Argued June 20, 2003
Judge Leval concurs in the majority opinion and in a separate concurring opinion.
New York State sued National Service Industries, Inc. ("NSI") under the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 ("CERCLA"), 42 U.S.C. §§ 9601 et seq., to recover the costs of closure and remediation of the Blydenburg Landfill in Islip, New York. In granting summary judgment for New York, the District Court for the Eastern District of New York (Jacob Mishler, District Judge) applied the substantial continuity test for successor liability articulated in B.F. Goodrich v. Betkoski, 99 F.3d 505 (2d Cir. 1996), and held that NSI, which had purchased all of Serv-All Uniform Rental Corporation, Inc.'s ("Serv-All") assets, was the successor to Serv-All and was therefore responsible for Serv-All's CERCLA liability. See New York v. Nat'l Serv. Indus., 134 F. Supp. 2d 275, 280-81 (E.D.N.Y. 2001). NSI appeals the grant of summary judgment in favor of New York on the basis that, among other reasons, Betkoski is no longer good law after the Supreme Court's decision in United States v. Bestfoods, 524 U.S. 51 (1998). We agree that, after Bestfoods, the substantial continuity test cannot be applied to determine successor liability under CERCLA.
Because our resolution of this appeal does not involve analysis of most of the underlying facts of this case, we provide only a sketch of the events which led to the present suit and direct those who may be interested in greater detail to the opinion below. See Nat'l Serv. Indus., 134 F. Supp. 2d at 276-77.
From 1962 to 1988, Serv-All operated an industrial garment rental service. Id. at 276. As part of its business, it dry cleaned garments at a facility that was owned separately by Serv-All's two principals. In June 1978, Serv-All arranged for the disposal of several 55-gallon drums of liquid waste that contained perchloroethylene ("PCE"), a solvent used in the dry-cleaning process. The drums were deposited at the Blydenburgh Landfill in Islip, New York. In 1983 the landfill was listed in the New York Registry of Hazardous Waste Sites and in 1987 it was added to the federal National Priority List of contaminated sites. As of September 13, 2001, New York had spent at least $10,894,271 to clean up the landfill. The parties agree that because Serv-All arranged for disposal of wastes at the landfill, it is "potentially responsible" for the cost of cleaning up the landfill.
In 1988, NSI (then known as "Initial Service Investments") entered into an "Asset Sale Agreement," whereby it agreed to purchase Serv-All's customer contracts, customer lists, service inventory, accounts receivable, trucks, insulated jackets bearing the name Serv-All and the right to use or retire Serv-All's name. Id. at 278-79. Serv-All warranted that the asset transfer would be "sufficient to meet the needs" of Serv-All's customers. The sale price was based on Serv-All's revenues as an ongoing operation, as determined by the weekly revenues for the six weeks preceding and following the purchase. The purchase price was approximately $2,229,000. The owners of Serv-All covenanted that they would not compete with NSI for seven years.
On the date of the purchase, Serv-All adopted a plan of liquidation, which it carried out shortly thereafter. Id. at 279. NSI tried to make the transition from Serv-All to NSI as seamless as possible by employing the Serv-All drivers, using trucks which still bore the name Serv-All, using Serv-All on its letterhead and taking over the Serv-All telephone number. Id.
The district court held that "NSI's operations were a substantial continuation of Serv-All's business, and thus, NSI is subject to successor liability for Serv-All's environmental infractions." Id. at 280-81. In particular, the district court held that NSI was liable under CERCLA for the clean-up cost at the Blydenburgh Landfill. Id.
We are faced with the question of whether, in the context of CERCLA, the substantial continuity rule for successor liability that we adopted in Betkoski remains good law after the Supreme Court's decision in Bestfoods. We hold that it does not.
CERCLA makes any "person" who is the present or past owner or operator of a contaminated facility or a generator or transporter of the hazardous substances that pollute a facility, liable for the cost of cleaning up the facility. 42 U.S.C. § 9607(a). "Person" is defined by CERCLA to include corporations and other business organizations. See id. § 9601(20)-(21).
However, CERCLA does not provide rules for when corporations that are related to the responsible corporation should be held liable.
In the instant case, as it was in Betkoski, the question is whether a corporation that purchased the assets of a company liable for environmental response costs under CERCLA should be held liable for those same costs as a successor corporation. In Bestfoods, the question was somewhat different: it asked when a parent corporation should be held responsible for the CERCLA liabilities of its subsidiary. Bestfoods, 524 U.S. at 55.
