Saturday, May 15, 2010

Fair and fraudulent

Fair Isaac Corp. v. Experian Information Solutions Inc., 2010 WL 1875479 (D. Minn.)

Previous discussion. Fair Isaac lost summary judgment on its antitrust, breach of contract, and false advertising claims against Experian and the other credit bureaus, its competitors/customers for FICO scores. Following the recent tradition of weak trademark claims surviving more easily than other weak claims, the remaining trademark-related claims, along with a counterclaim for fraud on the PTO, proceeded to trial. At the close of evidence, the court ruled that because the keyword ad claim and the passing off claim would entitle Fair Isaac to equitable relief only, the court would decide them. The jury returned a verdict on the remaining claims in favor of defendants, finding that the alleged 350-800 marks, which the court had ruled descriptive, hadn’t acquired secondary meaning. The jury found for defendants on the counterclaim as well. Then the court ruled that Fair Isaac had failed to prove its keyword ad and passing off claims.

The fraud on the PTO counterclaim required the jury to find by clear and convincing evidence that (1) Fair Isaac made a false representation during the application process to the PTO for registrations of the 300-850 marks, (2) Fair Isaac knew that representation to be false when it was made and intended to deceive the PTO, and (3) the PTO relied on the false representation in deciding to issue the registrations.

The counterclaim was premised on two statements in response to the PTO’s initial denial on mere descriptiveness grounds. First, there was a declaration by a Fair Isaac employee stating that “[t]o the best of my knowledge, only the FICO score uses the 300-850 range as a unique identifier for credit bureau risk scores.” Fair Isaac argued that the statement was true because, though others may have used the term “300-850,” none of them claimed to have used it as a unique identifier. Defendants conceded this, but argued that the statement was still false because Fair Isaac didn’t use the term as a mark either. Fair Isaac rejoined that defendants didn’t make this argument at trial, and that it used a seal with 350-800 at its center as of 2004, making the statement true.

The second statement was from Fair Isaac’s outside legal counsel: “300-850 is the credit scoring scale only for Applicant’s credit bureau-based risk products and not for ... other credit bureau-based risk products that competitors develop.” Fair Isaac argued that this statement was taken out of context, which explained that the term was not per se descriptive.

The court upheld the jury’s verdict as based on sufficient evidence. The jury heard evidence that competitors sold credit bureau-based risk products that did in fact use the same, or nearly the same, scoring range and Fair Isaac knew these competitors sold such products. This was sufficient to justify an inference of fraudulent intent, and the statements were material in that they were the only intervening event between the PTO’s initial rejection of the registration application and its subsequent issuance of a notice of allowance. Defendants also offered testimony from a former Deputy Assistant Commissioner for Trademarks that the question of whether competitors were also using the 300-850 scoring range would have been important to a reasonable examiner in deciding whether to issue the registration.

Fair Isaac argued that the expert testimony shouldn’t have been allowed, because the test is not whether the information would have been important to a reasonable examiner but whether the registration would not have issued but for the false information. The latter standard, however, is only for damages, not for when the fraud claim seeks only to declare the trademark invalid.

Fair Isaac further argued licensee estoppel: defendants Experian and Trans Union should’ve been precluded from challenging the validity of a mark they’d licensed. Defendants contended that defendant VantageScore was not a licensee, and thus the jury would’ve heard the same evidence in any event. Fair Isaac replied that VantageScore is under the absolute control of the credit bureaus, meaning that agency principles and equity demand the application of licensee estoppel to it as well. But McCarthy says that other parties, even those “closely affiliated” with a licensee, are not foreclosed from challenging the validity of a mark, and there are several cases rejecting the agency argument. Thus, the court agreed that VantageScore could challenge the mark and so the issue would have been submitted to the jury regardless of licensee estoppel.

