N.D. Cal Rejects Preemption and Standing Defenses Against Claims Under CA Spam Statute — Asis Internet Servs. v. Subscriberbase Inc.
[Post by Venkat]
Asis Internet Services v. Subscriberbase Inc., (N.D. Cal.) Case No. 09-3505 SC; April 1, 2010 [scribd]
Judge Conti (in the Northern District of California) issued a potentially significant decision last week that keeps the door open for plaintiffs alleging claims under California’s spam statute. This is good news for anti-spam plaintiffs, and comes on the heels of last month’s state court trial win for another spam plaintiff which resulted in an award of $7000. (Balsam v. Trancos)
Background: The Ninth Circuit in Gordon v. Virtumundo held that veteran spam plaintiff James Gordon could not sue under CAN-SPAM because he was not a bona-fide ISP (or provider of an “internet access service”) and because he did not suffer any “adverse effects” from the spam he received. He was harmed as any email recipient would be and thus was not among the class of plaintiffs entitled to sue under CAN-SPAM. The court also held that CAN-SPAM preempted Gordon’s claims under Washington’s spam statute. Here is Prof. Goldman’s post discussing that ruling. The post asked whether Virtumundo would spell an end to “spam litigation factories”. Following Virtumundo, courts knocked out Gordon’s many pending cases. Plaintiffs who appear similarly situated to Gordon have persevered in California, trying to assert claims under California’s spam statute.
One issue that was left unresolved was whether CAN-SPAM preempted California’s anti-spam statute (and how much of it was preempted). A few courts have dealt with this issue. One court held that a plaintiff must satisfy the “actual fraud” standard in order to assert a claim under a state spam statute. (Hoang v. Reunion.com, discussed by Ethan here and here.) Other courts (Asis Internet v. Vistaprint and Asis Internet v. Consumerbargaingiveaways, discussed by Ethan here) have taken a less restrictive view, rejecting Reunion.com’s premise that only claims satisfying the traditional “actual fraud” standard survive CAN-SPAM’s broad preemption clause.
This case takes a fresh look at the issue and rejects the Reunion.com view. Regardless of the outcome of this dispute, portions of California’s spam statute, such as the provisions imposing a blanket ban and those requiring labeling are preempted. The issue is whether there’s room to allege that emails are misleading based on information in the subject lines, from lines, or headers, and make a claim under California’s spam statute.
The Court’s Ruling:
1. Were the Subject Lines Were Likely to Deceive Recipients: Defendants argued that the subject lines were not likely to deceive recipients. Section 17529.5(a)(3) prohibits subject lines that are likely to mislead “about a material fact regarding the contents or subject matters of the message.” The court predictably refuses to conclude as a matter of law that the email subject lines are not deceptive. I don’t know if anyone reasonably believes that email subject lines which offer free products (for example: “Review & Keep Designer Handbags worth $1500 Dollars-guys invited too”) actually offer free products. But given the statutory language, it would have been tough for the court to conclude as a matter of law that the subject lines were not “likely to mislead.” Once you use the word “free,” if something is not actually free, you will have an uphill battle getting a court to conclude that the marketing copy is not misleading as a matter of law.
[Tip to marketers: avoid using "free" unless something truly is free, particularly when this claim is made in an email or online!]
2. Did Defendants Have Knowledge That the Subject Lines Were Likely to Deceive: Defendants also argued that the complaint did not sufficiently allege that defendants knew the subject lines were likely to deceive, in light of the fact that defendants did not transmit the messages. Plaintiffs earlier lost a case on this precise issue, where a district court held that the defendants could not be held liable for spam received by plaintiffs because the defendants in that case neither sent nor “knowingly procured” the messages. (Here’s Professor Goldman’s post discussing that ruling, along with other affiliate liability issues. As mentioned below, this ruling was affirmed by the Ninth Circuit.) In light of the fact that plaintiffs previously lost on this issue, it’s irritating for plaintiffs to get another chance here. Defendants will likely have to fight this one out on summary judgment, rather than at the motion to dismiss stage. It would have been nice for the court to require some specific allegations regarding defendants’ supposed knowledge.
3. Do Plaintiffs have Standing Under California Law: Defendants’ first standing argument was that Proposition 64, which amended California’s false advertising and unfair competition laws to provide that only plaintiffs who “suffered injury in fact and has lost money or property” could bring claims, applied to section 17529.5 and barred the claims asserted by plaintiffs. The court rejected this argument, noting that section 17529.5 included “independent, non-exclusive standing and remedy provisions” which authorize ISPs (defined as “electronic mail service providers”) to bring suit. The court acknowledged the effect of Prop 64 was unclear, and reasoned that Prop 64 was designed to curb the practice of members of the general public bringing lawsuits to enforce advertising and unfair competition rules, rather than change the standing requirements of statutes which only applied to a narrower (more specific) class of plaintiffs. According to the court, since California’s spam statute only authorized a narrow class of plaintiffs to bring claims to begin with, the statute was left unaffected by Prop 64.
4. Are Plaintiffs’ Claims Preempted by CAN-SPAM: Defendants relied on Gordon v. Virtumundo and argued that plaintiffs’ claims were preempted by CAN-SPAM. The court acknowledged that other courts in California had reached differing results on the preemption issue. Hoang v, Reunion.com held that CAN-SPAM’s preemption clause only leaves room for state law causes of action based on “common law fraud,” which requires allegations of reliance and actual harm. On the other hand, two judges in the Northern District of California (Asis Internet v. Vistaprint and Asis Internet v. Consumerbargaingiveaways) came to a different conclusion, finding that section 17529.5 was not preempted, even though this statute does not require a showing of reliance or damages. (Odd sidenote: the Reunion plaintiffs are represented by the same lawyers who are representing defendants in this case.)