Because the analysis of parent-subsidiary liability in Bestfoods bears on our analysis of successor liability here, we consider whether the rule of successor liability we adopted in Betkoski is consistent with the Supreme Court's analysis in Bestfoods.
A. Bestfoods --
Bestfoods involved an effort by the United States to recover under CERCLA the cost of remediating a facility near Muskegon, Michigan. Id. at 57-58. In 1957, Ott Chemical Company had begun to manufacture chemicals at the facility. Id. at 56. In 1965, the Bestfoods corporation (then known as CPC International, Inc.), created a subsidiary corporation that purchased Ott Chemical. The subsidiary corporation (also called Ott Chemical) continued to manufacture chemicals until 1972, when Bestfoods sold it to an unrelated company. Id. The district court held that Bestfoods was liable under CERCLA because it had "operated" the facility through its subsidiary. CPC Int'l, Inc. v. Aerojet-Gen. Corp., 777 F. Supp. 549, 572 (W.D. Mich. 1991). In particular, the district court reasoned that Bestfoods had operated the facility through its subsidiary because it had selected the subsidiary's board and populated its executive ranks with Bestfoods officials. Id. at 573. The Sixth Circuit reversed en banc. United States v. Cordova Chem. Co., 113 F.3d 572 (1997). It held, seven judges to six, that a parent corporation could only be liable for the actions of its subsidiary if the state common law rules for piercing the corporate veil were met or if the parent had directly participated in the operation of the facility. Id. at 580.
The Supreme Court vacated the circuit's en banc opinion and remanded. The Court held that "CERCLA is  like many another congressional enactments in giving no indication that 'the entire corpus of state corporation law is to be replaced simply because a plaintiff's cause of action is based upon a federal statute.'"
Bestfoods, 524 U.S. at 63 (quoting Burks v. Lasker, 441 U.S. 471, 478 (1979)). The "failure of [CERCLA] to speak to a matter as fundamental as the liability implications of corporate ownership demands application of the rule that 'in order to abrogate a common-law principle, the statute must speak directly to the question addressed by the common law.'" Id. (quoting United States v. Texas, 507 U.S. 529, 534 (1993)). The Supreme Court rejected the district court's test for a parent corporation's derivative liability because that test "disregard[ed] entirely . . . time-honored common law rule[s]." Id. at 70.
B. Betkoski and the Substantial Continuity Test --
In considering the substantial continuity test, we take from 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. Because the substantial continuity test adopted in Betkoski departs from the common law rules of successor liability, Betkoski is no longer good law.
As we described in Betkoski, the "traditional common law rule states that a corporation acquiring the assets of another corporation only takes on its liabilities if any of the following apply: the successor expressly or impliedly agrees to assume them; the transaction may be viewed as a de facto merger or consolidation; the successor is the 'mere continuation' of the predecessor; or the transaction is fraudulent." 99 F.3d at 519 (quoting United States v. Carolina Transformer Co., 978 F.2d 832, 838 (4th Cir. 1992)); see also Fletcher, Cyclopedia of the Law of Private Corporations § 4892.75. The exception at issue here is the "mere continuation" exception. See Betkoski, 99 F.3d at 519.
Betkoski described the "mere continuation" test as the "common law test (sometimes referred to as the identity test), which requires the existence of a single corporation after the transfer of assets, with an identity of stock, stockholders, and directors between the successor and predecessor corporations." Id. at 519 (citing Carolina Transformer, 978 F.2d at 838).
Noting that the substantial continuity test (sometimes referred to as the "continuity of enterprise" approach) had been followed by the Supreme Court in the labor law context, in Betkoski we contrasted that doctrine with the mere continuation rule. Rather than considering ownership, the substantial continuity test focuses on the continuity of the business: Whether the "successor maintains the same business, with the same employees doing the same jobs, under the same supervisors, working conditions, and production processes, and produces the same products for the same customers." Id.
We adopted the substantial continuity test because "CERCLA is entitled to a construction that advances its primary goals [and] the substantial continuity test is more consistent with [those] goals [and is thus] superior to the older and more inflexible 'identity' rule." Id. Accordingly, we held the substantial continuity test to be "the appropriate legal test for successor liability under CERCLA." Id. In Betkoski we were not applying general common law principles of corporate law but rather adopting a special rule for use in CERCLA cases that departed from those principles.
New York argues that the substantial continuity test is simply a refined version of the traditional mere continuation exception. New York is correct that, unlike the novel rule adopted by the district court in Bestfoods, there is some prior precedent for the substantial continuity rule and the rule is generally presented as a variant of the mere continuation exception. However, the substantial continuity test is not a sufficiently well established part of the common law of corporate liability to satisfy Bestfoods' dictate that common law must govern. We briefly describe the current state of the substantial continuity doctrine.
Within federal law, the substantial continuity doctrine is well established in the area of labor law. See, e.g., Fall River Dyeing & Finishing Corp. v. NLRB, 482 U.S. 27, 43-45 (1987).
However, the labor law cases are particular to the labor law context and therefore have not been and cannot easily be extended to other areas of federal common law. In Textile Workers Union v. Lincoln Mills, the Supreme Court held that section 301 of the Labor Management Relations Act "authorizes federal courts to fashion a body of federal law for the enforcement of  collective bargaining agreements." 353 U.S. 448, 451 (1957).
Thus, the rules that federal courts and the Supreme Court have adopted in the labor law context do not necessarily reflect general federal common law but rather represent law fashioned for the purpose of enforcing section 301. See Golden State Bottling Co. v. NLRB, 414 U.S. 168, 182 n.5 (1973) (contrasting the "perimeters of the labor-law doctrine of successorship" with the general rules governing asset purchase agreements).
Furthermore, some of the labor law substantial continuity cases have not involved asset purchases. See John Wiley & Sons, Inc. v. Livingston, 376 U.S. 543, 551 (1964) (using "substantial continuity" in the case of a merger to assess whether labor agreement should apply after the merger). Thus, in the labor law context the doctrine is not part of the common law "mere continuation" exception. Finally, even where the labor cases did involve an asset purchase, the focus was not on whether the successor corporation should be responsible for general corporate liabilities but whether the successor had a duty to bargain with the same workforce which had previously voted to bargain collectively with the predecessor corporation. Thus, in Fall River Dyeing & Finishing Corp v. NLRB, the test was employed to determine whether "those employees who have been retained will understandably view their job situations as essentially unaltered." 482 U.S. at 43 (1987)(quoting Golden State Bottling Co., 414 U.S. at 184).
The substantial continuity doctrine has also been applied in a few product liability cases. In Betkoski, we pointed to Mozingo v. Correct Mfg. Corp., where the Fifth Circuit applied the doctrine to hold an asset purchaser liable for an injury that occurred before the purchase. 752 F.2d 168, 175 (5th Cir. 1985).
The Fifth Circuit characterized the doctrine as "in the nature of an expansion of the mere continuance exception to the rule of non-liability" for asset purchasers. Id. But the Mozingo Court was applying Mississippi law and in Betkoski we were relying on federal common law. 1 See also Cyr v. B. Offen & Co., 501 F.2d 1145, 1153-54 (1st Cir. 1974) (adopting substantial continuity doctrine where New Hampshire law applied to hold asset purchaser liable for product injuries of predecessor). Perhaps if the substantial continuity doctrine were widely adopted among the states, it might be considered absorbed into federal common law and thus applied under CERCLA pursuant to the reasoning in Bestfoods in whatever states had not adopted it. However, at present the doctrine has been adopted by only a handful of states. See Fletcher, Cyclopedia of the Law of Private Corporations § 4892.75 ("A few states have loosened the traditional 'mere continuation' exception in strict liability cases to focus on continuity of the business or enterprise . . .
[but] [m]ost courts that have considered the continuation of enterprise exception have refused to adopt it.").
We thus find that the substantial continuity doctrine is not a part of general federal common law and, following Bestfoods, should not be used to determine whether a corporation takes on CERCLA liability as the result of an asset purchase. Our holding is consistent with dicta by the only other Court of Appeals that has considered whether the substantial continuity doctrine survives Bestfoods. See United States v. Davis, 261 F.3d 1, 54 (1st Cir. 2001) (noting that Bestfoods "left little room for the creation of a federal rule of liability under the statute").
Because we find that the substantial continuity doctrine of Betkoski is no longer good law and reject the argument that the doctrine is part of the federal common law for CERCLA purposes, we must vacate the district court's grant of summary judgment for plaintiff New York State.
For the forgoing reasons, we VACATE the grant of summary judgment for plaintiff and REMAND for further proceedings.
