PUBLISHED
UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT
No. 22-1216
ROEE KIVITI; ADIEL KIVITI, Plaintiffs - Appellants,
v. NAVEEN PRASAD BHATT, Debtor - Appellee.
Appeal from the United States District Court for the Eastern District of Virginia, at Alexandria. Anthony J. Trenga, Senior District Judge. (1:21-cv-00909-AJT-JFA)
Argued: March 7, 2023 Decided: September 14, 2023
Before RICHARDSON and RUSHING, Circuit Judges, and MOTZ, Senior Circuit Judge.
Vacated and remanded by published opinion. Judge Richardson wrote the opinion, in which Judge Rushing and Senior Judge Motz joined.
Maurice Belmont VerStandig, THE VERSTANDIG LAW FIRM, LLC, Henderson, Nevada, for Appellants. Justin Philip Fasano, MCNAMEE HOSEA, P.A., Greenbelt, Maryland, for Appellee.
2
RICHARDSON, Circuit Judge:
We must police Congress’s limits on judicial review, even when both parties would rather we not. Adiel and Roee Kiviti paid Naveen Bhatt to renovate their Washington, D.C. home. But he wasn’t properly licensed. So, according to D.C. law, there was a chance he owed them their money back. Rather than pay, he invoked the bittersweet sanctuary of bankruptcy. Refusing to be evaded that easily, the Kivitis pursued him, filing a two-claim complaint against him in bankruptcy court. Because of bankruptcy-law nuances, the suit was fruitful for the Kivitis only if they won on both claims.
One can thus imagine their frustration when the bankruptcy court dismissed one claim but not the other. The claims were meant to rise or fall together. It wasn’t worth the trouble to either party to litigate the remaining claim to completion without knowing if the dismissed one would be revived on appeal—the juice just wasn’t worth the squeeze. But their problem was that they could only appeal final orders. And the partial dismissal wasn’t final because one claim survived.
So they hatched a plan to make the bankruptcy court’s order final by voluntarily dismissing the surviving claim without prejudice. They could then immediately appeal the court-dismissed claim and decide afterward whether it was worth further litigating the party-dismissed claim.
At the district court, the plan went off without a hitch. The district judge accepted that the bankruptcy court’s partial dismissal was now final, and so reviewable, and then affirmed it. The Kivitis appealed to this Court, hoping to get the bankruptcy court’s order reversed and move forward on both claims. Meanwhile, Bhatt was content to not have to
3
litigate the claim that the bankruptcy court had left standing. Everything was unfolding as the parties had hoped.
Or so they thought. The thing is, we don’t allow parties to manufacture finality like this. Congress told federal courts to review only final bankruptcy orders, barring some exceptions not relevant here. And we zealously guard that boundary, rejecting clever gambits aimed at eroding the statutory line between us and plenary review. Since this is one of those gambits, we must reject it. The bankruptcy court’s partial dismissal was not a final order. Nor did the parties’ after-the-fact machinations make it one. We thus vacate the district court’s order for lack of jurisdiction. I. Background The Kivitis hired Bhatt to renovate their home in Washington, D.C. To renovate homes in D.C., contractors need to be licensed by the District. Because Bhatt told the Kivitis he was properly licensed, they thought everything was above board. Yet, delayed and defective, the renovations did not go well. And, as it turned out, Bhatt was not properly licensed. So the Kivitis sued him in D.C.’s Superior Court to the tune of $58,770—every penny they had paid him. But then Bhatt filed for Chapter 7 bankruptcy.
Chapter 7 bankruptcy liquidates the debtor’s estate. Subject to some exceptions, an appointed trustee identifies and liquidates the debtor’s assets. See 11 U.S.C. §§ 704(a), 725, 726. To permit the trustee’s orderly resolution, creditors are barred from seeking to recover outside the bankruptcy system. See 11 U.S.C. §§ 362(a), 727. Creditors, instead, file proofs of claim showing what they claim to be owed by the debtor. See Fed. R. Bankr. P. 3001(a). The allowed claims—those that are found valid—are then satisfied in order of
4
priority (secured creditors before unsecured, for example). If, as often is the case, there aren’t enough assets to satisfy all the unsecured claims, then the unsecured creditors split the remainder of the debtor’s assets pro rata. So those creditors often get little-to-none of their sought-after funds.
Once a debtor’s assets are liquidated and distributed, his debts are generally discharged by the bankruptcy court. Discharged debt cannot be collected outside of bankruptcy. This promotes Chapter 7’s goal of giving debtors a “fresh start” post- bankruptcy; debtors couldn’t start anew if they left bankruptcy only to face an onslaught of legal claims. See Marrama v. Citizens Bank of Ma., 549 U.S. 365, 367 (2007). But not all debts are discharged. As relevant here, a debt is non-dischargeable if it was obtained through “false pretenses, a false representation, or actual fraud.” See 11 U.S.C. § 523(a)(2)(A).
