Comments of the
Internet Committee of
the International Trademark Association
on the June 5, 2009 Report of Dennis Carlton
Regarding ICANN’s Proposed Mechanism for Introducing New gTLDs
The ICANN Board, in an October 18, 2006, Board resolution, directed the commission of an independent study “to deliver findings on economic questions relating to the domain registration market.” The Board provided examples of five specific questions that could be answered in the study, namely:
whether the domain registration market is one market or whether each TLD functions as a separate market;
whether registrations in different TLDs are substitutable;
what are the effects on consumer and pricing behavior of the switching costs involved in moving from one TLD to another;
what is the effect of the market structure and pricing on new TLD entrants; and
whether there are other markets with similar issues, and if so how are these issues addressed and by who?
The Internet Committee believes that an economic study of the domain registration market, such as the study outlined by the ICANN Board through its resolution, would provide needed empirical research and factual-based analysis that could inform future operational and policy matters concerning generic top-level domain names, including the development of registry agreements corresponding to those domain names.
While ICANN has been considering the introduction of new gTLDs since its inception, it never completed the comprehensive economic study of the domain name registration market and, as a result, did not develop a full understanding of the economic realities of the domain marketplace.
Instead, following its announcement of the Draft Applicant Guidebook (DAG) in October 2008, and the input of concerns from the community that ICANN’s process for introducing new gTLDs raises significant economic and consumer welfare concerns, ICANN solicited a report by Dennis Carlton on its proposed application process for introducing new gTLDs.
The June 5, 2009, Report of Dennis Carlton Regarding ICANN’s Proposed Mechanism for Introducing New gTLDs (hereinafter “Carlton Report”) does not specifically address many important economic issues, including the questions raised by the ICANN Board, a fact that is acknowledged in several places in the report. For example, the Carlton Report states that “while these issues are of economic interest, analysis of these questions is not necessary for evaluating ICANN’s gTLD proposal.” ( ¶¶ 34-35). Similarly, the Carlton Report acknowledges that the Board’s question as to whether the domain registration market is one market or whether each TLD operates as a separate market “remains an interesting question deserving of analysis” but, nevertheless finds that “evaluation of the impact of ICANN’s gTLD proposal on consumer welfare does not depend on the answers to this question.” (¶ 9).
The Internet Committee believes the Carton Report is based entirely upon select generalized economic theory and flawed assumptions that fail to consider the dynamic nature of the domain market. The Report reads more like an opinion and advocacy paper than anything else. In any case, a report is not the same as an independent economic study conducted in an unbiased fashion by an expert team not beholden to ICANN. An independent study would be based on empirical research into each of the questions posed, among others later determined to be relevant, and would require extensive consultations with a wide variety of different stakeholders in the marketplace. In contrast, the Carlton Report lacks any significant empirical data that might establish the market power, market forces or demand metrics that are critical to understanding economic theory in the context of the ICANN community. The report also appears to unjustly and willfully ignore and downplay the many commercial, government and consumer concerns raised about the introduction of new gTLDs. For example, the statement that the introduction of new gTLDs “raises no new issues” (¶ 55) is entirely unsupported and ignores the many serious concerns voiced in hundreds of pages of public comments submitted by numerous individuals, small businesses and the largest corporate entities in the world along with their trade associations, including the INTA Internet Committee. It is critical that all essential and appropriate regulatory/oversight policies be implemented before proceeding with major changes to the domain name space. Yet, the Carlton Report also fails to assess whether ICANN has the institutional capacity, resources, policies, and commitment required to address: 1) the extensive trademark infringement that experience to date has demonstrated will occur in new gTLDs; and 2) the Registry-Registrar cross ownership competitive abuses that are likely to occur in new gTLDs.
Specific Comments Regarding the Carlton Report:
The Carlton Report claims that a “generic” gTLD such as .CARS “may facilitate the ability of consumers to obtain both generic information about cars as well as the ability to access the websites of car manufacturers ...”. (¶ 33). It is already possible to obtain both “generic information about cars” and “access the websites of car manufacturers” using search engines. Moreover, it would still be necessary for consumers to type a second-level domain if they sought direct access to a second level domain under .cars, for example Ford.cars.
Similarly, the Carlton Report claims that the expansion of gTLDs “promotes consumer welfare by reducing search costs faced by internet users” (¶ 49). Yet, the report provides no evidence to substantiate the claim that more gTLDs, and the corresponding increase in second-level domains, would reduce search costs. In fact, introducing potentially thousands of new gTLDs into the domain name system may overwhelm Internet users forcing them to revert to search engines or other means to find their required online destination. Yet, the Carlton Report as an example, uses the FORD trademark to posit that searching for product information by FORD customers would be easier if the gTLD .FORD existed. (¶ 33). Today, customers searching for Ford can easily and intuitively receive product information about Ford cars by typing www.ford.com in fewer keystrokes that typing www.info.ford, for example. Not only is it already possible to search for information on Ford products using a variety of second-level domains and ccTLDs, but search engines already direct consumers to the relevant page within the ford.com website. Interestingly, this purported benefit is dependent on the introduction of “.brand” gTLDs, which, ICANN has still not specifically stated will be permitted as new gTLDs, much less shown that there is demand for such gTLDs that outweigh the concerns expressed by brand owners.
The Carlton Report contends that “new gTLDs would … likely promote future competition with .com and other TLDs.” (¶ 35). However, the report fails to substantiate why new gTLDs will succeed in creating meaningful competition with .com where other, more recently launched gTLDs have failed to do so -- a fact the Carlton Report itself acknowledges. (See ¶ 12).
With regard to issues of trademark protection, the Carlton Report takes the position -- without any substantiation or explanation -- that entry restrictions are likely to be an inefficient mechanism for protecting trademarks. (¶ 50). Professor Carlton is either unaware of trademark practices and procedures followed by most Trademark Offices throughout the world or is aware but believes that the relevant governments and officials are operating under mistaken assumptions of efficiency. Many Trademark Offices throughout the world impose entry restrictions in the form of statutory or regulatory prohibitions on granting registration to marks that are identical or likely to cause confusion with earlier registered marks. From a philosophical perspective, it seems far more likely that these restrictions embody the public policy goal of preventing consumer confusion than the economic goal of efficiency. However, these entry restrictions reflect a practical and fundamental economic consideration -- it is far more expensive to restrict trademark registrations post-entry. Private actions to oppose or cancel trademark registrations are far more costly and therefore inefficient than relying on the Trademark Office to prohibit confusing or misleading registrations in the first instance. As a result, consumer confusion is far more likely to occur when post-entry trademark registration restrictions are relied upon instead of pre-entry trademark registration restrictions. The same is true for domain names.
The reference to the UDRP as a “mechanism currently existing for protecting the use of trademarks in domain names” (¶ 51) fails to note that the UDRP is a post-entry protection mechanism. The report completely overlooks the fact that costs related to initiating a UDRP proceeding have become burdensome for trademark owners dealing with the widespread proliferation of trademark infringement in the current gTLD space.
The proposal to enhance the UDRP by requiring that “the party that loses a challenge ... bears the legal costs of both parties” (¶ 52) does not take into account the practical realities of the cross-border nature of the domain name system and a dispute resolution mechanism in which the parties are, more often than not, located in different jurisdictions. The expense to the winner of collecting its costs are likely to far outweigh the costs themselves. In cases in which the losing party is located in an offshore jurisdiction, or where the domain name registrant has used fictitious registration information, the ability to recovery costs is likely to be very limited. Similarly, the requirement to post a bond, as suggested in paragraph 52, at the time a dispute is filed or a domain is registered, would have the effect of giving parties who have the most money the upper hand and restrict entry on an economic basis rather than an IP rights basis.
“The expansions in the number of TLDs under ICANN’s proposal would appear to raise no new issues (emphasis added) relating to enforcing and monitoring trademarks”. (¶ 55) Although paragraph 55 appears to acknowledge that the introduction of new gTLDs will, in fact, increase the required effort and associated costs of enforcing and monitoring trademarks, the report indicates that “it is unclear how much the introduction of a new gTLD…would further increase the required effort and associated costs…” Numerous trademark owners have contended in public comments that a large gTLD expansion will result in a corresponding increase in effort and costs associated with enforcing and monitoring trademarks. For example, watching services must be expanded to cover more gTLDs, and additional attorney and non-attorney time and costs must be devoted to monitoring and taking action with respect to the increased numbers of infringing domain names. Now is the appropriate time for ICANN to conduct the study necessary to quantify specifically the increased resources that trademark owners will be forced to expend to combat increased cybersquatting and abuse associated with the introduction of new gTLDs.
Although the Carlton Report acknowledges that the impact of gTLDs on the need for defensive registrations merits close attention (¶ 56), it concludes erroneously that trademark owners’ concerns are overstated, but fails to substantiate this claims with any factual analysis or empirical data.
In numerous instances, the Carlton Report refers to the increased competition for new registrants in new gTLDs and the positive effect on the marketplace of such competition. For example, despite the lack of empirical data in paragraph 9, the report concludes that it is “likely” that consumers will “realize significant benefits from new gTLDs due to increased competition for new registrants and increased innovation that would likely be fostered by entry.” However, from a trademark owner’s perspective, this competition has the potential to increase the frequency and scope of infringement. As owners of new gTLDs are desperate to attract new registrants in such a crowded environment, it is likely that such owners will resort to tactics that do not adequately take into account the rights of trademark owners and other third parties. Carlton assumes that introduction of new gTLDs will foster competition for domain registrants thus leading to unspecified benefits to consumers (presumably price and services benefits), but the report fails to consider the negative implications of such a “race to bottom” for new domain registrants.
The report discusses price capping and determines that “the ability of ICANN to protect trademark holders through existing dispute resolution mechanisms and appropriately-designed modifications of ICANN procedures for protecting intellectual property implies that price caps are not necessary to protect trademark holders.” (¶ 11) Setting aside the assumptions about the adequacy of the existing dispute resolution mechanisms discussed above, lack of price capping may exacerbate infringement because it will encourage owners of new gTLDs to sell their domain names to the highest bidder if conflicts arise at the time of registration. Moreover, anecdotal evidence suggests that registrars today charge excessive prices for domain names during the sunrise registration periods. They understand that the desire to protect trademarks during this period creates an artificial landrush bearing no relation to the need to use that particular name in a gTLD. This landrush allows registrars to charge whatever price they believe the market will bear. An independent empirical study would confirm that many trademark owners already pay excessive costs for sunrise registrations. A study could show that many domain names registered during the sunrise period do not resolve to live websites but remain as passive assets in large domain name portfolios. The report’s assertion that price-caps are inappropriate is inconsistent with the aforementioned argument that the introduction of new gTLDs will lead to competition for domain registrants and thus pricing benefits to consumers.
As we state throughout our comments, which are not exhaustive, the Carlton Report ignores the cautionary note in the ICANN Board resolution about the complexity of the domain registration market and the high levels of economic expertise required to produce reliable analysis and findings. As a result, the Report’s analysis does not reflect the dynamics of the domain name market. Further, the Carlton Report is unresponsive to demands for a true economic market study of the domain market and fails to provide the necessary empirical data essential to support reasoned decision making or understanding of the issues created by the intended introduction of new gTLDs. The INTA Internet Committee urges ICANN to conduct a full empirical study of the domain market and obtain community input on that study before finalizing any plans for releasing new gTLDs.
Thank you for considering our views on these important issues. If you have any questions regarding our submission, please contact External Relations Manager, Claudio DiGangi, at: firstname.lastname@example.org
About INTA and its Internet Committee
The International Trademark Association (INTA) is a 131 year-old not-for-profit membership association of more than 5,500 trademark owners, from more than 190 countries, dedicated to the support and advancement of trademarks and related intellectual property as elements of fair and effective national and international commerce. Over the last decade, INTA has been a leading voice for trademark owners on the future of the Internet DNS, and it is a founding member of ICANN's Intellectual Property Constituency (IPC). INTA’s Internet Committee consists of over 125 trademark professionals who evaluate treaties, laws, regulations and procedures relating to domain name assignment, use of trademarks on the Internet, and unfair competition on the Internet and develop and advocate policies to advance the balanced protection of trademarks on the Internet.