LEVAL, J., concurring
I agree completely with my colleagues that, after the Supreme Court's ruling in United States v. Bestfoods, 524 U.S. 51 (1998), the CERCLA liability of a seller of assets may no longer be imposed on the purchaser of those assets on the basis of the "substantial continuity" test several federal courts have adopted for such situations. See B.F. Goodrich v. Betkoski, 99 F.3d 505, 519 (2d Cir. 1996); United States v. Mexico Feed & Seed Co., 980 F.2d 478, 487 (8th Cir. 1992); United States v. Carolina Transformer Co., 978 F.2d 832, 838 (4th Cir. 1992). The Supreme Court made clear in Bestfoods that in the absence of a statutory authorization, courts are not free to impose federal statutory liability under standards that depart from the well established principles of common law. Under common law rules, a purchaser of assets does not succeed to the liabilities of the seller, except in narrow circumstances. The "substantial continuity" test fashioned by the federal courts for CERCLA liability is unquestionably a significant deviation from the common law rules. Nor is authority for it to be found in the terms of the CERCLA statute. The judgment of the district court therefore must be set aside. I agree with my colleagues that the Bestfoods principle is altogether adequate to dispose of this case. I write separately to note my belief that, even if the CERCLA substantial continuity test could survive Bestfoods, the judgment of the district court nonetheless could not stand because the court misunderstood the requirements of the test. I believe it is worthwhile noting this alternative basis for overturning the district court's ruling, for three reasons. First, notwithstanding our judgment that the test has been dispatched by Bestfoods, it could remain with us. In Bestfoods, the Supreme Court did not address the issue of successor liability; it ruled on a question of subsidiary liability. Although the principles of Bestfoods appear to us fully applicable to our question, there is as yet no holding of the Supreme Court that speaks directly to it. Second, and perhaps more important, even assuming that Bestfoods has laid this court-devised test to rest, the test might be resuscitated by legislative action. Courts developed the test because they believed it advanced the goals of CERCLA. If Congress shares that view, Congress might wish to give express statutory authorization for the test. Third, a substantial continuity test for successor liability might be devised by courts for use in connection with another federal statute that does confer proper authorization. In any of these circumstances, the test may be employed once again after Bestfoods. The test has caused substantial confusion. In view of the possibility of its return, it seems worth the trouble to explore how the test should be understood.
In a small number of contexts, courts have fashioned rules, notwithstanding the contrary principles of common law, to impose a liability incurred by a previous operator of a business on a successor in the operation of that business. Courts have described these rules in terms of a "substantial continuity" test for successor liability. The objective was in each case to further the policies of the particular statute under consideration, or prevent its circumvention. It would be quite mistaken to view these instances as involving the use of a pre-existing test of clear, well- understood contours, which courts have plugged into first one, then another statutory scheme. To the contrary, in the case of each statutory scheme, the courts, perceiving the inadequacy of the common law rules to support the objectives of the particular statute, have groped case by case toward a new standard, sometimes following the lead of the administrative agency charged with front-line administration of the statute. The test developed for use with each statute reflects considerations peculiar to that statute. As a result, the test developed for one statute differs from the test developed for another. Although courts have often identified these as tests of "substantial continuity," substantial continuity is not the entirety of any such test, but merely an element. Nor is the concept of substantial continuity satisfied in the same way in each circumstance, regardless of the statute whose objectives are being advanced. Notwithstanding some common ground, it is a different test, designed to achieve different purposes, depending on the statute for which it was developed.
