05-0514-cv, 05-0702-cv, 05-0706-cv, 05-0708-cv Marsh, et al. v. New York, et al.

UNITED STATES COURT OF APPEALS

FOR THE SECOND CIRCUIT

August Term, 2005

(Argued: February 6, 2006 Decided: August 28, 2007)

05-0514-cv (Lead),-0702-cv (XAP),-0706-cv (XAP),-0708-cv (XAP)

----------------------------------------

Langdon Marsh, as Acting Commissioner of the New York State Department of Environmental Conservation and Trustee of the Natural Resources and Michael D. Zagata, as Commissioner of the New York State Department of Environmental Conservation,

Plaintiffs,

State of New York and Denise M. Sheehan, as Acting Commissioner of the New York State Department of Environmental Conservation and Trustee of the Natural Resources,

Plaintiffs-Appellants-Cross-Appellees,

- v. -

Daniel Rosenbloom, Firmanco Associates, First Manhattan Company, as distributees of the assets of Panex Industries, Inc., Andreas Gal, Norman Halper and Oliver Lazare, in their capacities as co-executors of the Estate of Paul Lazare and Goldman Sachs & Company, as distributees of the assets of Panex Industries, Inc.,

Defendants-Cross-Defendants-Appellees- Cross-Appellants,

Panex Industries, Inc., Panex Industries, Inc. Liquidating Trust, Alpine Group, Inc., and Rochester Button Company, Inc.,

Defendant-Cross-Defendant,

Dresser Industries Inc., Intervenor-Plaintiff-Movant,

Turbodyne Electric Power Corporation, McGraw-Edison Company, Inc., Dresser-Rand Company, ABB Air Preheater, Inc., and Village of Wellsville,

Defendants-Cross-Claimants-Cross-Defendants,

Successor Panex Industries, Inc. Stockholders Liquidating Trust, Michael D. Debaecke, Esq., as Trustee of Successor Panex Industries, Inc. Stockholders Liquidating Trust,

Defendants,

Cooper Industries, Inc.,

Intervenor-Third Party-Defendant.

------------------------------------------

B e f o r e: JACOBS, POOLER, and JOHN R. GIBSON,* Circuit Judges.

Appeal from a judgment of the United States District Court

for the Western District of New York (Elfvin, J., District Judge)

following orders dismissing the State of New York's claims

against the dissolved corporation Panex's shareholder-

distributees and denying the Panex trustees' motion to dismiss

the State's CERCLA claims against Panex. We affirm the dismissal

of the State's claims against the shareholder-distributees and

reverse the judgment granted to the State on its CERCLA claims

against Panex.

* The Honorable John R. Gibson, Circuit Judge, United States Court of Appeals for the Eighth Circuit, sitting by designation.

2 RICHARD P. DEARING, Assistant Solicitor General of the State of New York, and EUGENE J. LEFF, Assistant Attorney General of the State of New York, for Plaintiff-Appellant-Cross-Appellee State of New York and Alexander Grannis**.

GITA F. ROTHSCHILD, law firm of McCarter & English, LLP, and MARK F. ROSENBERG, law firm of Sullivan & Cromwell LLP, for Defendants-Cross-Defendants-Appellees, Cross-Appellants Daniel Rosenbloom, Firmanco Associates, and First Manhattan Company.

ROBERT L. TOFEL and MARK A. LOPEMAN, Tofel & Partners, LLP, for Defendants- Cross-Defendants-Appellees, Cross- Appellants Andreas Gal, Estate of Paul Lazare, Norman Halper, Oliver Lazare.

BRIAN M. COGAN, Stroock & Stroock & Lavan LLP, for Defendant-Appellee-Cross- Appellant Goldman Sachs & Company.

JOHN R. GIBSON, Circuit Judge.

The State of New York appeals from orders of the United

States District Court for the Western District of New York

(Elfvin, J., District Judge) dismissing its claims against

shareholder-distributees of Panex Industries, Inc., a dissolved

Delaware corporation. The State asserted these claims several

years after Panex had been dissolved, outside the corporate wind-

up period established by Delaware General Corporation Law § 278

** Alexander Grannis succeeded Erin M. Crotty to the office of Commissioner of the New York State Department of Environmental Conservation and is named here pursuant to Federal Rule of Appellate Procedure 43(c)(2).

3 and before obtaining a judgment against Panex as required by

Delaware General Corporation Law § 325(b), but the State argues

that its claims are valid under the common law equitable trust

fund doctrine. The shareholder-distributees cross-appeal from

the district court's denial of a motion to dismiss the State's

CERCLA claims against Panex and the summary judgment granted to

the State on those claims. They argue that Delaware General

Corporation Law § 278 governs and that Panex lacked capacity to

be sued under the statute because it had been dissolved for over

three years by the time the State notified Panex of its claims

and filed suit. The district court found that CERCLA preempted

section 278 in this instance.

I.

The issues raised in this appeal are one chapter in a

complex tale involving numerous parties. At the heart of the

suit is the State's effort to recover $4.5 million in

unreimbursed environmental response costs that it has paid to

investigate and clean up the Wellsville-Andover Landfill site in

Allegany County, New York.1

Panex Industries, Inc., was formed in 1981 under Delaware

law as part of the reorganization plan of its predecessor

company, Duplan Corporation. One of Duplan's operating divisions

1 In all the State paid or raised costs of $10 million in connection with cleanup of the site, and the remaining sum is what is left after the State's settlements with other parties.

4 had been the Rochester Button Company, a manufacturing plant. In

the early 1970s, Rochester Button used the Wellsville-Andover

Landfill site to dispose of its industrial waste, placing much of

it in a special disposal pit designated for Rochester Button’s

exclusive use. There was abundant evidence that Rochester Button

made substantial deposits of hazardous waste at the landfill

during the course of its operations. The New York State

Department of Environmental Conservation ultimately determined

that the site presented a significant threat to the public health

and environment, and the State began incurring response costs in

connection with its investigation of contamination at the site in

April 1984.

