Amicus Curiae Brief — Territory of Guam, Petitioner v. United States

No. 20-382

In the Supreme Court of the United States

__________________

GOVERNMENT OF GUAM,

Petitioner,

v.

UNITED STATES OF AMERICA,

Respondent.

__________________

On Writ of Certiorari to the

United States Court of Appeals

for the District of Columbia

__________________

BRIEF FOR ATLANTIC RICHFIELD COMPANY

AS AMICUS CURIAE IN SUPPORT OF

RESPONDENT

__________________

ELISABETH S. THEODORE

STEPHEN K. WIRTH

ARNOLD & PORTER KAYE

SCHOLER LLP

601 Massachusetts Ave.

NW

Washington, DC 20001

(202) 942-5000

SHANNON WELLS STEVENSON

Counsel of Record

KENZO S. KAWANABE

BENJAMIN B. STRAWN

ADITI KULKARNI-KNIGHT

DAVIS GRAHAM & STUBBS LLP

1550 17th St., Suite 500

Denver, CO 80202

(303) 892-9400

shannon.stevenson@dgslaw.com

Counsel for Amicus Curiae

March 31, 2021

Becker Gallagher · Cincinnati, OH · Washington, D.C. · 800.890.5001

i

TABLE OF CONTENTS

TABLE OF AUTHORITIES....................................... ii

INTEREST OF THE AMICUS CURIAE................... 1

INTRODUCTION AND SUMMARY OF ARGUMENT ......................................................................... 3

ARGUMENT .............................................................. 6

I. The United States’ Proposed Rule Provides Critical Clarity and Consistency to Private Parties

With Potential Contribution Liability ................. 6

II. Guam’s Proposed Rule Will Undermine the PurPoses of CERCLA’s Statute of Limitations for

Contribution Claims ........................................... 14

CONCLUSION ......................................................... 22

ii

CASES

TABLE OF AUTHORITIES

Asarco LLC v. Atl. Richfield Co.,

866 F.3d 1108 (9th Cir. 2017) ................ 8, 9, 10, 17

Atlantic Richfield Co. v. Christian,

140 S. Ct. 1335 (2020) ............................................1

Bernstein v. Bankert,

733 F.3d 190 (7th Cir. 2012) ........................ 7, 8, 17

Bowsher v. Merck & Co.,

460 U.S. 824 (1983) .............................................. 13

Fla. Dep’t of Revenue v. Piccadilly Cafeterias,

Inc., 554 U.S. 33 (2008) ........................................ 13

Fla. Power Corp. v. FirstEnergy Corp.,

810 F.3d 996 (6th Cir. 2015) ...................... 8, 17, 20

ITT Indus. v. BorgWarner, Inc.,

506 F.3d 452 (6th Cir. 2007) ..................................8

Morrison Enters., LLC v. Dravo Corp.,

638 F.3d 594 (8th Cir. 2011) ................................ 20

Refined Metals Corp. v. NL Indus.,

937 F.3d 928 (7th Cir. 2019) .............................. 8, 9

United States v. Marion,

404 U.S. 307 (1971) .............................................. 14

iii

STATUTES

42 U.S.C. § 9613(f)(3)(B) ................................... passim

42 U.S.C. § 9613(g)(2)............................................... 16

REGULATIONS

63 Fed. Reg. 8473 (Feb. 19, 1998) ...............................9

OTHER AUTHORITES

H.R. Rep. No. 99-253 (1985), reprinted in 1986

U.S.C.C.A.N. 2835.......................................... 20, 21

1

INTEREST OF THE AMICUS CURIAE

Atlantic Richfield Company has a significant interest in how this Court interprets section 113 of the

Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (“CERCLA”). 1 In 1977,

Atlantic Richfield acquired and later merged with The

Anaconda Company, which for decades had operated

copper and other mining and mineral processing sites

throughout the United States. As a consequence, Atlantic Richfield found itself inheriting responsibility

for cleaning up pollution in cooperation with the Environmental Protection Agency (“EPA”) at CERCLA

“Superfund” sites across the country. Atlantic Richfield was a party to this Court’s most recent decision

involving CERCLA, Atlantic Richfield Co. v. Christian, 140 S. Ct. 1335 (2020). And Atlantic Richfield is

also both the proponent and the target of contribution

claims under § 113(f)(3)(B) of CERCLA for and

against other private parties involved with cleanups

at polluted sites. As both a party seeking contribution,

and a party from whom contribution is sought, Atlantic Richfield has substantial financial and legal interests in the resolution of this case and an important,

balanced perspective on the consequences the Court’s

ruling will have for private parties involved in environmental cleanups.

Atlantic Richfield also has a related case pending

in this Court, Atlantic Richfield Co. v. Asarco LLC,

Case No. 20-1142, the outcome of which will turn on

the Court’s decision here. Atlantic Richfield’s case involves a lead smelting facility at the East Helena

1 All parties consented to the filing of this brief. Counsel for At-

lantic Richfield authored this brief in whole. No one other than

Atlantic Richfield funded the preparation of the brief.

2

Superfund Site in Montana that plaintiff Asarco and

its predecessors owned and operated for more than

100 years. Atlantic Richfield’s predecessor Anaconda

had leased land from Asarco at the site to operate a

zinc fuming plant between 1927 and 1972, and then

sold the plant to Asarco, which continued to operate it

for another decade. In 1998, Asarco entered a judicially approved Consent Decree with EPA that required it to perform a comprehensive cleanup at East

Helena (“1998 Consent Decree”). But Asarco failed to

perform the cleanup and entered bankruptcy in 2005.

To resolve EPA’s claims in the bankruptcy, Asarco

committed to fund its obligations under the 1998 Consent Decree and turned the remediation over to a custodial trustee. In 2012—fourteen years after promising to clean up the site—Asarco brought a contribution claim against Atlantic Richfield.

Atlantic Richfield argued that the three-year statute of limitations in CERCLA § 113 barred Asarco’s

contribution claim because Asarco had “resolved” its

“liability for some or all of a response action” in its

1998 Consent Decree. The district court agreed and

dismissed Asarco’s claim. But the Ninth Circuit reversed, holding that the 1998 Consent Decree, while

requiring Asarco to perform multiple response actions, did not “resolve [Asarco’s] liability” for some or

all of a response action because one paragraph of the

Decree released claims for civil penalties without releasing claims for injunctive relief. On remand, Asarco

obtained a multi-million-dollar judgment against Atlantic Richfield based on its 14-year-old claim. Atlantic Richfield has now petitioned for certiorari and

asked this Court to hold the petition pending the decision in this case. See No. 20-1142.

3

If this Court adopts the United States’ proposed

interpretation of § 113(f)(3)(B), Atlantic Richfield will

prevail on remand in the Ninth Circuit. The United

States correctly contends that “[a] settlement resolves

a person’s liability for a response action (or response

costs) if it definitively requires the person to perform

(or pay for) that action.” U.S. Br. 11. It is undisputed

that Asarco’s 1998 Consent Decree did just that.

Atlantic Richfield thus has a direct interest in the

outcome of this case. And Atlantic Richfield’s case—as

well as Atlantic Richfield’s experience with CERCLA

contribution claims more generally—provides an essential perspective that is missing in this dispute between two government actors: that of private parties

which are most often subject to CERCLA contribution

claims.

In particular, Atlantic Richfield’s case underscores the adverse effects of Guam’s proposed approach to § 113(f)(3)(B). It illustrates that deciding

whether a party has resolved its liability based on the

nuances of release language in a settlement agreement leads to unpredictable and inequitable results.

And it illustrates the significant unfairness in letting

contribution claims linger, with no statute of limitations running, even after a party has agreed to perform or pay for environmental cleanup in a settlement

with the United States or a State. This brief will accordingly aid the Court in interpreting the words “resolve [its] liability.”

INTRODUCTION AND

SUMMARY OF ARGUMENT

The fundamental policy objectives of CERCLA are

to promote timely cleanup of contaminated Sites and

4

to ensure that parties that caused contamination pay

for its cleanup. The fundamental policy objective of a

statute of limitations is to ensure that claims are

promptly pursued so that parties have certainty and

access to relevant evidence.

As Atlantic Richfield’s case and others illustrate,

Guam’s proposed rule undermines all of these policies.

It will lead to uncertainty regarding parties’ rights to

contribution, and it will allow settling parties to delay

indefinitely before pursuing contribution, prejudicing

non-settling PRPs and jeopardizing the availability of

the evidence necessary to fairly determine parties’ relative contributions.

First, the facts of Atlantic Richfield’s case illustrate the folly of deciding whether a party has resolved

its liability based on the nuances of the specific terms

of a particular settlement agreement. Adopting a rule

that allows different results based on immaterial differences in parties’ respective settlement agreements

will create uncertainty as to when parties have a right

to seek contribution, and as to when their statutes of

limitations have run. That uncertainty, in turn, will

frustrate CERCLA’s goal of incentivizing early settlements to fund environmental cleanups. All parties

would benefit from a simple test to determine when a

CERCLA contribution claim arises and the statute of

limitations on that claim begins to run.

Second, Atlantic Richfield’s case illustrates the

problems caused by letting contribution claims linger

with no statute of limitations running. Statutes of limitations exist to ensure certainty about parties’ potential liabilities and access to the witnesses and evidence necessary to fairly decide cases. By the time

5

Asarco filed its complaint in 2012, forty years had

passed since Atlantic Richfield had conducted any operations at the Site; thirty years had passed since EPA

started its investigation of the Site; and fourteen

years had passed since Asarco had entered a judicially

approved settlement with EPA, in which Asarco promised to perform a complete and comprehensive

cleanup of the Site. By the time the case was finally

tried in 2018, Atlantic Richfield was handicapped by

its inability to undertake its own investigation of the

Site before significant remedial work was done, as

well as its limited access to witnesses and documents.

In light of these considerations, Atlantic Richfield

urges the Court to adopt the United States’ position

and hold that a party has “resolved [its] liability” for

a specific response action when it enters a judicially

approved settlement that conclusively determines its

obligation to perform or pay for that response action.

This simple rule will promote certainty regarding the

existence of contribution claims, as well as the timely

resolution of those claims. Because many remediations take place in phases, however, the Court should

make sure to clarify that a party that settles its responsibility for one particular response action or part

of a response action—for example, cleaning up one

portion of a site—does not trigger the statute of limitations for contribution for other response actions it

has not yet agreed to perform or pay for in its settlement agreement.

6

I.

ARGUMENT

The United States’ Proposed Rule Provides

Critical Clarity and Consistency to Private

Parties With Potential Contribution Liability.

The United States urges the Court to adopt a

straightforward rule: when a party enters a judicially

approved settlement agreement that conclusively determines the party’s obligation to perform a particular

response action, the party has resolved its liability for

that response action and has a right to seek contribution for it. If the parties have entered into a “conclusive agreement about what [one] party must do,” U.S.

Br. 38, the settlement has resolved that party’s liability for at least “some” of a response action.

By contrast, Guam urges this Court to hold that

whether a party has “resolved its liability” is a recordintensive inquiry that requires examining lengthy settlement documents and evaluating myriad provisions,

including the existence and extent of any “release,”

the contours of any reservations of rights, the extent

of a party’s admission of liability, and presumably

many other provisions that might appear in a settlement agreement. Pet. Br. 41. Indeed, Guam argues

that any of many “magic words” can defeat the resolution of liability, including disclaimers or non-admissions of liability; reservations of rights by EPA to enforce compliance with the agreement, to require the

Potentially Responsible Party (“PRP”) to take additional response actions not addressed in the agreement, or to ensure the PRP complies with other applicable laws; or release terms that provide less than a

7

complete release for the PRP, or delay such release

until the PRP’s work is complete.

The Court should adopt the clear, bright-line rule

proposed by the United States. That rule is consistent

with the statutory text, U.S Br. 34-38, and advances

CERCLA’s underlying purposes. And the rule offers

critical advantages to the private sector parties who

most often find themselves at odds over CERCLA contribution claims. It offers certainty and predictability,

not only to parties asserting claims for contribution,

but also for third parties who have potential liability

for a site and must evaluate their own exposure to contribution claims based on agreements to which they

are not parties and over which they have no say.

It is easy to figure out whether a PRP has committed to EPA that it will undertake a particular response action: the consent decree will say so on its

face. Thus, for example, no one disputes that Guam

conclusively agreed to undertake response actions in

its settlement with the EPA. Likewise, in Atlantic

Richfield’s dispute with Asarco, no one disputes that

Asarco conclusively agreed to undertake particular

cleanup activities—response actions—in the 1998

Consent Decree. A PRP that enters such a settlement

will know when its right to contribution arises and expires under the United States’ rule with perfect clarity. Likewise, a third party like Atlantic Richfield will

know when the time has passed for a settling PRP to

pursue contribution claims; it will not be perpetually

waiting for the hammer to drop.

Guam’s proposed rule, by contrast, is exceptionally malleable and unpredictable. Cases from courts

8

applying that rule or versions of it illustrate as

much—as Guam put it in its petition for certiorari, the

courts are “expressly divided” on what sort of language in a Consent Decree or settlement defeats contribution. Pet. 17. Thus, for example, the Seventh Circuit held in one case that a settlement did not resolve

liability within the meaning of § 113(f) because it did

not “immediately resolv[e] all liability,” Bernstein v.

Bankert, 733 F.3d 190, 213 (7th Cir. 2012), while concluding in another case that a covenant not to sue did

resolve a party’s liability even though it was “not comprehensive,” see Refined Metals Corp. v. NL Indus.,

937 F.3d 928, 931 (7th Cir. 2019); see also Fla. Power

Corp. v. FirstEnergy Corp., 810 F.3d 996, 1004 (6th

Cir. 2015) (holding a consent order stating satisfaction

of its requirements “shall . . . resolve” the responsible

party’s “liability to EPA” nevertheless did not resolve

“some or all” of that party’s liability for purposes of

§ 113(f) because of the “context in which the reference

[to resolution of liability] [was] made”); ITT Indus. v.

BorgWarner, Inc., 506 F.3d 452, 459-60 (6th Cir. 2007)

(holding liability was not resolved where EPA had reserved the right to adjudicate liability for failure to

comply with the terms of the settlement agreement itself).

In Atlantic Richfield’s case, the Ninth Circuit followed a basic approach similar to that advocated by

Guam. The court first acknowledged that Asarco’s

1998 Consent Decree “clearly required Asarco to take

response actions to clean up hazardous waste at the

Site,” Asarco LLC v. Atl. Richfield Co., 866 F.3d 1108,

1114 (9th Cir. 2017), meaning that it would have triggered the statute of limitations under the rule the

United States proposes. But the court held that

9

Asarco’s agreement to undertake response actions

was not enough, and instead engaged in an intensive

analysis of various collateral provisions of the settlement agreement—agreeing with its sister circuits

about the import of certain particular terms, and disagreeing about others. For example, the Ninth Circuit

rejected the Sixth Circuit’s conclusion in ITT that a

reservation of the right to enforce the settlement itself

defeated contribution. But the Ninth Circuit also rejected the Seventh Circuit’s conclusion in Refined Metals that a settlement release need not be “comprehensive” to trigger contribution.

Ultimately, the Ninth Circuit held that Asarco’s

1998 Consent Decree did not trigger contribution because Paragraph 209 of the Decree released only

claims for civil penalties, but not claims for injunctive

relief. Asarco, 866 F.3d at 1126 (noting that Paragraph 209 “is expressly limited to liability with regards to the United States’ claims for civil penalties”).

But Paragraph 214 of the agreement barred EPA from

initiating “a separate action under Sections 3008(h)

and 3013 of RCRA … for work to be performed at the

Facility,” App. 252, ¶ 214 2; in other words, EPA did

release its claims for injunctive relief so long as Asarco

performed the corrective action work it agreed to undertake. 3

2 “App.” citations refer to Atlantic Richfield’s appendix to its Pe-

tition for Certiorari in Atlantic Richfield Co. v. Asarco LLC, Case

No. 20-1142.

3 Lest there be any doubt, EPA confirmed this when it published

its Public Notice of the Consent Decree in the Federal Register,

explaining that the decree “resolve[d] civil penalty and injunctive

10

The Ninth Circuit further opined that the Consent Decree did not trigger contribution because it

contained assorted “references to Asarco’s continued

legal exposure,” including (1) a provision that did not

limit Asarco’s obligation to perform work outside the

facility’s boundaries, even if it lacked access (¶ 122);

(2) a paragraph setting forth a limited covenant not to

sue (¶ 214); and (3) other paragraphs setting forth the

scope of the release (¶¶ 216-17). Asarco, 866 F.3d at

1126.

