This appeal presents a question of successor liability under the Comprehensive Environmental Response Compensation and Liability Act of 1980 ("CERCLA"), 42 U.S.C. §§ 9601-9675 (2000). Specifically, the question before us is whether federal common law for purposes of determining corporate successor liability under CERCLA incorporates state law — in this case, New York law — or displaces state law in favor of a uniform national rule derived from traditional common-law principles. We must then determine whether, under the applicable governing law, defendant-counterclaimant-appellee National Service Industries, Inc. ("NSI") is liable as the legal successor to Serv-All Uniform Rental Corp. ("Serv-All"). For the reasons that follow, we find that we need not decide whether federal common law under CERCLA would displace state law because the claim of plaintiff-counter-defendant-appellant the State of New York ("the State") would fail under either New York or traditional common law. That is, we hold that New York follows the traditional common-law rules of successor liability, which would govern if state law did not provide the rule of decision, and which require some evidence of continuity of ownership to find the existence of a de facto merger, and thus, successor liability. For that reason, we also hold that the State's common-law unjust enrichment and restitution claims fail.

The facts underlying this appeal are undisputed and have been set forth in New York v. National Services Industries, Inc., 134 F. Supp. 2d 275, 276-77 (E.D.N.Y.2001) ("NSI I") and New York v. National Service Industries, Inc., 380 F. Supp. 2d 122, 123-26 (E.D.N.Y.2005) ("NSI III"), familiarity with which is presumed. We repeat only those facts necessary to understand this appeal.

From 1962 to 1988, Serv-All operated a uniform rental business that served customers on Long Island. It delivered its clients clean uniforms, and picked up and cleaned the soiled ones. Until the mid-1980s, Serv-All cleaned the uniforms itself using a dry-cleaning process. In 1978, it arranged to have several dozen drums of perchloroethelyne, a hazardous substance used in dry cleaning, disposed of at the Blydenburgh Landfill in Islip, New York. In soil, perchloroethelyne decomposes into several chemicals, including vinyl chloride. Both perchloroethelyne and vinyl chloride are considered hazardous substances under CERCLA. See 42 U.S.C. § 9602; 40 C.F.R. § 302.4 (2005). Since 1983, the Blydenburgh Landfill has been listed in the New York Registry of Hazardous Waste Sites and in 1987 was added to the National Priority List of contaminated sites. Because Serv-All arranged for the disposal of the perchloroethelyne, the parties do not dispute that Serv-All was a potentially responsible party under CERCLA. See 42 U.S.C. § 9607(a)(3).

In October 1988, Serv-All sold almost all of its assets, including its customer list, current contracts, trucks, good will, and the right to use its name to Initial Service Investments ("Initial") for approximately $2.2 million. Ralph Colantuni and William Lepido, Serv-All's owners, also signed a covenant not to compete with Initial in the area for seven years. Initial operated the uniform rental business in large measure as Serv-All had done before the sale. It used the Serv-All name and trucks, and employed some of Serv-All's management and support personnel and all but one of Serv-All's former drivers. Initial did not continue Serv-All's practice of dry-cleaning uniforms, however, but instead laundered them in water at one of its facilities.

After the transaction, Serv-All changed its name to C-L Dissolution Corporation and adopted a liquidation plan. On January 27, 1989, C-L Dissolution Corporation formally dissolved. In 1992, NSI bought all shares of Initial's stock. On August 31, 1995, Initial merged into NSI.

Meanwhile, the State conducted a cleanup of the Blydenburgh Landfill, for which it incurred response costs that, as of 2002, exceeded $12 million with interest.1

In 1999, the State sued NSI to recover its response costs under CERCLA § 107(a), 42 U.S.C. § 9607(a), and for unjust enrichment and restitution under state law. Although a corporation that purchases another corporation's assets generally is not liable for the debts of the seller, the State asserted that NSI was liable because it fit into one of the four exceptions to this rule. See New York v. Nat'l Serv. Indus., Inc., 352 F.3d 682, 683 (2d Cir.2003) ("NSI II"). Specifically, the State argued that NSI was the "mere continuation" of Serv-All pursuant to the "substantial continuity" test we adopted in B.F. Goodrich v. Betkoski, 99 F.3d 505 (2d Cir.1996). In Betkoski, we concluded that "the substantial continuity test [is] the appropriate legal test" for determining whether a corporation that purchased another's assets is a mere continuation of the prior business and thus liable for its debts. Id. at 519. The substantial continuity test was a CERCLA-specific rule that we held displaced the narrow and more restrictive state common-law rules for determining whether an asset purchaser is a mere continuation of the seller. Id.

