This guest post was written by Sean Costello. Mr. Costello is an associate resident in the Atlanta office of Jones Day. This post is entirely his work. It, of course, represents only his views, and not the views of his clients or firm:
The Class Action Fairness Act has succeeded in shifting more putative class actions from state to federal court. Whether we’re talking about original federal filings or removals from state court, in the post-CAFA world – and, particularly, in the months immediately after CAFA became law – there were more putative class actions in federal courts than before CAFA.
That’s the conclusion reached by the Federal Judicial Center (FJC) researchers who have been studying the statistics since CAFA took effect. Last July, I guest-posted about the FJC’s comparison of pre and post CAFA numbers. The FJC’s April 2007 report showed a dramatic increase in putative class actions filed in or removed to federal court.
In November 2007, the FCJ issued a short progress report on CAFA, which got about as much attention as the last one, which is to say, it got hardly any attention at all. But, like the last report, it deserves a look. This time, the FJC researchers took a harder look at what happened in state courts after CAFA, and the extent to which CAFA had an effect on its true target: state-law based “consumer protection” class actions. The FJC found that state-court class actions declined post-CAFA. Because the data on state-court class actions is so hard to come by – as the FJC writes in its report, “reliable data on class action activity in most state court systems simply do not exist” – the FJC researchers focused on one big state, California. Their preliminary conclusion? Class actions in California’s state superior courts went down, while class actions in California’s federal courts went up. Indeed, the FJC found a “marked increase in class action activity in 2005” in the California federal courts studied. The FJC’s analysis was limited to seven California superior courts and four California federal courts, but encompassed Los Angeles County, which has long had a hyperactive class action docket, and the FJC issued the usual caveats. But the findings are consistent with what happened in Madison County, Illinois and with anecdotal reports from lawyers defending putative class actions in other hot spots. My money, therefore, is on CAFA.
The November 2007 progress report also took a harder look at “consumer protection class actions.” The FJC’s earlier report lumped federal “consumer protection” class actions (which would include Fair Debt Collection Practices Act, Fair Credit Reporting Act, and Truth in Lending Act cases) together with state consumer protection actions in the “other fraud” category. That made drawing conclusions about CAFA’s effect on state “consumer fraud” class actions – which were a, if not the, primary impetus behind CAFA – difficult. This time, the FJC distinguished between state and federal “consumer protection” class actions, and the only conclusion one can reasonably draw from that analysis is that CAFA has delivered the goods when it comes to these bloated, lawyer-driven nuisance suits. In the years before CAFA, state “consumer protection” class actions pending in federal court fluctuated from year to year within a fairly narrow range, never increasing by more than 15 percent, and reaching a high of 30 in 2004. But in just the first six months of 2005, there were 43 such cases, and 37 of them were post-CAFA. For all of 2005, 88 state “consumer protection” class actions found their way into federal court, and 82 of them were post-CAFA. All of this added up to a nearly four-fold increase (193% including all cases and 173% including just post-CAFA cases) in state “consumer protection” class actions in federal court in 2005, compared with 2004.
That more of these state “consumer protection” actions are winding up in federal court is welcome news, but it’s not really surprising. The more interesting question is, what happens to these cases once they make their way out of the clerk’s office? After all, it has been almost three years since CAFA went into effect, and the cases that survived the initial jurisdictional challenges are actually getting litigated in federal court. The FJC has promised further reports analyzing remand rates, class certification decisions and other matters.
But we don’t have to await the next FJC report to identify one issue that deserves a closer look: what effect does the denial of class certification have on a federal court’s jurisdiction under CAFA? As putative class actions filed or removed under CAFA snake their way through the federal system, this issue undoubtedly will become more common. By my count, there are about a dozen published (electronically or otherwise) decisions on the question. Thus far, only the federal district courts have weighed in; there’s no federal appellate decision yet, at least as far as I know. Despite the issue’s potential significance, it hasn’t received much attention from the law reviews, mainstream legal press or blogs thus far. This is a first stab.