EPA consent decrees are notoriously long and

complicated. Guam’s petition asks this Court to hold

that three specific features present in its own settlement—the absence of an admission of liability, the

U.S.’s reservation of rights, and a release that is conditioned on successful completion of the response action—mean that the settlement did not resolve some

or all of Guam’s liability. Pet. Br. 41-43. But these are

hardly the only potentially relevant features, as the

Ninth Circuit’s decision in Atlantic Richfield’s case illustrates. Guam’s fluid position would force litigants

and the courts to interpret pages upon pages of EPA

consent decrees—a daunting prospect that is further

complicated by the fact that these agreements are

subject to frequent revision and amendment. Adopting Guam’s position thus dooms regulated businesses

to continued uncertainty as to which magic words

relief claims of the United States against ASARCO under

RCRA.” 63 Fed. Reg. 8473 (Feb. 19, 1998 (emphasis added); see

id. (“The consent decree . . . resolves civil penalty claims of the

United States against [ASARCO] under the CWA . . . [and] also

resolves civil penalty and injunctive relief claims of the United

States against ASARCO under RCRA….” (emphasis added)).

11

sufficiently describe the required degree of resolution,

and thus to continued uncertainty about the extent of

their ongoing exposure to CERCLA contribution actions.

This uncertainty is especially pernicious because

the private parties that are most often the target of

CERCLA contribution actions have no ability to affect

the terms of the settlement agreement between another private party and the government. Atlantic

Richfield had no ability back in 1998 to require Asarco

and the EPA to include language that would have

clarified whether the settlement triggered the statute

of limitations for bringing contribution claims. And

settling parties have no particular incentive to push

for clarity, since they can always choose to bring a contribution action within three years. The parties with

the greatest interest in clarity—the defendants in future contribution actions—have no seat at the table.

Indeed, Asarco actually excluded Atlantic Richfield

from its settlement negotiations with the EPA.

ER003046. 4

A bright-line rule that focuses on whether the settlement requires a response action also prevents the

disparate treatment of similarly-situated parties

based on minor, non-substantive discrepancies in the

particular terms of their respective agreements. Contrasting the D.C. Circuit’s decision in this case with

the Ninth Circuit’s decision in Atlantic Richfield’s

case shows why. In each of these cases, EPA ordered

the primary PRP (Asarco and Guam) to investigate

4 “ER” citations are to the Excerpts of Record (“ER”) filed in Case

No. 18-35934 (9th Cir.).

12

contamination at a Superfund site. In each case, EPA

sued that PRP in a complaint seeking injunctive relief, and in each case the PRP entered a judicially approved consent decree with EPA. In each case, that

decree required the PRP to pay civil penalties and to

undertake specific response actions to clean up the respective sites. In each case, the consent decree also

stated that the United States reserved the right to

pursue claims for violations unrelated to the claims in

its complaint; that the settling party would be released from the United States’ claims when it complied with the decree’s requirements; and that the

parties had entered the decree without admitting liability.

Despite these materially identical facts, the Ninth

Circuit’s parsing of the scope of a few scattered provisions led to a different and unpredictable outcome.

Thus, for example, the Ninth Circuit focused on

whether Asarco used the specific word “release” with

respect to injunctive relief, even though the Consent

Decree in substance clearly released the claims for injunctive relief provided Asarco faithfully fulfilled its

commitments.

And more generally, none of the provisions that

Guam says should govern the inquiry have anything

to do with the goals of the statutory provision for CERCLA contribution. The goal is to compensate settling

parties who agree to perform remediations. So it

makes sense to conclude that a settling party has resolved its liability for “some” of a “response action” or

“some” of the “costs” of a “response action,” once it has

agreed to perform that response action or undertake

that cost. And a settling party may then seek

13

contribution for that particular response action or

cost. The manifest purpose of this language is to both

allow and require the settling party, once it knows it

will have to pay for any part of a response (either

through direct performance or by funding EPA’s

cleanup), to seek contribution for that specific part of

the response.

It is worth emphasizing that, when a party resolves its liability for only “some” of a response action

or cost, it acquires a right of contribution only for that

specific response action or cost. Resolving liability for

“some” of a response action does not create a right of

contribution for other potential response actions that

are not required by the party’s settlement. This is an

important distinction for PRPs who, unlike Asarco or

Guam, agree to undertake response actions specific to

particular geographic areas, particular contaminants,

or particular media (like groundwater or soil). Thus,

for example, a PRP’s agreement to undertake response actions to remediate soil at a site would not

trigger its right to seek contribution for a groundwater

cleanup that has not even begun and that might involve different responsible parties. Because many

EPA remediation agreements at particular sites are

phased, the contribution trigger in those cases must

be specific to a particular response action, not sitewide.

The Court has long recognized the advantages of

bright-line rules over fuzzy standards like that proposed by Guam. See Bowsher v. Merck & Co., 460 U.S.