Relying on our decision in Betkoski, the district court (Mishler, J.) held that NSI was the legal successor to Serv-All and therefore responsible for Serv-All's CERCLA liability. See NSI I, 134 F.Supp.2d at 280-81. It consequently entered judgment for the State. NSI filed a timely appeal, maintaining that the substantial continuity test adopted in Betkoski was not good law after the Supreme Court's decision in United States v. Bestfoods, 524 U.S. 51, 118 S. Ct. 1876, 141 L. Ed. 2d 43 (1998). In Bestfoods, a case involving parent-subsidiary liability under CERCLA, the Supreme Court rejected the test that the district court had applied in that case because it "disregard[ed] entirely . . . time-honored common-law rule[s]." Id. at 70, 118 S. Ct. 1876. The Court held that nothing in CERCLA indicated that Congress intended to abrogate the common law, and thus a parent corporation could be charged with its subsidiary's wrongdoing only when the corporate veil could be pierced. Id. at 63-64, 70, 118 S. Ct. 1876. While the Supreme Court noted that "[t]here is significant disagreement among courts and commentators over whether, in enforcing CERCLA's indirect liability, courts should borrow state law, or instead apply a federal common law of veil piercing," it did not resolve the question. Id. at 63 n. 9, 118 S. Ct. 1876.

In considering NSI's first appeal, we took from the Supreme Court's decision in Bestfoods "the principle that when determining whether liability under CERCLA passes from one corporation to another, we must apply common law rules and not create CERCLA-specific rules." NSI II, 352 F.3d at 685. "Because the substantial continuity test in Betkoski departs from the common law rules of successor liability," we held, "Betkoski is no longer good law." Id. As we explained in NSI II, the traditional common-law rules provide that a corporation that acquires another corporation's assets assumes the latter's liabilities only if (1) the successor corporation expressly or impliedly agrees to do so, (2) the transaction may be viewed as a de facto merger, (3) the successor is the mere continuation of the predecessor, or (4) the transaction is fraudulent. Id. Accordingly, we vacated the district court's grant of summary judgment and remanded the case to the district court for further proceedings.

In doing so, we declined to decide whether courts should apply state law or create a federal rule based on traditional common-law principles for purposes of determining successor liability under CERCLA. See id. at 687 n. 1. Nevertheless, we observed in NSI II that, although we had chosen a federal rule in Betkoski, the analysis for determining whether federal common law would absorb or displace state law might produce a different result after Bestfoods. See id.

On remand, NSI moved for summary judgment on the ground that it was not the legal successor to Serv-All. The State responded that NSI was liable pursuant to the "de facto merger" theory of successor liability. See NSI III, 380 F.Supp.2d at 128-29. The de facto merger doctrine creates successor liability when the transaction between the purchasing and selling companies is in substance, if not in form, a merger.

In granting NSI's motion, the district court (Townes, J.), considered the choice-of-law question left open in NSI II: whether CERCLA requires the displacement of state law for purposes of determining successor liability. Id. at 128. The court concluded that it did not need to decide the question, however, because the State's claim would fail regardless of which law applies. Id. The court found that general common law principles require continuity of ownership before a de facto merger can be found to have occurred. Id. at 129-30. Although the New York Court of Appeals has not explicitly held that continuity of ownership is required to find a de facto merger, the district court predicted that it too would require continuity of ownership under state law. See id. at 131. Because there is no evidence of continuity of ownership in this case, the district court held that NSI is not Serv-All's successor for purposes of the State's common-law or CERCLA claims. See id. at 134. The district court therefore entered judgment for NSI dismissing the State's claims.

On appeal, the State argues that New York law should govern whether NSI is the legal successor to Serv-All with respect to its CERCLA and common law claims. Because it contends that New York law is unsettled with respect to whether continuity of ownership is required to find a de facto merger, it asks us to certify this question to the New York Court of Appeals.

We review de novo a district court's grant of summary judgment, construing all evidence and drawing all reasonable inferences in favor of the State, the non-moving party. See Syms v. Olin Corp., 408 F.3d 95, 100 (2d Cir.2005).

The first question before us is what law provides the rule of decision governing corporate successor liability under CERCLA. We conclude, like the district court did, that we need not decide this question because the outcome would be the same whether we apply state law or a national rule derived from traditional common-law principles. To reach this conclusion, however, we review the analysis by which CERCLA choice-of-law questions should be resolved.