The issue generally arises in one of three ways--
First, the plaintiff files a putative class action premised on CAFA jurisdiction in federal court, class certification is denied (or the class allegations are dismissed), and the defendant moves to dismiss under Fed. R. Civ. P. 12(b)(1) (lack of subject matter jurisdiction). Though I haven’t done exhaustive research, in every case I found, courts concluded that they lacked subject matter jurisdiction over the case when the issue arose in these circumstances. See Falcon v. Philips Electronics North America Corp., 489 F. Supp.2d 367 (S.D.N.Y. 2007) (no jurisdiction under CAFA after finding that named plaintiff was inadequate); Hoffer v. Cooper Wiring Devices, Inc., No. 06-763, 2007 WL 2891401 (N.D. Ohio Sept. 28, 2007) (no jurisdiction under CAFA after finding that plaintiff was precluded from bringing class action under Ohio Consumer Sales Practices Act); and Arabian v. Sony Electronics, Inc., No. 05-1741, 2007 U.S. Dist. LEXIS 67769 (S.D. Cal. Sept. 13, 2007) (no jurisdiction under CAFA after finding that plaintiff failed typicality and predominance tests under Fed. R. Civ. P. 23(a) and 23(b)); Giovanniello v. N.Y. Law Pbl’g Co., No. 07-1990, 2007 U.S. Dist. LEXIS 56694 (S.D.N.Y. Aug. 6, 2007) (no jurisdiction under CAFA after dismissing class allegations because New York statute precluded class action in the circumstances); McGaughey v. Treistman, No. 05-7069, 2007 U.S. Dist. LEXIS 126 (S.D.N.Y. Jan. 4, 2007) (no jurisdiction under CAFA after finding that numerosity requirement of Fed. R. Civ. P. 23(a) could not be satisfied).
In the second scenario, the defendant removes a putative class action from state court under CAFA, class certification is denied (or the class allegations are dismissed, or some other post-removal event occurs), and the plaintiff moves to remand. In all but one of these cases, the court concluded that it still had jurisdiction. See Colomar v. Mercy Hospital, Inc., No. 05-22409, 2007 WL 2083562 (S.D. Fla. July 20, 2007) (retaining jurisdiction after dismissing the diverse co-defendant and rejecting claim that “home-state controversy exception” applied); Genenbacher v. Centurytel Fiber Co. II, LLC, 500 F. Supp.2d 1014 (C.D. Ill. 2007) (retaining jurisdiction after denying class certification); Giannini v. Schering-Plough Corp., No. 06-06823, 2007 WL 1839789 (N.D. Cal. June 26, 2007) (declining to exercise jurisdiction after dismissing class allegations); Garcia v. Boyar & Miller, P.C., No. 06-1936, 2007 U.S. Dist. LEXIS 39072 (N.D. Tex. May 30, 2007) (retaining jurisdiction after plaintiffs withdrew request for class certification); and Davis v. Homecomings Financial, 05-1466, 2007 WL 905939 (W.D. Wash. Mar. 22, 2007) (retaining jurisdiction after certifying statewide rather than nationwide class, and rejecting plaintiffs’ argument that, because the reduction in class size reduced the amount in controversy below $5 million, CAFA jurisdiction no longer existed).
In the third scenario, the case is filed in federal court or removed from state court under CAFA, class certification is denied (or the class allegations are dismissed), neither party raises the issue of jurisdiction, and the court raises the issue on its own and either assumes the answer or asks for briefs. See Good v. Ameriprise Financial, Inc., No. 06-1027, 2008 U.S. Dist. LEXIS 3982 (Jan. 18, 2008) (noting the split and asking for briefs, which are due February 2008); and In re: Welding Fume Prods. Liab. Litig., 245 F.R.D. 279 (N.D. Ohio 2007) (asking the parties to address the court’s jurisdiction, but assuming that it continued to have jurisdiction over the case).
