Chapter 4: Waste Management and Pollution Prevention

Waste Management and Pollution Problems p. 310 (5^th edition):

The volume and composition of domestic waste streams have changed dramatically over time. In 1910, Americans produced about a half-pound of municipal garbage per person per day, twice the per capita amount produced in Europe. By 1960, the volume of municipal waste in the United States had grown to more than 88 million tons per year, 2.7 pounds of waste per person per day. Waste volumes continued to grow during the next five decades, reaching approximately 254 million tons in 2007, 4.6 pounds per person per day. Paper and paper products are the largest elements in the municipal waste stream today (32.7%) followed by yard waste (12.8%) and food waste (12.5%); plastics (12.1%), metals (8.2%), rubber, leathers, and textiles (7.6%), wood (5.6%), and glass (5.3%) also occupy prominent portions of the waste stream.

Today most municipal solid waste (54%) is sent to landfills. As environmental standards have tightened, the number of operating landfills that dispose of municipal waste has declined sharply from 8,000 in 1988 to only 1,754 in 2007. Approximately 12.6 percent of the municipal waste stream is incinerated, a substantial decline from the 30 percent of the waste stream combusted in 1960. A total of 87 municipal incinerators have the capacity to generate electricity, while burning a total of up to 95,000 tons of municipal solid waste each day. In recent years there has been a substantial resurgence of recycling and composting of waste. Only 6.4 percent of the municipal waste stream was recycled or composted in 1960, but this number increased to 16.8 percent in 1990 and 33.4 percent in 2007. A total of 63 million tons of waste was recycled  in 2007, nearly double the 34 million tons recycled in 1990. Auto batteries were the materials recycled most regularly (99%), followed by steel cans (64%), yard trimmings (64%) and paper and paperboard (54%).

Waste Management Facilities p. 312 (5^th edition):

While the Superfund program concentrates on remediating contamination problems caused by past waste management practices, contemporary management of hazardous waste is regulated under RCRA. A total of 16,349 entities reported that they generated 46.7 million tons of RCRA hazardous waste in 2007. EPA, The National Biennial RCRA Hazardous Waste Report, at 1-1 (2008). In 2007, 1,395 facilities managed RCRA hazardous wastes, 516 were treatment, storage, or disposal facilities (TSDs); most of these were storage-only facilities. More than 57 percent of this waste was managed by facilities in two states—Texas and Louisiana. Most waste was managed by generators on site. A total of 7.2 million tons of hazardous waste were shipped offsite; 4.1 million tons of this crossed state lines. Id. at 3-1, 4-1.

Deepwell or underground injection is the principal method for managing hazardous waste, accounting for more than 42 percent of waste management in 2007. Only 11.9 million tons of hazardous waste were managed in landfills or surface impoundments in 2007, id. at 2-5; 3 million tons were incinerated. Id.

Regulation of Coal Combustion Waste p. 313 (5^th edition):

In 1980 Congress temporarily exempted ash generated by coal combustion from regulation as hazardous waste under RCRA pending an EPA study that was required to be completed by October 1982.  RCRA §8002(n), 42 U.S.C. 6982(n).  EPA agreed to complete this study only after a lawsuit to require completion of the study was filed in 1991.  After the EPA study finally was completed in March 1999, EPA considered regulating some coal combustion wastes as hazardous under Subtitle C of RCRA because they often contain arsenic, lead, mercury, and selenium and had been the cause of several incidents of environmental damage. However, after vigorous lobbying from the utility industry, EPA decided not to regulate the waste as hazardous.  The issue came to national attention on December 22, 2008 when the collapse of an earthen containment wall at the Kingston powerplant in Tennessee buried 400 acres of eastern Tennessee in coal ash sludge to a depth of six feet. The utility industry generates more than 130 million tons of this ash annually and much of it is stored in unlined ponds at 1,300 sites across the U.S. Shaila Dewan, Hundreds of Coal Ash Dumps, With Virtually No Regulation, N.Y. Times, Jan. 7, 2009, p. A1.  In light of the Tennessee disaster, EPA Administrator Sheila Jackson has pledged that EPA will consider regulating coal combustion waste.

E-waste Recycling Legislation update p. 313 (5^th edition):

By the end of 2008 at least 18 states and New York City had adopted e-waste recycling legislation. Most of these laws require product manufacturers to develop plans to collect and recover their products from consumers.

What Substances Are Solid Wastes? p. 341 (5^th edition):

In October 2008, EPA finally adopted new regulations amending its definition of solid waste to address when recyclable materials are subject to regulation under RCRA.  73 Fed. Reg. 64668 (2008), The regulations create two new exclusions from RCRA -- the "Generator Controlled Exclusion" and the "Transfer-Based Exclusion."  The Generator Controlled Exclusion excludes hazardous secondary materials that are legitimately reclaimed under the control of the generator. The exclusion covers materials generated and reclaimed on-site in the same facility, materials generated in one facility and reclaimed at a different facility under the control of the generator, and materials reclaimed by a contractor who certifies that it retains ownership of and liability for the materials.  The Transfer-Based Exclusion exempts materials transferred to a third party for recycling or reclamation. To qualify for this exclusion generators have to show that the materials will not be part of a sham recycling operation.  This requires a showing: (1) "that hazardous secondary materials being recycled provide a useful contribution to the recycling process or to the product of the recycling process" and (2) "that the product of the recycling process is valuable." Generators also have to (1) consider whether the hazardous secondary material is managed liked a normal feedstock process for the manufacturing process, and (2) to compare concentrations of the hazardous constituents found in the product made from the materials with "concentrations of hazardous constituents in analogous products."

EPA's new regulations defining solid waste are analyzed in depth in Jeffrey M. Gaba, Rethinking Recycling, 38 Envt'l L. 1053 (2008).  Professor Gaba concludes that they perpetuate many of the existing flaws in EPA's approach to regulation of recyclable materials.  He criticizes the regulations as "incoherent," "poorly drafted and confusing," and argues that they "may unnecessarily include materials involved in legitimate recycling within the coverage of Subtitle C."  Reviewing the prior caselaw, Gaba concludes that: (1) "EPA may not regulate materials as solid wastes if they are still part of a continuous process within the generating industry itself" (AMC I and Assn'n of Battery Recyclers) and (2) "once generated, a material may be a statutory solid waste regardless of any subsequent act of recycling" (American Petroleum Institute and AMC II).  Noting that in Safe Food & Fertilizer v. EPA, 350 F.3d 1263 (D.C. Cir. 2003), the court approved EPA's exclusion of certain wastes from Subtitle C regulation based on the agency's assessment of the likely environmental harm, Gaba makes the following proposal. "First, materials that are recycled through activities involving land application or burning of wastes should be classified as solid wastes and subject to Subtitle C regulation" because this activity "involves the same environmental harms as direct disposal of hazardous wastes. Second, EPA should classify all materials involved in 'sham' recycling as solid wastes." Id. at 1100. Gaba proposes that EPA should issue new criteria defining what constitutes "sham recycling," require generators to bear the burden of establishing that recycling of their byproducts is not a sham, and create a set of "safe harbor" provisions for discharging this burden similar to the provisions contained in CERCLA §127's exemption from liability for recycling.

RCRA’s “Imminent Hazard” Provisions p. 366 (5^th edition):

RCRA contains two important provisions that can be used to require responsible parties to bear the cost of cleaning up environmental contamination.  Sections 7003 and 7002(a)(1)(b) of RCRA authorize lawsuits to require the cleanup of wastes that "may present an imminent and substantial endangerment to health or the environment". Section 7003 of RCRA authorizes the EPA Administrator to sue any person "who has contributed or who is contributing" to "handling, storage, treatment, transportation or disposal of any solid waste or hazardous waste" that meets the "imminent and substantial endangerment" standard.  Section 7002(a)(1)(b) authorizes citizen suits to be brought in the same circumstances.  The statute specifies that eligible defendants may include "any past or present generator, past or present transporter, or past or present owner or operator" of a facility that treats, stores, or disposes of waste.  