For example, over a period of years, in the interpretation of the National Labor Relations Act ("NLRA"), 29 U.S.C. §§ 151 et seq., the Supreme Court, generally following the lead of the National Labor Relations Board ("NLRB"), developed a doctrine shaping the bargaining obligations of an employer who takes over a business formerly conducted by another employer, together with a significant part of the predecessor's workforce. Where the predecessor had been obligated to bargain with the union certified as the representative of its workforce, the Court imposed on the successor the predecessor's duty to bargain with that union. The facts and reasoning of these cases clearly indicate that they are responsive to the specific concerns of § 8(a)(5) of the NLRA, and are not intended to fashion a general principle of successor liability applicable in other contexts. 2 See NLRB v. Burns Int'l Sec. Svcs., Inc., 406 U.S. 272 (1972); Fall River Dyeing & Finishing Corp. v. NLRB, 482 U.S. 27 (1987). 3 In Burns International, Burns had won a competitive bid to oust another company, Wackenhut, as provider of security services to a Lockheed plant. Within the previous few months, the Wackenhut employees had selected the United Plant Guard Workers Union ("UPG") as their bargaining representative, and the UPG had been certified by the NLRB as their exclusive bargaining agent. 406 U.S. at 274-75. To perform the job it had won, Burns hired many of the workers Wackenhut was laying off, enough to make up a majority of its workforce assigned to that job. Burns regarded the newly hired workers as covered by its collective bargaining agreement with a different union, which encompassed workers in similar employment at other sites. It refused to bargain with UPG. The NLRB found that this refusal violated §§ 8(a)(5) and 8(a)(1) of the NLRA, 29 U.S.C. §§ 158(a)(5) and (a)(1), and ordered that Burns bargain with UPG with respect to the former Wackenhut employees. The Supreme Court upheld the order. The Court focused on whether the Wackenhut "bargaining unit," which the NLRB had certified for representation by UPG, remained intact after the change in employer, and found that it did because the same employees were performing the same tasks under the same conditions. Id. at 278-80. The Court explained, "The source of [Burns'] duty to bargain with the union is . . . the fact that it voluntarily took over a bargaining unit that was largely intact and that had been certified within the past year." Id. at 287. The court noted furthermore, "It would be a wholly different case if the [NLRB] had determined that because Burns' operational structure and practice differed from those of Wackenhut, the Lockheed bargaining unit was no longer the appropriate one." Id. at 280. Just as the considerations guiding the test are distinct in the context of duty to bargain under the NLRA, so the remedy is also particular to that context. The Court explicitly ruled that Burns was not bound to the substantive terms of the collective bargaining agreement the UPG had negotiated with Wackenhut. Its sole "successor" obligation was to bargain with UPG. It remained free to fashion its own demands over such substantive terms as "changes in corporate structure, composition of the labor force, work location, task assignment, and nature of supervision." Id. at 287-88. It is quite clear that the considerations that guided the Court in fashioning both the test and the remedies under the test were uniquely tailored to the particular labor law context and are of little relevance to a CERCLA dispute over succession to cleanup liabilities.
Fifteen years later, the Supreme Court refined the test in Fall River Dyeing, making even more clear its intent to rest the analysis of successor duties on the specific concerns of the NLRA. Fall River, a newly formed company, had purchased a portion of the assets of a recently shut- down dyeing works, and hired enough of the former company's unionized employees to make up a majority of its workforce. The NLRB ordered Fall River to bargain with the union that had the workers had chosen while employed by the predecessor company. The Supreme Court affirmed the NLRB's order. While employing what it called a "substantial continuity" test and noting that the duty to bargain was inherited when the successor company "acquired substantial assets of its predecessor and continued, without interruption or substantial change, the predecessor's business operations," 482 U.S. at 43, the Court focused on the fact that where the successor company retained the same employees in the same positions, "those employees who have been retained will understandably view their job situations as essentially unaltered." Id. The Court explained that the viewpoint of the workers provided the touchstone for application of the test, because
emphasis on the employees' perspective furthers the Act's policy of industrial peace. If the employees find themselves in essentially the same jobs after the employer transition and if their legitimate expectations in continued representation by their union are thwarted, their dissatisfaction may lead to labor unrest.
Id. at 43-44. As in Burns, the application of a "substantial continuity" test focused on the composition of the workforce, and the employees' "legitimate expectations" of being represented by the union they had selected under the terms of the NLRA, while employed together in the same positions and in the same location, although under a different employer.
In Golden State Bottling Co. v. NLRB, 414 U.S. 168 (1973), the Supreme Court considered a different question also arising under the National Labor Relations Act: "whether the bona fide purchaser of a business, who acquires and continues the business with knowledge that the predecessor had committed an unfair labor practice in the discharge of an employee, may be ordered by the [NLRB] to reinstate the employee with backpay." Id. at 170. All American Beverages, Inc., had purchased the assets of Golden State's soft drink bottling and distribution business with knowledge of the NLRB's prior order to Golden State to remedy an unfair labor practice it had committed in discharging an employee. All American had continued Golden State's operations without interruption and employed its labor force. The Board ordered All American to reinstate with back pay the employee wrongfully dismissed by Golden State. The Supreme Court affirmed the Board's order. The judgment was based on considerations particular to the labor context. First was the primary interpretive role of the NLRB, which had issued the reinstatement order, and enjoyed "broad discretion to fashion . . . relief adequate to achieve the ends, and effectuate the policies, of the [NLRA]." Id. at 176. Second, the "objectives of national labor policy," which include "protection to the employees from a sudden change in the employment relationship," appropriately "balanced" the "prerogative of owners independently to rearrange their businesses." Id. at 182. Third, as in Fall River, "[T]he employees' legitimate expectation is that the unfair labor practices will be remedied [and thus] a successor's failure to do so may result in labor unrest." Id. at 184. In sum, then,
Avoidance of labor strife, prevention of a deterrent effect on the exercise of rights guaranteed [by] the Act, and protection for the victimized employee - all important policies subserved by the [NLRA] - are achieved at a relatively minimal cost to the bona fide successor.