CERCLA does not specifically provide that a successor corporation may be held liable for response costs. Nevertheless, we have held that CERCLA encompasses successor liability. See NSI II, 352 F.3d at 685. Thus, strictly speaking, federal law (i.e., federal common law) governs successor liability under CERCLA. See United States v. Kimbell Foods, Inc., 440 U.S. 715, 727, 99 S. Ct. 1448, 59 L. Ed. 2d 711 (1979). In giving content to that law, however, we must decide whether to "adopt state law or ... fashion a nationwide federal rule." Id. at 728, 99 S. Ct. 1448. Because federal common law governs the State's claim whether we apply state law or a national rule, the terminology can be confusing. To clarify, we use the term "traditional common law" to refer to the law that would govern were we to hold that state law should be displaced in favor of a national rule. See Bestfoods, 524 U.S. at 70, 118 S. Ct. 1876 (explaining that federal common law under CERCLA should not depart from traditional common law principles); NSI II, 352 F.3d at 685 (concluding that federal common law for purposes of successor liability under CERCLA should apply traditional common law principles). We use "state law" or "New York law" to refer to the rule that would govern were we to hold that federal common law would not displace state law.

As an initial matter, NSI maintains that we do not need to undertake a new choice-of-law analysis. It contends that Betkoski's holding that state law should be displaced is still good law even if its adoption of the substantial continuity test is not. That is, NSI asserts that we are bound by our decision in Betkoski that CERCLA requires a national rule for purposes of determining whether a corporation is a successor-in-interest even though the national rule we adopted in that case has been overruled. We disagree.

In Betkoski, we held that CERCLA required a uniform federal rule for successor liability because doing so allowed us to adopt the substantial continuity test. See Betkoski, 99 F.3d at 519; accord B.F. Goodrich v. Betkoski, 112 F.3d 88, 91 (2d Cir.1997) ("Betkoski II") (clarifying upon the denial of rehearing that "our primary reason for adopting a federal common law rule [in Betkoski] was our concern that allowing state law rules such as the inflexible and easily evaded `identity' rule to control the question of successor liability would defeat the goals of CERCLA"). Because we determined that the substantial continuity test could not stand after Bestfoods, our rationale for displacing state law has been overruled. We should therefore revisit our choice-of-law decision in Betkoski. See BankBoston, N.A. v. Sokolowski (In re Sokolowski), 205 F.3d 532, 534-35 (2d Cir.2000) (holding that we are "bound by a decision of a prior panel unless and until its rationale is overruled, implicitly or expressly, by the Supreme Court or this court en banc" (quotation marks and citation omitted)).

Generally, "[i]n the absence of congressional guidance on the issue of what law to apply, we look to the three-part test enunciated by the Supreme Court in United States v. Kimbell Foods, Inc." VKK Corp. v. Nat'l Football League, 244 F.3d 114, 122 (2d Cir.2001). Under Kimbell Foods, courts determining whether federal common law should displace state law must consider: (1) whether the federal program, by its very nature, requires uniformity; (2) whether application of state law would frustrate specific objectives of the federal program; and (3) whether application of a uniform federal rule would disrupt existing commercial relationships based on state law. 440 U.S. at 728-29, 99 S. Ct. 1448. As the Supreme Court noted recently, the essence of this test is "whether the relevant federal interest warrants displacement of state law." Empire Healthchoice Assurance, Inc. v. McVeigh, 547 U.S. ___, 126 S. Ct. 2121, 2132, 165 L. Ed. 2d 131 (2006) (citing Boyle v. United Techs. Corp., 487 U.S. 500, 507 n. 3, 108 S. Ct. 2510, 101 L. Ed. 2d 442 (1988)).