In addition to a proof of claim, the bankruptcy code also permits a creditor to seek relief by filing what is known as an adversary proceeding. Adversary proceedings resemble civil suits and take place within a broader bankruptcy case. See In re Boca Arena, Inc., 184 F.3d 1285, 1286 (11th Cir. 1999) (“In bankruptcy, adversary proceedings generally are viewed as ‘stand-alone lawsuits’. . .”). They pit some parties involved in the bankruptcy against each other, resolving some of their discrete issues.
When Bhatt filed for Chapter 7 bankruptcy, the code automatically stayed the Kivitis’ D.C. Superior Court action. Automatic bankruptcy stays stop any collection efforts, including lawsuits, outside the bankruptcy system. So it was the Chapter 7 process
5
or nothing. The Kivitis entered the bankruptcy fray by filing both an adversary proceeding and a proof of claim.
The Kivitis’ adversary proceeding brought two counts: Count I asked the bankruptcy court to declare Bhatt owed them $58,770 under D.C. law; Count II asked it to pronounce that debt nondischargeable. In other words, tell Bhatt he owes us money and that—despite the bankruptcy—he must pay in full.
The bankruptcy court rejected Count II, finding that, if a debt existed, it was dischargeable. So it partially dismissed the adversary proceeding. But it allowed Count I to proceed toward trial to determine whether Bhatt owed the Kivitis any money. It never got there.
As an alternative to the adversary proceeding, the Kivitis also filed a proof of claim—in the broader bankruptcy case, yet outside the adversary proceeding—stating they had a right to payment of the $58,770 from Bhatt’s estate. When the Kivitis filed their proof of claim, they became eligible for some portion of that distribution unless an interested party successfully objected to it. See 11 U.S.C. § 502(a). Given the fact their claim was unsecured (i.e., low priority) and the state of Bhatt’s finances, they seemed unlikely to recover their full claim this way.
The Kivitis’ strategy was a hedge. The proof of claim advanced the same legal theory to obtain the same money as they sought in Count I of the adversary proceeding. But the proof of claim was limited to some share of whatever assets were left in the bankruptcy estate. The adversary proceeding, in contrast, might permit full recovery. But that full recovery existed only outside the bankruptcy estate. And to get outside, the Kivitis
6
had to establish that the claim was not dischargeable. Otherwise, the adversary proceeding was limited to the same share of the assets that the proof of claim would cover. So—unless the bankruptcy judge’s ruling that the debt to the Kivitis was dischargeable was overturned—any right to recover from the adversary proceeding was duplicative of their proof of claim. The Kivitis’ proof of claim was thus a fallback: If the ruling on dischargeability stood, then the proof of claim gave Kivitis a chance to receive some of their money, even if they were unlikely to receive all their money.
But if the Kivitis had to rely on their fallback, both parties preferred to know right away. They would rather understand how appellate courts felt about dischargeability (Count II) before deciding whether they should expend the resources to litigate the debt (Count I). If Count II could not be saved, the real fight would move to the proof-of-claim process and effort spent adjudicating Count I would be largely wasted.
So the parties struck a deal. They voluntarily dismissed Count I, without prejudice, “so as to give rise to a final order from which an appeal of the dismissal of Count II of the Complaint may be taken.” J.A. 63. Having done so, the Kivitis appealed their Count II loss to the district court, who affirmed it. They then appealed to this Court. II. Discussion The district court did not have jurisdiction over the Kivitis’ appeal. District courts have jurisdiction to hear appeals from “final judgments, orders, and decrees” entered in bankruptcy “cases and proceedings.” 1 28 U.S.C. § 158(a). The bankruptcy court’s order
1
They also sometimes have jurisdiction to hear interlocutory bankruptcy appeals.
28 U.S.C. § 158(a)(2), (3). This is not one of those times.
7
dismissing Count II was not a final order when entered because Count I remained. And the parties cannot collude to create finality after the fact through a voluntary dismissal without prejudice. See Waugh Chapel S. v. United Food and Com. Workers Union Local 47, 728 F.3d 354, 359 (4th Cir. 2013). So there was no final order, which means the district court had no jurisdiction.
A. The Order was not final when entered We, as appellate judges, are used to weeding out interlocutory appeals masquerading as final-order appeals. That is because many appeals we get purport to be final-order appeals and it is up to us to make sure they are. See 28 U.S.C. § 1291 (“The courts of appeals . . . shall have jurisdiction of appeals from all final decisions of the district courts.”); Va. Dep’t of Corr. v. Jordan, 921 F.3d 180, 187 (4th Cir. 2019) (“[C]ourts must always assure themselves of subject matter jurisdiction before reaching the merits.”). And while we sometimes struggle to describe district-court finality at the margins, at base, a district-court order is final when it “ends the litigation on the merits and leaves nothing for the court to do but execute the judgment.” Britt v. DeJoy, 45 F.4th 790, 792 (4th Cir. 2022) (en banc) (quoting Catlin v. United States, 324 U.S. 229, 233 (1945)).
Finality plays a similar role in bankruptcy. Parties to a bankruptcy can appeal bankruptcy-court decisions to district courts. But they generally can appeal only “final” judgments and orders. See § 158(a). 2 So a district-court judge considering whether she
2
Section 158(a) governs appeals from dispositive bankruptcy-court decisions. See 28 U.S.C. § 157(b)(1), (c)(2). A separate provision controls “objections” to non-binding, “proposed findings and conclusions.” See id. § 157(c)(1). This opinion deals with only the former.
8
can review a bankruptcy-court order engages in an analysis similar to our analysis when considering whether we can review a district-judge’s order.
Similar. Not identical. The fact is that “the concept of finality in bankruptcy cases”
is “more pragmatic and less technical.” McDow v. Dudley, 662 F.3d 284, 287 (4th Cir. 2011). Bankruptcy cases are often a conglomeration of multiple discrete disputes that “but for the status of the bankrupt party . . . would be separate, stand-alone lawsuits.” In re James Wilson Assocs., 965 F.2d 160, 166 (7th Cir. 1992). So we call an order that “definitively” disposes of one of these discrete disputes “final,” and allow a party to immediately appeal it, even if it does not dispose of the broader bankruptcy case. Ritzen Grp., Inc. v. Jackson Masonry, LLC, 140 S. Ct. 582, 586–87 (2020). But we only do so because the Bankruptcy Code allows us to. While § 1291 facilitates review of final decisions in district-court cases, § 158(a) allows for appeals from final judgments, orders, and decrees entered in both bankruptcy “cases and proceedings,” 28 U.S.C. § 158(a) (emphasis added). The rules are different. Bullard v. Blue Hills Bank, 575 U.S. 496, 501 (2015). Said another way, to be appealable, the challenged bankruptcy decision does not have to end the entire bankruptcy case; it just has to end a proceeding inside the case. See In re Boca Arena, 184 F.3d at 1286; Bullard, 575 U.S. at 501–02; cf. Britt, 45 F.4th at 792.
The initial step in the analysis is perhaps the hardest, if only because it is the most novel. To decide whether a bankruptcy-court order is final, a court first needs to know what the proper scope of the proceeding, or “discrete dispute,” adjudicated below was. See Bullard, 575 U.S. at 501–02 (quoting Howard Delivery Serv., Inc. v. Zurich Am. Ins. Co., 547 U.S. 651, 657 n.3 (2006)). By this we mean the court needs to find the “appropriate
9
procedural unit for determining finality.” Ritzen, 140 S. Ct. at 588. Once that unit is determined, the process becomes more familiar: we apply the finality principles from our § 1291 jurisprudence to that unit. See 1 Collier on Bankruptcy ¶ 5.08[1][b] (2023) (“Once the courts [determine the appropriate procedural unit] . . ., the principles developed under section 1291 will control the determination of finality.”); In re Integrated Res., Inc., 3 F.3d 49, 52–53 (2d Cir. 1993).
The appropriate procedural unit for determining finality here is the adversary proceeding. Adversary proceedings are discrete disputes. They are cabined off from the main bankruptcy case by their own complaint, filing fee, motions, subset of parties, docket numbers, and judgment. In re Ayre, 360 B.R. 880, 885 (C.D. Ill. 2007). As a result, adversary proceedings resemble “stand-alone lawsuits” brought inside the bankruptcy. See In re Boca Arena, 184 F.3d at 1286. So an order ending the litigation on the merits in an adversary proceeding is immediately appealable even when “the umbrella bankruptcy case remains pending.” See Ritzen, 140 S. Ct. at 586–87; In re Boca Arena, 184 F.3d at 1286; In re AroChem Corp., 176 F.3d 610, 619 (2d Cir. 1999); In re La. World Exposition, Inc., 832 F.2d 1391, 1396 (5th Cir. 1987).
