Superfund Cleanup
()

What happens to a Superfund contribution claim when the contribution plaintiff—the party that spent money to clean up or to reimburse the government—collects on its insurance? Superfund practitioners have often ignored this question, but the traction it receives and the confusion it causes when someone raises it suggest that careful lawyers ought to attend to it.

Section 113 of the Comprehensive Environmental Response, Compensation and Liability Act, or CERCLA, 42 U.S.C. § 9613, provides the principal mechanisms through which responsible parties may reallocate among themselves the costs of responding to releases of hazardous substances. Section 113(f)(1) allows a contribution claim to be brought “during or after” any enforcement claim brought against the contribution plaintiff. Alternatively, Section 113(f)(3)(B) allows any person who has resolved any part of his or her responsibility to the United States or a state in a settlement approved by a court or an administrative agency to bring a contribution claim.

The Pennsylvania Hazardous Sites Cleanup Act contains similar provisions in Section 705, 35 Pa. Stat. Ann. § 6020.705. Section 705(a) parallels CERCLA Section 113(f)(1), and Section 705(c)(2) parallels CERCLA Section 113(f)(3)(B).

At common law, the “collateral source rule directs that payments from a collateral source shall not diminish the damages otherwise recoverable from the wrongdoer,” as in Tannenbaum v. Nationwide Insurance, 992 A.2d 859, 863 n.4 (Pa. 2010). So, a tort plaintiff that recovers from its insurer typically recovers in full from the tortfeasor without any set-off of the insurance recovery. The innocent injured party should recover twice before a tortfeasor should escape the full cost of its actions by taking advantage of its victim’s insurance, per the Restatement (Second) of Torts § 920A(2), Comment (b).

However, those federal courts that have rendered opinions have suggested that the collateral source rule does not apply to a CERCLA contribution action. The authorities are a little thin. Only one court of appeals has addressed the issue, and it held that a plaintiff who admitted to a complete recovery from insurance and other settlements could not recover further from another contribution defendant, in Friedland v. TIC-The Industrial Co., 566 F.3d 1203 (10th Cir. 2009).

Nevertheless, district court decisions have begun to accumulate. While they often state that the collateral source rule does not apply, the courts primarily apply an equitable analysis to avoid an unfair double recovery by a contribution plaintiff that has already received a reimbursement from its insurers. Among such decisions are American Premier Underwriters v. General Electric, No. 1:05-cv-437 (S.D. Ohio Mar. 15, 2013); Yankee Gas Services v. UGI Utilities, 852 F. Supp. 2d 229 (D. Conn. 2012); New York State Electric & Gas v. FirstEnergy, 808 F. Supp.2d 417 (N.D.N.Y. 2011); United Alloys v. Baker, 797 F. Supp. 2d 974 (C.D. Cal. 2011); Appleton Papers v. George A. Whiting Paper, 776 F. Supp. 2d 857 (E.D. Wis. 2011), appeal pending, No. 13-2447 (7th Cir. argued Feb. 28, 2014); Basic Management v. United States, 569 F. Supp. 2d 1106 (D. Nev. 2008); Raytheon Aircraft v. United States, 66 Env’t Rep. Cas. (BNA) 1758 (D. Kan. Dec. 8, 2007); and Vine Street v. Keeling, 460 F. Supp. 2d 728 (E.D. Tex. 2006).

Notice that these cases are all decided in the last eight years, and there are fewer than a dozen of them. Now consider the hundreds of CERCLA contribution cases that have been decided with no consideration at all of setting off any party’s insurance or other indemnification recoveries. It is hard to call the inapplicability of the collateral source rule settled, even if the reported opinions seem to line up.

Even assuming that it were resolved that insurance payments ought to be taken into account in contribution claims under CERCLA, how that accounting would work remains uncertain. Some advocate mechanical offset rules. As discussed below, making mechanical rules work fairly presents challenges when the liability of the insured is not determined all at once but often unfolds over years or decades.

More importantly, the statute provides specifically nonmechanical direction to courts on how they ought to arrive at allocations of Superfund costs: “In resolving contribution claims, the court may allocate response costs among liable parties using such equitable factors as the court determines are appropriate.” That is, the court in each case should do what is fair. One might argue that the court ought to do what is fair, taking evidence of insurance and insurance coverage settlement payments into account, or not, depending on whether the collateral source rule applies.

Mechanical rules seem easy if you think of a typical automobile body damage insurance claim. If an insurer accepts coverage and pays bills as they come due, and if the collateral source rule in fact does not apply, then the insured cannot be said to have incurred a response cost if the cost was paid by its insurer. So, if an insurer pays the March cleanup contractor’s bill directly, the insured probably cannot seek contribution for the costs on that bill, assuming the collateral source rule does not apply.

A more complicated problem arises when the insurer does not accept coverage and then settles a coverage dispute. In that case, the insured—the contribution plaintiff—will receive a check for costs already incurred. Some of those costs may properly be the subject of a CERCLA contribution claim and some may not. Friedland addressed the assertion by Robert M. Friedland that some of the insurance coverage settlement payments were on account of his attorney fees, and therefore they should not be set off from his contribution recovery. Attorney fees would typically not be recoverable as CERCLA costs of response, per Key Tronic v. United States, 511 U.S. 809 (1994). The U.S. Court of Appeals for the Tenth Circuit held that if Friedland and his insurers wanted to allocate some of the coverage settlements to “defense” rather than “indemnity,” they should have done so explicitly, and it set the insurance settlements off against the contribution recovery.

Friedland was easy—the contribution claim was brought after Friedland had resolved all of his liability to the United States. His exposure was fixed, and it was just a matter of toting up his contribution and insurance coverage settlements to make sure that he did not make a profit on the case.

In a more typical contested coverage situation, the insurer ultimately resolves its liability and the resolution may come before the insured has resolved its liability to the government. It may come before the insured even knows the total costs of response or, in a matter with more than one responsible party, how much of the total the insured will have to pay.

In those cases—often resulting in insurance coverage settlements for lump-sum policy buybacks or site releases—the insurer and the insured intend the lump sum to cover the expected share of costs that the insured will ultimately have to pay, discounted by the litigation risks of the underlying CERCLA case and the coverage dispute, plus an amount to cover any non-CERCLA costs that the insurer might have to pay (like defense costs or personal injury claims).

Thus, the very nature of that settlement suggests that none of the insurance coverage settlement proceeds reimbursed the share of liability properly assigned to anyone other than the insured; none of it covers the portion of any payments the insured might have made that it can get back in a contribution case. So, there should be no offsets.

But if that settlement is being considered in a contribution case, then it must have been arrived at before anyone knew the insured’s fair share of the site. Moreover, it may have been impossible at the time of settlement to allocate the insurance settlement to any specific costs or risks; it is just a check for a release. What if the insurer paid too much? Should there be no offset? What if it paid too little? Should there be any offset at all? And then there is the issue of whether the insurer retained any subrogation rights. Should insurance be treated differently from any other contractual indemnification?

These complexities suggest that a more nonmechanical, equitable approach would be better. In any event, each party’s insurance and indemnification coverage for CERCLA costs could become relevant in a contribution case. Careful lawyers will take discovery and plan cases with that in mind.  

David G. Mandelbaum is national co-chair of the environmental practice group of Greenberg Traurig. His principal office is in Philadelphia. He teaches “Oil and Gas Law” and “Environmental Litigation: Superfund” in rotation at Temple University’s Beasley School of Law, and serves as vice chair of the Pennsylvania Statewide Water Resources Committee. Follow him on Twitter @mandelbaumd.