2013 RAA: Summary of Economic Analysis Supporting Removal of
Bulk Whois Access Provision
During the Negotiation Process for the 2013 RAA, the registrars approached ICANN about removing the bulk access requirement from the RAA (see Section 3.3.6 of the 2009 RAA). The 2009 RAA states, at Section 3.7.7 , that:
Registrar's obligations under Subsection 3.3.6 shall remain in effect until the earlier of (a) replacement of this policy with a different ICANN policy, established according to Section 4, governing bulk access to the data subject to public access under Subsection 3.3.1, or (b) demonstration, to the satisfaction of ICANN, that no individual or entity is able to exercise market power with respect to registrations or with respect to registration data used for development of value-added products and services by third parties.
In support of the registrars’ request, they referred ICANN to a 23 January 2010 letter from Statton Hammock of Network Solutions to Kurt Pritz regarding RAA section 3.7.7 as well as a 6 May 2011 letter from Jeff Eckhaus of Demand Media to Kurt Pritz. Each of these letters asserted that the threshold set out in section 3.3.7(b) had been satisfied in today’s market conditions, and no individual or entity is able to exercise market power with respect to registrations or with respect to registration data used for development of value-added products and services by third parties. In July 2012, ICANN provided both of these letters to economist Joshua D. Wright, Professor, George Mason University School of Law and Department of Economics, to assess the registrar assertions, and also pointed him to the language at Section 3.7.7 of the RAA.
In early August 2012, Professor Wright, in a discussion with ICANN counsel, stated his opinion that no single registrar possesses market power on the “buy-side” (which would be seen in control of shelf space to extract concessions from registries, such as favorable placement or the exclusion of rival registrars) or on the “sell-side” (which would be demonstrated by an ability to raise the market price for registering domain names and providing other services). Professor Wright explained that among the differing market share percentages that support conclusions of market power across jurisdictions, no single registrar meets those percentage thresholds. Further, to support a finding of market power, those threshold market share percentages normally must be observed along with significant barriers to entry into the market and/or with evidence of a single firm being able to raise prices substantially above the competitive level.
Focusing on GoDaddy as ICANN’s largest accredited registrar, Professor Wright used publicly available data such as that available at webhosting.info to analyze overall market share as well as GoDaddy’s share of creating new registrations. He noted that GoDaddy’s overall market share of 32% of all total registrations has remained constant since 2009, and there is little evidence to suggest that in the near future it will rise to the 60% market share threshold that is generally required in the United States to presume market power. Further, evidence does not support that there are high barriers to entry in the registrar market; in fact, the number of small registrars shows the exact opposite. In terms of pricing, GoDaddy’s pricing appears to be competitive. Another possible indicia of market power could be in requiring longer-term contracts, however information available from webhosting.info supports that for new registrations, GoDaddy actually enters into more single-year registration agreements than the next three largest registrars (ENOM, Tucows and Network Solutions).
To demonstrate market power, the anticompetitive concerns of lock-in effects or high switching costs for consumers are also reviewed. The prevalence of one-year registration agreements for GoDaddy, however, supports that it does not exercise market power in forcing lock-in effects. In addition, the statistics show that transfers among registrars are frequent, which means that for consumers, the costs of switching between registrars are low. Analyzing statistics on transfers shows that GoDaddy indeed loses registrations through transfer, which helps support that there are not only low costs in switching from GoDaddy, but that competition against GoDaddy is vigorous.
Professor Wright also looked at GoDaddy’s share of registrations in .COM as compared to other registrars. While not suggesting that any single TLD represents a relevant antitrust market, Professor Wright noted that the percentage of GoDaddy registrations that are in .COM, as well as in other existing gTLDs, are similar to the percentages of .COM registrations for each of GoDaddy’s three largest competitors. This equal distribution of registrations among registrars suggests that GoDaddy lacks market power in any single TLD, as across all TLDs.
Based on Professor Wright’s observations, ICANN determined that the requirements of Section 3.7.7(b) of the 2009 RAA had been met, and that no individual or entity is able to exercise market power with respect to registrations or with respect to registration data used for development of value-added products and services by third parties. Based on the satisfaction of this requirement, ICANN agreed to relieve registrars that sign onto the 2013 RAA of the bulk Whois requirement, while maintaining language within the 2013 RAA that if such market power could be later demonstrated, the bulk access provision could be reinstated.