Longhorn is long shot for federal fame

Board of Regents v. KST Electric, Ltd., 2008 WL 577215 (W.D. Tex.)

This is a magistrate’s report and recommendation; it says some interesting things about federal trademark dilution. The University of Texas sued KST under state and federal trademark law, alleging infringement and dilution of its registered trademark – a longhorn steer in silhouette.

Among other things, KST sought summary judgment on federal trademark dilution, arguing that UT hadn’t shown the requisite fame. KST’s expert conducted a national survey that, he said, showed that only 5.8% of respondents associated the longhorn logo with UT alone, and only 21.1% of Texas respondents did so. Because the current standard requires a mark to be “widely recognized by the general consuming public of the United States,” 15 U.S.C. § 1125(c)(2)(A), the court only briefly mentioned a UT study conducted solely in the metropolitan Austin area. Rather, UT attacked KST’s survey and submitted circumstantial evidence of the logo’s fame based on the idea that UT “has permeated the national consciousness primarily through the success of its football program.”

KST’s survey tested the UT longhorn logo in white with a white background, rather than in its usual burnt orange color (the logo itself is “almost invariably” either burnt orange or displayed against a burnt orange background). The control group was shown the longhorn from the Longhorn World Championship Rodeo logo, to control for guessing or for the possibility that any depiction of a longhorn steer will be associated with UT.

Slightly less than half of the total respondents were from Texas; almost 22% of Texas respondents shown the control logo associated it with UT alone, and 26.5% associated it with UT and another “company, organization, or place.”Among non-Texas respondents, 5.3% in the control associated the control logo with UT alone, and 6.1% with UT and some other entity.

UT claimed the logo presentation was biased because of the absence of burnt orange. The expert claimed that UT’s mark is used in a wide variety of contexts, so picking any one would bias the results. The magistrate found this response disingenuous. The mark should be shown in its typical manner, rather than an unlikely manner.

The expert’s coding was also suspect. He didn’t count as associations favoring UT responses such as “Texas football” or “the Texas Longhorns” because they didn’t include reference to an academic institution. He also admitted to numerous coding errors, almost invariably in KST’s favor. A subject who associated the UT logo with “Texas college,” and gave the reason for doing so “That’s the image on helmets” was coded as not making a UT association.

UT also argued that the question was leading because it asked for associations with any “company, organization, or place,” which could lead respondents to simply say “Texas,” which would then be coded against UT.

Though UT didn’t raise the issue, the magistrate was troubled by KST’s sample size, 454 people, which “seems small for a national sample,” and “would not pass muster for a reputable national poll (like a Gallup or Zogby). Gallup and other major polling and survey organizations typically have a sample size of 1,000-1,500 individuals for national polls” (citation omitted). Comment: Ulp! The magistrate seems to be unfamiliar with trademark/advertising litigation surveys, as to which 454 is a pretty decent sample – and a pretty expensive one. Moreover, the sample size will affect the confidence interval, but not the best estimate of actual fame. For a clear explanation of this, as well as discussion of the case law on sample size, see Jacob Jacoby, Amy H. Handlin, & Alex Simonson, Survey Evidence in Deceptive Advertising Cases Under the Lanham Act: An Historical Review of Comments From the Bench, 84 Trademark Reporter 541 (1994).

In any event, the magistrate found that KST’s survey didn’t show absence of fame. UT’s affirmative evidence of fame, however, was insufficient to avoid summary judgment, so KST still prevailed on the federal dilution claims. UT’s evidence appeared impressive “at first blush,” but upon examination was evidence of niche fame in sports, clearly excluded by the TDRA. UT games are regularly televised nationwide and watched by millions of viewers; Sports Illustrated regularly features UT and the logo; UT has many official corporate sponsorships for its teams; UT has one of the nation’s most successful collegiate licensing programs, just behind first-place Notre Dame.

But UT’s evidence didn’t show that someone who was not a college football fan (or perhaps, “to a lesser extent,” college baseball or basketball fan) would recognize the longhorn logo as associated with UT. The magistrate was “well aware” of the popularity of college football, and noted that “the Court counts itself as a more than casual fan of Saturday afternoon football in the Fall.” But it was unwilling to make the leap to the “general consuming public,” a group it defined as “nearly the entire population of the United States.” UT athletics have “a level of national prominence,” but the longhorn still doesn’t “stand toe-to-toe with Buick or KODAK.” (Now there’s a mental image for you; a little like the image at issue in Deere v. MTD, actually.)

Quoting Barton Beebe, the magistrate concluded that the “‘TDRA is simply not intended to protect trademarks whose fame is at all in doubt,’” and recommended summary judgment for KST on the federal dilution claim.