Research Support' School of Business Administration July 1993 BIFURCATION OF OWNER AND OPERATOR ANALYSIS UNDER CERCLA: FINDING ORDER IN THE CHAOS OF PERVASIVE CONTROL Working Paper #717 Lynda J. Oswald University of Michigan

i l l

BIFURCATION (:)F THE OWNER ANI) OPERATO'I(.)R ANALYSIS UNDER CERCLA: FINDING ORDER IN THE CHAOS OF PERVASIVE C(:)NTROL+ Since its enactment in 1980, the Comprehensive Environmental Response, Compensation, and Liability Act1 ("CERCLA" or "Act") has been the focus of attention by commentators who criticize both its statutory provisions and the interpretations given thein by the courts.2 These commentators complain, and rightfully so, that the statute is vague and incomplete.3 They fear that CERCLA liability is too expansive, and they worry that the courts have inappropriately extended CERCLA liability to hold parties such as corporate officers, individual shareholders, and parent corporations liable in instances where traditional corporate law doctrines would have shielded thel.4 Many of these fears are justified. Liability under CERCLA is Draconian, and deliberately so.5 Congress' intent in enacting CERCLA was to ensure that everyone who' was potentially responsible for hazardous waste contamination be held responsible for the costs of cleaning it up,6 and Congress cast a wide net in an effort to achieve its objectives.7 However, Congress deliberately left unanswered many questions regarding the statute's liability provisions, intending instead that those gaps be filled "under common law principles... [by] a court performing a case-by-case evaluation."8 Although the courts agree that Congress intended CERCLA's liability scheme to be interpreted broadly,9 they have not agreed on what those interpretations should be. As a result, much of the case law interpreting CERCLA is analytically confused and inconsistent.10 Many of the judicial pronouncements found in these cases could be read broadly as expanding liability in precisely the ways that cause these commentators such anxiety.

Bifurcation Under CERCLA Page 2 Despite these legitimate concerns regarding CERCLA and its application by the courts, the outlook is not as bleak as many commentators would paint it. CERCLA is a relatively new statute. The case law under CERCLA is newer still -- the first major case addressing liability under CERCLA was not decided unLil 194. 1 Thus, the courts have had relatively little time to identify, interpret, and apply CERCLA's liability schemnes. To a large exctnt, CERCLA jurislrudence is suffering growing pains. Thei courts alre still sorting their way through the statute's obscure provisions. Their efforts have resulted in a myriad of tests for evaluating the liability of corporate actors, tests whose conflicting and contradictory provisions confound commentators.12 The rules regarding CERCLA liability are gradually becoming clearer, however, and will undoubtedly continue to solidify over time as the courts have additional opportunities to explore the issues raised by CERCLA's statutory schemie. At the moment, however, CERCLA can charitably be described only as being in a state of flux. The current disarray of the legal rules developed under CERCLA is, in large part, the cause of these fears of expanded liability. The uncertainty of the rules, and the inconsistencies in their judicial applications, lead commentators alnd potential defendants to suspect the worst in terms of doctrinal development and liability rules. Certainly, the most effective (and logical) way of correcting the problems that have arisen under CERCLA would be to redraft the statute so as to simplify and clarify the principles of CERCLA liability. These issues are complex and politically charged,13 however, and it seems improbable that Congress will undertake any drastic revisions in this aiea.14 Instead, CERCLA rules are likely to continue to be developed in the courts through case-by-case analysis. Much of the doctrinal confusion that currently surrounds CERCLA could be alleviated if courts, conmmentators, and practitioners would focus on the big picture of CERCLA liability rather than on the minutiae of the statute and judicial interpretations of it. Perhaps nowhere is this more true than in the area of individual shareholder and

Bifurcation Under CERCLA Page 3 parent corporation liability. Courts and commentators alike have devoted significant amounts of energy to trying to determine whether shareholder liability should be direct (i.e., based on CERCLA's statutory language) or indirect (i.e., based on traditional notions of piercing the corporate veil). All too often, these theories are regarded as being mutually exclusive,15 rather than complementary. Even more distressing, the courts are beginning to merge the tests for evaluating direct and indirect liability into a single, imprecise standard -- the pervasive control test16 -- further clouding the question of liability under CERCLA. This Article focuses primarily on the liability of parent corporations under CERCLA, in part because the cases involving parent corporations raise typical types of issues in relatively uncomplicated factual circumstances. In addition, the cases involving parent corporations are plentiful enough to provide a good sample for evaluation, but limited enough to be manageable.17 This Article argues that the courts could bring order to the current chaos of CERCLA owner and operator liability by implementing two simple steps: (1) the courts need to bifurcate the owner and operator analyses, so as to explicitly acknowledge that owner liability is indirect, and operator liability is direct; and (2) the courts need to abandon the pervasive control test, which has tended to discourage courts from engaging in careful, reasoned analyses of owner and operator liability. Bifurcation of owner and operator liability, coupled with an explicit rejection of the pervasive control test, are necessary to ensure that courts reach the correct outcome in cases involving CERCLA liability of parent corporations and other shareholders. Part I begins by furnishing a brief overview of CERCLA. It also identifies and discusses the issue of owner and operator liability under the Act. Part II describes how and why the case law in this areas got off to such a bad start, and why those early mistakes contribute so heavily to the confusion that persists in this area of the law today. Part III addresses the indirect liability of parent corporations, i.e., the effect of holding a

Bifurcation Under CERCLA Page 4 parent corporation liable under the common law theory of piercing the corporate veil. It examines the question of whether state or federal law should be used in CERCLA piercing cases, and analyzes the approaches used by courts imposing indirect liability under the Act, including the development of the pervasive control test. Part IV examines the issues raised by imposition of direct liability upon parent corporations, and analyzes the two alternative tests used to assess such liability (capacity to control versus exercise of control). The pervasive control test's role in direct liability cases is examined. Finally, Part V illustrates how an explicit segregation of the owner and operator liability analyses, when linked with careful and comprehensive liability standards, can lead to more doctrinally defensible outcomes in CERCLA parent corporation cases. Concluding remarks are made in Part IV. I. OWNER AND OPERATOR LIABILITY UNDER CERCLA: THE ISSUE DEFINED Congress enacted CERCLA in 1980 in an attempt to address the environmental issues posed by hazardous waste disposal. A number of environmental incidents, including Love Canal in New Yorkl8 and the James River kepone contamination in Virginia'9 convinced Congress that existing federal statutes were unable to effectively address hazardous waste disposal problems.20 Unlike its statutory predecessors,21 CERCLA is a remedial statute. It was designed primarily to rectify environmental problems posed by hazardous waste produced and abandoned in the past, rather than operating prospectively to prevent future problems.22 The problem of cleaning up existing contamination is a staggering one: commentators estimate that the entire cleanup bill for hazardous waste improperly disposed of in the past will ultimately reach hundreds of billions of dollars.23 Because Congress' goal in enacting CERCLA was to ensure that the parties responsible for hazardous waste contamination bear the costs of its cleanup,24 liability under CERCLA is

Bifurcation Under CERCLA Page 5 deliberately broad: liability is retroactive,25 joint and several,26 and strict.27 Persons found liable under CERCLA are responsible for both cleanup costs and damages,28 and the statute permits only three narrow defenses, for acts of God, war, or unrelated third parties.29 Imposition of liability under CERCLA generally requires findings that: (1) the contaminated property or site is a "facility;" (2) a release or threatened release of a hazardous substance from the facility has occurred; (3) response costs have been incurred as a result of the release or threatened release; and (4) the party to be held liable falls within one of the four classes of potentially responsible parties ("PRPs") described in section 107 of CERCLA.30 Each of these four elements is interpreted broadly under the Act. "Facility" essentially means "any site or area where hazardous waste has... come to be located."31 "Release" is defined to include spills, leaks, dumping, emissions, or any other means by which a hazardous substance is released into surface or subsurface water or land, or the ambient air.32 "Hazardous substance" includes a host of substances defined as hazardous or toxic under a variety of federal environmental statutes.33 Section 107 is the relevant provision for purposes of this Article, however. This section establishes four categories of PRPs: (1) the current owner and operator of the facility; (2) any person who formerly owned or operated a facility at the time of disposal of any hazardous substance; (3) persons who arranged for the disposal or treatment of hazardous substances (commonly known as "generators"); and (4) transporters of hazardous waste.34 CERCLA defines "person" to include corporations and other business entities as well as individuals,35 but makes no specific reference to parent corporation or shareholder liability.36 This Article focuses on the issue of "owner" and "operator" liability under CERCLA.37 As noted above,38 the statute specifically refers to the liability of both present and past owners and operators.39 The phrase "owner or operator" itself is defined circularly in the statute as "any person owning or operating a facility."40 CERCLA

Bifurcation Under CERCLA Page 6 explicitly excludes from liability as an owner or operator any person who, without participating in the management of the firm, holds indicia of ownership primarily to protect a security interest in the facility (the "secured creditor exemption").41 It is hornbook law that shareholders are, in effect, merely investors in the corporation in which they own stock.42 The shareholders typically are not liable for the debts and liabilities of the corporation beyond their contribution to capital.43 The corporation is regarded as an entity "separate and apart" from its shareholders.44 Thus, the corporation can, in its own right, own property, enter into contracts, incur debts, and commit torts and crimes.45 This limited liability extends not only to individual shareholders, but also to corporations who own shares in another corporation.46 The two affiliated corporations are generally regarded as separate and distinct legal entities.47 Even a parent corporation, which by definition owns more than fifty percent of the voting stock of its subsidiary48 and so can exercise control over it, is protected by the rule of limited liability, absent fraud or other abuse of the corporate form.49 Courts typically hold parent corporations liable for the CERCLA violations of their subsidiaries under one of two theories: direct liability under the Act, or indirect liability under piercing doctrine. These two bases of liability parallel those found in other, non-environmental, legal contexts.50 The courts have expended a fair amount of energy in discussing the rationale behind their choice of one theory over the other, seemingly ignoring the fact that the theories are not mutually exclusive. Nothing in CERCLA's statutory provisions dictate that only one of these two traditional forms of parent liability may apply. The courts' struggles with these two theories stem directly from their failure to understand and confront the key distinction under the Act -- the difference between "owner" and "operator" liability.51 Operator liability is a direct liability based upon the parent's own activities with respect to a facility. Owner liability, on the other hand, is an

Bifurcation Under CERCLA Page 7 indirect liability that falls under traditional corporate law doctrines, such as piercing of the corporate veil.52 CERCLA's statutory provisions are very s a clear as the nature of the ownership interest required. "Owner" under section 107(a) refers to the owner of the facility at which hazardous substances are released.53 Parent corporations own stock in their subsidiaries; they have no direct ownership interest in the facility.54 In order to hold a parent corporation liable as an owner of the facility, therefore, the subsidiary's separate existence must be disregarded pursuant to a piercing analysis. In contrast, "operator" refers to an individual or entity who operates a facility. In a typical parent-subsidiary relationship, the subsidiary operates the facility. In order for the parent corporation to be held liable as an operator, it must somehow be directly involved in the management and operation of that facility.55 While corporate status does not shield a parent from operator liability,56 neither does it subject such a party to liability. Some sort of affirmative parent action is required before operator liability can attach. It is impossible for a parent corporation as a shareholder to engage in behavior that would lead to it being held liable as an operator under CERCLA. Shareholders are, by definition, investors in the firm. Their authority is limited to electing the board of directors and voting on major corporate actions; they have no ability to engage in day-today management of the firm, its operations, or its hazardous waste practices.57 It is only when the parent acts outside the role of shareholder and exercises direct control over the facility that operator liability can arise. Thus, the parent corporation's liability as an operator must be grounded in its own affirmative acts. Courts tend to link the terms "owner" and "operator" as a single phrase -- "owner or operator" -- and they typically do not distinguish between the different bases of liability for each. While these liability theories are not mutually exclusive, neither are they synonymous. A parent corporation can be liable as an operator under CERCLA only if it actually operated the facility.58 Similarly, a parent corporation can be held liable as an owner under CERCLA only if the corporate veil is pierced.59 Thus, the

Bifurcation Under CERCLA Page 8 choice between direct and indirect liability necessarily depends upon the category of PRP under which the parent is being held liable, not upon a determination that one theory is philosophically or analytically superior to the other. Much of the current confusion in CERCLA law can be traced to the failure of early courts to recognize this critical distinction. II. NEPACCO: A FIRST STEP IN THE WRONG DIRECTION Much of the parent corporation case law under CERCLA draws heavily upon a case that did not even address parent corporation liability: United States v. Northeastern Pharmaceutical & Chemical Co.60 (NEPACCO). The case involved a Delaware corporation that manufactured hexachlorophene, a principal by-product of which is 2,3,7,8-tetrachlorodibenzo-p-dioxin ("dioxin"), at a leased chemical manufacturing facility in Verona, Missouri.61 The case arose out of contamination caused by the corporation's illegal burial of eighty-five 55-gallon drums of toxic wastes on a farm owned by James Denney. Edwin Michaels, the president and a shareholder of NEPACCO, was present at the Verona plant during its first year of construction and operation. He then returned to Connecticut.62 In 1971, two years after NEPACCO began its operations at the Veronaplant, Ronald Mills, a shift supervisor, proposed to Bill Ray, the plant manager, that Mills be hired as an independent contractor to dispose of the drums of toxic waste. Ray discussed the issue with John Lee, who, in addition to being a shareholder in the corporation, was the vice-president with direct management responsibilities for the plant.63 Lee defined the characteristics that the disposal site should possess;64 evidence indicated that Lee knew that Mills proposed to bury the drums on the Denney farm.65 Ray visited the site prior to disposal and reported his observations to Lee.66 In 1979, the EPA investigated the site and discovered the dioxin contamination. It cleaned up the site and brought suit for recovery of the response costs.

Bifurcation Under CERCLA Page 9 In a 1984 opinion commonly known as NEPACCO 1,67 the District Court for the Western District of Missouri held that Lee and Michaels were directly liable as owners and operators under CERCLA.68 The court based its findings upon two lines of reasoning: (1) that the logical extension of CERCLA's secured creditor exemption is that active shareholders can be held personally liable under the Act, and (2) that "owner or operator" should be defined in terms of the capacity to control, rather than actual exercise of control.69 Unfortunately for the reasoned development of CERCLA jurisprudence, both of these arguments are flawed. The NEPACCO I court began by examining the secured creditor exemption. Section 101(20)(A) of CERCLA defines "owner or operator" as including "any person owning or operating [a] facility," but as excluding "a person, who, without participating in the management of a vessel or facility, holds indicia of ownership primarily to protect his security interest in the vessel or facility."70 The court thus concluded that section 101 (20)(A) "literally reads that a person who owns interest [sic] in a facility and is actively participating in its management can be held liable for the disposal of hazardous waste."71 The court believed that this construction promoted Congress' intent that CERCLA "insure, so far as possible, that the parties responsible for the creation of hazardous waste sites be liable for the response costs of cleaning them up."72 Because Lee and Michaels were in the "unique position" of being both active participants in the management of the plant and major shareholders in the corporation, the NEPACCO I court found that they met the statutory definition of "owner and operator."73 The NEPACCO I court then stated that it was adopting the reasoning of United States v. Mobil Oil Corp.,74 in which, according to the NEPACCO I court, the Fifth Circuit held that "owner or operator" was defined under the Clean Water Act as the person "who has the capacity to make timely discovery of oil discharges,... to direct the activities of persons who control the mechanisms causing pollution,... [and] to prevent and abate damage."75 The NEPACCO 1 court applied this "person-in-charge test" to hold

Bifurcation Under CERCLA Page 10 Lee and Michaels liable as owners or operators under CERCLA because they had the capacity to control disposal, had the power to direct negotiations regarding the disposal of hazardous substances, and had the capacity to prevent and abate the damage caused by disposal.76 On appeal, the Eighth Circuit, in a 1986 opinion commonly known as NEPACCO II, affirmed Lee's liability under section 107(a)(3) as a generator,77 but reversed the lower court's finding that the two officers were liable as owners or operators, noting that owner or operator liability under CERCLA attaches only to owners or operators of the disposal facility.78 The "facility" in NEPACCO was a farm field belonging to a third party; neither of the individuals nor NEPACCO itself owned or operated the site. Although the distinction may appear, at first glance, to be a rather technical one, it has firm footing in the statutory language. Mere operation of the corporation itself is insufficient; the officers must have been involved in the operation of the farm that constituted the facility. Although the distinction drawn by the Eighth Circuit is of crucial importance to the correct development of CERCLA jurisprudence, it was virtually ignored by later courts and commentators. Despite the Eighth Circuit's reversal of the district court's holding on the owner or operator liability issues, the NEPACCO I court's analysis has provided the basis for many of the court opinions addressing parent corporation liability as owners or operators under CERCLA.79 This reliance can be attributed in part to timing: a number of owner and operator cases were decided in the two-year timespan between the NEPACCO I and ANEPACCO II decisions.80 By citing to and relying upon the NEPACCO I court's analysis of the owner or operator issue, these later courts legitimized that analysis. In addition, in overturning NEPACCO I, the Eighth Circuit did not comment on the district court's substantive reasoning regarding the type of behavior that could lead to owner and operator liability, thus leaving the impression that NEPACCO rs reasoning might be accurate even though its specific application was flawed.

Bifurcation Under CERCLA Page 11 Despite their popularity, both of the tests set forth by the NEPACCO I court are faulty. First, as several commentators have pointed out,81 the NEPACCO I court's decision rests upon an interpretation of Mobil Oil that is fundamentally incorrect. The NEPACCO I court was wrong when it stated that Mobil Oil addressed the question of the definition of "owner or operator."82 Rather, Mobil Oil was concerned with whether a corporate owner was the "person in charge" under the Clean Water Act so as to qualify for a statutory immunity.83 The key question in evaluating operator liability, under both the Clean Water Act and under CERCLA,84 is whether actual exercise of control over the facility is necessary, or whether mere capacity to exercise control will suffice. While the language of Mobil Oil seems to adopt the latter position,85 that issue was not before that court, and so the court's views on this important subject can scarcely be said to be carefully considered.. In blithely adopting this flawed construction of Mobil Oil's owner or operator analysis, the NEPACCO I court set the stage for a definition of owner and operator liability under CERCLA that makes no distinction between "owner" and "operator" and that is based upon the existence of capacity to control, as opposed to actual exercise of control.86 NEPACCO I's erroneous interpretation was quickly picked up by other courts,87 and has apparently found a permanent, albeit undeserved, place in CERCLA jurisprudence.88 Second, NEPACCO Ts "secured creditor exemption" test is based upon an interpretation of the Act that is little short of ludicrous. Section 101 (20)(A) of CERCLA specifically excludes from the definition of "owner or operator" any person "who, without participating in the management of... a facility, holds indicia of ownership primarily to protect his security interest in the... facility."'89 The legislative history of CERCLA clearly reveals that Congress' intent in enacting this section was to immunize secured creditors from the CERCLA liabilities of their debtors.90 However, both NEPACCO I and several later courts have interpreted this exemption as necessarily holding liable those persons who do hold indirect ownership interests and who do

Bifurcation Under CERCLA Page 12 participate, at least peripherally, in the management of the firm -- e.g., active shareholders, including parent corporations.91 These courts misunderstand the fundamental nature of the property interests at issue. A person holding a security interest has a direct property interest, which may be manifested through such things as a mortgage, deed of trust, lien, title held pursuant to a lease-financing transaction, legal or equitable title obtained pursuant to foreclosure, assignments, or pledges.92 For example, a secured creditor who holds a mortgage on property used as a disposal plant has a property interest in that facility. Absent the restriction of section 101(20)(A), that property interest could cause the secured creditor to be deemed to be an owner of the facility. However, only when coupled with "active participation" in management will this ownership interest result in the secured creditor being held liable for the CERCLA cleanup costs.93 As long as no such active participation in management is present, the secured creditor is insulated from liability,94 despite its property interest in the facility. A shareholder, whether an individual or a parent corporation, has no such ownership interest in thefacility. Rather, the shareholder has an ownership interest in the corporation; the corporation, in turn, has an ownership interest in the facility. Absent a piercing of the corporate veil, the shareholder cannot be deemed to have an ownership interest in the facility itself.95 Thus, even active management in the subsidiary cannot render a shareholder liable under this provision. Instead, the shareholder must have actively participated in the operation of the facility itself -- and, at that point, the shareholder's liability is based solely in its own affirmative acts as an operator, and not in its status as an owner of the corporation.96 Thus, NEPACCO 's secured creditor exemption test has two major flaws: it incorrectly focuses on management of the subsidiary instead of management of the facility, and it places unwarranted emphasis on the shareholder's status as a shareholder as opposed to its active participation in the operation of the facility.

