LOWER COLORADO RIVER AUTHORITY v. PAPALOTE CREEK II II 

United States Court of Appeals, Fifth Circuit.

LOWER COLORADO RIVER AUTHORITY, Plaintiff – Appellee v. PAPALOTE CREEK II, L.L.C., formerly known as Papalote Creek Wind Farm II, L.L.C., Defendant – Appellant

No. 16-50317

    Decided: May 31, 2017

Before KING, JOLLY, and PRADO, Circuit Judges.

This appeal concerns the district court’s grant of a petition to compel arbitration. Defendant–Appellant Papalote Creek II, L.L.C. argues that that the district court did not have jurisdiction to compel arbitration because the underlying dispute between the parties was not ripe, and even if the district court did have jurisdiction, the underlying dispute was outside the scope of the arbitration provision. Because we conclude that the district court lacked jurisdiction to compel arbitration, we VACATE and REMAND.

I. FACTUAL AND PROCEDURAL BACKGROUND

A. The Power Purchase Agreement

Plaintiff–Appellee Lower Colorado River Authority (LCRA) is a conservation and reclamation district based in Austin, Texas, and a political subdivision of the State of Texas. LCRA sells wholesale electric power to municipal-owned utilities and electric cooperatives in Texas. In December 2009, LCRA entered into a Power Purchase Agreement (PPA) with Defendant–Appellant Papalote Creek II, L.L.C. (Papalote), a Delaware limited liability company that builds and operates wind farms. Papalote planned to build an 87-turbine wind farm in Texas (the Project), and under the PPA, LCRA agreed to purchase all of the energy generated by the Project at a fixed price for an 18-year term.

Relevant to this appeal are four sections of the PPA: § 4.3, § 9.3, § 13.1, and § 13.2. First, § 4.3, which is entitled “Liquidated Damages Due to [LCRA’s] Failure to Take,” provides a formula for how to calculate the liquidated damages that LCRA would owe to Papalote in the event that LCRA failed to take all of the Project’s energy. As noted above, LCRA is required to take all of the energy generated by the Project. However, should LCRA fail to do so, § 4.3 details how to calculate Papalote’s “exclusive remedy” of liquidated damages. This liquidated damages calculation would depend in part on the difference between the PPA’s fixed price and the price that Papalote is otherwise able to obtain in selling the energy.

Second, § 9.3, which is entitled “Limitation on Damages for Certain Types of Failures,” provides the following:

Notwithstanding anything to the contrary in [the PPA], [Papalote’s] aggregate liability for (i) failure of [Papalote] to construct the Project and/or (ii) failure of one hundred percent (100%) of the Project’s Turbines to achieve the Commercial Operation Date on the Scheduled COD and/or (iii) failure of one hundred percent (100%) of the Project’s Turbines to achieve the Commercial Operation Date on June 1, 2011 and/or (iv) a Termination Payment, shall be limited in the aggregate to sixty million dollars ($60,000,000). [LCRA’s] damages for failure to perform its material obligations under [the PPA] shall likewise be limited in the aggregate to sixty million dollars ($60,000,000).

(Emphasis added.) Notably, § 9.3 refers to Papalote’s “liability” and LCRA’s “damages,” and the parties’ underlying dispute is based, in part, on this word choice.

Finally, § 13.1 and § 13.2 provide a two-step arbitration procedure. The first step, as dictated in § 13.1, requires, inter alia, that “[i]f any dispute arises with respect to either Party’s performance hereunder,” the senior officers of LCRA and Papalote meet in an attempt to resolve the dispute. Under the second step, as outlined in § 13.2, if the dispute is not resolved through the first step within a certain timeframe, either party may submit that dispute “to binding arbitration in accordance with the Commercial Arbitration Rules of the American Arbitration Association ․ effective at the time of the dispute.” Section 13.2 also provides additional details on the arbitration, including that the arbitrator shall use a “baseball” procedure (in which each party puts forth an offer and the arbitrator is limited to choosing one of the two offers).

B. Negotiations and Petition to Compel Arbitration

Papalote completed construction of the Project in 2010, and in the ensuing years, LCRA complied with its obligations under the PPA by purchasing all of the energy generated by the Project. In April 2015, however, LCRA initiated discussions with Papalote regarding the PPA. Although the parties dispute the precise nature of these discussions,¹ neither party appears to have threatened to breach the PPA. Ultimately, in June 2015, LCRA sent Papalote a letter stating that, pursuant to § 13.2, LCRA was “initiat[ing] the arbitration process to resolve the dispute between LCRA and Papalote regarding LCRA’s limitation of liability under the PPA and its impact on LCRA’s performance obligations.” LCRA also noted that it “intends to continue to fully perform its obligations under the PPA during this arbitration process.” After Papalote requested more information about the purported dispute, LCRA sent another letter explaining that the dispute was “whether LCRA’s liability is limited to $60,000,000 under the PPA.” Papalote rejected LCRA’s request to proceed to arbitration, reasoning that “[a]n academic question about the damages LCRA might owe for a hypothetical breach simply does not constitute a ‘dispute’ that is proper for arbitration under the PPA.” Papalote also argued that a dispute over LCRA’s potential liability limitation was not covered by the arbitration provision in the PPA, which was limited to disputes regarding performance obligations.

Following Papalote’s refusal to arbitrate, LCRA filed a petition to compel arbitration in Texas state court on June 30, 2015. Papalote timely removed the petition to federal district court on the basis of diversity jurisdiction. In the district court, Papalote opposed the petition to compel by arguing that the dispute at issue did not fall within the scope of the arbitration provision. According to Papalote, the arbitration provision only includes disputes related to the parties’ performance obligations, and LCRA’s dispute regarding whether its liability is capped under § 9.3 is not a performance obligation. Papalote also challenged the ripeness of the dispute in passing, arguing that, “if the Court would prefer to deny the [petition to compel] based on the lack of a ripe dispute, it may do so consistent with the facts presented and without running afoul of any binding authority.”

In February 2016, the district court granted LCRA’s petition to compel arbitration. As an initial matter, the district court noted that both parties agreed that the PPA’s arbitration provision was valid and enforceable, and thus, the only question was whether the dispute fell within the scope of the arbitration provision. The district court then rejected LCRA’s argument that the arbitration provision covered any dispute arising under the PPA. Instead, the district court found that, under § 13.1, the parties had agreed to arbitrate only disputes that “arise[ ] with respect to either Party’s performance.” (Alteration in original.) Thus, the district court framed the question as “whether the dispute LCRA seeks to arbitrate—whether or not LCRA’s liability would be capped at $60 million in the event it elected to purchase from Papalote less than the total amount of energy it contracted to buy—qualifies as a dispute ‘with respect to either Party’s performance’ under the PPA.” In answering that question, the district court recognized that, “in a certain sense, one could understand ‘performance’ to concern only those promises which were the essence of the PPA—the sale and production of wind energy—and conceptualize the buyer’s obligation to pay for failing to take as compensation for its failure to perform, rather than as an independent performance obligation.” The district court reasoned, however, that “the better view here ․ is that LCRA’s bargained-for obligation to pay Papalote a specified sum if LCRA takes less than all of the energy produced is itself a performance obligation under the PPA.” According to the district court, LCRA’s failure to take all of the Project’s energy was not necessarily itself a breach giving Papalote the right to terminate the PPA. Instead, the PPA allows the parties’ obligations to continue so long as LCRA pays liquidated damages, and if LCRA fails to make the necessary liquidated damages payment, only then would “that failure ․ constitute an ‘Event of Default’ permitting Papalote to suspend its performance and terminate the [PPA].” As for Papalote’s ripeness argument, the district court recognized that “ripeness questions plainly loom.” However, the district court declined to further address whether ripeness should be decided by the arbitrator or the court because the parties had failed to adequately brief the issue. Papalote timely appealed.²

C. Arbitration and Subsequent Developments

Following the district court’s order compelling arbitration, the parties proceeded through arbitration. Besides its arguments on the merits, Papalote also argued that the arbitrator should dismiss the claim because it was not ripe. On June 28, 2016, the arbitrator issued a decision in favor of LCRA. Specifically, the arbitrator found that, “[u]nder [§] 9.3 of the ․ [PPA], LCRA’s liability for liquidated damages and/or a Termination Payment for its failure to take power under the agreement is limited to $60,000,000.” The arbitrator did not directly address Papalote’s ripeness argument.

On October 10, 2016, LCRA notified Papalote that it would cease taking energy under the PPA beginning on October 12, 2016, and that its resulting liquidated damages would be capped at $60 million per § 9.3.

II. RIPENESS

Papalote argues that the district court erred in compelling arbitration because the issue was not ripe, and thus, the district court lacked jurisdiction to compel arbitration. We first address whether the underlying dispute between the parties must be ripe in order for the district court to have jurisdiction to compel arbitration. Finding that the answer is yes, we next turn to whether the underlying dispute was ripe. We review the jurisdictional issue of ripeness de novo. See Choice Inc. of Tex. v. Greenstein, 691 F.3d 710, 714 (5th Cir. 2012). “As the party asserting federal jurisdiction,” LCRA has “the burden of demonstrating that jurisdiction is proper.” See Stockman v. FEC, 138 F.3d 144, 151 (5th Cir. 1998).

A. Jurisdiction to Compel Arbitration

Although the district court recognized that “ripeness questions plainly loom,” the district court declined to address whether the underlying dispute was ripe because the parties had failed to adequately brief whether ripeness should be determined by the district court or the arbitrator. Papalote’s purported failure to adequately brief the ripeness issue, however, does not result in waiver here. Even assuming that Papalote had failed to adequately raise the ripeness issue in the district court, we still must consider whether the district court had jurisdiction to compel arbitration: “[E]very federal appellate court has a special obligation to ‘satisfy itself not only of its own jurisdiction, but also that of the lower courts in a cause under review,’ even though the parties are prepared to concede it.” United Transp. Union v. Foster, 205 F.3d 851, 857 (5th Cir. 2000) (alteration in original) (quoting Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 95 (1998)). Under Article III of the Constitution, federal courts are confined to adjudicating “cases” and “controversies.” Id. And to be a case or controversy for Article III jurisdictional purposes, the litigation “must be ripe for decision, meaning that it must not be premature or speculative.” See Shields v. Norton, 289 F.3d 832, 835 (5th Cir. 2002); see also Choice Inc. of Tex., 691 F.3d at 715 (“The justiciability doctrines of standing, mootness, political question, and ripeness ‘all originate in Article III’s case or controversy language ․’ ” (omission in original) (quoting DaimlerChrysler Corp. v. Cuno, 547 U.S. 332, 352 (2006))). In other words, “ripeness is a constitutional prerequisite to the exercise of jurisdiction.” Shields, 289 F.3d at 835.

Thus, we must confront the question that the district court declined to address: Was there a ripe controversy between LCRA and Papalote such that the district court had jurisdiction over the petition to compel arbitration? But this question raises another question that must be answered first: Which dispute matters for the purpose of determining whether there is a ripe controversy? In this context, there are effectively two potential disputes that a court could consider in determining whether there is a sufficiently ripe controversy. On the one hand, LCRA argues that the dispute should be viewed as whether arbitration should be compelled under the Federal Arbitration Act (FAA). In this sense, the dispute was almost certainly a ripe controversy because LCRA was seeking to compel arbitration immediately, not at some hypothetical future date. On the other hand, Papalote contends that the dispute should be viewed as whether the underlying dispute presents a ripe controversy. Put another way, Papalote’s position is that a court must “look through” the petition to compel arbitration in order to determine whether the underlying dispute—in this case, the contractual disagreement over whether § 9.3 caps LCRA’s liability at $60 million—presents a sufficiently ripe controversy.

Under the Supreme Court’s decision in Vaden v. Discover Bank, 556 U.S. 49 (2009), we must follow the second approach—i.e., we must “look through” the petition to compel arbitration in order to determine whether the underlying dispute presents a sufficiently ripe controversy to establish federal jurisdiction. At issue in Vaden, as in this case, was § 4 of the FAA, which “provides for United States district court enforcement of arbitration agreements.” Id. at 58. Under § 4, a petition to compel arbitration may be brought before “any United States district court which, save for [the arbitration] agreement, would have jurisdiction under Title 28 ․ of the subject matter of a suit arising out of the controversy between the parties.” 9 U.S.C. § 4; see also Vaden, 556 U.S. at 58. In Vaden, the Supreme Court addressed how a district court should determine if it has jurisdiction over a § 4 petition to compel arbitration. Vaden, 556 U.S. at 53, 62–65; see also Volvo Trucks N. Am., Inc. v. Crescent Ford Truck Sales, Inc., 666 F.3d 932, 936 (5th Cir. 2012). In answering this question, the Supreme Court first emphasized that the FAA is “ ‘something of an anomaly’ in the realm of federal legislation: It ‘bestow[s] no federal jurisdiction but rather requir[es] [for access to a federal forum] an independent jurisdictional basis’ over the parties’ dispute.” Vaden, 556 U.S. at 59 (alterations in original) (quoting Hall Street Assocs., L.L.C. v. Mattel, Inc., 552 U.S. 576, 581–82 (2008)). Turning to the statutory language of § 4, the Supreme Court reasoned that “[t]he phrase ‘save for [the arbitration] agreement’ indicates that the district court should assume the absence of the arbitration agreement and determine whether it ‘would have jurisdiction under title 28’ without it.” Id. at 62 (second alteration in original). And the Supreme Court held that § 4’s reference to “the controversy between the parties” means the underlying substantive conflict between the parties. Id. In other words, the Supreme Court held that a district court must “look through” a § 4 petition to determine whether the district court would have jurisdiction over the underlying substantive controversy. See id. at 53, 62.

Although Vaden concerned whether there was federal question jurisdiction, we see no reason that the holding is limited to only that specific jurisdictional issue. For example, the Supreme Court summed up its holding by stating that “§ 4 of the FAA does not enlarge federal-court jurisdiction; rather, it confines federal courts to the jurisdiction they would have ‘save for [the arbitration] agreement,’ ” and “[m]indful of that limitation, we read § 4 to convey that a party seeking to compel arbitration may gain a federal court’s assistance only if, ‘save for’ the agreement, the entire, actual ‘controversy between the parties,’ as they have framed it, could be litigated in federal court.” Id. at 66 (first alteration in original). Accordingly, Vaden’s holding necessarily implies that any of the reasons that a federal court may lack subject matter jurisdiction over the underlying dispute—e.g., ripeness—would similarly prevent a district court from having jurisdiction to compel arbitration. Put another way, given Vaden’s holding that the district court should assume the absence of an arbitration agreement in determining whether there is jurisdiction over the underlying dispute, it necessarily follows that, if the underlying dispute is not ripe, then the district court would not have jurisdiction to compel arbitration.

In response, LCRA largely fails to address Papalote’s jurisdictional argument, and indeed, LCRA (appellee here) does not address Vaden at all in its brief. Instead, LCRA makes a series of arguments that skip the threshold question of whether the district court had jurisdiction to compel arbitration. For example, LCRA contends that whether an arbitration provision requires a claim to be ripe is a question for the arbitrator, not the district court. Specifically, LCRA contends that, under Howsam v. Dean Witter Reynolds, Inc., 537 U.S. 79 (2002), ripeness is a question of procedural arbitrability to be resolved by the arbitrator, and even if ripeness is a question of substantive arbitrability, there is clear and unmistakable evidence here that the parties agreed to delegate the issue to the arbitrator. But we need not reach that issue because it does not change the fact that the district court must have jurisdiction in the first instance to compel arbitration, and a ripe controversy is a necessary component of subject matter jurisdiction.³ See, e.g., Lopez v. City of Houston, 617 F.3d 336, 341 (5th Cir. 2010) (“Ripeness is a component of subject matter jurisdiction, because a court has no power to decide disputes that are not yet justiciable.”). Additionally, LCRA argues that the arbitration provision does not require a dispute to be ripe. Once again, however, whether an arbitration provision requires a ripe dispute does not change the fact that the district court still must have jurisdiction to compel arbitration.

B. Ripeness of the Underlying Dispute

Having established that the district court should have determined whether the underlying dispute was ripe, we now turn to the merits of the ripeness issue. As discussed above, a claim must be ripe for a federal court to have jurisdiction, see, e.g., Shields, 289 F.3d at 835, and “[a] court should dismiss a case for lack of ‘ripeness’ when the case is abstract or hypothetical,” Orix Credit All., Inc. v. Wolfe, 212 F.3d 891, 895 (5th Cir. 2000) (quoting New Orleans Pub. Serv., Inc. v. Council of New Orleans, 833 F.2d 583, 586 (5th Cir. 1987)). In determining whether a case is ripe, there are two key considerations: “the fitness of the issues for judicial decision and the hardship to the parties of withholding court consideration.” Id. (quoting New Orleans Pub. Serv., 833 F.2d at 896).

At the outset, it is helpful to frame the underlying dispute. Although neither party was in breach of the PPA, the parties disagreed about whether § 9.3 capped LCRA’s aggregate liability at $60 million. Thus, the underlying claim that LCRA sought to arbitrate is effectively one for a declaratory judgment that its interpretation of § 9.3 is correct. “In the declaratory judgment context, whether a particular dispute is ripe for adjudication turns on whether a substantial controversy of sufficient immediacy and reality exists between parties having adverse legal interests.” Venator Grp. Specialty, Inc. v. Matthew/Muniot Family, LLC, 322 F.3d 835, 838 (5th Cir. 2003). “Whether particular facts are sufficiently immediate to establish an actual controversy is a question that must be addressed on a case-by-case basis.” Orix, 212 F.3d at 896. Notably, “[t]he threat of litigation, if specific and concrete, can indeed establish a controversy upon which declaratory judgment can be based.” Id. at 897. The fact that future litigation may be contingent upon certain factors occurring does not necessarily defeat jurisdiction over a declaratory judgment action, but “a district court must take into account the likelihood that these contingencies will occur.” See id. Accordingly, we have described the ripeness inquiry as “focus[ing] on whether an injury that has not yet occurred is sufficiently likely to happen to justify judicial intervention.” See id. (quoting Chevron U.S.A., Inc. v. Traillour Oil Co., 987 F.2d 1138, 1153 (5th Cir. 1993)); see also Shields, 289 F.3d at 835 (“We look to the practical likelihood that a controversy will become real.”).

Here, LCRA has failed to show that its claim was ripe at the time the district court compelled arbitration. The sole dispute between the parties centered on their differing interpretations of whether § 9.3 capped LCRA’s liability at $60 million, but as relevant to this appeal, the differing interpretations of § 9.3 would not need to be resolved unless LCRA first decided to stop taking energy. LCRA argues that this dispute was sufficiently ripe because it was a purely legal issue and LCRA “had a direct and immediate dilemma” because it could not determine, without knowing if its liquidated damages would be capped by § 9.3, whether it should continue to take all of the Project’s energy or opt instead to pay liquidated damages. We disagree. LCRA was fully performing under the PPA at the time the district court compelled arbitration. Although it is not an absolute prerequisite for ripeness for there to be a contractual breach, there is no evidence that LCRA threatened to stop taking energy or that such a decision was even likely. To the contrary, LCRA appears to have consistently maintained that, even if it received a favorable ruling, there was only a possibility that it would opt to stop taking energy. See Orix, 212 F.3d at 896 (“Such unasserted, unthreatened, and unknown claims do not present an immediate or real threat to [the plaintiff] such that declaratory relief is proper.”). For example, in opposing Papalote’s motion to stay arbitration pending appeal, LCRA argued that any harm to Papalote absent a stay was only a “mere possibilit[y]” that depended on “a myriad of uncertainties,” one of which was whether LCRA would decide to stop taking energy. LCRA may have been able to establish that the issue was ripe, but on this record, it has failed to do so.⁴ LCRA simply points to no other evidence or allegations that would support the idea that it was sufficiently likely to decide to stop taking energy, nor does it point to anything that would contradict its own statements to the contrary. See id. at 897 (“[The plaintiff] simply fails to allege facts that show that these contingencies are likely to occur.”); see also Texas v. United States, 523 U.S. 296, 300 (1998) (“A claim is not ripe for adjudication if it rests upon ‘contingent future events that may not occur as anticipated, or indeed may not occur at all.’ ” (quoting Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568, 580–81 (1985))).⁵

LCRA’s reliance on Venator Group Specialty is misplaced. In Venator Group Specialty, we held that a declaratory judgment action was ripe even though it sought a ruling on the effect of terms in a commercial lease that had not yet been triggered. 322 F.3d at 839–40. Whether one of the terms would ever be triggered depended on a specific contingency occurring in the future—the lessor invoking a certain right in the lease. Id. Despite this contingency, we found that, based on the facts of that case, the lessor was “very likely” to invoke the right at issue, and thus, the dispute was ripe. Id. Unlike the circumstances in Venator Group Specialty, however, the circumstances in this case, on the record before us, do not demonstrate that the contingency at issue—i.e., LCRA deciding to stop taking energy from the Project and paying liquidated damages instead—was likely to occur at the time the district court compelled arbitration. As discussed above, LCRA itself described this decision as merely an uncertainty.

However, this case presents a unique circumstance: LCRA prevailed in arbitration and, several months later, decided to cease taking energy under the PPA. If the district court were deciding the ripeness issue today, the underlying dispute would be ripe given that LCRA has, in fact, stopped taking energy. The contingency at the time the district court compelled arbitration—i.e., whether LCRA would ultimately decide to stop taking energy—has occurred. Thus, LCRA argues that, in light of the recent developments, this court should affirm the order compelling arbitration because the underlying dispute is now ripe. LCRA’s argument has some initial appeal. Indeed, there are a number of cases discussing how, “[i]n weighing a ripeness claim, an appellate court may properly consider events occurring after the trial court’s decision.” New Orleans Pub. Serv., 833 F.2d at 586 n.2; see also, e.g., Blanchette v. Conn. Gen. Ins., 419 U.S. 102, 140 (1974) (“[S]ince ripeness is peculiarly a question of timing, it is the situation now rather than the situation at the time of the District Court’s decision that must govern.”).

But the fact that an appellate court may consider subsequent events when assessing ripeness does not necessarily dictate the result in this case. Instead, the unique procedural context here is fundamentally different from the typical situation in which an appellate court considers subsequent events when assessing ripeness. In the typical situation in which subsequent events support a ripeness finding, the underlying merits are part of the case, and the appellate court (or the district court on remand) can evaluate the merits in light of the fact that the case is now ripe. See, e.g., Skull Valley Band of Goshute Indians v. Nielson, 376 F.3d 1223, 1238 & n.2 (10th Cir. 2004); Syron v. ReliaStar Life Ins., 506 F. App’x 500, 503–05 (6th Cir. 2012). Here, however, the underlying merits are not part of this appeal. Instead, the “merits” of this case are unique in that they concern whether the district court properly compelled arbitration (whereas the underlying merits were addressed in the actual arbitration). This distinction is key because the order compelling arbitration has already completed its intended effect in a way that a typical merits ruling would not—arbitration has already occurred. This case would be similar to the typical ripeness cases addressing subsequent events if we (or the district court) compelled arbitration now in light of the fact that the underlying dispute is now ripe, but that is not the situation given that arbitration has already occurred. And if we (or the district court) were to compel arbitration now, it does not necessarily have the same effect as affirming the district court’s order compelling arbitration because a new arbitration could conceivably result in a different outcome.

To frame the issue more concretely, we must decide whether we can affirm the prior order compelling arbitration, which was made without jurisdiction at the time, because the dispute has since become ripe or whether we must vacate the prior order compelling arbitration and reconsider the petition to compel now. This question hinges on whether the subsequent developments here can somehow retroactively bestow jurisdiction on the district court’s prior order compelling arbitration. We hold that they cannot. Although this appears to be a matter of first impression, we find most applicable the general rule that judgments made by a district court without subject matter jurisdiction are void. See, e.g., Brumfield v. La. State Bd. of Educ., 806 F.3d 289, 298 (5th Cir. 2015) (“An order ‘is void only if the court that rendered it lacked jurisdiction of the subject matter, or of the parties ․’ ” (quoting Williams v. New Orleans Pub. Serv., Inc., 728 F.2d 730, 735 (5th Cir. 1984))); Hill v. McDermott, Inc., 827 F.2d 1040, 1043 (5th Cir. 1987) (“A judgment is void on jurisdictional grounds if the [district] court lacked jurisdiction over the subject matter or over the parties.”); cf. United Student Aid Funds, Inc. v. Espinosa, 559 U.S. 260, 270 (2010) (“A void judgment is a legal nullity.”). Under this reasoning, although subsequent events have made it such that the district court would now have jurisdiction to compel arbitration, these events do not retroactively cure the void order compelling Papalote to an arbitration that it should not have been forced to attend at the time. We recognize that the parties have already fully arbitrated the underlying dispute once while this appeal was pending. However, we cannot evade the fact that the district court lacked jurisdiction when it compelled arbitration, and the fact that a court would have jurisdiction now to compel arbitration does not retroactively bestow jurisdiction on the prior order.