Judge Conti found that Virtumundo “did not clearly resolve this split.” In his view, the Ninth Circuit in Virtumundo found the plaintiff’s claims under Washington’s spam statute preempted because those claims reached “non-deceptive statements.” The court focused on the fact that in Virtumundo, plaintiffs were complaining about immaterial errors in emails (the fact that an advertiser’s name did not “expressly appear in the ‘from lines,’” and Virtumundo’s use of “fanciful domain names”). Judge Conti viewed Virtumundo (and Mummagraphics) as standing for the proposition that only state law claims that reach immaterial errors and omissions were preempted. The court’s discussion on preemption is in depth, and worth reading. Ultimately, the court concludes that requiring the elements of reliance and damages would have one practical consequence:
[i]t would limit the scope of entities that are entitled to bring suit under section 17529.5. It would restrict enforcement suits to the (presumably more rare) case where a plaintiff actually been duped by a misleading subject line. The bulk of deceptive email would go unpunished, until such an email happened to mislead someone with the resources and wherewithal to pursue a private claim.
The court found that Congress could not have intended this result under CAN-SPAM.
5. Whether the Enforcement Mechanism of Section 17529.5 Conflicts With CAN-SPAM: The final issue (which it sounds like defendants didn’t press, but which to me was the most interesting) was whether the enforcement mechanism of the California spam statute conflicts with CAN-SPAM. CAN-SPAM only allows for enforcement by two classes of entities: (1) certain government actors and agencies and (2) bona-fide ISPs who have been been “adversely affected”. The California spam statute on the other hand allows the Attorney General, ISPs, and recipients of emails to bring suit. As the court notes, allowing a broader class of plaintiffs to sue under state law may “disturb . . . the fine balance struck by Congress . . . .” However, the court concludes that it would not upset the balance intended by Congress because CAN-SPAM’s preemption savings clause only speaks to the subject of state law (rather than the class of plaintiffs authorized to sue). The court also noted that there was no evidence of Congressional intent to “occupy the field.”
My Reaction: The court is right that Virtumundo and Mummagraphics found the state law claims in those cases to be preempted by CAN-SPAM because those claims relied on immaterial errors. However, California plaintiffs typically raise similar claims, and courts have not been the most vigilant about scrutinizing them. Plaintiffs seem to construe California’s spam statute pretty broadly. For example, in Trancos, the recent decision following trial, and Kleffman v. Vonage, a case pending in front of the California Supreme Court, some of the arguments raised by plaintiffs were similar to the arguments raised by Gordon in Virtumundo. The preemption structure only works if courts dealing with the state law claims use the same standard for what is false or deceptive.
The practical consequence of the court’s ruling is to open a gaping exception to the enforcement structure that Congress may have intended. Congress intended only two classes of actors to enforce CAN-SPAM: (1) the authorities and government agencies and (2) bona-fide ISPs who were adversely effected. CAN-SPAM does not allow for enforcement by mere recipients of email, and doesn’t allow for enforcement where the recipient does not suffer any injury, and the court’s ruling opens the doors to this enforcement. This is particularly troubling, given that many plaintiffs do not take steps to avoid email and may even invite the harm in question. To the extent they are trolling for spam emails in order to bring lawsuits, they are really no different from a member of the general public.
Another issue to consider is the scope of affiliate liability. It wouldn’t make sense for state laws to hold a broader category of actors liable (under a lower standard) than CAN-SPAM, but this could be a practical result of allowing these types of claims to proceed under California’s spam statute.
Finally, I’m not sure there’s anything wrong with allowing ISPs and those with sufficient resources to sue, when they actually have been harmed. I doubt we’ve derived much benefit from the proliferation of the anti-spam litigation “cottage industry.”
What Does the Ruling Mean For Spam Litigation: The ruling gives spam plaintiffs in California some breathing room. With respect to subject line violations at least, plaintiffs can continue to bring claims under California’s spam statute, and unfortunately, can continue to try to sue affiliates as well. They can bring claims in state court (as in Trancos) or in federal court, to the extent they are able to independently satisfy jurisdictional requirements. There is a case pending in front of the California Supreme Court (Kleffman v. Vonage – oral argument is set for May) which should clarify the scope of California’s anti-spam statute and whether it’s preempted, and Reunion.com is likely to be appealed as well, once it makes its way out of the trial court. But until then, plaintiffs will probably cite to this case for the proposition that subject line claims and other claims based on allegedly misleading aspects of emails that don’t necessarily violate CAN-SPAM are not knocked out by CAN-SPAM’s preemption clause.
Other Asis Cases: Asis has filed many many different spam lawsuits. 13 alone according to a Justia search. There’s been ongoing activity on several of those cases recently:
Asis Internet . Azoogle.com, Case Nos. 08-15979 and 08-17779 (9th Cir.) (Dec. 2, 2009): 9th Cir. affirms dismissal of claims brought against various entities for lack of standing and because Asis failed to prove affiliate liability.
Asis Internet v. Member Source Media, Case No. C-08-1321 EMC (Jan. 28, 2010): CAN-SPAM claims dismissed.
Asis Internet v. Active Response Group, Case No. C07-06211 TEH (Feb. 9, 2010): dismissed for lack of standing under CAN-SPAM; state law claims dismissed without prejudice.