824, 841 n.18 (1983) (“Bright-line rules upon which

the parties’ expectations may be firmly established

are preferable to . . . protracted litigation . . . .”); Fla.

14

Dep’t of Revenue v. Piccadilly Cafeterias, Inc., 554 U.S.

33, 52 (2008) (noting the advantages of a “simple,

bright-line rule instead of [a] complex, after-the-fact

inquiry”). This Court should not endorse a rule that

leads to idiosyncratic results that are impossible for

parties to predict. This is even more critical where this

determination impacts not just the rights of the parties to the Consent Decree, but also the rights of third

parties from whom contribution may be sought. This

Court should adopt the simple, clear, and decisive approach the United States proposes, a rule that is consistent with the text and the purpose of CERCLA.

II.

Guam’s Proposed Rule Will Undermine

the Purposes of CERCLA’s Statute of Limitations for Contribution Claims.

Guam’s rule is also irreconcilable with the policy

rationales underlying statutes of limitations generally, and CERCLA’s contribution statute of limitations specifically. Guam’s rule would indefinitely defer triggering the limitations period, allowing settling

parties who know that they have contribution claims

to delay bringing those claims. And because settling

parties can negotiate the terms of their settlements,

Guam’s rule would create massive opportunities for

gamesmanship and sandbagging. As Atlantic Richfield’s case illustrates, allowing a settling PRP to delay bringing contribution claims for decades is fundamentally unfair to the defendants in those contribution actions. Delay leads to lost witnesses, evidence,

and opportunity for investigation. Further, allowing

parties to pursue contribution claims decades after

promising to undertake response actions undermines

15

CERCLA’s policy of incentivizing prompt cleanups

and erodes judicial efficiency.

Civil statutes of limitations “represent a public

policy about the privilege to litigate,” and “their underlying rationale is to encourage promptness in the

bringing of actions, that the parties shall not suffer by

loss of evidence . . . .” United States v. Marion, 404

U.S. 307, 322 n.14 (1971) (internal citations omitted).

Limitations periods “are founded upon the general experience of mankind that claims, which are valid, are

not usually allowed to remain neglected, they promote

justice by preventing surprises through the revival of

claims that have been allowed to slumber until evidence has been lost, memories have faded, and witnesses have disappeared, and they are primarily designed to assure fairness to defendants.” Id. (citations

omitted). “The theory is that even if one has a just

claim it is unjust not to put the adversary on notice to

defend within the period of limitation and that the

right to be free of stale claims in time comes to prevail

over the right to prosecute them.” Id. Statutes of limitations also ensure that courts are “relieved of the

burden of trying stale claims when a plaintiff has slept

on his rights.” Id.

In the context of CERCLA contribution, it is the

agreement to undertake or fund a response action—

and not a release from liability—that reasonably puts

a settling PRP on notice that it is time to seek contribution from other PRPs. The purpose of the contribution remedy is to obtain contribution for response actions, so it only makes sense that the claim would accrue when the settling PRP understands that it will

be obligated to take a response action. Nothing about

16

the presence or absence of specific terms of a release

from liability from the United States—or any of the

other settlement provisions that Guam discusses—is

at all relevant to the basic question of whether a settling PRP is aware that it has a contribution claim

against a different entity.

Nor is the presence or absence of a release relevant to the prosecution of a settling PRP’s contribution claim. CERCLA already contemplates that settling PRPs will litigate contribution claims before the

final cost of response actions is known, and whether

or not the United States has “cap[ped] the settling

party’s liability.” Cf. Pet. Br. 39. That’s clear from section 113(f)(3)(B) itself, which states that a PRP can

bring a contribution claim once it has resolved its liability for “some” of a response action. CERCLA accordingly permits settling PRPs to litigate the liability percentage of each potential contributor to a particular

response action, and then obtain a declaratory judgment requiring the contributor to pay, on an ongoing

basis, that percentage of the ongoing response costs.

42 U.S.C. § 9613(g)(2). Thus, on the settling PRP’s

side of the balance sheet, there is no policy reason why

Congress would have wanted to delay accrual of a contribution claim pending some supposed perfectly

worded “release” from EPA.

By contrast, letting settling PRPs wait years, or

even decades, before pursuing claims for contribution

will cause immense prejudice to defendants in contribution cases. Under Guam’s proposal, settling PRPs

essentially can elect to trigger the contribution period

at their option. All they have to do is enter into a settlement agreement that requires response actions, but

17

decline to admit liability. Years down the line, if the

settling PRP decides it is time to pursue a claim for

contribution, it can enter into another agreement with

the United States that admits liability, thus starting

the statute of limitations.