The choice-of-law question is a complicated one that has led our sister circuits to reach different answers. See, e.g., N. Shore Gas Co. v. Salomon, Inc., 152 F.3d 642, 650 (7th Cir.1998) (noting the circuit split). On one hand, in United States v. Davis, 261 F.3d 1 (1st Cir.2001), for example, the First Circuit found "no evidence that application of state law to the facts of this case would frustrate any federal objective," and held that federal common law therefore would absorb state law to determine the scope of successor liability under CERCLA. Id. at 54. The court concluded that this approach "heed[s] the Supreme Court's warning that courts should presume that matters left unaddressed are subject to state law when a `comprehensive and detailed' federal statutory regime is at issue, and that cases in which the creation of a `special federal rule would be justified' generally are `few and restricted.'" Id. at 53 (quoting O'Melveny & Myers v. Fed. Deposit Ins. Corp., 512 U.S. 79, 85, 87, 114 S. Ct. 2048, 129 L. Ed. 2d 67 (1994)); see also Redwing Carriers, Inc. v. Saraland Apartments, 94 F.3d 1489, 1501 (11th Cir. 1996) (finding that "[a]bsent a showing that state partnership law is inadequate to achieve the goals of CERCLA, we discern no imperative need to develop a general body of federal common law to decide cases such as this" (internal quotation marks omitted)); City Mgmt. Corp. v. U.S. Chem. Co., 43 F.3d 244, 253 (6th Cir.1994) (explaining that state law governs the question of corporate successor liability under CERCLA); Atchison, Topeka & Santa Fe Ry. Co. v. Brown & Bryant, Inc., 159 F.3d 358, 363 (9th Cir.1998) (noting that "[a]lthough often invoked in this context, there has been no real explanation of the need for uniformity in this particular area of successor liability — especially since state law will in many other instances determine whom the EPA may or may not look to for compensation"); Anspec Co., Inc. v. Johnson Controls, Inc., 922 F.2d 1240, 1248 (6th Cir.1991) (Kennedy, J. concurring) (arguing that corporations are creatures of state law and therefore state law should determine their liability absent a conflict with federal law).

On the other hand, in a split decision, the Third Circuit recently reaffirmed its pre-Bestfoods precedent that successor liability for purposes of CERCLA should be determined by a uniform, national rule. See United States v. Gen. Battery, 423 F.3d 294 (3d Cir.2005). In doing so, the court observed that no Supreme Court case had explicitly overruled its prior holding. See id. at 299-300. It also concluded that the Supreme Court's approach in Bestfoods supported the use of a national rule based on traditional common law. Id. at 300. In Bestfoods, the Third Circuit found, the Supreme Court had applied "fundamental" and "hornbook" principles of corporate liability, not the law of any particular state. Id. As the Third Circuit explained, "the Supreme Court declined to apply Michigan law and instead looked to the general `hornbook' rule of veil-piercing." Id. Moreover, the General Battery court rejected the reasoning of the First and Ninth Circuits, which, as noted, found no conflict between CERCLA's purposes and the use of state law. Id. at 301-02, 303. By contrast, the Third Circuit concluded that "[a] more uniform and predictable federal liability standard corresponds with specific CERCLA objectives by encouraging settlements and facilitating a more liquid market in corporate and `brownfield' assets." Id. at 302. Similarly, the Fourth Circuit concluded that a national rule is required by "[t]he national interest in the uniform enforcement of CERCLA and the same interest in preventing evasion by a responsible party." United States v. Carolina Transformer Co., 978 F.2d 832, 837 (4th Cir.1992).

As we observed in NSI II, the Kimbell Foods factors appear to favor the absorption (non-displacement) of state law. See NSI II, 352 F.3d at 687 n. 1. Although CERCLA is a federal statute for which there is presumably an interest in uniform application, where there is no conflict between federal policy and the application of state law, "a mere federal interest in uniformity is insufficient to justify displacing state law in favor of a federal common law rule." Betkoski II, 112 F.3d at 91 (citing O'Melveny & Myers, 512 U.S. at 87-88, 114 S. Ct. 2048); VKK Corp., 244 F.3d at 121-22; see also Kamen v. Kemper Fin. Servs., 500 U.S. 90, 98, 111 S. Ct. 1711, 114 L. Ed. 2d 152 (1991) ("The presumption that state law should be incorporated into federal common law is particularly strong in areas in which private parties have entered legal relationships in the expectation that their rights would be governed by state-law standards.... Corporate law is one such area.").

Here, NSI points to no conflict between the application of state law and the federal interests at issue in CERCLA, and we fail to see one. See Empire HealthChoice Assurance, Inc. v. McVeigh, 396 F.3d 136, 141 (2d Cir.2005) (finding no reason to displace state law where "Empire's briefs on appeal fail to mention a single state law or state-imposed duty that runs contrary to the federal interests asserted in this case"), aff'd, ___ U.S. ___, 126 S. Ct. 2121, 165 L. Ed. 2d 131 (2006). New York law embraces successor liability and the tests it uses to define successor liability are not unduly restrictive. In fact, according to the State, New York's standards for finding successor liability under the de facto merger exception are more expansive than traditional common-law principles. Nevertheless, because the State's claim fails under either New York law or traditional common-law principles, we need not, and thus do not, decide whether CERCLA requires the displacement of state law in this instance.

