The report “Success by Default”
focuses on default rates among various types of complainants and respondents.
The default rate was 52% for the study period and seems to have grown since.
Default per se is not necessarily a bad thing. The report argues that UDRP has
been an effective remedy for cybersquatting primarily because it makes it economically
inefficient for abusive registrants to defend their names. Known cybersquatters
default (i.e., fail to defend the name) 70 – 100% of the time. For example,
the notorious typosquatter John Zuccarini was challenged under the UDRP 61 times
and defaulted in 44 (72%) of the cases.
Unfortunately, many seemingly good faith registrants default, too, and some examples are provided. The degree to which Respondents are able to defend themselves
is the single most significant factor in determining the outcome of UDRP cases, with defaulters losing 96% of the time (99% at NAF). The report proposes modifying the UDRP to require Complainants to post a $1,000 bond in addition to the costs of filing a complaint. The refundable bond would
discourage merit-less claims and help legitimate Respondents to participate
in the proceeding rather than defaulting. If the Complainant wins the case the
money would be refunded; if the complaint fails, the money is given to the Respondent
to defray the costs of a defense. If the Respondent defaults, the bond is refunded
and the procedure should be truncated so that Complainants’ arbitration costs
can be further reduced.
The report’s authors were particularly interested in cases that involve free
expression issues. The report identified 85 cases that involve domain names
that were either used for or reflect criticism and commentary. It found that
these cases are the least likely to involve defaults (only 7%, compared to 52%
overall), and that the decisions are inconsistent but often hostile to free
speech. One key problem revolves around the standard for a finding of “confusing
similarity” to a trademark. Some panelists have shown deference to free expression
values while others have concocted bizarre rationales for findings of “confusing
similarity” and “bad faith”
The study calls for resolving this inconsistency in a liberal way. Domain names
that signal criticism, parody or commentary upon products and companies (e.g., icannwatch.org) should not be classified as “confusing” unless they
are used in ways that actively promote fraud, deception or confusion. Precedents
that stretch notions of confusing similarity to include any incorporation
of a trademark in a domain name should be repudiated and the policy modified
to prevent such findings.
The report also shows how UDRP panelists (and the parties themselves) rely
heavily on “precedent.” A quantitative analysis of precedents cited shows that
of the top 20 cases UDRP panelists cite most often, all were won by Complainants.
Indeed, 16 of the 20 were Respondent defaults.
The report indicates that the decision to make the list of “bad faith” factors
“without limitation” gives panelists too much discretion and facilitates poor
or inconsistent decisions, because they can just make up bad faith factors.
The report calls for adding new factors to the list of bad faith items, but
limiting the factors to the specified ones.