discouraging start for the GSE shareholders in what is sure to be an uphill legal battle between them and the Treasury. For now, the “net-worth sweep” seems, legally speaking, safe.

Implications of the Sweep

With that being said, aside from the legality of the net-worth sweep, there are legitimate reasons the Treasury should wean itself off of the huge profits the GSEs provide and allow private money to regain a more classical sense of ownership in the mortgage giants. For one, the profit sweep greatly deters any prospective investors to invest in a GSE or distressed financial institution under conservatorship. It is hard to imagine private-sector investors feeling comfortable partnering with government entities when courts say that the government is able to “sweep” all profits indefinitely. Secondly, due to the expropriation of profits, the GSEs are currently unable to build any cash reserves from their sizeable profits to hold against their trillions in outstanding liabilities. Thus, if and when the mortgage securities markets take another turn for the worse, American taxpayers will immediately be on the hook to bear the burden of keeping the mortgage market afloat. In order for the private market to share in the risk of a downturn, they must be allowed to share in the profits of an upswing as well. In sum, while the Treasury is likely feeling comfortable with the legality of their “net-worth” sweeps, perhaps they should not feel as comfortable with the prudential implications of a continued profit-sweeping arrangement.