Bifurcation Under CERCLA Page 13 Thus, NEPACCO I, the first major CERCLA liability case, started CERCLA jurisprudence off on the wrong foot. The opinion incorrectly focused on capacity to control instead of actual control, and on control of the subsidiary instead of control of the facility, and it failed to distinguish between liability as an owner versus liability as an operator. Each of these analytical flaws has been magnified in the subsequent case law, contributing greatly to the current disarray in CERCLA liability rules. III. INDIRECT LIABILITY OF PARENT CORPORATIONS: PIERCING THE CORPORATE VEIL Under traditional corporate law doctrine, a court will pierce the veil of a corporation to hold the shareholders liable for the acts of the corporation where the corporation was formed to perpetrate a fraud,97 or where the shareholders actively controlled or dominated the corporation such that the corporation is the shareholders' "mere instrumentality" or "alter ego."98 Piercing is considered an extraordinary form of relief,99 and courts are hesitant to pierce a corporate veil unless circumstances clearly indicate that such an action is warranted.100 Moreover, courts that engage in piercing almost invariably confine their efforts to either close corporations or to affiliated corporate groups, such as parent-subsidiary relationships.101 In his ground-breaking 1931 treatise, Frederick J. Powell provided a "laundry list" of factors that indicate when a subsidiary's veil ought to be pierced so as to hold a parent corporation liable.102'These factors concentrate on the degree of control that the parent exercises over the subsidiary and the extent to which the parent has respected the corporate formalities of the subsidiary.103 The factors have been adopted (and adapted) by the courts to the extent that they now form the basis of virtually all parent-subsidiary piercing law.104 Under modem rules, neither the mere fact of ownership nor the general ability to control that normally accompanies such ownership is sufficient to support a piercing of the veil.105 Rather, some additional evidence of domination by the parent

Bifurcation Under CERCLA Page 14 over the subsidiary, whether evidenced through intermingling of management or finances or through disregard of corporate formalities, must be shown. Were the rule otherwise, every subsidiary would be susceptible to piercing because every parent corporation, by virtue of its controlling interest in its subsidiary, is in a position to exercise a certain degree of control over its subsidiary. Analytically, the imposition of indirect liability upon parent corporations through piercing of the corporate veil is both easier and more difficult than the imposition of direct liability. It is easier in the sense that piercing is a well-established corporate law doctrine. The theory is familiar to both courts and commentators, and so the use of the theory in the CERCLA context does not cause the same degree of consternation that direct liability seems to engender. It is more difficult in the sense that piercing doctrine is by no means precise or predictable. As a common law doctrine based in notions of equity,;06 piercing doctrine has been criticized as "defy[ing] any attempt at rational explanation"o17 and has been described as being "rare, severe, and unprincipled."108 A. Indirect Liability of Parent Corporations Under CERCLA Most of the CERCLA cases that have examined the piercing issue have looked at it in the context of determining whether jurisdiction could be exercised over a parent in a state where, but for the presence of the subsidiary, jurisdiction would not exist. 109 Relatively few cases examining the piercing of the corporate veil in the environmental context have been decided on the merits.110 Because the case law on piercing under CERCLA has arisen in this rather sterile analytical environment, it is not as fully developed or as well-reasoned as one might hope. 1. Piercing Under CERCLA: State or Federal Common Law? One of the first issues that a court analyzing the indirect liability of a parent corporation under CERCLA must decide is whether to apply state piercing doctrine or a

Bifurcation Under CERCLA Page 15 federal common law rule of piercing. Federal courts are permitted to formulate federal common law in cases arising under federal statutes.11 If a federal statute contains standards for disregard of the corporate form, the court must follow those standards, not conflictingt state standards.2 Where the federal statute is silent on the issue, the court may choose to either adopt state law or federal common law on the issue. In making its choice, the court must balance state and federal interests, including factors such as: (1) whether application of state law would frustrate the objectives of the federal statute; (2) the need for national uniformity on the issue; and (3) the extent to which a federal rule would disrupt commercial relationships predicated on state law.113 In general, the courts may develop federal common law only where Congress has expressly granted them the power to do so, or where a federal rule of decision is "necessary to protect uniquely federal interests."114 As with so many other issues, Congress did not address whether federal or state common law should be used in piercing cases under CERCLA. Most of the courts that have addressed the issue have determined that federal common law should apply,l15 although they often formulate that rule in different ways.116 A few courts have applied state piercing doctrine. 117 As a practical matter, the outcome is unlikely to differ whether a state rule or a federal rule is applied. In Joslyn Manufacturing Co. v. TL. James & Co., 1 18 the most influential CERCLA piercing case to date, the Fifth Circuit did not indicate whether it was applying state or federal law as the rule of decision. The court applied a piercing test from an earlier, non-environmental case, however, in which it had stated: [W]e find no need to determine whether a uniform federal alter ego rule is required since the federal and state alter ego tests are essentially the same. Our non-diversity alter ego cases have rarely stated whether they were applying a federal or state standard, and have cited federal and state cases interchangeably. 119 Other courts have also noted the similarity between state and federal piercing standards. As one court noted, the federal common law draws heavily upon state law for guidance

Bifurcation Under CERCLA Page 16 "and for that reason the choice between state and federal law may in many cases present questions of academic interest, but little practical significance."120 Thus, the key issue under CERCLA is not whether state or federal common law should apply, but rather, how the statute should interact with traditional piercing law. 2. Piercing Under CERCLA: The Substantive Rules As noted above,l21 very few courts have grappled with the issue of the indirect. liability of parent corporations under CERCLA in a substantive context, and only one federal appellate court has addressed this issue. Thus, the Fifth Circuit's 1990 decision in Joslyn Manufacturing Co. v. T.L. James & Co.122 has become a bellwether for CERCLA piercing analysis. The facts in Joslyn were undisputed. The case arose as a result of discharges from a creosating plant constructed by Lincoln Creosating Company, Inc. ("Lincoln") in 1935. Lincoln was, from the very beginning, wholly under the control of T.L. James & Company ("T.L. James").123 Lincoln was sold to Joslyn Manufacturing Company ("Joslyn") in 1950, which owned and operated the plant until 1969.124 The property passed through a number of owners in subsequent years. In response to the EPA's cleanup order,l25 Joslyn brought suit for contribution against T.L. James, arguing that T.L. James was liable as an "owner or operator" under CERCLA because of its relationship with its subsidiary, Lincoln.126 The district court granted T.L. James' motion for summary judgment. The court analyzed the issue of parent liability under CERCLA as an either/or issue: either CERCLA permitted direct liability of parent corporations, or it permitted indirect liability.127 The court apparently did not contemplate the possibility that both types of liability could arise under the statute. The district court noted that CERCLA does not specifically address the direct liability of parent corporations. It declined to follow the analyses of courts that had

Bifurcation Under CERCLA Page 17 found that corporate officers could be held directly liable because it felt that such analyses "ignore the corporate form without an express congressional directive."128 The court discussed the important role that limited liability and the corporate form play within American jurisprudence,129 and concluded that "[n]either the clear language of CERCLA nor its legislative history provides authority for imposing individual liability on corporate officers or direct liability on corporations."l30 Thus, the court determined that parent corporations could only be held liable under CERCLA if the circumstances warranted a piercing of the subsidiary's corporate veil.131 The district court then applied the Fifth Circuit's version of Powell's laundry list for piercing and concluded that the facts before it would not support a piercing of the corporate veil to hold T.L. James liable for its subsidiary's CERCLA violations.132 The court found that the subsidiary strictly observed corporate formalities by keeping its own books and records and by holding frequent shareholder and director meetings, that the operations of the parent and subsidiary were kept separate, that neither of the two key persons in the subsidiary were employed by the parent, that the subsidiary owned the property on which the facility was located and that that property was not used by the parent, that none of the subsidiary's employees were on the payroll of the parent, and that the subsidiary filed separate tax returns, paid its own bills, and made its own arrangements for employee benefits.133 Thus, the district court determined that there was "simply no proof that [the parent] had complete domination of finances, policies and practices to cause [the subsidiary] to be not a separate business entity but a mere conduit of" the parent.l34 The district court's rejection of the notion of direct liability under CERCLA likely resulted from its failure to recognize that direct liability does not mean liability based merely upon status -- i.e., a parent corporation is not held liable merely because of its ownership interest in the subsidiary -- but that rather, direct liability is based upon affirmative acts undertaken by the parent with regard to the operation of the facility. The

Bifurcation Under CERCLA Page 18 facts in Joslyn do not indicate that any such affirmative acts of control that would lead to direct liability were actually present in that case. While it would appear, therefore, that the district court reached the correct outcome, its rejection of the existence of direct liability under CERCLA created a disturbing ripple in the pattern of CERCLA liability. On appeal, the Fifth Circuit affirmed,135 adopting much of the district court's analysis. The appellate court agreed that absent "an express Congressional directive to the contrary, common law principles of corporation law, such as limited liability," must govern judicial analysis.136 Like the district court, the Fifth Circuit explicitly rejected the analyses of courts that had allowed parent corporations to be held directly liable under CERCLA as owners or operators, noting that neither the language of CERCLA nor its legislative history explicitly includes parent corporations within the definition of owner or operator.137 To do so, in the Fifth Circuit's view, "would dramatically alter traditional concepts of corporation law."'138 The court concluded that any such redrafting of corporate law must come from Congress, not the courts. Instead, the Fifth Circuit held that a parent corporation is liable for the CERCLA violations of its subsidiary only where the facts support a piercing of the corporate veil.139 Although the Joslyn court's ultimate conclusion regarding direct liability was incorrect, its analysis resulted in large part from its proper refusal to read CERCLA's remedial purposes broadly in order to hold a parent corporation liable based solely upon its status as an owner of the offending corporation.140 Other courts have shown a similar reticence to use mere status, in the absence of affirmative acts of operation, as a basis for liability. For example, in In re Acushnet River & New Bedford Harbor,141 the United States District Court for the District of Massachusetts rejected the government's argument that the purposes of CERCLA were so paramount "that the most punctilious and complete corporate separateness must be observed" and that "the point of piercing occurs just as soon as the parent's contact with the subsidiary transcends a 'pure investment relationship."l142 Rather, the court found that imposition of "CERCLA liability on parent

Bifurcation Under CERCLA Page 19 corporations for no reason other than the fact that they did not ignore the performance of their subsidiary" would so drastically undermine traditional doctrine as to eradicate it -- a result it felt that Congress had not intended.143 The Acushnet court specifically rejected the government's argument that the subsidiary's veil could be pierced simply because the parent organized the subsidiary with the intent of limiting its own liability for environmental contamination. 144 As the court recognized, the broad interpretation of parent liability urged by the government would wreak disaster upon normal business investment practices.145 One of the major purposes in allowing limited liability is to restrict the amount of risk that shareholders face and so to encourage them to engage in risky ventures.146 Noth in CERCLA's statutory language or legislative history indicates that Congress intended to override this basic premise of corporate law in enacting the statute.147 If a parent corporation were to be held liable for its subsidiary's activities based upon an exercise of a minimal degree of control, the parent would be unlikely to invest in ventures that pose a risk of environmental harm, even though the risk might be low and the potential benefit of the activity high.'48 This rejection of an expansive reading of the Act's remedial purposes is a laudable addition to CERCLA case law. Much of CERCLA's language is cloudy and ambiguous.149 Some of this confusion is the result of Act's last minute passage by a lame duck Congress,150 but other parts of the statute were left deliberately unclear or incomplete because of political exigencies.151 Too many courts have used the resulting imprecise statutory language as a license to legislate -- many of the more expansive interpretations of CERCLA rest in little more than judicial attempts to flesh out CERCLA's broad but vague objectives.152 To a certain extent, the courts' response is understandable: protection of the environment is an undeniably worthy goal and the "polluter pays" principle, at a gut level, evokes a sense of fairness. Nonetheless, the role of the courts is to give effect to

Bifurcation Under CERCLA Page 20 the language that Congress has written, not to rewrite that language in broader terms, or to further unspoken congressional policy objectives. The courts do neither the public nor Congress a favor when they step to fill the gaps left by a Congress unwilling or unable to do its job because of political considerations or political paralysis. By carefully confining CERCLA's scope to the language set forth by Congress, the courts cannot only ensure that congressional intent is not exceeded, they can give an incentive to Congress to express its intent more clearly and completely. The refusal of courts like Joslyn and Acushnet to use CERCLA as a tool for weakening traditional corporate law doctrines is thus based in sound statutory interpretation principles. What these courts fail to realize, however, is that direct liability of parent corporations under CERCLA can be supported by the statute's reference to "operator" liability and by traditional corporate law notions of direct liability. No expansive interpretation of CERCLA is necessary to support this result. The liability standard created by the Acushnet court presents two analytical problems of its own, however. The Acushnet court distilled the following test for purposes of piercing the veil under CERCLA: "the Court looks closely for suggestions of pervasive control by [the parent] over [the subsidiary's] hazardous waste disposal policies, or for an indication that [the parent] treats [the subsidiary] as a mere instrumentality with respect to the hazardous waste of [the parent]." 153 The first half of the Acushnet test, the pervasive control test, also arises in the context of direct liability'54 and, in fact, was first articulated in a direct liability case. 155 It poses much the same dangers in both contexts -- the patness of its name encourages courts to use the test as a shortcut to liability and so deters the court from engaging in a full discussion of the factors mitigating for or against parent liability. 156 For example, the Acushnet court extracted the pervasive control test from a list of seven factors it found were commonly considered in evaluating a piercing decision. 157 The court specifically noted that no single factor was "either necessary or sufficient" to support a piercing of the

Bifurcation Under CERCLA Page 21 subsidiary's veil,158 yet it then went on to, in effect, state a test based upon a single one of those factors. The second half of the Acushnet test appears at first glance to be merely an iteration of standard piercing doctrine.'59 However, traditional doctrine generally looks for several signs of parental domination over the subsidiary before finding that piercing is appropriate. By suggesting that parental involvment in hazardous waste activities alone can form the basis for indirect liability, the Acushnet court articulated a version of the piercing doctrine that is broader than the general rule typically applied in non-CERCLA cases. A fair reading of the court's opinion suggests that perhaps the court did not intend to apply a one-factor test for evaluating piercing in the context of CERCLA. The court noted, for example, that "[t]he policies underlying the statute in question can direct the emphasis that a court will place on the various factors examined in deciding whether to pierce the corporate veil."160 Certainly, the parent's involvement in the hazardous waste activities or practices of the subsidiary can be one important factor in determining whether a subsidiary's veil should be pierced for purposes of CERCLA. Such involvement alone should not support such a piercing, however, nor should the absence of such involvement automatically insulate a parent from liability. Despite the emphasis its test placed on hazardous waste activities, in actually evaluating the parent's liability, the Acushnet court focused exclusively upon the financial relationship between the parent and the subsidiary. It noted that while the parent may have influenced the management and philosophy of the subsidiary, it did so within the normal constraints of a parent-subsidiary relationship. 161 The financial integrity and independent corporate existence of the subsidiary had not been impugned by the parent such that the subsidiary's corporate separateness could be ignored.162 The Acushnet court's analysis was carried out in terms of the overall relationship between the two entities, with no special emphasis upon hazardous waste issues; in fact, in discussing the

Bifurcation Under CERCLA Page 22 actual parent-subsidiary relationship, the court did not even touch upon that specific topic. Thus, the court actually applied the traditional type of piercing analysis; it did not actually look to see whether the subsidiary was a "mere instrumentality" of the parent in the context of hazardous waste, or whether the parent exercised "pervasive control" over the subsidiary's hazardous waste practices. The danger in Acushnet, then, lies not in the manner in which the court decided the case, but rather in the test that the court articulated. There is no reason to have a specific piercing test for CERCLA liability as the court seems to suggest -- the statute does not mandate such a test, and the statutory objectives can be fulfilled under traditional tests. B. Indirect Liability of Parent Corporations Under CERCLA: A Summary Joslyn and Acushnet both recognize that in order to hold the parent liable under CERCLA as an owner, the subsidiary's veil must be pierced. Both cases treat indirect and direct liability as mutually exclusive, however -- a result which is analytically unsatisfying and unsupported by the statutory language. In instances where the parent has, in effect, become the alter ego of its subsidiary, it should (and can) be held liable as an owner. There are other instances, however, where the parent may not have overstepped the bounds of corporate separateness to the extent that piercing is warranted, yet was still involved enough in the facility's activities that it should be deemed an operator. Imagine, for example, a parent who strictly observed corporate formalities, avoided intertwining officers and directors, and adequately capitalized its subsidiary, yet provided active, daily supervision and control over hazardous waste disposal activities of the subsidiary. Such a parent should not escape liability just because its activities do not justify a piercing of the subsidiary's veil. It is in just such instances that direct liability of corporations becomes important.

Bifurcation Under CERCLA Page 23 IV. DIRECT LIABILITY OF PARENT CORPORATIONS The concept of direct parent liability is neither novel nor new. Under traditional corporate law doctrine, a parent can be held directly liable where the parent was itself involved in the activitiesat gave rise to the claim against the subsidiary. For example, where the parent has engaged in an independent act of negligence, the parent can be held directly liable for the damages ensuing; it is not necessary to pierce the subsidiary's corporate veil to reach the parenta63 Direct actions can be based upon a number of different types of parental involvement in the activities giving rise to the claim, including an affirmative undertaking by the parent,164 the good samaritan doctrine,165 agency theory,166 or the parent's involvement in fraud'67 or misrepresentation.168 Because imposition of direct liability upon a parent corporation requires, in most instances, just a mundane and conventional application of well-accepted traditional legal rules, it has been subject to far less scholarly scrutiny than its more controversial cousin, the piercing doctrine.169 Nonetheless, the application of direct liability to parent corporations in the CERCLA arena has been fraught with contention and dissent. Commentators have been quick to criticize the application of this theory in this context as an unwarranted expansion of parent corporation liability and an unjustified assault upon the traditional protections of corporate law and the corporate form.170 The fault lies not in the theory of direct liability -- certainly, a parent corporation can, and should, be held liable for its own activities that lead to a CERCLA violation, even if that violation occurs under the feigned auspices of the subsidiary. Rather, the fault lies with the manner in which the courts have applied direct parent liability under CERCLA. The courts have been unable to articulate a clear, rational test to support direct parent liability under the Act. Because of their failure to separate owner and operator liability, the courts have imported certain of the elements of piercing analysis into the direct liability test. The resulting confusion is reflected in the two competing

Bifurcation Under CERCLA Page 24 theories articulated by the courts: (1) the capacity to control theory and (2) the active exercise of control theory.171 The mere existence of two rival theories has contributed greatly to the uncertainty that surrounds this area of CERCLA jurisprudence. When these competing theories are coupled with the cloudy rationales and analyses found in early cases, it is little wonder that commentators are so troubled by the notion of direct liability under CERCLA. The discussion that follows treats the capacity to control and exercise of control tests as independent theories. While it is easier from a conceptual viewpoint to treat these two tests as discrete and to categorize the cases as falling into one camp or the other, realistically, the situation is not that clear. Although the opinions tend to adopt one theory or the other in general terms, at the case-specific level, the opinions tend to blur the distinctions between the two theories -- further evidence of the confusion that prevails in this area. Thus, the categorization of the cases given below is valid only in general terms, and not in terms of the specifics of each individual opinion. A. The Capacity to Control Test The capacity to control test, because it is based upon mere status as opposed to affirmative acts, conflicts sharply with traditional notions of parent corporation liability.'72 Every parent corporation, by virtue of the power it wields over its subsidiary, is in a position where it could, if it so chose, control that subsidiary's activities, including those activities relating to its environmental matters and the operation of a facility.173 A literal application of the capacity to control test would thus lead to a finding of parent liability in every case involving a CERCLA violation by a subsidiary. Fortunately, the courts have shied away from such a rigid approach to this test. Idaho v. Bunker Hill Co.,174 the first case to address parent corporation liability under CERCLA, instead adopted a narrower definition. In its 1986 opinion, the United States District Court for the District of Idaho held that a parent corporation could be directly

Bifurcation Under CERCLA Page 25 liable under CERCLA as an owner or operator because of its capacity to control the hazardous waste practices at the facility owned or operated by its subsidiary.175 In so holding, the Bunker Hill court correctly established that the parent's relationship to the facility, not to the subsidiary, was the crucial determinant in evaluating the CERCLA liability of a parent corporation. 176 Unfortunately, the court did not emphasize this point, and later courts failed to accurately read its subtle signal. The Bunker Hill court explicitly adopted for purposes of determining parent corporation liability both NEPACCO Ts "secured creditor exemption" argument177 and its "person-in-charge" test.178 The facts before the court indicated that the parent at times had controlled the subsidiary's board, the parent received weekly reports of the day-today operations of the subsidiary, the subsidiary was not permitted to spend more than $500 on pollution matters without the parent's approval, and capital expenditures by the subsidiary required the parent's approval. Finally, the subsidiary was undercapitalized with only $1,100 even though the parent received $27 million in dividends from the subsidiary over a 6-year period -- a fact that the court found particularly relevant "to Congress' intent that those who bore the fruits must also bear the burdens of hazardous waste disposal... "180 Thus, the Bunker Hill court found that the parent: was in a position to be, and was, intimately familiar with hazardous waste disposal and releases at the... facility; had the capacity to control such disposal and releases; and had the capacity, if not total reserved authority, to make decisions and implement actions and mechanisms to prevent and abate the damage caused by the disposal and releases of hazardous wastes at the facility.l81 While the specific outcome reached in Bunker Hill may be satisfactory, two parts of the court's analysis are disturbing. First, the facts of Bunker Hill suggest that not only did the parent retain substantial capacity to control the subsidiary's environmental activities, it actually exercised that control to an extent that would exceed the normal oversight functions of a parent.l82 However, instead of adopting a narrower test based

