Opinion
No. 18-1095
12-13-2018
In re: SETTLEMENT FACILITY DOW CORNING TRUST, Debtor. DOW SILICONES CORPORATION; DEBTOR'S REPRESENTATIVES, Interested Parties-Appellants, v. FINANCE COMMITTEE; CLAIMANTS' ADVISORY COMMITTEE, Interested Parties-Appellees.
NOT RECOMMENDED FOR PUBLICATION
File Name: 18a0619n.06 ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF MICHIGAN OPINION BEFORE: ROGERS, STRANCH, and THAPAR, Circuit Judges
ROGERS, Circuit Judge. In the mid-1990s, Dow was facing potential mass-tort liability in the several billions of dollars stemming from injuries allegedly caused by breast implants it had manufactured. Dow and representatives of the tort claimants eventually agreed on a Chapter 11 plan of reorganization creating a nearly two-billion-dollar settlement facility to pay out breast-implant claims. The agreement governing that facility provides for guaranteed base payments (or first-priority payments) made as claims are proven, and for potential supplemental payments (or second-priority payments) made down the road, if funds allow. Because the settlement facility is set to wrap up in June 2019, with an estimated surplus of over $150 million after paying all projected first-priority claims, the Finance Committee running the facility sought approval from the district court to pay fifty percent of second-priority payments before all first-priority claims had been made. As interpreted by an earlier panel of this court, the facility agreement allows the district court to authorize early second-priority payments so long as all first-priority payments are "virtually guaranteed." Based on projections by the Independent Assessor (in essence, the facility's financial consulting firm), and after hearing competing expert testimony on the methodology used to calculate those projections, the district court found that all first-priority payments were virtually guaranteed and authorized the requested second-priority payments. This determination was not clearly erroneous and accordingly must be upheld, despite Dow's arguments below and here that a virtual guarantee has not been shown.
Dow Corning Corporation changed its name to Dow Silicones Corporation effective February 1, 2018, so for convenience, we refer to appellants as Dow throughout.
All dollar amounts discussed in this opinion are used in terms of net present value, which the plan of reorganization determines as of the date the plan went into effect, using a discount rate of 7% compounded annually. Thus, in terms of today's dollars, the figures are much larger.
I.
The plan of reorganization, effective June 2004, gave claimants the option of settling their breast-implant claims through a $1.95 billion settlement facility or litigating them against a $400 million litigation facility. The settlement facility is open to any of the more than 100,000 people who submitted a bare-bones proof of claim during the bankruptcy proceedings, and allows claimants to submit claims over a sixteen-year period ending June 2019 (with interim claim-specific deadlines along the way). Claimants who choose to settle can seek up to three types of base compensation: explant benefits to offset the cost of removing a Dow breast implant; rupture benefits to compensate for a Dow implant that ruptured while implanted; and either expedited release benefits, which provide a fixed payment for any claimant who used a Dow implant (and forgoes disease benefits), or disease benefits, which provide scaled amounts to claimants who used a Dow implant and can show a qualifying disease. These are all first-priority payments.
Second-priority payments take several forms as well. Premium payments provide recipients of first-priority disease or rupture benefits an additional payment worth a percentage of their claim, increased severity payments provide increased amounts for disease claimants whose condition worsens after an initial base payment, and class 16 payments reimburse Dow Chemical for settlement amounts it paid before the settlement facility kicked off.
To ensure that all successful claimants received at least their base (first-priority) payments, the settlement agreement requires as a default that all base payments are made before any second-priority payments are disbursed. The Finance Committee overseeing the facility may, however, seek authorization from the district court to distribute second-priority payments early by filing a motion. That motion must be supported by a detailed accounting of the status of distributions, including an "accounting of pending claims and projections and analysis of the cost of resolution of such pending Claims as described in [the section of the agreement governing the Independent Assessor's quarterly reports]." All parties, including Dow, are provided an "opportunity to be heard" on the motion. After that opportunity, the district court may authorize the requested second-priority payments so long as "adequate provision has been made to assure" payment of all first-priority claims.
In 2011, the Finance Committee sought authorization to make fifty percent of historical and future premium payments (a class of second-priority payments). In support of its motion, the Finance Committee relied on projections by the Independent Assessor estimating that the facility would have a surplus of $68 million after paying all pending and projected first-priority claims and fifty percent of premium payments. The district court authorized the payment but, in doing so, read the agreement to require only "adequate assurance"—something akin to a strong likelihood—that first-priority payments would not be jeopardized by the requested disbursement. See In re Settlement Facility Dow Corning Tr., 2013 WL 6884990, at *7 (E.D. Mich. Dec. 31, 2013). This court reversed on appeal and held that the agreement requires a "virtual guarantee" (as opposed to merely "adequate assurance") that higher priority payments will be made in order to authorize lower priority payments. See In re Settlement Facility Dow Corning Tr., 592 F. App'x 473, 478-80 (6th Cir. 2015). "[T]his standard does not require absolute certainty, [but] it is nonetheless stricter than the 'strong likelihood' or 'more probable than not' levels of confidence that describe 'adequate assurance.'" Id. at 480. This court also held that it was error not to consider Dow's expert reports and testimony criticizing the methodology underlying the Independent Assessor's projections. See id. at 480-81.