Fair Isaac also challenged the court’s ruling that 300-850 is merely descriptive, arguing that the categorization of the mark should have been submitted to the jury. Descriptive v. suggestive/arbitrary is typically a question of fact, but can sometimes be determined on summary judgment, as here. Though Fair Isaac argued that it arbitrarily chose the scoring range, that was unpersuasive: a manufacturer can’t claim a trademark in “Red Bike” because it arbitrarily chose to paint its bicycle product red instead of some other color. Nor is it persuasive to argue that the term is suggestive because the actual scoring range goes beyond 300-850. Analogously, “Red Bike” would still be descriptive if it were approximate rather than a precise description of a crimson color. (If the deviation were far enough, it would be misdescriptive, and still treated as descriptive for these purposes.)

Fair Isaac then argued that it should get JMOL or a new trial on secondary meaning. The court found that sufficient evidence supported the jury’s verdict, despite the existence of evidence favorable to Fair Isaac, including length of time in the market and deliberate copying of scoring ranges. The use of the score range in ads, for example, was descriptive. Fair Isaac argued that the court erred by failing to instruct the jury that the federal registration required a presumption of secondary meaning. The court had ruled, however, that the registrations placed a burden of production on the defendants to show that the term lacked secondary meaning; defendants met that burden, and therefore the burden of persuasion went to Fair Isaac.

Defendants also succeeded in their motion to amend the judgment to direct cancellation of the 300-850 registration based on lack of secondary meaning, stayed pending appeal.

Defendants moved for attorneys’ fees on the Lanham Act claims, and on the coordinate state law claims based on a provision of the Minnesota deceptive trade practices act allowing fees if the party complaining of a deceptive trade practice has brought an action knowing it to be “groundless.” The court held that this was a similar standard to the Lanham Act, which allows fee awards in “exceptional cases.”

The court declined to find that the infringement claims were groundless, unreasonable, or pursued in bad faith. Although the court earlier noted that this case seems to have been brought to deter entry of a new competitor into the market, Fair Isaac’s claims did survive a hotly contested motion for summary judgment and a motion for judgment as a matter of law at the close of a three-week trial. The jury verdict suggested that Fair Isaac “badly misjudged” the strength of its claims. But they weren’t wholly without merit, given uncertainties about how the court would rule on various motions in limine, the role of registration, and issues including licensee estoppel and spoliation. Fraud on the PTO didn’t necessarily make this an exceptional case either; courts still have discretion on fee awards. The hypothetical the court used to prove that not all cases of fraud on the PTO justify fee awards was one in which there was fraud on the PTO but the plaintiff won on common-law claims. Though that doesn’t seem very closely connected to this case, the court went on to note that defendants here would’ve had to defend against the common-law claims regardless of the registration—and of course since the standards for finding infringement are the same for registered and unregistered marks, they would have incurred substantial attorneys’ fees.

I’m not sure this really addresses the problem, though: in this case, the marginal cost of defending against the fraudulently obtained federal registration was clearly huge, given that a substantial amount of the litigation focused on fraud on the PTO, and on the related issue of secondary meaning. One can sue under the Lanham Act with an unregistered mark, of course; but I can certainly imagine a court finding that proceeding under one head of jurisdiction was groundless while one wasn’t—for example, trademark claims but not false advertising claims, as in the recent Franklin Mint case.

2 comments:

Charles Pergiel said...

I'm reading some of your posts and I am thinking that I would not want to be involved in any of these disputes in any way, shape or form.

They must take up a lot of people's time, and since time is a lawyer's stock in trade, they must likewise be very expensive.

What would be nice is if you could give us some idea of relative costs & benefits involved in these things.

Rebecca Tushnet said...

Litigation is certainly expensive. A heavily litigated Lanham Act case like this one could easily see a million dollars in lawyers' fees, possibly for each party, though the defendants may have shared some costs. In theory, the private costs should generate public benefit, and when there are truly false claims that can actually be true. I do think there's a general benefit to having a truthfulness requirement in advertising, so that we can largely trust the claims we see, but there is certainly a lot of expense in enforcing that requirement as well.