Bifurcation Under CERCLA Page 26 upon actual exercise of control (which would have led the court to the same outcome), the court opted for the broader capacity to control test. The Bunker Hill court's broad premise that direct liability of the parent can be based on nothing more than unexercised ability to control the subsidiary's facility, if widely adopted, would radically rewrite corporate law doctrine on parent-subsidiary relationships. There are indications in the opinion that the court was aware of the dangers inherent in the capacity to control test; at one point, the court emphasized that it was "mindful that in adopting the [NEPACCO I] test, care must be taken so that 'normal' activities of a parent with respect to its subsidiary do not automatically warrant finding the parent an owner or operator."183 However, the court failed to discuss what those permitted "normal" activities might be or how they could be distinguished from activities that would support a finding that the parent possessed a capacity to control sufficient to render it liable as an operator. Second, the Bunker Hill court held merely that the parent was "in this case, an owner or operator for purposes of CERCLA,"184 making no effort to distinguish whether the parent was being held liable in its capacity as an owner, or whether it was being held liable for actions undertaken by it as an operator. The court's failure to draw a clean distinction between owner and operator liability indicates the court's fundamental confusion regarding the different bases of liability for each category. 185 Because Bunker Hill was applying a direct liability theory, the parent could only be liable in its role as an operator. The Bunker Hill court did not engage in the piercing analysis necessary to hold the parent liable as an owner. A year later, in 1987, the United States District Court for the District of Colorado also addressed parent corporation liability under CERCLA, in Colorado v. Idarado Mining Co.'186 Although the Idarado case followed closely on the heels of the Bunker Hill opinion, the Idarado court did not refer to it. Instead, the Idarado court relied specifically upon NEPACCO Ts "person-in-charge" test187 and a 1986 opinion by the

Bifurcation Under CERCLA Page 27 United States District Court for the Western District of Missouri, United States v. Conservation Chemical Co.,188 which the Idarado court found required the following factors to be considered in evaluating whether a parent corporation was an owner or operator of a facility: the percentage of the subsidiary's stock owned by the parent, whether and to what extent the parent controls the subsidiary's marketing, whether the parent has or exercises authority to execute contracts on behalf of the subsidiary, and whether the parent controls selection, supervision, transfer and similar aspects of employment for those normally employed by the subsidiary.189 Conservation Chemical dealt with the individual liability of a president and stockholder'90 who was held directly liable based upon his own personal participation in the activities that led to the CERCLA violation.191 Corporate officers are subject to personal liability for their own torts, even if they were acting on behalf of the corporation and in an official capacity when they committed the torts. 192 Their liability is direct and, in this sense, is analogous to direct parent liability. The "factors" isolated from Conservation Chemical by the Idarado court are deceptive, however. The Idarado court focused on those facts cited by the Conservation Chemical court that indicated the individual's general control over the subsidiary, and ignored the facts that clearly revealed the defendant was actively involved in the operation of the facility, including the corporation's waste treatment and environmental practices.193 Moreover, the factors cited by the Idarado court address parent involvement in the operation of the subsidiary, not in the involvement of the i facility, as required by CERCLA. Under the facts before it, the Idarado court's incorrect focus on operation of the subsidiary as opposed to operation of the facility probably did not affect the outcome. In addition to fulfilling the Conservation Chemical criteria,194 the parent in Idarado provided advice to the subsidiary on environmental issues (for which it was paid),195 the parent knew of and attempted to address the subsidiary's environmental problems beginning in 1944, and commissioned engineering reports on environmental problems at -

Bifurcation Under CERCLA Page 28 the facility.196 Thus, it would appear that the parent was actively involved in the operation of the facility in addition to the general operation of the subsidiary. Nonetheless, the Idarado court articulated an incorrect standard, thus creating the opportunity for future courts to find parent liability in inappropriate circumstances. A final noteworthy point about the Idarado decision is that the court characterized the parent's relationship with its subsidiary as being one of "pervasive control."'97 This appears to be the first time that this phrase appeared in the context of CERCLA liability. The phrase has surfaced in several direct and indirect liability cases since then, and, as mentioned earlier,198 is a significant contributor to the current confusion in CERCLA case law. Because of the significance of this test in the development of CERCLA jurisprudence, this test is discussed separately below.199 B. The Actual Exercise of Control Test Most courts that have examined the direct liability of parent corporations under CERCLA have adopted a narrower approach than that articulated in Bunker Hill and Idarado. They reject the notion that mere capacity to control the affairs of a subsidiary is sufficient to support direct liability under CERCLA, and look instead for some evidence of actual exercise of control over the subsidiary and its activities.200 As with indirect liability as an owner,201 there is a single appellate decision in this area -- United States v. Kayser-Roth Corp.,202 where the First Circuit held a parent corporation liable as an "operator." The district court's decision in that case203 is almost as interesting as the circuit court's, however, for it illustrates both the progress made in the development of parent liability rules under CERCLA, and the problems that remain. Kayser-Roth arose out of a fairly straightforward set of circumstances. Stamina Mills, a wholly-owned subsidiary of Kayser-Roth Corporation prior to Stamina's dissolution in 1977, ran a textile manufacturing operation from 1952 to 1975. Accidental and deliberate spills of tricholorethylene ("TCE") on Stamina Mills' property resulted in

Bifurcation Under CERCLA Page 29 aquifer contamination and, ultimately, an EPA cleanup action.204 In evaluating KayserRoth's liability for the cleanup costs, the district court carefully distinguished between direct and indirect liability of parent corporations, correctly noting that operator liability is direct, while owner liability is indirect.205 The court looked first at parent corporation liability as an operator. Unfortunately, its analysis of this issue was only partially right. The court correctly stated that the parent's mere status as an owner of a subsidiary was insufficient to render it liable as an operator under CERCLA.206 The court then went on to state, however, that evidence of "pervasive control"207 by the parent over the subsidiary's "management and operations"208 was necessary to es to tablish the parent's liability as an operator. While the court did recognize that actual exercise of control was necessary to support parent liability,209 it incorrectly characterized that control as being over the subsidiary, as opposed to over the facility itself.210 Although the parties stipulated that Stamina Mills was a facility for the purposes of CERCLA,211 a facility is a physical site, not an intangible entity, such as a corporation.212 General managerial control over the facility, absent evidence of exercise of control over the operations of the facility at which the contamination occurred, should not result in a finding of parent liability. Although the district court cited a number of factors that indicated the parent's actual exercise of general managerial control,21 many of the factors it cited in support of its finding that the parent controlled environmental matters specifically were framed in terms of capacity to control, indicating the court's confusion over the nature of the test it was applying.214 In addition, the court went on to hold the parent liable as an owner as well under piercing theory,215 using the same phrase -- "pervasive control" -- in evaluating the parent's indirect liability as an owner as it did in evaluating its direct liability as an operator.216 The district court noted that many of the same factors that supported operator liability also supported owner liability.217 It is frustrating that the first

Bifurcation Under CERCLA Page 30 court to explicitly distinguish between owner and operator liability indelibly blurred that distinction by adopting the same test for liability for both. On appeal, the First Circuit affirmed the district court's finding that the parent was liable as an operator, and so found it unnecessary to address the parent's liability as an owner.218 The appellate court found that the parent exercised "pervasive control" over the subsidiary's operation,219 as evidenced by the parent's total control over the subsidiary's budget, capital expenditures, and real estate transactions, the parent's placement of its own personnel in virtually all of the subsidiary's director and officer positions, and the requirement that the subsidiary direct all government contacts, including those involving environmental matters, to the parent.220 Thus, the First Circuit adopted the district court's determination that pervasive control was sufficient to support direct liability under CERCLA. While the First Circuit declined to articulate "the exact standard necessary" to hold a parent liable as an operator under CERCLA, it noted "that it is obviously not the usual case" for a parent to be the operator of its wholly-owned subsidiary.221 The court then articulated an actual exercise of control test: "To be an operator requires more than merely complete ownership and the concomitant general authority or ability to control that comes with ownership. At a minimum it requires active involvement in the activities of the subsidiary.222 Both the statements indicate that the h First Circuit was making the same error that the district court did, for both incorrectly focus on the parent's activities vis-a-vis the subsidiary, as opposed to vis-a-vis the facility. The question is not whether the parent operates the subsidiary, but whether it operates the facility, and that operation is evidenced by participation in the activities of the facility, not the subsidiary. Control of the subsidiary, if extensive enough, gives rise to indirect liability under piercing doctrine, not direct liability under the statutory language. Later courts have struggled with the same question the Kayser-Roth court faced: how can reasonable curbs be placed on the extent of direct parent corporation liability

Bifurcation Under CERCLA Page 31 under CERCLA? Unfortunately, most of these courts confront the same stumbling-block that the Kayser-Roth court tripped over; they fail to recognize the distinction between operating a subsidiary and operating a facility. For example, in 1991, the United States District Court for the Western District of Michigan, in CPC International, Inc. v. AerojetGeneral Corp.,223 also applied the actual exercise of control test. The court noted that "a parent's mere oversight of a subsidiary's business in a manner appropriate and consistent with the investment relationship between a parent and its wholly owned subsidiary" should not give rise to parent liability as an "operator;"224 rather, a parent may be held directly liable as an operator "only when it has exerted power or influence over its subsidiary by actively participating in and exercising control over the subsidiary's business during a period of disposal of hazardous waste."225 The court then listed activities that would support the imposition of such liability: Factors to consider in assessing whether a parent corporation operated its subsidiary include the parent's participation in the subsidiary's board of directors, management, day-to-day operations, and specific policy matters, including areas such as manufacturing, finances, personnel and waste disposal. In addition, determining the origin and business function of the subsidiary in the context of the parent corporation's business may be helpful in determining whether the parent has operated a wholly owned subsidiary.226 The court specifically noted that evidence "indicative of the actions of a prudent investor, rather than an active operator, including monitoring of a subsidiary's financial performance, consolidation of corporate business matters such as accounting and legal work, and cooperation between the subsidiary and the parent in research," would not, by itself, support a finding that the parent was directly liable under CERCLA.227 Similar factors have been considered by other courts. For example, in denying a parent corporation's motion for summary judgment in Rockwell International Corp. v. IU International Corp.,228 the United States District Court for the Northern District of Illinois ruled that mere capacity to exercise control is insufficient to support liability;

Bifurcation Under CERCLA Page 32 rather, the parent must actually exercise control in order to be held liable.229 Factors that supported a finding of actual exercise of authority in this case included the parent's hiring or approving the hiring of certain officers of the subsidiaries that owned and operated the facility, interlocking officers of the parent and subsidiaries, recommendations by the parent's auditors regarding hazardous waste disposal by the subsidiaries, review of environmental purchases by parent personnel, parent control and monitoring of operational plans and procedures for the facility, and public announcements by the parent that it operated the facility.230 C. Direct Liability of Parent Corporations: A Summary Many courts justify holding parent corporations or other shareholders directly liable for CERCLA cleanup costs as a means of furthering CERCLA's goal of "plac[ing] the ultimate responsibility for cleanup on those responsible" for hazardous waste contamination.231 Although the judiciary's desire to fulfill the intent of Congress is both understandable and laudable, a real danger lies in not inquiring closely into who is actually "responsible" for those problems.232 Certainly, parent corporations should not be able to hide behind their subsidiary's corporate form.233 Nonetheless, a parent corporation's status as a controlling shareholder should not be enough to render it automatically and directly liable for its subsidiary's actions; rather, its liability must rest solely upon a finding that its own actions make it an operator of the offending facility. The courts are making progress, albeit slowly, in clarifying the standards for direct liability under CERCLA. While some courts persist in believing that they must apply a capacity to control test in order to fulfill the public policy purposes of CERCLA,234 most are beginning to recognize that actual control, rather than mere ability to control, is necessary to support direct liability under CERCLA.235 Even though the actual exercise of control test is not always applied properly by the courts,236 their growing recognition that it is the proper standard is reassuring.

Bifurcation Under CERCLA Page 33 Unfortunately, the courts' tendency to discuss direct liability in terms of "pervasive control" has set the actual exercise test skidding on a collision course with the capacity test. The danger of using this type of idiomatic phrase is that it discourages courts from scrutinizing closely the parent's activities for evidence of actual control over the facility, and encourages them instead to settle for evidence of general, unspecified control over the subsidiary, whether exercised or not.237 If the courts were to recognize just two key points, direct liability under CERCLA would be much clearer. First, direct liability of parent corporations is not a novel idea. No broad sweeping statements regarding the remedial purposes of CERCLA are necessary in order to support the imposition of direct liability in an appropriate case, and courts should not regard such liability with suspicion as though it were a foreign or unusual theory of parent liability. Courts need not go through analytical contortions in evaluating the direct liability of parent corporations. They need only turn to traditional, tried-and-true direct liability rules. Second, the courts must stop focusing on the relationship between the parent and the subsidiary, and instead focus on the relationship that is critical under the statute -- that between the parent and thefacility.238 Even those courts that recognize that actual control and not mere capacity is required tend to ignore the parent's role in the operation of the facility.239 A facility is a physical site; it is the "area" where hazardous wastes have "come to be located."240 A subsidiary, on the other hand, is a corporation, an intangible legal entity.241 In order to be an operator, the defendant must exercise control over the actual area where the hazardous substances are located.242 Discussions of the parent's operation of the "subsidiary," as opposed to the facility, are thus inapposite. Although a parent's exercise of complete and tenacious control over the subsidiary may well indicate an exercise of control over the facility, a lesser degree of control over the subsidiary might not lead to the same result.

Bifurcation Under CERCLA Page 34 Control over the subsidiary is an important issue in terms of indirect (piercing) liability. The danger in CERCLA case law is that evidence of a degree of control over the subsidiary that is insufficient to support direct liability as an owner may be coupled with a degree of control over operations of the facility that is similarly insufficient to support a finding of direct liability as an operator. The result of this fuzzy analysis may well be a finding of parental liability, whereas separate analyses of owner and operator liability might well result in a finding of no liability. Courts that focus on capacity to control rather than actual exercise of control and courts that focus on the parent's relationship with the subsidiary rather than the facility are thus apt to make overly broad determinations of direct liability. V. FINDING ORDER IN CHAOS: DEVELOPING A WORKABLE STANDARD FOR OWNER AND OPERATOR LIABILITY UNDER CERCLA It is fruitless to engage in prolonged discussions of congressional intent in trying to interpret and apply CERCLA. The statute is so poorly written243 and so internally inconsistent that conflicting interpretations are rampant.244 The legislative history sheds little light and is virtually useless in divining congressional intent.245 Congress simply failed to address many of the pressing questions regarding PRP liability. The best that courts and commentators can do is try to decipher the overall scheme intended by Congress, and to attempt to fashion liability rules that further that scheme within the bounds of sound public policy. Despite the statutory deficiencies, the problems presented by CERCLA's liability scheme are by no means insurmountable. As the above discussion illustrates, the courts are starting to articulate clearer, more rational bases for finding liability under the statute. Although they have not yet reached the proper endpoint, their movement is going in the right direction. Courts could speed their journey toward a coherent, fair scheme of CERCLA liability if they were to focus on two simple steps: (1) bifurcating owner and

Bifurcation Under CERCLA Page 35 operator liability under CERCLA, and (2) articulating precise, consistent standards for each basis of liability. The First Circuit was the first to recognize the need to bifurcate owner and operator liability. Kayser-Roth246 was decided six months after Joslyn.247 The First Circuit thus had the opportunity to examine the Fifth Circuit's reasoning in handing down its own decision, yet it deliberately rejected the analysis found in Joslyn. The KayserRoth court distinguished Joslyn248 by noting that Joslyn dealt with owner liability, whereas Kayser-Roth addressed operator liability.249 Thus, the First Circuit found that its decision in Kayser-Roth was not in conflict with the Fifth Circuit's decision in Joslyn. The distinction drawn by the Kayser-Roth court is a correct one, and one that should guide future courts in evaluating parent liability under CERCLA.250 A few courts have picked up on Kayser-Roth's owner/operator distinction, and have explicitly adopted separate standards for analyzing a parent corporation's status as an owner versus as an operator.251 In Jacksonville Electric Authority v. Eppinger & Russell Co.,252 for example, the United States District Court for the Middle District of Florida held that a parent corporation could be held liable as an "owner" only where the facts support a piercing of the corporate veil.253 The court quoted the usual "laundry list" of factors that support a piercing,254 and concluded that aside from common stock ownership, common officers, and common directors, the facts before it did not indicate that the parent used the subsidiary as a sham to avoid direct liability.255 The parent and subsidiary "strictly observed" corporate formalities, and "did not have common business departments, did not file consolidated financial statements or tax returns, and did not combine daily operations."256 The parent did not finance the subsidiary, cause its incorporation, use its property as its own, or "pillage" the subsidiary's assets.257 Thus, the court found that the subsidiary's corporate veil could not be pierced and, consequently, the parent could not be held liable as an owner under CERCLA.258

Bifurcation Under CERCLA Page 36 The Jacksonville Electric court then went on to consider (and reject) the parent's liability as an operator.259 While the court recognized the need for bifurcation, its opinion nonetheless reveals some of the same analytical errors made by earlier courts. For example, the Jacksonville Electric court fell into the same trap that snared the Acushnet court20 -- it focused on the parent's involvement in the operations of the subsidiary rather than its involvement in the operations of the facility. The court drew from the analyses of earlier courts261 in finding that operator liability could be imposed: on the parent corporation of a wholly-owned subsidiary when the parent exercises actual and pervasive control of the subsidiary to the extent of actually involving itself in the daily operations of the subsidiary. Actual involvement in the decisions regarding the disposal of hazardous substances is a sufficient, but not a necessary, condition to the imposition of operator liability.262 The standard articulated by the court clearly indicates that operator liability could attach even in the absence of parent involvement in the subsidiary's environmental practices or the running of the facility. However, the court also noted that "[a] parent corporation will almost always have some authority to control a subsidiary's actions (including the decisions about hazardous waste), yet it will not usually be the case where the parent is actively involved in the subsidiary's operations to the point of actually participating in and controlling the conduct which is subject to CERCLA liability."263 This suggests that the court was aware that some sort of involvement in waste disposal activities (and hence in the operation of the facility) was necessary -- that mere involvement in the operation of the subsidiary itself was insufficient. The Jacksonville Electric court did correctly perceive the need to limit operator liability in some ways. The court recognized that the parent's mere ownership of all, or substantially all, of the subsidiary's stock was insufficient to support liability as an "operator" under CERCLA;264 to hold otherwise "would impose automatic liability on the parent corporation of a wholly-owned subsidiary, a result which is not dictated by the plain language of CERCLA."265 Likewise, the mere existence of common officers and

Bifurcation Under CERCLA Page 37 directors, and the parent's receipt of financial status reports from the subsidiary "merely demonstrate[] the concomitant general authority and ability to control that comes with ownership."266 Thus, the court seems to grasp the difference between activities that lead to operator liability and activities that lead to owner liability at some intuitive level; nonetheless, the standard it articulated did not make this critical distinction.267 Similarly, in John Boyd Co. v. Boston Gas Co.,268 the United States District Court for the District of Massachusetts noted that the parent corporation in question could not be held liable under CERCLA as an operator because it did not exercise the "pervasive control" over the subsidiary that the Kayser-Roth court had stated was necessary to support operator liability.269 The facts in John Boyd indicated that the parent corporation had essentially acted as a conduit for the sale of the subsidiary, had been the parent corporation for only one day, and had acted only to vote its shares to liquidate the subsidiary and sell its assets to a third party.270 The court found that "[a] single such corporate act, unconnected in any way to decisions about the operation of the facility in question," would not support operator liability.271 Thus, the John Boyd court (unlike most earlier courts) correctly focused on the parent's relationship with the facility, as opposed to the subsidiary, and correctly recognized that mere ministerial acts, in the absence of any active participation in the facility's activities, could not lead to operator liability. The John Boyd court also examined the parent's liability as an owner, and concluded that under appropriate circumstances, the corporate veil could be pierced to hold a parent corporation liable for the CERCLA violations of its subsidiary.272 The court noted that the statutory goals of CERCLA -- to provide the federal government with the means needed to respond quickly and effectively to hazardous waste problems and to ensure that those responsible for cleanup costs pay them -- indicated that the mere presence of the corporate form was insufficient to shield responsible parties from CERCLA liability.273 By the same token, however, "CERCLA does not make the

Bifurcation Under CERCLA Page 38 corporate form irrelevant."274 The John Boyd court was much persuaded by the Acushnet court's argument that limited liability promotes social utility by encouraging corporations that want to put a waste or generation site to productive use to create "well-capitalized, non-fraudulent, separate corporate subsidiaries."275 As both the Acushnet and John Boyd courts emphasized, no matter how important the policy objectives of CERCLA, there is nothing in the statutory language and nothing inherent in the nature of hazardous waste disposal that supports a policy abolishing the protection of limited liability in all instances.276 Both courts indicated, however, that the objectives of the statute can influence the emphasis that a court will place on the various factors that a court considers in determining whether to pierce a corporate veil,277 and that in CERCLA cases, the extent of the parent's control over the subsidiary's hazardous waste disposal practices is key.278 Because of the absence of facts indicating such an extensive degree of control, the John Boyd court declined to pierce the subsidiary's veil.279 Thus, the first step courts must take in articulating a clear, workable standard for evaluating parent liability under CERCLA is to recognize the distinction between owner and operator liability and to clarify the standards that should be used for each. The second step is to clearly identify the theory under which the parent's liability is being assessed so that the correct standard may be applied. Owner liability, because it is an indirect liability that arises under traditional, common law corporate doctrine, should be based upon the traditional corporate law doctrine of piercing the corporate veil. Neither the Act's statutory language nor public policy considerations dictate that piercing in the CERCLA context should differ in any way from piercing in any other context. The usual factors, such as fraud, inadequate capitalization, intertwining directorates, and failure to observe corporate formalities,280 should prevail. While the threshold for piercing in the CERCLA context should not be higher than it is in other, non-environmental contexts, neither should it be lower. CERCLA's policy objective of making the polluter pay should not be used to overcome