III. CONCLUSION

In sum, we have made clear that a district court must have subject matter jurisdiction over the underlying dispute in order to compel arbitration under 9 U.S.C. § 4. Because the matter was not ripe—i.e., there was no Article III “case” or “controversy”—at the time the district court entered judgment in this case, the district court’s judgment is void for lack of subject matter jurisdiction and is vacated.⁶ Even though subsequent intervening events have created a controversy that is now ripe, we cannot retroactively resurrect the district court’s void judgment under the facts of this case. Nevertheless, because the basic underlying controversy, originally raised and pursued by these same parties, is now ripe, we remand the case to the district court for such orders and proceedings as the district court deems necessary and appropriate. We leave to the discretion of the district court whether it should consider anew the petition to compel arbitration or conduct other proceedings.

For the foregoing reasons, we VACATE the judgment of the district court and REMAND the case for further proceedings consistent with this opinion.

FOOTNOTES

1.   LCRA claims that these discussions centered on whether § 9.3 capped its aggregate liability at $60 million. Conversely, Papalote claims that these discussions were on a more general level, such as exploring alternative pricing options.

2.   Papalote also filed a motion to stay arbitration pending appeal, arguing that it was likely to prevail on appeal because the dispute was not ripe. In opposition, LCRA contended that Papalote’s appeal was not likely to succeed because ripeness was an issue for the arbitrator to decide. Moreover, LCRA contended that Papalote had failed to show that it would suffer an irreparable injury, in part, because the effect of any arbitration award would ultimately depend on “a myriad of uncertainties,” including whether LCRA would decide to stop taking energy under the PPA. The district court denied the motion to stay, reasoning, inter alia, that Papalote had not shown that it was likely to succeed on its ripeness argument because “the weight of authority on the topic states ripeness is a question for the arbitrator to decide.” However, the district court noted that there was “ample room for disagreement on the question [of] whether courts or arbitrators should decide ripeness issues,” but it concluded that the potential for disagreement “does not affect the Court’s conclusion that a stay is not warranted in this case.”

3.   LCRA points to district court cases holding that ripeness is a question for the arbitrator, not the district court. However, many of these cases predate Vaden. See, e.g., Albritton v. W.S. Badcock Corp., No. 1:02-CV378, 2003 WL 21018636, at *4 (N.D. Miss. Apr. 7, 2003) (citing Howsam for the proposition that “procedural questions such as ripeness are for an arbitrator, not the court, to decide”). Regarding the district court cases decided after Vaden, we disagree with their holdings. See, e.g., Transp. Workers Union of Am. v. Veolia Transp. Servs., Inc., 24 F. Supp. 3d 223, 229–30 (E.D.N.Y. 2014) (holding that ripeness argument was a question for the arbitrator). Indeed, none of the district court cases relied on by LCRA for this proposition even cites Vaden. See id.; Grant v. Brown, No. 4:14CV01395, 2014 WL 6389577, at *2 (E.D. Mo. Nov. 14, 2014); Local Union No. 13417 of the United Steel Workers v. Kan. Gas Serv. Co., No. 12-1003, 2012 WL 1435305, at *7 n.3 (D. Kan. Apr. 25, 2012); Milliman, Inc. v. Health Medicare Ultra, Inc., 641 F. Supp. 2d 113, 119 (D.P.R. 2009).

4.   Presumably the prices of other sources of energy were among the primary factors affecting whether LCRA would decide to stop taking the Project’s energy in the event that it received a favorable ruling.

5.   LCRA’s hardship argument is also unavailing. LCRA claims that it faced the difficult decision of continuing to take all of the Project’s energy or opting instead to breach the PPA and risk a damages award exceeding $60 million. Moreover, according to LCRA, the PPA required it to continue performing while the dispute was being resolved. But this overstates the steps that LCRA was required to take in order to make the underlying dispute ripe. As highlighted above, LCRA was not required to breach the PPA to make the dispute ripe.

6.   We therefore do not reach Papalote’s argument that the underlying dispute was outside the scope of the arbitration provision.

KING, Circuit Judge: