New class certification decision: Sony Electronics, Inc. v. Superior Court

In Sony Electronics, Inc. v. Superior Court, ___ Cal.App.4th ___ (Nov. 28, 2006), which was published yesterday, the Court of Appeal (Fourth Appellate District, Division One) reversed an order granting class certification of UCL, CLRA, and Song-Beverly Consumer Warranty Act claims, holding that the class, as certified, was not ascertainable. The trial court had certified a class consisting of: "All persons and entities who [bought certain Sony laptops] in which the memory connector pins for either of the two memory slots were inadequately soldered." Slip op. at 4-5. The Court of Appeal relied on the rule that a class must be defined in terms of objective facts, not in terms of liability:

Unfortunately, because there is no evidence showing that this alleged manufacturing defect is universal to all GRX Series Notebook computers, the class definition requires a merits-based determination in order to establish whether a particular GRX Series Notebook owner is a member of the class. The members of such a class are thus not readily identifiable so as to permit appropriate notice to be given and the definition would not permit persons who receive notice of this action to determine whether they are part of the class.

Id. at 12. The opinion concludes by directing the trial court "to conduct further proceedings on the issue of whether an alternative class is properly certifiable." Id. at 14.

Comments

[TommyK]

I was unclear on the court's reasoning. Are they enshrining the notion that if the merits need to be decided to determine who is a class member (other than from objective records), class cert is inappropriate? Or was the standard that the people getting notice have no way to readily determine they are class members? They also enshrined some concepts from the Suzuki case that I had thought were questionable law post Linder. Hmmmm . . .

[John Hurley]

Tommy -- the court was not advocating an inquiry into the merits on class certification. The issue in the case was that the class definition constituted an improper "fail-safe" class.

The class was defined as those persons who bought computers where the pins were "inadequately" soldered -- but the "adequacy" of the soldering was also the basis of the class claims. So if a class were certified and Sony prevailed on the merits, it would be a hollow victory since no class of plaintiffs would be bound by the result since there would not be anyone with a computer with inadequate solder.

To avoid this type of problem, the class in this case could have been defined as all purchasers of the particular model of computer, without reference to whether or not the computers of the individual class members had the claimed defect. Of course, the plaintiff can then run into serious commonality/typicality problems if the defect is not widespread.

[Kimberly A. Kralowec]

I did not think this case was particularly noteworthy, except insofar as it construes the ascertainability element of class certification, which few cases do. As John points out, the primary problem with the class definition was its use of the word "inadequately," which requires a factual finding that should be reserved for the trier of fact.

[Lex Aquila]

John Hurley is incorrect, I believe, in assuming you could certify a 'no injury' class of people who bought a certain model of computer. Such an 'all purchasers' class definition sweeps within the "class" consumers who purchased computers without the alleged defect. In other words, it makes the class one of uninjured and maybe some injured consumers. Because an injury would be required for the underlying claims, this particular class definition should not stand. In fact, what the Judge certified was closer to an 'incidents' class, which would at least contain only injured purchasers. However, this particular class was probably not ascertainable. I didn't read the decision, but there are perhaps other ways to define the class with objective facts that are not contingent on legal findings (adequacy, etc.).