A Good Deed May Go Unpunished: Parties That Voluntarily Clean Up Sites Can Sue for Cost Recovery Under CERCLA

Atlantic Research Case

The facts of Atlantic Research are straightforward. Atlantic Research Corporation ('ARC') leased property at a facility owned by the Department of Defense, which ARC used to retrofit rocket motors for the United States. A high-pressure water spray was used to remove spent rocket propellant, which was then burned. Some of the wastewater and residue from the burnt propellant resulted in contamination to soil and groundwater. ARC voluntarily investigated and cleaned up the contamination, and then sued the government under ”107(a) and 113 to recover a portion of its costs.

The argument before the Court centered on the meaning of '107(a)(4)(B), which allows the recovery of 'any other necessary costs of response incurred by any other person,' in addition to the costs of removal or remedial action incurred by the government or Indian tribes, as set forth in the previous sub-paragraph, '107(a)(4)(A).

ARC argued that 'any other person' meant anyone other than the government or Indian tribes, i.e., private parties who had incurred response costs. The United States argued that the phrase referred to anyone other than the four categories of PRP previously listed in '107(a). It reasoned that ARC's position would render the 'other' in 'any other person' superfluous because the phrase 'any other necessary costs of response' in '107(a)(4)(B) already served to preclude governmental and tribal entities (whose remedy is set forth in '107(a)(4)(A)) from recovering under sub-paragraph (B). In addition, the government argued that, in providing for a limited right of contribution in '113 while leaving '107 untouched, Congress expressed its intention not to provide any cause of action to PRPs beyond that set forth in '113.

Justice Clarence Thomas, writing for the Court, dealt primarily with the parties' argument over the language of '107(a). Explaining that '[s]tatutes must 'be read as a whole,” and noting the 'remarkably similar structures' of sub-paragraphs (A) and (B), the Court concluded that the language of subparagraph (B) could be understood only with reference to sub-paragraph (A). Both paragraphs concern costs incurred by certain entities that bear a specified relationship to the national contingency plan (a plan promulgated by the Environmental Protection Agency that specifies how to respond to contamination). Moreover, the Court observed, the use of the phrase 'any other necessary costs' establishes a link between the two provisions. In light of that relationship, it is natural to read 'any other person' as referring to the immediately preceding sub-paragraph (A), i.e., anyone other than a government or tribal entity.

The Court found the United States' interpretation 'makes little textual sense.' Its reading of '107(a)(4)(B), which contrasts 'any other necessary costs' to the immediately preceding sub-paragraph (A) while contrasting 'any other person' to the four categories of PRP previously listed in the statute, 'would destroy the symmetry' of sub-paragraphs (A) and (B) 'and render sub-paragraph (B) internally confusing.' Additionally, the statute's broad definition of PRPs includes virtually all persons likely to incur cleanup costs, even so-called 'innocent' PRPs that were not responsible for any contamination. The government's interpretation of 'any other person' would, in the Court's view, 'reduce the number of potential plaintiffs to almost zero, rendering '107(a)(4)(B) a dead letter.'

As for the government's argument that the word 'other' in 'any other person' would be rendered superfluous under ARC's interpretation, the Court responded that it was not necessary 'to avoid surplusage at all costs' in statutory construction. 'It is appropriate to tolerate a degree of surplusage rather than adopt a textually dubious construction that threatens to render the entire provision a nullity.'

The Court also made short work of the United States' argument regarding the interplay of ”107 and 113, distinguishing between the remedy of cost recovery under '107(a) and contribution under '113(f) as 'two 'clearly distinct' remedies.' Contribution involves a tort-feasor's right to collect from others responsible for the same tort after the tort-feasor has paid more than its proportionate share. Cost recovery, by contrast, allows a party to recover without any establishment of liability to a third party. Put another way, a plaintiff suing under '107 seeks to recover 'the costs it has 'incurred' in cleaning up a site. ' When a party pays to satisfy a settlement agreement or a court judgment, it does not incur its own costs of response. Rather, it reimburses other parties for costs that those parties incurred.'

Holding that the plain language of '107(a)(4)(B) allows a PRP to recover costs from other PRPs, the Court upheld the Eighth Circuit's decision, allowing ARC to maintain its cost recovery action against the United States.

After Atlantic Research

The result in Atlantic Research represents a victory of sorts for state governments that increasingly have adopted voluntary cleanup programs that allow parties (PRP or otherwise) to clean up contaminated land pursuant to an agreement with a modicum of state oversight. These programs, which typically do not constitute the type of approved settlements referred to in '113, allow the states to preserve their limited resources by reducing the amount of involvement in enforcement, litigation, and negotiation with PRPs. If the participants in the programs did not have the incentive of recovering their costs under CERCLA, however, the efficacy of such programs would be diminished.

From the standpoint of PRPs, Atlantic Research basically preserves the status quo, in which PRPs are free to sue other PRPs under CERCLA, regardless of whether they had first been sued under CERCLA or settled with the government. If anything, the recent decision mainly represents a change in form over substance. While before, courts 'directed traffic' by considering all PRP suits as claims for contribution under '113, courts will now have to consider the plaintiff's procedural posture ' i.e., sued or not sued? settled or not settled? ' before deciding how to treat its CERCLA claim. In any event, as both the Eighth Circuit and Supreme Court observed, once sued under CERCLA, a defendant PRP can counterclaim for contribution under '113, which will, as a practical matter, result in a court's equitably allocating responsibility among all PRPs, including the plaintiff.

Nevertheless, while superficially little appears to have changed, Atlantic Research leaves several ancillary questions unanswered, some of which may offer plaintiff PRPs additional tactical benefits in litigation, depending on how those questions are resolved. For example, even if a defendant PRP files a '113 counterclaim for contribution, whether the plaintiff's claim is treated as a cost recovery action or contribution claim could affect the parties' respective burdens of proof, and could affect the ultimate allocation of costs, including the allocation of so-called 'orphan shares' (percentages of fault that are attributable to PRPs who are judgment-proof). How the plaintiff's claim is treated may also determine whether it is governed by a three-year or six-year statute of limitations.

In addition, the protection from contribution actions afforded to settling parties under '113(f)(2) does not extend to protection against '107 cost recovery actions. Thus, a PRP that believes it has purchased finality through a settlement with the government may suddenly find itself the target of another PRP's lawsuit, artfully pleaded as '107 cost recovery.

Even if, as the Court predicted, a district court were to apply traditional rules of equity by considering a party's prior settlement as a part of the liability calculus, the litigation costs to get to that point would not be insignificant.

Furthermore, while 'traffic directing' under Atlantic Research may appear fairly straightforward, requiring only a determination of whether the plaintiff has been sued or settled, there is a significant gray area left unresolved. Whereas '113(f)(1) provides for contribution 'during or following' a ”106 or 107 action, '113(f)(3)(B) provides for a right of contribution to any person who enters into an administrative or judicially approved settlement with the government. Often, those types of settlements are embodied in consent decrees pursuant to which the PRP agrees to undertake a cleanup at its own expense pursuant to government oversight. On the one hand, the costs of performing under the consent decree are clearly 'costs of response' incurred by the PRP, which, under Atlantic Research, are the type of costs subject to '107 cost recovery. On the other hand, an action to recover these costs appears to be exactly the kind of contribution action contemplated by '113(f)(3)(B). In such a case, does the legislative intent of '113 prevail over the plain meaning of '107? The Supreme Court declined to address the issue, other than to acknowledge in a footnote that it was not addressing the issue.

Quite apart from these unanswered questions, another footnote in the Court's opinion has the potential to change a widely held understanding of liability under '107 completely. Justice Thomas stated, 'We assume without deciding that '107(a) provides for joint and several liability.' Obviously, if '107(a) cost recovery does not impose joint and several liability, many of the assumptions regarding the perceived advantages of an action for cost recovery over one for contribution would have to be re-examined.

Given the costs associated with the remediation of the most complex of contaminated sites as well as the resources available to the large corporations that are typically the parties involved in those cleanups, it is surely just a matter of time before the Supreme Court is asked to revisit these issues.

Jason L. Jurkevich is an associate with Sills Cummis & Gross P.C., of Newark, NJ. The views and opinions expressed in this article are those of the author and do not necessarily reflect those of the firm. This article originally appeared in the New Jersey Law Journal, a sister publication of this newsletter.

One U.S. Supreme Court decision this past term brought welcomed news to those labeled 'potentially responsible parties' ('PRPs') under the Comprehensive Environmental Response, Compensation and Liability Act, 42 U.S.C. '9601 et seq. ('CERCLA'). On June 11, 2007, the Court unanimously agreed, in United States v. Atlantic Research, No. 06'562, that PRPs that voluntarily clean up contaminated property may bring suit for cost recovery against other PRPs under '107 of CERCLA. In classic style, however, the Court's opinion left certain questions unanswered and even raised one or two new questions.

In the first few years after CERCLA's passage, it was unclear whether a PRP that had cleaned up contaminated property could bring an action for cost recovery against other PRPs under '107. The question appeared to become moot after 1986, when Congress passed the Superfund Amendment and Reauthorization Act ('SARA'), which added a number of new provisions to CERCLA, including '113, a provision expressly allowing PRPs to assert claims for contribution against other PRPs. Following the SARA amendments, every federal circuit adopted the approach that, whether or not plaintiff PRPs who brought CERCLA actions asserted separate claims under '107 and '113, the claims were uniformly treated as contribution claims.

However, in Cooper Indus. v. Aviall Servs. , 543 U.S. 157 (2004), the Supreme Court narrowly interpreted the statutory contribution right to apply only to PRPs that had actually been sued under CERCLA or entered into certain judicially approved or administrative settlements. As a result, whether PRPs that voluntarily undertook to clean up contaminated land could bring suit for cost recovery under CERCLA '107 ' a question that had basically lain dormant for nearly two decades ' was suddenly of primary concern to many voluntary remediators that viewed '107 as their sole route to financial recovery. A split developed among the Circuits. The Second, Seventh, and Eighth Circuits held that PRPs could assert CERCLA claims under '107(a); the Third Circuit said they could not.

Atlantic Research Case

The facts of Atlantic Research are straightforward. Atlantic Research Corporation ('ARC') leased property at a facility owned by the Department of Defense, which ARC used to retrofit rocket motors for the United States. A high-pressure water spray was used to remove spent rocket propellant, which was then burned. Some of the wastewater and residue from the burnt propellant resulted in contamination to soil and groundwater. ARC voluntarily investigated and cleaned up the contamination, and then sued the government under ”107(a) and 113 to recover a portion of its costs.

The argument before the Court centered on the meaning of '107(a)(4)(B), which allows the recovery of 'any other necessary costs of response incurred by any other person,' in addition to the costs of removal or remedial action incurred by the government or Indian tribes, as set forth in the previous sub-paragraph, '107(a)(4)(A).

ARC argued that 'any other person' meant anyone other than the government or Indian tribes, i.e., private parties who had incurred response costs. The United States argued that the phrase referred to anyone other than the four categories of PRP previously listed in '107(a). It reasoned that ARC's position would render the 'other' in 'any other person' superfluous because the phrase 'any other necessary costs of response' in '107(a)(4)(B) already served to preclude governmental and tribal entities (whose remedy is set forth in '107(a)(4)(A)) from recovering under sub-paragraph (B). In addition, the government argued that, in providing for a limited right of contribution in '113 while leaving '107 untouched, Congress expressed its intention not to provide any cause of action to PRPs beyond that set forth in '113.

Justice Clarence Thomas, writing for the Court, dealt primarily with the parties' argument over the language of '107(a). Explaining that '[s]tatutes must 'be read as a whole,” and noting the 'remarkably similar structures' of sub-paragraphs (A) and (B), the Court concluded that the language of subparagraph (B) could be understood only with reference to sub-paragraph (A). Both paragraphs concern costs incurred by certain entities that bear a specified relationship to the national contingency plan (a plan promulgated by the Environmental Protection Agency that specifies how to respond to contamination). Moreover, the Court observed, the use of the phrase 'any other necessary costs' establishes a link between the two provisions. In light of that relationship, it is natural to read 'any other person' as referring to the immediately preceding sub-paragraph (A), i.e., anyone other than a government or tribal entity.

The Court found the United States' interpretation 'makes little textual sense.' Its reading of '107(a)(4)(B), which contrasts 'any other necessary costs' to the immediately preceding sub-paragraph (A) while contrasting 'any other person' to the four categories of PRP previously listed in the statute, 'would destroy the symmetry' of sub-paragraphs (A) and (B) 'and render sub-paragraph (B) internally confusing.' Additionally, the statute's broad definition of PRPs includes virtually all persons likely to incur cleanup costs, even so-called 'innocent' PRPs that were not responsible for any contamination. The government's interpretation of 'any other person' would, in the Court's view, 'reduce the number of potential plaintiffs to almost zero, rendering '107(a)(4)(B) a dead letter.'

As for the government's argument that the word 'other' in 'any other person' would be rendered superfluous under ARC's interpretation, the Court responded that it was not necessary 'to avoid surplusage at all costs' in statutory construction. 'It is appropriate to tolerate a degree of surplusage rather than adopt a textually dubious construction that threatens to render the entire provision a nullity.'

The Court also made short work of the United States' argument regarding the interplay of ”107 and 113, distinguishing between the remedy of cost recovery under '107(a) and contribution under '113(f) as 'two 'clearly distinct' remedies.' Contribution involves a tort-feasor's right to collect from others responsible for the same tort after the tort-feasor has paid more than its proportionate share. Cost recovery, by contrast, allows a party to recover without any establishment of liability to a third party. Put another way, a plaintiff suing under '107 seeks to recover 'the costs it has 'incurred' in cleaning up a site. ' When a party pays to satisfy a settlement agreement or a court judgment, it does not incur its own costs of response. Rather, it reimburses other parties for costs that those parties incurred.'