This is not a farfetched scenario; when EPA works

with private parties to remediate polluted sites, it is

not a one-agreement-and-you-are-out situation. EPA

and the private party often work hand-in-glove for

years as EPA oversees the response actions, and enter

into multiple successive Administrative Orders on

Consent, Unilateral Administrative Orders, or judicial consent decrees governing the cleanups. Indeed,

courts that follow Guam’s proposed rule have

acknowledged that it permits the settling party to essentially decide whether and when to trigger the statute of limitations. See Bernstein, 733 F.3d at 213 (stating that the settling party and the EPA “may choose

to structure their contract” to resolve liability or not).

A settlement that conditions a “release” on performance might not resolve liability, while a settlement

that does not contain such a limitation could, see Fla.

Power Corp., 810 F.3d at 1003-04, even though the two

settlements are substantively identical, because in

the latter situation the EPA can still bring an enforcement action for nonperformance. Or a settling party

might avoid resolving its liability simply by including

language indicating that the settlement “shall not

constitute any admission of liability.” Id. at 1003.

The consequence for the unwitting nonparty is

that years down the line, after documents are lost and

memories have faded, it may be surprised with an untimely contribution action. That is exactly what

18

happened to Atlantic Richfield. Asarco had worked

with EPA since the early 1980s to investigate the East

Helena Site’s environmental condition. Because

Asarco had leased land to Atlantic Richfield, had acquired the zinc fuming operation in 1972, and had run

the operation itself for at least a decade, it had extensive knowledge of any contribution the zinc fuming

plant made to the Site’s contamination. Throughout

the 1980s and 1990s, Asarco conducted numerous Site

investigations to characterize the contamination at

the Site. Although EPA identified Atlantic Richfield

as a PRP in the late 1980s, Asarco, 866 F.3d at 1114,

neither EPA nor Asarco ever once identified the zinc

fuming operation as a meaningful contributor of any

contamination and never pursued any relief from Atlantic Richfield. ER003800 (Asarco’s calculation of total percent of arsenic contamination attributable to

Atlantic Richfield’s zinc fuming operation was

0.710%).

With the knowledge gathered from its investigations, Asarco in 1998 entered a comprehensive settlement agreement with EPA, in which Asarco undertook to perform a complete remediation of the Site. At

the time it entered this settlement, Asarco was of

course well aware of Atlantic Richfield’s operations at

the Site.

Seven years after entering the 1998 Consent Decree, Asarco filed for bankruptcy. In the bankruptcy

proceedings, Asarco fully acknowledged its obligations, undertaken in the 1998 Consent Decree, to perform response actions and clean up the Site and

agreed to fund that cleanup with a cash payment of

approximately $100 million. The Ninth Circuit held

19

that the bankruptcy settlement, unlike the 1998 Consent Decree, did resolve Asarco’s liability.

In 2012, three years after entering its bankruptcy

settlement, 14 years after committing to undertake a

sitewide cleanup, and forty years after Atlantic Richfield had ceased operating at the Site, Asarco for the

first time contended that Atlantic Richfield’s operation of the zinc fuming plant had contributed to the

pollution and demanded contribution from Atlantic

Richfield for the work that Asarco had originally committed to perform in the 1998 Consent Decree.

Because Asarco waited decades before ever indicating that it believed Atlantic Richfield bore any responsibility for the contamination at the Site, Atlantic

Richfield faced significant prejudice at trial:

-

Atlantic Richfield never had the opportunity

to conduct sampling at the Site to perform

its own investigation of impacts from the

zinc fuming plant because remedial work

was already under way by the time Asarco

brought its claim. Instead, Atlantic Richfield had to rely on isolated data points from

Asarco’s historical sampling.

-

Although these limited data points indicated that the zinc fuming plant did not contribute to the contamination, the court

noted the “sparse historical record” “which

does not include . . . very much in the way of

detailed information that would allow a fact

finder to determine the precise nature and

amount of pollutants that were released to

20

the environment by Asarco’s and Anaconda’s operations.” App. 65 n.10.

-

Asarco’s expert admitted that the historical

record was “meager” and incomplete such

that his allocation method was calculated

“on a more subjective basis.” ER001683:1-6;

1717:23-25.