Under both New York law and traditional common law, a corporation that purchases the assets of another corporation is generally not liable for the seller's liabilities. See Gen. Battery, 423 F.3d at 305 (applying traditional common law principles); Schumacher v. Richards Shear Co., 59 N.Y.2d 239, 244-45, 464 N.Y.S.2d 437, 451 N.E.2d 195 (1983) (applying New York law). Both New York law and traditional common law, however, recognize certain exceptions to this rule. Gen. Battery, 423 F.3d at 305; Schumacher, 59 N.Y.2d at 245, 464 N.Y.S.2d 437, 451 N.E.2d 195. Hence, as noted above, a buyer of a corporation's assets will be liable as its successor if: "(1) it expressly or impliedly assumed the predecessor's tort liability, (2) there was a consolidation or merger of seller and purchaser, (3) the purchasing corporation was a mere continuation of the selling corporation, or (4) the transaction is entered into fraudulently to escape such obligations." Schumacher, 59 N.Y.2d at 245, 464 N.Y.S.2d 437, 451 N.E.2d 195; accord N. Shore Gas Co., 152 F.3d at 651 (traditional common-law principles).

The exception for the "consolidation or merger of seller and purchaser" is at issue here. Schumacher, 59 N.Y.2d at 245, 464 N.Y.S.2d 437, 451 N.E.2d 195. The State claims that Serv-All's sale of its assets to Initial (now NSI) was, in fact, a de facto merger of Serv-All into Initial. As noted, "[a] de facto merger occurs when a transaction, although not in form a merger, is in substance `a consolidation or merger of seller and purchaser.'" Cargo Partner AG v. Albatrans, Inc., 352 F.3d 41, 45 (2d Cir.2003) (quoting Schumacher, 59 N.Y.2d at 245, 464 N.Y.S.2d 437, 451 N.E.2d 195). At common law, the hallmarks of a de facto merger include: (1) continuity of ownership; (2) cessation of ordinary business and dissolution of the acquired corporation as soon as possible; (3) assumption by the purchaser of the liabilities ordinarily necessary for the uninterrupted continuation of the business of the acquired corporation; and (4) continuity of management, personnel, physical location, assets, and general business operation. See generally N. Shore Gas Co., 152 F.3d at 652 & n. 6; accord In re Augie/Restivo Baking Co., Ltd., 860 F.2d 515, 520 (2d Cir.1988); 15 William Meade Fletcher et al., Fletcher Cyclopedia of the Law of Private Corporations § 7124.20 (3d ed.2002) (citing the majority rule).

NSI does not dispute, for purposes of this appeal, that the transaction between Initial (now NSI) and Serv-All satisfies the last three elements of the test: Initial assumed Serv-All's customer contracts, kept the Serv-All name, uniforms, vehicles, and many of its employees, and the former Serv-All company ceased to exist soon after the sale. Both parties also agree that there is no continuity of ownership between Initial, the purchasing corporation, and Serv-All, the selling corporation, because the transaction was for cash and there is no evidence that the previous owners of Serv-All have continued to profit from the company in any way.2 Moreover, the State does not dispute that if traditional common law were to supply the rule of decision, the four-part de facto merger test would apply. See NSI I, 352 F.3d at 685 (concluding that the federal common law of CERCLA must not "depart[] from common law rules"). Nor does it challenge the district court's conclusion that under the four-part test there has been no de facto merger because there is no evidence of continuity of ownership between NSI and Serv-All. See NSI III, 380 F.Supp.2d at 129. Instead, the State asserts that New York law is broader than traditional common law and is unsettled with respect to whether a de facto merger requires continuity of ownership. In consequence, it asks this Court to certify this question to the New York Court of Appeals.

The New York Court of Appeals has not directly addressed whether a de facto merger creates liability for a successor corporation, but this Court and lower New York courts have held that it does. See, e.g., Cargo Partner, 352 F.3d at 45-46; Nettis v. Levitt, 241 F.3d 186, 193-94 (2d Cir.2001) (per curiam); AT & S Transp., LLC v. Odyssey Logistics & Tech. Corp., 22 A.D.3d 750, 803 N.Y.S.2d 118, 120 (2d Dep't 2005); In re New York City Asbestos Litig., 15 A.D.3d 254, 789 N.Y.S.2d 484, 486 (1st Dep't 2005); Sweatland v. Park Corp., 181 A.D.2d 243, 587 N.Y.S.2d 54, 55 (4th Dep't 1992). Generally, New York courts have looked to the four traditional common-law factors: whether there is continuity of ownership, continuity of management, a dissolution of the selling corporation, and the assumption of liabilities by the purchaser. See, e.g., In re New York City Asbestos Litig., 789 N.Y.S.2d at 486; Fitzgerald v. Fahnestock & Co., 286 A.D.2d 573, 730 N.Y.S.2d 70, 71 (1st Dep't 2001); Sweatland, 587 N.Y.S.2d at 56; accord Cargo Partner, 352 F.3d at 46 (applying New York law). We must determine, however, whether New York would recognize the existence of a de facto merger where there is no evidence of continuity of ownership.

1. Standards for Ascertaining State Law & Certification

"In the absence of authoritative law from the state's highest court, we must either (1) predict how the New York Court of Appeals would resolve the state law question, or, if state law is so uncertain that we can make no reasonable prediction, (2) certify the question to the New York Court of Appeals for a definitive resolution." DiBella v. Hopkins, 403 F.3d 102, 111 (2d Cir.2005). In attempting to ascertain how the New York Court of Appeals would rule, we give "due regard to the rulings of other state courts." Pinto v. Allstate Ins. Co., 221 F.3d 394, 402 (2d Cir.2000); see Comm'r v. Estate of Bosch, 387 U.S. 456, 465, 87 S. Ct. 1776, 18 L. Ed. 2d 886 (1967) (stating that, although not authoritative, the judgment of an intermediate appellate state court "is a datum for ascertaining state law which is not to be disregarded by a federal court unless it is convinced by other persuasive data that the highest court of the state would decide otherwise") (quoting West v. Am. Tel. & Tel., Co., 311 U.S. 223, 237, 61 S. Ct. 179, 85 L. Ed. 139 (1940) (emphasis omitted)).

We may, in our discretion, certify a question to the New York Court of Appeals where there is "an unsettled and significant question of state law that will control the outcome of a case pending before this Court." 2d Cir. R. § 0.27; see also McGrath v. Toys "R" Us, Inc., 356 F.3d 246, 249 (2d Cir.2004). Certification is not to be used routinely, see Elliott Assocs., L.P. v. Banco de la Nacion, 194 F.3d 363, 370 (2d Cir.1999), but only in those cases "where there is a split of authority on the issue, where [a] statute's plain language does not indicate the answer, or when presented with a complex question of New York common law for which no New York authority can be found." McCarthy v. Olin Corp., 119 F.3d 148, 153 (2d Cir.1997) (quoting Riordan v. Nationwide Mut. Fire Ins. Co., 977 F.2d 47, 51 (2d Cir.1992) (alteration in original)). With these standards in mind, we turn to our examination of New York law.

In Cargo Partner AG v. Albatrans, Inc., we held that the New York Court of Appeals would require a showing of continuity of ownership to establish the existence of a de facto merger in a contract case. 352 F.3d at 46. In that case, we considered whether the defendant, Albatrans, could be liable for the debts of the former Chase-Leavitt Company, a shipping company whose assets Albatrans had bought. Id. at 43. Cargo Partner did not allege that there was any continuity of ownership between Chase-Leavitt and Albatrans. Id. at 46. Instead, it argued that New York applies a flexible analysis that does not necessarily require proof of each of the four factors to establish a de facto merger. Id. We concluded that we did not need to "determine whether all four factors must be present for there to be a de facto merger" because "[w]hichever test applies, we are confident that the doctrine of de facto merger in New York does not make a corporation that purchases assets liable for the seller's contract debts absent continuity of ownership." Id. As we explained, "continuity of ownership is the essence of a merger" and is what helps us distinguish a merger from an asset sale. Id. at 47. That is, "the nature of an asset sale is that the seller's ownership interest in the entity is given up in exchange for consideration." Id. Because there was no proof of continuity of ownership, "the asset purchase was not `a merger ... called something else.'" Id. (quoting Cargo Partner AG v. Albatrans, Inc., 207 F. Supp. 2d 86, 104 (S.D.N.Y.2002) (alteration in original)).

Cargo Partner explicitly limited its holding to contract cases, id. at 46, and is therefore not binding here. Nevertheless, its reasoning applies equally in cases involving tort claims or other claims by involuntary creditors. The continuity-of-ownership element "is designed to identify situations where the shareholders of a seller corporation retain some ownership interest in their assets after cleansing those assets of liability." Gen. Battery, 423 F.3d at 306. This factor serves the same purpose in tort cases, i.e., it distinguishes an asset purchase from a de facto merger.3 Accordingly, several New York courts have held in tort cases that, because continuity of ownership is the essence of a merger, a de facto merger will not be found in the absence of this element. See, e.g., In re New York City Asbestos Litig., 789 N.Y.S.2d at 487; Buja v. KCI Konecranes, Int'l Plc, 12 Misc. 3d 859, 815 N.Y.S.2d 412, 415-16 (N.Y. Sup.Ct., 2006).