Holding that the plain language of '107(a)(4)(B) allows a PRP to recover costs from other PRPs, the Court upheld the Eighth Circuit's decision, allowing ARC to maintain its cost recovery action against the United States.

After Atlantic Research

The result in Atlantic Research represents a victory of sorts for state governments that increasingly have adopted voluntary cleanup programs that allow parties (PRP or otherwise) to clean up contaminated land pursuant to an agreement with a modicum of state oversight. These programs, which typically do not constitute the type of approved settlements referred to in '113, allow the states to preserve their limited resources by reducing the amount of involvement in enforcement, litigation, and negotiation with PRPs. If the participants in the programs did not have the incentive of recovering their costs under CERCLA, however, the efficacy of such programs would be diminished.

From the standpoint of PRPs, Atlantic Research basically preserves the status quo, in which PRPs are free to sue other PRPs under CERCLA, regardless of whether they had first been sued under CERCLA or settled with the government. If anything, the recent decision mainly represents a change in form over substance. While before, courts 'directed traffic' by considering all PRP suits as claims for contribution under '113, courts will now have to consider the plaintiff's procedural posture ' i.e., sued or not sued? settled or not settled? ' before deciding how to treat its CERCLA claim. In any event, as both the Eighth Circuit and Supreme Court observed, once sued under CERCLA, a defendant PRP can counterclaim for contribution under '113, which will, as a practical matter, result in a court's equitably allocating responsibility among all PRPs, including the plaintiff.

Nevertheless, while superficially little appears to have changed, Atlantic Research leaves several ancillary questions unanswered, some of which may offer plaintiff PRPs additional tactical benefits in litigation, depending on how those questions are resolved. For example, even if a defendant PRP files a '113 counterclaim for contribution, whether the plaintiff's claim is treated as a cost recovery action or contribution claim could affect the parties' respective burdens of proof, and could affect the ultimate allocation of costs, including the allocation of so-called 'orphan shares' (percentages of fault that are attributable to PRPs who are judgment-proof). How the plaintiff's claim is treated may also determine whether it is governed by a three-year or six-year statute of limitations.

In addition, the protection from contribution actions afforded to settling parties under '113(f)(2) does not extend to protection against '107 cost recovery actions. Thus, a PRP that believes it has purchased finality through a settlement with the government may suddenly find itself the target of another PRP's lawsuit, artfully pleaded as '107 cost recovery.

Even if, as the Court predicted, a district court were to apply traditional rules of equity by considering a party's prior settlement as a part of the liability calculus, the litigation costs to get to that point would not be insignificant.

Furthermore, while 'traffic directing' under Atlantic Research may appear fairly straightforward, requiring only a determination of whether the plaintiff has been sued or settled, there is a significant gray area left unresolved. Whereas '113(f)(1) provides for contribution 'during or following' a ”106 or 107 action, '113(f)(3)(B) provides for a right of contribution to any person who enters into an administrative or judicially approved settlement with the government. Often, those types of settlements are embodied in consent decrees pursuant to which the PRP agrees to undertake a cleanup at its own expense pursuant to government oversight. On the one hand, the costs of performing under the consent decree are clearly 'costs of response' incurred by the PRP, which, under Atlantic Research, are the type of costs subject to '107 cost recovery. On the other hand, an action to recover these costs appears to be exactly the kind of contribution action contemplated by '113(f)(3)(B). In such a case, does the legislative intent of '113 prevail over the plain meaning of '107? The Supreme Court declined to address the issue, other than to acknowledge in a footnote that it was not addressing the issue.

Quite apart from these unanswered questions, another footnote in the Court's opinion has the potential to change a widely held understanding of liability under '107 completely. Justice Thomas stated, 'We assume without deciding that '107(a) provides for joint and several liability.' Obviously, if '107(a) cost recovery does not impose joint and several liability, many of the assumptions regarding the perceived advantages of an action for cost recovery over one for contribution would have to be re-examined.

Given the costs associated with the remediation of the most complex of contaminated sites as well as the resources available to the large corporations that are typically the parties involved in those cleanups, it is surely just a matter of time before the Supreme Court is asked to revisit these issues.

Jason L. Jurkevich is an associate with Sills Cummis & Gross P.C., of Newark, NJ. The views and opinions expressed in this article are those of the author and do not necessarily reflect those of the firm. This article originally appeared in the New Jersey Law Journal, a sister publication of this newsletter.