Both the district court and this court denied motions to stay distribution of payments pending appeal, and the facility paid out a substantial portion of the requested premium payments before this court reversed. --------
Rather than relitigate the prior authorization motion on remand, the Finance Committee filed a new motion in December 2016 for authorization to pay fifty percent of all outstanding and future second-priority payments. The motion was premised on the Independent Assessor's 2016 report, which projected a remaining cushion of $100.4 million even after all first- and requested second-priority payments were made. Those projections were based on extrapolations from claims processing and payment history over the life of the facility using a series of what the district court found to be conservative assumptions about the filing and success rates of future claims.
As of the time of the Independent Assessor's 2016 report, there were about 72,000 domestic breast-implant claimants who remained eligible to file claims for first-priority disease or expedited release benefits. There were also about 17,000 people with pending first-priority claims with various deficiencies that might be cured with additional evidence. Extrapolating from historical filing and cure rates, the Independent Assessor projected that slightly less than 3% of the 72,000 remaining claimants would file successful claims for disease benefits and that about 2% of the 17,000 claimants would cure the deficiencies in their pending claims. If those estimates held up, the report concluded that all future first-priority claims would cost about $65 million, leaving $159 million in the facility. The contemplated fifty percent of second-priority payments are comprised of several different types of payments, some of which are fixed and some of which can only be projected. The Independent Assessor estimated that the requested second-priority payments could be made with a sizable cushion of about $100.4 million left over.
Before the district court, Dow introduced expert testimony challenging the Independent Assessor's methodology and argued that the projections did not reach a virtual guarantee that first-priority payments would be funded. Dow's expert testified that if 12-17%, rather than the projected 3%, of the 72,000 remaining eligible claimants filed successful claims, then liability would exceed the $100.4 million cushion and some first-priority payments would not be made. Thus, Dow argued, without any sort of statistical risk analysis quantifying the probability that the larger percentage of claimants would file and thereby jeopardize first-priority payments, first-priority payments were not virtually guaranteed. According to Dow, the Independent Assessor's methodology could not reach that level of confidence because it is based only on historical filing experience and assumptions about future filings rather than any epidemiological data or other study of the likely merit of outstanding claims. Finally, Dow argued that there was a real possibility that the impending closure of the facility would result in a massive spike in the number of claims, which was unsatisfactorily accounted for by the projections.
In response, the Finance Committee and the Claimants' Advisory Committee (which has taken up the cause on appeal) stressed two critical features supporting the accuracy of the Independent Assessor's projections. First, they emphasized that the Independent Assessor's quarterly projections over the life of the facility had always relied on conservative assumptions and so consistently overestimated the number of successful first-priority disease claims. Second, they noted that the projections assumed a higher rate of future claims than were actually filed during the wind down of a very similar breast-implant settlement program funded by other manufacturers—accounting for the likely spike that Dow was concerned about. In response to Dow's other methodological criticisms, the Claimants' Advisory Committee introduced expert testimony that the projections used industry-standard methodologies and that an epidemiological study of the kind suggested would not be possible in the context of a closed-claimant universe like this one. At bottom, the Claimants' Advisory Committee argued that the $100.4 million projected cushion provided such a large margin for error that there was no reasonable possibility that first-priority payments would be jeopardized.
After a hearing, the district court found that the Independent Assessor's report shows to a virtual guarantee that the settlement facility can afford the fifty percent disbursement of second-priority payments without jeopardizing future first-priority payments. Quoting the report, the court concluded that the Independent Assessor "has utilized conventional statistical and actuarial techniques to estimate the number, dollar amount and timing of these liabilities," has "relie[d] heavily on the Trust's historical experience to determine many of the components of this analysis," and "has scrutinized the supporting data to ensure that the information critical to this analysis is consistent and reliable." The court found that Dow's experts, despite their many methodological criticisms, failed to offer their own superior methodology or suggest any scenario in which the Independent Assessor's projections could be off by the full $100.4 million cushion. The mere fact that 72,000 claimants could theoretically file claims did not trouble the court given that, among other reasons, many of those claimants had taken no action since filing their proofs of claim more than twenty years before. The district court therefore dismissed Dow's criticisms as an attempt to, in effect, transform the standard from one of "virtual" to "actual" guarantee.
Instead, the court relied on the Independent Assessor's analysis, which it characterized as the methodology incorporated into the facility agreement and the "only reasonably available method of predicting future claims in a closed pool of claimants." The court found "that claim filing has [indisputably] slowed considerably," and that comparison to the most analogous mass-tort settlement facility indicates that wind-down behavior "is predictable." Thus, the district court was satisfied that the $100.4 million projected cushion was large enough to virtually guarantee all first-priority payments, and authorized the requested second-priority payments.
II.
A.