Prior to the enactment of CERCLA in 1980, §7003 was one of the few authorities available to EPA to respond to dumpsite contamination problems.  As initially enacted in 1976, §7003 imposed liability on a defendant who "is contributing" to contamination that "is presenting" the requisite "imminent and substantial endangerment.”  When Congress amended RCRA in 1980 it changed this language to clarify that it also applied to past actions that "may present" such endangerment, easing the government's burden of proof.  Section 7003 specifies that defendants may be restrained from contributing to endangerment, ordered "to take such other action as may be necessary, or both."  This has enabled the government to obtain a wide range of equitable relief, including orders to defendants to remediate contamination.  See, e.g., United States v. Price, 688 F.2d 204 (3d Cir. 1982).

When it amended RCRA in 1984, Congress added a parallel "imminent hazard" provision to the citizen suit provisions in §7002. Section 7002(a)(1)(b) authorizes "any person" to bring such an "imminent and substantial endangerment" action, tracking the same language used in §7003.  Section 7002 also authorizes courts to issue the same broad equitable relief as §7003.  While courts generally have interpreted §7002(a)(1)(b) coextensively with § 7003, citizens suing under the former must comply with additional requirements before filing suit, such as providing 90 days of notice of intent to sue (unless a violation of RCRA Subtitle C is alleged).  Citizen suits under §7002(a)(1)(b) also can be precluded if federal or state authorities are diligently prosecuting their own enforcement actions or using CERCLA authorities to cleanup the contamination.

Although the "imminent and substantial endangerment" language may suggest that actions under §§ 7003 and 7002(a)(1)(b) can be used only to respond to the most serious contamination problems, courts have broadly interpreted the "imminent hazard" provisions to provide relief in a wide range of situations.  Focusing on the "may present" language, courts have held that these provisions need not be confined to emergency situations.  Proof of actual or immediate harm is not required, but rather only a showing of a risk of harm.  United States v. Waste Industries, Inc., 734 F.2d 159 (4th Cir. 1984); Interfaith Community Organization v. Honeywell Int'l, Inc., 399 F.3d 248 (3d Cir. 2005). To be held liable for "contributing" to waste management that gave rise to the risk of harm, it has been held to be sufficient for a landowner to have been aware of contamination and to have failed to abate it.  United States v. Price, 688 F.2d 204 (3d Cir. 1982).  However, the mere sale of an industrial property that contains asbestos insulation has been held not to be the "handling, storage, treatment, transportation or disposal" of waste that could give rise to liability.  Sycamore Industrial Park Associates v. Ericsson, Inc., 546 F.3d 847 (7th Cir. 2008).

Because §§7002 and 7003 apply to activity involving not only hazardous waste, but also  "any solid waste," they can be used to respond to contamination problems involving non-hazardous wastes and substances that are excluded from CERCLA's coverage. However, in Meghrig v. RFC Western, Inc., 516 U.S. 479 (1966), the Supreme Court unanimously held that § 7002(a) does not authorize a private cause of action under RCRA to recover the costs of past efforts to clean up a petroleum leak because the "may present" and "imminent" language clearly exclude waste that already has been cleaned up.  The plaintiffs had sought to use RCRA, rather than CERCLA, to recover for the clean up costs because petroleum is excluded from the list of hazardous substances whose release or threat or release can trigger CERCLA liability.  Unlike CERCLA, which expressly authorizes contribution actions under §113(f), RCRA has no express contribution provision.  For an argument that RCRA should be construed to authorize an implied private right of action for contribution see Kenneth K. Kilbert, Re-Exploring Contribution Under RCRA's Imminent Hazard Provisions, 87 Neb. L. Rev. 420 (2008).

CERCLA Liability and the AAI Rules p. 380 (5^th edition):

Although the AAI rules resolved some issues, industry officials continue to press for more clarification.  The federal rules also do not protect against liability under state Superfund statutes, which may impose differing requirements.  EPA maintains a website providing information on AAI compliance. http://www.epa.gov/brownfields/regneg.htm.

Arranger (Generator) Liability: The Burlington Northern Decision p. 391 (5^th edition):

Many “arrangers” will be the companies who produced hazardous wastes as a byproduct to their production processes and then arranged for the disposal of that waste.  However, the statutory term extends more broadly – or at least it appears to do so.  For instance, an early influential case is United States v. Aceto Agricultural Chemical Corp., 872 F.2d 1373 (8^th Cir. 1989).  In that case, companies that manufactured pesticides had shipped raw materials to a company, Aidex, which mixed the materials together to produce particular pesticide formulations.  Aidex then shipped the formulated products either back to the manufacturers or on to their customers.  After the government spent $10 million to clean up the Aidex facility, the United States sued the pesticide manufacturers as arrangers.  The district court found that spills of hazardous raw materials were an “inherent part” of the formulation process, and that the manufacturers were aware that they were occurring.  The Eight Circuit concluded that CERCLA was broadly designed to hold those responsible for hazardous wastes to be also responsible for cleaning them up.  It sustained the district court’s finding that the manufacturers were liable as arrangers. 

In tying any particular arranger to a particular release of hazardous substances, the statutory language provides little guidance regarding what the government must prove. Many waste sites end up containing a "chemical soup" which may make it impossible to prove that an arranger's waste is actually part of the contamination. Because the language of section 107(a)(3) refers to “any facility…containing such hazardous substances” (emphasis supplied), the statute suggests that the government has some obligation to prove that the generator’s waste is at a facility where releases occur, but it initially was unclear if this obligation required the government to prove that the generator’s wastes were part of the release in question.

Not all circuit courts followed the approach of Aceto in developing “arranger” liability.  Amcast Industrial Corp. v. Detrex Corp., 2 F.3d 746 (7th Cir. 1993), took a different course.  The case involved a facility run by Elkhart Products Corp., a subsidiary of Amcast, for the manufacture of copper fittings. The manufacturing process involved the use of trichloroethylene (TCE), which Elkhart purchased from Detrex. Some 800 gallons of TCE were discovered in the groundwater adjacent to Elkhart’s plant, and investigation revealed that the truck drivers accidentally spilled TCE on Elkhart’s premises while filling Elkhart’s tanks. Holding that the words “arranged for” “imply intentional action,” the court held that Detrex was not liable because it did not hire the trucker for the purpose of spilling TCE. 3 F.3d at 751.

South Florida Water Management Dist. v. Montalvo, 84 F.3d 402 (11th Cir. 1996) adopted a “totality of the circumstances” approach to arranger liability. It stated that “while factors such as a party’s knowledge (or lack thereof) of the disposal, ownership of the hazardous substances, and intent are relevant to determining whether there has been an ‘arrangement’ for disposal, they are not necessarily determinative of liability in every case.” 84 F.3d at 407. The Third Circuit canvassed the cases on arranger liability and noted that there was wide agreement that the liability determination was fact-sensitive and not dependent upon the defendant’s characterization of the arrangement or transaction—but little agreement beyond these two points. In the end, the Third Circuit concluded: “After carefully examining the language of the statute and considering the standards adopted by other courts, we conclude that the most important factors in determining “arranger liability” are: (1) ownership or possession; and (2) knowledge; or (3) control. Ownership or possession of the hazardous substance must be demonstrated, but this factor alone will not suffice to establish liability. A plaintiff must also demonstrate either control over the process that results in a release of hazardous waste or knowledge that such a release will occur during the process.” Morton Int’l., Inc. v. A.E. Staley Mfg. Co. 343 F.3d 669, 677-678 (3rd Cir. 2003).

A number of courts, including Aceto and Montalvo, have recognized that the seller of a “useful product” is not subject to arranger liability, although they will look through “sham” transactions. In some cases, courts have held that even bona fide sales of spent materials can subject the seller to liability. Noting that “all that is necessary is that the [disposal or] treatment be inherent in the particular arrangement, even though the arranger does not retain control over its details,” the Ninth Circuit held liable a seller of spent automotive batteries to a lead reclamation plant. Catellus Development Corp. v. United States, 34 F.3d 748, 753 (9th Cir. 1994).

In 2009, the United States Supreme Court took up the issue of arranger liability for the very first time. 

Burlington Northern and Sante Fe Railway Co. v.
United States, 129 S.Ct. 1870 (2009)

JUSTICE STEVENS delivered the opinion of the Court.

I

In 1960, Brown & Bryant, Inc. (B&B), began operating an agricultural chemical distribution business, purchasing pesticides and other chemical products from suppliers such as Shell Oil Company (Shell). Using its own equipment, B&B applied its products to customers’ farms. B&B opened its business on a 3.8 acre parcel of former farmland in Arvin, California, and in 1975, expanded operations onto an adjacent .9 acre parcel of land owned jointly by the Atchison, Topeka & Santa Fe Railway Company, and the Southern Pacific Transportation Company (now known respectively as the Burlington Northern and Santa Fe Railway Company and Union Pacific Railroad Company) (Railroads). Both parcels of the Arvin facility were graded toward a sump and drainage pond located on the southeast corner of the primary parcel. Neither the sump nor the drainage pond was lined until 1979, allowing waste water and chemical runoff from the facility to seep into the ground water below.

During its years of operation, B&B stored and distributed various hazardous chemicals on its property. Among these were the herbicide dinoseb, sold by Dow Chemicals, and the pesticides D-D and Nemagon, both sold by Shell. Dinoseb was stored in 55-gallon drums and 5-gallon containers on a concrete slab outside B&B’s warehouse. Nemagon was stored in 30-gallon drums and 5-gallon containers inside the warehouse. Originally, B&B purchased D-D in 55-gallon drums; beginning in the mid-1960’s, however, Shell began requiring its distributors to maintain bulk storage facilities for D-D. From that time onward, B&B purchased D-D in bulk.

When B&B purchased D-D, Shell would arrange for delivery by common carrier, f.o.b. destination.² When the product arrived, it was transferred from tanker trucks to a bulk storage tank located on B&B’s primary parcel. From there, the chemical was transferred to bobtail trucks, nurse tanks, and pull rigs. During each of these transfers leaks and spills could-and often did-occur. Although the common carrier and B&B used buckets to catch spills from hoses and gaskets connecting the tanker trucks to its bulk storage tank, the buckets sometimes overflowed or were knocked over, causing D-D to spill onto the ground during the transfer process.

Aware that spills of D-D were commonplace among its distributors, in the late 1970’s Shell took several steps to encourage the safe handling of its products. Shell provided distributors with detailed safety manuals and instituted a voluntary discount program for distributors that made improvements in their bulk handling and safety facilities. Later, Shell revised its program to require distributors to obtain an inspection by a qualified engineer and provide self-certification of compliance with applicable laws and regulations. B&B’s Arvin facility was inspected twice, and in 1981, B&B certified to Shell that it had made a number of recommended improvements to its facilities.

Despite these improvements, B&B remained a “ ‘[s]loppy’ [o]perator.” Over the course of B&B’s 28 years of operation, delivery spills, equipment failures, and the rinsing of tanks and trucks allowed Nemagon, D-D and dinoseb to seep into the soil and upper levels of ground water of the Arvin facility. In 1983, the California Department of Toxic Substances Control (DTSC) began investigating B&B’s violation of hazardous waste laws, and the United States Environmental Protection Agency (EPA) soon followed suit, discovering significant contamination of soil and ground water. Of particular concern was a plume of contaminated ground water located under the facility that threatened to leach into an adjacent supply of potential drinking water.

Although B&B undertook some efforts at remediation, by 1989 it had become insolvent and ceased all operations. That same year, the Arvin facility was added to the National Priority List, see 54 Fed. Reg. 41027, and subsequently, DTSC and EPA (Governments) exercised their authority under 42 U. S. C. §9604 to undertake cleanup efforts at the site. By 1998, the Governments had spent more than $8 million responding to the site contamination; their costs have continued to accrue.

In 1991, EPA issued an administrative order to the Railroads directing them, as owners of a portion of the property on which the Arvin facility was located, to perform certain remedial tasks in connection with the site. The Railroads did so, incurring expenses of more than $3 million in the process. Seeking to recover at least a portion of their response costs, in 1992 the Railroads brought suit against B&B in the United States District Court for the Eastern District of California. In 1996, that lawsuit was consolidated with two recovery actions brought by DTSC and EPA against Shell and the Railroads.

The District Court conducted a 6-week bench trial in 1999 and four years later entered a judgment in favor of the Governments. In a lengthy order supported by 507 separate findings of fact and conclusions of law, the court held that both the Railroads and Shell were potentially responsible parties (PRPs) under CERCLA-the Railroads because they were owners of a portion of the facility, see 42 U. S. C. §§9607(a)(1)-(2), and Shell because it had “arranged for” the disposal of hazardous substances through its sale and delivery of D-D, see §9607(a)(3).  …

The Court of Appeals acknowledged that Shell did not qualify as a “traditional” arranger under §9607(a)(3), insofar as it had not contracted with B&B to directly dispose of a hazardous waste product. 520 F. 3d 918, 948 (CA9 2008). Nevertheless, the court stated that Shell could still be held liable under a “ ‘broader’ category of arranger liability” if the “disposal of hazardous wastes [wa]s a foreseeable byproduct of, but not the purpose of, the transaction giving rise to” arranger liability. Ibid. Relying on CERCLA’s definition of “disposal,” which covers acts such as “leaking” and “spilling,” 42 U. S. C. §6903(3), the Ninth Circuit concluded that an entity could arrange for “disposal” “even if it did not intend to dispose” of a hazardous substance. 520 F. 3d, at 949.

Applying that theory of arranger liability to the District Court’s findings of fact, the Ninth Circuit held that Shell arranged for the disposal of a hazardous substance through its sale and delivery of D-D:

“Shell arranged for delivery of the substances to the site by its subcontractors; was aware of, and to some degree dictated, the transfer arrangements; knew that some leakage was likely in the transfer process; and provided advice and supervision concerning safe transfer and storage. Disposal of a hazardous substance was thus a necessary part of the sale and delivery process.” Id., at 950.

Under such circumstances, the court concluded, arranger liability was not precluded by the fact that the purpose of Shell’s action had been to transport a useful and previously unused product to B&B for sale.

II

. . .In these cases, it is undisputed that the Railroads qualify as PRPs under both §§9607(a)(1) and 9607(a)(2) because they owned the land leased by B&B at the time of the contamination and continue to own it now. The more difficult question is whether Shell also qualifies as a PRP under §9607(a)(3) by virtue of the circumstances surrounding its sales to B&B.

To determine whether Shell may be held liable as an arranger, we begin with the language of the statute. As relevant here, §9607(a)(3) applies to an entity that “arrange[s] for disposal . . . of hazardous substances.” It is plain from the language of the statute that CERCLA liability would attach under §9607(a)(3) if an entity were to enter into a transaction for the sole purpose of discarding a used and no longer useful hazardous substance. It is similarly clear that an entity could not be held liable as an arranger merely for selling a new and useful product if the purchaser of that product later, and unbeknownst to the seller, disposed of the product in a way that led to contamination. See Freeman v. Glaxo Wellcome, Inc., 189 F. 3d 160, 164 (CA2 1999); Florida Power & Light Co. v. Allis Chalmers Corp., 893 F. 2d 1313, 1318 (CA11 1990). Less clear is the liability attaching to the many permutations of “arrangements” that fall between these two extremes -- cases in which the seller has some knowledge of the buyers’ planned disposal or whose motives for the “sale” of a hazardous substance are less than clear. In such cases, courts have concluded that the determination whether an entity is an arranger requires a fact-intensive inquiry that looks beyond the parties’ characterization of the transaction as a “disposal” or a “sale” and seeks to discern whether the arrangement was one Congress intended to fall within the scope of CERCLA’s strict-liability provisions. See Freeman, 189 F. 3d, at 164; Pneumo Abex Corp. v. High Point, Thomasville & Denton R. Co., 142 F. 3d 769, 775 (CA4 1998) (“ ‘[T]here is no bright line between a sale and a disposal under CERCLA. A party’s responsibility … must by necessity turn on a fact-specific inquiry into the nature of the transaction’ ” (quoting United States v. Petersen Sand & Gravel, 806 F. Supp. 1346, 1354 (ND Ill. 1992))); Florida Power & Light Co., 893 F. 2d, at 1318.

Although we agree that the question whether §9607(a)(3) liability attaches is fact intensive and case specific, such liability may not extend beyond the limits of the statute itself. Because CERCLA does not specifically define what it means to “arrang[e] for” disposal of a hazardous substance, we give the phrase its ordinary meaning. In common parlance, the word “arrange” implies action directed to a specific purpose. See Merriam-Webster’s Collegiate Dictionary 64 (10th ed. 1993) (defining “arrange” as “to make preparations for: plan[;] . . . to bring about an agreement or understanding concerning”). Consequently, under the plain language of the statute, an entity may qualify as an arranger under §9607(a)(3) when it takes intentional steps to dispose of a hazardous substance.

The Governments do not deny that the statute requires an entity to “arrang[e] for” disposal; however, they interpret that phrase by reference to the statutory term “disposal,” which the Act broadly defines as “the discharge, deposit, injection, dumping, spilling, leaking, or placing of any solid waste or hazardous waste into or on any land or water.” 42 U. S. C. §6903(3); see also §9601(29) (adopting the definition of “disposal” contained in the Solid Waste Disposal Act). The Governments assert that by including unintentional acts such as “spilling” and “leaking” in the definition of disposal, Congress intended to impose liability on entities not only when they directly dispose of waste products but also when they engage in legitimate sales of hazardous substances knowing that some disposal may occur as a collateral consequence of the sale itself. Applying that reading of the statute, the Governments contend that Shell arranged for the disposal of D-D within the meaning of §9607(a)(3) by shipping D-D to B&B under conditions it knew would result in the spilling of a portion of the hazardous substance by the purchaser or common carrier. See Brief for United States 24 (“Although the delivery of a useful product was the ultimate purpose of the arrangement, Shell’s continued participation in the delivery, with knowledge that spills and leaks would result, was sufficient to establish Shell’s intent to dispose of hazardous substances”). Because these spills resulted in wasted D-D, a result Shell anticipated, the Governments insist that Shell was properly found to have arranged for the disposal of D-D.

While it is true that in some instances an entity’s knowledge that its product will be leaked, spilled, dumped, or otherwise discarded may provide evidence of the entity’s intent to dispose of its hazardous wastes, knowledge alone is insufficient to prove that an entity “planned for” the disposal, particularly when the disposal occurs as a peripheral result of the legitimate sale of an unused, useful product. In order to qualify as an arranger, Shell must have entered into the sale of D-D with the intention that at least a portion of the product be disposed of during the transfer process by one or more of the methods described in §6903(3). Here, the facts found by the District Court do not support such a conclusion.

Although the evidence adduced at trial showed that Shell was aware that minor, accidental spills occurred during the transfer of D-D from the common carrier to B&B’s bulk storage tanks after the product had arrived at the Arvin facility and had come under B&B’s stewardship, the evidence does not support an inference that Shell intended such spills to occur. To the contrary, the evidence revealed that Shell took numerous steps to encourage its distributors to reduce the likelihood of such spills, providing them with detailed safety manuals, requiring them to maintain adequate storage facilities, and providing discounts for those that took safety precautions. Although Shell’s efforts were less than wholly successful, given these facts, Shell’s mere knowledge that spills and leaks continued to occur is insufficient grounds for concluding that Shell “arranged for” the disposal of D-D within the meaning of §9607(a)(3). Accordingly, we conclude that Shell was not liable as an arranger for the contamination that occurred at B&B’s Arvin facility.

JUSTICE GINSBURG, dissenting.

Although the question is close, I would uphold the determinations of the courts below that Shell qualifies as an arranger within the compass of the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA). See 42 U. S. C. §9607(a)(3). As the facts found by the District Court bear out, Shell “arranged for disposal … of hazardous substances” owned by Shell when the arrangements were made.

In the 1950’s and early 1960’s, Shell shipped most of its products to Brown and Bryant (B&B) in 55-gallon drums, thereby ensuring against spillage or leakage during delivery and transfer. Later, Shell found it economically advantageous, in lieu of shipping in drums, to require B&B to maintain bulk storage facilities for receipt of the chemicals B&B purchased from Shell. By the mid-1960’s, Shell was delivering its chemical to B&B in bulk tank truckloads. Id., at 89a, 115a. As the Court recognizes, “bulk storage of the chemical led to numerous tank failures and spills as the chemical rusted tanks and eroded valves.” Ante, at n. 1.

Shell furthermore specified the equipment to be used in transferring the chemicals from the delivery truck to B&B’s storage tanks. ² In the process, spills and leaks were inevitable, indeed spills occurred every time deliveries were made.  See also App. to Pet. for Cert. in No. 07-1601, pp. 119a-122a (“It is undisputed that spills were inherent in the delivery process that Shell arranged … .”).

Strict, Joint, and Several Liability: The Implications of Burlington Northern p. 401 (5^th edition):

While oil spills typically come from a single source, hazardous substances released at dump sites can be a complex mixture of wastes from many sources, so it was something of an open question how extending Section 311 liability to the typical multi-party CERCLA actions would translate.  All of the early decisions interpreting section 101(32) found that it authorized imposition of strict, joint, and several liability unless a defendant could demonstrate that the harm caused by its wastes is divisible.  United States v. Chem-Dyne Corp., 572 F. Supp. 802 (S.D. Ohio 1983); United States v. Bliss, 667 F. Supp. 1298 (E.D. Mo. 1987). This approach is firmly ground in common law principles reflected in the Second Restatement of Torts.  At common law, a finding of joint and several liability still leaves any tortfeasor who believes he has paid more than his fair share the option of filing a contribution action against the other tortfeasors to recover from them. When all PRPs are solvent, the net effect of this system is to facilitate more expeditious payment of cleanup costs at the front end, leaving the sorting out of relative liability to the backend contribution action.  When PRPs who would otherwise shoulder significant liability are bankrupt or insolvent, however, the effect can be to leave tortfeasors who were relatively minor contributors – but solvent – bearing a far greater share of financial responsibility.  In those cases, a contribution action against an insolvent joint tortfeasor is essentially useless.  The prospect of being responsible for these “orphan shares” of liability has been one of the spurs to litigation over when the liability is properly apportionable, rather than joint and several.  Despite a number of district court and court of appeals decisions on apportionment over the years, the Supreme Court had never weighed in on the issue until the 2009 Burlington Northern decision, below.

The Court of Appeals criticized the evidence on which the District Court’s conclusions rested, finding a lack of sufficient data to establish the precise proportion of contamination that occurred on the relative portions of the Arvin facility and the rate of contamination in the years prior to B&B’s addition of the Railroad parcel. The court noted that neither the duration of the lease nor the size of the leased area alone was a reliable measure of the harm caused by activities on the property owned by the Railroads, and-as the court’s upward adjustment confirmed-the court had relied on estimates rather than specific and detailed records as a basis for its conclusions.

Despite these criticisms, we conclude that the facts contained in the record reasonably supported the apportionment of liability. The District Court’s detailed findings make it abundantly clear that the primary pollution at the Arvin facility was contained in an unlined sump and an unlined pond in the southeastern portion of the facility most distant from the Railroads’ parcel and that the spills of hazardous chemicals that occurred on the Railroad parcel contributed to no more than 10% of the total site contamination, see id., at 247a-248a, some of which did not require remediation. With those background facts in mind, we are persuaded that it was reasonable for the court to use the size of the leased parcel and the duration of the lease as the starting point for its analysis. Although the Court of Appeals faulted the District Court for relying on the “simplest of considerations: percentages of land area, time of ownership, and types of hazardous products,” 520 F. 3d, at 943, these were the same factors the court had earlier acknowledged were relevant to the apportionment analysis. See id., at 936, n.18 (“We of course agree with our sister circuits that, if adequate information is available, divisibility may be established by ‘volumetric, chronological, or other types of evidence,’ including appropriate geographic considerations” (citations omitted)).

The Court of Appeals also criticized the District Court’s assumption that spills of Nemagon and dinoseb were responsible for only two-thirds of the chemical spills requiring remediation, observing that each PRP’s share of the total harm was not necessarily equal to the quantity of pollutants that were deposited on its portion of the total facility. Although the evidence adduced by the parties did not allow the court to calculate precisely the amount of hazardous chemicals contributed by the Railroad parcel to the total site contamination or the exact percentage of harm caused by each chemical, the evidence did show that fewer spills occurred on the Railroad parcel and that of those spills that occurred, not all were carried across the Railroad parcel to the B&B sump and pond from which most of the contamination originated. The fact that no D-D spills on the Railroad parcel required remediation lends strength to the District Court’s conclusion that the Railroad parcel contributed only Nemagon and dinoseb in quantities requiring remediation.

The District Court’s conclusion that those two chemicals accounted for only two-thirds of the contamination requiring remediation finds less support in the record; however, any miscalculation on that point is harmless in light of the District Court’s ultimate allocation of liability, which included a 50% margin of error equal to the 3% reduction in liability the District Court provided based on its assessment of the effect of the Nemagon and dinoseb spills. Had the District Court limited its apportionment calculations to the amount of time the Railroad parcel was in use and the percentage of the facility located on that parcel, it would have assigned the Railroads 9% of the response cost. By including a two-thirds reduction in liability for the Nemagon and dinoseb with a 50% “margin of error,” the District Court reached the same result. Because the District Court’s ultimate allocation of liability is supported by the evidence and comports with the apportionment principles outlined above, we reverse the Court of Appeals’ conclusion that the Railroads are subject to joint and several liability for all response costs arising out of the contamination of the Arvin facility.

JUSTICE GINSBURG, dissenting.

As to apportioning costs, the District Court undertook an heroic labor. The Railroads and Shell, the court noted, had pursued a “‘scorched earth,’ all-or-nothing approach to liability. Neither acknowledged an iota of responsibility … . Neither party offered helpful arguments to apportion liability.” Consequently, the court strived “independently [to] perform [an] equitable apportionment analysis.” Given the party presentation principle basic to our procedural system, it is questionable whether the court should have pursued the matter sua sponte.

The trial court’s mode of procedure, the United States urged before this Court, “deprived the government of a fair opportunity to respond to the court’s theories of apportionment and to rebut their factual underpinnings -- an opportunity the governmen[t] would have had if those theories had been advanced by petitioners themselves.” Brief for United States 41.³ I would return these cases to the District Court to give all parties a fair opportunity to address that court’s endeavor to allocate costs. Because the Court’s disposition precludes that opportunity, I dissent from the Court’s judgment.

NOTES AND QUESTIONS

1. What impact is joint and several liability likely to have on the willingness of defendants to settle? Will it make them more disposed to settlement for fear of being held liable for all the costs and damages or will it make them fight harder to avoid any liability? On the other hand, what effect does the belief that liability may be apportioned by the trial court have on a PRPs inclination to settle for amounts greater than what the PRP considers her fair share? 

2. As the operator of the site and the entity responsible for the actual improper disposal of wastes there, B&B would be expected to bear the greatest share of the clean up costs.  After Burlington Northern, who will bear those costs?  If joint and several liability had been sustained, who would bear them?

3. In Burlington Northern, the Railroads were being held liable as landowners who played no active role in creating the contamination.  In such cases, does the Supreme Court’s opinion suggest that apportionment will become the rule rather than the exception? 

4. Apportionment cases frequently involve arrangers.  How does Burlington Northern apply to them?  Consider, for instance, a case in which, a PRP argued that its waste had been dumped into a distinct area of the site and that a division based on volume of wastes at that location was appropriate. The district court, however, found the PRP had not sustained its burden on proving the first claim, because there just was not sufficient affirmative evidence to confirm it. It failed to sustain its burden on division by volume being appropriate because it had introduced no evidence of the “relative toxicity, migratory potential, degree of migration, and synergistic capacity” of the wastes at the site. New York v. Panex, 2004 WL 3008733 (W.D.N.Y. 2004).  After Burlington Northern, what facts would an arranger need to prove in order to apportion its liability?

Liability: Amount and Allocation p. 411 (5^th edition):

Private cost recovery has historically covered about 70 percent of total program costs, with the rest being covered by general appropriations and the Superfund tax. That tax expired in 1995, however. That fact plus a slow decline in private cost recovery has been drawing down program reserves and starving the program. From 1999 to 2007, EPA’s inflation-adjusted Superfund expenditures declined nearly 30%, from $1.8 billion in 1999 to $1.3 billion in 2007. GAO, Funding and Reported Costs of Enforcement and Administrative Activities (Aug. 17, 2008). (Private commitments for response and remediation averaged another $630 million annually.  EPA, FY 2007 Superfund Annual Report at 2 (April 2008).)  About ¾ of these expenditures go directly to removal and remediation activities, with 13 percent going to enforcement and the remainder for program administration. Id. Cost recoveries through enforcement actionshave returned approximately $30 billion to the Superfund program through 2007.  The vast majority of this amount is EPA’s estimate of the value of clean up commitments made by PRPs – some $22.5 billion.  Id.  The decline in expenditures has resulted in a scaling back of the pace of cleanup. Completed cleanups went from an average of about 75 a year from 1992 to 2000, to 47 in 2001, to an average of 42 annually  during the 2002-2007 period. Margaret Kriz, Superfund Slowdown, National Journal 1623, 1624 (June 1, 2002); GAO (2008).  The annual figure dropped to 24 in 2007. Center for Public Integrity, Superfund Progress Drops Off Under Bush (2007). The public/private balance in the program is also shifting. In fiscal year 2004, EPA reported spending $775 million on site cleanup, investigations, design of cleanup plans, and support for community, state, and local involvement while receiving private commitments of $680 million directed to future cleanup commitments and current cost recoveries. Superfund Annual Report at ii-iii.  

On February 17, 2009, President Obama signed into law the American Recovery and Reinvestment Act of 2009, Pub. L. 111-5, which provides $787 billion in federal spending to stimulate the economy.  In April 2009 EPA Administrator Lisa P. Jackson announced that $528 million of the stimulus money would be used to speed up the cleanup of 50 Superfund sites in 28 states. John M. Broder, WIthout Superfund Tax, Stimulus Aids Cleanup, N.Y. Times, April 25, 2009.

Allocating Liability: The Atlantic Research Decision p. 416 (5^th edition):

CERCLA has consistently been interpreted to impose joint and several liability on four classes of PRPs.  At the same time, there are a variety of means through which PRPs with certain characteristics can avoid being held liable for the entire costs of clean up. An important one for many small contributors is the de minimis settlement provision in section 122(g). Section 122(g) seeks to encourage prompt settlements between the government and PRPs that contributed small amounts of substances whose toxic or other hazardous effects are minimal in comparison with the contributions of others.

Another mechanism available to PRPs to reduce their liability is the statutory contribution provided by Section 113(f).  This section bars contribution from parties who have settled with the government for matters addressed in the settlement. This not only provides a degree of liability certainty for settling PRPs, but it also further enhances EPA’s leverage in negotiating settlements by diminishing the number of PRPs subject to later contribution actions by the nonsettlers. The Section 113(f) action was added to CERCLA in 1986.  It supplements the action that “any person” has always had under Section 107(B) against PRPs for “costs of response incurred … consistent with the national contingency plan.”

With the addition of §113(f), the choice of action – §113(f) or §107 – has significant ramifications.  First, liability is to be “equitably allocated” among PRPs under §113(f) in contrast to the joint and several liability of §107.  Second, the statute of limitations for §107 actions is three to six years from completion of removal work or initiation of remediation work, compared to three years from the date of judgment or settlement under §133(f).  §§113(g)(2) and (g)(3).  Third, “contribution protection” exists only for §113(f) actions.  §113(f)(2).  See Jeffrey M. Gaba, United States v. Atlantic Research:  The Supreme Court Almost Gets It Right, 37 Envtl. Law Rep. 10810, 10811-12 (2007). 

…

Appeals courts allowed PRP’s contribution claims under §113(f) for costs incurred from settlements, government orders or voluntary action, until 2004, when the Supreme Court caught many Superfund observers by surprise when it issued its decision in Cooper Industries, Inc. v. Aviall Services, Inc., 543 U.S. 157.

Cooper Industries sold four aircraft maintenance sites to Aviall Services, who eventually discovered that the ground and groundwater had been contaminated by leaks of petroleum and other hazardous chemicals.  The Texas Natural Resource Conservation Commission instructed Aviall to remediate the site, and threatened, but did not pursue, judicial action if Aviall refused.  After spending at least $5 million in remediation costs, Aviall sued Cooper to recover remediation costs, initially filing claims under both §107(a) and §113(f)(1).  It later dropped the §107 claim.  Ultimately, on motion for summary judgment, the district court dismissed Aviall’s §113(f)(1) claim, holding that the statutory contribution action was only available to persons who had been sued for cost recovery under § 107.  The Fifth Circuit en banc, reversed, finding that §113(f)(1) allowed contribution claims even when the PRP had not been sued. 

The Supreme Court then reversed the Fifth Circuit, holding that PRPs could utilize §113 to seek contribution only after they had been sued for cost recovery.  Speaking for seven justices, Justice Thomas explained:

§113 provides two express avenues for contribution:  §113(f)(1) (“during or following” specified civil actions) and §113(f)(3)(B) (after an administrative or judicially approved settlement that resolves liability to the United States or a State). . . . Notably absent from §113(g)(3) is any provision for starting the limitations period if a judgment or settlement never occurs, as is the case with a purely voluntary cleanup.  The Supreme Court declined to decide whether a person in Aviall’s situation could sue for cost recovery under § 107.  In dissent, Justice Ginsburg argued that such a cost recovery suit should be available to persons who engage in voluntary clean ups. Justice Ginsburg pointed to Key Tronic Corp. v. United States, 511 U.S. 809, 818 (1994) in support of her claim that §107(a)(4)(B) “allows any person who has incurred costs for cleaning up a hazardous waste site to recover all or a portion of those costs from any other person liable under CERCLA.”

In declining to reach the §107 issue, Justice Thomas did note that the First, Second, Third, Fourth, Sixth, Seventh, Tenth, and Eleventh Circuit Courts of Appeals decisions have held that a private party that is itself a PRP may not pursue a section 107(a) action against other PRPs for joint and several liability. One reason for these results is that section 107(a)(4)(B) apparently authorizes a PRP who has incurred costs to recover all of them from the other PRPs, rather than being expressly limited to some appropriate share of those costs. At the same time, however, the circuit decisions that ruled against a § 107 action for a potentially responsible party did so pre-Aviall, when it was widely thought that the §113 contribution action was available to PRPs who cleaned up on a voluntary basis. 

After Aviall, commentators worried that the decision would hamper voluntary clean ups and settlements.  In addition, it spawned renewed interest in the use of §107 as a means of providing equitable sharing of costs.  Three years after the decision, the following case reached the Supreme Court.

United States v. Atlantic Research
127 S. Ct. 2331 (2007)

Justice Thomas delivered the opinion of the Court.

Two provisions of the CERCLA - §§107(a) and 113(f)-allow private parties to recover expenses associated with cleaning up contaminated sites. In this case, we must decide a question left open in Cooper Industries, Inc. v. Aviall Services, Inc.: whether §107(a) provides so-called potentially responsible parties (PRPs), 42 U.S.C. §§9607(a)(1)-(4), with a cause of action to recover costs from other PRPs. We hold that it does.

I A

Courts have frequently grappled with whether and how PRPs may recoup CERCLA-related costs from other PRPs. The questions lie at the intersection of two statutory provisions-CERCLA §§107(a) and 113(f). Section 107(a) defines four categories of PRPs and makes them liable for, among other things:

“(A) all costs of removal or remedial action incurred by the United States Government or a State or an Indian tribe not inconsistent with the national contingency plan; [and]

“(B) any other necessary costs of response incurred by any other person consistent with the national contingency plan.” §9607(a)(4)(A)-(B).

§113(f) authorizes one PRP to sue another for contribution in certain circumstances.. Prior to the advent of §113(f)'s express contribution right, some courts held that §107(a)(4)(B) provided a cause of action for a private party to recover voluntarily incurred response costs and to seek contribution after having been sued. . . .

In Cooper Industries, we held that a private party could seek contribution from other liable parties only after having been sued under §106 or §107(a). This narrower interpretation of §113(f) caused several Courts of Appeals to reconsider whether PRPs have rights under §107(a)(4)(B), an issue we declined to address in Cooper Industries. . . . Today, we resolve this issue.

B

In this case, respondent Atlantic Research leased property at the Shumaker Naval Ammunition Depot, a facility operated by the Department of Defense. At the site, Atlantic Research retrofitted rocket motors for petitioner United States. Using a high-pressure water spray, Atlantic Research removed pieces of propellant from the motors. It then burned the propellant pieces. Some of the resultant wastewater and burned fuel contaminated soil and groundwater at the site.

Atlantic Research cleaned the site at its own expense and then sought to recover some of its costs by suing the United States under both §107(a) and §113(f). After our decision in Cooper Industries foreclosed relief under §113(f) . . . . [t]he United States moved to dismiss, arguing that §107(a) does not allow PRPs (such as Atlantic Research) to recover costs. The District Court granted the motion to dismiss . . . .

The Court of Appeals for the Eighth Circuit reversed. . . . [I]t held that §107(a)(4)(B) provides a cause of action to Atlantic Research. To prevent perceived conflict between §107(a)(4)(B) and §113(f)(1), the Court of Appeals reasoned that PRPs that “have been subject to §§106 or 107 enforcement actions are still required to use §113, thereby ensuring its continued vitality.” We . . .  affirm.

II A

The parties' dispute centers on what “other person[s]” may sue under §107(a)(4)(B). The Government argues that “any other person” refers to any person not identified as a PRP in §§107(a)(1)-(4). . . .  In accord with the Court of Appeals, Atlantic Research believes that subparagraph (B) provides a cause of action to anyone except the United States, a State, or an Indian tribe-the persons listed in subparagraph (A). We agree with Atlantic Research.

Statutes must “be read as a whole.” Applying that maxim, the language of subparagraph (B) can be understood only with reference to subparagraph (A). The provisions are adjacent and have remarkably similar structures. . . . Bolstering the structural link, the text also denotes a relationship between the two provisions. By using the phrase “other necessary costs,” subparagraph (B) refers to and differentiates the relevant costs from those listed in subparagraph (A).

In light of the relationship between the subparagraph, it is natural to read the phrase “any other person” by referring to the immediately preceding subparagraph (A), which permits suit only by the United States, a State, or an Indian tribe. The phrase “any other person” therefore means any person other than those three. See 42 U.S.C. §9601(21) (defining “person” to include the United States and the various States). Consequently, the plain language of subparagraph (B) authorizes cost-recovery actions by any private party, including PRPs. . . .

Moreover, the statute defines PRPs so broadly as to sweep in virtually all persons likely to incur cleanup costs. Hence, if PRPs do not qualify as “any other person” for purposes of §107(a)(4)(B), it is unclear what private party would. . . .

B

The Government also argues that our interpretation will create friction between §107(a) and §113(f), the very harm courts of appeals have previously tried to avoid. In particular, the Government maintains that our interpretation, by offering PRPs a choice between §107(a) and §113(f), effectively allows PRPs to circumvent §113(f)'s shorter statute of limitations. Furthermore, the Government argues, PRPs will eschew equitable apportionment under §113(f) in favor of joint and several liability under §107(a). Finally, the Government contends that our interpretation eviscerates the settlement bar set forth in §113(f)(2).

We have previously recognized that §§107(a) and 113(f) provide two “clearly distinct” remedies. Cooper Industries. “CERCLA provide[s] for a right to cost recovery in certain circumstances, §107(a), and separate rights to contribution in other circumstances, §§113(f)(1), 113(f)(3)(B).” (emphases added). The Government, however, uses the word “contribution” as if it were synonymous with any apportionment of expenses among PRPs. This imprecise usage confuses the complementary yet distinct nature of the rights established in §§107(a) and 113(f).

Section 113(f) explicitly grants PRPs a right to contribution. Contribution is defined as the “tortfeasor's right to collect from others responsible for the same tort after the tortfeasor has paid more than his or her proportionate share, the shares being determined as a percentage of fault.” Nothing in §113(f) suggests that Congress used the term “contribution” in anything other than this traditional sense. The statute authorizes a PRP to seek contribution “during or following” a suit under §106 or §107(a). 42 U.S.C. §9613(f)(1). Thus, §113(f)(1) permits suit before or after the establishment of common liability. In either case, a PRP's right to contribution under §113(f)(1) is contingent upon an inequitable distribution of common liability among liable parties.

By contrast, §107(a) permits recovery of cleanup costs but does not create a right to contribution. A private party may recover under §107(a) without any establishment of liability to a third party. Moreover, §107(a) permits a PRP to recover only the costs it has “incurred” in cleaning up a site. When a party pays to satisfy a settlement agreement or a court judgment, it does not incur its own costs of response. Rather, it reimburses other parties for costs that those parties incurred.

Accordingly, the remedies available in §§107(a) and 113(f) complement each other by providing causes of action “to persons in different procedural circumstances.” Section 113(f)(1) authorizes a contribution action to PRPs with common liability stemming from an action instituted under §106 or §107(a). And §107(a) permits cost recovery (as distinct from contribution) by a private party that has itself incurred cleanup costs. Hence, a PRP that pays money to satisfy a settlement agreement or a court judgment may pursue §113(f) contribution. But by reimbursing response costs paid by other parties, the PRP has not incurred its own costs of response and therefore cannot recover under §107(a). As a result, though eligible to seek contribution under §113(f)(1), the PRP cannot simultaneously seek to recover the same expenses under §107(a). Thus, at least in the case of reimbursement, the PRP cannot choose the 6-year statute of limitations for cost-recovery actions over the shorter limitations period for §113(f) contribution claims.

For similar reasons, a PRP could not avoid §113(f)'s equitable distribution of reimbursement costs among PRPs by instead choosing to impose joint and several liability on another PRP in an action under §107(a). The choice of remedies simply does not exist. In any event, a defendant PRP in such a §107(a) suit could blunt any inequitable distribution of costs by filing a §113(f) counterclaim. Resolution of a §113(f) counter-claim would necessitate the equitable apportionment of costs among the liable parties, including the PRP that filed the §107(a) action. 42 U.S.C. §9613(f)(a) (“In resolving contribution claims, the court may allocate response costs among liable parties using such equitable factors as the court determines are appropriate”).

Finally, permitting PRPs to seek recovery under §107(a) will not eviscerate the settlement bar set forth in §113(f)(2). That provision prohibits §113(f) contribution claims against “[a] person who has resolved its liability to the United States or a State in an administrative or judicially approved settlement ... .” 42 U.S.C. §9613(f)(2). The settlement bar does not by its terms protect against cost-recovery liability under §107(a). For several reasons, we doubt this supposed loophole would discourage settlement. First, as stated above, a defendant PRP may trigger equitable apportionment by filing a §113(f) counterclaim. A district court applying traditional rules of equity would undoubtedly consider any prior settlement as part of the liability calculus. Second, the settlement bar continues to provide significant protection from contribution suits by PRPs that have inequitably reimbursed the costs incurred by another party. Third, settlement carries the inherent benefit of finally resolving liability as to the United States or a State.

III

Because the plain terms of §107(a)(4)(B) allow a PRP to recover costs from other PRPs, the statute provides Atlantic Research with a cause of action. We therefore affirm the judgment of the Court of Appeals.

NOTES AND QUESTIONS

1.  Does Atlantic Research discourage settlement?  The settlement bar under §113(f) prevents non-settling PRPs from filing §113 claims against settling PRPs.  However, §107 claims are not similarly barred, so PRPs who settle with EPA may still find themselves a target for liability claims following Atlantic Research.  See Mark Yeboah, United States v. Atlantic Research:  Of Settlement and Voluntary Incurred Costs, 32 Harvard Envtl. L.R. 279 (2008); Jeffrey M. Gaba, United States v. Atlantic Research:  The Supreme Court Almost Gets It Right, 37 Envtl. Law Rep. 10810, 10815-16 (2007).  In the Atlantic Research opinion, Justice Thomas discounted this possibility, arguing “we doubt this supposed loophole would discourage settlement [because] . . . [a] district court applying traditional rules of equity would undoubtedly consider any prior settlement as part of the liability calculus.”

2.  After this decision, can PRPs recover costs in response to an administrative order or a settlement agreement?  In Aviall, the Court explicitly chose not to consider whether PRPs subject to an administrative order could make a contribution claim under §113(f)(1).  Lower courts have since decided they cannot, since an administrative order is not a civil action.  In Atlantic Research, Justice Thomas presumed settlement agreements would not fall under §107:

§107(a) permits a PRP to recover only the costs it has “incurred” in cleaning up a site.  When a party pays to satisfy a settlement agreement or a court judgment, it does not incur its own costs of response. Rather, it reimburses other parties for costs that those parties incurred. 

It may be that the settling PRP would have to file two claims:  a §113(f)(3)(B) contribution claim for costs reimbursed to the government and a §107 claim for costs directly incurred. 

3.  Three years passed between the Court’s 2004 decision in Aviall, eliminating a cause of action in contribution under §113(f) for PRPs who voluntarily clean up property, and the Court’s 2007 decision in Atlantic Research, finding a direct right of cost recovery for those same PRPs under §107(a)(4)(B).  For PRPs whose voluntary action cases were decided within that time frame may have some interesting procedural hurdles to overcome. For example, in 1991, Beazer East, Inc., after voluntarily starting a remediation process pursuant to an EPA Administrative Order to clean up hazardous waste on its property, sued The Mead Corporation, the former owner, in contribution for investigation and remediation costs.  In 1996, a Pennsylvania District Court found Beazer’s §§ 107 and 113(f) claims duplicative, and dismissed the §107 claim.  The Third Circuit heard oral argument on the case in fall 2004 and in June 2005 rejected the liability allocation by the magistrate judge and remanded for a new allocation proceeding before the District Judge.  However, in December 2004, the Aviall decision came down, which at least implicitly precluded §113(f) suits based on Administrative Orders.  Mead failed to raise this defense in its appeal, waiting until January 2006 to raise its objection to the §113(f) claim.  After filing an interlocutory appeal on the question of the §113(f) claim, the Supreme Court decided Atlantic Research.  Unfortunately for Mead, its appeal came too late:  the Third Circuit found that Mead waived its defense when it failed to raise the issue on its 2005 appeal.  Beazer East, Inc. v. The Mead Corp. (Beazer III), 525 F.3d 255, 258 (3d Cir. 2008).  The legal result could be described as paradoxical:   the §107 claim, under which Mead may have been liable after Atlantic Research, was dismissed, but the §113(f) claim, under which Mead was not liable after Aviall, was upheld.  The question remains:  for cases where a court has (erroneously) dismissed a §107 claim as duplicative with a §113(f) claim, what is the correct outcome when it is later determined that the PRP may only be legally liable under §107?  Should procedure take precedence over equity and fairness?  Other courts have suggested that allowing an amended complaint is the correct recourse. See E.I. DuPont de Nemours & Co. v. United States, 508 F.3d 126, 136 n.6 (3d Cir. 2007); Montville Township v. Woodmont Builders, LLC, 244 Fed. Appx. 514, 518-19 (3d Cir. 2007); ITT Industries, Inc. v. BorgWarner, Inc., 506 F.3d 452, 458 (6th Cir. 2007).

4.  Champion Laboratories, Inc. v. Metex Corp., Civ. No. 02-5284 (WHW), slip op., 2008 WL 1808309 (D. N.J. Apr. 21, 2008), is another recent case affected by the Atlantic Research decision.  Champion and Metex own neighboring sites in New Jersey.  Both sites are contaminated, and the Champion site groundwater contamination was caused by contamination migration from the Metex site.  Champion filed suit against Metex in 2002, but amended pleadings were filed after the Atlantic Research decision in 2007.   Champion seeks contribution under §113(f)(1) and §113(f)(3)(B).  In its 2008 opinion, the district court dismissed Champion’s §113(f)(1) claim because Champion had not been sued under §106 or §107.  Champion’s §113(f)(3)(B) claim was upheld because Champion has entered into settlement with New Jersey DEP.  Metex filed a §113(f)(1) counterclaim, seeking contribution from Champion.  Interestingly, the district court upheld Metex’s counterclaim because Champion’s 107(a) claim against Metex provides the basis for Metex’s §113(f)(1) counterclaim (§113 claims are allowed during or following a §106 or §107(a) civil action).

C & A Carbone, Inc. v. Town of Clarkstown & United Haulers p. 449:

Some municipalities had responded to the local garbage crisis by creating municipal entities to process and dispose of solid waste.  Two counties in New York followed this route.  Aftter the Carbone decision, New York state legislation created the Oneida-Herkimer Solid Waste Management Authority to manage all solid waste within the two counties, pursuant to a management agreement entered into between the counties and the Authority.  Private haulers could still pick up solid waste, but the waste had to be delivered to the Authority, which undertook to build disposal, processing and recycling facilities to handle the waste.  The tipping fees ranged from $86/ton to $172/ton.  Private haulers challenged the arrangement, arguing that it was prohibited by Carbone.    The district court agreed, but the Second Circuit reversed.  The case then reached the Supreme Court. 

United Haulers Ass’n, Inc. v. Oeida-Herkimer Solid Waste
Management Authority, 550 U.S. 330 (2007)

To determine whether a law violates this so-called “dormant” aspect of the Commerce Clause, we first ask whether it discriminates on its face against interstate commerce.   In this context, “‘discrimination’ simply means differential treatment of in-state and out-of-state economic interests that benefits the former and burdens the latter.”   Discriminatory laws motivated by “simple economic protectionism” are subject to a “virtually per se rule of invalidity,” Philadelphia v. New Jersey, which can only be overcome by a showing that the State has no other means to advance a legitimate local purpose.

This Court [in Carbone] struck down the ordinance, holding that it discriminated against interstate commerce by “hoard[ing] solid waste, and the demand to get rid of it, for the benefit of the preferred processing facility.”    The dissent pointed out that all of this Court's local processing cases involved laws that discriminated in favor of private entities, not public ones.   According to the dissent, Clarkstown's ostensibly private transfer station was “essentially a municipal facility,” and this distinction should have saved Clarkstown's ordinance because favoring local government is by its nature different from favoring a particular private company. The majority did not comment on the dissent's public-private distinction.

The parties in this case draw opposite inferences from the majority's silence. The haulers say it proves that the majority agreed with the dissent's characterization of the facility, but thought there was no difference under the dormant Commerce Clause between laws favoring private entities and those favoring public ones. The Counties disagree, arguing that the majority studiously avoided the issue because the facility in Carbone was private, and therefore the question whether public facilities may be favored was not properly  before the Court.

We believe the latter interpretation of Carbone is correct. As the Second Circuit explained, “in Carbone the Justices were divided over the fact of whether the favored facility was public or private, rather than on the import of that distinction.”    The Carbone dissent offered a number of reasons why public entities should be treated differently from private ones under the dormant Commerce Clause. It is hard to suppose that the Carbone majority definitively rejected these arguments without explaining why.

The Carbone majority viewed Clarkstown's flow control ordinance as “just one more instance of local processing requirements that we long have held invalid.”    Id., at 391, 114 S.Ct. 1677.   It then cited six local processing cases, every one of which involved discrimination in favor of private enterprise. The Court's own description of the cases acknowledges that the “offending local laws hoard a local resource-be it meat, shrimp, or milk-for the benefit of local businesses that treat it.”    Id., at 392, 114 S.Ct. 1677 (emphasis added). If the Court were extending this line of local processing cases to cover discrimination in favor of local government, one would expect it to have said so. Cf. United States v. Burr, 25 F. Cas. 55, 165 (No. 14,693) (CC Va. 1807) (Marshall, C.J.) (“[A]n opinion which is to ... establish a principle never before recognized, should be expressed in plain and explicit terms”).

The Carbone majority stated that “[t]he only conceivable distinction ” between the laws in the local processing cases and Clarkstown's flow control ordinance was that Clarkstown's ordinance favored a single local business, rather than a group of them.  . If the Court thought Clarkstown's processing facility was public, that additional distinction was not merely “conceivable”-it was conceived, and discussed at length, by three Justices in dissent.     Carbone cannot be regarded as having decided the public-private question.

The flow control ordinances in this case benefit a clearly public facility, while treating all private companies exactly the same. Because the question is now squarely presented on the facts of the case before us, we decide that such flow control ordinances do not discriminate against interstate commerce for purposes of the dormant Commerce Clause.

Compelling reasons justify treating these laws differently from laws favoring particular private businesses over their competitors…. States and municipalities are not private businesses-far from it. Unlike private enterprise, government is vested with the responsibility of protecting the health, safety, and welfare of its citizens.  These important responsibilities set state and local government apart from a typical private business.

Given these differences, it does not make sense to regard laws favoring local government and laws favoring private industry with equal skepticism. As our local processing cases demonstrate, when a law favors in-state business over out-of-state competition, rigorous scrutiny is appropriate because the law is often the product of “simple economic protectionism.”  Laws favoring local government, by contrast, may be directed toward any number of legitimate goals unrelated to protectionism. Here the flow control ordinances enable the Counties to pursue particular policies with respect to the handling and treatment of waste generated in the Counties, while allocating the costs of those policies on citizens and businesses according to the volume of waste they generate.

The contrary approach of treating public and private entities the same under the dormant Commerce Clause would lead to unprecedented and unbounded interference by the courts with state and local government. The dormant Commerce Clause is not a roving license for federal courts to decide what activities are appropriate for state and local government to undertake, and what activities must be the province of private market competition.

We should be particularly hesitant to interfere with the Counties' efforts under the guise of the Commerce Clause because “[w]aste disposal is both typically and traditionally a local government function.” 

Finally, it bears mentioning that the most palpable harm imposed by the ordinances-more expensive trash removal-is likely to fall upon the very people who voted for the laws. Our dormant Commerce Clause cases often find discrimination when a State shifts the costs of regulation to other States, because when “the burden of state regulation falls on interests outside the state, it is unlikely to be alleviated by the operation of those political restraints normally exerted when interests within the state are affected.”  Here, the citizens and businesses of the Counties bear the costs of the ordinances. There is no reason to step in and hand local businesses a victory they could not obtain through the political process.

We hold that the Counties' flow control ordinances, which treat in-state private business interests exactly the same as out-of-state ones, do not “discriminate against interstate commerce” for purposes of the dormant Commerce Clause.

The Counties' flow control ordinances are properly analyzed under the test set forth in Pike v. Bruce Church, which is reserved for laws “directed to legitimate local concerns, with effects upon interstate commerce that are only incidental.”   Philadelphia v. New Jersey.  Under the Pike test, we will uphold a nondiscriminatory statute like this one “unless the burden imposed on [interstate] commerce is clearly excessive in relation to the putative local benefits.”   

After years of discovery, both the Magistrate Judge and the District Court could not detect any disparate impact on out-of-state as opposed to in-state businesses. The Second Circuit alluded to, but did not endorse, a “rather abstract harm” that may exist because “the Counties' flow control ordinances have removed the waste generated in Oneida and Herkimer Counties from the national marketplace for waste processing services.”  We find it unnecessary to decide whether the ordinances impose any incidental burden on interstate commerce because any arguable burden does not exceed the public benefits of the ordinances.

The ordinances give the Counties a convenient and effective way to finance their integrated package of waste-disposal services. While “revenue generation is not a local interest that can justify discrimination against interstate commerce,”  Carbone (emphasis added), we think it is a cognizable benefit for purposes of the Pike test.

At the same time, the ordinances are more than financing tools. They increase recycling in at least two ways, conferring significant health and environmental benefits upon the citizens of the Counties. First, they create enhanced incentives for recycling and proper disposal of other kinds of waste. Solid waste disposal is expensive in Oneida-Herkimer, but the Counties accept recyclables and many forms of hazardous waste for free, effectively encouraging their citizens to sort their own trash. Second, by requiring all waste to be deposited at Authority facilities, the Counties have markedly increased their ability to enforce recycling laws. If the haulers could take waste to any disposal site, achieving an equal level of enforcement would be much more costly, if not impossible. For these reasons, any arguable burden the ordinances impose on interstate commerce does not exceed their public benefits.

NOTES AND QUESTIONS

1. In Lebanon Farms Disposal, Inc. v. County of Lebanon, 538 F.3d 241 (3d Cir. 2008), the Third Circuit relied on United Haulers in refusing to invalidate a flow control ordinance requiring waste to be disposed at a landfill owed by a municipal waste authority.  Pursuant to United Haulers, the court noted that a flow control ordinance that benefits a "clearly public facility" and that treats "private business interests exactly the same as out-of-state ones," does not discriminate against interest commerce.  The court remanded the case to the district court to apply the Pike v. Bruce Church balancing test comparing the local benefits to the burdens on interstate commerce.

Yucca Mountain p. 463 (5^th edition):

Consistent with his campaign promise, President Obama has sharply scaled back funding for the Yucca Mountain project in his first budget.  This action has been interpreted by some as sounding the death knell for a project on which almost $8 billion has been spent by the federal government.  If the Yucca Mountain project is killed, what should be done with the volumes of high-level radioactive waste accumulating at nuclear power plants?  In the absence of a permanent repository, should any new nuclear power plants be licensed?