Bifurcation Under CERCLA Page 39 traditional protections of the corporate form. Moreover, the parent's activities with respect to the facility should be regarded as irrelevant to piercing analysis. Rather, the question should be whether the parent would be held liable for any liability of the subsidiary, whether environmental in nature or not. Operator liability, on the other hand, is a direct liability that arises from CERCLA's statutory language and so is constrained by statutory definitions. Thus, operator liability can only be evidenced by extensive parental involvement in and control over the facility; the parent's involvement in the activities of the subsidiary itself are not of importance. Factors that tend to support piercing, such as common directors or inadequate capitalization, but are unrelated to the operation of the facility, should not give rise to liability. Likewise, activities that involve the facility but which are consistent with the parent's investor status, such as monitoring of the subsidiary's performance, supervision of the subsidiary's finance and capital budget decisions, and articulation of general policies and procedures, should not give rise to direct liability of the parent. Rather, operator liability should flow from the parent's active management of the facility as evidenced by its involvement in day-to-day operations of the facility. Much of the current confusion that flows from the courts' failure to distinguish these two grounds of liability could be alleviated if the courts were to eradicate the "pervasive control" test. This test has been used both by courts evaluating a parent's direct liability281 and those evaluating a parent's indirect liability,282 sometimes even appearing in both contexts in a single case.283 The pervasive control test is dangerous for several reasons. First, because the test arises in both the direct and indirect liability contexts, it contributes to the confusion surrounding these two separate theories of liability. The net result of using a single test for liability is that the factors courts consider in evaluating whether a parent corporation is directly or indirectly liable under CERCLA are converging. The extent of parental control over subsidiary finances and operations, intertwining of officers and directors,

Bifurcation Under CERCLA Page 40 parental control over day-to-day operations of the subsidiary, and parental involvement in subsidiary environmental practices are now considered to be relevant to the inquiry under both liability theories.284 This merging of the two underlying theories of liability is at least partially the cause of fears of expanded liability. In addition, by framing parent liability in terms as amorphous as "control," the courts fail to focus on the specific factors that should lead to liability under each theory. This is evident in a 1989 case decided by the United States District Court for the District of Pennsylvania, United States v. Nicolet, Inc.285 In articulating a federal rule for determining when a veil can be pierced under CERCLA, the Nicolet court stated: Where a subsidiary is or was at the relevant time a member of one of the classes of persons potentially liable under CERCLA; and the parent had a substantial financial or ownership interest in the subsidiary; and the parent corporation controls or at the relevant time controlled the management and operations of the subsidiary, the parent's separate corporate existence may be disregarded.286 Of course, every parent corporation, by definition, has "a substantial financial or ownership interest" in its subsidiary. As discussed above, that fact alone is insufficient to support a piercing of the subsidiary's veil.287 The second part of the test, which focuses on the control exercised by the parent over the subsidiary, is essentially a paraphrase of the actual exercise of control test articulated by a number of courts in the direct liability context.288 The Nicolet test thus devolves into a control test for piercing, and it blurs the distinction between direct and indirect liability. If the court had focused upon the traditional multi-factor tests used for determining whether piercing was appropriate,289 the distinction between operator liability, which is based upon control of the facility, and owner liability, which is based upon domination of the subsidiary, would be clearer. The greater problem with the pervasive control test, however, is that it allows courts to engage in an abbreviated analysis in evaluating parent liability. Instead of evaluating the "laundry list" of factors supporting piercing of the corporate veil or delving into the degree of parent involvement in the facility's activities, the court can take

Bifurcation Under CERCLA Page 41 a shortcut to liability by simply looking for evidence of "pervasive control." Thus, the test discourages courts from engaging in a close scrutiny of parent-subsidiary relationships and eliminates the need for a careful articulation of the factors relevant to making a determination of parent liability under each theory. Moreover, "pervasive control," as applied by the courts, doesn't mean pervasive control at all. Factors such as parental control over the subsidiary's hazardous waste practices,290 the subsidiary's financial and/or operational structure,291 and the intertwining of directors and officers292 have all, either singly or in some combination, been considered evidence of pervasive control sufficient to support either direct or indirect liability. As applied by the courts, the pervasive control test simply requires some evidence of control that is inconsistent with the proper separation of corporate entities. However, that improper control may well be at a level that no court would ever normally (in a non-CERCLA context) consider would allow either piercing of the subsidiary's veil or imposition of direct liability upon the parent. As a result, the pervasive control test increases the likelihood that parent corporations will be held liable in inappropriate circumstances. Instead of looking for "pervasive" control, the courts should be looking for evidence of actual control. The type of control necessary differs, however, depending upon the theory being used. Inordinate parental control over the subsidiary's corporate functions and existence as evidenced by the usual laundry list of factors, can result in indirect liability, even in the absence of facts that would show parental involvement in the operation of the facility itself. Direct liability, on the other hand, requires a showing of actual control over and involvement in the activities of the facility that gives rise to the CERCLA violation. The absence of the usual laundry list of factors is irrelevant to the determination of direct parent liability.

Bifurcation Under CERCLA Page 42 VI. CONCLUSION Parent corporation liability under CERCLA need not be the morass that it currently is. The courts simply need to recognize that parent corporation liability can be either direct, based upon the parent's role as an operator, or indirect, based upon its role as an owner. Direct liability requires an actual exercise of control over the facility; indirect liability requires an application of traditional piercing law. Thus, there are two clear, separate tests to use, both reasonably easy to apply, both grounded in traditional legal doctrines. Much of the current confusion in CERCLA case law, and the concomitant fears of expanded liability, can be traced to the extreme positions taken by earlier courts. For example, Joslyn's view that direct liability is never permitted is excessively narrow and in conflict with CERCLA's statutory scheme. Likewise, NEPACCO I's broad reading that virtually everyone and anyone can be liable under CERCLA leads to overly expansive findings of liability. By adhering to the middle ground and carefully articulating the bases and standards for liability, the courts can do much to alleviate the concern commentators and practitioners feel regarding parent liability under CERCLA. Explicit bifurcation of owner and operator liability is necessary to ensure that parent corporations are not held liable for the CERCLA violations of their subsidiaries in inappropriate circumstances. Nothing in CERCLA's language or legislative history indicates that Congress intended to force parent corporations to become insurers of their subsidiaries, or that protections extended to parent corporations under traditional doctrine were to be eradicated. Only by bifurcating owner and operator liability analysis under CERCLA can courts be certain of getting the results that Congress intended.

Bifurcation Under CERCLA Page 43 1. Pub. L. No. 96-510, 94 Stat. 2767 (1980) (codified as amended in scattered sections of 26 U.S.C. and 42 U.S.C. ~~ 9601-9657 (1988) [hereinafter "CERCLA"], amended by the Superfund Amendments and Reauthorization Act of 1986, Pub. L. No. 99-499, 100 Stat. 1613 [hereinafter "SARA"]. The Omnibus Budget Reconciliation Act of 1990, Pub. L. No. 101-508, ~ 6301, 104 Stat. 1388, 1388-319 (1990) (codified at 42 U.S.C. ~ 9611(a)) reauthorizes CERCLA until September 30, 1994, and provides for up to $5.1 billion in funding. 2. See, e.g., Percy L. Angelo & Lynn L. Bergeson, The Expanding Scope of Liability for Environmental Damage and Its Impact on Business Transactions, 8 CORP. L. REV. 101 (1985); John J. Little, Towards Respectfor Corporate Separateness in Defining the Reach of CERCLA Liability, 44 Sw. L.J. 1499 (1991); Tom McMahon & Katie Moertl, The Erosion of Traditional Corporate Law Doctrines in Environmental Cases, 3 NAT. RESOURCES & ENV'T 29 (1988); Gregory P. O'Hara, Minimizing Exposure to Environmental Liabilities for Corporate Officers, Directors, Shareholders, and Successors, 6 SANTA CLARA COMPUTER & HIGH TECH. L.J. 1 (1990); Todd W. Rallison, Comment, The Threat to Investment in the Hazardous Waste Industry: An Analysis of Individual and Corporate Shareholder Liability Under CERCLA, 1987 UTAH L. REV. 585. 3. Their complaints are echoed by the courts. Courts have repeatedly criticized CERCLA both for its sparse legislative history, see, e.g., New York v. Shore Realty Corp., 759 F.2d 1032, 1040 (2d Cir. 1985) (CERCLA is "an eleventh hour compromise" and has an inadequate legislative history); Chemical Waste Mgt., Inc. v. Armstrong World Indus., 669 F. Supp. 1285, 1290 n.6 (E.D. Pa. 1987) ("CERCLA's legislative history is sparse and generally uninformative" and "last-minute additions and deletions to the statute render its legislative history of little practical use"), and for its vague and imprecise statutory language. See, e.g., United States v. Mottolo, 605 F. Supp. 898, 902

Bifurcation Under CERCLA Page 44 (D.N.H. 1985) ("CERCLA has acquired a well-deserved notoriety for vaguely-drafted provisions and an indefinite, if not contradictory, legislative history."); United States v. Price, 577 F. Supp. 1103, 1109 (D.N.J. 1983) (describing CERCLA as "inadequately drafted"); United States v. Wade, 577 F. Supp. 1326, 1331 (E.D. Pa. 1983) (CERCLA "leaves much to be desired from a syntactical standpoint"). 4. Generally, traditional corporate law doctrine shields corporate officers and shareholders from liability for the corporation's actions, unless the officer participated in the tortious or illegal act, or unless the circumstances warrant a piercing of the corporate veil. See, e.g., Frank H. Easterbrook & Daniel R. Fischel, Limited Liability and the Corporation, 52 U. CHI. L. REV. 89, 89-90 (1985) ("The rule of limited liability means that the investors in the corporation are not liable for more than the amount they invest.... The managers and the other workers are not vicariously liable for the firm's deeds."). 5. See infra notes 25-27 and accompanying text (discussing CERCLA's retroactive, strict, joint and several liability scheme). 6. See, e.g., Pennsylvania v. Union Gas Co., 491 U.S. 1, 21 (1989) ("The remedy that Congress felt it needed in CERCLA is sweeping: everyone who is potentially responsible for hazardous-waste contamination may be forced to contribute to the costs of cleanup."); H.R. REP. No. 253 (III), 99th Cong., 1st Sess. 15, reprinted in 1986 U.S.C.C.A.N. 3038, 3038 (noting that Congress' goals in enacting CERCLA were: "(1) to provide for cleanup if a hazardous substance is released into the environment or if such release is threatened and (2) to hold responsible parties liable for the costs of these clean-ups"). 7. See generally infra notes 25-27 and accompanying text (discussing CERCLA's retroactive, strict, and joint and several liability scheme). 8. United States v. Chem-Dyne Corp., 572 F. Supp. 802, 808 (S.D. Ohio 1983). See also Smith Land & Improvement Corp. v. Celotex Corp., 851 F.2d 86, 91 (3d Cir.) ("Congress expected the courts to develop a federal common law to supplement the statute.")

Bifurcation Under CERCLA Page 45 (citations omitted), cert. denied, 488 U.S. 1029 (1988). The legislative history of CERCLA makes it clear that Congress intended the courts to fill in gaps in the statutory scheme. See, e.g., 126 CONG. REC. 31,965 (1980) (statement of Rep. Florio), reprinted in SENATE COMM. ON ENV'T & PUBLIC WORKS, 97TH CONG., 2D SESS., A LEGISLATIVE HISTORY OF THE COMPREHENSIVE ENVIRONMENTAL RESPONSE, COMPENSATION, AND LIABILITY ACT OF 1980 (SUPERFUND), vol. I, at 778 (Comm. Print 1983) [hereinafter A LEGISLATIVE HISTORY] ("Issues of joint and several liability not resolved by this shall be governed by traditional and evolving principles of common law."); id. at 30,932 (statement of Sen. Randolph), reprinted in A LEGISLATIVE HISTORY, supra, vol. I, at 686 ("It is intended that issues of liability not resolved by this act, if any, shall be governed by traditional and evolving principles of common law."). 9. See, e.g., Tanglewood East Homeowners v. Charles-Thomas, Inc., 849 F.2d 1568, 1572 (5th Cir. 1988) (declining to adopt a "crabbed" reading of the statute); Dedham Water Co. v. Cumberland Farms Dairy, Inc., 805 F.2d 1074, 1081 (1st Cir. 1986) (noting that "CERCLA is essentially a remedial statute" and that the court "therefore [is] obligated to construe its provisions liberally to avoid frustration of the beneficial legislative purposes"). 10. Perhaps nowhere is this clearer than in the context of parent corporation liability, where the only two circuit court cases decided to date on the issue are in conflict. Compare Joslyn Mfg. Co. v. T.L. James & Co., 893 F.2d 80 (5th Cir. 1990) (holding that a parent corporation can be held liable for its subsidiary's CERCLA violation only where the circumstances permit a piercing of the corporate veil to reach the parent as an "owner"), cert. denied, 498 U.S. 1108 (1991) (discussed infra notes 122-39 and accompanying text), with United States v. Kayser-Roth Corp., 910 F.2d 24 (ist Cir. 1990) (holding that a parent can be held directly liable as an "operator" for its subsidiary's CERCLA violation), cert. denied, 498 U.S. 1084 (1991) (discussed infra notes 202-22 and accompanying text).

Bifurcation Under CERCLA Page 46 11. See United States v. Northeastern Pharmaceutical & Chem. Co. (NEPACCO), 579 F. Supp. 823 (W.D. Mo. 1984), affd in part, rev'd in part, 810 F.2d 726, 732-33 (8th Cir. 1986), cert. denied, 484 U.S. 848 (1987) (discussed infra Part I1). The first parent corporation case was not decided until 1986. See Idaho v. Bunker Hill Co., 635 F. Supp. 665 (D. Idaho 1986) (discussed infra notes 174-85 and accompanying text). 12. For a discussion of the rules in the contexts of corporate officers, individual shareholders, parent corporations, and successor corporations, see generally Lynda J. Oswald & Cindy A. Schipani, CERCLA and the "Erosion" of Traditional Corporate Law Doctrine, 86 Nw. L. REV. 259 (1992). 13. CERCLA was enacted by a lame-duck Congress, and was the result of a number of last-minute political compromises -- a fact which undoubtedly explains at least in part the statute's poor drafting. For example, two of CERCLA's precursors, the Environmental Emergency Response Act, S. 1480, 96th Cong., 1st Sess. (1979), and the Comprehensive Oil Pollution Liability and Compensation Act, H.R. 85, 96th Cong., 1st Sess. (1979), contained a strict liability standard, but that standard was omitted from the final version of CERCLA as a result of opposition by a number of legislators. See, e.g., 126 CONG. REC. 26,786 (1980), reprinted in A LEGISLATIVE HISTORY, supra note 8, vol. II, at 358-59 (statement of Rep. Stockman) (arguing that strict liability was an inappropriate standard for CERCLA). As a compromise, section 101(32) of CERCLA, 42 U.S.C. ~ 9601(32) (1988), provides that the standard of liability under CERCLA is the same as the standard of liability under section 311 of the Clean Air Act, 33 U.S.C. ~ 1321 (1988), which is not explicit but which has been interpreted by the courts as being strict liability. See generally Lynda J. Oswald, Strict Liability of Individuals Under CERCLA: A Normative Analysis, 20 B.C. ENVTL. AFF. L. REV. 579, 598-603 (1993) (discussing the development of strict liability under CERCLA). While the strict liability standard was thus never stated explicitly in the enacted version of CERCLA, proponents of the strict liability standard emphasized that it remained the operative standard for determining liability.

Bifurcation Under CERCLA Page 47 See, e.g., 126 CONG. REC. 30,932 (1980), reprinted in A LEGISLATIVE HISTORY, supra note 8, vol. I, at 686 (statement of Sen. Randolph) (noting that "changes were made in recognition of the difficulty in prescribing in statutory term liability standards which will be applicable in individual cases," but emphasizing that the standard remained strict liability). 14. CERCLA is up for reauthorization in 1994, and numerous bills have been introduced in the 103rd Congress as a result. These bills all deal with discrete, incremental changes in CERCLA, however, such as limiting municipal liability for solid waste, see H.R. 541, 103d Cong., 1st Sess. (1993), clarifying the "innocent landowner" defense, see H.R. 570, 103d Cong., 1st Sess. (1993), addressing State contribution rules, H.R. 768, 103d Cong., 1st Sess. (1993), and providing for interim response actions, see H.R. 1125, 103d Cong., 1st. Sess. (1993). The Clinton Administration has vehemently opposed proposals for extensive redrafting of CERCLA's retroactive, strict, and joint and several liability provisions. See Raymond B. Ludwieszewski, Superfund Liability at Issue, NAT'L L.J. (June 14, 1993), at p.29. 15. See, e.g., Joslyn Mfg. Co. v. T.L. James & Co., Inc., 893 F.2d 80, 82 (5th Cir. 1990) (holding that parent corporations may only be held indirectly liable under CERCLA, and that direct liability is not permitted by the Act), cert. denied, 498 U.S. 1108 (1991) (discussed infra notes 122-39 and accompanying text). 16. See generally infra notes 281-92 and accompanying text (discussing the pervasive control test). 17. Cases involving individual shareholders are relatively rare, however. In addition, to date, every case imposing liability upon an individual shareholder has involved an active shareholder of a closely-held corporation. See, e.g., New York v. Shore Realty Corp., 759 F.2d 1032 (2d Cir. 1985); United States v. Carolina Transformer Co., 739 F. Supp. 1030 (E.D.N.C. 1989); United States v. Bliss, 20 Envtl. L. Rep. (Envtl. L. Inst.) 20,879 (E.D. Mo. 1988); United States v. Northernaire Plating Co., 670 F. Supp. 742 (W.D.

Bifurcation Under CERCLA Page 48 Mich. 1987); United States v. Conservation Chem. Co., 628 F. Supp. 391 (W.D. Mo. 1986); United States v. Ward, 618 F. Supp. 884 (E.D.N.C. 1985). The liability of these individuals is more correctly based in their actions as officers of the corporation, not in their status as shareholders. See generally Oswald, supra note 13, at 630-32. 18. See, e.g., 126 CONG. REC. 30,931 (1980) (remarks of Sen. Randolph, co-sponsor of CERCLA), reprinted in A LEGISLATIVE HISTORY, supra note 8, vol. I, at 684 (CERCLA was the result of a public concern regarding hazardous waste disposal that had been brought to the "national consciousness, largely as a result of the severe public health problems discovered at Love Canal"); S. REP. No. 848, 96th Cong., 2d Sess. 8-10 (1980), reprinted in A LEGISLATIVE HISTORY, supra note 8, vol. I, at 315-17. 19. See S. REP. No. 848, supra note 18, at 7, reprinted in A LEGISLATIVE HISTORY, supra note 8, vol. I, at 314. 20. Other major federal environmental statutes addressing hazardous waste disposal issues include the Resource Conservation and Recovery Act ("RCRA"), 42 U.S.C. ~~ 6901-92k (1988); the Clean Water Act, 33 U.S.C. ~~ 1251-1387 (1988); the Clean Air Act, 42 U.S.C. ~~ 7401-7642 (1988); the Toxic Substances Control Act, 15 U.S.C. ~~ 2601-29 (1988); and the Rivers and Harbors Appropriation Act of 1899, 33 U.S.C. ~~ 401-67 (1988). 21. For example, RCRA, 42 U.S.C. ~~ 6901-92K (1988), which was the primary statute dealing with hazardous waste issues prior to enactment of CERCLA, "is prospective and applies to past sites only to the extent that they are posing an imminent hazard." H.R. REP. No. 1016, 96th Cong., 2d Sess., pt. 1, at 22, reprinted in 1980 U.S.C.C.A.N. 6119, 6125. 22. See, e.g., United States v. Shell Oil Co., 605 F. Supp. 1064, 1072 (D. Colo. 1985) (CERCLA "is by its very nature backward looking. Many of the human acts that have caused the pollution already had taken place before its enactmentll").