Likewise, an order only partially ending the adversary proceeding is not. 1 Collier on Bankruptcy ¶ 5.08[5] (2023) (“[O]nce one identifies the adversary proceeding . . . that is a separate judicial unit for purposes of determining finality, it becomes clear that orders entered during the course of that proceeding that leave the merits to be determined are interlocutory.”). Since adversary proceedings are “essentially a separate civil proceeding within the bankruptcy proceeding,” In re Ayre, 360 B.R. at 885, many of the Federal Rules
10
of Civil Procedure apply to them, including those governing finality, see Fed. R. Bankr. P. 7054(a). And under those Rules—unless a judge “expressly determines” otherwise—an order dismissing fewer than all the claims against a defendant is not final. Fed. R. Civ. P. 54(b) (absent specific finding “any order or other decision, however designated, that adjudicates fewer than all the claims . . . does not end the action as to any of the claims . . . .”); see also Fox v. Baltimore City Police Dep’t, 201 F.3d 526, 530 (4th Cir. 2000) (“Ordinarily, a district court order is not ‘final’ until it has resolved all claims as to all parties.”).
Put simply, “the ‘discrete dispute’ is the adversary proceeding itself, not a particular claim within that proceeding.” Ayers v. U.S. Dep’t of Def., 819 F. App’x 180, 181 (4th Cir. 2020) (per curiam). So an order dismissing only one claim in a multi-claim adversary proceeding does not amount to a final order. Id.; In re Esteva, 60 F.4th 664, 672–73 (11th Cir. 2023).
The bankruptcy court’s order was thus not final when entered. The order dismissed Count II with prejudice, ending litigation on the merits for that claim. Yet it did not dismiss Count I. That count had a ways to go before it was finally adjudicated. Consequently, the court had more to do than “execute the judgment.” Britt, 45 F.4th at 792. So the partial dismissal was not an appealable final order. 3
3
This conclusion stands even though the order determined that any debt was dischargeable. Dischargability orders are core proceedings under § 157(b)(2), which is a “‘textual clue’ that Congress viewed” them as discrete disputes. See Ritzen, 140 S. Ct. at 590. But that fact alone does not “clinch the matter.” Id. (cleaned up). Since the dischargability decision did not “conclusively resolve” the adversary proceeding, id. at 588, it was not a final order.
11
B. The parties cannot manufacture finality The parties will be unsurprised by our holding that the bankruptcy court’s order was not final—and so not appealable—when entered. After all, that is why they agreed to dismiss Count I. They thought that, by doing so, they would make the order final because there would no longer be anything left to adjudicate in the adversary proceeding. J.A. 63 (agreeing to dismiss Count I “so as to give rise to a final order from which an appeal . . . may be taken”); J.A. 73 (“The instant appeal is from an interlocutory order that became a final order upon dispensation of the remaining cause of action below.”).
They were wrong. They cannot “use voluntary dismissals as a subterfuge to manufacture jurisdiction for reviewing otherwise non-appealable, interlocutory orders.” See Waugh Chapel S., 728 F.3d at 359. When Congress requires finality, we must ensure that “every matter in the controversy . . . [is] decided in a single appeal.” Microsoft Corp. v. Baker, 582 U.S. 23, 36 (2017) (quoting McLish v. Roff, 141 U.S. 661, 665–66 (1891)). Yet, if we allowed the parties to appeal Count II now, there would be nothing to stop them from reinstating—and then separately appealing—Count I down the line. See J.A. 63 (agreeing that the dismissal is “without prejudice to the [Kivitis’] right to amend their pleadings so as to seek a finding of liability should there be an appellate remand”). Such tactics impermissibly “erode the finality principle” Congress enacted. See Microsoft, 582 U.S. at 37; see also In re AroChem Corp., 176 F.3d at 619 (explaining that the “flexible” nature of bankruptcy finality does not “overcome the general aversion to piecemeal appeals” (cleaned up)). So the voluntary dismissal did not make the bankruptcy court’s earlier, partial dismissal final.
12
True, some partial dismissals are final and appealable after the parties voluntarily dismiss the remaining claims. See Affinity Living Grp., LLC v. StarStone Specialty Ins. Co., 959 F.3d 634, 638–39 (4th Cir. 2020). But that’s only when the district court dismisses some claims and, in the process, makes it legally impossible to prevail on the remaining claims, even while allowing them to limp on. 4 Id. In this sense, the litigants do not impermissibly create finality by voluntarily dismissing the doomed claims; they merely recognize that it already effectively exists. See id. at 639 (hearing an appeal from a case that was “legally over” even before the voluntary dismissal).
That is not what happened here. These parties set out to create finality, not recognize it. The adversary proceeding was not “legally over” after the bankruptcy court’s partial dismissal. See id. Count II’s dismissal did not mean the Kivitis could not prevail on Count I. Count I asked whether Bhatt was in debt to the Kivitis—i.e., whether Bhatt violated D.C. law by renovating the Kivitis’ house without a license and thus owed them money. Count II asked whether any such debt was dischargeable—i.e., whether Bhatt obtained such a debt through “false pretenses, a false representation, or actual fraud.” See 11 U.S.C. § 523(a)(2)(A). Just because the bankruptcy court found that Bhatt had obtained
4
This distinction is more than semantic. To illustrate, compare Microsoft with Affinity Living Group. In Microsoft, the district court’s partial order denied the plaintiffs’ motion for class certification. 582 U.S. at 27. But that denial had no effect on the merits of the claims the parties voluntarily dismissed. Legal success was just as likely as it had been before the order, any resulting pay out would just be less lucrative. But, in Affinity Living Group, the district court’s partial order dismissed two of the plaintiffs’ four claims on the ground the defendant had no duty under the policy to the plaintiff. 959 F.3d at 639. And that duty was an essential element of the two claims that remained. Id. So the order meant there was no way for the plaintiff to legally succeed on the claims—they were “legally over” before they were voluntarily dismissed. Id.
13
no debt through fraud did not mean the Kivitis were wrong that he violated D.C. law. Count I was still very much alive.
The parties agree that the Kivitis could have prevailed on Count I’s merits even after Count II’s dismissal. In fact, they maintained their proof of claim which sought the same monies. Yet the Kivitis still argue that their voluntary dismissal recognized rather than created finality because, they say, their Count II loss rendered the adversarial proceeding moot. According to them, a moot proceeding is “legally over” and so this appeal falls within Affinity Living Group’s safe harbor.
The root of their mootness argument is that, without Count II, any judgment they won via Count I could not be collected outside bankruptcy and their $58,770 would be recovered—if at all—only through the bankruptcy’s proof-of-claims process. So once Count II failed, they argue, Count I became legally moot. A case is moot when “it is impossible for a court to grant any effectual relief whatever” to the complaining party. 5 Mission Prod. Holdings, Inc. v. Tempnology, LLC., 139 S. Ct. 1652, 1660 (2019) (quoting Chafin v. Chafin, 568 U.S. 165, 172 (2013)). Recall that if a debt is discharged, it cannot be collected outside the bankruptcy proceedings. So if the bankruptcy court agreed Bhatt owed the Kivitis money (success on Count I), but determined that debt dischargeable
5
To be clear, we are describing Article III mootness and not “equitable mootness.”
So-called “equitable mootness” is not real mootness but a pragmatic doctrine particular to bankruptcy under which appellate courts dismiss an appeal when “changes to the status quo following the order being appealed make it impractical or inequitable to ‘unscramble the eggs.’” In re Castaic Partners II, LLC, 823 F.3d 966, 968 (9th Cir. 2016); In re Bate Land & Timber, LLC, 877 F.3d 188, 195 (4th Cir. 2017). It is not implicated here.
14
(failure on Count II), the Kivitis could not directly use the judgment they won from Count I to force Bhatt to pay them outside of bankruptcy. Their only means of recovery would be within bankruptcy, via their already filed proof of claim. 6 The Kivitis thus contend that they could not get “any effectual relief” from the adversary proceeding, rendering it moot. Appellant’s Supp. Br. at 10 (“[T]he Bankruptcy Court’s dismissal of Messrs. Kiviti’s claim . . . caused the remaining case to become moot, and to accordingly strip the Bankruptcy Court of Article III jurisdiction.”).
The Kivitis’ argument is clever; but it misses at least one critical link: Mootness is an Article III doctrine, and bankruptcy courts are not Article III courts. Mootness arises out of Article III’s “case-or-controversy” requirement. The United States’s judicial Power extends only to cases or controversies. U.S. Const. art. III, § 2. To be a case or controversy, parties must have a “‘personal stake in the outcome’ of the lawsuit,” at each stage of the litigation. Lewis v. Cont’l Bank Corp., 494 U.S. 472, 478 (1990) (quoting Los Angeles v. Lyons, 461 U.S. 95, 101 (1983)); Campbell-Ewald Co. v. Gomez, 577 U.S. 153, 161 (2016). If they lose that stake, their case drifts beyond the judicial Power and becomes moot. See United States v. Payne, 54 F.4th 748, 751 (4th Cir. 2022). But since bankruptcy courts are not Article III courts, they do not wield the United States’s judicial Power. Stern v.
6
That is not to say that the Kivitis or Bhatt could not use the resolution of Count I indirectly, inside the greater bankruptcy case. Depending on who prevailed on Count I, that ruling could affect the proof of claim—filed before the same judge, in the same case, and based on the same legal theory. Such indirect use of the resolution of Count I, however, does not impact the Kivitis’ argument that the adversary proceeding was moot.
15
Marshall, 564 U.S. 462, 503 (2011). 7 So they can constitutionally adjudicate cases that would be moot if heard in an Article III court. 8 To be sure, a bankruptcy case must—at the start—be within the judicial Power.
Section 1334 grants near-exclusive jurisdiction over bankruptcy matters to federal district courts. 28 U.S.C. § 1334(a). The district court may then refer a bankruptcy case to a bankruptcy judge (who serves as a unit of the district court). 28 U.S.C. §§ 157(a), 151. 9 But, of course, a district court can only refer a case that it has jurisdiction over. See Ex parte McCardle, 74 U.S. 506, 514 (1868) (“Without jurisdiction the court cannot proceed at all in any cause.”). So before a bankruptcy case is referred to a bankruptcy court, the case must satisfy Article III. See In re Curtis, 571 B.R. 441, 447 (B.A.P. 9th Cir. 2017) (“[T]he bankruptcy court’s power to hear, or to hear and determine, as the case may be,
7
At least, they do not do so lawfully. See Stern, 564 U.S. at 469. That is not to say they do not adjudicate matters that Article III courts could also adjudicate. They do. See id. at 488 (citing Murray’s Lessee v. Hoboken Land & Improvement Co., 59 U.S. 272, 284 (1856)). Our point here is that they are not constitutionally limited to deciding only matters an Article III court could adjudicate.
8
The harder question may be why they can constitutionally adjudicate cases that are within the judicial Power and so could be heard in Article III courts. See Oil States Energy Servs., LLC v. Greene’s Energy Grp., LLC, 138 S. Ct. 1365, 1372–73 (2018) (“Congress cannot confer the Government’s judicial Power on entities outside Article III.” (cleaned up)); cf. Granfinanciera, S.A. v. Nordberg, 492 U.S. 33, 56 n.11 (1989); Stern, 564 U.S. at 492 n. 7; Wellness Int’l Network, Ltd. v. Sharif, 575 U.S. 665, 678–79 (2015). But we need not dive into this question here.
9
In practice, referrals occur automatically because virtually all district courts have entered standing orders of reference. 1 Collier on Bankruptcy ¶ 3.02[1] (2023). These orders refer every bankruptcy case to bankruptcy judges so that district courts don’t need to make case-by-case referrals.
16
bankruptcy cases and proceedings is entirely dependent upon the referral by the district court.”).
So too must Article III be satisfied after the bankruptcy court acts and the case is returned to the district court from the bankruptcy court. Every action by a district court is constrained by Article III, including reviewing a bankruptcy court order. So a district court has no authority to act without an existing constitutional case or controversy. See In re Thorpe Insulation Co., 677 F.3d 869, 880 (9th Cir. 2012); In re Croniser, No. 22-1227, 2022 WL 7935991, at *1 (4th Cir. Oct. 14, 2022) (unpublished).
But that limit on the district court’s authority does not constrain the bankruptcy court. Once a case is validly referred to the bankruptcy court, the Constitution does not require it be an Article III case or controversy for the bankruptcy court to act. See In re Technicool Sys., Inc., 896 F.3d 382, 385 (5th Cir. 2018) (“Bankruptcy courts are not Article III creatures bound by traditional standing requirements.”). 10 That requirement comes from the Constitution’s limits on the judicial Power. Bankruptcy courts do not wield judicial Power. End of story.
At least that is the end of the constitutional story. The statutory story—i.e., whether bankruptcy courts have statutory authority to decide a constitutionally moot matter—is
10
Through a separate mechanism, district courts can refer cases or proceedings to a bankruptcy court while retaining the exclusive authority to enter “any final order[s] or judgment[s].” See 28 U.S.C. § 157(c)(1). In that situation, the bankruptcy court can only recommend a disposition. Id. Since that case or proceeding is truly being adjudicated by the district court, Article III’s constraints may apply. See, e.g., In re Kaiser Gypsum Co., 60 F.4th 73, 81 (4th Cir. 2023). That is not this case’s posture.
17
more complex. Still, the tale concludes the same way: a bankruptcy court can adjudicate a constitutionally moot matter.
Bankruptcy courts, as statutory creatures, have whatever power Congress lawfully gives them. So to see if bankruptcy courts can decide matters outside the judicial Power, we check to see if Congress has given them that power. And Congress has said that bankruptcy courts “may hear and determine all [bankruptcy] cases . . . and all core proceedings . . . referred” to them by a district court. 28 U.S.C. § 157(b)(1) (emphasis added). Not just those that could be fully adjudicated in district court. Once a bankruptcy case lands in bankruptcy court, any number of things could happen before the estate’s distribution is settled. As relevant here, the parties could embroil themselves in an adversary proceeding. But whether that proceeding could itself be adjudicated in an Article III court is of no moment. By § 157’s text, a bankruptcy court’s jurisdiction requires only that the case or core proceeding arise under Title 11 and be referred to the bankruptcy court. § 157(b)(1). Section 157 does not require every “discrete dispute[ ],” see Ritzen, 140 S. Ct. at 587, arising post-referral to satisfy Article III. Nor does any other provision.
Against this silence, Congress has elsewhere explicitly imported some Article III-
type requirements onto bankruptcy courts. For example, it created so-called “bankruptcy standing” by giving “parties in interest” a right to be heard, at least in Chapter 11 bankruptcies. See 11 U.S.C. § 1109(b); In re Capital Contracting Co., 924 F.3d 890, 895
18
(6th Cir. 2019) (discussing § 1109(b)). 11 And it also codified some version of mootness for real-property cases. See 11 U.S.C. § 363(m); In re Rare Earth Mins., 445 F.3d 359, 363 (4th Cir. 2006) (explaining that § 363(m) “creates a rule of ‘statutory mootness’”). Yet these principles are not the same as Article III’s limits and Congress has never imported all those limitations. We refuse to make that choice for it; indeed we cannot do so. See Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 96 (2012) (“[W]hat a text does not provide is unprovided.”).
A few bankruptcy courts confronting this issue have disagreed. They point to a district court’s authority to “withdraw, in whole or in part, any case or proceeding” referred to a bankruptcy court. See 28 U.S.C. § 157(d). They also think it significant that Congress called bankruptcy courts “unit[s] of the district court,” with judges that are “judicial officer[s]” thereof. See § 151. Following these breadcrumbs, those courts have held that a bankruptcy court’s power depends on a district court’s power—and thus evaporates if the
11
“Bankruptcy standing” under § 1109(b) refers to a party’s ability to object in the bankruptcy court. See In re Cap. Contracting Co., 924 F.3d at 894–96; In re Kaiser Gypsum Co., 60 F.4th at 81–82. This is distinct from “bankruptcy appellate standing,” which concerns a party’s ability to appeal a bankruptcy order. In re Kaiser Gypsum Co., 60 F.4th at 81–82. Courts have split over whether bankruptcy standing simply incorporates Article III’s requirements or imposes more stringent limitations. Compare In re Global Indus. Techs., Inc., 645 F.3d 201, 211 (3d Cir. 2011) (bankruptcy standing is coextensive with Article III standing) with In re Tower Park Props., LLC, 803 F.3d 450, 456–57 & n.6 (9th Cir. 2015) (bankruptcy standing is more restrictive than Article III standing); see also In re Cap. Contracting Co., 924 F.3d at 895. They are also split over whether it applies outside of Chapter 11 bankruptcies. See id. And whether bankruptcy appellate standing survives the Supreme Court’s decision in Lexmark International v. Static Control Components, Inc., 572 U.S. 118 (2014), is an open question. See In re Kaiser Gypsum Co., 60 F.4th at 82. We need not confront these issues today. We simply reference § 1109(b) because it is evidence that when Congress wants to impose certain Article III-esque requirements on bankruptcy courts, it does so explicitly.
19
case strays outside of Article III’s bounds. See, e.g., In re Kilen, 129 B.R. 538, 542–43 (Bankr. N.D. Ill. 1991) (“In light of the derivative nature of the bankruptcy court’s power, it is obvious that the constitutional standards of Article III which bind the district court also bind the bankruptcy court.”); In re Interpictures, Inc., 86 B.R. 24, 28–29 (Bankr. E.D.N.Y. 1988); but cf. Stern, 564 U.S. at 500–01. 12 We are unconvinced. Congress requires bankruptcy courts to get cases from district courts and allows district courts to withdraw cases. And the Constitution limits the cases a district court can refer or withdraw. But these two facts do not add up to a limit on the types of matters a bankruptcy court can adjudicate after referral, nor a requirement that those matters be eligible for withdrawal. That could be how it works—if Congress said so. Yet Congress has not said so. So that is not how it works.
In the same vein, we refuse to overread Congress’s designation of bankruptcy courts as “unit[s]” of the district court. See 28 U.S.C. § 151. Certainly, district courts oversee many aspects of bankruptcy courts. For example, when bankruptcy judges consider matters within the judicial Power, district courts can—indeed sometimes must—review those actions. § 157(a), (b)(1), (c)(1). But bankruptcy courts are not “mere adjuncts” of
12
Although it did not, language in our opinion, In re Grewe, 4 F.3d 299 (4th Cir.
1993), could be read as endorsing this view. In deciding whether bankruptcy courts are “court[s] of the United States” as the term is used in an attorney’s fees provision, we said that “bankruptcy courts are, for jurisdictional purposes, inseparable from the district court.” See, e.g., id. at 304. But there we were concerned with the statutory power to award attorney’s fees, not Article III’s case-or-controversy limitation. So that case does not answer this question. It merely reminds us that “jurisdiction” “is a word of many, too many, meanings.” Arbaugh v. Y&H Corp., 546 U.S. 500, 510 (2006) (quoting Steel Co. v. Citizens for Better Env’t, 523 U.S. 83, 90 (1998)).
20
district courts. Stern, 564 U.S. at 487. They exercise “broad powers” under their own statutory grant of jurisdiction. See id. at 488. Congress did not impliedly limit that express grant through cryptic labelling in a separate provision.
The argument to the contrary depends on finding something inherent about the words “cases” and “proceedings” in the bankruptcy jurisdictional provisions that brings them necessarily within the judicial Power. It assumes that when Congress gave federal courts (Article III and non-Article III alike) jurisdiction over bankruptcy “cases” and “proceedings,” it imbued those words with Article III’s limits. So that, even absent Article III, the statutes themselves would require the same level of adversariness.
But we do not ordinarily interpret a jurisdictional statute’s constraints to be the same as the Constitution’s. In fact, we often go out of our way to create differences and draw distinctions even where—unlike here—the statute uses Article III’s precise language. See Navy Fed. Credit Union v. LTD Fin. Servs., LP, 972 F.3d 344, 352 (4th Cir. 2020) (“Unlike the constitutionally permitted ‘minimal diversity’ jurisdiction, diversity must be ‘complete’ to satisfy this Congressional grant.” (citing Strawbridge v. Curtiss, 7 U.S. 267, 267 (1806)).
If we did, our caselaw would look very different. The constitutional-jurisdiction inquiry would often collapse into the statutory one. Yet it doesn’t. For example, litigants can have statutory jurisdiction to sue States even if the Constitution forbids it. See, e.g., Hans v. Louisiana, 134 U.S. 1, 10 (1890). More to the point, we dismiss moot cases because they are no longer Article III cases, not because they fall outside the scope of a statute. See, e.g., Eden, LLC v. Justice, 36 F.4th 166, 169–72 (4th Cir. 2022). So we refuse to read in Article III’s limits.
21
To recap, bankruptcy courts are not Article III courts. So Article III constraints—
such as mootness—do not apply to them as a matter of constitutional law. They only apply if Congress said so in a statute. But it hasn’t. And that means whether Count I was constitutionally moot is beside the point. The bankruptcy court could still adjudicate it.
Since the Kivitis cannot argue that their adversary proceeding was constitutionally moot when Count II was dismissed, they have not shown the proceeding was legally doomed when they dismissed Count I. 13 They are thus left arguing the order was final because Count I was practically over post-dismissal. See, e.g., Appellant’s Supp. Br. at 4 (claiming there was “no judicial economy” to pursuing Count I before appealing Count II). Yet the Supreme Court rejected this exact reasoning in Microsoft. The Microsoft plaintiffs also thought it “economically irrational” to litigate their still-legally-viable claims to final judgment. Microsoft Corp., 582 U.S. at 34. Still, the Court refused to let them create finality by voluntarily dismissing those claims. Id. at 36; see also Affinity Living Grp., 959 F.3d at 639 (distinguishing between claims that are “legally” and merely “practically” over). Here too, it is irrelevant that the parties think there is “no judicial economy” in litigating Count I to final judgment before they appeal the order dismissing only Count II. Count I is legally viable, and its dismissal was without prejudice, so that dismissal did not
13
The Kivitis raised no other arguments for why the proceeding was legally over when Count II was dismissed and so we do not address any. This means we do not consider any of the Bankruptcy Code’s codified “standing-esque requirements.” See In re Cap. Contracting Co., 924 F.3d at 895. We thus offer no opinion on whether a bankruptcy court may dismiss a claim in an adversary proceeding that it views—in a non-constitutional sense—as “moot.” The Kivitis have not argued it could.
22
create a final order under § 158(a). And that means the district court lacked jurisdiction to review it.
* * *
Lower federal courts have only the power given to them by Congress. But they
have every inch of that power. That is true for us. And it is true for bankruptcy courts. For district courts, that means—irrelevant exceptions aside—they can hear appeals only from final bankruptcy orders. For bankruptcy courts, that means they are essentially unencumbered by Article III’s case-or-controversy requirement. The Kivitis would have us ignore Congress on both fronts. They, like Bhatt, implore us to review the bankruptcy court’s non-final order despite Congress’s explicit instructions to the contrary. They then beseech us to etch one limit on our power into the bankruptcy code. We decline their entreaties. Having done so, it is clear that the district court reviewed a non-final order. Since it had no jurisdiction to do so, its order is
VACATED
AND REMANDED.