The State argues that this reasoning is not applicable here and that, at least in tort actions, intermediate New York courts have issued conflicting decisions. It contends that these cases suggest that the Court of Appeals might adopt a more flexible test. The State first points to City of New York v. AAER Sprayed Insulations, Inc., 281 A.D.2d 228, 722 N.Y.S.2d 20 (1st Dep't 2001), a case in which, it claims, the First Department found the existence of a de facto merger in the absence of continuity of ownership. In that case, Basic Corporation and Kelley Island entered into an asset purchase agreement whereby Basic bought Kelley Island's building materials product line and the factories used to produce them. Id. at 20-21. The opinion does not indicate whether there was continuity of ownership, but the State asserts in its brief in this case that the parties in AAER Sprayed Insulations conceded a lack of continuity of ownership. Nevertheless, the court found the allegation of successor liability sufficient to survive summary judgment because, "[a]t a minimum, there was a cessation of Kelley Island's ordinary business coupled with a continuation by Basic of the purchased business at the same physical location, and with the same assets, general business operation, and many of the same personnel." Id. at 21.

Even assuming that there was no evidence of continuity of ownership in that case, AAER Sprayed Insulations does not indicate a change in the law. The opinion itself is not explicit about whether there was proof of continuity of ownership, and the court discussed only Basic's objection on appeal that there could be no de facto merger because Basic did not buy all of Kelley Island's assets. See id. Moreover, even if the First Department had adopted the view that a de facto merger could exist in the absence of continuity of ownership, it has since rejected this view. In its recent decision in In re New York City Asbestos Litigation, a tort case, the First Department recognized the flexibility of the de facto merger standard, but explicitly held that continuity of ownership is the touchstone of the concept. As it explained, continuity of ownership is the "essence of a merger" and thus a necessary predicate to a finding of a de facto merger. 789 N.Y.S.2d at 487; see also Subramani v. Bruno Mach. Corp., 289 A.D.2d 167, 736 N.Y.S.2d 315, 316 (1st Dep't 2001) (finding no de facto merger in a tort action in the absence of continuity of ownership); accord Kretzmer v. Firesafe Prods. Corp., 24 A.D.3d 158, 805 N.Y.S.2d 340, 341 (1st Dep't 2005) (stating that a showing of continuity of ownership is required in contract actions).

The State next points to the Fourth Department's decision in Sweatland, in which the court considered whether defendant Park Corporation was the successor-in-interest to Bertsch & Company, the manufacturer of the industrial machine that caused the plaintiff's injury. 587 N.Y.S.2d at 55. After filing for bankruptcy protection, Bertsch entered into an asset purchase agreement with Park Corporation whereby Park Corporation bought its tangible assets, the sole and exclusive right to use the trade name Bertsch, all customer lists, and all other files relating to the business. Id. The opinion does not indicate, however, whether Park Corporation bought the assets for cash or stock. Thus, there is no indication whether there was continuity of ownership between the two companies. In concluding that an issue of fact remained as to whether the transaction was a de facto merger, the court listed the four traditional factors but noted that "[n]ot all of these factors are needed to demonstrate a merger; rather, these factors are only indicators that tend to show a de facto merger." Id. at 56 (internal quotation marks omitted; alteration in original). It noted further that "[p]ublic policy considerations dictate that, at least in the context of tort liability, courts have flexibility in determining whether a transaction constitutes a de facto merger. While factors such as shareholder and management continuity will be evidence that a de facto merger has occurred, those factors alone should not be determinative." Id. (citation omitted). The court therefore concluded that the trial court had not erred in denying defendant's motion for summary judgement and that "plaintiff should be allowed to conduct further discovery to determine whether the transaction constituted a de facto merger." Id. at 55.

Although some cases have cited Sweatland for the proposition that all four factors need not be present to establish a de facto merger, see, e.g., AT & S Transp., 803 N.Y.S.2d at 120, Fitzgerald, 730 N.Y.S.2d at 70, no New York court has applied it to hold explicitly that a de facto merger may be found in the absence of evidence of continuity of ownership.4 Moreover, although its statement that a court has the flexibility to find a de facto merger in the absence of continuity of ownership appears, at first glance, absolute, the court simply held that the plaintiff should be allowed further discovery to determine whether there had been a de facto merger.

