FILED
                                                         United States Court of Appeals
                                                                 Tenth Circuit

                                   PUBLISH                     December 20, 2016
                                                               Elisabeth A. Shumaker
                   UNITED STATES COURT OF APPEALS                  Clerk of Court

                                TENTH CIRCUIT


 ELIZABETH HAMMOND,

       Plaintiff - Appellee,
                                                       No. 16-2243
 v.

 STAMPS.COM, INC.,

       Defendant - Appellant.


                 Appeal from the United States District Court
                        for the District of New Mexico
                    (D.C. No. 1:15-cv-00605-JCH-SCY)


Seth E. Pierce of Mitchell Silberberg & Knupp, LLP, Los Angeles, California
(Gilbert S. Lee of Mitchell Silberberg & Knupp, LLP, Los Angeles, California,
and Andrew G. Schultz of Rodey, Dickason, Sloan, Akin & Robb, P.A.,
Albuquerque, New Mexico, with him on the briefs), for Defendant-Appellant.

Sid Childress, Sante Fe, New Mexico, for Plaintiff-Appellee.


Before KELLY, GORSUCH, and McHUGH, Circuit Judges.


GORSUCH, Circuit Judge.


      The Class Action Fairness Act promises a federal forum for proposed class

actions if (among other things) the amount “in controversy” exceeds $5 million.

28 U.S.C. § 1332(d)(2). But though it has made appearances in many and even
pretty ancient federal jurisdictional statutes, the term “in controversy” remains a

source of dispute for the parties before us and gives rise to this appeal. Affording

the phrase its traditional meaning, we find federal jurisdiction here beyond doubt.

      The case comes to us this way. In exchange for a monthly subscription fee,

Stamps.com lets customers print their own postage from home, saving them a trip

to a post office or the purchase of a self-metering device. But Ms. Hammond says

Stamps.com’s website disclosures didn’t clearly explain its monthly subscription

charges. She says she thought that Stamps.com would charge her only for those

months when she actually used its service, not every single month. The

representations on Stamps.com’s website, she alleges, were misleading and

amounted to an unlawful trade practice. Claiming that “many members of the

general public” were similarly deceived, Ms. Hammond seeks to pursue a class

action in New Mexico state court on behalf of everyone in the country who, like

her, called to cancel their subscriptions after “discovering” that Stamps.com “was

taking money from them” every month. Ms. Hammond alleges that this class

includes “hundreds or thousands of persons.” And while she doesn’t allege a total

damages amount, she contends that she is entitled to $300 in statutory damages

and that other members of the proposed class should “likely” receive damages of

$31.98, representing two monthly subscription charges ($15.99 x 2), based on her

estimate of how long customers could have reasonably failed to notice the




                                        -2-
monthly charges before calling to cancel. Ms. Hammond also seeks punitive

damages for herself and other class members.

      Unsurprisingly, Stamps.com sought to remove the case to federal court.

The company presented uncontested declarations showing that in the last four

years (corresponding to a likely statute of limitations period) at least 312,680

customers called to cancel their subscriptions. The company observed that, if

each of these persons were to win the same $300 in damages Ms. Hammond seeks

for herself, the value of this case would exceed $93 million. And even if other

class members could secure only $31.98 in damages, the company noted, the

case’s potential value would still lie at almost $10 million. Well above the $5

million threshold Congress ordained.

      All the same, the district court refused jurisdiction. It held that

Stamps.com failed to meet its burden of showing that over $5 million was “in

controversy.” The reason? The court faulted Stamps.com for failing to

disaggregate from the total number of customer cancellations those customers

who “felt duped” by Stamps.com’s website disclosures. As the district court

noted, customers could have cancelled their accounts because of “any of a myriad

of . . . reasons.” Not everyone was deceived. Put more pointedly, without proof

from Stamps.com establishing how many of its customers were actually deceived,

the district court thought the company couldn’t satisfy the $5 million “in

controversy” requirement.

                                         -3-
      This conclusion rests on a legal error about the meaning of a key statutory

term. In using the phrase “in controversy” CAFA borrowed a term heavily

encrusted with meaning. A term that’s long appeared in our federal diversity

statute and, indeed, traces its lineage all the way back to the Federal Judiciary Act

of 1789, if not beyond. See An Act to Establish the Judicial Courts of the United

States, ch. 20, 1 Stat. 73, 78 (1789) (codified as amended at 28 U.S.C. § 1332(a));

Thomas E. Baker, The History and Tradition of the Amount in Controversy

Requirement: A Proposal to “Up the Ante” in Diversity Jurisdiction, 

, 302-03 (1984). When Congress chooses to employ a term of legal art like

this we typically assume it is employing its accepted meaning. As Justice

Frankfurter put it, “if a word is obviously transplanted from another legal source,

whether the common law or other legislation, it brings the old soil with it.” Felix

Frankfurter, Some Reflections on the Reading of Statutes, 47 Colum. L. Rev. 527,

537 (1947); see also Antonin Scalia & Bryan A. Garner, Reading Law 73 (2012).

      That old soil reveals much here. As historically used, the term “in

controversy” has never required a party seeking to invoke federal jurisdiction to

show that damages “are greater” or will likely prove greater “than the requisite

amount” specified by statute. Hartis v. Chi. Title Ins. Co., 

, 944 (8th

Cir. 2012). Instead, the term has required a party seeking federal jurisdiction to

show only and much more modestly that “a fact finder might legally conclude”

that damages exceed the statutory amount. 

Supreme Court has

                                        -4-
explained, to justify dismissal under this standard “it must appear to a legal

certainty that the claim is really for less than the jurisdictional amount.” St. Paul

Mercury Indem. Co. v. Red Cab Co., 

, 289 (1938). And this court

has repeatedly “caution[ed] counsel and courts” against pursuing any other

understanding or test. Frederick v. Hartford Underwriters Ins. Co., 

, 1248 n.4 (10th Cir. 2012); see also McPhail v. Deere & Co., 

,

955 (10th Cir. 2008); Gibson v. Jeffers, 

, 220 (10th Cir. 1973);

14AA Charles Alan Wright et al., Federal Practice and Procedure § 3702, at 306-

08 (4th ed. 2011). Of course, all these expositions about the meaning of the term

“in controversy” have come in the course of interpreting earlier federal

jurisdictional statutes and it is at least conceivable Congress could have meant

something different in CAFA. Our presumption of consistent usage is just that, a

presumption. But we cannot think of — and the parties do not even attempt to

give us — any reason to suppose that in using the term “in controversy” Congress

in CAFA meant anything at odds with our traditional understanding.

      To know that much is to know how our case must come out. Applying the

traditional understanding of the term “in controversy” to the undisputed facts

before us yields no doubt: federal jurisdiction exists here as a matter of law.

After all and by everyone’s admission, actual damages run at least $31.98 and

perhaps $300 per person (before punitive damages). And, as everyone

acknowledges, at least 312,000 people cancelled their subscriptions during the

                                         -5-
class period. All this leads to the possibility that a jury might lawfully award

relief between nearly $10 million and $93 million. A legal possibility that is

more than enough to trigger federal jurisdiction.

      The district court employed a different and mistaken understanding of the

statutory term. It effectively required Stamps.com to prove that a factfinder

would (or probably would) find damages in excess of the statutory amount. In

reaching its decision to deny jurisdiction, the district court emphasized that it’s

unlikely Ms. Hammond will be able to show all 312,000 customers cancelled their

accounts because of the company’s alleged misrepresentations. For our part, we

don’t doubt that probabilistic judgment for a moment. The problem is that it

answers the wrong question. For the question at this stage in the proceedings

isn’t what damages the plaintiff will likely prove but what a factfinder might

conceivably lawfully award. And on that question — the only legally salient

question before us — no one has identified any legal impediment precluding a

jury from finding all 312,000 persons entitled to relief. At the end of the day,

“[e]ven if it is highly improbable that the Plaintiffs will recover the amounts

Defendants have put into controversy, this does not meet the legally impossible

standard.” Raskas v. Johnson & Johnson, 

, 888 (8th Cir. 2013).

(And though it’s beside the point, it is perhaps noteworthy that even Ms.

Hammond’s own probabilistic estimate about the likely outcome in this case

appears at odds with the district court’s, for she has volunteered her belief that

                                         -6-
maybe half of the 312,000 customers will ultimately prove entitled to recovery —

an estimate that would yield damages of between $4.98 million and $46 million

before punitive damages, and no one disputes punitive damages could exceed

$20,000.)

      Usually history offers its lessons for good reason and our case supplies no

exception. For the law’s traditional course here is not some empty formalism.

The jurisdictional test history suggests serves a useful purpose, helping to keep

cases from bogging down in mini-trials before they’ve even begun. At this stage,

we’re just trying to decide the forum for the dispute, not liability or damages.

And a more aggressive inquiry into the likelihood of success on the merits would

invite delays and costs more appropriately reserved for adjudicating the merits

than choosing the forum. Such an inquiry would, as well, force the proponent of

jurisdiction to argue against himself, tasking him with the job of proving his own

likely liability in a sufficient number of individual cases simply to get a foot in

the door of the federal courthouse. A result awfully unlikely on its face, one

made all the more so when (again) we’re unaware of any other arena in which the

term “in controversy” has been interpreted to exact such a heavy price for federal

jurisdiction. See 

(concluding that a party should not be

forced to concede liability to win jurisdiction); 

(same);

Lewis v. Verizon Commc’ns, Inc., 

, 400 (9th Cir. 2010) (same);

Spivey v. Verture, Inc., 

, 986 (7th Cir. 2008) (same).

                                         -7-
      Neither do Ms. Hammond’s rejoinders persuade us. She suggests that any

error here was merely factual, not legal, so our review should be more deferential.

But we do not question any fact found by the district court. For that matter and

again, we agree with the district court’s probabilistic judgment that it is very

unlikely all 312,000 persons who cancelled service during the relevant period will

prove entitled to damages in the end. We acknowledge, too, that in its recitation

of the applicable legal standards the district court cited the correct legal test for

jurisdiction (focusing on what’s legally possible, not what’s likely). The problem

is that in applying the law to the facts the court proceeded to employ a more

demanding test than the law allows, mistakenly focusing not on the legally

possible but the factually probable. And that is an error of law we are not free to

disregard.

      Alternatively, Ms. Hammond suggests we should wait. Once the class is

certified and the number of class members is readily apparent, she says, we’ll all

have a better idea of the number of individuals who relied on the company’s

alleged misrepresentations. So Stamps.com can and should simply proceed in

state court until that point and then remove the case to federal court. But in

adopting CAFA, Congress didn’t say that a federal forum should await class

certification or trial or some mini-trial concerning likely success on the merits

(something Congress certainly could have done if it wished: consider the

preliminary injunction context). To the contrary, Congress expressly provided

                                          -8-
that it’s enough to establish that over $5 million is “in controversy.” It provided,

too, that parties wanting to remove cases to federal court generally must do so

within 30 days after service of the complaint. 28 U.S.C. § 1446(b)(1). And if a

district court (as here) remands the case to state court, Congress granted the party

pursuing federal jurisdiction the right to an immediate appeal and a decision on

the merits of that appeal within 60 days. 28 U.S.C. § 1453(c)(1)-(2). So it is that

Congress chose a particular term and developed a comprehensive statutory

scheme that consistently bespeaks a wish to resolve the forum question early and

quickly in the litigation. For a court to delay a statutory entitlement to a federal

forum without legal justification is no more defensible than for a court to deny

that entitlement without legal justification. Our job is to abide Congress’s policy

directions, not replace them with others of our own hand.

      Holding for Ms. Hammond in this case would place us at odds not just with

the language and structure of the statute but with the considered judgment of

many other circuits. Already several courts have held as we do today, explaining

that federal jurisdiction under CAFA doesn’t depend on how much the plaintiff is

likely to recover but on the amount the plaintiff’s allegations suggest she might

lawfully recover. And neither, again, are we given any reason to doubt the

collective wisdom of our colleagues. It is our judgment, as it was theirs, that

“[o]nce the proponent of federal jurisdiction has explained plausibly how the

stakes exceed $5 million . . . the case belongs in federal court unless it is legally

                                          -9-
impossible for the plaintiff to recover that much.” 

; see

also 

; 

; 

.

      The remand order is vacated and the case is remanded for further

proceedings consistent with this opinion.




                                       - 10 -