Id. at 185. The particular substantial continuity test applied was thus substantially shaped by the objectives of the NLRA.
Against this background, and in similar fashion, after the passage of CERCLA courts began to perceive a need to extend successor liability beyond the scope of the conventional common law rules lest the objectives of the Act be frustrated. The courts reasoned in a manner similar to the opinions reviewed above, with the important distinction that their reasoning concerned the objectives of CERCLA, rather than those of the labor relations Act. Carolina Transformer dealt with successor liability in a case in which the parties had manipulated the corporate form to continue a company's operations while evading its liability. Carolina Transformer was solely owned by one individual, who was its president until being succeeded by his son. In its business of repairing and rebuilding electrical transformers, the company disposed of chemical wastes in a manner that caused it to accrue CERCLA liability. After the Environmental Protection Agency performed a CERCLA cleanup on its site, Carolina Transformer sold most of its assets to FayTranCo, a company founded and owned by the children of the Carolina Transformer owner, including the son who had succeed his father as president of Carolina Transformer. FayTranCo substantially continued the operations of Carolina Transformer. The district court ruled that FayTranCo was a successor of Carolina Transformer for purposes of CERCLA liability, and the Fourth Circuit affirmed.
The court began by enumerating eight factors, which dealt exclusively with the degree of continuity of the enterprise. It went on from there, however, to focus on whether "the transfer . . . was part of an effort to continue the business of the former corporation yet avoid its existing or potential state or federal liability." Carolina Transformer, 978 F.2d at 838 (emphasis added). In its recitation of the facts justifying imposition of liability, the court clearly focused on the fact that the form of succession was designed to continue the business in largely the same hands while effectively defeating the CERCLA liability the predecessor corporation had incurred. The court's explanation emphasized that assets of the predecessor corporation were transferred to the successor at below-market prices or with only nominal compensation. Id. at 838-41. The transaction was seen as a manipulation of the corporate form to defeat existing and potential liabilities. The court's reasoning in no way suggests that continuity alone would have been sufficient for the imposition of successor liability, if the facts had not included manipulation to defeat the CERCLA liability of the predecessor corporation.
United States v. Mexico Feed & Seed Co. adhered even more clearly to the same approach. The asset purchaser and the seller were both in the business of collecting and hauling waste oil. The purchaser acquired the assets of the seller and continued its operations. Among the assets sold were a handful of tanks that had previously leaked waste oil onto a CERCLA cleanup site. Id. at 482-84. Despite the buyer's substantial continuation of the seller's business, the court declined to impose successor liability. The Eighth Circuit stressed that "[t]he purpose of corporate successor liability . . . is to prevent corporations from evading their liabilities through changes of ownership." Id. at 487. The court noted, "The cases imposing 'substantial continuation' successorship have correctly focused on preventing those responsible for the wastes from evading liability through the structure of subsequent transactions." Id. at 488. The court contrasted Carolina Transformer, where successor liability had been imposed upon facts suggesting a manipulative effort to escape liability by creating a new corporate entity in substantially the same hands as the predecessor. Id. at 488-89. The Mexico Feed court concluded, notwithstanding the new owner's substantial continuation of the seller's business, that no successor liability would be imposed where the purchaser had "no knowledge of the offending tanks" or of its seller's potential CERCLA liability, and where the sale of assets, on which the successor liability was assertedly predicated, had been an arm's-length transaction for "adequate consideration" and "between two competitors, not a cozy deal where responsible parties merely changed the form of ownership yet in substance remained the same." Id. at 489-91. Nor was the transaction one "where a purchasing corporation either in collusion with the seller, or independently, bought only 'clean' assets, and knowingly left 'dirty' assets behind with an insufficient asset pool to cover any potential liability." Id. at 489-90. While such manipulations of the corporate form might have justified imposing liability, the court concluded that mere continuation of an enterprise by a good-faith purchaser of assets did not.
