The PHH Case, Part I: Retroactivity and Liability Exposure

A little over a year ago, the Bureau determined that PHH Corporation had violated section 8 of the Real Estate Settlement Procedures Act, the prohibition on payments for referrals, and ordered PHH to disgorge just over $109 million. The D.C. Circuit heard arguments on PHH’s challenge to that order on April 4. The case has proved to be a major confrontation, raising the oft-repeated questions about the Bureau’s constitutionality in an appeals court that seems likely to decide them and contesting the Bureau’s interpretation of section 8 in some fundamental ways. Much has been written about the case; but in this and the next couple posts, I’d like to focus on a few aspects that have attracted less attention but could still be quite significant. Assuming the D.C. Circuit doesn’t eliminate the Bureau—and, given how the Supreme Court dealt with the last major separation-of-powers challenge to an agency’s structure, that assumption seems fair—these other issues will continue to be important even if the Bureau loses the PHH case.

The first topic is the temporal scope of liability. This is particularly significant because the Bureau evidently believes many of its administrative adjudications are not subject to a statute of limitations. This point is worth stressing because it is not limited to the kinds of claims PHH faced. The Bureau’s theory was that RESPA’s statute of limitations applies to “actions,” and “action” refers to a court case rather than an administrative proceeding. The Bureau is authorized to bring administrative enforcement proceedings for any of the laws it enforces (except to the extent other laws specifically limit that authority)–TILA, the FDCPA, Regulation O, and all the rest. Many of those laws state limit periods for “actions” just like RESPA does.  Presumably in cases under those laws the Bureau would take the same position it did here, that an administrative enforcement proceeding is not subject to the time limit.

If the Bureau prevails in this view—and that is at least a substantial possibility—it will be important for any entity facing Bureau enforcement to understand what other time limits there could be on liability.

The Director’s decision in the PHH case recognized one such limit. An administrative proceeding without a statute of limitations became possible only thanks to subtitle E of the Dodd-Frank Act, which became effective on July 21, 2011. By July 20, 2011, PHH’s potential RESPA liability for payments on July 20, 2008, or before had elapsed. For the Bureau to revive that liability would have amounted to a retroactive application of the law under which the Bureau (according to the Director’s decision) enjoys no statute of limitations for administrative adjudications. Retroactive operation is disfavored, so he imposed liability only back to July 21, 2008.

But one could argue for later dates, even if the Bureau is right that its administrative adjudication faces no statute of limitations. In this post I’ll discuss an argument for July 21, 2011, as the limit on PHH’s liability to the Bureau. (In a later post I’ll discuss July 16, 2013, as a possible limit.) This timing issue can be quite significant. While the public materials in the PHH case don’t reveal the detailed time series of its payments, it’s possible to estimate that if liability went back only to July 21, 2011, PHH would only have had to disgorge about $40 million. Much less than the $109 million the Bureau ordered. It is worth thinking about timing issues beyond the statute of limitations itself!

Why July 21, 2011? This involves that same issue of retroactivity. The Supreme Court has said that in general a statute does not operate retroactively unless it is clear Congress intended it to do so. The Bureau seems to take for granted that when it comes to Bureau enforcement, the Dodd-Frank Act should not have retroactive effect. I’ll take that for granted too. That’s why the Director did not impose liability for payments before July 21, 2008; doing so would have been a retroactive effect. But what about the payments between that date and July 2011? Would it be retroactive for the Bureau to require disgorgement of those?

The Director seems to have assumed, in passing, that it would not. The clue to this is one line in the decision that notes it is not a retroactive effect for a statute to modify procedural rules, “including changes to the forum which charges are prosecuted.” But it may not be so simple. Before the Dodd-Frank Act, the Department of Housing and Urban Development could enforce RESPA section 8 by suing in court. Now the Consumer Financial Protection Bureau can enforce it by having its enforcement office file charges before an administrative law judge and then seek an order from the Director. There are at least a few ways in which this is more than just a procedural change.

A “change in forum,” one imagines, is just a move from one court to another. The old court and the new court would apply the same decision rules, and the outcome ought, in the main, to be the same regardless of the change in forum. A change that can “affect substantive entitlement to relief” would arguably be more than just a change in forum. Moving from courts to administrative adjudication could have that kind of effect. For example, the Director of the Bureau might well have a statutory interpretation different from what a court would reach. A court’s task in statutory interpretation is to find a single best reading, even if the statute is ambiguous. By contrast, if a statute is ambiguous an agency might legitimately inject its policy preferences to choose an interpretation that is not the best one a court would reach but is still permissible.  So an agency as decisionmaker on statutory issues is different from a court.

On questions of remedy as well, an agency may be different. A court deciding whether to grant disgorgement is exercising its equitable discretion, subject to the standards applicable to that sort of authority. An agency head deciding whether to order disgorgement will presumably also take account of various policy matters—the value of deterrence, the significance of the issue to a broader industry or economy, the conduct of the respondent during the investigation, and much more—that would not be as relevant to a court. I’m not suggesting it would be improper for an agency to think about such things. But the fact that it might can make it a substantially different kind of forum from a court.

Besides allowing a new adjudicator, the Dodd-Frank Act also changed the complaining party. A change in plaintiff can be quite significant. For example, in Hughes Aircraft Co. v. United States, the Supreme Court held “permitting actions by an expanded universe of plaintiffs with different motives” would “essentially create[] a new cause of action, not just an increased likelihood that an existing cause of action will be pursued.” In that case the expanded plaintiffs were private qui tam relators who, under the prior law, would not have been able to bring the claims at issue. Adding private plaintiffs seeking financial gain is obviously not the same as changing from HUD to the Bureau. Still, the Bureau is not quite the same. Its policy mandates are not necessarily the same, and its enforcement priorities may be different. Moreover, the Bureau does have a financial interest different from HUD’s, because the Bureau can impose civil penalties that go into the Bureau’s Civil Penalty Fund rather than into the general fund at Treasury.  (True, the Bureau would not be able to impose those penalties on pre-enactment conduct, regardless how the broader retroactivity argument that I’m discussing plays out. But in a case where conduct spans the pre-enactment and post-enactment periods, the availability of civil penalties might encourage the Bureau to bring a case, even if it couldn’t obtain those periods for the pre-enactment conduct.)

These issues are complicated, and there are arguments to make both ways. I’ve only offered one side here just to sketch the ideas. The bottom line of the argument might be that the Bureau cannot obtain relief for RESPA violations predating July 21, 2011; or at least that it can attack such conduct only by suing in court. Would this be a windfall for RESPA violators? Not exactly. The Dodd-Frank Act left in place HUD’s authority to enforce section 8 by suing in court, with a three-year statute of limitations, just as it did before July 2011.


The Bureau’s Arbitration Proposal: Giving substantive effect to class actions?

On May 24, the Consumer Financial Protection Bureau published its proposal to restrict the use of arbitration agreements in a broad range of consumer financial products and services. “Predispute” arbitration agreements—typically entered as part of general product agreements—have been a topic of interest at the Bureau for some time, as it carried out research on how they are used. Section 1028 of the Dodd-Frank Act, which directed the Bureau to do those studies, empowers the Bureau to prohibit predispute arbitration agreements (with respect to consumer financial products and services) or restrict their use.

The Bureau does not propose to prohibit arbitration agreements entirely. Instead, it is focused on how arbitration agreements affect class actions. According to the Bureau’s proposal, class actions are an important mechanism for securing companies’ compliance with consumer financial law, especially because in many cases the damages available to a given individual consumer would be too small to warrant individual litigation. So the Bureau’s proposed rule would prohibit a company from using a predispute arbitration agreement to avoid a class action in court. If a company and its product are covered by the rule (a complicated question for another time), the company would not be allowed to “rely in any way on a pre-dispute arbitration agreement . . . with respect to any aspect of a class action.”  “Class action,” for these purposes, would mean not only certified class actions (those in which the court has ruled that the case can proceed on a class basis), but also cases in which the plaintiffs seek class status but the courts have not yet decided. In future agreements entered after the rule is effective, companies would also have to include a clause promising not to use the agreement to stop a class action in court.

The Bureau’s proposal includes an extensive discussion of the legal authorities on which the rule would rely. But one important discussion seems to be missing: Does the Bureau actually have the authority to privilege class actions in this way?

Class actions in federal court are, of course, a feature of federal judicial procedure—in particular Rule 23 of the Federal Rules of Civil Procedure. The federal courts establish procedure under the Rules Enabling Act, which says federal procedural rules cannot “abridge, enlarge, or modify any substantive right.”  The Supreme Court has repeatedly warned that Rule 23 must be interpreted carefully to avoid overstepping that bound.

The Bureau’s proposed rule would mean that when a plaintiff in federal court is a representative of a certified class, the defendant cannot assert a predispute arbitration agreement to compel the plaintiff to arbitrate claims that are “related” to the class action. (The arguments would be about the same for pre-certification cases, in which plaintiffs ask for class treatment but no decision has been made. For simplicity I’ll just discuss certified cases.) Notably, the Bureau’s proposal would not otherwise call consumer arbitration agreements into question. Absent a class action, the defendant would be free to enforce the arbitration agreement. But it would be unable to do so in a class action.

It seems likely to me that the federal courts could not properly establish a prohibition like that on their own. Of course the Federal Arbitration Act would preclude them from limiting the use of arbitration clauses. But the Rules Enabling Act on its own would also be an obstacle. Whether to arbitrate or litigate is a key question about a case that, as far as the federal rules are concerned, should not depend on whether the case proceeds on a class basis. Perhaps one could say arbitration is not a “substantive right”—what the Rules Enabling Act says federal procedural rules cannot affect. But the Supreme Court has interpreted that category broadly with respect to Rule 23. For example, the Court recently observed (although in dictum) that “evidence cannot be deemed improper merely because the claim is brought on behalf of a class.”  (That statement actually seems to go a bit far. The Rules Enabling Act is also the basis for the Federal Rules of Evidence, the very nature of which is to determine which evidence is proper or improper for a case in federal court.) Moreover, the promise to arbitrate is a contractual obligation; it would be hard to justify treating that obligation as inferior to other contractual promises with respect to the Rules Enabling Act.

So, can the Bureau achieve by regulation the result that the Rules Enabling Act would preclude for courts themselves? That is the question that the Bureau should contemplate.

It would be tempting to say there is no issue because, in the Bureau’s scheme, Rule 23 would not be affecting substantive rights; the Bureau’s regulation, and behind it section 1028, would be doing the work. But that answer only begs the question. Does section 1028 indeed authorize the Bureau to prescribe different outcomes for consumer class actions? It doesn’t say so explicitly. It allows the Bureau to restrict the use of arbitration agreements, but the Bureau and the courts will need to interpret that authority in light of other relevant law, such as the Rules Enabling Act. The Rules Enabling Act articulates a preference that the substantive rights of the parties to a case shouldn’t depend on whether they litigate in federal or state court. One could argue that a later statute, addressing an unrelated area, should not be read to permit exactly that outcome.

As an aside, note that although the Bureau’s proposed rule would also apply to state class actions, the enforceability of an arbitration clause could still depend on the difference between state and federal court. State class actions can be different from, and more limited than, federal class actions. For example, Mississippi has no class actions at all, and Virginia only on a few limited topics; South Carolina doesn’t allow class actions for monetary relief below a threshold of $100 per class member. South Carolina seems to rule out exactly the sorts of small-dollar class actions that the Bureau says it wants to protect!

Some hypotheticals might help illustrate why it is not self-evident that section 1028 would permit the Bureau’s proposed rule. The Bureau can identify and prohibit unfair, deceptive, or abusive acts or practices (“UDAAP”). Could the Bureau declare that a defendant in a class action about marketing of financial products is not allowed to introduce evidence of its communications with individual consumers? Assume for the moment that such a rule could be justified, on its own terms, under the UDAAP standards. Would UDAAP authority permit the Bureau to, in effect, revise the evidence rules for class actions?

As another example, under the Fair Debt Collection Practices Act (“FDCPA”) the Bureau can generally regulate “the collection of debts by debt collectors.”  Sometimes when a consumer complains that a debt collector is trying to collect an incorrect amount, it turns out that the collector has already obtained (perhaps by default) a state-court judgment for that amount. In that situation, a federal court will often refuse to entertain the consumer’s claim under the FDCPA, on the ground that the Rooker-Feldman doctrine precludes a lower federal court from questioning a state-court judgment.  Could the Bureau overcome that obstacle by issuing a rule stating that a state-court judgment does not conclusively show that a debt collector is claiming the right amount?

I raise these questions here, but I don’t mean to be suggesting an answer. These are interesting issues of statutory interpretation, and the best approach presumably depends on the details. (The Bureau’s proposal cites a few existing regulations from other agencies that restrict the use of arbitration agreements. But so far as I can tell, among those examples only the SEC’s involved a comparable preference for class actions; and that rulemaking did not consider the issue. Moreover, the SEC was only approving a rule issued by FINRA, a private industry self-regulatory body, rather than imposing the restriction itself. That fact might make the analysis different.) With respect to the Bureau’s arbitration proposal, there are surely arguments to be made that the Bureau does have the authority to privilege class actions against arbitration. And it might not be surprising if historical materials associated with the enactment of the Dodd-Frank Act reveal that section 1028 was meant to permit exactly that sort of restriction. My point here is simply to observe that this is an important and non-trivial issue, one that the Bureau and anybody interested in the Bureau’s arbitration proposal should think about.