Meanwhile, unaware of the contamination at the landfill site

or of the State's recently commenced investigation, Panex’s

shareholders voted to dissolve the corporation on September 24,

1984. Panex filed its Certification of Dissolution effecting its

formal dissolution under Delaware law on April 15, 1985. To

facilitate the corporate wind-up, Panex's liquidation plan

created a Stockholder’s Liquidating Trust, which was intended in

part to reduce tax liability arising after dissolution, see City

Investing Co. Liquidating Trust v. Continental Casualty Co.,

(Del. 1993). Panex's former shareholders had

received liquidating distributions totaling $64 million before

the Trust was created. The Trust received $6 million in funding

5 at its inception, and it distributed about $4.5 million to former

shareholders in July 1987 when the statute of limitations had run

on its 1982 and 1983 tax years and there were no other known

Panex liabilities. In all, the shareholder-distributees received

over $68 million in distributions. The defendant-appellees in

this action were among those distributees.

Delaware General Corporation Law § 278 generally establishes

a three-year continuation period, beginning at dissolution, for

dissolved corporations to wind up their affairs and for unknown

claimants to assert claims against the corporation. After this

period, the corporation ceases to exist and lacks capacity to be

sued. The State sent Panex formal notice of its claim for

response costs at the landfill site in March 1988, but Panex did

not receive the notice until April 25, 1988--just over three

years after its dissolution (which occurred on April 15, 1985),

thus just after the wind-up period expired. Upon receipt of this

notice, the trustees of the Panex Trust extended the life of the

Trust and postponed further distributions. For the next several

years, the State conducted investigations at the site and, in

1994, formulated a remediation plan.

After adopting the remediation plan, the State filed this

action in the Western District of New York against Panex, the

Panex Trust, and the purchasers of the Rochester Button assets,

among others, asserting federal claims under CERCLA and nuisance

6 claims under New York law. On behalf of Panex, its trustees

moved to dismiss, arguing that Delaware General Corporation Law §

278 barred all claims against Panex because the suit was filed

more than three years after its dissolution. The district court

dismissed the state-law nuisance claims but denied the motion to

dismiss the CERCLA claims, holding that CERCLA preempted

Delaware's statutory limit on the dissolved corporation's

capacity to be sued.

In March 1997, the costs of defending this and another

CERCLA lawsuit2 had depleted the Panex Trust further, and the

district court granted the State leave to join Panex's

shareholder-distributees as defendants in this action. The State

asserts claims under the common law equitable trust fund

doctrine, which allows claimants against a dissolved or insolvent

corporation to follow the distributed assets of the corporation

into the hands of its shareholders in order to satisfy the

corporation’s liability. See, e.g., Koch v. United States,

(10th Cir. 1943).

2 Panex and the shareholder-distributees were involved in similar environmental litigation in the Virgin Islands, and the Third Circuit concluded that Panex lacked capacity to be sued under Delaware General Corporation Law § 278 and that Delaware General Corporation Law § 325(b) barred suit against the former shareholders. In re Tutu Wells Contamination Litig., No. 95- 7280, slip op. at 9 (3d Cir. Dec. 21, 1995) (noted in table at

). The shareholder-distributees have argued that the State participated in that litigation and is bound by the outcome in that case, but we need not reach that argument to resolve the instant appeal and cross-appeal.

7 Panex's shareholder-distributees moved to dismiss the claims

against them under Federal Rules of Civil Procedure 12(b)(1) and

12(b)(6). The district court granted the motion on October 2,

1997, ruling that the trust fund doctrine did not survive

Delaware's enactment of section 278, which barred the State’s

claims because they were not brought within three years of

Panex’s dissolution. The district court also concluded that the

State’s claim against the shareholder-distributees was premature

on the ground that Delaware General Corporation Law § 325(b)

required the State to obtain a judgment against Panex and the

Panex Trust, and have that judgment returned unsatisfied, before

pursuing recovery from the shareholder-distributees, which the

State had not done. The court rejected the State's argument that

it should adopt the trust fund doctrine as a matter of federal

common law under CERCLA, which would in turn preempt the Delaware

statutes. As a result of this ruling, the shareholder-

distributees were dismissed as defendants.

Seven years later, the district court granted summary

judgment to the State on its CERCLA claims against Panex and the

successor trust that had succeeded the Panex Trust, concluding

that CERCLA preempts the Delaware statutory limits that otherwise

would bar suit against the dissolved corporation. The district

court's judgment held Panex and the successor trust jointly and

severally liable to the State for $4,558,034.83 under CERCLA §

8 107,

, and declared that those entities were

jointly and severally liable for all future response costs

incurred by the State in cleaning up the site under CERCLA §§

113(g)(2). Neither Panex nor the successor trust has any assets

to pay the judgment, so, if the State is going to recover from

anyone, it must be the shareholder-distributees. Thus, the State

appeals the district court's 1997 order dismissing its claims

against the shareholder-distributees. The shareholder-

distributees cross-appeal the 2004 grant of summary judgment

against Panex and the denial of an earlier motion to dismiss the

claims against Panex in light of the State's failure to file suit

within the three-year wind-up period established by Delaware

General Corporation Law § 278.

The State advances four arguments in its appeal:

1. The district court erred in determining that section 278

bars the State’s claim against the shareholder-distributees,

because the common law trust fund doctrine survives enactment of

the statute;

2. The district court erred in holding that the State’s

claims against the shareholder-distributees were premature under

section 325(b) because it had not first obtained an unsatisfied

judgment against Panex;

3. The district court correctly held that CERCLA preempts

any time limits that Delaware General Corporation Law § 278 would

9 place on the State’s claims against Panex, and the court should

have allowed its claims against the shareholder-distributees to

proceed on the same grounds; and

4. The district court erred in refusing to recognize that

the trust fund doctrine applies in any timely-filed CERCLA suit

as a matter of federal common law.

On cross-appeal, the shareholder-distributees argue that the

district court erred in finding that CERCLA preempts Delaware

law's limitation on the dissolved corporation Panex's capacity to

be sued after the expiration of the wind-up period.

II.

The State argues that the district court erred in holding

that Delaware General Corporation Law §§ 278 and 325(b) bar its

claims against the Panex shareholder-distributees. According to

the State, the trust fund doctrine permits claims against

dissolved corporations to go forward with no special time limit,

and sections 278 and 325(b) have no effect upon its continued

relevance. We review the district court's decision to grant the

motion to dismiss de novo. See Cooper v. Parsky,

(2d Cir. 1998).

A. Section 278

We first address the State's argument that the trust fund

doctrine survives enactment of Delaware General Corporation Law,

, allowing its claims against Panex

10 and the shareholder-distributees to proceed even though suit was

filed more than three years after Panex’s dissolution. Under the

common law, dissolution of a corporation terminated its existence

as a legal entity, thus abating all pending actions by and

against it and terminating its capacity to sue or be sued. In re

Citadel Indus., Inc.,

(Del. Ch. 1980). The

trust fund doctrine first arose, in part, to compensate for this

rather harsh rule, giving creditors some protection in the event

of a corporate dissolution. In re RegO Co.,

(Del. Ch. 1992). Essentially, the trust fund doctrine gave

creditors an equitable right to follow corporate assets after

dissolution, such that the assets are held like a trust in which

the creditors have a claim superior to that of the shareholders.

Id.; see also Koch v. United States,

(10th Cir.

1943); Snyder v. Nathan,

(7th Cir. 1965).

Several states have enacted statutes that continue the

existence of corporations for a definite period of time following

dissolution, thereby providing a statutory remedy for the

difficulties associated with the common law abatement rule.

Considering that the equitable remedy arose in order to supply

relief where none existed, it may be argued that adequate

statutory remedies deprive courts of equitable jurisdiction. See

George I. Wallach, Products Liability: A Remedy in Search of a

Defendant-The Effect of a Sale of Assets and Subsequent

11 Dissolution on Product Dissatisfaction Claims,

, 332 (1976). Indeed, several courts construing such statutes

have concluded that the statutory remedies available to creditors

obviate reliance upon equitable remedies, thereby precluding

their use by the courts. See, e.g., Reconstruction Fin. Corp. v.

Teter,

(7th Cir. 1941) (holding that Illinois

statutes "completely regulate and control both the substantive

and procedural rights" of a corporation's creditors); Hunter v.

Fort Worth Capital Corp.,

(Tex. 1981)("The

effect of these statutes was to supplant the equitable trust fund

theory by declaring a statutory equivalent."). But see Green v.

Oilwell, Div. of U.S. Steel Corp.,

(Okla.

1989) (holding that state law did not provide a direct remedy for

creditors and therefore did not displace the trust fund

doctrine). The Delaware Court of Chancery has addressed this

issue briefly, explaining that "the problem that the trust fund

doctrine addresses has been ameliorated by provisions in the

corporate codes of most or all jurisdictions that continue the

existence of the corporation as a jural entity for limited

purposes following dissolution." In re RegO Co.,

.

Delaware's post-dissolution statute, section 278 of the

Delaware General Corporation Law3, was enacted in order "to

3

provides:

All corporations, whether they expire by their own

12 formalize the continued existence of corporate assets and to

provide a mechanism for the assertion of claims as part of the

'winding up' process . . . [continuing] the corporation's

existence by operation of law." City Investing Co. Liquidating

Trust v. Continental Casualty Co.,

(Del.

1993). Like other post-dissolution statutes, section 278

provides that "any suit against the corporation, which was filed

before dissolution or during the three year statutory wind-up

period, does not abate, even on the expiration of the wind-up

period." In re RegO Co.,

. When the wind-up

period expires, however, so does the corporation’s capacity to be

limitation or are otherwise dissolved, shall nevertheless be continued, for the term of 3 years from such expiration or dissolution or for such longer period as the Court of Chancery shall in its discretion direct, bodies corporate for the purpose of prosecuting and defending suits, whether civil, criminal or administrative, by or against them, and of enabling them gradually to settle and close their business, to dispose of and convey their property, to discharge their liabilities and to distribute to their stockholders any remaining assets, but not for the purpose of continuing the business for which the corporation was organized. With respect to any action, suit or proceeding begun by or against the corporation either prior to or within 3 years after the date of its expiration or dissolution, the action shall not abate by reason of the dissolution of the corporation; the corporation shall, solely for the purpose of such action, suit or proceeding, be continued as a body corporate beyond the 3-year period and until any judgments, orders or decrees therein shall be fully executed, without the necessity for any special direction to that effect by the Court of Chancery.

13 sued.

The initial question before this court is whether section

278 supersedes the trust fund doctrine, preventing the State's

claims against Panex’s shareholder-distributees from going

forward because the State filed suit after the expiration of the

three-year wind-up period. The State argues that because section

278 does not explicitly address the remedies available to

creditors against shareholder-distributees, section 278 does not

supersede the trust fund doctrine as to these defendants. The

district court noted, however, that the trust fund doctrine has

never been used by a Delaware law court to circumvent section 278

in any situation. Other courts also have recognized that the

trust fund doctrine has been superseded by wind-up statutes, and

the district court cited three cases to support this proposition:

Pacific Scene, Inc. v. Penasquitos Inc.,

(Cal.

1985); Hunter,

; and Blankenship v. Demmler

Manufacturing Co.,

(Ill. App. Ct. 1980).

The State argues that the district court's reliance upon

these cases is misplaced because they involve statutes that

provide specific statutory remedies against shareholder-

distributees, unlike section 278, thereby limiting their

applicability. Thus, California Corporations Code § 2009

"restored to creditors a direct remedy against the former

shareholders of dissolved corporations," Pacific Scene,

; in Texas, Article 7.12 of the Texas Business Corporation

Act "applies to officers, directors, and shareholders of a

dissolved corporation," Hunter,

; and in

Illinois, the two-year survival statute provided that corporate

dissolution "shall not take away or impair any remedy available

to or against such corporation, its directors, or shareholders"

for claims accruing before dissolution as long as suit was filed

within the two year period, Blankenship,

.

These cases support the conclusion that section 278 applies

to this case. First, in concluding that statutory remedies

supersede the common law trust fund doctrine, all three cases

address as a policy matter the necessity of protecting

shareholders, together with officers and corporations, from

uncertain liability; this reduces the significance of differences

in statutory language. See, e.g., Pacific Scene,

(stating that "shareholders nonetheless possess an important

statutory interest in the final and certain termination of their

involvement with the affairs of a dissolving corporation");

Hunter,

(stating that "Article 7.12 expresses a

legislative policy to restrict the use of the trust fund theory

to pre-dissolution claims, and to protect shareholders, officers

and directors of a dissolved corporation from prolonged and

uncertain liability"). We recognize that shareholders, officers,

and corporations all have an interest in certainty and finality.