These two decisions make several things clear. First, while courts have used the term "substantial continuity test" to designate tests in both labor law and CERCLA, the test is different in each context. In the labor law context, it is guided by the question how the imposition or non-imposition of successor liability would affect important policies of the national labor law. Crucial to such decisions are issues such as the structure of the bargaining unit, the legitimate expectations of employees, and the goal of labor peace - issues that have no bearing on whether to impose successor CERCLA liability. 4 Likewise, in the CERCLA context, courts have relied on factors that have no bearing on imposition of successor liability under labor law. Second, while the courts have used the shorthand term "substantial continuity" to refer to the test that causes a predecessor's CERCLA liability to be visited on a successor, that is merely a shorthand name. The test is not merely one of substantial continuity. More than substantial continuity is required to place the predecessor's liability on the successor to its business. Although (as in Carolina Transformer) the courts have recited a series of factors relating solely to the issue of continuity, those factors were only part of the test. The courts' ultimate decisions whether to impose successor liability have turned on additional factors involving whether failure to impose liability on the successor would undermine the objective of the Act to impose liability on responsible parties. 5
The district court in the case under review misunderstood what conditions are necessary for imposition of successor liability under the substantial continuity test. According to the district court's view the eight factors cited in Carolina Transformer, which relate solely to continuity, are alone sufficient to make the purchaser liable for the seller's CERCLA obligations. Because NSI continued Serv-All's business making little change, this justified imposing Serv- All's liability on NSI as its "legal successor." 6
If a seller's CERCLA liability could be imposed on a purchaser of assets merely because it continued in substantially unchanged form the operations of the seller, the rule would result in disastrously unfair consequences, which furthermore would be harmful to the economy as a whole. In many circumstances, a purchaser of assets, buying with the intention of continuing the business being sold, would have no knowledge, and no reasonable way of learning, that the assets were burdened with liability. Businesses that engage in illegal dumping of toxic wastes often do so in secret, arranging for furtive nocturnal dumping by unlicensed carriers. Such a seller will have no records of its illegal dumping and will certainly not volunteer the information to the purchaser of its assets. In addition, in many instances the dumping that gives rise to liability was done long in the past, at a time when such activities were unregulated and no records were maintained. Even assuming good faith among all participants in the purchase and sale, there may be no reasonable way for the purchaser to learn about the seller's history of disposal of toxic wastes because none of the present employees of the seller were employed when the disposal occurred and the seller has no records of the disposal.
Thus, a good faith buyer of assets, who does all reasonable due diligence, and has no reason to suspect liability, may learn too late that assets purchased at full going-business value came with a hidden liability that might vastly exceed their value. Business assets purchased in good faith for $1 million might easily bring liabilities of $5 million or $10 million in cleanup costs.
A rule of successor liability that threatened good-faith buyers with huge, unpredictable liability would also impose serious systemic costs on the economy. Such a rule would depress the price purchasers would be willing to pay for assets, as buyers would risk acquiring massive hidden liability. At the very least, buyers would need to purchase expensive liability insurance, which would reduce the market value of assets offered for sale. Asset purchasers might also favor piecemeal purchases at breakup value, in preference to purchasing at going-business value to continue the operation, resulting in both depression of the market value of the assets and the destruction of the jobs of those previously employed in the seller's operation. Nor is the unfairness that would result from the imposition of such capricious, arbitrary liabilities on innocent good faith purchasers nullified by the theoretical availability of insurance. This is so for at least three reasons. First, as noted, insurance would impose premium costs on all purchasers, thus diminishing the value of the assets being sold. Second, insurance policies have limits on the amount of coverage. Cleanup costs can be so extraordinarily high as to exceed the available coverage. Third, an arbitrary and unfair imposition of a substantial liability on a blameless party is no less unfair or arbitrary (although less drastic) when its impact is dispersed by the insurance mechanism among numerous blameless parties.
In conclusion, both the governing caselaw and good sense teach that, as to CERCLA liability, the so-called "substantial continuity" test for successor liability requires more than substantial continuity in the seller's business operations. It requires also the presence of factors justifying the conclusion that the objectives of CERCLA would be improperly circumvented if such liability were not imposed. The exact nature of such liability-justifying facts was not fully developed in the few cases that had considered the question before the Supreme Court's Best Foods decision brought the process to a halt. If the so-called "substantial continuity" test should either survive Bestfoods or be reincarnated in new legislation, it is of great importance that it be correctly understood as requiring not merely continuity but also the presence of such CERCLA- defeating conduct as justified imposition of liability in Carolina Transformer and guided the refusal to impose liability in Mexico Feed.