Even if Sweatland could be read to permit a finding of a de facto merger in the absence of continuity of ownership, its rationale for departing from the traditional common-law rule has been undermined by the recent decision of the New York Court of Appeals in Semenetz v. Sherling & Walden, Inc., 7 N.Y.3d 194 (2006). In that case, defendant S & W Edger Works, Inc. ("Edger Works"), an Alabama company, sold a band sawmill to Semenetz Lumber Mill, Inc. in 1998. Id. at 196. On July 26, 1999, Sean Semenetz, an infant, was injured when his right hand and fingers got caught between a sprocket and chain apparatus in the sawmill. Id. at 197. In October 2000, "Edger Works sold most of its assets, including real property, good will, trade names and inventory to Sawmills and Edgers, Inc." ("Sawmills") for $300,000. Id. The purchase agreement explicitly specified that the purchasing company assumed none of the seller company's liability, except with respect to sawmills that had been ordered but not delivered. Id. Edger Works then changed its name to Sherling & Walden, Inc. Sawmills, the purchasing company, began manufacturing sawmills in the same plant that Edger Works had used and retained some of Edger Works' former employees. Id. It also began to advertise itself as "formerly S & W Edger Works." Id.

In 2002, Semenetz brought suit on behalf of her infant son against Sawmills, Edger Works, and Sherling & Walden. Id. Sawmills, which was also an Alabama company, answered the complaint by asserting that the New York court lacked personal jurisdiction over it. It later moved for summary judgment on this ground. Id. Because there was personal jurisdiction over Edger Works, which had sold and shipped the offending sawmill to Semenetz Lumber in 1998, the question was whether Sawmills could be liable for Edger Works' allegedly tortious conduct. Id. at 198.

Semenetz argued that Sawmills was liable pursuant to the "product line" theory of successor liability, which is not part of the traditional common law but which Semenetz asked the Court of Appeals to adopt. Under the product line theory, a corporation that buys another's product line and continues production of that line can be liable for an injury caused by that product even if it was produced by the predecessor corporation. The California Supreme Court first adopted the product line theory in Ray v. Alad Corp., 19 Cal. 3d 22, 136 Cal. Rptr. 574, 560 P.2d 3 (1977). The Ray court concluded that this theory of successor liability was appropriate because "`a party which acquires a manufacturing business and continues the output of its line of product under the circumstances here presented assumes strict tort liability for defects in units of the same product line previously manufactured and distributed by the entity from which the business was acquired.'" Semenetz, 7 N.Y.3d at 199 (quoting Ray, 136 Cal. Rptr. 574, 560 P.2d at 11).

The New York Court of Appeals rejected this approach to successor liability. Considering one of the rationales given by the California Supreme Court in Ray — that the sale of assets and dissolution of the original company destroys the products-liability plaintiffs' remedies — the court concluded that this was merely a statement of the problem and therefore not sufficient to justify a change in the law. Id. at 200. The court further observed that "the `product line' exception threatens `economic annihilation for small businesses'" and could deter small businesses from purchasing on-going businesses. Id. at 200-01 (quoting Bernard v. Kee Mfg. Co., Inc., 409 So. 2d 1047, 1049 (Fla.1982)). The court also concluded that "extending liability to the corporate successor places responsibility for a defective product on a party that did not put that product into the stream of commerce" and is thus inconsistent with the justification for strict products liability. Id. at 201. In consequence, the court found that "adoption of the `product line' exception would mark `a radical change from existing law implicating complex economic considerations better left to be addressed by the Legislature.'" Id. (quoting City of New York v. Pfizer, 260 A.D.2d 174, 688 N.Y.S.2d 23, 25 (1st Dep't 1999)).

Although Semenetz concerned a different theory of successor liability, it clearly suggests that the New York Court of Appeals will not eviscerate traditional common-law norms of successor liability in tort cases. That is, it suggests that the court does not find the public policy considerations at issue in tort cases sufficient to justify the departure from the common-law standards that would be necessary to find the existence of a de facto merger in the absence of any evidence of continuing ownership.5 Accordingly, because we find that New York would not depart from the traditional common law to find a de facto merger in the absence of any evidence of continuity of ownership, we are not presented with an "exceptional" circumstance that would require certification of this question to the New York Court of Appeals. McGrath, 356 F.3d at 250.

In sum, the State has failed to point to any evidence of continuity of ownership and therefore has failed to raise an issue of fact that NSI is liable as Serv-All's corporate successor. This failure is fatal to both the State's CERCLA and common-law claims.

For the foregoing reasons, we AFFIRM the judgment of the district court dismissing the State's claims.