15 See 15A William M. Fletcher, Cyclopedia of the Law of

Corporations § 7373 (2006)("The trust fund doctrine is fuzzy;

statutes by contrast are sharp. Accordingly, the adoption of

corporate dissolution statutes has supplanted the equitable trust

theory in most jurisdictions.").

Second, all three cases deal with post-dissolution claims,

so the courts were addressing the availability of the trust fund

doctrine despite statutory schemes that limit remedies to pre-

dissolution claims. The cases question whether to apply the

trust fund doctrine in order to provide extra-statutory remedies,

which explains the emphasis on statutory construction and whether

the statutes regulate corporate liability to the point of

superceding the trust fund doctrine. See, e.g., Hunter,

(stating that "no real purpose would be served by .

. . permitting suits against officers, directors, and

shareholders of a dissolved corporation, unless the legislature

intended for the statute to bar resort to the trust fund theory

apart from the statute in order to enforce post-dissolution

claims. To hold otherwise would violate the rule of statutory

construction that the legislature is never presumed to do a

useless act"); Pacific Scene,

("Courts and

commenters . . . have been troubled by [the] implication that

legislators uselessly created a redundant statutory remedy for a

subclass of claims concurrently remediable in equity.").

16 In contrast, the instant case involves a claim accruing

before Panex’s dissolution. Therefore the key question is not

whether section 278 completely supersedes the trust fund

doctrine, but is instead the narrower question of whether section

278 provides a remedy for the State’s pre-dissolution claim

against Panex’s former shareholders.4 Contrary to the State's

assertions, several Delaware cases suggest that the Court of

Chancery interprets section 278 as applying to claims against

both corporations and shareholders that arise before dissolution.

The court in In re RegO Co. briefly discussed the relationship

between the trust fund doctrine and statutory remedies. It

acknowledged that section 278 addresses the same problems as the

trust fund doctrine, but it also recognized that the "modern

scheme still leaves open the question, what, if any, rights are

afforded to persons who have no claim against a corporation at

the time of its dissolution, or during the statutory wind-up

period, but who do thereafter acquire such a claim." In re RegO

Co.,

. The court concluded that a corporation

could not be liable for a post-dissolution claim, but

4 Even if section 278 does not encompass such a remedy, the differences in statutory language between section 278 and other state statutes are not necessarily dispositive. As the decisions cited by the district court indicate, the important issue is whether the State of Delaware would have enacted section 278 while preserving a subclass of claims, those against shareholders, remediable in equity. We believe it would not.

17 characterized the possibility that shareholders and directors may

be liable as "an unclear and troubling question."

The crucial question for the court in analyzing statutory

remedies was whether a claim arose before or after the statutory

wind-up period, not whether the defendant was the corporation,

the directors, or the shareholders.5 By addressing the fact that

shareholders and directors may be liable in the modern scheme,

but only in the context of post-dissolution claims, the court was

implicitly recognizing that section 278 covers all potential pre-

dissolution claims, regardless of which corporate constituent is

named as the defendant.

Similarly, in In re Citadel Industries, Inc. the court made

no distinction between potential defendants when it analyzed the

5 The distinction between pre-dissolution and post- dissolution claims articulated in In Re RegO Co. also allows us to address another of the State's arguments. Section 282(b) of the Delaware General Corporation Law provides that if a corporation chooses to distribute its assets in accordance with procedures described in section 281(a), then its shareholders "shall not be liable for any claim against the corporation on which an action, suit or proceeding is not begun prior to the expiration of the period described in § 278 of this title." The State argues that this demonstrates the continued vitality of the trust fund doctrine in Delaware law. As the Court of Chancery explained, however, sections 280-282 were passed in order to address the uncertainty associated with dissolving a corporation that faces potential future claimants. See In re RegO Co.,

. In other words, sections 280-282 do not recognize the continued vitality of the trust fund doctrine, but rather foreclose the use of the trust fund doctrine for post-dissolution claims, provided dissolved corporations follow the procedures outlined in section 281(a). This is of no consequence for determining whether section 278 has an effect on the trust fund doctrine's applicability in pre-dissolution claims.

18 expiration of the three-year statutory wind-up period. The court

concluded that the corporation "no longer existed as a body

corporate. It no longer had legal existence as a corporation. .

. . [T]here was no longer a legal entity which could be continued

through its officers, directors and shareholders." In re Citadel

Industries, Inc.,

. The court also expressed

concern that "[o]nce a corporation is dissolved, the following

three-year period run, all known debts paid, [and] all remaining

assets distributed to shareholders, . . . [h]ow can a vast number

of former shareholders be compelled to return any final

distribution of assets, etc.?"

. Although section 278

does not set forth specific remedies against shareholders, there

is evidence that the Court of Chancery considers section 278 to

be a comprehensive statutory remedy available to creditors for

claims against all potential defendants, including the

corporation, its officers, and its shareholders.6

6 The Court of Chancery's decisions in City Investing Co. Liquidating Trust

, and Rosenbloom v. Esso Virgin Islands, Inc.,

(Del. 2000), are of no avail to the State in this case. Neither case makes any reference to the availability of the trust fund doctrine for pursuing claims against dissolved corporations and their former shareholders. City Investing Co. Liquidating Trust held that "a liquidating trust is the successor of the corporation whose assets it administers" and thus subject to creditors' claims despite the corporation's dissolution.

. Similarly, in Rosenbloom, the Court of Chancery recognized that the creation of a successor trust was necessary to preserve the claimants' rights and "complete the winding up process."

. Instead of making any reference to the trust fund doctrine, both cases delineate the role of trusts in Delaware's statutory framework.

19 We are persuaded by the general consensus that modern

statutory remedies have effectively replaced the trust fund

doctrine and that there are sound reasons for abiding by the

wind-up period established by section 278. We therefore conclude

that the district court correctly held that the State's claims

against the shareholder-distributees are barred by section 278.

B. Section 325(b)

Having concluded that section 278 applies to the State's

claims against Panex and its former shareholders, we must

likewise conclude that Delaware General Corporation Law § 325(b)7

bars the State's claims against the shareholder-distributees. As

the State points out, Delaware law requires that for section

325(b) to apply, the defendants must be liable "by the provisions

of this chapter." Del. Code Ann. tit. § 325(a). The State

argues that this precludes application of section 325(b) in this

case because it is a suit arising in equity.

In light of our conclusion that section 278 provides the

only basis for liability, the State's argument must fail.

