by Mark Moller
In a recent post, Stephen Gardner makes some interesting comments about my recent article, Class Action Lawmaking: An Administrative Law Model. Before I respond, let me quickly summarize what the article is, and isn't, trying to do: Stephen says I am advocating "turning federal courts into little more than administrative agencies that defer to the Legislative Branch as much as possible." That's not quite what I'm arguing and my apologies if I wasn't clearer.
To clarify, my article tries to draw out some parallels between class actions and agency proceedings in the field of statutory interpretation. In particular the article makes the following related claims: First, federal courts in the class context often assume far more discretion to "play" with the meaning of statutes passed by Congress than they would consider appropriate if the interpretive liberties were taken by an agency. These courts assume this extra interpretive "play" or "license" when the interpretive dispute affects the discretion of the court to certify a class; trial courts often play with the statute's meaning in these questionable ways in order to pave the way for certification.
If, however, federal courts followed the interpretive guidelines that apply to agencies, they would act with more restraint when interpreting such statutes, by hewing to the probable or "clear" meaning of the statute, even if selecting that meaning would mean rejecting certification and even if there's a plausible but not probable contrary reading that would permit the class action to go forward. That's how a good agency, one that follows the rules of interpretation that the Supreme Court commands agencies to follow, would do things.
To make this a little more concrete, say, for example, that a plaintiff's lawyer, we'll call him Joe Wheat, brings a class of hundreds of thousands of statutory fraud claims against Acme Corporation under a federal consumer protection statute. Acme argues this statute envisions case by case inquiries into the subjective reliance of individual class members. Wheat argues, instead, that the court should adopt a presumption of reliance that excuses plaintiffs from meeting such a burden. The courts I'm criticizing might, for example, pick Wheat's version based on a policy preference in favor of class certification-even if Acme's contrary version would be selected as the "clear" meaning under the interpretive rules that govern in administrative law.
Stephen's main contention, if I read him right, is that the choice between Acme's and Wheat's readings of the reliance requirement should, as a separation of powers matter, be left to courts' own policy preferences. If that's his argument, I disagree.
The choice between flavors of reliance is a choice that Congress is authorized to make. No one would argue that courts could simply ignore Congress if it commanded that "in claims under this statute, reliance must be proven based on individual, case-by-case inquiries into the totality of the circumstances; in no circumstances may courts presume reliance or use circumstantial evidence to estimate the frequency of reliance across a class of claimants." I don't read Stephen to believe that courts could ignore such a command, either.
But, when Congress isn't completely crystal clear about how causation must be proven, Stephen, however, seems to assume courts can simply fill in the gaps as they like. Article III, however, confers no federal common law power on federal courts; if courts fill gaps in federal statutes, that gap-filling must rest on a default assumption about Congress's preferences or on a legislative delegation of policymaking authority to the court. There's no evidence Congress wants all ambiguities to be resolved in favor of class actions (quite the contrary, if the steady drumbeat of federal class action reform is any indication). And, there's no reason to think Congress would want to delegate more interpretive flexibility to class action courts than they do to agencies. If they did, the Chevron framework in administrative law would be hard to justify-at least if Chevron is to be justified, as it is in United States v. Mead Corp.., as a rule that accords with Congress's intent.
So what to do? While there's perhaps a case to be made that courts should interpret ambiguities against certification, that's a pipe dream. So, my proposal is much more modest (perhaps entirely too modest for the taste of many scorched earth class action reformers): When faced with a statutory interpretation question that affects whether a class can be certified, courts should measure the scope of their interpretive discretion based on the interpretive practices that courts demand of agencies. When deciding between Joe Wheat's version of reliance and Acme's, in other words, courts should take a dose of the medicine they prescribe agencies. If courts took that administrative law baseline seriously, it would be a modest move in the direction of judicial restraint. Of course, in so saying, I don't mean to compliment agencies-the fact that courts in the class context exceed even agencies in their zest for creative interpretation shows just how far class management has pulled courts from the faithful agent model of interpretation.
Comments