The Supreme Court’s Decision in
Burlington Northern & Santa Fe Railway Company v. United States
Speech by John C. Cruden
Acting Assistant Attorney General
Environment and Natural Resources Davison
Before the Environmental Law Institute
May 29, 2009
I want to thank the Environmental Law Institute for holding this timely conference on the Supreme Court’s much talked-about decision in Burlington Northern & Santa Fe Railway v. United States.
My remarks today offer the perspective of the Justice Department’s Environment and Natural Resources Division on the decision. That perspective is informed by the United States’ role not only as an affirmative CERCLA litigant, but also as a defendant in some CERCLA actions. To truly assess the significance of the decision, it is important to view the opinion in the context of the United States’ overall CERCLA practice. I want to say at the outset, however, that the position of the Department of Justice on matters in litigation will be found in the briefs that we file in court, not this speech. We already have filed several briefs discussing Burlington Northern and I am confident that there will be many more such briefs in the near future. This speech is designed to give you our initial views about the decision, but it is not a comment on any litigation.
Let me start by saying that we at the Department of Justice have been litigating CERCLA cases for many years, and we have heard with many new decisions that the world of Superfund enforcement was coming to an end. We heard that about retroactive liability after the district court ruling in United States v. Olin Corporation, about parent liability after the Supreme Court spoke in the Bestfoods case, and about contribution actions after the Supreme Court’s decisions in Aviall and Atlantic Research Corporation. After reviewing some of the recent law firm comments about Burlington Northern, I am reminded of Mark Twain’s famous quote: “The reports of my death are greatly exaggerated.” While Burlington Northern is an important case, and it will be cited often, it clearly is not the death knell of Superfund enforcement. In fact, as I will outline in this speech, it is a decision that arose from an unusual fact pattern. It also is a decision that reaffirms the law of joint and several liability.
Burlington Northern addresses two areas of CERCLA litigation – the standard for what we all call “arranger” or “generator” liability under Section 107(a)(3), and the circumstances under which a party is jointly and severally liable to the United States for response costs. The case involved a Superfund site in Arvin, California. In 1960, Brown & Bryant began operating an agricultural chemical distribution business on its four-acre property. In 1975, Brown & Bryant expanded its operation onto a one-acre parcel that it leased from a predecessor to Burlington Northern & Santa Fe Railway Company. Brown & Bryant stored and distributed various pesticides on the Site, including the herbicide dinoseb, sold by Dow Chemicals, and the pesticides D-D and Nemagon, both sold by Shell. Over the course of Brown & Bryant’s 28 years of operation, 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 groundwater at the Site. Leaks and spills of D-D, which Shell required Brown & Bryant to purchase in bulk, occurred in part during the transfer process from delivery tanker trucks to Brown & Bryant’s bulk storage tank. Investigations of the Site in the 1980s by both the California Department of Toxic Substances Control and the United States Environmental Protection Agency revealed significant contamination of soil and groundwater. A plume of contaminated groundwater located under the Site threatened to leach into an adjacent drinking water supply.Brown & Bryant became insolvent and ceased all operations in 1989. EPA and the State undertook cleanup efforts at the Site and by 1998 had incurred more than $8 million in response costs. The United States and the State sued both Burlington Northern and Shell for cost recovery pursuant to Section 107 of CERCLA.
In its opinion, the Supreme Court first addressed the standard for arranger liability. The Court held that “the word ‘arrange’ implies action directed to a specific purpose” and that, to qualify as an arranger, a party must have intended that “at least a portion of the product be disposed of.” The Court recognized that in some cases – such as those involving transactions whose sole purpose is to get rid of used and no longer useful materials – the party's intent is clear and liability will attach. In cases in which a party's purposes are less than clear, the Court noted that the determination of whether a party is liable as an arranger will require a fact intensive and case specific inquiry that looks beyond the party's characterization of the transaction.
At trial in Burlington Northern, the United States showed that Shell knew some of its product would spill as an incidental part of the delivery process, that Shell continued its relationship with Brown & Bryant despite such knowledge, and that Shell set some requirements for the handling of its product during delivery. The United States believed those facts were sufficient to establish Shell's liability as an arranger because a party should be found to intend the foreseeable consequences of its actions. The Supreme Court disagreed, holding that under the facts of that case – involving spills that were “a peripheral result of the legitimate sale of an unused, useful product” and where the party took numerous steps to reduce the likelihood of spills – knowledge of spills alone was insufficient to prove that the party “planned for” the disposal. Importantly, however, the Court recognized that knowledge that a party's product was likely to be spilled may still be probative of that party's intent to dispose of hazardous substances, even in the case of a useful product.
Any discussion of the significance of Burlington Northern must consider its practical impact and take note of the very unusual facts involved in the case. The disposal in Burlington Northern consisted of unintended leaks and spills of unused pesticide products during their transfer from Shell to Brown & Bryant for eventual use. In addition, as the Supreme Court noted, Shell actually took steps to reduce the likelihood of spills. Very few of the United States’ cases involve these types of facts. As you would imagine, it is somewhat unusual for us to litigate a case in which a party disposed of an unused and useful product. Recognizing this, the Supreme Court noted that the “Court of Appeals acknowledged that Shell did not qualify as a ‘traditional’ arranger.”
In our more traditional cases in which we are pursuing a party that has arranged to get rid of waste or byproducts generated during its business operations, there is little question that the party intended to arrange for disposal. In those cases, it is unlikely that Burlington Northern will have much direct impact. For example, Burlington Northern clearly recognizes liability in the normal disposal cases, such as where a generator has arranged to send “used and no longer useful” hazardous substances to a landfill or unauthorized dump site. Burlington Northern also does not prevent liability for the disposal of a partially useful product. As the Supreme Court stated, to be liable as an arranger, a party need only have intended that “a portion of the product be disposed of.” Burlington Northern will, of course, be directly relevant when there is a sale of an unused and 100% useful product. However, even before Burlington Northern, the United States had to address CERCLA's so-called "useful product" defense in cases involving the alleged sale of hazardous materials that have some value.
With respect to joint and several liability, although we did not prevail, Justice Stevens’ opinion clearly reaffirmed the basic principles that we have advocated for years. My own view is that while the first part of the opinion – on arranger liability – is largely legal, the second part of the opinion – concerning joint and several liability – is predominantly factual. In large part, the Court simply affirmed the factual determination of the district court, applying the clearly erroneous standard of appellate review.
The Court clearly re-affirmed the legal standards for joint and several liability that virtually every court of appeals has applied, and that the United States has advocated in numerous briefs. Moreover, the Supreme Court cited favorably decisions from the First, Third, Fourth, Fifth, Sixth, Eighth, and District of Columbia Circuits concerning the law of joint and several liability. The governing principles, in the Court’s own words, are:
“[A]pportionment is proper when ‘there is a reasonable basis for determining the contribution of each cause to a single harm’” (slip op. at 14).
“Not all harms are capable of apportionment, however, and CERCLA defendants seeking to avoid joint and several liability bear the burden of proving that a reasonable basis for apportionment exists” (slip op. at 14).
“When two or more causes produce a single, indivisible harm, ‘courts have refused to make an arbitrary apportionment for its own sake, and each of the causes is charged with responsibility for the entire harm’” (slip op. at 14).
Finally, “[e]quitable considerations play no role in the apportionment analysis; rather, apportionment is proper only when the evidence supports the divisibility of the damages jointly caused by the PRPs” (slip op. at 15 n.9).
The Court cited the “seminal opinion” of United States v. Chem-Dyne for the proposition, based on the Restatement (Second) of Torts, that multiple tortfeasors each are responsible for an entire harm unless they cause distinct harms or a single harm for which there is a reasonable basis for apportionment of the harm. The Court also noted that the “Chem-Dyne approach has been fully embraced by the Courts of Appeals.” Such opinions include:
O’Neil v. Picillo, in which the First Circuit observed that the “practical effect of placing the burden on defendants has been that responsible parties rarely escape joint and several liability, courts regularly finding that where wastes of varying (and unknown) degrees of toxicity and migratory potential commingle, it simply is impossible to determine the amount of environmental harm caused by each party” (883 F.2d at 178-79).
United States v. Alcan, in which the Third Circuit cited the Restatement in holding that “the burden of proving that the harm is capable of apportionment is placed on the tortfeasor” because “‘[a]s between the proved tortfeasor who has clearly caused some harm, and the entirely innocent plaintiff, any hardship due to lack of evidence as to the extent of the harm should fall upon the former’” (964 F.2d at 269).
United States v. Hercules, in which the Eighth Circuit stated that “proving divisibility is a ‘very difficult proposition,’ and the Restatement recognizes that some harms, ‘by their nature, are normally incapable of any logical, reasonable, or practical division’” (247 F.3d at 717).
As I stated earlier, the facts in Burlington Northern were unusual. The case involved a small number of defendants and only a few relevant products. The Site was relatively small and involved a single operator whose operations were fairly consistent over time. The contamination of primary concern was only a single contaminant plume in the groundwater and, as Burlington Northern’s reply brief to the Supreme Court noted, there were “no synergistic effects at issue” and “the EPA remedy ‘[could] treat all three chemicals with a common process’” (Reply Brief at 20-21). Those facts are not typical of the United States’ CERCLA enforcement cases. Our cases often involve a large number of parties, numerous contaminants of varying levels of toxicity and mobility resulting from the combination of multiple pollutants, and variations in contaminant levels within and among the soil and groundwater at a site. As the trial court in Alcan noted in just such a case:
“Where, as here, hazardous wastes are commingled, apportionment of liability must take into account ‘evidence disclosing the individual and interactive qualities of the substances deposited there.’ Among the matters to be considered are the ‘relative toxicity, migratory potential and synergistic capacity’ of the various substances present. Generally speaking, an inquiry into apportionment of liability is ‘intensely factual.’ Moreover, the burden on the defendant is ‘substantial.’” (892 F. Supp. at 656 (citations omitted)).
The Supreme Court did not change the burden on defendants to prove divisibility. The Court simply found that the record in the case reasonably supported the district court’s apportionment of liability. The Court specifically recognized the district court's "detailed findings" in affirming that court’s apportionment analysis. In fact, the district court made over 500 findings of fact and conclusions of law. Those findings included over 50 addressing the flow of surface water and groundwater at the Site, an additional 50 addressing the handling of materials at the Site, and numerous findings about the transport of hazardous substances once they were spilled and the migration of such contamination into the groundwater. It was against this backdrop that the Supreme Court found the district court's analysis reasonably supported. Every record will be different, and each court faced with a divisibility defense will have to decide whether, in light of all the evidence, the defendant has met its burden of showing that there is a reasonable basis to apportion liability. We think that will continue to be difficult to do in most cases.
Since the Supreme Court affirmed the decision-making process of the district court, I also believe that it is relevant to discuss what that court said about joint and several liability. Here are some relevant statements by the district court:
·“Proving divisibility is a ‘very difficult proposition,’ and the Restatement recognizes that some harms, ‘by their nature, are normally incapable of any logical, reasonable, or practical division’” (# 454 (citations omitted)).
·“The evidence supporting divisibility must be concrete and specific” (# 456).
·“Where causation is unclear, courts should not hasten to ‘split the difference’ in an attempt to achieve equity. ‘If they are in doubt, district courts should not settle on a compromise amount that they think best approximates the relative responsibility of the parties.’ ‘In such circumstances, courts lacking a reasonable basis for dividing causation should avoid apportionment altogether by imposing joint and several liability’” (# 461 (citations omitted)).
Finally, I want to offer three practice pointers.
First, the United States does not always seek joint and several liability in the CERCLA enforcement actions that we bring. When we do, we affirmatively allege joint and several liability in our complaint. However, it always has been the case that we consider an allegation of joint and several liability carefully before filing a complaint and we assert it only when we feel the facts support it. That will continue to be our approach, taking into account the Burlington Northern decision. When we do assert joint and several liability, however, you can be assured that we will vigorously litigate that principle. Moreover, unlike Burlington Northern, in which, due to the unusual procedural posture of the case, we never had the opportunity to offer proof to the district court of the defendants’ joint and several liability, when we make that assertion in future cases, we will be prepared to back up our position with scientific evidence.
Second, even if we do not seek joint and several liability, or a party successfully proves that the harm at a site is divisible, a party found liable still will have to pay its share. In fact, our Division has litigated cases in which a defendant that successfully proved divisibility still ended up paying more after trial than we offered in settlement before litigation. Additionally, at any given site, joint and several liability is not necessarily an all-or-nothing proposition – while the harm at a portion of a site may be divisible, the liable parties still may be held jointly and severally liable for the remainder of the site.
Third, parties involved in settlement negotiations with the Department of Justice or the Environmental Protection Agency should know that, while our settlement offers obviously will be informed by the Burlington Northern decision, our practice will remain essentially the same. We will continue to seek RD/RA consent decrees in which parties agree to perform the response work, consistent with the principles of “enforcement first” and “polluter pays.” At the same time, for settlement purposes, we will continue to make offers that take into account the equities of a case and all other relevant circumstances. If a case does not settle, as I said, the United States’ complaint will allege joint and several liability where appropriate and we will vigorously litigate that proposition as we have in the past.
In conclusion, the rumors of CERCLA’s demise are greatly exaggerated. Since Burlington Northern was decided, I have approved several multi-million dollar CERCLA settlements which were absolutely consistent with the settlements we achieved before this decision. While I am sure that more parties will be raising joint and several liability issues in some of our cases, I am equally confident that we will prevail in those cases in which we seek joint and several liability.