Indeed, the State has an unchallenged judgment against Panex’s liquidating trust; the difficulty is that its assets are insufficient to satisfy the claim. 7

(b) provides:

No suit shall be brought against any officer, director or stockholder for any debt of a corporation of which such person is an officer, director or stockholder, until judgment be obtained therefor against the corporation and execution thereon returned unsatisfied.

20 Section 278 applies to claims against shareholders that arise

before dissolution, so therefore section 325(b) also applies and

the State must obtain judgment against Panex before pursuing its

claim against the shareholder-distributees.

III.

As we have concluded that the Delaware statutes bar the

State's claims against Panex and its shareholder-distributees, we

turn to the State's argument that those statutes should not apply

in the face of CERCLA. The district court accepted this argument

as to Panex, holding that CERCLA preempted Delaware General

Corporation Law § 278's three-year limitation on Panex's capacity

to be sued. It rejected the argument as to the shareholder-

distributees, however, reasoning that Delaware law controls

because any liability of the shareholder-distributees would arise

from their amenability to suit under Delaware law, not from

CERCLA or federal common law. We hold that CERCLA does not

require displacement of Delaware law in this case, and the suits

against both the shareholder-distributees and Panex are barred.

The State first contends that Delaware law conflicts with

the federal policy expressed in CERCLA, such that the six-year

CERCLA limitations period set forth at

(g)(2)(B)

preempts the three-year corporate wind-up period established by

Delaware General Corporation Law § 278. Alternatively, the State

urges this Court to displace the Delaware statutes and apply the

21 trust fund doctrine as a matter of federal common law in CERCLA

cases, allowing it to pursue Panex assets that have been

distributed to Panex's former shareholders.

We begin with the observation that corporate law is

overwhelmingly the province of the states. See Kamen v. Kemper

Fin. Servs., Inc.,

(1991). The Federal Rules

of Civil Procedure provide that state law governs a corporation's

capacity to be sued, Fed. R. Civ. P. 17(b), and the Supreme Court

has held that "[h]ow long and upon what terms a state-created

corporation may continue to exist is a matter exclusively of

state power," with the federal government "powerless to resurrect

a corporation which the state has put out of existence for all

purposes." Chicago Title & Trust Co. v. Forty-One Thirty-Six

Wilcox Bldg. Corp.,

(1937); see also Melrose

Distillers, Inc. v. United States,

(1959)

(state law determines the question of corporate existence).

Whether framed in terms of conflict preemption or in terms of the

creation of federal common law, the Supreme Court expressly has

cautioned against displacement of state law in areas

traditionally occupied by the states. See, e.g., English v. Gen.

Elec. Co.,

(1990) (warning that preemption of

"areas that have been traditionally occupied by the States" is

inappropriate absent "clear and manifest" congressional intent to

supersede state law); Atherton v. FDIC,

(1997)

22 (stating that Congress legislates against the background of state

law, so a "significant conflict" between federal policy and state

law must be specifically shown before the creation of federal

common law is justified). Keeping these principles in mind, we

address in turn each of the State's arguments for displacement of

Delaware corporate law.

A. Conflict Preemption

To determine whether CERCLA preempts the Delaware statutes,

we must ascertain the intent of Congress. Cal. Fed. Sav. & Loan

Ass'n v. Guerra,

(1987). The Supreme Court has

identified three situations that show congressional intent to

preempt state law: (1) where Congress expressly states its intent

to preempt; (2) where Congress's scheme of federal regulation is

sufficiently comprehensive to give rise to a reasonable inference

that it leaves no room for the state to act; and (3) where state

law actually conflicts with federal law.

. CERCLA

does not expressly state an intent to preempt state law across

the board. See Bedford Affiliates v. Sills,

(2d Cir. 1998). While CERCLA does state that it applies

"[n]otwithstanding any other provision or rule of law,"

(a), this clause refers only to substantive liability and

does not express congressional intent to preempt state rules on

how litigation proceeds, including a party's amenability to suit.

See Citizens Elec. Corp. v. Bituminous Fire & Marine Ins. Co., 68

(7th Cir. 1995). Nor is the CERCLA regulatory

scheme so comprehensive that we reasonably can infer an intent to

preempt; in fact, state corporate law can supplement CERCLA in

several situations. Bedford Affiliates,

.

Our inquiry therefore focuses on the third preemption

scenario, whether Delaware law actually conflicts with CERCLA.

An actual conflict between state and federal law exists when

"compliance with both federal and state regulations is a physical

impossibility," Guerra,

(quoting Fla. Lime &

Avocado Growers, Inc. v. Paul,

(1963)), or

when state law is "an obstacle to the accomplishment and

execution of the full purposes and objectives of Congress,"

Guerra,

(quoting Hines v. Davidowitz,

(1941)). Absent clear congressional intent to the

contrary, federal preemption of state law is not favored,

see English,

, especially in areas of law

traditionally occupied by the states. As we observed above,

corporate law is one of these areas. Kamen,

. For

preemption to occur in this instance, then, the conflict between

state law and federal policy must be a "sharp" one. See Boyle v.

United Tech. Corp.,

(1988).

The "physical impossibility" form of conflict does not exist

here, because it is certainly possible to comply with the CERCLA

limitations period and Delaware's limits on the amenability to

24 suit of dissolved corporations and their shareholder-

distributees. As long as a CERCLA plaintiff files its claim

within three years of the corporation's dissolution as required

by Delaware General Corporation Law § 278, or seeks extension of

the wind-up period from the Court of Chancery within that time as

section 278 allows, it also meets CERCLA's six-year limitations

period,

(g)(2). See Witco Corp. v. Beekhuis,

(3d Cir. 1994) (finding no actual conflict where it

was physically possible for CERCLA claimant to file within three-

year CERCLA limitations period and eight-month period established

by Delaware nonclaim statute, which also contained a mechanism to

preserve contingent CERCLA contribution claims). That a CERCLA

plaintiff, like the State here, might find it impossible to

comply with both statutes in some circumstances is not enough to

establish an actual conflict between the two in this case. See

. The question of preemption thus turns on whether

Delaware law presents an obstacle to the accomplishment of

CERCLA's objectives.

CERCLA manifests Congress's intent that hazardous waste

sites should be cleaned up and that those responsible for the

contamination should bear the costs. Pennsylvania v. Union Gas

Co.,

(1989). To effectuate these goals, CERCLA

looks backward in time and imposes wide-ranging liability. It

allows the government to recover remediation costs directly from

25 parties responsible for contamination,

(a)(4)(A);

it allows private parties to seek indemnification and

contribution for clean-up costs from potentially responsible

parties,

(a)(4)(B); and it imposes strict

liability on owners and operators of contamination sites,

(a)(1). See B.F. Goodrich Co. v. Murtha,

(2d Cir. 1992). Even so, CERCLA's statutory scheme

anticipates that, in some situations, it will be impossible to

recover from responsible parties. See Commander Oil Corp. v.

Barlo Equip. Corp.,

(2d Cir. 2000) ("neither

does CERCLA automatically assign liability to every party with

any connection to a contaminated facility"). In other words,

CERCLA's cost-recovery objective, while strong, is not absolute

and may yield to countervailing considerations. As the Supreme

Court has stated, "there is no federal policy that the fund

should always win," and "'more money' arguments" alone are

insufficient to justify displacement of state law. O'Melveny &

Myers v. FDIC,

(1994) (discussing federal common

law in the context of the Financial Institutions Reform,

Recovery, and Enforcement Act).

We cannot conclude that Delaware law is an obstacle to the

accomplishment of CERCLA's objectives in this instance. The

State's strongest argument on this point is that CERCLA aims to

hold corporations financially responsible for the environmental

26 damage their activities cause, and dismissal of its claims under

Delaware law will require taxpayers to pay for the Rochester

Button cleanup, even though millions of dollars in Panex assets

are traceable to the shareholder-distributees. CERCLA

recognizes, however, that recovery will not always be possible

and manifests no intent that funds that once belonged to a party

responsible for contamination should be frozen indefinitely or

traced infinitely. See Onan Corp. v. Indus. Steel Corp.,

(D. Minn. 1989) (recognizing need to construe

CERCLA broadly to achieve environmental and cost-assignment goals

but stating that CERCLA's reach is "not unlimited"), aff'd,

(8th Cir. 1990). That Delaware law, in affording

dissolved corporations and their shareholders a measure of

finality, operates to leave the State with no source of recovery

in this case amounts to the type of "more money" argument the

Supreme Court rejected in O'Melveny,

. It is not

the sort of sharp conflict between state law and federal policy

that justifies preemption.

The district court documented the disagreement among federal

courts on this issue.8 Many of the cases that hold that CERCLA

8 Several district courts, like the court below, have held that CERCLA preempts state limits on the capacity of dissolved corporations to be sued in light of congressional intent that CERCLA impose broad-ranging liability. See, e.g., United States v. Sharon Steel Corp.,

(D. Utah 1987); BASF Corp. v. Cent. Transp., Inc.,

(E.D. Mich. 1993) (collecting cases); Idylwoods Assocs. v. Mader

27 preempts state limits predate Supreme Court precedent strongly

admonishing courts against displacing state law lightly, see,

e.g., O'Melveny,

. In light of that precedent and

CERCLA's limits we join those Courts of Appeals that have held

that CERCLA does not preempt state statutes that limit a party's

capacity to be sued. See Levin Metals Corp. v. Parr-Richmond

Terminal Co.,

(9th Cir. 1987); Onan Corp. v.

Indus. Steel Corp.,

(D. Minn. 1989), aff'd,

(8th Cir. 1990); Witco Corp. v. Beekhuis,

(3d Cir. 1994).

A conflict could exist if the Delaware statutes would thwart

CERCLA's goals by encouraging corporations responsible for

contamination to dissolve and distribute assets to avoid CERCLA

liability, but this simply is not the case. States have

incentives not to enact laws that would inspire such a "race to

the bottom." Anspec Co. v. Johnson Controls, Inc.,

(6th Cir. 1991) (Kennedy, J., concurring) ("States

have a substantial interest in protecting their citizens and

state resources."). Delaware law protects against this result,

requiring dissolving corporations to provide security that will

be "reasonably likely to be sufficient" to cover claims that have

not been made known to the corporation or that, based on facts

Capital,

(W.D.N.Y. 1996) (collecting cases).

28 known to the corporation, are likely to arise within a period

after dissolution.

(c)(3), 281(b).

See also Bradford C. Mank, Should State Corporate Law Define

Successor Liability?: The Demise of CERCLA's Federal Common Law,

, 1160 (2000) (corporate successor

liability laws of most states "generally prevent corporations

from using sham transactions to escape CERCLA liability"). In

addition, section 278 provides for extension of the wind-up

period beyond three years "as the Court of Chancery shall in its

discretion direct," which could give a potential CERCLA plaintiff

time to investigate the contamination site while preserving its

ability to make a claim against the dissolving corporation.

Finally, we are not persuaded by the district court's rationale

that it is "unlikely that Congress intended that CERCLA treat

differently and inconsistently corporations in identical

positions based upon the state of their incorporation";

preemption is not favored absent clear congressional intent to

the contrary, see English,

, and Congress's

probable desire for uniformity is not enough to justify

displacement of state law where that state law does not actually

conflict with the "clear and manifest purpose of Congress,"

Witco,

. See also New York v. Nat'l Serv. Indus.,

Inc.,

, 208 (2d Cir. 2006).

On a fundamental level, the CERCLA statute of limitations

29 and Delaware's corporate wind-up period serve different purposes,

reinforcing our conclusion that they do not actually conflict.

CERCLA's statute of limitations "extinguishes the right to

prosecute an accrued cause of action after a period of time."

Burlington N. & Santa Fe Ry. Co. v. Poole Chem. Co.,

(5th Cir. 2005). In contrast, Delaware's section 278

defines a dissolved corporation's capacity to be sued, creating a

right for dissolved corporations and their former shareholders to

be free from suit after a period of time. See

As the State

points out, an important goal of CERCLA's statute of limitations

is to allow time for parties to gauge response costs before the

suit is filed, H.R. Rep. No. 99-253, pt. 3, at 20 (1985),

reprinted in 1986 U.S.C.C.A.N. 3038, 3043; thus, the applicable

CERCLA statute of limitations in this case did not begin to run

until "6 years after initiation of physical on-site construction

of the remedial action."