You can certainly see why a federal court might want to get rid of a CAFA case after denying class certification, particularly where (as with most of the above cases) consumer fraud claims based on relatively small individual amounts are involved. What was once a potential multi-million or billion dollar case when the claims of thousands of people were aggregated has become a trivial dispute over pocket change or a few hundred bucks. That’s not the stuff of which federal cases are made.
First things first. CAFA itself doesn’t supply a ready answer, at least not explicitly. There is not a provision in CAFA that says, for instance, “the court shall retain jurisdiction over a case even if class certification is denied.” Some courts have suggested that language in CAFA providing that “[t]his subsection shall apply to any class action before or after the entry of a class certification order by the court with respect to that action,” 28 U.S.C. §1332(d)(8), can be read to support continued jurisdiction. See, e.g., In re: Welding Fume Prods. Liab. Litig., 245 F.R.D. at 317. It can’t be read that way, no matter how hard you squint. “[C]lass certification order” is expressly defined in CAFA to mean “an order issued by a court approving the treatment of some or all aspects of a civil action as a class action.” 28 U.S.C. §1332(d)(1)(C). In other words, only orders granting class certification (in some respect) fit. In fact, by negative implication, that language could arguably be read to support the view that denial of class certification means that CAFA no longer applies to the case. A better language-based rationale is that CAFA grants federal diversity jurisdiction over any “class action” satisfying its minimal diversity and amount in controversy requirements, and defines “class action” to mean “any civil action filed under rule 23 of the Federal Rules of Civil Procedure or similar State statute or rule of judicial procedure authorizing an action to be brought by 1 or more representative persons as a class action.” 28 U.S.C. §§1332(d)(1)(B),(2). In other words, CAFA recognizes that putative class actions are, by definition, aspirational – when the case grows up it wants to be a class action, but it isn’t there yet – and, thus, what matters is whether that aspiration is stated on the pleadings and papers.
Though to some extent courts rely on CAFA’s structure and language, including the “class action” definition, for the most part the results are a function of the general jurisdictional principle the court applies. The courts that retain jurisdiction say rely on the principle that, generally, events subsequent to removal (or filing) do not deprive the federal court of diversity jurisdiction. This is, of course, settled, black letter law. See, e.g., Garcia, 2007 U.S. Dist. LEXIS 39072, at *6-*7 (citing, among other cases, St. Paul Mercury Indemn. Co. v. Red Cab Co., 303 U.S. 283 (1938)); see also Genenbacher, 500 F. Supp. 2d at 1017 (same). Under this view, a post-removal class certification denial no more deprives the federal court of jurisdiction than a post-removal reduction in the amount of controversy. Congress is presumed to have known the law at the time it passed CAFA, so even if it is not enshrined in the statute’s language, it applies. See Davis, 2007 WL 905939, at *2; see also Miedema v. Maytag Corp., 450 F.3d 1322, 1331 (11th Cir. 2006).
The courts adopting the view that post-removal class certification denials do not deprive the court of jurisdiction make some other sound points. For instance, class certification is not a final judgment. It is interlocutory, and Fed. R. Civ. P. 23(c)(1)(C) expressly provides that “[a]n order that grants or denies class certification may be altered or amended before final judgment.” See, e.g., Genenbacher, 500 F. Supp. 2d at 1017 (“The denial of certification is an interlocutory order that may be altered or amended before final judgment.”). There’s also Fed. R. Civ. P. 23(f), which allows the losing side on class certification to seek an appeal. To send a case back to state court after denying class certification, therefore, could have some troubling implications. Since class certification is interlocutory, the state court arguably would be free to take an independent look, particularly where the state class certification rule differs from the federal rule, which is often the case. Suppose the state court comes to the opposite conclusion. Can the case then be re-removed on an “order or other paper” ground, even though it is well beyond the initial 30 day window? See Genenbacher, 500 F. Supp. 2d at 1017 n.3 (raising this issue). Maybe; maybe not. But even assuming that the case gets removed again? Then what? The federal district court might reinstate its class certification denial and reject the state court’s reasoning, but if the federal district court sends the case back again, the very same thing might happen. The case would become a ping pong ball, at least until one of the courts or one of sides gave up. (Eventually, the statute of limitations would probably come into play, but that, too, would raise a host of issues.) The federal district court might try to enjoin the state court, but that may or may not succeed. In most jurisdictions, it would probably fail.
