Appeals Court Finds Agreements Restricting Search Engine Keyword Bidding Not Anti-Competitive
In 1-800 Contacts Inc v FTC 18-3848 the Court of Appeals for the Second Circuit (CA2C) has reversed the controversial majority decision of the Federal Trade Commission (FTC), which held that agreements that prevent competitors from bidding for their competition’s trademarked terms as search engine ad generating keywords are anti-competitive.
As noted in the earlier reporting of the FTC’s decision, 1-800 Contacts Inc, which is the largest online retailer of contact lenses in the USA, actively sought to protect its trademarks by entering into agreements with rival online contact lens sellers who had used 1-800 Contacts’ trademarks as search engine ad generating keywords. The agreements prohibit the parties thereto from using the other party’s trademarks as search engine keywords and also require each party to designate the other party’s trademarks as negative search engine keywords – which prevents the search engine from generating an ad even though the advertiser did not select the keyword. However, most of the agreements allow the use of the other party’s trademarks for non-infringing uses, such as comparative advertising and parodies.
The FTC majority gave preference to open competition in online search bidding and held that the agreements constitute unfair methods of competition on the basis that consumers would be deprived of ads informing them that identical products are available at lower prices. The higher prices 1-800 Contacts charged compared to other online competitors was considered to evidence the anticompetitive effect of the agreements. It was also held that by harming competition in bidding for search engine keywords 1-800 Contacts benefited by paying artificially reduced prices for its ads.
The FTC majority held that trademark’s pro-competitive functions of reducing the likelihood of consumer confusion, assuring consumers of consistent quality and reducing consumer costs of making purchasing decisions are outweighed by the benefits from consumers having ready access to market information. However, as noted in the dissenting judgment people who search with a trademarked term most likely do so for navigational rather than cost comparison reasons and, by allowing for comparative advertising, the agreements achieve their goal of preventing trademark infringement while balancing the need to permit non-infringing advertising.
On appeal the CA2C noted that trademark settlement agreements are not automatically immune from antitrust scrutiny and that what must be determined is whether the restraints in the agreements are reasonable in light of their actual effects on the market and their pro-competitive justifications. It was also noted that the level of proof required is dependent on where the alleged anticompetitive agreement falls between agreements that are per se illegal and those that have strong pro-competitive justifications. If the likelihood of anticompetitive effects can easily be ascertained, courts apply an abbreviated rule of reason analysis known as the “inherently suspect” framework and only minimal evidence or analysis is required before the burden shifts to the defendant to provide procompetitive justifications for the agreement. In contrast, if the likelihood of anticompetitive effects cannot easily be ascertained, then a full rule of reason analysis is required starting with establishing a prima facie case of anticompetitive conduct before the burden shifts to the defendant. If procompetitive justifications for the agreement can be provided, then the burden shifts back to the plaintiffs to prove that any legitimate competitive benefits offered by defendants could have been achieved through less restrictive means.
The CA2C held that the FTC majority was wrong to apply the inherently suspect framework to the agreements as agreements to protect trademarks should not be presumed to be anticompetitive, and their balancing of the need to permit non-infringing advertising meant they were not obviously anticompetitive.
The CA2C found the evidence for difference in prices to competitor’s products to be theoretical and anecdotal and not sufficiently direct to show that prices would have been different in the absence of the agreements. While there was evidence showing that prices for certain keywords dropped, that was found not to be direct evidence of the effect on the market as a whole.
The CA2C held that the FTC majority gave insufficient weight to the pro-competitive functions of trademarks and found the agreements to simply be the product of negotiations regarding trademark rights, and that those provisions were not auxiliary to an underlying illegal agreement.
The FTC argued that the agreements could have instead had a less restrictive provision requiring the generated ads to clearly show the identity of the rival seller. However, the CA2C held that courts should give substantial weight and deference to agreements determined between parties. The CA2C considered forcing companies to be less aggressive in enforcing their trademarks is antithetical to the procompetitive goals of trademark policy. The FTC majority was also found to have not considered the downstream effects of requiring less aggressive enforcement, and so had not established that the same procompetitive effects would be achieved thereby.
Back here in New Zealand the Commerce Commission has advised it has just filed proceedings against consumer loan provider Moola, alleging that it has engaged in cartel conduct by entering into agreements with other consumer loan providers regarding online advertising with Google Ads. As with the 1-800 Contacts agreements, the agreements prevent the parties from bidding on each other’s brand names and require the use of negative keywords. It is not yet clear whether the agreements allow for comparative advertising, although presumably they will as that is provided for under New Zealand’s Trade Marks Act 2002.
Author: Quinn Miller