-

Atlantic Richfield had no opportunity to obtain evidence from witnesses with

knowledge of its historical operations or its

communications with Asarco and EPA during the investigations of 1980s-90s. Atlantic

Richfield’s sole fact witness at trial had only

ever visited the site one time, in 1971. App.

96.

-

And, due to the passage of time, many relevant documents were lost—some destroyed

in a fire, and many of those remaining deteriorated to the point they were illegible.

Asarco’s delay of more than 14 years disadvantaged Atlantic Richfield by unfairly limiting the evidence and witnesses it had access to. By the time

Asarco pursued its claim, the trustee responsible for

the Site had already begun remediation efforts, App.

59, ¶ 67, making any current sampling irrelevant to

establishing the parties’ relative contributions. Had

the case been brought no later than 2001, Atlantic

Richfield would have had better access to data, witnesses, and documents, as well as an opportunity to

investigate the Site itself.

21

Adopting Guam’s proposed rule will lead to similar problems in other cases. As the Eighth Circuit has

observed, an interpretation of § 113 under which a

PRP could “be subject to new private-party claims for

decades” would “frustrate Congress’s intent to assure

that evidence concerning liability and response costs

i[s] fresh . . . and to provide some measure of finality

to affected responsible parties.” See Morrison Enters.,

LLC v. Dravo Corp., 638 F.3d 594, 610 (8th Cir. 2011)

(internal citation omitted)); Fla. Power Corp., 810

F.3d at 1001-09 (laboriously parsing the language of

multiple successive Administrative Orders on Consent and concluding that a settling party’s contribution claim was timely even though it was filed more

than 10 years after remediation began); see also H.R.

Rep. No. 99-253 pt. 1, at 138 (1985), reprinted in 1986

U.S.C.C.A.N. 2835, 2920. As these cases illustrate, allowing parties to commit to perform response actions,

and then to wait decades before alerting other PRPs

of their potential liability for contribution undermines

the purposes of statutes of limitations. It leads to surprise claims against the non-settling PRPs, and deprives them of access to evidence and witnesses to defend those claims.

In addition to undermining the purposes of the

statute of limitations, the result in Atlantic Richfield’s

case also illustrates how Guam’s proposed rule undermines CERCLA’s policies of promoting prompt cleanups and encouraging responsible parties to settle. In

that case, Asarco was rewarded for not performing the

response actions it promised to undertake in 1998 and

ultimately filing for bankruptcy. Had Asarco performed its obligations under the 1998 Consent Decree,

the 2009 bankruptcy court Consent Decree—which

22

the Ninth Circuit determined was the trigger for

Asarco’s limitations period—never would have existed

and Asarco would have had no basis on which to extend the limitations period. By contrast, a rule that

gives a party an immediate right to seek contribution

as soon as it enters a settlement agreement and commits to perform a response action promotes CERCLA’s

policies. A PRP that has stepped forward to commit to

undertake response actions should have an immediate right to seek contribution from other PRPs for that

response action so that it can know as soon as possible

the extent of the liability it will bear and so that it can

recover from those parties before time drags on and

those other parties become insolvent or otherwise

judgment-proof.

CERCLA’s fundamental policy objectives are to

promote timely cleanup of contaminated sites and to

ensure that parties that caused the contamination

pay for its cleanup. As illustrated by Atlantic Richfield’s case and others cited above, Guam’s proposed

rule advances neither of these policies. Allowing a settling PRP to wait years before seeking contribution

does not incentivize a prompt cleanup, and it may often result in a loss of evidence that precludes a reliable determination of which party caused the contamination. For these reasons, the Court should reject

Guam’s proposed rule.

CONCLUSION

The Court should hold that a claim for contribution accrues when a party agrees with the EPA that it

will undertake a response action or cover the costs of

a response action.

23

Respectfully submitted,

ELISABETH S. THEODORE

STEPHEN K. WIRTH

ARNOLD & PORTER KAYE

SCHOLER LLP

601 Massachusetts Ave.

NW

Washington, DC 20001

(202) 942-5000

SHANNON WELLS STEVENSON

Counsel of Record

KENZO S. KAWANABE

BENJAMIN B. STRAWN

ADITI KULKARNI-KNIGHT

DAVIS GRAHAM &

STUBBS LLP

1550 17th St., Suite 500

Denver, CO 80202

(303) 892-9400

shannon.stevenson@dgslaw.com

Counsel for Amicus Curiae

March 31, 2021