Courts that send CAFA cases packing once the class allegations or aspects of the case have vanished feel constrained by the Federal Rules’ admonition that, “[w]henever it appears by suggestion of the parties or otherwise that the court lacks jurisdiction of the subject matter, the court shall dismiss the action.” Giovanniello, 2007 U.S. Dist. LEXIS 56694, at *3 n.11 (quoting Fed. R. Civ. P. 12(h)(3) (emphasis added)). These courts conclude that they should dismiss or remand CAFA cases after denying class certification (or dismissing class allegations) because either (i) Fed. R. Civ. P. 12(h)(3) gives them no choice, or, (ii) to the extent they have a choice, it is by virtue of the supplemental jurisdiction statute, 28 U.S.C. § 1367, and principles of economy, convenience and fairness and comity counsel dismissal. See, e.g., Giannini, 2007 WL 1839789, at *4. Speaking in terms of “supplemental jurisdiction” and relying on the supplemental jurisdiction statute doesn’t make a whole lot of sense. In these cases, the only basis for original jurisdiction was CAFA. There was nothing “supplemental.” At least one case has questioned this rationale on just this basis. See, e.g., Colomar, 2007 WL 2083562, at *3 (“[I]t is not readily apparent how a court can retain original jurisdiction of a case under CAFA, yet remand under a supplemental jurisdiction theory….”).
A dismissal under Fed. R. Civ. P. 12(b)(1) generally is without prejudice, so, like cases that are remanded, it well may be filed again in state court. The courts that go down this road are not oblivious to the potential consequences. To deal with them, these courts have fashioned a “rule” of sorts to distinguish cases that warrant dismissal (or remand) from run-of-the-mill class certification denials that might not. Here’s the rule (approach might be a better word): “[I]f class certification is subsequently denied on a basis that precludes even the reasonably foreseeable possibility of subsequent class certification in the future, the Court may lose jurisdiction.” Falcon, 489 F. Supp. 2d at 368.
Other courts have taken this a step further. They attempt to reconcile their dismissal (or remand) following denial of class certification with the principle that post-removal/filing events do not oust a court of diversity jurisdiction. In their view, if class certification is denied on a basis that is so clear that there is absolutely no “reasonably foreseeable possibility” of a class action, jurisdiction has not been lost. Instead, jurisdiction never existed in the first place—“there is not – and never was – diversity jurisdiction” under CAFA. Arabian, 2007 U.S. Dist. LEXIS 67769, at *15. This is a sort of fraudulent joinder doctrine in reverse, where post-removal evidence is used to show that jurisdiction did not exist even at the time of filing or removal.
While it may have its place in certain cases (e.g., where unequivocal, settled state substantive law says that no class actions may be brought, as in Hoffer and Giovanniello), I’m not sure the “reasonably foreseeable possibility” test is workable or desirable as a general matter, at least not without some limitations and caveats. It seems to raise more questions than it answers. For instance, if this is the test to be applied, why shouldn’t it be applied at the very beginning of the case? Upon filing or removal of a case in which jurisdiction is based on CAFA, should the court require the parties to address whether there is a “reasonably foreseeable possibility” of a class action? If a case is removed to federal court under CAFA, does the defendant have the burden of demonstrating that the plaintiff has a genuine shot at establishing a class action, since the defendant has the burden of establishing jurisdiction, and, if so, won’t that put the defendant in a very awkward position? Will the court need to decide, or at least make preliminary findings regarding, class certification in the context of a remand motion? How is “reasonably foreseeable possibility” defined?
I’m not sure what the answers are. Context and circumstance are, obviously, critical, and this might be the sort of issue in which nuance trumps general principles. As more and more federal courts begin reaching class certification in CAFA cases, the question, “what now?,” is sure to arise more frequently. It’s something to keep an eye on.