Tuesday, October 15, 2013

Software update defeats class certification

Waller v. Hewlett–Packard Co., No. 11 cv0454, 2013 WL 5551642 (S.D. Cal. Sept. 29, 2013)

So I’m trying to cut down on California coverage and just give the highlights.  That said, I recommend this case for some of the most detailed discussion of the Article III standing v. class action mechanism issue that’s percolating (or perhaps just thrashing around) in the California district courts.  Here’s my previous discussion of the claim that HP misleadingly advertised hands-free automatic backup, when in fact backing up certain file types required individual customer programming.

As the court explained, it had stayed Waller’s motion for class certification pending the appeal of a different case, expecting that “the Ninth Circuit would confront and resolve a question that has become a kind of spike strip in the class certification of lawsuits” under the UCL—whether all members of a putative class must have Article III standing and what that would mean, given that the underlying claim has no injury requirement but Article III standing does.  The potential consequence of requiring Article III standing for every unnamed class member is that “[s]imple removal by the defendant would be a game-changer.”  Plus, a lot turns on whether actual reliance is required, or merely a purchase of a product with misleading labeling: the former would generally preclude certification.  Here, the court reconsidered its decision to stay the case and denied Waller’s motion with prejudice, while leaving him with individual claims.

The Ninth Circuit has held that UCL claims are governed by the reasonable consumer standard, meaning that individualized proof of reliance and causation isn’t necessarily required. Stearns v. Ticketmaster Corp., 655 F.3d 1013 (9th Cir. 2011).  Relatedly, there was no Article III problem as long as class members “were relieved of their money”; and only the standing of the representative party mattered anyway. Mazza v. American Honda Motor Co., Inc., 666 F.3d 581 (9th Cir. 2012), contained a throwaway line “[N]o class may be certified that contains members lacking Article III standing,” citing a Second Circuit case but then immediately citing Stearns and not explaining the contradiction, and then setting a low bar for standing: as the court here summarized, “[s]imply spending money on something that doesn’t do what it claims to do is all the injury absent class members need” under Mazza.  The district court opinions are also in disarray.

Considering the precedent, the court here concluded that where a class representative has standing, arguments about unnamed class members’ injuries should be analyzed through Rule 23, not through examining the Article III standing of unnamed class members.  For one thing, Mazza didn’t acknowledge what it was arguably saying or deal with earlier conflicting circuit and Supreme Court authority.  Lewis v. Casey, 518 U.S. 343, 395 (1996) (“One class representative has standing, and with the right to sue thus established ... the propriety of awarding classwide relief ... does not require a demonstration that some or all of the unnamed class could themselves satisfy the standing requirements for named plaintiffs.”) (Souter, J., concurring).  Also, HP removed this case to federal court, and it would be unfair to let HP do that and then “seize on federal constitutional standing requirements to say the case cannot proceed as a class action.… HP, for perfectly understandable reasons, doesn’t like that relief is available under the UCL without any proof of reliance or injury, and it is trying to backdoor those elements into this case with an Article III standing requirement.”

Even if the court was wrong about Article III, it would find that absent class members had Article III standing here.  The meaning of “injury in fact” isn’t entirely clear in a misrepresentation-based UCL case. Some cases suggest that “simply buying a product that doesn’t do what it claims is an automatic economic injury, even if the misrepresentation is irrelevant to the purchaser’s usage.”  But arguably the injury must be traceable to the defendant’s conduct “in some thicker sense: the plaintiff must have actually focused on the alleged misrepresentations in deciding to buy the product, and suffered some injury by relying on the misrepresentation.”  If misrepresentation is enough, then standing isn’t much of a barrier as long as absent class members bought a product of diminished capabilities and therefore diminished economic value.  But if reliance is required, that’s almost invariably a highly individualized inquiry. 

The court concluded that Article III would be satisfied by the purchase of a product containing the alleged misrepresentations.  First, a contrary rule would make class actions of this kind “dead on arrival in federal court.”  All a defendant would have to do would be to remove under CAFA and kill them.  Second, Article III’s requirements turn on the nature of the claim asserted, and in California UCL and FAL cases don’t require individualized proof of deception, reliance, and injury. The UCL is focused “on the defendant’s conduct, rather than the plaintiff’s damages, in service of the statute’s larger purpose of protecting the general public against unscrupulous business practices.”  “That being the law of the substantive claims at issue in this case, it makes little sense to concoct and impose a standard for Article III standing that effectively contains the very reliance and deception requirements the actual claims do not.”

Third, the court had a (relatedly) broader conception of Article III injury than HP argued for.  Reliance on misrepresentation causes loss.  But there’s also an economic loss “where a product simply doesn’t do something it purports to do—irrespective of whether that hampers the consumer’s intended use and irrespective of its hypothetical, retrospective impact on the consumer’s purchasing decision.”  That’s just basic economics: the market price has been inflated.  “A car with dysfunctional headlights, for example, is simply less useful and therefore less valuable than one with working headlights; it’s no rebuttal to say that the owner of the former doesn’t drive the car at all or drives it only in broad daylight and never in tunnels. Everyone who pays for a car is paying some amount of money for working headlights.”  Fourth, the court was unwilling to draw the fine distinction between product defects “that are imminent, unmitigatable, and pose a constant risk of harm, and defects that are hypothetical and may never cause a consumer any injury or loss in value.”  Defective airbag sensors, for example, only harm drivers who actually get in accidents.  “[O]n some level, the question of injury caused by a product will always be hypothetical; it will always turn on an individual’s particular usage of that product.”