Bifurcation Under CERCLA Page 49 23. See RICHARD H. GASKINS, ENVIRONMENTAL ACCIDENTS: PERSONAL INJURY AND RESPONSIBILITY 64-65, 231 (1989). At the time of CERCLA's enactment in 1980, the Environmental Protection Agency ("EPA") estimated that the United States produced 57 million metric tons of hazardous waste per year -- about 600 pounds per citizen. S. REP. No. 848, supra note 18, at 3, reprinted in A LEGISLATIVE HISTORY, supra note _, vol. I, at 310. A recent study by the University of Tennessee's Waste Management Research and Education Institute estimates that cleanup of Superfund sites is likely to cost between $106 and $302 billion over the next 30 years (measured in 1990 dollars). See MILTON RUSSELL ET AL., HAZARDOUS WASTE REMEDIATION: THE TASK AHEAD 16 (Dec. 1991). The EPA has estimated that cleanup will cost $300 to $500 billion, excluding Department of Energy sites. See OFFICE OF TECHNOLOGY ASSESSMENT, ASSESSING CONTRACTOR USE IN SUPERFUND 1 n.l (1989), reprinted in 17 CHEM. WASTE LITIG. REP. 715 (1989). 24. See, e.g., supra note 6 (quoting H.R. REP. No. 253 (III)). 25. See United States v. NEPACCO, 810 F.2d 726,732-33 (8th Cir. 1986) ("Although CERCLA does not expressly provide for retroactivity, it is manifestly clear that Congress intended CERCLA to have retroactive effect."), cert. denied, 484 U.S. 848 (1987); See generally Michael P. Healy, Direct Liability for Hazardous Substance Cleanups Under CERCLA: A Comprehensive Approach, 42 CASE W. RES. L. REV. 65, 82-85 (1992). 26. Although the statute does not specifically provide for joint and several liability, the courts have determined that such liability is appropriate in cases of indivisible harm. See, e.g., O'Neil v. Picillo, 883 F.2d 176, 178-79 (1st Cir. 1989), cert. denied, 493 U.S. 1071 (1990); United States v. Monsanto Co., 858 F.2d 160, 171 (4th Cir. 1988), cert. denied, 490 U.S. 1106 (1989); United States v. Chem-Dyne Corp., 572 F. Supp. 802, 810-11 (S.D. Ohio 1983); see generally Alice T. Valder, Note, The Erroneous Site Selection Requirement for Arranger and Transporter Liability Under CERCLA, 91 COLUM. L. REV. 2074, 2079-81 (1991) (discussing joint and several liability under CERCLA). Costs may be allocated among PRPs based upon principles of contribution or equitable

Bifurcation Under CERCLA Page 50 apportionment. See Southland Corp. v. Ashland Oil, Inc., 696 F. Supp. 994, 1003 (D.N.J. 1985). SARA added an express right of contribution among PRPs. See 42 U.S.C. ~ 9613(f)(1) (1988) ("Any person may seek contribution from any other person who is liable or potentially liable under CERCLA.") See generally F. James Handley, CERCLA Contribution Protection: How Much Protection?, 22 Envtl. L. Rep. (Envtl. L. Inst.) 10,542 (1992); Stephen Baird Russo, Note, Contribution Under CERCLA: Judicial Treatment After SARA, 14 COLUM. J. ENVTL. L. 267 (1989). 27. See, e.g., Tanglewood East Homeowners v. Charles-Thomas, Inc., 849 F.2d 1568, 1572 (5th Cir. 1988); New York v. Shore Realty Corp., 759 F.2d 1032, 1042 (2d Cir. 1985). For a discussion of the legislative history of the strict liability standard of CERCLA, see Oswald, supra note 13, at 598-603. 28. 42 U.S.C. ~ 9607(a) (1988). CERCLA states that responsible persons shall be liable for "all costs of removal or remedial action incurred by the United States Goverment"; "any other necessary costs of response incurred by any other person;" "damages for injury to, destruction of, or loss of natural resources;" and "the costs of [certain specified] health assessment or health effects stud[ies]." Id. 29. Id. ~ 9607(b) provides: (b) There shall be no liability under subsection (a) of this section for a person otherwise liable who can establish by a preponderance of the evidence that the release or threat of release of a hazardous substance and the damages resulting therefrom were caused solely by -- (1) an act of God; (2) an act of war; (3) an act or omission of a third party other than an employee or agent of the defendant, or than one whose act or omission occurs in connection with a contractual relationship, existing directly or indirectly, with the defendant (except where the sole contractual arrangement arises from a published tariff and acceptance for carriage by a common carrier by rail), if the defendant establishes by a preponderance of the evidence that (a) he exercised due care with respect to the hazardous substance concerned, taking into consideration the characteristics of such hazardous substance, in light of all relevant facts and circumstances, and (b) he took precautions against foreseeable acts or omissions of any such third party and the consequences that could foreseeably result from such acts or omissions; or (4) any combination of the foregoing paragraphs.

Bifurcation Under CERCLA Page 51 For a discussion of these defenses, see generally Frank P. Grad, A Legislative History of the Comprehensive Environmental Response, Compensation and Liability ("Superfund") Act of 1980, 8 COLUM. J. ENVTL. L. 1, 9 (1982); Developments in the Law -- Toxic Waste Litigation, 99 HARV. L. REV. 1458, 1518 (1986) [hereinafter Developments]. 30. See, e.g., Amoco Oil Co. v. Borden, Inc., 889 F.2d 664, 668 (5th Cir. 1989); United States v. Aceto Agric. Chems. Corp., 872 F.2d 1373, 1379 (8th Cir. 1989); Ascon Properties, Inc. v. Mobil Oil Corp., 866 F.2d 1149, 1152-53 (9th Cir. 1989); Developments, supra note 29, at 1511-42. 31. CERCLA defines "facility" as: (A) any building, structure, installation, equipment, pipe or pipeline (including any pipe into a sewer or publicly owned treatment works), well, pit, pond, lagoon, impoundment, ditch, landfill, storage container, motor vehicle, rolling stock, or aircraft, or (B) any site or area where a hazardous substance has been deposited, stored, disposed of, or placed, or otherwise come to be located; but does not include any consumer product in consumer use or any vessel. 42 U.S.C. ~ 9601(9) (1988). The courts have applied the definition of "facility" broadly. See, e.g., Tanglewood East Homeowners v. Charles-Thomas, Inc., 849 F.2d 1568, 1573 (5th Cir. 1988) (residential subdivision is a "facility"); Washington v. Time Oil Co., 687 F. Supp. 529, 532 (W.D. Wash. 1988) (filter cake); United States v. Bliss, 667 F. Supp. 1298, 1305 (E.D. Mo. 1987) (stable, road, and spray trucks); United States v. Ward, 618 F. Supp. 884, 895 (E.D.N.C. 1985) (roadside); New York v. General Elec. Co., 592 F. Supp. 291, 295-97 (N.D.N.Y. 1984) (dragstrip); United States v. Metate Asbestos Corp., 584 F. Supp. 1143, 1148 (D. Ariz. 1984) (trailer park). 32. 42 U.S.C. ~ 9601(22), (8) (1988). 33. Section 101(14) of CERCLA defines "hazardous substances" to include substances defined as hazardous or toxic under section 311 of the Clean Water Act, 33 U.S.C. ~ 1321(b)(2)(A) (1988); section 307(a) of the Clean Water Act, 33 U.S.C. ~ 1317(a) (1988); section 112 of the Clean Air Act, 42 U.S.C. ~ 7412 (1988); and section 7 of the Toxic Substances Control Act, 15 U.S.C. ~ 2606 (1988). CERCLA specifically does not

Bifurcation Under CERCLA Page 52 apply to the release of petroleum, crude oil, or natural gas. See 42 U.S.C. ~ 9601(14) (1988). 34. Section 107 of CERCLA states in relevant part: (a) Notwithstanding any other provision or rule of law, and subject only to the defenses set forth in subsection (b) of this section -- (1) the owner and operator of a vessel or facility, (2) any person who at the time of disposal of any hazardous substance owned or operated any facility at which such hazardous substances were disposed of, (3) any person who by contract, agreement, or otherwise arranged for disposal or treatment, or arranged with a transporter for transport for disposal or treatment, of hazardous substances owned or possessed by such person, by any other party or entity, at any facility or incineration vessel owned or operated by another party or entity and containing such hazardous substances, and (4) any person who accepts or accepted any hazardous substances for transport to disposal or treatment facilities, incineration vessels or sites selected by such person, from which there is a release, or a threatened release which casuses the incurrence of response costs, of a hazardous substance, shall be liable for -- (A) all costs of removal or remedial action incurred by the United States Government or a State or an Indian tribe not inconsistent with the national contingency plan; (B) any other necessary costs of response incurred by any other person consistent with the national contingency plan; (C) damages for injury to, destruction of, or loss of natural resources, including the reasonable costs of assessing such injury, destruction, or loss resulting from such release; and (D) the costs of any health assessment or health effects study carried out under section 9604(i) of this title. 42 U.S.C. ~ 9607(a) (1988). 35. "The term 'person' means an individual, firm, corporation, association, partnership, consortium, joint venture, commercial entity, United States Government, State, municipality, commission, political subdivision of a State, or any interstate body." Id. ~ 9601(21). 36. Early cases quickly disposed of the argument that shareholders could not be held liable under CERCLA. The rationale offered by the court in United States v. McGrawEdison Co., 718 F. Supp. 154, 157 n.l (W.D.N.Y. 1989) (citation omitted), is typical: [B]oth individuals and corporations are included within the definition of "person" under section 101(21) of CERCLA. Accordingly, if an individual stockholder can be liable under CERCLA for his corporation's

Bifurcation Under CERCLA Page 53 disposal, a corporation which holds stock in another corporation (e.g., a subsidiary) and actively participates in its management can be held liable for cleanup costs incurred as a result of that corporation's disposal. 37. Few cases have addressed parent corporation liability as generators or transporters. In City of New York v. Exxon Corp., 112 B.R. 540, 547 (S.D.N.Y. 1990), however, the court determined that the standard for holding a parent corporation directly liable as a generator or transporter should be the same as that used to hold a parent corporation directly liable. 38. See supra note 30 and accompanying text. 39. Section 107(a)(1) (addressing the liability of present owners and operators) is written in the conjunctive; nonetheless, the courts have consistently interpreted the phrase in the disjunctive, extending CERCLA liability to entities who currently own or operate a facility. See, e.g., United States v. Fleet Factors Corp., 901 F.2d 1550, 1554 n.3 (11th Cir. 1990) ("Although the 'owner and operator' language of ~ 9607(a)(1) is in the conjunctive, we construe this language in the disjunctive in accordance with the legislative history of CERCLA and the persuasive interpretations of other courts.") (citing United States v. Maryland Bank & Trust Co., 632 F. Supp. 573, 577-78 (D. Md. 1986); Guidice v. BFG Electroplating & Mfg. Co., 732 F. Supp. 556, 561 (W.D. Pa. 1989); Artesian Water Co. v. New Castle County, 659 F. Supp. 1269, 1280 (D. Del. 1987), affd, 851 F.2d 643 (3d Cir. 1988), cert. denied, 111 S. Ct. 752 (1991)), cert. denied, 111 S. Ct. 752 (1991); Tanglewood East Homeowners v. Charles-Thomas, Inc., 849 F.2d 1568, 1573 (5th Cir. 1988) ("[C]ourts addressing this issue have rejected this argument... that liability may be imposed only on those persons who both own and operate polluted property."). Mere status as the owner or operator of a facility is sufficient to result in liability under this provision; no affirmative act is required. Thus, current owners or operators of a facility that releases or threatens to release hazardous substances are liable for cleanup even if the disposal of the substance occurred prior to their ownership. See, e.g., Tanglewood East, 849 F.2d at 1572.

Bifurcation Under CERCLA Page 54 40. 42 U.S.C. ~ 9601(20)(A) (1988) provides: The term "owner or operator" means (i) in the case of a vessel, any person owning, operating, or chartering by demise, such vessel, (ii) in the case of an onshore facility or an offshore facility, any person owning or operating such facility, and (iii) in the case of any facility, title or control of which was conveyed due to bankruptcy, foreclosure, tax delinquency, abandonment, or similar means to a unit of State or local government, any person who owned or operated, or otherwise controlled activities at such facility immediately beforehand. Such term does not include a person who, without participating in management of a vessel or facility, holds indicia of ownership primarily to protect his security interest in the vessel or facility. 41. See id. This exception is discussed further infra notes 89-94 and accompanying text. 42. See, e.g., United States v. Jon-T Chems., Inc. 768 F.2d 686, 690 (5th Cir. 1985), cert. denied, 475 U.S. 1011 (1986) (citations omitted): Under the doctrine of limited liability, the owner of a corporation is not liable for the corporation's debts. Creditors of the corporation have recourse only against the corporation itself, not against its parent company or shareholders. It is on this assumption that 'large undertakings are rested, vast enterprises are launched, and huge sums of capital are attracted.' See also Easterbrook & Fischel, supra note 4, at 89-90. 43. See, e.g., Krivo Indus. Supply Co. v. National Distillers & Chem. Corp., 483 F.2d 1098, 1102 (5th Cir. 1973), modified per curiam, 490 F.2d 916 (5th Cir. 1974): Basic to the theory of corporation law is the concept that a corporation is a separate entity, a legal being having an existence separate and distinct from that of its owners. This attribute of the separate corporate personality enables the corporation's stockholders to limit their personal liability to the extent of their investment.... The corporate form, however, is not lightly disregarded, since limited liability is one of the principal purposes for which the law has created the corporation. See also HENRY BALLANTINE, BALLANTINE ON CORPORATIONS ~ 1, at 1,4, ~ 1 19, at 288 (1946); HARRY G. HENN & JOHN R. ALEXANDER, LAWS OF CORPORATIONS ~ 73 (3d ed. 1983). Several theoretical advantages have been advanced in support of limited liability, including minimizing risk exposure of absentee investors, encouraging very large scale enterprises and portfolio diversification, minimizing agency costs, maintaining efficiency of the capital market, and minimizing creditors' collection costs and the costs of

Bifurcation Under CERCLA Page 55 contracting around liability. See PHILLIP BLUMBERG, THE LAW OF CORPORATE GROUPS: TORT, CONTRACT AND OTHER COMMON LAW PROBLEMS IN THE SUBSTANTIVE LAW OF PARENT AND SUBSIDIARY CORPORATIONS ~ 4.02 (1987); Easterbrook & Fischel, supra note 4, at 89-97; Richard Posner, The Rights of Creditors of Affiliated Corporations, 43 U. CHI. L. REV. 499, 502-13 (1976). It has been argued that these justifications are not relevant to affiliated corporations, see BLUMBERG, supra, at ~ 5.01, leading some commentators to argue that unlimited liability would be more appropriate for affiliated corporations. See Paul Halpern et al., An Economic Analysis of Limited Liability in Corporation Law, 30 U. TORONTO L.J. 117, 148, 150 (1980). Nonetheless, limited liability persists as the rule for affiliated corporations. 44. 1 WILLIAM FLETCHER, FLETCHER CYCLOPEDIA OF THE LAW OF PRIVATE CORPORATIONS ~ 1.07, at 25 (rev. perm. ed. 1986); see also HENN & ALEXANDER, supra note _, ~ 68, at 127 ("For most purposes, [a corporation] is a person separate and apart from the persons who compose it."). 45. HENRY BALLANTINE, BALLANTINE ON CORPORATIONS ~ 118 (rev. ed. 1946). 46. See generally HENN & ALEXANDER, supra note 43, ~ 148, at 355. 47. HENN & ALEXANDER, supra note 43, ~ 148, at 355 ("[T]he parent corporation and its subsidiary are treated as separate and distinct legal persons even though the parent owns all the shares in the subsidiary and the two enterprises have identical officers and directors.") (footnote omitted). See also id. ~ 146, at 347 ("The prevailing rule is that where corporate formalities are substantially observed, initial financing reasonably adequate, and the corporation not formed to evade an existing obligation or a statute or to cheat or to defraud, even a controlling shareholder enjoys limited liability.") (footnote omitted); 1 FLETCHER, supra note 44, ~ 33, at 568 ("Neither does the mere fact that there exists a parent-subsidiary relationship between two corporations make the one liable for the torts of the affiliate.") (footnote omitted).

Bifurcation Under CERCLA Page 56 48. See BLACK'S LAW DICTIONARY 1004, 1280 (5th ed. 1979) (a parent corporation owns more than fifty percent of the stock of another corporation). A subsidiary is defined as an entity "controlled by another corporation by reason of the latter's ownership of at least a majority of the shares of the capital stock." 18 AM. JUR. 2D Corporations ~ 35 (1985) (footnote omitted). 49. Generally, the separate existence of the subsidiary or other affiliated corporation will be recognized unless: (a) The business transactions, property, employees, bank and other accounts and records of the corporation are intermingled; (b) The formalities of separate corporate procedures for each corporation are not observed (where the directors and officers of each corporation are common, separate meetings and delineation of the respective capacities in which the common directors and officers are acting should be observed); (c) The corporation is inadequately financed as a separate unit from the point of view of meeting its normal obligations forseeable in a business of its size and character, because of either initial inadequate financing or having its earnings drained off so as to keep it in a condition of financial dependency; (d) The respective enterprises are not held out to the public as separate enterprises; (e) The policies of the corporation are not directed to its own interests primarily but rather to those of the other corporation. 50. See infra notes 97-108 and accompanying text (discussing indirect liability under traditional law) and notes 163-68 and accompanying text (discussing direct liability under traditional law). 51. See Kathryn R. Heidt, Liability of Shareholders Under the Comprehensive Environmental Response Compensation and Liability Act (CERCLA), 52 OHIO ST. L.J. 133, 174-75 (1991) (noting that over one-half of courts have confused the terms "owner" and operator"); Oswald & Schipani, supra note 12, at 300 ("[c]ourts typically connect the two terms as a single phrase -- 'owner or operator' -- and fail to distinguish between the

Bifurcation Under CERCLA Page 57 grounds supporting the imposition of liability upon the two categories of potentially responsible parties"). 52. See Joslyn, 893 F.2d at 83. 53. See 42 U.S.C. ~ 9607(a) (1988) (quoted in note 29 supra). 54. The Fifth Circuit explicitly recognized this point in the context of an individual shareholder. See Riverside Market Dev. Corp. v. International Bldg. Prods., Inc., 931 F.2d 327, 330 (5th Cir. 1991) (noting that "'[t]he property of the corporation is its property, and not that of the stockholders, as owners") (quoting 1 FLETCHER, supra note 44, ~ 31, at 555), cert. denied, 112 S. Ct. 636 (1991). See also Amcast Industrial Corp. v. Detrex Corp., No. S88-620, 1990 U.S. Dist. LEXIS 15191, *8 (N.D. Ind. 1990) (holding that former shareholders of a corporation were not liable as owners of a "facility" owned by the corporation because the statute does not "even remotely suggests a congressional intent to abrogate the common law of corporations by subjecting stockholders to liability as the 'owners' of corporation property."). See generally HENN & ALEXANDER, supra note 43, ~ 71 (shareholders do not own assets of the corporation); Oswald, supra note, at 631 (same). 55. See Allen Kezsbom et al., "Successor" and "Parent" Liability for Superfund Cleanup Costs: The Evolving State of the Law, 10 VA. ENVTL. L.J. 45, 60 (1990) (noting that "[o]perator liability holds a parent liable for its own conduct in relation to the operation of the facility through the subsidiary, not for the subsidiary's conduct"). 56. Id. at 66. 57. See HENN & ALEXANDER, supra note 43, at ch. 9(D) (discussing rights of shareholders). 58. See infra notes 238-42 and accompanying text. See also Healy, supra note 25, at 119 ("The distinction between the direct and derivative liability of shareholders is plain; direct liability is based on the parties' actual conduct while derivative liability depends on formal relationships.").

Bifurcation Under CERCLA Page 58 59. See Healy, supra note 25, at 119. 60. 579 F. Supp. 823 (W.D. Mo. 1984), affd in part, rev'd in part, 810 F.2d 726 (8th Cir. 1986), cert. denied, 484 U.S. 848 (1987). 61. 579 F. Supp. at 828. 62. Id. 63. Id. 64. Id. at 830. 65. Id. 66. Id. 67. 579 F. Supp. 823 (W.D. Mo. 1984). 68. In addition, the court found that Mills was liable as a transporter under CERCLA ~~ 101(26), 104, 106(a), and 107(a)(4), 42 U.S.C. ~~ 9601(26), 9604, 9606(a), 9607(a)(4) (1988), and that NEPACCO was liable under CERCLA ~~ 104, 106(a), and 107(a)(1) & (3), 42 U.S.C. ~~ 9604, 9606(a), 9607(a)(l) & (3) (1988). 69. See Healy, supra note 25, at 110-15. 70. 42 U.S.C. ~ 9601(20)(A) (1988), quoted in NEPACCO I, 579 F. Supp. at 848. 71. 579 F. Supp. at 848. 72. Id. (footnote omitted). 73. Id. at 848 (discussing Lee's liability) & 849 (discussing Michaels' liability). 74. 464 F.2d 1124, 1127 (5th Cir. 1972) (quoted in Apex Oil Co. v. United States, 530 F.2d 1291 (8th Cir. 1976)). 75. Id. at 1293, quoted in NEPACCO 1, 579 F. Supp. at 848. 76. 579 F. Supp. at 848-49. The court noted that: Defendant Lee had the capacity to control the disposal of hazardous waste at the NEPACCO plant; the power to direct the negotiations concerning the disposal of wastes at the Denney farm site; and the capacity to prevent and abate the damage caused by the disposal of hazardous wastes at the Denney farm site. Finally, Lee was a major stockholder in NEPACCO and activally participated in the management of NEPACCO in his capacity as vice-president.