(g)(2)(B). The

Delaware statute limiting capacity to be sued does not

necessarily interfere with this goal, however, because it

provides for potentially indefinite extension of the three-year

wind-up period in the discretion of the Court of Chancery.

. Delaware's statute limits capacity to

be sued, not liability, and thus does not conflict with CERCLA's

statute of limitations--even if in operation the state law

precludes the CERCLA plaintiff from recovering in some

30 circumstances. See Witco,

; Louisiana-Pac. Corp.

v. ASARCO, Inc.,

(9th Cir. 1993); Onan,

; Levin Metals Corp. v. Parr-Richmond Terminal

Co.,

(9th Cir. 1987).

The State analogizes this case to Bedford Affiliates, where

we held that CERCLA preempted state-law claims for restitution

and indemnification.

. In that case, however,

the plaintiffs' state-law claims for restitution and

indemnification stood in the way of CERCLA's objective of

encouraging settlement, which it achieves by restricting the

availability of contribution actions. The restitution and

indemnification claims would have given the plaintiffs an

alternative to the contribution action withheld by CERCLA,which

flies in the face of the federal goal of encouraging settlements.

CERCLA's statute of limitations and Delaware's corporate wind-up

period present no such direct conflict.

In sum, the State has not shown such a conflict between

Delaware law and the congressional policy manifested in CERCLA as

to lead us to conclude that Congress intended to preempt

Delaware's corporate wind-up period, which protects dissolved

corporations' and their former shareholders' interests in

finality. CERCLA does not suggest that "the entire corpus of

state corporation law is to be replaced simply because a

plaintiff's cause of action is based upon a federal statute,"

31 Burks v. Lasker,

(1979), or because it would

net the government more money, O'Melveny,

, which

is essentially all the State has shown here. That is not

sufficient to justify preemption.

B. Federal Common Law

Having held that the CERCLA statute of limitations does not

preempt Delaware law in this instance, we address the State's

argument that we should create a rule of federal common law based

on the equitable trust fund doctrine for CERCLA cases. The

State's proposed rule would displace Delaware law and allow the

State to pursue the assets Panex distributed to its former

shareholders.

The Supreme Court has sharply curtailed the federal courts'

ability to create rules of federal common law. To justify

creation of a rule of federal common law, the State must show

specifically a "significant conflict between some federal policy

or interest and the use of state law." Wallis v. Pan Am.

Petroleum Corp.,

(1966). Cases that call for the

creation of federal common law are "few and restricted," see

Atherton v. FDIC,

(1997), and it is difficult

to prove the need for a federal common law rule, see Atchison,

Topeka & Santa Fe Ry. Co. v. Brown & Bryant, Inc.,

,

364 (9th Cir. 1998). The existence of a complex federal

statutory scheme does not automatically show that Congress

32 intended courts to fill its gaps with rules of federal common

law. Id. at 362 (citing O’Melveny). Rather, where federal

statutory regulation is "comprehensive and detailed," as CERCLA

is, we presume that matters left unaddressed are "left subject to

the disposition provided by state law." O'Melveny,

. We strongly presume that state law should be determinative

where "private parties have entered legal relationships with the

expectation that their rights and obligations would be governed

by state-law standards," as is the case with most corporate

matters. Kamen v. Kemper Fin. Servs., Inc.,

(1991).

While recognizing that Congress intended CERCLA to provide a

sweeping remedy that requires responsible parties to bear the

costs of cleaning up environmental contamination they cause, the

Supreme Court has advised courts not to create CERCLA-specific

rules to displace well-settled state corporate law just because a

case involves CERCLA. United States v. Bestfoods,

(1998). In Bestfoods, the Court refused to adopt a "relaxed,

CERCLA-specific rule of derivative liability that would banish

traditional standards and expectations from the law of CERCLA

liability."

. This Circuit has thus interpreted

Bestfoods as a clear warning against creating CERCLA-specific

federal common law rules. New York v. Nat'l Serv. Indus., Inc.,

(2d Cir. 2003) (holding that Bestfoods required

33 us to overrule B.F. Goodrich v. Betkoski,

(2d Cir.

1996), which had applied a federal common law rule of substantial

continuity for purposes of determining corporate successor

liability under CERCLA). The First Circuit likewise has

interpreted Bestfoods as leaving "little room for the creation of

a federal rule of [corporate] liability" under CERCLA. United

States v. Davis,

(1st Cir. 2001). A split

decision of the Third Circuit, in contrast, read Bestfoods to

favor a uniform federal standard. United States v. Gen. Battery

Corp.,

(3d Cir. 2005), cert. denied, Exide

Tech. v. United States,

(2006).

While Bestfoods stops short of expressly instructing courts

to apply state law, it clearly admonishes courts to refrain from

creating CERCLA-specific rules in the face of applicable long-

standing common law principles. Compare, Gen. Battery Corp.,

, with

(Rendell, J., concurring and

dissenting). As a practical matter, those principles typically

come from state law. See Ronald H. Rosenberg, The Ultimate

Independence of the Federal Courts: Defying the Supreme Court in

the Exercise of Federal Common Law Powers,

,

455 (2004).

Having concluded that we cannot create a rule of federal

common law on the basis of the mere involvement of CERCLA in the

case, we proceed to the traditional analysis. As discussed in

34 Part III.A above, there is no significant conflict between state

law and federal policy in this instance. The absence of such a

conflict weighs heavily against creation of a rule of federal

common law as a threshold matter, because a significant conflict

is "normally a 'precondition'" to the creation of federal common

law. Atherton,

(citing O'Melveny,

).

In light of these principles, it is questionable whether we

even need to entertain the additional considerations set forth in

United States v. Kimbell Foods, Inc.,

(1979), to analyze the State's invitation to apply the trust fund

doctrine as a matter of federal common law, in place of the

Delaware statutes, to permit it to recover CERCLA cleanup costs

from Panex's shareholder-distributees. In Bestfoods, the Supreme

Court flatly rejected the creation of federal common law without

even citing the Kimbell Foods test. See Gen. Battery,

n.19 (Rendell, J., concurring and dissenting).

Nonetheless, the parties have addressed the Kimbell Foods

considerations, and we typically examine them in deciding whether

to draw from state law or to create a rule of federal common law

in cases involving federal programs. See Nat'l Serv. Indus.,

Inc., 460 F.3d at 207. Under Kimbell Foods, we consider: (1) the

"need for a nationally uniform body of law"; (2) "whether

application of state law would frustrate specific objectives of

35 the federal programs"; and (3) "the extent to which application

of a federal rule would disrupt commercial relationships

predicated on state law."

.

First, the State argues that there is a strong need for a

uniform body of federal law, because diverse state rules will

frustrate CERCLA's goals of cleaning up environmental

contamination and making sure that responsible parties, rather

than taxpayers, bear the costs. "Although CERCLA is a federal

statute for which there is presumably an interest in uniform

application, where there is no conflict between federal policy

and the application of state law, 'a mere federal interest in

uniformity is insufficient to justify displacing state law in

favor of a federal common law rule.'" Nat'l Serv. Indus., 460

F.3d at 208. "To invoke the concept of 'uniformity' . . . is not

to prove its need." Atherton,

. The need for

uniformity is weak in this case. No state provides a safe

"haven" for polluters. Atchison, 159 F.3d at 364. The State

expresses concern that states' different corporate wind-up

periods could prove troublesome, but the fifty states' laws on

corporate dissolution are "largely uniform" already. Anspec,

(Kennedy, J., concurring). Cf. Gen. Battery,

. Moreover, variations in rules among states do not

prove a need for uniformity "as long as the applicable standard

is applied evenhandedly to particular disputes." Wilson v. Omaha

36 Indian Tribe,

(1979) (internal punctuation

omitted).

Next, we inquire whether application of state law will

frustrate federal objectives. Our preemption analysis in Part

III.A faced the same question and concluded that Delaware law

does not significantly frustrate the objectives manifested in

CERCLA. We acknowledge that "corporate law's preference for

limited corporate liability is theoretically at odds with

CERCLA's broad remedial goals." Mank, supra, 68 U. Cin. L. Rev.

at 1160. Nonetheless, CERCLA recognizes that recovery will not

always be possible and does not mandate recovery from every

person with any connection to a contaminated site. See Commander

Oil Corp.,

. That the fund would win under the

State's proposed standard is not sufficient to justify adopting a

rule of federal common law to expand the standard of liability

for shareholder-distributees of a dissolved corporation whose

predecessor owned a company that was responsible for

environmental contamination. See O'Melveny,

; see

also Mank, supra, 68 U. Cin. L. Rev. at 1159 (stating that in

O'Melveny the Supreme Court "rejected the view that the

government is entitled to an expansive federal common law

standard just because the government would win more often").

Finally, we inquire whether adoption of the rule the State

proposes as a matter of federal common law would disrupt

37 commercial relationships predicated on state law. Kimbell Foods,

. We easily conclude that it would. The

presumption that state law should be determinative "is

particularly strong in areas in which private parties have

entered legal relationships with the expectation that their

rights and obligations would be governed by state-law standards."

Kamen,

; see also Kimbell Foods,

.

Shareholders have the expectation of limited liability under

state law when they invest in corporations. Delaware General

Corporation Law §§ 278 and 325(b) protect shareholders from

buying into a perpetual threat of liability by providing that,

after a period of time after their corporation dissolves, they no

longer face liability on claims against the dissolved corporation

and are free to conduct their financial affairs. The alternative

would be unworkable. Dissolved corporations might delay

distributions indefinitely, diminishing shareholders' incentive

to invest. If distributions were made, shareholder-distributees

would have to hold onto them, just in case an unknown claim

arises at some point in the future, or else come up with funds to

replace any received distributions that have been expended once

the claim and demand for disgorgement arise. See City of

Philadelphia v. Stepan Chem. Co.,

(E.D.

Pa. 1989); In re Citadel Indus., Inc.,

. Hanging

a cloud of perpetual uncertainty over the former shareholders of

38 dissolved business entities in the name of CERCLA would impair

the finality that allows parties to proceed with confidence into

new transactions. "Major economic decisions, critical to

society, are best made in a climate of relative certainty and

reasonable predictability." Polius v. Clark Equip. Co.,

, 83 (3d Cir. 1986) ("Unforeseeable alterations in successor

liability principles complicate transfers and necessarily

increase transaction costs."); Perry E. Wallace, Jr., Liability

of Corporations and Corporate Officers, Directors, and

Shareholders under Superfund: Should Corporate and Agency Law

Concepts Apply?,

, 842 (1989) (An unchecked

interpretation of CERCLA liability engenders "uncertainties and

fears" that "unnecessarily diminish the affected industries'

contributions to certain basic economic and business functions in

society.").

The absence of a significant conflict between Delaware law

and CERCLA's goals directs our conclusion that we must not create

a federal common law version of the trust fund doctrine in this

case, a conclusion that is reinforced by the weak need for

uniformity and the strong need to protect existing commercial

relationships based on state law. We affirm the district court's

holding that Delaware General Corporation Law §§ 278 and 325(b)

must govern the shareholder-distributees' amenability to the

State's CERCLA claims, and, accordingly, we hold that those

39 claims were properly dismissed.

IV.

Finally, we consider the implications of our holding for the

cross-appeal of the shareholder-distributees, acting as Panex

trustees, of the district court's entry of summary judgment

against Panex on the State's CERCLA claims. In refusing to

dismiss those claims, the district court concluded that CERCLA

preempted Delaware's limits on Panex's capacity to be sued. As

discussed in Part III.A, we reject that finding of preemption and

reverse the judgment of the district court on the CERCLA claims

against Panex.

The shareholder-distributees had suggested that we need not

reach their cross-appeal if we were to uphold the dismissal of

the State's trust fund doctrine claims against them, because as a

practical matter this holding absolves them of liability whether

or not the State has a viable claim against the defunct and

penniless Panex. Nonetheless, the district court's conclusion

that CERCLA preempts Delaware General Corporation Law § 278 for

purposes of the CERCLA claims against Panex is not without

consequence. As a matter of principle, displacement of state law

is not favored under recent Supreme Court precedent, and, as a

matter of practicality, refusal to apply state law in this

instance would have unsettling implications for commercial

relationships as discussed above. The claims against both the

40 shareholder-distributees and Panex should be dismissed as both

lack capacity to be sued under Delaware law.

V.

We affirm the portion of the district court's order

dismissing the State's claims against the shareholder-

distributees. We reverse the denial of the motion to dismiss the

State's CERCLA claims against Panex, as well as the summary

judgment granted against Panex on those claims.

41