For the reasons well expressed in the opinion of Chief Judge Walker, as well as those expressed in this concurrence, I join in the court's ruling that the judgment of the district court must be vacated.
1. In Bestfoods, the Supreme Court expressly declined to decide whether courts should apply state or federal common law under CERCLA. See 524 U.S. at 64 n.9. In a per curiam opinion issued shortly after our opinion in Betkoski, we explained that we had selected federal law rather than borrowing state law because of "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." B.F. Goodrich v. Betkoski, 112 F.3d 88, 91 (1997) (per curiam). Because we find that, as a matter of federal law, the substantial continuity doctrine cannot be applied, the analysis prescribed by United States v. Kimball Foods, Inc., 440 U.S. 715 (1979), to determine whether state or federal common law should be used would likely come out differently now. In particular, a state's "mere continuity" rule is unlikely to "frustrate specific objectives of the federal program," id. at 728-29, because it is likely to be the same as the federal rule. In fact, if the state had adopted the substantial continuity test, as the Mozingo court found that Mississippi had, then the state common law might actually facilitate CERCLA's objectives.
2. Section 8(a)(5), 29 U.S.C. § 158(a)(5), makes it an "unfair labor practice" for an employer "to refuse to bargain collectively with the representatives of his employees."
3. Earlier in John Wiley and Sons, Inc. v. Livingston, 376 U.S. 543 (1964), the Supreme Court held that where a company obligated to arbitrate disputes under a collective bargaining agreement merged with another company which had no such duty before the merger, the company resulting from the merger was bound by the predecessor's duty to arbitrate.
4. For example, it is inconceivable that courts would impose CERCLA successor liability on the facts of the Burns case, where the Supreme Court found that the successor company inherited the predecessor's duty to bargain with the employees' designated union. The successor company in Burns had purchased no assets of the predecessor, but had simply ousted the predecessor as provider of security services in a competitive bidding process, then hired many of the predecessor's laid-off employees to fulfill its newly assumed contractual obligations. Such facts satisfied "substantial continuity" in the context of the duty to bargain, but would not in the CERCLA context. Assume that a dry-cleaning company, A, wins a contract previously held by B, which has incurred CERCLA liability. A hires some of B's employees to service its new contract. Nothing in the CERCLA cases suggests that A would succeed to B's CERCLA liability, even though it might be obligated to bargain with the union chosen by B's employees. The tests are different, despite the shared name.
5. This circuit's decision in Betkoski to adopt a "substantial continuity" test did not discuss what are the essential elements of the test. We stated, citing Carolina Transformer, Mexico Feed, and other cases, that courts "have read CERCLA as making successor corporations liable, in certain circumstances, for their predecessors' acts." 99 F.3d at 518. As justification for the doctrine, we explained that "absent successor liability, a predecessor could benefit from the illegal disposal of hazardous substances and later evade responsibility for remediation simply by changing the form in which it does business, thereby subverting the Act's purpose of holding responsible parties liable for cleanup costs." Id. at 519. While we made no effort to specify the elements of the test of substantial continuity, the above-quoted discussion makes clear that, like the Carolina Transformer and Mexico Feed courts, we were concerned with manipulative contrivances designed to defeat CERCLA's objectives.
6. The district court noted in addition that because the New York State Department of Environmental Conservation found in a public proceeding that Serv-All had arranged for the disposal of hazardous material at the Blydenburgh Landfill, and that this was recorded in a judicial decision of a New York State court, NSI was "on notice" of Serv-All's potential environmental liabilities at the time of its purchase of Serv-All's assets. The court noted that "this factor further militates" in favor of imposing successor liability on Serv-All. The district court, however, made no finding whether NSI was in fact aware of Serv-All's liability, or whether its being technically "on notice" was pertinent, in the manner found crucial in Carolina Transformer and Mexico Feed, to a finding that the objectives of the of the Act required imposition of successor liability. On the district court's reasoning, the substantial continuity of operations was alone sufficient to justify the purchaser's liability, and that the legal "notice" derived from the fact of official publications only added to an already sufficient basis of liability.