Bifurcation Under CERCLA Page 59 Id. at 849. Michaels, on the other hand: was the founder and president of the NEPACCO operation in Versona, as well as a major stockholder in the corporation. Michaels had the capacity and general responsibility as president to control the disposal of hazardous wate at the NEPACCO plant; the power to direct the negotiations concerning the disposal of wastes at the Denney farm site; and the capacity to prevent and abate the damage caused by the disposal of hazardous wastes at the Denney farm site. Id. 77. 810 F.2d at 743. The Eighth Circuit, like the district court, rejected Lee's contention that he could not be liable as a generator under CERCLA ~ 107(a)(3), 42 U.S.C. ~ 9607(a)(3), because the corporation owned or possessed the hazardous substances, not him. The court stated that "[i]t is the authority to control the handling and disposal or hazardous substances that is critical under the statutory scheme," and that "requiring proof of personal ownership or actual physical possession of hazardous substances as a precondition for liability under CERCLA ~ 107(a)(3)... would be inconsistent with the broad remedial purpose" of the statute. 810 F.2d at 743. 78. Id. at 742-43. 79. See, e.g., Colorado v. Idarado Mining Co., 18 Envtl. L. Rep. (Envtl. L. Inst.) 20,578 (1987), rev'd on other grounds, 916 F.2d 1486 (10th Cir. 1990), cert. denied, 111 S. Ct. 1584 (1991). 80. See, e.g., New York v. Shore Realty Corp., 759 F.2d 1032 (2d Cir. 1985); Idaho v. Bunker Hill Co., 635 F. Supp. 665 (D. Idaho 1986); United States v. Conservation Chem. Co., 628 F. Supp. 391 (W.D. Mo. 1986). 81. See, e.g., Richard G. Dennis, Liability of Officers, Directors and Shareholders Under CERCLA: The Casefor Adopting State Law, 36 VILL. L. REV. 1367, 1385-87 (1991); Healy, supra note 25, at 111-13; Donald B. Mitchell, Jr., Joslyn Manufacturing Co. v. T.L. James & Co., Inc.: Integrating CERCLA with the American Common-Law Tradition, 17 COLUM. J. ENVTL. L. 53, 58-59 (1992).

Bifurcation Under CERCLA Page 60 82. 579 F. Supp. at 848. In rejecting the government's contention that statutory immunity is extended only to natural persons and that a corporation was not therefore within the class of persons to whom statutory immunity was available, the Mobil Oil court stated: The owner-operator of a vessel or a vacility [sic] has the capacity to make timely discovery of oil discharges. The owner-operator has power to direct the activities of persons who control the mechanisms causing the pollution. The owner-operator has the capacity to prevent and abate damages. Accordingly, the owner-operator of a facility governed by the [Clean Water Act]... must be regarded as as "person in charge" of the facility for the purposes of ~ 1161. 464 F.2d at 1127. The court went on to state: "A more restrictive interpretation would frustrate congressional purpose by exempting from the operation of the Act a large class of persons who are uniquely qualified to assume the burden imposed by it." Id. The Mobil Oil court thus was concerned with defining who was the "person in charge" and not with the characteristics of an "owner-operator." Moreover, "owner or operator" and "person in charge" are not synonymous terms. See Apex Oil Co. v. United States, 530 F.2d 1291, 1294 n.8 (8th Cir. 1975) ("We do not suggest that the class of persons included within the meaning of 'owner or operator' and 'person in charge' are identical."). 83. Under section 1161 of the Clean Water Act, 33 U.S.C. ~ 1 161(b)(4) (remanded and currrently codified at 33 U.S.C. ~ 1321(b)(5)), "persons in charge" of a vessel or facilitymust notify the appropriate federal agency of oil discharges into surrounding water. The statute also provides that information acquired through mandatory disclosure cannot be used for criminal prosecution of the discloser. 84. The definition of "owner or operator" under the Clean Water Act is remarkably similar to that under CERCLA. 33 U.S.C. ~ 1321(a)(6) (1988) states: "[O]wner or operator" means (A) in the case of a vessel, any person owning, operating, or chartering by demise, such vessel, and (B) in the case of an onshore facility, and an offshore facility, any person owning or operating such onshore facility or offshore facility, and (C) in the case of any abandoned offshore facility, the person who owned or operated such facility immediately prior to such abandonment....

Bifurcation Under CERCLA Page 61 CERCLA's definition of "owner or operator" is found at 42 U.S.C. ~ 9601(20)(A) (quoted in supra note 40). 85. See 464 F.2d at 1127 (quoted in supra note 82). 86. See Healy, supra note 25, at 113-14 & n.201. 87. See, e.g., Idaho v. Bunker Hill Co., 635 F. Supp. 665, 672 (D. Idaho 1986); Colorado v. Idarado Mining Co., 18 Envtl. L. Rep. (Envtl. L. Inst.) 20,578, 20,578 (D. Colo. 1987), rev'd on other grounds, 916 F.2d 1486 (10th Cir. 1990), cert. denied, 111 S. Ct. 1584 (1991). See generally Healy, supra note 25, at 115-18 (discussing cases). 88. Although NEPACCO dealt with the liability of individual shareholders and corporate officers, its test has been extended to parent corporations as well. See, e.g., Bunker Hill, 635 F. Supp. at 672 (noting that the "test... may properly be employed to determine when a parent corporation becomes an owner or operator with respect to a subsidiary's facilities"). 89. 42 U.S.C. ~ 9601(20)(A) (1988). 90. See H.R. REP. No. 172, 96th Cong., 2d Sess., pt. 1, at 36, reprinted in A LEGISLATIVE HISTORY, supra note 8, vol. II, at 546 (exemption was drafted to protect financial institutions that "hold title either in order to secure a loan or in connection with a lease financing agreement under the appropriate banking laws, rules, or regulations"). 91. See, e.g., New York v. Shore Realty Corp., 759 F.2d 1032, 1052 (2d Cir. 1985) ("The use of this exception implies that an owning stockholder who manages the corporation... is liable under CERCLA as an 'owner or operator."'); United States v. Nicolet, Inc., 712 F. Supp. 1193, 1203 (E.D. Pa. 1989) ("[A] corporation which holds stock in another corporation (e.g., a subsidiary) and actively participates in its management can be held liable for cleanup costs incurred as a result of that corporation's disposal."); Idaho v. Bunker Hill Co., 635 F. Supp. 665, 671-72 (D. Id. 1986); NEPACCO I, 579 F. Supp. at 848 ("The statute literally reads that a person who owns interest [sic] in a facility and is actively participating in its management can be held

Bifurcation Under CERCLA Page 62 liable for the disposal of hazardous waste."); United States v. Mirabile, 15 Envtl. L. Rep. (Envtl. L. Inst.) 20, 994, 20,995 (E.D. Pa. 1985) ("Courts have generally concluded that the exemption from liability gives rise to an inference that an individual who owns stock in a corporation and who actively participates in its management can be held liable for clean up costs incurred as a result of improper disposal by the corporation."). 92. The EPA recently issued a rule regarding the liability of secured creditors, see 57 Fed. Reg. 18,382 (1992) (codified at 40 C.F.R. ~ 300.1100(a)), that suggests that mere capacity to control may be insufficient to support parent liability. Although the EPA rule does not address parent corporation liability, it does address an analogous issue: the liability of secured creditors as owners or operators under CERCLA when property in which they hold a security interest is contaminated with hazardous substances. The EPA's rule is intended to alleviate the uncertainty felt by lending institutions after the Eleventh Circuit held in United States v. Fleet Factors Corp., 901 F.2d 1550 (11th Cir. 1990), cert. denied, 498 U.S. 1046 (1990), that a secured creditor could be held liable under CERCLA if the extent of the creditor's participation in the financial management of the debtor's business indicated that the creditor had the capacity to influence the debtor's treatment of hazardous waste. The rule defines the three key provisions of the secured creditor exemption found in CERCLA ~ 107(20)(A): (1) indicia of ownership; (2) protection of a security interest; and (3) participation in management. With regard to the latter, the rule states that the "mere capacity to influence, or ability to influence, or the unexercised right to control facility operations" does not constitute participation in the management of the facility. 57 Fed. Reg. at 18,383. Rather, participation in management for purposes of ~ 101(20)(A) consists of the exercise of decisionmaking control over the overall day-to-day decisionmaking of the enterprise with respect to either environmental compliance or substantially all of the operational aspects of the enterprise. Id.

Bifurcation Under CERCLA Page 63 It could be argued by analogy that parent corporation liability likewise should be based upon actual exercise of control, as opposed to mere capacity to control. 93. See 57 Fed. Reg. at 18,383 (codified at 40 C.F.R. ~ 300.1100(b)(1)). See generally supra note 93 (discussing the EPA rule). 94. See 42 U.S.C. ~ 9601((20)(A) (1988). 95. See, e.g., Riverside Market Dev. Corp. v. International Bldg. Prods., Inc., 931 F.2d 327, 330 (5th Cir. 1991) (individual's position as a majority stockholder in the corporation that purchased the facility did not make him an owner or operator of the facility because "[t]he property of the corporation is its property, and not that of the stockholders, as owners") (quoting 1 FLETCHER, supra note 44, ~ 31, at 555), cert. denied, 112 S. Ct. 636 (1991); see also HENN & ALEXANDER, supra note 43, ~ 71 (shareholders do not own assets of corporation, although they may exercise limited control over those assets); Heidt, supra note 25, at 174 ("A shareholder of a corporation which owns the facility, is not an 'owner' of the facility -- unless the corporate veil is pierced."); Oswald & Schipani, supra note 12, at 299 ("Shareholders are investors in the corporation engaging in hazardous waste activities; the corporation itself is the 'owner' of the facility."). 96. See generally Oswald, supra note 13, at 631-32. 97. See, e.g., United States v. Jon-T Chems., Inc., 768 F.2d 686, 691 (5th Cir. 1985) ("While limited liability remains the norm in Amercian corporation law, certain equitable exceptions to the doctrine have developed. The most common exception is fraud."), cert. denied, 475 U.S. 1011 (1986); Irwin & Leighton, Inc. v. W.M. Anderson Co., 532 A.2d 983, 987 (Del. Ch. 1987) ("The paradigm instance [for piercing the corporate veil] involves the use of a corporate form to perpetrate a fraud."); 1 FLETCHER, supra note 44, ~ 41, at 603 ("In short, the corporate veil may be pierced upon a showing of improper conduct of that the corporation was either formed or used for some illegal, fraudulent, or unjust purpose.").

Bifurcation Under CERCLA Page 64 98. See 1 FLETCHER, supra note 44, ~ 41.10, at 615 ("The doctrine of alter ego fastens liability on the individual who uses a corporation merely as an instrumentality to conduct his or her own personal business, and such liability arises from fraud or injustice perpetrated not on the corporation but on third persons dealing with the corporation."); see generally PHILLIP BLUMBERG, THE LAW OF CORPORATE GROUPS: SUBSTANTIVE LAW ~ 6.01, at 111 (1987). Generally, factors that would tend to indicate domination include extent of stock ownership, identity of officers and directors, financing of the corporation, responsibility for day-to-day operations, arrangements for payment of salaries and expenses, and origin of the subsidiary's business and assets. See, e.g., Phoenix Canada Oil Co. v. Texaco, Inc., 658 F. Supp. 1061, 1084 (D. Del. 1987), affd in part, rev'd in part, 842 F.2d 1466 (3d Cir.), cert. denied, 488 U.S. 908 (1988). 99. See Zubik v. Zubik, 384 F.2d 267, 273 (3d Cir. 1967) ("[A]ny court must start from the general rule that the corporate entity should be recognized and upheld unless specific, unusual circumstances form an exception."), cert. denied, 390 U.S. 988 (1968); 1 FLETCHER, supra note 44, ~ 41.10, at 614-15 ("The standards of application of alter ego principles are high, and the imposition of liability notwithstanding the corporate shield is to be exercised reluctantly and cautiously."). 100. See, e.g., I FLETCHER, supra note 44, ~ 41.10, at 614-15 ("[T]he imposition of liability notwithstanding the corporate shield is to be exercised reluctantly and cautiously. Some courts speak of disregard of the corporate entity as requiring exceptional circumstances.") (footnotes omitted). 101. Empirical data indicate that the courts restrict the use of piercing doctrine to close corporations and to parent-subsidiary corporations or other affiliated corporate groups. See Robert B. Thompson, Piercing the Corporate Veil: An Empirical Study, 76 CORNELL L. REV. 1036, 1047 (1991). Piercing has apparently never been used in the context of a publicly-held corporation. Id. 102. These factors are:

Bifurcation Under CERCLA Page 65 (a) The parent corporation owns all or most of the capital stock of the subsidiary. (b) The parent and subsidiary corporations have common directors or officers. (c) The parent corporation finances the subsidiary. (d) The parent corporation subscribes to all the capital stock of the subsidiary or otherwise causes its incorporation. (e) The subsidiary has grossly inadequate capital. (f) The parent corporation pays the salaries and other expenses or losses of the subsidiary. (g) The subsidiary has substantially no business except with the parent corporation or no assets except those conveyed to it by the parent corporation. (h) In the papers of the parent corporation or in the statements of its officers, the subsidiary is described as a department or division of the parent corporation, or its business or financial responsibility is referred to as the parent corporation's own. (i) The parent corporation uses the property of the subsidiary as its own. (j) The directors or executives of the subsidiary do not act independently in the interest of the subsidiary but take their orders from the parent corporation in the latter's interest. (k) The formal legal requirements of the subsidiary are not observed. FREDERICK J. POWELL, PARENT AND SUBSIDIARY CORPORATIONS ~ 6, at 9 (1931). Not all of these indicia of control need be present in order to support a finding that piercing is appropriate, provided that enough of them are satisfied as to indicate that the parent does indeed control the subsidiary. 103. See David H. Barber, Piercing the Corporate Veil, 7 WILLAMETTE L. REV. 371, 397 (1981). There are two general versions of piercing doctrine: the instrumentality theory and the alter ego theory. The instrumentality theory is usually articulated as three-prong test that examines: (1) excessive exercise of control by the parent; (2) wrongful or inequitable conduct; and (3) causal relationship to the plaintiffs loss. See POWELL, supra note 102, ~~ 5, 6. The alter ego theory is a two-pronged test that requires: "(I) that there

Bifurcation Under CERCLA Page 66 be such unity of interest and ownership that the separate personalities of the corporation and the individual [shareholders] no longer exist and (2) that, if the acts are treated as those of the corporation alone, an inequitable result will follow." Autonotriz del Golfode California v. Resnick, 306 P.2d 1, 3 (Cal. 1957). The doctrines are, in effect, the same. See BLUMBERG, supra note 98, ~ 6.01, at 111; 1 FLETCHER, supra note 44, ~ 43.10, at 759; Barber, supra note 103, at 397. 104. See STEPHEN B. PRESSER, PIERCING THE CORPORATE VEIL ~ 1.03[4], at p. 1-28 n.47 (1992) ("Powell's specific tests are often applied without even attributing them to their author, so common and accepted have they become."). 105. See, e.g., Baker v. Raymond Int'l, Inc., 656 F.2d 173, 180 (5th Cir. 1981) ("Ownership of a controlling interest in a corporation entitles the controlling shareholder to exercise the normal incidents of stock ownership... without forfeiting the protection of limited liability."), cert. denied, 456 U.S. 983 (1982); Wehner v. Syntex Agribusiness, Inc., 616 F. Supp. 27, 29-30 (D.C. Mo. 1985) ("'[T]he fact that the parent may own all of the stock of the subsidiary and even maintain control incident to stock ownership does not justify ignoring the separateness of the two corporations."') (quoting Escude Cruz v. Ortho Pharmacuetical Corp., 619 F.2d 902, 905 (1st Cir. 1980)); see also supra notes 46 -49 (discussing limited liability of parent corporations). 106. See Irwin & Leighton, Inc. v. W.M. Anderson Co., 532 A.2d 983, 987 (Del. Ch. 1987) ("The protection offered by the corporate form, however, is not absolute; equity has long acted to extend a corporate liability to those in control of the corporation in appropriate circumstances."); 1 FLETCHER, supra note 44, ~ 141; HENN & ALEXANDER, supra note 43, ~ 146, at 250; LEWIS D. SOLOMON, DONALD E. SCHWARTZ & JEFFREY D. BAUMAN, CORPORATIONS, LAW AND POLICY 241 (3d ed. 1986) ("Where justice requires, courts occasionally disregard the corporate entity and allow the plaintiffs to reach the assets of the shareholders. This result is metaphorically described as 'piercing the corporate veil."').

Bifurcation Under CERCLA Page 67 107. Jonathan M. Landers, A Unified Approach to Parent, Subsidiary and Affiliate Questions in Bankruptcy, 42 U. CHI. L. REV. 589, 620 (1975). 108. Easterbrook & Fischel, supra note 4, at 89. The authors went on to state: "There is a consensus that the whole area of limited liability, and conversely of piercing the corporate veil, is among the most confusing in corporate law." Id. 109. See, e.g., Wehner v. Syntex Agribusiness, Inc., 616 F. Supp. 27 (D.C. Mo. 1985); United States v. Bliss, 108 F.R.D. 127 (E.D. Mo. 1985); cf., United States v. Arkwright, Inc., 690 F. Supp. 1133 (D.N.H. 1988) (Clean Air Act case). 110. The most notable exception is Joslyn Mfg. Co. v. T.L. James & Co., 893 F.2d 80 (5th Cir. 1990), cert. denied, 498 U.S. 1108 (1991), in which the court determined the facts did not support a piercing of the corporat veil. See generally infra notes 122-39 and accompanying text (discussing Joslyn). See also Jacksonville Elec. Auth. v. Eppinger & Russell Co., 776 F. Supp. 1542 (M.D. Fla. 1991) (discussed infra notes 252-67 and accompanying text); John Boyd Co. v. Boston Gas Co., 775 F. Supp. 435 (D. Mass. 1991) (discussed infra notes 268-79 and accompanying text); Trinity Indus., Inc. v. Dixie Carrier, Inc., 23 Envtl. L. Rep. (Envtl. L. Inst.) (E.D. La. 1992). In addition, even in cases where piercing is raised, it is often only one of several grounds of liability asserted against the parent. See, e.g., CPC International, Inc. v. Aerojet-General Corp., 777 F. Supp. 549, 572 (W.D. Mich. 1991); United States v. Nicolet, Inc., 712 F. Supp. 1193, 1196-97 (E.D. Pa. 1989); City of New York v. Exxon Corp., 112 B.R. 540, 546 (S.D.N.Y. 1990). 111. See United States v. Kimbell Foods, Inc., 440 U.S. 715, 726 (1979). 112. See Anderson v. Abbott, 321 U.S. 349, 364-65 (1944). 113. See Kimbell Foods, 440 U.S. at 728-29. 114. Texas Indus., Inc. v. Radcliff Materials, Inc., 451 U.S. 630, 640 (1980). 115. See, e.g., In re Acushnet River & New Bedford Harbor, 675 F. Supp. 22, 31 (D. Mass. 1987), where the court stated:

Bifurcation Under CERCLA Page 68 One can hardly imagine a federal program more demanding of national uniformity than environmental protection. Congress did not intend that the ability of the executive to fund the clean up of hazardous waste sites should depend upon the attitudes of the several states toward parentsubsidiary liability in general, or CERCLA in particular. The need for a uniform federal rule is especially great for questions of piercing the corporate veil, since liability under the statute must not depend upon the particular state in which a defendant happens to reside. 116. In Acushnet, id. at 33 (citations omitted), the court listed the following factors, "in approximate descending order of importance," to be considered in determining whether a subsidiary's veil should be pierced: (1) inadequate capitalization in light of the purposes for which the corporation was organized; (2) extensive or pervasive control by the shareholder or shareholders, (3) intermingling of the corporation's properties or accounts with those its owner; (4) failure to observe corporate formalities and separatenenss, (5) siphoning of funds from the corporation, (6) absence of corporate records, and (7) nonfunctioning of officers or directors. In short, the Acushnet court applied Powell's laundry list. See supra note. In United States v. Nicolet, Inc., 712 F. Supp. 1193, 1202 (E.D. Pa. 1989), on the other hand, the court articulated a standard based on parental control of the subsidiary, coupled with "a substantial financial or ownership interest." Where a subsidiary is or was at the relevant time a member of one of the classes of persons potentially liable under CERCLA; and the parent had a substantial financial or ownership interst in the subsidiary; and the parent corporation controls or at the relevant time controlled the management and operations of the subsidiary, the parent's separate corporate existence may be disregarded. 117. See, e.g., CPC Int'l, Inc. v. Aerojet-General Corp., 777 F. Supp. 549, 574 (W.D. Mich. 1991). There have been more cases applying state law in the individual liability context. See, e.g., New York v. Shore Realty Corp., 759 F.2d 1032 (2d Cir. 1985); United States v. Carolawn Corp., 14 Envtl. L. Rep. (Envtl. L. Inst.) 20,698 (D.S.C. 1984); United States v. Mottolo, 629 F. Supp. 56 (D.N.H. 1984); cf. Anspec Co. v. Johnson Controls, Inc., 922 F.2d 1240, 1248 (6th Cir. 1991) (Kennedy, J., concurring) (applying state common law to successor corporation liability issues). 118. 893 F.2d 80 (5th Cir. 1990), cert. denied, 498 U.S. 1108 (1991).

