Fair Isaac Corp. v. Experian Information Solutions, Inc., --- F.3d ----, 2011 WL 3586429 (8th Cir.)
The court of appeals affirmed the district court’s grant of summary judgment against Fair Isaac (FICO)’s antitrust and false advertising claims and its ruling that FICO’s trademark was merely descriptive, as well as the jury verdict that FICO obtained its trademark registration through fraud on the PTO. It also affirmed the denial of defendants’ attorneys’ fees.
FICO credit scores are produced using information from three credit bureaus, Experian, Equifax, and TransUnion. Its algorithm generates credit scores that fall within a credit-score range of 300-850, and FICO registered a trademark for "300-850." The credit bureaus developed their own credit scores, based only on their data, which are less desirable to lenders, then began to develop a joint venture to create a tri-bureau algorithm that could compete with FICO (and let them pay less for the use of FICO’s algorithms). The result, VantageScore, was offered cheaply to key lenders in return for its adoption, in the hope of creating industry momentum.
I won’t review the antitrust claims in detail; suffice it to say that there’s not much left of antitrust law in this country.
FICO argued that when it adopted the 300-850 term it had no meaning in the credit scoring industry, and that there was no competitive need for that score range (or even any numeric score range). Moreover, FICO argued that defendants failed to meet their burden to show mere descriptiveness. However, they presented evidence “that the mark conveyed the approximate range of FICO's credit scores and that FICO had selected the mark for that reason,” as well as evidence of FICO’s own use of the mark to inform consumers that their scores would fall within that range. This was sufficient evidence of descriptiveness, even without consumer surveys: descriptiveness can be proved with the same variety of evidence that can be used to show genericness or secondary meaning. Consumers would immediately understand 300-850 to describe the qualities and characteristics (here, the range) of the score.
Fraud on the PTO requires knowingly false material representations of fact. Intent can be inferred from indirect and circumstantial evidence, but must be clear and convincing. Given the standard for reviewing jury verdicts, the court of appeals affirmed. Defendants pointed to two statements made in response to the PTO’s initial mere descriptiveness refusal. (1) A FICO employee stated that "[t]o the best of [her] knowledge, only the FICO score uses the 300-850 range as a unique identifier for credit bureau risk scores." FICO said this was true because no one else used the range as a “unique identifier,” but defendants argued that FICO didn’t use it as a mark either but only as a credit score range like other ranges. FICO responded that its specimen showed use as a mark, but defendants’ expert disagreed. A reasonable jury could have found falsity.
(2) FICO’s outside legal counsel stated that the mark was not descriptive because "300-850 is the credit scoring scale only for [FICO's] credit bureau-based risk products and not for ... other credit bureau-based risk products that competitors develop." FICO argued that this was true in context because TransUnion, the only other credit bureau using 300-850 as its range, wasn’t using the term as a mark. Still, a reasonable jury could have found falsity.
As to knowing falsity/intent to deceive, defendants presented evidence that FICO’s employee was aware that others were using the same range for the same purpose and that she knew that FICO wasn’t using the term as a mark. The “artful” use of the phrase “unique identifier” could count against her. The jury could have found similarly with respect to outside counsel.
FICO argued that even if the statements about third-party use of the range were intentionally false, they wouldn’t affect descriptiveness and were thus not material to the issuance of the registration. Third party uses, according to FICO, would have been material only if someone else had superior rights in the mark, and TransUnion didn’t, given its disclaimer of any use of the range as a mark. Because it filed an ITU, FICO argued, whether TransUnion previously used the mark "could never be material." Defendants presented a PTO expert who testified that a reasonable examiner “would consider it important in deciding whether to allow the registration to know whether others were using 300 to 850 as a score range for credit scoring services.” Also, since the application was initially rejected for mere descriptiveness and the PTO didn’t issue the registration until the two statements had been made, a reasonable jury could find reliance. (Indeed I am hard pressed to see how a reasonable jury could find otherwise.)
FICO argued licensee estoppel: Experian had signed an agreement that permitted it to use FICO’s 300-850 trademark and that had a no-contest provision barring Experian from challenging the validity of FICO’s exclusive rights to its marks. VantageScore was not a licensee, so Experian’s ability to raise the issue in its own name was irrelevant. An alter ego of the licensee may also be estopped by licensee estoppel, but that wasn’t the situation here. A mark that’s invalid can’t be infringed; since VantageScore successfully challenged the mark, it can’t serve as the basis of an infringement action against anyone.
The basis of the false advertising claim was Experian advertises an in-house credit score, the PLUS Score, with a 330-830 range and the claim “[s]ee the same type of score that lenders see.” FICO argued that this would mislead consumers into thinking they were either buying the FICO score or a score widely used by lenders, neither of which was true. Experian argued that “same type” meant “a three-digit numerical representation of an individual's credit risk based on credit report information and calculated using a credit-score algorithm.” FICO pointed out that no lenders use the PLUS Score and argued that Experian’s own documents demonstrated confusion by consumers over whether they were buying a score used by lenders.
The court of appeals agreed that the statement was neither literally nor implicitly false. The score was one of the “same type” that lenders see, “namely a score indicative of how lenders would assess an individual's creditworthiness.” FICO’s only extrinsic evidence was Experian’s call scripts, which didn’t demonstrate that an implicitly false message was being communicated.
Defendants argued that the district court abused its discretion by declining to find that this was an exceptional case for fees purposes. Exceptional cases mean that the plaintiff's action was groundless, unreasonable, vexatious, or pursued in bad faith. Defendants contended that the jury finding of fraud on the PTO weighed heavily in favor of finding exceptionality, along with FICO’s allegedly substantial delay in asserting trademark rights in the score range and other factors indicating that the score range was not a mark. The district court disagreed, refusing to find FICO’s claims “wholly without merit.” FICO managed to survive motions for summary judgment and judgment as a matter of law. This was not an abuse of discretion.