Turning to class status, the court easily found numerosity and commonality (what HP represented its SimpleSave hard drive would do and whether that was misleading to a reasonable consumer).  Waller’s claim was typical: he alleged that he relied on HP’s misrepresentations about SimpleSave and lost money thereby; it didn’t matter that not every member of the class would’ve suffered a data loss, and some might not have had the “outlier” file types that weren’t automatically saved.  The representation at issue—that SimpleSave automatically backed up all file types—was either absolutely true or absolutely false.  “It may back up all of a purchaser’s files, but that just means the purchaser is lucky, and it doesn’t mean that he possesses a product that’s of less utility than is advertised.” Even customers with no outlier file types “bought and now own a device that just doesn’t do what it’s alleged to do, and is therefore of diminished utility and value.”

Waller and his counsel were also adequate representatives.

The court rejected HP’s arguments that individualized questions of reliance and injury defeated predominance because different users might’ve understood how SimpleSave worked differently.  This misunderstood what a UCL and FAL claim required: “very little.”  Deceptiveness was a question of materiality, which was for a jury to decide rather than a court at the certification stage.  As a policy matter, California focuses on the defendant’s objective conduct. “If a product is advertised with a misrepresentation, that is more or less the end of it; it can be presumed at the class certification stage that the consumer would have paid less for the product, or not purchased it at all, with more accurate information.”  HP argued that this produced a windfall to unharmed or oblivious consumers; this was true “to the extent that the UCL and FAL are extremely consumer-friendly statutes that require very little to establish liability.”  But there were checks on runaway liability: the UCL is an equitable statute generally limiting plaintiffs to injunctive relief and restitution, not damages or attorneys’ fees, and HP had “presentable” arguments for a small, relatively painless restitution award given that the files not automatically backed up were used by less than 1% of the population; studies showed that most people would’ve bought the product anyway; and the user manual also disclosed that the product could be programmed to back up the outlier files.  A trier of fact might well find no material misrepresentation or only a tiny one.  “Of course, Waller puts the value differential somewhere between $12 and $30 per hard drive, but these are all questions for the trier of fact to resolve.” 

A second check on runaway liability “is the requirement that class members at least have been exposed to the misrepresentation,” per Mazza.  HP argued that putative class members weren’t exposed to a uniform representation, since the 1 TB drive had different packaging from the 320 GB drive Waller bought, and since online purchasers didn’t all see the same representations: one website “neither described the SimpleSave or included text from the product packaging in the listing,” while Amazon apparently included the misrepresentations at issue but it wasn’t clear that the text was identical to the actual packaging or how responsible HP was for that description.  HP’s point could be addressed by modifying the class definition, possibly with a website-by-website analysis.  At a minimum, certification of a class of all in-store purchasers of SimpleSave devices could be appropriate.

However, “several months after Waller purchased his SimpleSave and after he filed his original complaint in this case, HP released a free software update for the SimpleSave that makes it automatically back up all file types.”  And already-purchased SimpleSave devices automatically checked for updates when connected to the internet.  The update remains available even though SimpleSave devices are no longer made or sold.  Indeed, Waller’s own SimpleSave had the update soon after it was released.  Waller conceded at his deposition that the update addressed his grievance.

HP argued that common questions couldn’t predominate given that individualized inquiries would be needed to determine when class members took advantage of the upgrade.  “Not only do individual issues potentially predominate when there’s an available remedy for the grievance of the putative class, but the availability of that remedy calls into question the Court’s very benefit-of-the-bargain theory of injury in this case.”  Once purchasers got a device that automatically saved all their files, they then had exactly what they paid for.

Relatedly, HP argued that Waller didn’t fairly and adequately protect the class’s interests given the remedy already available without a costly lawsuit.  When disappointed class members would be better off with refunds or replacements, the high transaction costs—notice and attorneys’ fees—arent’ justified.  This tied in to HP’s argument that the upgrade was superior to a class action.  Although Rule 23(b)(3) speaks of “other available methods for fairly and efficiently adjudicating the controversy,” and upgrades aren’t a form of “adjudication,” remedial measures taken by a defendant can be taken into account in superiority analysis. 

Even if the update didn’t affect predominance, it did bear on adequacy and superiority.  The upgrade also went to the core of the economic injury argument at the heart of UCL certification as the court interpreted it. “The UCL may not require any individualized proof of deception, reliance, and injury, but what that really means is that it doesn’t require proof of deception, reliance, and injury beyond a buyer paying more for a product than he otherwise would have.”  Without that threshold economic injury, there’s no UCL claim. Waller’s extra injuries—the cost of retrieving files that weren’t backed up—weren’t enough to satisfy the injury and causation requirements for standing in a restitution class action.  Thus, the update caused certification to fail.

The court did reject HP’s argument that Waller didn’t offer a reliable method for calculating restitution, something that need not be shown with certainty at the certification stage. The financial injury could be measured by calculating the cost differential between the price of the SimpleSave and the price of a device that had to be manually configured to save certain file types.

No comments: