[PUBLISH]

                  IN THE UNITED STATES COURT OF APPEALS
                                                                                FILED
                            FOR THE ELEVENTH CIRCUIT                  U.S. COURT OF APPEALS
                                                                        ELEVENTH CIRCUIT
                               ________________________                      JAN 04 2001
                                                                         THOMAS K. KAHN
                                      No. 99-12833                             CLERK
                                _______________________

                           D.C. Docket No. 97-00102-CV-D-S

CHESTER SMITH, individually and on behalf of all
others similarly situated, MERLE FISHER, individually
and on behalf of all others similarly situated, et al.,

                                             Plaintiffs counterdefendants-Appellants,
                                             Cross-Appellees,
       versus

GTE CORPORATION, GTE SOUTH, INC.,

                                             Defendants-counterclaimants-Appellees,
                                             Cross-Appellants.

                             __________________________

                     Appeals from the United States District Court
                         for the Middle District of Alabama
                           _________________________
                                 (January 4, 2001)


Before CARNES and BARKETT, Circuit Judges, and POLLAK*, District Judge.
_________________________
        * Honorable Louis H. Pollak, U.S. District Judge for the Eastern District of Pennsylvania,
sitting by designation.

CARNES, Circuit Judge:
      In this putative class action suit, the plaintiffs assert various state law claims

based on an alleged scheme by the defendants, GTE Corporation and GTE South,

Inc. (collectively “GTE”), to defraud their customers into leasing telephones and

paying exorbitant lease charges. GTE filed a motion to dismiss, contending that

the Alabama Public Service Commission (“APSC”), which regulates public

utilities operating in that state, has exclusive jurisdiction over the plaintiffs’ claims.

Alternatively, GTE argued APSC has primary jurisdiction over the claims and that

the district court should abstain until the plaintiffs’ claims were presented to and

reviewed by the APSC. Relying on the primary jurisdiction doctrine, the district

court concluded that the plaintiffs should first present their claims to the APSC and

for that reason dismissed the suit without prejudice. The plaintiffs appealed.

       We vacate the district court’s order and remand with directions that the case

be dismissed on the grounds that federal courts lack subject matter jurisdiction over

this state law case because there is an insufficient amount in controversy for

diversity jurisdiction to exist, and no federal law question in the complaint for

federal question jurisdiction to exist.




                                 I. BACKGROUND


                                            2
       The origin of this lawsuit lies in the deregulation of “customer premises

equipment” (“CPE”). GTE,1 in addition to providing telecommunications services,

leases telephones and related equipment, collectively referred to as CPE, to some

of its telecommunications services customers. Before 1988, the leasing activity of

GTE and other telecommunications providers, including the amount of the lease

rates, was subject to federal and state regulation. In the early 1980s, the Federal

Communications Commission (“FCC”) decided to deregulate the CPE activity of

these providers, thereby allowing them to compete freely with other non-

telecommunications providers in the market for CPE while the regulation of their

telecommunications services continued.

       As part of the deregulation plan, the FCC found it necessary to preempt state

regulation of CPE activity, but it allowed states to develop their own deregulation

plans provided that those plans were implemented by December 31, 1987. In the

Matter of Procedures for Implementing the Detariffing of Customer Premises

Equipment and Enhanced Services, 99 F.C.C.2d 354 (1984). Working with the

telecommunications providers in Alabama, the APSC followed the directive of the

FCC and achieved the deregulation of the providers’ CPE activity before 1988.


       1
         GTE Corporation is the holding company parent of GTE South, Inc., a local exchange
carrier which provides telecommunications services to customers in Alabama, among other
states.

                                              3
       In January of 1997, Chester Smith and three other Alabama residents filed

this putative class action lawsuit against GTE. According to the plaintiffs, after its

CPE activity had been deregulated, GTE offered to sell at “artificially high prices”

phones that were then being leased by its customers and treated a customer’s lack

of response as “an agreement to continue leasing,” which the plaintiffs refer to as

“an unlawful negative option.” GTE allegedly treated a customer’s lack of

response to the offer as “an agreement to continue leasing,” which the plaintiffs

refer to as “an unlawful negative option.” The plaintiffs further allege that during

the “Deregulation Period” – which they define as January 2, 1988 until the present

– GTE has carried out a “fraudulent scheme,” which includes charging its

customers exorbitant fees for leased telephones, concealing the existence and

amount of those charges, failing to inform customers they would be better off

purchasing phones from third parties, and in some instances, charging customers

for phones that no longer worked or had been returned to GTE.2




       2
          As an example of GTE’s alleged deceptive nature of lease charges, the plaintiffs state
that the charge for a leased phone was listed as “desk phone” under the heading of “local
services” and then listed under the heading of “GTE Basic Service” in the monthly bills sent to
customers, thereby implying that the charge was part of the regulated basic service and that non-
payment for the charge would result in disconnection of the customer’s phone service. They
contend that not until October 1996 did GTE first use the word “rental” to refer to the lease
charges.

                                                4
       In their amended complaint, the plaintiffs assert state law claims for fraud,

unjust enrichment, breach of contract, and breach of warranty. In addition, the

plaintiffs seek equitable relief in Count VI of their complaint, including

an injunction preventing GTE from misrepresenting its lease charges on monthly

bills and a declaration that the lease agreements for telephones are “null and void

from their inception.”

      The plaintiffs contend that diversity jurisdiction exists over their state law

claims, and they seek to certify the following two classes: (1) the “Damages

Class,” which consists of “all persons who presently reside in Alabama who have

leased telephone equipment for residential use from [GTE] at any point in time

between January 2, 1988 and the date of this suit,” and (2) the “Injunctive Class,”

the composition of which is identical to the “Damages Class” except that it also

includes residents of Kentucky, North Carolina, and Virginia. Although the

complaint does not allege the number of members in each of the proposed classes,

it does allege that “[i]n mid-1993 [GTE] leased telephone equipment to 36,065

residents of the state of Alabama.” Consequently, the “Damages Class” alone

consists of more than 36,000 members.

      On August 14, 1997, GTE moved for judgment on the pleadings. In its

motion, GTE argued that the APSC, which supervises and regulates public utilities


                                          5
in Alabama, had either exclusive or primary jurisdiction over the plaintiffs’ claims.

GTE requested that the district court dismiss the plaintiffs’ suit and effectively

require them to present their claims to the APSC. The motion was referred to the

magistrate court who ultimately recommended that the district court, under the

primary jurisdiction doctrine, stay the proceedings in the lawsuit until the

plaintiffs’ claims could be heard and decided by the APSC.3

       Agreeing that the APSC should exercise primary jurisdiction over the

plaintiffs’ claims, the district court adopted the recommendation of the magistrate

court except that it did not order a stay. Instead, the court dismissed the case

without prejudice to allow the plaintiffs to “assert their claims before the APSC,

with leave to any Party to move the court to reinstate the action on the active

docket of the court, if and when appropriate and necessary.” The plaintiffs

appealed.



       3
          “Primary jurisdiction is a judicially created doctrine whereby a court of competent
jurisdiction may dismiss or stay an action pending a resolution of some portion of the actions by
an administrative agency.” Wagner & Brown v. ANR Pipeline Co., 

, 201 (5th Cir.
1988). Even though the court is authorized to adjudicate the claim before it, the primary
jurisdiction doctrine “comes into play whenever enforcement of the claim requires the resolution
of issues which, under a regulatory scheme, have been placed within the special competence of
an administrative body; in such a case the judicial process is suspended pending referral of such
issues to the administrative body for its views.” United States v. Western Pac. R.R. Co., 

, 64, 

, 165 (1956) (explaining that, like an exhaustion of remedies
requirement, the primary jurisdiction doctrine “is concerned with promoting proper relationships
between the courts and administrative agencies charged with particular regulatory duties”).

                                                6
      On appeal, we raised the question of whether this case involved a sufficient

amount in controversy to establish federal diversity jurisdiction under 28 U.S.C. §

1332. The parties submitted supplemental briefing on that issue and addressed it at

oral argument. Both the plaintiffs and GTE contend that there is a sufficient

amount in controversy. However, we conclude that there is not, and that federal

question jurisdiction also is not present, with the result that the district court lacked

subject matter jurisdiction over the case.

                                  II. DISCUSSION

      “Federal courts have limited subject matter jurisdiction, or in other words,

they have the power to decide only certain types of cases.” Morrison v. Allstate

Indem. Co., 

, 1260-61 (11th Cir. 2000) (citing University of South

Alabama v. American Tobacco Co., 

, 409-10 (11th Cir. 1999)).

Lower federal courts can exercise this power only over cases for which there has

been a congressional grant of jurisdiction, see 

the Constitution

unambiguously confers this jurisdictional power to the sound discretion of

Congress, federal courts should proceed with caution in construing constitutional

and statutory provisions dealing with [their] jurisdiction.” University of South

(citations and internal marks omitted).




                                             7
        In this case, neither party has challenged federal court jurisdiction.

However, because a federal court is powerless to act beyond its statutory grant of

subject matter jurisdiction, a court must zealously insure that jurisdiction exists

over a case, and should itself raise the question of subject matter jurisdiction at any

point in the litigation where a doubt about jurisdiction arises. See Fitzgerald v.

Seaboard Sys. R.R., Inc., 

, 1251 (11th Cir. 1985) (“A federal court

not only has the power but also the obligation at any time to inquire into

jurisdiction whenever the possibility that jurisdiction does not exist arises.”); see

also 

(raising sua sponte on appeal the issue of

whether the case involved a sufficient amount in controversy for diversity

jurisdiction); Laughlin v. K-Mart Corp., 

, 873-74 (10th Cir. 1995)

(same).

       In their complaint, the plaintiffs allege that diversity jurisdiction exists over

this case, and neither the defendant nor the district court questioned that

allegation.4 Moreover, even after we raised the potential jurisdictional problem on

appeal, both parties maintained in their supplemental briefs that federal jurisdiction


       4
         When the plaintiffs filed their complaint, and when the district court dismissed the suit
on the basis of primary jurisdiction doctrine, some precedent in this circuit appeared to indicate
that a class claim for punitive damages could establish the requisite amount in controversy for
diversity jurisdiction over a class action. See infra Part II.A. That may be the reason that no one
questioned the district court’s jurisdiction.

                                                 8
exists over the case. But subject matter jurisdiction exists only where granted by

statute, and thus, a federal court is not bound by the jurisdictional contentions of

the parties. See Morrison,228 F.3d at 1261; Jackson v. Seaboard Coast Line R.R.

Co., 

, 1000-01 (11th Cir. 1982) (“The jurisdiction of a court over the

subject matter of a claim involves the court’s competency to consider a given type

of case and cannot be waived or otherwise conferred upon the court by the

parties.”).

       While both parties argue in their supplemental briefs that this case involves a

sufficient amount in controversy to sustain diversity jurisdiction, the plaintiffs also

raise the possibility that the FCC’s preemption of state regulation of CPE

establishes federal question jurisdiction over their claims, pursuant to 28 U.S.C. §

1331. Because the plaintiffs originally premised jurisdiction on the diversity of

citizenship, pursuant to 28 U.S.C. § 1332, and because the parties primarily focus

on that basis for jurisdiction, we will address the issue of diversity jurisdiction first.



                          A. DIVERSITY JURISDICTION

       The foundation for federal court diversity jurisdiction – the power to decide

cases between citizens of different states – is Article III of the United States

Constitution. See U.S. Const. art. III, § 2. However, when Congress created lower


                                            9
federal courts, it limited their diversity jurisdiction to cases in which there was a

minimum monetary amount in controversy between the parties. See Snyder v.

Harris, 

, 334, 

, 1056 (1969). Today, the threshold

amount in controversy for diversity jurisdiction, excluding interests and costs, is

$75,000. See 28 U.S.C. § 1332.

       Regarding the amount in controversy requirement, the plaintiffs allege in

their complaint the following: “Although the actual damages claimed by the named

Plaintiffs are less than $75,000.00, Plaintiffs also claim punitive damages against

[GTE]. Therefore, the matter in controversy exceeds, exclusive of interest and

costs, the sum of $75,000.00.”5 In making this allegation, the plaintiffs appear to

have relied on this Court’s holding in Tapscott v. MS Dealer Serv. Corp., 

(11th Cir. 1996), that a class claim for punitive damages under Alabama law

could be viewed in the “aggregate” for amount in controversy purposes, which

meant that the amount of punitive damages claimed could be attributed in toto to

each member of the class in order to establish diversity jurisdiction over the claims


       5
         The compensatory, or actual, damages of the class members appear to be relatively
small. According to the plaintiffs, the allegedly fraudulent and excessive lease rates charged by
GTE ranged from $3 to $10 a month. Over a twelve year period (1988 to the present), the class
members would have paid approximately between $440 and $1,400 in lease charges, well below
the $75,000 required for diversity jurisdiction. Moreover, the compensatory damages claims, as
well as the unjust enrichment claim, of the class members arise from each plaintiff’s individual
agreements with GTE, and thus, those claims may not be aggregated for amount in controversy
purposes. See 

.

                                               10
of the entire class. See 

However, after the district court in this

case dismissed the plaintiffs’ suit and they appealed, this Court decided Cohen v.

Office Depot, Inc., 

(11th Cir. 2000) (“Cohen II”). In Cohen II, we

held that a prior panel decision by the Former Fifth Circuit, Lindsey v. Alabama

Tel. Co., 

(5th Cir. 1978),7 precluded aggregating punitive damages to

establish diversity jurisdiction over a class action, and that decision is to be

followed notwithstanding a contrary holding by the subsequent panel in Tapscott.

See Cohen 

.8

       6
           Generally, when plaintiffs join in one lawsuit, the value of their claims may not be
added together, or “aggregated,” to satisfy the amount in controversy requirement for diversity
jurisdiction. See Zahn v. International Paper Co., 

, 295, 

, 508 (1974);

, 89 S.Ct. at 1056. However, “when several plaintiffs unite to enforce a
single title or right, in which they have a common and undivided interest, it is enough if their
interests collectively equal the jurisdictional amount.” 

, 94 S.Ct. at 508
(distinguishing between the “common and undivided” interest, for which aggregation is
permissible, and “separate and distinct” interests, for which aggregation is not permissible)
(emphasis added) (quoting Troy Bank of Troy, Indiana v. G.A. Whitehead & Co., 

,
40-41, 

(1911)).
       7
         Decisions of the Fifth Circuit issued prior to October 1, 1981 are binding precedent on
this Court. See Bonner v. City of Prichard, 

, 1207 (11th Cir. 1981) (en banc).
       8
          Under the well-established prior panel precedent rule of this Circuit, the holding of the
first panel to address an issue is the law of this Circuit, thereby binding all subsequent panels
unless and until the first panel’s holding is overruled by the Court sitting en banc or by the
Supreme Court. See Cargill v. Turpin, 

, 1386 (11th Cir. 1997).

        In the initial panel decision in Cohen v. Office Depot, Inc., 

(11th Cir.
1999) (“Cohen I”) this Court reversed the district court’s order of dismissal for lack of subject
matter jurisdiction. In that case, after the district court had stricken the Florida plaintiffs’
punitive damage claim because it had not been properly pled in accordance with Fla. Stat. §
768.72, the court concluded that the plaintiff’s class action suit did not satisfy the amount in
controversy requirement. See 

On appeal, the Court held in Cohen I that Fla. Stat. §

                                                11
       In light of Cohen II and Lindsey, which require that any class claim for

punitive damages under Alabama law be divided pro rata among all of the class

members for amount in controversy purposes, we put to the parties the question of

whether the present case involves a sufficient amount in controversy for purposes

of diversity jurisdiction. They responded that, despite the decisions in Cohen II

and Lindsey, punitive damages may still be viewed in the aggregate to satisfy the

requisite amount in controversy. In addition, the plaintiffs contend that their claim

for attorney’s fees and the value of the requested injunctive relief may be viewed in

the aggregate, thereby providing additional bases for establishing a sufficient

amount in controversy. We address their contentions in that order.

                                     1. Punitive Damages

       In arguing that Tapscott continues to be controlling authority on the issue of

aggregating a class claim for punitive damages, the parties primarily rehash the

same arguments considered and rejected by this Court in Cohen II. They maintain



768.72 did not apply to cases filed in federal court, see 

and relying on Tapscott,
concluded that the punitive damages claim could be aggregated to satisfy the amount in
controversy requirement for diversity jurisdiction. See 

         The defendants in Cohen filed a petition for rehearing in which the Lindsey decision was
“belatedly” brought to the Court’s attention. See Cohen 

. Concluding that
Lindsey was controlling on the punitive damages issue under our prior panel precedent rule, we
held in Cohen II that the punitive damages claim could not be aggregated, and we affirmed the
district court’s dismissal for lack of subject matter jurisdiction. See Cohen 

-
77, 83.

                                                 12
that no conflict exists between Lindsey and Tapscott because the Lindsey Court did

not address the same issue as the Tapscott Court – whether class members have a

“common and undivided interest” in a claim for punitive damages under Alabama

law, thereby permitting a class claim for punitive damages to be viewed in the

aggregate for amount in controversy purposes. See 

, 89

S.Ct. at 1056; see also Zahn v. International Paper Co., 

, 294, 

, 508 (1974) (explaining that “when several plaintiffs unite to enforce a single

title or right, in which they have a common and undivided interest,” their claims

may be viewed in the aggregate to satisfy the amount in controversy requirement)

(citations omitted).

      According to the parties, the Lindsey Court simply assumed that punitive

damages could not be viewed in the aggregate under the Supreme Court’s decision

in Snyder, and failed to consider the “common and undivided interest” exception to

the general rule prohibiting aggregation acknowledged by the Supreme Court in

Snyder, as well as in its later decision in Zahn. Because the Tapscott Court

subsequently concluded that Alabama plaintiffs do have a common and undivided

interest in a class claim for punitive damages, the plaintiffs contend that holding

from Tapscott is the controlling law on the issue of whether such damages can be




                                          13
viewed in the aggregate. We rejected that very contention in Cohen 

, and we are bound by Cohen II to reject it again here.

      Recently, in H & D Tire and Automotive-Hardware, Inc. v. Pitney Bowes,

Inc., 

, 329-30 (5th Cir. 2000), the Fifth Circuit addressed a similar

conflict between Lindsey and the subsequent panel decision in Allen v. R&H Oil &

Gas Co., 

, 1329 (5th Cir.1995). Concluding that the result in Lindsey

implicitly requires that a class claim for punitive damages be distributed pro rata

among the class members for amount in controversy purposes, which is the same

conclusion we reached in Cohen II, the Fifth Circuit in Pitney Bowes adhered to

the prior panel precedent rule and held that a class claim for punitive damages

could not be viewed in the aggregate to satisfy the requisite amount in controversy.

See 

words, the Fifth Circuit in Pitney Bowes reached the same

conclusion about Lindsey that we did: Lindsey establishes binding circuit law

against aggregation of punitive damages for amount in controversy purposes.

      Like the plaintiff in Cohen II, the parties in this case are arguing simply that

the decision in Lindsey was wrong because, according to them, class members

have a “common and undivided interest” in a class claim for punitive damages.

See Cohen 

. But “[e]ven if we thought Lindsey wrong, the

prior panel precedent rule is not dependent upon a subsequent panel’s appraisal of


                                          14
the initial decision’s correctness.” Id.; see also United States v. Steele, 

, 1317-18 (11th Cir. 1998) (en banc) (“Under our prior precedent rule, a panel

cannot overrule a prior one’s holding even though convinced it is wrong.”)

(citations omitted). Simply put, Lindsey precludes viewing a punitive damages

claim in the aggregate for amount in controversy purposes, and because that panel

was the first in this Circuit to address the issue, Lindsey is the law of this Circuit.

See Cohen 

-77; see also Pitney 

(“Because Lindsey is the earliest, and thus controlling, decision in this circuit, the

punitive damages claims of the putative class cannot be aggregated and attributed

to each plaintiff to meet the jurisdictional requirement.”).9

       Going beyond their argument that no conflict exists between Lindsey and

Tapscott, the parties contend that even if there is such a conflict, an exception to

the prior panel precedent rule exists where the first panel to address an issue failed

to follow and apply controlling Supreme Court precedent. In support of this



       9
         The plaintiffs point out that the Tapscott Court, in considering the nature of punitive
damages under Alabama law in or to conclude that the class members had a common and
undivided interest in those damages, relied on Alabama Supreme Court cases that were decided
after Lindsey. But there is no indication in Lindsey that the Court based its decision on the
particular nature of Alabama punitive damages. See Pitney 

(“Lindsey's
reasoning did not rely on a characterization of punitive damages under Alabama law ... .”).
Thus, there is no basis for concluding that the subsequent Alabama decisions relied on by
Tapscott, or any subsequent Alabama decision addressing punitive damages issues, have
undermined or “overruled” Lindsey.

                                               15
contention, GTE points to Tucker v. Phyfer, 

(11th Cir. 1987), in

which the panel expressly reached a holding contrary to that of a previous panel.

See 

n.7. In doing so, the Tucker Court noted that the prior panel had

not referred to two Supreme Court cases that it thought compelled a different

result. See 

follow the prior panel’s holding, the Tucker Court

opined that if the two Supreme Court decisions had “been called to the attention of

the [prior] panel, the panel would have come to the conclusion we reach today.”

contend that the “exception” to the prior panel precedent rule

recognized in Tucker applies here and prevents Lindsey from binding subsequent

panels on the issue of aggregating punitive damages. Because the Lindsey Court

did not discuss the general aggregation standard noted in Snyder, which allows

aggregation of the claims of multiple plaintiffs to satisfy the amount in controversy

requirement when the plaintiffs assert a “single title or right in which they have a

common and undivided interest,” 

, 89 S.Ct. at 1056, the

parties argue that the Lindsey Court failed to follow the Supreme Court’s decision

in Snyder, and thus, that this Court’s subsequent decision in Tapscott controls.

That position is wrong from the beginning. In explaining why, we will start with

the supposed exception to the prior panel precedent rule.


                                          16
      If such an “exception” truly existed, it could end up nullifying the well-

established prior panel precedent rule that is an essential part of the governing law

of this Circuit. See Phillip M. Kannan, The Precedential Force of Panel Law, 76

Marq. L. Rev. 755, 763 (1993) (noting a similar exception to the prior panel

precedent rule for “serious or egregious errors” by the prior panel and explaining

that such an exception “is a prescription for rule-swallowing if ever there was

one”). To the extent Tucker supports such an exception, it is itself inconsistent

with prior precedent. In United States v. Bascaro, 

, 1343 (11th Cir.

1984), which preceded Tucker, we emphatically held that “the mere act of

proffering additional reasons not expressly considered previously...will not open

the door to reconsideration of the question by a second panel.” Bascaro involved

an issue of

statutory interpretation, but there is no principled basis for distinguishing between

that and the interpretation of governing decisions.

      Moreover, if there ever was an exception to the prior panel precedent rule

such as that expressed in Tucker, it did not survive the pronouncement of the en

banc Court in United States v. Steele, 

(11th Cir. 1998) (en banc),

that: “Under our prior precedent rule, a panel cannot overrule a prior one’s holding

even though convinced it is wrong.” 

- 18.


                                          17
       The idea of an exception to the prior panel precedent rule where a

subsequent panel is convinced the prior one reached the wrong result – for

whatever reason – is also inconsistent with a number of decisions in which panels

of this Court have obediently followed prior panel precedents they were convinced

were wrong. See, e.g., In re Dickerson, 

, 926 (11th Cir.

2000)(following the holding of an earlier panel decision even though “were we to

decide this issue on a clean slate, we would not so hold.”); United States v. Steele,

, 1234-35 (11th Cir. 1997) (reluctantly following a prior panel

decision even though convinced it conflicted with the plain language of the statute

in question);10 Turecamo of Savannah, Inc. v. United States, 

, 1087

(11th Cir. 1994) (following precedent thought to be wrong even though the prior

panel apparently overlooked important legislative history that would have led to a

different result); see also Wascura v. Carver, 

, 687 (11th Cir.

1999)(responding to the argument that the reasoning of a prior panel decision “is

unclear and inadequate to support its holding” by stating that “[w]e have no


       10
          The erroneous result reached by the prior panel whose decision bound the Steele panel
was corrected en banc, United States v. Steele, 

(11th Cir. 1998) (en banc), which
is how erroneous panel decisions should be corrected. See 

Before changing circuit
law, as it had the authority to do, the en banc court in Steele recognized that the panel in that
case had been bound to follow the prior panel decision even though it was wrong. See 

(the panel reached the result it did “because it was bound to do so by the prior panel
decision,” but “[b]ecause we are sitting en banc, we are not bound by the [prior panel]
decision.”).

                                               18
occasion to pass on that criticism, because we are bound by the [prior panel]

decision regardless of whether we agree with it.”).

       Permitting an “overlooked reason” exception would undermine the values

of stability and predictability in the law that the prior panel precedent rule

promotes. See Bonner v. City of Pritchard, 

, 1209-10 (11th Cir.

1981) (en banc); see also Jaffree v. Wallace, 

, 1533 (11th Cir.

1983)(“Judicial precedence serves as the foundation of our federal judicial system.

Adherence to it results in stability and predictability.”). For all of these reasons,

we categorically reject any exception to the prior panel precedent rule based upon a

perceived defect in the prior panel’s reasoning or analysis as it relates to the law in

existence at that time.11

       Even if the Tucker exception were engrafted onto the controlling law of this

circuit – instead of being rejected by it – that would do these parties no good. In

both Snyder and Zahn, the Supreme Court rejected attempts to aggregate claims of

class members, and there is no intimation in either of those decisions that members

of a class have a common and undivided interest in a class claim for punitive

       11
         Our use of the limiting phrase “as it relates to the law in existence at that time” is
deliberate. Subsequent panels are not bound by prior decisions where there has been a change in
the controlling law as a result of a subsequent en banc or Supreme Court decision or statutory
change. See United States v. Hanna, 

, 1288 (11th Cir. 1998)(“In this circuit, only
the court of appeals sitting en banc, an overriding United States Supreme Court decision, or a
change in the statutory law can overrule a previous panel decision.”).

                                              19
damages that would permit aggregating that claim. Thus, unlike the situation in

Tucker, there was no “clearly controlling Supreme Court precedent” on the issue of

aggregating punitive damages when Lindsey was decided. See 

n.7.

      Moreover, in attempting to justify the exception in Tucker, the panel in that

case noted that the prior panel whose decision it questioned had failed to mention

either of the two controlling Supreme Court decisions. See 

critical to

the Tucker panel, which said: “We hasten to add that had the [prior] panel

expressly considered [the two controlling Supreme Court decisions], we would be

bound by its interpretation and application of those decisions.” 

the Lindsey Court cited Snyder for the proposition that the claims of class

members may not be aggregated to satisfy the amount in controversy requirement,

and it also cited the Supreme Court’s subsequent decision in Zahn. See 

(citations omitted). These citations indicate that the Lindsey Court

was well aware of the “common and undivided interest” standard for aggregation

acknowledged in Snyder and Zahn but that it construed Snyder to preclude

aggregation of a class claim for punitive damages. So, even if the Tucker

exception were viable, it would have no application here.




                                         20
      As recently noted by the Fifth Circuit, “Lindsey's reasoning did not rely on a

characterization of punitive damages under Alabama law, but was instead based on

the principle that ‘the claims of several plaintiffs, suing as members of a class,

cannot be aggregated for the purpose of satisfying th[e] jurisdictional predicate.’”

Pitney 

(quoting 

) (citing Snyder

v. Harris, 

, 

(1969)); see also Ard v. Transcontinental

Gas Pipe Line Corp., 

, 601 (5th Cir. 1998) (“Lindsey therefore applies

Snyder’s reasoning that compensatory damage claims cannot be aggregated for

jurisdictional purposes to the context of punitive damage claims.”). The fact that

the Lindsey Court did not explicitly say “there is no common and undivided

interest in a class claim for punitive damages” does not undermine the clear

reasoning and result of that decision nor does it vitiate the resulting rule.

      In summary, the parties’ alternative argument boils down to the position that

Lindsey incorrectly interpreted and applied Snyder because, as the parties contend,

class members have a common and undivided interest in a claim for punitive

damages. The prior panel precedent rule clearly forecloses their position. See

Cohen 

; 

. Therefore, the class

claim for punitive damages under Alabama law may not be viewed in the aggregate




                                           21
for amount in controversy purposes, but instead must be divided pro rata among

each class member.

        As we have mentioned, supra note 5, the individual compensatory

damages of each class member appear to be, at most, approximately $1,400. That

means in order to satisfy the $75,000 amount in controversy requirement, each

class member would have to recover over $73,000 in punitive damages. For that to

be possible, the class, which appears to consist of at least 36,000 members, would

have to recover over $2.628 billion in punitive damages. That recovery is not

possible as a matter of law, see generally BMW of North America, Inc. v. Gore,

, 

(1996), and so, diversity jurisdiction cannot be

founded upon the claim for punitive damages.12


        12
           The named plaintiffs state that if class certification is denied, they will proceed in an
individual lawsuit, and in such a suit, they may realistically recover an amount of punitive
damages, especially when coupled with a recovery of attorney’s fees and the value of the
requested injunctive relief, that satisfies the amount in controversy requirement. Perhaps, but
this suit was filed as a class action and until the class certification is denied, we must treat it as a
class action. See Morrison, 228 F.3d at1263 n.7 (citing 3B J. Moore, Moore's Federal Practice, ¶
23.50 (2d ed. 1985)) ("In the interim between the commencement of the suit as a class action and
the court's determination as to whether it may be so maintained it should be treated as a class
suit."). If there is no diversity jurisdiction over a putative class action, it is a waste of the
resources for both the parties and the court to proceed through a class certification determination
on the basis that diversity jurisdiction would exist if the suit had been filed as an individual
action. See University of South Alabama,168 F.3d at 410 (“A necessary corollary to the concept
that a federal court is powerless to act without jurisdiction is the equally unremarkable principle
that a court should inquire into whether it has subject matter jurisdiction at the earliest possible
stage in the proceedings.”).



                                                  22
                                  2. Attorney’s Fees

      In the injunctive relief count of their complaint, the plaintiffs also ask the

court to ward them attorney’s fees. Alabama follows the “American Rule,” which

generally requires each party to pay its own attorney’s fees. See Ex Parte Horn,

, 702 (Ala. 1998). And “when there is no direct legal authority for

an attorney’s fee, a request for a fee [may not] be included in the computation o[f]

the jurisdictional amount.” GALT G/S v. JSS Scandinavia, 

, 1155

(9th Cir. 1998) (quoting Charles Alan Wright & Arthur R. Miller, Federal Practice

and Procedure § 3712, at 178 (3d ed. 1985)).

      There is an exception to the “American Rule,” and thus, “direct legal

authority” supporting inclusion of a claim for attorney’s fees in determining the

amount in controversy, when an award of fees is authorized either by statute or

contract. See Graham v. Henegar, 

, 736 n.9 (5th Cir. 1981) (citations

omitted). In this case, the plaintiffs do not claim a statutory or contractual right to

attorney’s fees. But a clearly established common law basis for awarding

attorney’s fees may also justify including a reasonable estimate of requested

attorney’s fees in determining the amount in controversy. See Ross v. Inter-Ocean

Ins. Co., 

, 661 (7th Cir. 1982) (noting that a reasonable estimate of

attorney’s fees may be included in the amount in controversy “where a litigant has


                                           23
a right, based on contract, statute, or other legal authority, to an award of attorney’s

fees if he prevails in the litigation”); see also 14B Charles Alan Wright & Arthur

R. Miller, Federal Practice and Procedure § 3702, at 92-93 & n.83 (3d. ed. 1985).

       As noted by this Court in Davis v. Carl Cannon Chevrolet-Olds, Inc., 

(11th Cir. 1999), Alabama recognizes two closely related equitable

doctrines providing for an award of attorney’s fees: the “common fund” doctrine

and the “common benefit” doctrine.13 See 

Under the common fund

doctrine, when litigation generates a fund benefitting a class of individuals, as in a

typical class action, the court may award fees to the plaintiff’s attorney, usually by

deducting a percentage from the fund in order to compensate the plaintiff’s

attorney. See Union Fidelity Life Ins. Co. v. McCurdy, 

, at * 2

(Ala. Sept. 22, 2000); see also Edelman & Combs v. Law, 

, 958

(Ala. 1995) (“[T]his Court, like the federal courts, has long recognized that a

lawyer who recovers an award for the benefit of a class of clients is entitled to a

reasonable fee from the amount recovered.”) (citations omitted). The common

benefit doctrine, under Alabama law, allows a court to require the defendant to pay

attorney’s fees, regardless of whether a fund has been generated by the litigation,

       13
          “In an ordinary diversity case where the state law does not run counter to a valid federal
statute or rule of court, and usually it will not, state law denying the right to attorney’s fees or
giving a right thereto . . . should be followed.” Alyeska Pipeline Service Co. v. Wilderness
Society, 

, 1622 n.31 (1975) (citations and quotations omitted).

                                                24
when the litigation has “conferred some kind of benefit on the public.” 

(citing Brown v. Alabama, 

, 592 (Ala. 1990)); see also

(indicating that fees may be awarded under the common

benefit doctrine when “the efforts of the plaintiff’s attorneys render a public

service or result in a benefit to the general public in addition to serving the interests

of the plaintiff”).

       In this case, the plaintiffs argue that they will be entitled to an award of

attorney’s fees under either the common fund doctrine or the common benefit

doctrine. In Davis, a diversity suit involving Alabama state law, this Court held

that a requested award of attorney’s fees deducted from the recovery of a common

fund could not be viewed in the aggregate to satisfy the amount in controversy

requirement for a class action. See 

. Thus, if the

common fund doctrine applies, the estimated amount of attorney’s fees in this case,

like the claim for punitive damages, must be divided pro rata among each class

member. Because the amount attributed to each of the more than 36,000 class

members would be well below $75,000, the attorney’s fee claim under the

common fund doctrine is not enough to establish diversity jurisdiction in this case.

       We now turn to the issue of whether the plaintiffs would be entitled to

attorney’s fees under the common benefit doctrine of Alabama and, if so, whether


                                           25
the estimated amount of those fees should be viewed in the aggregate for amount in

controversy purposes. In their complaint, the plaintiffs allude to the common

benefit doctrine as the basis for their attorney’s fees claim by requesting “a

reasonable attorney’s fee for the Plaintiffs for securing relief which will benefit the

general public and a large group of persons.” According to the plaintiffs, their

lawsuit against GTE will produce the type of public benefit for which Alabama

courts have awarded attorney’s fees under the common benefit doctrine because,

they say, their lawsuit will: (1) enjoin GTE’s alleged practice of targeting

minorities, the elderly, the poor, and similar demographic groups for its deceptive

leasing scheme; (2) enjoin unlawful practices affecting all of the residents of

communities in which GTE exercises a monopoly on telephone service; and (3)

determine if, and to what extent, the APSC has retained the authority to supervise

or regulate the CPE activity of telecommunications providers. It follows,

according to the plaintiffs, that their suit will result in substantial benefits to a large

group of persons, thereby justifying an award of common benefit attorney’s fees.

       In order to determine whether plaintiffs are correct, it is necessary to trace

the origins of the common benefit doctrine and its evolution under Alabama law.

The Supreme Court of Alabama first recognized the common benefit doctrine in

Miles v. Bank of Heflin, 

(1977). There, a group of shareholders


                                            26
brought suit against their corporation seeking an order allowing them to audit the

books and records of the corporation. See 

After holding that the

shareholders were so entitled, the Court turned to the question of whether or not to

award attorney’s fees to the plaintiffs on the grounds that they bestowed a non-

pecuniary benefit on the corporation and its other shareholders by bringing suit to

enforce shareholder rights. See 

The Court first noted that there “is no

inherent right to have an award of attorney’s fees in the absence of contract,

statute, or a recognized ground of equity.” 

recognized one such

equitable ground, holding that the plaintiffs, by bringing suit to enforce the rights

of shareholders, could be entitled to reimbursement for expenses incurred in the

course of the litigation, but only if they bestowed a substantial benefit on the

corporation. See 

Coupounas v. Morad, 

, 804 (Ala.

1980) (stating, “[i]t is an established rule of law that a minority stockholder is

entitled to be awarded a reasonable attorney’s fee by the corporation in which he

has an interest when a benefit has been conferred upon that corporation by the

stockholder’s derivative action.”). In Heflin, the plaintiffs were not reimbursed for

their litigation expenses because the Court decided that no such benefit had been

conferred by the plaintiffs’ suit.




                                          27
      In reaching its conclusion in Heflin that the bestowment of a substantial

benefit upon the corporation may entitle the plaintiffs to attorney’s fees, the

Alabama Supreme Court relied primarily on Mills v. Electric Auto-Lite Co., 

, 

(1970). Like Heflin, the Mills case was also a shareholder

derivative action. In regards to the fair apportionment of attorney’s fees, the Court

in Mills stated, “[t]he dissemination of misleading proxy solicitations was a deceit

practiced on the stockholders as a group, and the expenses of petitioners’ lawsuit

have been incurred for the benefit of the corporation and the other shareholders.”

(internal citations and quotations omitted). The Court concluded that the

fact the benefit conferred was non-pecuniary was of no moment, stating, “courts

increasingly have recognized that the expenses incurred by one shareholder in the

vindication of a corporate right of action can be spread among all shareholders

through an award against the corporation, regardless of whether an actual money

recovery has been obtained.” 

      The decisions and opinions in Heflin and Mills teach that the purpose of the

common benefit doctrine is to spread the cost of the litigation, including attorney’s

fees, among those who have benefitted from the litigation. It is not so much fee-

shifting as fee-spreading. This fee-spreading is accomplished, in the case of a

corporate defendant in a derivative suit, by assessing costs against the defendant


                                           28
corporation directly, because those costs are, in effect, incurred by the

shareholder/beneficiaries. See 

(explaining that the common

benefit doctrine is an extension of the common fund doctrine, but that in cases

involving the former “there was a ‘common fund’ only in the sense that the court’s

jurisdiction over the corporation as a nominal defendant made it possible to assess

fees against all of the shareholders through an award against the corporation” ).

Fee-spreading in such circumstances comports with the underlying purpose of both

the common fund and common benefit doctrines, which “is not to saddle the

unsuccessful party with the expenses but to impose them on the class that has

benefitted from them and that would have had to pay them had it brought the suit.”

See also Hall v. Cole, 

n.7, 

, 1946 n.7 (1973)

(noting that rationale underlying the common benefit doctrine represents an

extension of the common fund doctrine).

      Thus, under the common benefit doctrine, the focus is on the nature of the

relationship between the defendant and the class members, or, in the absence of a

certified class, those individuals who can be said to benefit from the litigation. If

the relationship is such that the costs can be spread indirectly to the beneficiaries

by imposing them on the defendant, then the relationship permits application of the

doctrine. As explained in Campbell v. General Motors Corp., 

                                          29
(N.D. Ala. 1998), “[g]enerally, federal courts have carefully hemmed in the

common benefit doctrine, applying it only where the fees, though paid for by the

defendant, are somehow distributed over those benefitted by the plaintiffs’

actions.” 

n. 9 (emphasis and citations omitted). As examples, the

Campbell court cited 

, 93 S.Ct. at 1945-48, in which the

United States Supreme Court permitted an award of attorney’s fees against a union

defendant where the suit had resulted in non-monetary benefits to the union

members, and Barton v. Drummond Co., 

, 982-85 (5th Cir. Unit B

1981), in which this Court permitted an award of attorney’s fees against a

corporation when the suit resulted in non-monetary benefits for its stockholders.

See 

n.9; see also 

, 90

S.Ct. at 628.

      Since its decision in Heflin, the Alabama Supreme Court has awarded

attorney’s fees under the common benefit doctrine in the context of only one

particular sort of relationship – that of citizens and their governmental entity. See

generally Brown v. State, 

, 591-92 (Ala. 1990); Ex Parte Horn, 

, 706 (Ala. 1998). In such lawsuits, the governmental entity is the

defendant and the citizens of that entity are the incidental beneficiaries of the suit.

Although federal courts have yet to construe the common benefit doctrine quite so


                                           30
broadly,14 application of the doctrine to the citizen/government relationship is

entirely consistent with its theoretical underpinnings. The relationship of citizens

to a governmental entity is such that imposing litigation costs on the defendant,

payable from the government’s coffers, is, in effect, obtaining those fees from the

citizen beneficiaries of the litigation.15

       Although the Alabama Supreme Court has never explicitly based its

common benefit analysis on the nature of the relationship between a defendant and

the beneficiaries of the lawsuit, that rationale has clearly animated the court’s

decisions. For example, in Ex Parte Horn, 

(Ala. 1998), the question

was whether the plaintiffs had conferred a sufficiently ‘public’ benefit that it was

equitable to impose attorney’s fees on the city of Birmingham. The plaintiffs had

successfully brought an action preventing the construction of a sanitary waste

transfer station adjacent to their neighborhood in the western part of the city. See



       14
           See 

n.39 (noting that in the Court’s previous common
benefit cases, application of the doctrine was limited to cases where the benefits of the litigation
“could be traced with some accuracy, and there was reason for confidence that the costs could
indeed be shifted with some exactitude to those benefitting. In this case however, sophisticated
economic analysis would be required to gauge the extent to which the general public, the
supposed beneficiary . . . would bear the costs.”).
       15
         That is not to say that the Alabama Supreme Court would only apply the common
benefit doctrine to governmental defendants. The relationship between a defendant who is a
governmental entity and the citizens who stand to benefit from the lawsuit brought against that
defendant is probably not the only type of relationship where application of the common benefit
doctrine is appropriate.

                                                 31


The city argued that “the facility would have had no impact on residents

of eastern Birmingham and, thus, . . . a benefit was not conferred on all the

residents of Birmingham in the same manner as it was conferred on [the

plaintiffs].” 

The underlying premise of the city’s argument was

that the costs of the litigation would be shouldered by all of the Birmingham

residents if the city was forced to pay, and that it would be inequitable to impose

those costs on the entire city of Birmingham because all of its residents did not

benefit equally from the litigation. The Alabama Supreme Court ultimately

rejected the city’s argument, reasoning that “the plaintiffs’ efforts clearly resulted

in an increased level of due process protection to all residents of Birmingham ...

and all of its residents have received a common benefit . . . .” 

What

matters for purposes of our analysis, though, is that in Horn the Alabama Supreme

Court accepted the underlying legal premise of the city’s argument – application of

the common benefit doctrine turns on whether requiring the defendant to pay the

plaintiffs’ litigation costs would spread those costs equitably among the

beneficiaries of the lawsuit.

      Returning to the facts of our case, the relationship between GTE and the

proposed beneficiaries of this lawsuit does not warrant application of the common

benefit doctrine. Imposing on GTE litigation costs, including attorney’s fees,


                                          32
would not operate to spread those costs among the purported beneficiaries of the

litigation, the class members. Those beneficiaries do not have any apparent

economic or other connection with GTE – like shareholders to a corporation, union

members to a union or citizens to a governmental entity – that would result in those

costs being shared by the class members. Even assuming that this lawsuit will

confer a benefit on the general public, requiring GTE to pay attorney’s fees under

the common benefit doctrine would not serve the doctrine’s purpose of spreading

the costs of this litigation among those beneficiaries. Instead, it would merely

saddle GTE as the unsuccessful defendant with those costs, in clear contravention

of the American Rule.

       Therefore, we conclude that any award of attorney’s fees in this case would

be paid out of the common fund created by the litigation and not by GTE on the

basis of a common benefit theory. An award of common fund attorney’s fees may

not be viewed in the aggregate, see 

, and from that it

follows the requested attorney’s fees in this case do not establish the requisite

amount in controversy for diversity jurisdiction.16

       16
         The plaintiffs in this case have also asserted claims of fraud against GTE. The Alabama
Supreme Court has, on one occasion, recognized an exception to the American Rule where
“fraud, willful negligence or malice has been practiced.” See Reynolds v. First Alabama Bank of
Montgomery, N.A., 

, 1242-43 (Ala. 1985). However, the Alabama courts have
apparently not extended the application of Reynolds beyond the facts of that case. Notably, the
Reynolds court emphasized that case involved fraud on the part of a trustee which caused losses

                                              33
                                      3. Injunctive Relief

       For amount in controversy purposes, “the value of the requested injunctive

relief is the monetary value of the benefit that would flow to the plaintiff if the

injunction were granted.” Cohen 

(discussing Ericsson GE

Mobile Communications, Inc. v. Motorola Communications & Elecs., Inc., 

, 218-20 (11th Cir. 1997)). In this case, the plaintiffs seek to enjoin GTE

from concealing or misrepresenting its lease charges and to require GTE to notify

its customers that failure to pay the lease charges cannot result in termination of

telephone service. Essentially, the monetary value of the injunction to the

plaintiffs is the present value of the future lease charges that the class members

could avoid by being clearly informed of the allegedly excessive cost of those

charges and of the fact that continuation of their telephone service is not contingent

on paying those charges.

       As to any individual class member, that monetary benefit would be well

below $75,000. For example, a class member who paid the highest current



to the individual trusts of the class members, and thus, that the suit was “essentially an equitable
proceeding.” 

Moreover, the plaintiffs have not cited, and this Court has not found,
any decision in the fifteen years since Reynolds where an Alabama appellate court has shifted
attorney’s fees to a defendant because it had committed fraud. For these reasons, we conclude
that application of the “fraud” exception alluded to in Reynolds is too speculative to serve as a
basis for including an award of attorney’s fees in determining the amount in controversy in this
case.

                                                 34
monthly lease rate for a telephone (about $10) for even as long as sixty years

would end up paying only $7,200, and the present value of that amount would be

far less. Of course, if the value of the injunction could be aggregated, given the

size of the class the injunctive relief sought in this case would satisfy the amount

in controversy requirement. But as explained recently in the Morrison decision,

aggregation depends on the rights asserted by the plaintiffs and not the particular

type of relief sought. See 

.

      Aggregation is permissible only when multiple plaintiffs seek to “enforce a

single title or right, in which they have a common and undivided interest,” 

, 94 S.Ct. at 508, and thus, “when an injunction protects rights that

are separate and distinct among the plaintiffs, the value of the injunction to the

individual plaintiffs may not be aggregated to sustain diversity jurisdiction.”

(citing Alfonso v. Hillsborough County Aviation

Auth., 

(5th Cir. 1962)). In this case, the requested injunction would

protect the rights of the class members that arise from their individual lease

agreements with GTE. When plaintiffs assert rights that arise from individual

contracts with a defendant, those rights are separate and distinct, and thus, their

claims may not be aggregated. See, e.g., Oliver v. Alexander, 

, 145-48

(1832) (joint suit by seamen to collect wages under their individual employment


                                          35
contracts); 

(class action involving individual insurance

policies); Alvarez v. Pan American Life Ins. Co., 

, 993 (5th Cir. 1967)

(same).

      The allegations that the lease agreements were induced and maintained by a

common scheme of fraud does not change the separate and distinct nature of the

rights asserted in this case. See Eagle Star Ins. Co. v. Maltes, 

, 780

n.4 (5th Cir. 1963) (“Federal Rule of Civil Procedure, [R]ule 20 allows the joinder

of parties plaintiff when there is a common question of law or fact and the claims

of all plaintiffs arose out of the same transaction o[r] occurrence. However, this

joinder for convenience of the court affects in no way the entirely separate question

of aggregation of claims to satisfy the jurisdictional amount.”). The plaintiffs

allege that, in those communities where it had a monopoly, GTE took advantage of

its monopoly status in order to accomplish its fraudulent scheme, and that its

customers were not in a position to individually negotiate their lease agreements.

If true, those facts may be relevant to the egregiousness of GTE’s alleged fraud,

but it remains true that the rights asserted arise from the individual lease agreement

of each class member. Part of the relief sought by the plaintiffs is for the court to

declare “the lease agreements null and void from their inception.”




                                          36
      It is also worth mentioning that since GTE raised the specter of the primary

jurisdiction of the APSC, the plaintiffs have insisted that the CPE activity

challenged in this case is not part of GTE’s public utility service but instead is a

free market, non-regulated business. If true, that is a further indication that the

rights which would be protected or vindicated by the requested injunction are the

separate and distinct rights of the class members in their lease agreements with

GTE, and as a result, the value of the injunctive relief may not be aggregated to

satisfy the amount in controversy requirement for diversity jurisdiction.

                    B. FEDERAL QUESTION JURISDICTION

      We turn now to plaintiffs’ fallback argument, which is that the district court

had federal question jurisdiction over this case, pursuant to 28 U.S.C. § 1331,

because this lawsuit seeks injunctive relief based on the FCC’s preemption of state

regulation of the CPE activity of telecommunications providers. Recall that the

plaintiffs seek an injunction to prevent GTE from misrepresenting its telephone

lease charges as part of its regulated, telecommunications service and thereby

wrongfully implying that it may disconnect a customer’s telephone service for

failure to pay the lease charges. The plaintiffs also seek a declaration that the lease

agreements “are null and void from their inception.”

      Under the federal question jurisdiction statute, 28 U.S.C. § 1331, a district


                                          37
court has subject matter jurisdiction over “all civil actions arising under the

Constitution, laws, or treaties of the United States.” Whether a claim arises under

federal law for purposes of 28 U.S.C. § 1331 is generally determined by the well-

pleaded complaint rule, “which provides that federal jurisdiction exists only when

a federal question is presented on the face of the plaintiff’s properly pleaded

complaint.” Caterpillar, Inc. v. Williams, 

, 392, 

, 2429

(1987). A well-pleaded complaint presents a federal question where it “establishes

either that federal law creates the cause of action or that the plaintiff’s right to

relief necessarily depends on resolution of a substantial question of federal law.”

Franchise Tax Bd. v. Construction Laborers Vacation Trust for S. Cal., 

,

27-28, 

, 2856 (1983).

       The complaint in this case establishes neither basis for federal question

jurisdiction, because it contains only state law causes of action and does not show

that any “substantial question of federal law” is necessary for the plaintiffs to

obtain their requested relief. See 

of whether the Alabama Public

Service Commission retains some regulatory power over the CPE activity of GTE,

or whether such power was preempted, arose not from the complaint but rather

from

GTE’s invocation of the primary jurisdiction doctrine in defense. As the Supreme


                                           38
Court stated in Franchise Tax Bd., federal question jurisdiction exists only when

“the plaintiff’s complaint establishes that the case ‘arises under’ federal law.” 

103 S. Ct. at 2847 (emphasis in original) (“‘[A] right or immunity created by

the Constitution or laws of the United States must be an element, and an essential

one, of the plaintiff’s cause of action.’”) (citation omitted).

      A federal question is presented when a suit “seeks injunctive relief from

state regulation[] on the ground that such regulation is pre-empted by a federal

statute which, by virtue of the Supremacy Clause of the Constitution, must prevail .

. . .” Shaw v. Delta Air Lines, Inc., 

, 96 n.14, 

, 2899

n.14 (1983). But the complaint in this case seeks injunctive relief against GTE, not

against the regulatory activity of the Public Service Commission. Again, the issue

of the Commission’s regulatory authority arises only from GTE’s defensive

assertion of the purported primary or exclusive jurisdiction of the APSC over the

plaintiffs’ claims.

      The plaintiffs argue that their claim for injunctive relief is essentially a

federal claim because federal law has completely preempted states’ regulation of

the CPE market and the claim for injunctive relief is based on that preemption.

“One corollary of the well-pleaded complaint rule ... is that Congress may so

completely pre-empt a particular area that any civil complaint raising this select


                                           39
group of claims is necessarily federal in character.” Metropolitan Life Ins. Co v.

Taylor, 

, 63-64, 

, 1546 (1987); Franchise Tax 

, 103 S. Ct. at 2854 (“[I]f a federal cause of action completely preempts a

state cause of action any complaint that comes within the scope of the federal cause

of action necessarily ‘arise under’ federal law.”).

      The Supreme Court has recognized that complete preemption occurs when

“the pre-emptive force of a statute is so ‘extraordinary’ that it ‘converts an ordinary

state common-law complaint into one stating a federal claim for purposes of the

well-pleaded complaint rule.’” 

, 107 S. Ct. at 2430

(citing Metropolitan 

, 107 S. Ct. at 1547). Under these rare

circumstances, “any claim purportedly based on that pre-empted state law is

considered, from its inception, a federal claim.” 

noted that “the [Supreme] Court has revisited the complete

preemption doctrine only sparingly” and has found complete preemption only in

the context of two federal statutes -- the Labor Management Relations Act

(LMRA), 29 U.S.C. § 141, et seq., and the Employee Retirement Income Security

Act (ERISA), 29 U.S.C. § 1001, et seq. BLAB T.V. of Mobile, Inc. v. Comcase

Cable Communications, Inc., 

, 855 (11th Cir. 1999). Furthermore,

“although the Supreme Court recognizes the existence of the complete preemption


                                          40
doctrine, the Court does so hesitatingly and displays no enthusiasm to extend the

doctrine into areas of law beyond the LMRA and ERISA.” 

Similarly,

this Court has yet to find complete preemption outside the context of the LMRA

and ERISA. 

which is the only Eleventh Circuit case to address directly the

application of the complete preemption doctrine outside of the context of the

LMRA and ERISA, we discussed the various tests employed by other courts to

determine whether complete preemption exists. See 

(concluding that

the Cable Communications Policy Act did not completely preempt state law).

Although we refused to adopt any particular approach, we summarized the various

factors considered by other courts as follows:

      These cases reveal a varying emphasis on such questions as whether
      the state claim is displaced by federal law under an ordinary
      preemption analysis, whether the federal statute provides a cause of
      action, what kind of jurisdictional language exists in the federal
      statute, and what kind of language is present in the legislative history
      to evince Congress’s intentions.

Furthermore, we concluded that the focus of each of the tests was “to

determine whether Congress not only intended a given federal statute to provide a

federal defense to a state cause of action that could be asserted either in a state or

federal court, but also intended to grant a defendant the ability to remove the

adjudication of the cause of action to a federal court by transforming the state

                                           41
cause of action into a federal [one].” 

quotation omitted).17 In

other words, the “touchstone” of federal jurisdiction under the complete

preemption doctrine is “Congress’s intent.” 

of the factors that we considered in BLAB, we are persuaded that the

complete preemption doctrine does not provide a basis for federal jurisdiction in

this case. The plaintiffs have not shown, and our review has not found, any

indication that either Congress or the FCC intended completely to preempt state

law in deregulating the CPE activity of telecommunications providers. Therefore,

federal question jurisdiction is not present as to the plaintiffs’ claim for injunctive

relief.

          In arguing that we should find complete preemption, the plaintiffs point to

FCC orders which expressly preempt some state regulation of CPE activities.

However, a review of those orders reveals that complete preemption was not

intended. Although the FCC’s orders concerning deregulation recognized that

          17
         The fact that, in BLAB, we refer to federal defenses rather than causes of actions, and to
“removal” jurisdiction rather than “subject matter” jurisdiction, reflects that the doctrine of
complete preemption has generally been considered in the context of a defendant seeking to
remove an action that, on its face, only asserts state law claims. In light of our conclusion that
the plaintiffs incorrectly relied on diversity jurisdiction as a basis for the district court’s subject
matter jurisdiction over this action, it is the plaintiffs rather than the defendant who argue that
the complete preemption doctrine provides an alternative basis for federal jurisdiction over this
action. Nonetheless, the same principles apply in this case as would apply if we were
determining the propriety of removal jurisdiction. See 

(noting that
removal jurisdiction is available only where federal courts have original jurisdiction over an
action pending in state court).

                                                  42
preemption of many state regulations would occur, the FCC stated that “we

preempt the states here only to the extent that their . . . regulation is at odds with

the regulatory scheme set forth.” In re Amendment of Section 64.702 of the

Commission’s Rules and Regulations, 84 F.C.C.2d 50, 103 (1980). Similarly, in a

subsequent order addressing the deregulation process, the FCC concluded that

“[t]he framework we adopt here provides maximum flexibility for states to

[deregulate] embedded CPE” and that “it is in the public interest that states be

given a substantial role in developing procedures for [deregulation].” Detariffing

of Customer Premises Equipment and Enhanced Services, 99 F.C.C.2d at 355 &

363 (1984). See also Computer and Communications Indus. Ass’n v. FCC, 

, 205 (D.C. Cir. 1982) (noting that “the Commission was careful to limit

the area of preemption”). Statements like those are not the language of complete

preemption, but are instead inconsistent with it. The FCC did not intend to

completely preempt all state laws concerning CPE regulation.

      Likewise, a review of the Federal Communications Act, 47 U.S.C. § 151, et

seq., pursuant to which the FCC was acting in deregulating CPE activity, reveals

that Congress also did not intend to preempt completely state causes of action or

remedies concerning the subject matter of the Act. In the Communications Act,

Congress provided that:


                                           43
      Any person claiming to be damaged by any common carrier subject to
      the provisions of [the Communications Act] may either make
      complaint to the Commission as hereinafter provided for, or may
      bring suit for the recovery of the damages for which such common
      carrier may be liable under the provisions of [the Communications
      Act] , in any district court of the United States of competent
      jurisdiction . . . .

47 U.S.C. § 207. Despite this provision permitting federal jurisdiction over well-

pleaded actions under the Communications Act, Congress further provided that

“[n]othing in this chapter contained shall in any way abridge or alter the remedies

now existing at common law or by statute, but the provisions of this chapter are in

addition to such remedies.” 47 U.S.C. § 414. As we stated in BLAB, the existence

of this type of “savings” clause which “contemplate[s] the application of state-law

and the exercise of state-court jurisdiction to some degree . . . counsels against a

conclusion that the purpose behind the . . . Act was to replicate the ‘unique

preemptive force’ of the LMRA and 

.

      Furthermore, the plaintiffs have not pointed to any provision in the

Communications Act itself or in its legislative history which would indicate “that

Congress intended state law causes of action within the scope of [the

Communications Act] to be federalized.” Sanderson, Thompson, Ratledge &

Zimny v. AWACS, Inc., 

, 958 (D. Del. 1997) (finding no

complete preemption in context of the Communications Act). Therefore, we


                                          44
conclude that the plaintiffs’ request for injunctive relief is not subject to the

complete preemption exception to the well-pleaded complaint rule. Consequently,

the district court had no subject matter jurisdiction over this action.

      Although the complete preemption doctrine does not apply in this case, we

recognize that use of the term “preemption” in this context has caused “a

substantial amount of confusion between the complete preemption doctrine and the

broader and more familiar doctrine of ordinary preemption.” 

. For that reason, it is worth pointing out that:

      complete preemption functions as a narrowly drawn means of
      assessing federal removal jurisdiction, while ordinary preemption
      operates to dismiss state claims on the merits and may be invoked in
      either federal or state court.

In other words, our conclusion that the complete preemption

doctrine does not provide a basis for federal jurisdiction in this action does not

preclude the parties from litigating about the preemptive effect, if any, of the

FCC’s orders or the Communications Act in any subsequent state court action.

                                 III. CONCLUSION

      Because the district court lacked subject matter jurisdiction over this case,

we VACATE its order dismissing the case on abstention grounds and REMAND

with directions that the case be dismissed for lack of jurisdiction.

      VACATED AND REMANDED.

                                           45