Bifurcation Under CERCLA Page 69 119. United States v. Jon-T Chems., Inc., 768 F.2d 686, 690 (5th Cir. 1985) (cited in Joslyn, 893 F.2d at 83), cert. denied, 475 U.S. 1014 (1986). 120. 675 F. Supp. at 33. See generally Note, Piercing the Corporate Veil Under CERCLA: Toward a Uniform Federal Rule of Decision, 22 PAC. L.J. 854 (1991); Note, To Pierce or Not to Pierce? When is the Question: Developing a Federal Rule of Decisionfor Piercing the Corporate Veil Under CERCLA, 68 WASH. U.L.Q. 733 (1990). 121. See supra note 115-17 and accompanying text. 122. 893 F.2d 80 (5th Cir. 1990), cert. denied, 498 U.S. 1108 (1991). See also United States v. Nicolet, Inc., 712 F. Supp. 1193, 1202 (E.D. Pa. 1989); United States v. KayserRoth Corp., 724 F. Supp. 15, 20 (D.R.I. 1989), affd, 910 F.2d 24 (1st Cir. 1990), cert. denied, 498 U.S. 1084 (1991). 123. 893 F.2d at 81. James Company owned all of the non-voting preferred stock and 60% of the common stock of Lincoln. Because the other two stockholders endorsed their shares over to James Company as security for their unpaid capital subscription, James Company was able to control 100 percent of Lincoln's stock. Id. 124. Id. at 82. 125. Other parties were also ordered to take part in the cleanup action. 696 F. Supp. at 224. 126. 893 F.2d at 82. 127. It seems likely that the district court was led astray by the inaccurate phrasing of the issue presented to it: "whether CERCLA imposes direct liability upon a parent corporation or requires a claimant to pierce the corporate veil before liability may attach." 696 F. Supp. at 224. 128. Id. at 225. 129. Id. at 226. The court stated: "this court holds that the corporate form, including limited liability for shareholders, is a doctrine firmly entrenched in American jurisprudence that may not be disregarded absent a specific congressional directive." Id.

Bifurcation Under CERCLA Page 70 130. Id. (footnote omitted). 131. Although the district court's opinion was based solely on the theory of indirect liability, and in fact, the court explicitly denied that direct liability could attach to parent corporations, see supra note 130 and accompanying text, the court contradicted itself in the final footnote of its opinion, 696 F. Supp. at 232 n.20. The court noted that under common law, an officer could be held liable for his or her personal participation in a wrongful corporate act. Id. (The court incorrectly characterized this rule as an exception to the common law rule of limited liability, however. Id. In point of fact, limited liability is a common law corporate notion that protects the owners of a corporation. Officer and director, on the other hand, arises from traditional notions of agency and tort law. See RESTATEMENT (SECOND) OF AGENCY ~ 343 (1958); JOSEPH W. BISHOP, JR., THE LAW OF CORPORATE OFFICERS & DIRECTORS: INDEMNIFICATION AND INSURANCE ~ 3.13 (1982); 3A FLETCHER, supra note 44, ~ 1135 (perm. ed. rev. vol. 1986)). The district court thus concluded: "If T.L. James & Company and its officers and directors had been actively involved in the day-to-day operations of Lincoln, including the disposal of hazardous waste, then, arguably, liability would attach." 696 F. Supp. at 233 n.20. This is, of course, a statement of the direct liability theory that the court had previously denounced. 132. The Fifth Circuit had set forth a test for determining whether a subsidiary's veil should be pierced in a non-environmental case called United States v. Jon-T Chemicals, Inc., 768 F.2d 686, 691-92 (5th Cir. 1985), cert. denied, 475 U.S. 1014 (1986). These factors include whether: (1) the parent and subsidiary have common stock ownership; (2) the parent and the subsidiary have common directors or officers; (3) the parent and the subsidiary have common business departments; (4) the parent and the subsidiary file consolidated financial statements and tax returns;

Bifurcation Under CERCLA Page 71 (5) the parent finances the subsidiary; (6) the parent caused the incorporation of the subsidiary; (7) the subsidiary operates with grossly inadequate capital; (8) the parent pays the salaries and other expenses of the subsidiary; (9) the subsidiary receives no business except that given to it by the parent; (10) the parent uses the subsidiary's property as its own; (11) the daily operations of the two corporations are not kept separate; and (12) the subsidiary does not observe the basic corporation formalities, such as keeping separate books and records and holding shareholder and board meetings. Joslyn, 696 F. Supp. at 227 (citing Jon-T Chemicals, 768 F.2d at 691-92). 133. Id. at 231. Joslyn set forth a number of factors that it claimed supported a piercing of Lincoln's corporate veil, such as the existence of common stock ownership and common directors, the exercise of control over the subsidiary's finances by two of the parent's directors, the provision of substantial loans to the subsidiary by the parent (all of which were repaid), the hiring and firing of executive officers of the subsidiary by the parent, and use of subsidiary office space by individuals who served as officers of both corporations. Id. at 230-31. The court refused to find that these factors were sufficient to support a piercing of the corporate veil. 134. Id. at 231. 135. 893 F.2d 80 (5th Cir. 1990), cert. denied, 498 U.S. 1108 (1991). 136. Id. at 83. 137. Id. at 82-83. 138. Id. at 82. 139. Id. The Fifth Circuit reaffirmed the piercing test used by the district court, id. at 83 (approving of the district court's use of the Jon-T Chemicals factors cited in supra note

Bifurcation Under CERCLA Page 72 132), and agreed with its conclusion that the facts at hand did not support a piercing of the subsidiary's veil. 140. Id. at 83. 141. 675 F. Supp. 22 (D. Mass. 1987). 142. 635 F. Supp. at 31-32. 143. Id. at 32. 144. Id. at 34 ("'There is nothing fraudulent or against public policy in limiting one's liability by the appropriate use of corporate insulation."') (citations omitted). Because none of the facts indicated that the subsidiary was in any way a shell corporation, the court granted the parent's motion to dismiss for lack of personal jurisdiction. Id. at 35. 145. See Acushnet, 675 F. Supp. at 32: Under traditional principles, a corporation which wants to put a waste site or past generation site to productive use can do so by creating a well capitalized, non-fraudulent, separate corporate subsidiary. The ability to work through the subsidiary justifies the initial investment, which will delimit the extent of the risk. Under the sovereigns' proposed rule, a corporation which wanted to reclaim and make productive a waste site could not do so without risking all its corporate assets if it appeared to be more than passively interested in the performance of its subsidiary. Patently, the sovereigns' rule would discourage investors, and reduce the number of solvent parties from which the sovereign will be able to seek clean up costs and damages. 146. See United States v. Jon-T Chems., Inc., 768 F.2d 686, 690 (5th Cir. 1985), cert. denied, 475 U.S. 1014 (1986) (citation omitted): Under the doctrine of limited liability, the owner of a corporation is not liable for the corporation's debts. Creditors of the corporation have recourse only against the corporation itself, not against its parent company or shareholders. It is on this assumption that 'large undertakings are rested, vast enterprises are launched, and huge sums of capital attracted.' 147. See generally Oswald, supra note 13, at 625. 148. Commentators have suggested that the broad liability provisions of CERCLA have already distorted investment patterns. See, e.g., Perry E. Wallace, Jr., Liability of Corporations and Corporate Officers, Directors, and Shareholders Under Superfund: Should Corporate and Agency Law Concepts Apply?, 14 J. CORP. L. 839, 842 (1989);

Bifurcation Under CERCLA Page 73 Cynthia S. Korhonen & Mark W. Smith, Note, CERCLA Defendants: The Problem of Expanding Liability and Diminishing Defenses, 31 WASH. U.J. URB. & CONTEMP. L. 289, 315-16 (1987); Rallison, supra note 2, at 622-23. 149. See, e.g., United States v. Maryland Bank & Trust Co., 632 F. Supp. 573, 578 (D. Md. 1986) ("The structure of [CERCLA] Section 107(a), like so much of this hastily patched together compromise Act, is not a model of statutory clarity."); United States v. Mottolo, 605 F. Supp. 898, 902 (D.N.H. 1985) ("CERCLA has acquired a well-deserved notoriety for vaguely-drafted provisions and an indefinite, if not contradictory, legislative history."). 150. For a comprehensive review of the legislative history of CERCLA, see Grad, supra note 29. 151. CERCLA's legislative history is replete with examples of political compromise. See generally supra note 13. 152. See supra note 13 (discussing CERCLA's legislative history). Thus, for example, the courts have fashioned the strict liability standard for CERCLA, even though the standard itself is not expressed in the statute. See supra note 27. Likewise, the courts have formulated various rules for determining the liability of corporate individuals under CERCLA, see Oswald & Schipani, supra note 12, at 275-94, even though, again, the statute does not address this issue. 153. 675 F. Supp. at 33-34. 154. Id. 155. See Colorado v. Idarado Mining Co., 18 Envtl. L. Rep. (Envtl. L. Inst.) 20,578, 20,579 (D. Colo. 1987), rev'd on other grounds, 916 F.2d 1468 (10th Cir. 1990), cert. denied, 111 S. Ct. 1584 (1991). 156. See infra notes 229-91 and accompanying text. 157. See Acushnet, 675 F. Supp. at 33 (citations omitted):

Bifurcation Under CERCLA Page 74 These factors include, in approximate descending order of importance, (1) inadequate capitalization in light of the purposes for which the corporation was organized, (2) extensive or pervasive control by the shareholder or shareholders, (3) intermingling of the corporation's properties or accounts with those of its owner, (4) failure to observe corporate formalities and separateness, (5) siphoning of funds from the corporation, (6) absence of corporate records, and (7) nonfunctioning officers or directors. 158. Id. at 33. 159. See supra note 103 and accompanying text (regarding instrumentality doctrine). 160. 675 F. Supp. at 33 (citations omitted). 161. Id. at 35. 162. Id. 163. See, e.g., HENN & ALEXANDER, supra note 43, ~ 146, at 347 ("[A] shareholder, whether a natural person or a corporation, may be liable on the ground that such shareholder's activity resulted in the liability."). 164. See, e.g., Anglo Eastern Bulkships, Ltd. v. Ameron, 556 F. Supp. 1198, 1202 (S.D.N.Y. 1982); Far West Financial Corp. v. D & S Co., Inc., 760 P.2d 399, 410 (Cal. 1988); Blaustein v. Pan American Petroleum & Transp. Co., 21 N.Y.S.2d 651, 751-52 (1940). 165. Under the Restatement (Second) of Torts, the "good Samaritan" doctrine imposes liability where an actor failes to exercise reasonable care when it undertakes the performance of a duty owed to a third party. See RESTATEMENT (SECOND) OF TORTS ~ 324A (1986). The doctrine has been used several times in recent years by employees of subsidiaries seeking to hold the parents liable for failing to provide a safe workplace. See generally Andrew J. Natale, Note, Expansion of Parent Corporate Shareholder Liability Through the Good Samaritan Doctrine, 57 U. CIN. L. REV. 717 (1988). 166. See generally BLUMBERG, supra note 43, ~ 14.03. The parent can be either the principal or the agent, see RESTATEMENT (SECOND) OF AGENCY ~ 14M comment a (1958), and the agency itself can be either express or implied. See generally Jackam v. Hospital Corp. of America Mideast, Ltd., 800 F.2d 1577 (11th Cir. 1986); Japan Petroleum Co.

Bifurcation Under CERCLA Page 75 (Nigeria) Ltd. v. Ashland Oil, Inc., 456 F. Supp. 831 (D. Del. 1978). In addition, apparent authority or agency can arise through estoppel. See, e.g., RESTATEMENT (SECOND) OF AGENCY ~ 8B (1958). "[B]ecause common law agency requires a consensual understanding between the parties," it is rarely found in parent-subsidiary relationships. BLUMBERG, supra note 43, ~ 6.06.1, at 126 (citing Kingston Dry Dock Co. v. Lake Champlain Transp. Co., 31 F.2d 265 (2d Cir. 1929); Bendix Corp. v. Adams, 610 P.2d 24, 32-33 (Alaska 1980)). It appears that the only CERCLA case to date decided under an agency theory is FMC Corp. v. Northern Pump Co., 668 F. Supp. 1285 (D. Minn. 1987). The court there found that, as a matter of law, the facts did not evidence "the degree of control necessary to establish an agency relationship." Id. at 1293. Analysis of agency theory is made difficult by the propensity of courts to use the term "agency" incorrectly in piercing cases, as a metaphor for common identity between the two affiliated corporations. Used in this context, the term is really just another way of expressing "alter ego" or "instrumentality" theory. See PHILLIP BLUMBERG, PROCEDURAL PROBLEMS IN THE LAW OF PARENT AND SUBSIDIARY CORPORATIONS ~ 1.02 (1983 & Supp. 1992). The term is not appropriately used in the context of piercing. See RESTATEMENT (SECOND) OF AGENCY ~ 14M, Reporter's Note, at 67-72 (1957) ("It is useful to distinguish situations in which liability is imposed on a parent because of the existence of the agency relation in our common law understanding of that relation, from cases in which the corporate veil of the subsidiary is pierced for other reasons of policy. Unfortunately, however, the courts have not always observed the distinction between these two separate bases for parent's liability."); see also BLUMBERG, supra note 43, ~ 6.06.2 (discussing the inappropriate use of "agency" in piercing cases); Cathy S. Krendl & James R. Krendl, Piercing the Corporate Veil: Focusing the Inquiry, 55 DENVER L.J. 1, 3 n.9 (1978).

Bifurcation Under CERCLA Page 76 167. The parent's involvement may arise through its own participation or acquiescence in the subsidiary's fraud, see, e.g., Stephens v. National Distillers & Chem. Corp., 811 F. Supp. 937, 955 (S.D.N.Y. 1993), or the parent's own affirmative fraud or misrepresentation. See, e.g., Daher v. G.D. Searle & Co., 695 F. Supp. 436, 437 (D. Minn. 1988). These instances of fraud are distinct from the situation in which a parent creates a subsidiary in order to perpetrate fraud. This latter situation would give support a piercing of the corporate veil, and the imposition of indirect liability upon the parent. See, e.g., Irwin & Leighton, Inc. v. W.M. Anderson Co., 532 A.2d 983, 987 (Del. Ch. 1987) ("The paradigm instance [for piercing the corporate veil] involves the use of a corporate form to perpetrate a fraud."); 1 FLETCHER, supra note 44, ~ 43. 168. The misrepresentation may relate to the identity or the economic condition of the subsidiary. See generally BLUMBERG, supra note 43, ~ 14.01.1. 169. See supra notes 97-101 and accompanying text. 170. See, e.g., McMahon & Moertl, supra note 2, at 29; Mitchell, supra note 81, at 70 -71. 171. See infra Parts TV.A (discussing the capacity to control theory) and IV.B (discussing the actual exercise of control theory). Corporate officers have been held directly liable under CERCLA as well, under theories somewhat analogous to the direct liability theories discussed here. See generally Oswald & Schipani, supra note 12, at 275-97 (discussing cases). 172. See generally supra notes 97-101, 163-70 and accompanying text (discussing traditional theories of parent corporation liability). 173. See supra note 48 (definition of a subsidiary). 174. 635 F. Supp. 665 (D. Idaho 1986). The court provides very few facts regarding the CERCLA violation in this case. Gulf Resources & Chemical Corporation ("Gulf') was the parent corporation of Bunker Hill Company. Gulf and Bunker Hill merged in 1968, and Bunker Hill became a wholly-owned subsidiary. Id. at 670. Bunker Hill was sold to

Bifurcation Under CERCLA Page 77 Bunker Limited Partnership in 1982. Id. at 676. The opinion says nothing about the nature of either the facility or the activities leading to the CERCLA violation. 175. The capacity to control test was articulated in a number of cases involving the individual liability of officers prior to appearing in Bunker Hill. See, e.g., United States v. Carolawn Co., 14 Envtl. L. Rep. (Envtl. L. Inst.) 20, 699, 20,700 (D.S.C. 1984) (noting that "to the extent that an individual has control or authority over the activities of a facility from which hazardous substances are released or participates in the management of such a facility, he may be held liable for response costs incurred at the facility notwithstanding the corporate character of the business"). See generally Dennis, supra note 81, at 1378-1381 (discussing cases). The only appellate opinion among these cases was New York v. Shore Realty Corp., 759 F.2d 1032 (2d Cir. 1985). Donald LeoGrande formed Shore Realty Corp. for the purpose of purchasing a site on which he knew hazardous wastes were stored. Id. at 1037. It took Shore Realty three months in which to evict the tenants, during which time 90,000 additional gallons of hazardous substances were disposed of on the site. Shore Realty took no steps to prevent the additional disposal, nor did it attempt to deal with the hundreds of thousands of gallons of contaminants already on the site. In holding LeoGrande, an officer and shareholder of Shore Realty, personally liable, the court first adopted the secured creditor exemption argument articulated in NEPACCO I, 579 F. Supp. at 848 (discussed supra note 89-96 and accompanying text). See 759 F.2d at 1052. Second, the Shore Realty court held that LeoGrande's position as the person in charge of the facility was sufficient to render him liable as an "operator." Id. Although the Shore Realty court spoke in terms of capacity to control, a careful reading of the facts of this case indicates that LeoGrande exercised actual control over the events leading to the CERCLA violations. Specifically, the court noted that LeoGrande made, directed, and controlled all corporate decisions and actions. Id. at 1052. See generally Oswald & Schipani, supra note 12, at 284-86 (discussing cases).

Bifurcation Under CERCLA Page 78 176. See infra notes 238-42 and accompanying text. The Bunker Hill court's emphasis on the parent corporation's involvement in the subsidiary's environmental practices, 635 F. Supp. at 672, and its discussion of the parent's capacity to control facility activities, id. at 670 (quoted in text supra note 181), indicates that the court correctly recognized that control over the running of thefacility, not general control over the subsidiary, is required under CERCLA. The court's reliance upon capacity to control, rather than actual exercise of control, was incorrect, however. 177. Id. at 671-72. 178. Id. (citing NEPACCO I, 579 F. Supp. at 848-49). 179. Id. at 672. The Bunker Hill court had first considered the issue of personal jurisdiction, noting that the court could not acquire long-arm jurisdiction over a parent based solely on the local activities of the subsidiary where "the subsidiary is carrying on its own business and the corporate formalities have been observed." Id. at 670. Many of the same facts that lead the court to hold the parent directly liable under CERCLA, see supra notes 179-80 and accompanying text, supported its exercise of personal jurisdiction over the parent. Id. Additional facts considered by the court in deciding the jurisdiction issue were the consolidation of the parent's and subsidiary's tax returns and the parent's ability to overrule any transactions or management decisions made by the subsidiary. Id. 180. Id. at 672. 181. Id. The parent was held liable as a past owner or operator under CERCLA section 107(a)(2), 42 U.S.C. ~ 9607(a)(2). 182. For example, the parent's demand for weekly reports of daily operations and its restrictions on spending on pollution matters by the subsidiary, 635 F. Supp. at 672, both suggest the subsidiary was not permitted to act with the usual amount of corporate independence. 183. Id.

Bifurcation Under CERCLA Page 79 184. Id. Although it is clear from the opinion that the court was evaluating the parent's liability under a direct liability, the opinion also reveals the court's confusion regarding the differences between direct and indirect liability. In finding the parent liable, the court noted that "[t]o hold otherwise... would allow the corporate veil to frustrate congressional purpose." Id. The "corporate veil," of course, is a reference to piercing doctrine, which is an indirect liability theory. 185. Fortunately, relatively few courts have adopted Bunker Hill's test. See, e.g., United States v. Nicolet, Inc., 712 F. Supp. 1193 (E.D. Pa. 1989) (adopting the Bunker Hill theory as one of three theories of direct liability in denying the parent's motion for summary judgment); Vermont v. Staco, Inc., 684 F. Supp. 822 (D. Vt. 1988) (finding parent corporation liable under RCRA and CERCLA without any demonstration of exercise of control), vacated in part, No. 86-190 (D. Vt. Apr. 20, 1989) (1989 U.S. Dist. LEXIS 17341). Most courts have applied the actual exercise of control test discussed below. It would appear that the capacity to control test is on the wane and that it is very weak today, if it survives at all. 186. 18 Envtl. L. Rep. (Envtl. L. Inst.) 20,578 (D. Colo. 1987), rev'd on other grounds, 916 F.2d 1468 (10th Cir. 1990), cert. denied, 111 S. Ct. 1584 (1991). Idarado Mining Company admitted that it was the owner and operator of the Idarado Mine, which was the site of cleanup action under CERCLA. Id. at 20,578. Newmont Mining Company, the parent and major shareholder of Idarado Mining Co., disputed its liability as an owner and operator of the Mine, id., as did Newmont Services Limited, a subsidiary of Newmont Mining Co. formed to provide management and other services to Newmont Mining Co.'s subsidiaries, including Idarado Mining Co. Id. at 20,579. 187. Id. at 20,578 (quoting NEPACCO 1, 579 F. Supp. at 848). 188. 628 F. Supp. 391 (W.D. Mo. 1985). 189. 18 Envtl. L. Rep. (Envtl. L. Inst.) at 20,578.

Bifurcation Under CERCLA Page 80 190. The individual held liable was the president and controlling shareholder of a chemical company that violated CERCLA. The individual was a trained engineer, personally designed the facility and oversaw its construction, designed several of its waste treatment processes, and was actively involved in its day-to-day management. He researched treatment processes and marketed the company, administered the corporation, executed contracts on its behalf, and hired and supervised employees. 628 F. Supp. at 420. 191. The Conservation Chemical court noted that "corporate officials who actively participate in the management of a disposal facility can be held personally liable" under CERCLA, 628 F. Supp. at 419, and that the defendant's "high degree of personal involvement in the operation and the decision-making process" rendered him liable. Id. at 420. The individual's status as a stockholder was essentially irrelevant to the determination of liability; his activities as an officer were key. See Oswald & Schipani, supra note 12, at 281. 192. See 3A FLETCHER, supra note 44, ~ 1135: It is thoroughly well-settled that a person is personally liable for all torts committed by him... notwithstanding he may have acted as the agent or under directions of another. And this is true to the full extent as to torts committed by the officers or agents of a corporation in the management of its affairs. Id. at 267; see also Donsco, Inc. v. Casper Corp., 587 F.2d 602, 606 (3d Cir. 1978); Escude Cruz v. Ortho Pharmaceutical Corp., 619 F.2d 902, 907 (1st Cir. 1980); BALLANTINE, supra note 45, ~ 112, at 275 (rev. ed. 1946). 193. See supra note 190. 194. The opinion indicates that Newmont Mining owned 80.1% of Idarado's stock since 1962, that "the vast majority" of Idarado's officers since 1939 were also officers or directors of Newmont Mining, that Idarado paid a fee to Newmont Mining for certain legal, management, and technical services and advice, that Newmont Mining controlled

Bifurcation Under CERCLA Page 81 Idarado's recruitment and reimbursement procedures, and that Newmont Mining entered into contracts on Idarado's behalf. Id. at 20,578-79. 195. Id. at 20,578. 196. Id. 197. Id. at 20579. 198. See supra note 191. 199. See infra notes 281-92 and accompanying text. 200. The analysis of the court in City of New York v. Exxon Corp., 112 B.R. 540 (S.D.N.Y. 1990), is typical: [W]hile the parent corporation's capacity to discover in a timely fashion the release or threat of release of hazardous substances, the parent corporation's power to direct mechanisms causing the release, and the parent corporation's capacity to prevent and abate damage, are certainly relevant and material factors to consider in evaluating a parent corporation's potential liability, the corporation must exercise its power or capacity to control its subsidiary in order to be held liable under Section 107(a). Id. at 548 n.9 (citations omitted). See also Mobay Corp. v. Allied-Signal, Inc., 761 F. Supp. 345, 354 (D.N.J. 1991) (requiring an actual exercise of control by the parent corporation); United States v. Nicolet, Inc., 712 F. Supp. 1193, 1203 (E.D.Pa. 1989) (same). 201. See supra notes 122-39 and accompanying text (discussing Joslyn). 202. 910 F.2d 24 (1st Cir. 1990), cert. denied, 498 U.S. 1084 (1991). 203. 724 F. Supp. 15 (D.R.I. 1989). 204. Id. at 17-18. 205. Id. at 22-23. 206. Id. at 22. 207. Id. at 23. 208. Id. at 22.

Bifurcation Under CERCLA Page 82 209. Id. ("The question becomes whether Kayser-Roth exercised control over Stamina Mills management and operations sufficient to find that Kayser-Roth was a de facto operator."). 210. Id. 211. Id. at 21. 212. See supra note 31 (discussing definition of a "facility"). 213. The court listed the following examples of the parent's "practical total control" over the subsidiary's operations: Kayser-Roth exercised pervasive control over Stamina Mills through, among other things: 1) its total monetary control including collection of accounts payable; 2) its restrictions on Stamina Mills' financial budget; 3) its directive that the subsidiary-governmental contact, including environmental matters, be funneled directly through Kayser-Roth; 4) its requirement that Stamina Mills' leasing, buying or selling of real estate first be approved by Kayser-Roth; 5) its policy that Kayser-Roth approve any capital transfer or expenditures greater than $5,0000; and finally, 6) its placement of Kayser-Roth personnel in almost all Stamina Mills' director and officer positions, as a means of totally ensuring that KayserRoth corporate policy was implemented and precisely carried out. 724 F. Supp. at 22. 214. The court stated: Kayser-Roth had the power to control the release or threat of release of TCE, had the power to direct the mechanisms causing the release, and had the ultimate ability to prevent and abate damage. Kayser-Roth knew that Stamina Mills employed a scouring system that used TCE; indeed, Kayser-Roth approved the installation of that system after mandating that a cost-benefit study by made by Stamina Mills. Kayser-Roth not only had the capacity to determine the use of TCE but was also able to direct Stamina Mills on how the TCE should have been handled. Id. at 22-23 (footnote omitted). 215. Id. at 23-24. 216. Id. at 24. 217. The court stated: Kayser-Roth's control over environmental matters; its policy of approving all capital expenditures of greater than $5,000; its stranglehold on income and expenses; its practice of placing Kayser-Roth personnel in Stamina Mills' director positions, thereby precluding other Stamina Mills

Bifurcation Under CERCLA Page 83 executives from significant daily decisionmaking; and its overwhelming control over Stamina Mills' financial and operational structure add flesh to the skeletal proposition that Kayser-Roth's corporate existence should be disregarded. Id. (footnote omitted). 218. 910F.2dat28n.ll. 219. Id. at 27 (quoting 724 F. Supp. at 18). 220. Id. (quoting 724 F. Supp. at 22). The First Circuit did not find the parent's exercise of control over the subsidiary's environmental practices to be dispositive: "Although indicia of ability to control decisions about hazardous waste are indicative of the type of control necessary to hold a parent corporation liable as an operator, we do not think the presence of such indicia is essential, assuming there are other indicia of the pervasive control necessary to prove operator status." Id. at 27 n.8. 221. Id. at 27. 222. Id. However, the First Circuit also stated that "indicia of ability to control decisions about hazardous waste are indicative of the type of control necessary to hold a parent corporation liable as an operator," 910 F.2d at 27 n.8, and it quoted the district court's language regarding the parent's "power to control" certain environmental matters, id. at 28 (quoting 724 F. Supp. at 22 (quoted supra note 214)), concluding that "[s]uch control is more than sufficient to be liable as an operator under CERCLA." Id. These statements suggest that the First Circuit too was confusing the power to control and the actual exercise of control tests. In light of the parent's actual exercise of control over its subsidiary's environmental decisions, these statements are probably best viewed as dicta. See Jacksonville Elec. Auth. v. Eppinger & Russell Co., 776 F. Supp. 1542, 1547 n.4 (M.D. Fla. 1991). 223. 777 F. Supp. 549 (W.D. Mich. 1991). 224. Id. at 573. 225. Id. 226. Id.

Bifurcation Under CERCLA Page 84 227. Id. The facts before the CPC International court indicated that: (1) the subsidiary was wholly owned; (2) the parent actively participated in, and at times controlled, the subsidiary's board of directors; (3) the intertwining of management officials enabled the parent to become involved in both major decisionmaking and day-to-day operations of the subsidiary; (4) the parent's officials were actively involved in and controlled various affairs of the subsidiary, including environmental and labor issues; and (5) the parent controlled the subsidiary's budgets and major capital expenditures. Id. at 575. The CPC International court thus held the parent directly liable as an operator, noting that: "When a parent corporation permeates the board, management and decision-making of a wholly owned subsidiary that was disposing of hazardous waste, operator liability directly attaches under CERCLA." Id. 228. 702 F. Supp. 1384 (N.D. Ill. 1988). 229. Id. at 1390. 230. Id. at 1390-91. See also Mobay Corp. v. Allied-Signal, Inc., 761 F. Supp. 345, 354 (D.N.J. 1991); United States v. Nicolet, Inc., 712 F. Supp. 1193, 1203 (E.D. Pa. 1989); United States v. McGraw-Edison Co., 718 F. Supp. 154, 157 (W.D.N.Y. 1989) (holding that a corporation's minority shareholder status (49% ownership) in the corporation operating the facility will not automatically relieve the shareholder of liability as an "owner" or "operator"). 231. Dedham Water Co. v. Cumberland Farms Dairy, Inc., 805 F.2d 1074, 1081 (1st Cir. 1986). 232. As the United States Supreme Court has stated, a remedial purpose alone is not enough to "add a gloss to the operative language of [a] statute quite different from its commonly accepted meaning." Ernst & Ernst v. Hochfelder, 425 U.S. 185, 198-99 (1976). 233. See Colorado v. Idarado Minging Co., 18 Envtl. L. Rep. (Envtl. L. Inst.) 20578, 20579 (D. Colo. 1987) ("As a practical matter, if mere interposition of a separate

Bifurcation Under CERCLA Page 85 corporate entity could insulate against CERCLA liability, it would be far to easy to evade the statute."). 234. See, e.g., Nurad, Inv. v. William Hooper & Sons, Co., 996 F.2d 837, 842 (4th Cir. 1992) (noting that authority to control, not actual control, "is the definition of the word 'operator' that most courts have adopted, and it is one which properly declines to absolve from CERCLA liability a party who possessed the authority to abate the damage caused by the disposal of hazardous substances but who declined to actually exercise that authority"). 235. See, e.g., Rockwell International Corp. v. IU International Corp., 702 F. Supp. 1384, 1390 (N.D. SI. 1988); CPC International, Inc. v. Aerojet-General Corp., 777 F. Supp. 549, 573 (W.D. Mich. 1991) ("[T]he liability of a parent corporation cannot attach simply because a parent has had involvement with its subisidary in a manner merely consistent with their investment relationship. Rather, a parent must have actually operated the business of its subsidiary."); City of New York v. Exxon Corp., 112 B.R. 540, 548 n.9 (S.D.N.Y. 1990) ("the corporation must exercise its power or capacity to control its subsididary in order to be held liable" as an owner or operator"); cf. United States v. Consolidated Rail Corp., 729 F. Supp. 1461, 1468 (D. Del. 1989) ("the statute requires that a person be actively participating in the management of the facility to be held liable for the disposal of hazardous wastes. The 'mere ability to exercise control as a result of the financial relationship of the parties is insufficient for liability to attach."' (quoting United States v. Mirabile, 10 Chem & Rad. Waste Lit. Rep. 688, 670-71 (E.D. Pa. 1988). As one court summed up the state of the law, "the weight of authority strongly favors application of the... exercise of control standard." Levin Metals, Corp. v. ParrRichmond Terminal Co., 781 F. Supp. 1454, 1456 (N.D. Cal. 1991). 236. See, e.g., supra note 222 (discussing confusion of First Circuit in Kayser-Roth). 237. Perhaps the greater danger is that the courts tend to refer to "pervasive control" in evaluating both direct and indirect liability, creating the possibility that the distinctions

Bifurcation Under CERCLA Page 86 between these two bases of liability will become even more blurred than they already are. See, e.g., City of New York v. Exxon Corp., 112 B.R. 540, 550, 553 (S.D.N.Y. 1990). See generally infra note 284 and accompanying text. 238. As one court noted, the cases indicate that the courts will impose liability upon an officer or shareholder if that person: "(1) actually participated in the facility's operations; or (2) actually exercised control over, or was otherwise intimately involved in the operations of the corporation immediately responsible for the operation of the facility." CBS, Inc. v. Henkin, 803 F. Supp. 1426, 1434 (N.D. Ind. 1992) (citing Levin Metals Corp. v. Parr-Richmond Terminal Co., 781 F. Supp. 1454, 1456-57 (N.D. Cal. 1991); Ronald Aronovsky & Lynn Fuller, Liability of Parent Corporations for Hazardous Substance Releases Under CERCLA, 24 U.S.F. L. REV. 421, 442 (1990). The second test is the one that is troubling, for it focuses on the wrong relationship. The NEPACCO II court correctly emphasized that operation of the facility, not the subsidiary, is key, see 810 F.2d at 742-43, but, unfortunately, its distinction was ignored by later courts. See supra note 78 and accompanying text. 239. See, e.g., CPC International, Inc. v. Aerojet-General Corp., 777 F. Supp. 549, 575 (W.D. Mich. 1991) ("When a parent corporation permeates the board, management and decision-making of a wholly owned subsidiary that was disposing of hazardous waste, operator liability directly attaches under CERCLA."). 240. 42 U.S.C. ~ 9601(9) (1988) (quoted in supra note 31). 241. See supra note 48 (discussing definition of a subsidiary). 242. Cf. Nurad, Inc. v. William E. Hooper & Sons Co., 966 F.2d 837 (4th Cir. 1992) ("[A] defendant operates a 'facility' only if it has authority to control the area where the hazardous substances were located."). 243. See supra note 3. 244. Compare Kayser-Roth, 910 F.2d at 26 (because Congress didn't expressly exclude parents from liability, it must have intended that they be held so liable) with Joslyn, 893

Bifurcation Under CERCLA Page 87 F.2d at 82 (because Congress did not expressly provide for parent liability in the statute, it must not have intended that they be held so liable). 245. See, e.g., New York v. Shore Realty Corp., 759 F.2d 1032, 1040 (2d Cir. 1985) (describing CERCLA as "an eleventh hour compromise" with an inadequate legislative history); Chemical Waste Mgmt., Inc. v. Armstrong World Indus., 669 F. Supp. 1285, 1290 n.6 (E.D. Pa. 1987) (noting that "CERCLA's legislative history is sparse and generally uninformative" and that "last-minute additions and deletions to the statute render its legislative history of little practical use"). 246. 910 F.2d 24 (1st Cir. 1990), cert. denied, 498 U.S. 1084 (1991). 247. 893 F.2d 80 (5th Cir. 1990), cert. denied, 498 U.S. 1108 (1991). 248. 893 F.2d at 81. 249. 910 F.2d at 27. The Kayser-Roth court's characterization of Joslyn's holding is misleading. In Joslyn, the Fifth Circuit clearly rejected the notion that parent corporations could be directly liable under CERCLA: Joslyn asks this court to rewrite the language of the Act significantly and hold parents directly liable for their subsidiaries' activities. To do so would dramatically alter traditional concepts of corporation law.... Any bold rewriting of corporation law in this area is best left to Congress. 893 F.2d at 82-83. 250. Although the distinction is an analytically compelling one, it is based upon a narrow reading of Joslyn, and it is almost certainly not a distinction that the Joslyn court either intended or would ratify. The Joslyn court clearly rejected the idea that direct liability was permissible under CERCLA as a matter of analysis and statutory interpretation; its refusal to find direct liability was not based upon the specific facts of the case before it. See, e.g., Joslyn, 893 F.2d at 82-3 (stating that imposing direct liability on parent corporations would drastically alter traditional law and the court declined to do so in the absence of any statutory language indicating that Congress intended such a result).

Bifurcation Under CERCLA Page 88 251. See Jacksonville Elec. Auth. v. Eppinger & Russell Co., 776 F. Supp. 1542 (M.D. Fla. 1991); John Boyd Co. v. Boston Gas Co., 775 F. Supp. 435 (D. Mass. 1991). Earlier courts also hinted at this distinction, but did not address it directly. See, e.g., CPC International, Inc. v. Aerojet-General Corp., 777 F. Supp. 549 (W.D. Mich. 1991). 252. 776 F. Supp. 1542 (M.D. Fla. 1991). See also Bronson Specialties Inc. v. Elmer Houghton, 1993 U.S. Dist. LEXIS 7714 (Apr. 12, 1993) (recognizing owner and operator distinction in context of motion for summary judgment). 253. Id. at 1545. 254. Id. at 1545 (quoting Jon-T Chemicals, 768 F.2d at 691-92; see supra note 132 and accompanying text (providing list and discussing its application in Joslyn)). 255. Id. at 1546. As the Jacksonville Electric court noted, these "are factors to be expected where the parent company owns virtually all of the subsidiary's stock." Id. 256. Id. 257. Id. 258. Id. 259. The Jacksonville Electric court found that the facts before it were substantially different from those present in Kayser-Roth, "where the parent corporation exercised pervasive control over the subsidiary's financial operations and gave final approval of the installation of the system that used the hazardous substance at issue in that case." Id. (citing Kayser-Roth, 910 F.2d at 27). The court emphasized that in the case before it, the parent was a college and the subsidiary was a creosoting business. Id. at 1549. In Kayser-Roth, on the other hand, the parent had been in the business of textile manufacturing for several years before it purchased the subsidiary, which was also a textile manufacturer. Id. (citing Kayser-Roth, 724 F. Supp. at 17-19). Thus, the parentsubsidiary relationships in the two cases were substantially dissimilar. Because the facts before it did not support an inference of pervasive control by the parent corporation, the Jacksonville Electric court granted the parent's motion for summary judgment. Id.

Bifurcation Under CERCLA Page 89 260. See supra notes 141-62 and accompanying text (discussing Acushnet) and infra notes 285-89 (discussing Nicolet). 261. The court cited Kayser-Roth, 910 F.2d 24 (1st Cir. 1990); Riverside Market Dev. Corp. v. International Bldg. Prod., Inc., 931 F.2d 327 (5th Cir.) (addressing operator liability of an individual shareholder), cert. denied, 112 S. Ct. 636 (1991); and United States v. Fleet Factors Corp., 901 F.2d 1550 (11th Cir. 1990) (addressing operator liability of a secured creditor), cert. denied, 498 U.S. 1046 (1991). 262. 775 F. Supp. at 1547-48. 263. Id. at 1547 n.4. 264. Id. 265. Id. at 1548. 266. Id. (emphasis in original). 267. See supra note 264 and accompanying text. 268. 775 F. Supp. 435 (D. Mass. 1991). 269. Id. at 441 (quoting Kayser-Roth, 910 F.2d at 27). 270. Id. at 437. 271. Id. at441. 272. Id. 273. Id. (quoting Dedham Water Co. v. Cumberland Farms Dairy, Inc., 805 F.2d 1074, 1081 (1st Cir. 1986)). The court's point here is unclear; the mere existence of the corporate form has never been enough to shield a parent from liability in the presence of factors that would indicate that the parent and subsidiary did not have a separate corporate existence in fact. See generally supra notes 97-104 and accompanying text. 274. 775 F. Supp. at 441. 275. Id. at 442 (citing Acushnet, 675 F. Supp. at 32). 276. See id.; John Boyd, 775 F. Supp. at 442. 277. Acushnet, 675 F. Supp. at 33; John Boyd, 775 F. Supp. at 442.

Bifurcation Under CERCLA Page 90 278. Acushnet, 675 F. Supp. at 33-34; John Boyd, 775 F. Supp. at 442. 279. The complaint did not allege that the parent corporation exercised any sort of financial or operational control over the subsidiary, but rather merely alleged that the parent's "control" consisted of voting to liquidate the subsidiary and to sell its assets. Id. at 443. The John Boyd court found that such an action was no more than an exercise of normal shareholder rights and so could not support an allegation of "pervasive control." Id. 280. See supra notes 97-104 and accompanying text (discussing traditional piercing doctrine). 281. See, e.g., United States v. Kayser-Roth Corp., 910 F.2d 24, 27 (1st Cir. 1990), cert. denied, 498 U.S. 1046 (1991); Jacksonville Elec. Auth. v. Eppinger & Russell Co., 776 F. Supp. 1542, 1546 (M.D. Fla. 1991); John Boyd Co. v. Boston Gas Co., 775 F. Supp. 435, 441 (D. Mass. 1991); City of New York v. Exxon Corp., 112 B.R. 540, 550 (S.D.N.Y. 1990). 282. See, e.g., John Boyd, 775 F. Supp. at 442; In re Acushnet & New Bedford Harbor, 675 F. Supp. 22, 33 (D. Mass. 1987). 283. See, e.g., Kayser-Roth, 724 F. Supp. at 24 (discussed supra note 217 and accompanying text). 284. See e.g., supra notes 281 & 282 and accompanying text (disucssing various cases). 285. 712 F. Supp. 1193 (E.D. Pa. 1989). 286. Id. at 1202. Nicolet examined the piercing issue the context of a a parent corporation's motion for summary judgment, which was denied. The complaint alleged that the subsidiary was the former owner and operator of the site, that the parent was first the majority, and then the sole, shareholder of the subsidiary, that the parent had actively participated in the management of the operations at the site while hazardous substances were being disposed of, that the parent was familiar with the subsidiary's disposal practices, could control disposal and its resulting releases, and could have abated

Bifurcation Under CERCLA Page 91 damages, and that the parent benefitted from the subsidiary's waste disposal practices. Id. at 1196-97. 287. See supra notes 105 and accompanying text. 288. See supra notes Part II.B and accompanying text (discussing the actual exercise of control test for direct liability). Nicolet correctly focuses on the control of the subsidiary, which is the relevant relationship for indirect liability. Direct liability, on the other hand, is based upon the parent's involvement in the facility. See supra notes 238-42 and accompanying text. Thus, the Nicolet test is not idential to the actual exercise of control test. 289. See supra notes 96-104 and accompanying text. 290. See, e.g., Acushnet, 675 F. Supp. at 33-34 (discussed supra notes 153-60 and accompanying text). 291. Kayser-Roth, 724 F. Supp. at 24 (discussed supra notes 208 and accompanying text). 292. Id.

— --'- ---— i- — 1-~'- *-"~-'