Affirmance is required in this case largely because we review the district court's factual findings under a deferential "clearly erroneous" standard. That standard applies even though the ultimate question—whether second-priority payments may be authorized under the settlement facility agreement—is a mixed question of law and fact. The first step in answering this mixed question is to determine what showing the contract requires for the district court to authorize the payments. That is the legal question that was answered by the earlier panel of this court: all first-priority payments must be "virtually guaranteed" to authorize second-priority payments. See In re Settlement Facility Dow Corning Tr., 592 F. App'x at 478-80. The district court correctly identified this as the controlling standard, and the parties agree that a "virtual guarantee" requires something like 99% certainty.
The next step is to make findings of "basic" or "historical" fact—typically questions like "who did what, when or where, how or why." See U.S. Bank Nat'l Ass'n v. Village at Lakeridge, LLC, 138 S. Ct. 960, 966 (2018). In this context, the district court made factual findings concerning the dramatic slowing of claims filings, the dormancy of many of the 72,000 remaining potential claimants, the accuracy of past projections, the generally accepted adequacy of the Independent Assessor's methodology, the lack of any relevant epidemiological data, and the existence of a conservatively projected cushion of $100.4 million. Factual findings such as these are reviewable only for clear error. See id.
Where the rubber meets the road is the application of that identified legal standard to those factual findings. Here, that question is whether the characteristics and content of the Independent Assessor's projections, as found by the district court, amount to a virtual guarantee that all first-priority payments will be made. This ultimate inquiry is a mixed question of law and fact. See id. (citing Pullman-Standard v. Swint, 456 U.S. 273, 289 n.19 (1982)). But not all mixed questions are created alike or deserve like review. Instead, the standard of review depends on "whether answering it entails primarily legal or factual work." See id. at 967.
The ultimate question here is entitled to review only for clear error because determining whether future first-priority payments are virtually guaranteed requires almost entirely factual work. To apply the "virtual guarantee" standard, the district court reviewed the content of complicated projection analyses, assessed the confidence in those projections by hearing and balancing competing expert evidence on the methodology employed, and ultimately determined the likely funds required to pay first- and second-priority payments over the remaining life of the facility. As was the case in Lakeridge, "[j]ust to describe that inquiry is to indicate where it (primarily) belongs: in the court that has presided over the presentation of evidence, that has heard all the witnesses, and that has both the closest and the deepest understanding of the record"—i.e., the district court. See id. at 968. This is particularly true where, as here, the governing legal standard is derived from the parties' contract and not more generally applicable law. See id. Moreover, the parties' arguments focus on the evidence and the record, and not on legal precedents and authority, thus indicating that the issues are factual rather than legal. See Indmar Prods. Co., Inc. v. Comm'r, 444 F.3d 771, 785-86 (6th Cir. 2006) (Rogers, J., concurring). The determinative issues on this appeal, in short, are factual and subject to deferential "clearly erroneous" review.
B.
The district court did not clearly err in finding that the Independent Assessor's projections supplied a virtual guarantee that the settlement facility can afford all future first-priority payments despite the early distribution of fifty percent of second-priority payments. Although different people might reasonably reach different conclusions on this record, we are not "left with the definite and firm conviction that a mistake has been committed," as would be required to reverse for clear error. See Easley v. Cromartie, 532 U.S. 234, 242 (2001) (quotation marks omitted).
The district court found a virtual guarantee based on the Independent Assessor's conservative projections of sufficient funds, expert evidence supporting the reliability of those projections, and the huge margin for error baked into the projected $100.4 million cushion. Dow's primary argument is that those projections could not supply a virtual guarantee without some sort of statistical risk analysis quantifying the probability that they might prove dramatically wrong. There is no doubt that the level of confidence in the Independent Assessor's projections is a relevant issue. See In re Settlement Facility Dow Corning Tr., 592 F. App'x at 480. But nothing in the governing agreement suggests that the Independent Assessor's projections must include a statistical risk analysis, and here there were sufficient indicia of reliability to support the district court's finding without one.
First, the Claimants' Advisory Committee presented expert evidence that the Independent Assessor "employed the most widely accepted and utilized method for forecasting future claims from data on past claim filings and past payments by the subject entity," and that the "method has been used and accepted as the basis of thousands of liability forecasts for mass tort trusts and defendants during the past 25 years." Second, expert evidence showed that the Independent Assessor's prior "liability forecasts were always significantly greater than the actual [facility] payments for the same period," and that the current projections were similarly based on "specific conservative assumptions." Third, the Claimants' Advisory Committee's expert showed that the projections conservatively accounted for a higher claims surge as the facility winds down than was actually seen in the most analogous breast-implant settlement fund. Together, this expert evidence supports the district court's reliance on the Independent Assessor's projection that sufficient funds are available to pay all first-priority claims.
The reliability of the projections is further backed up by the enormity of the margin for error provided for by the projected $100.4 million cushion. In order for first-priority payments to be jeopardized, the Independent Assessor's projections (which are designed to overestimate liability) must be wrong by more than $100.4 million. That is a lot of leeway. Dow argues that this margin for error is insufficient to virtually guarantee first-priority payments, because the cushion will be depleted if 12-17%, rather than the projected 3%, of the 72,000 remaining claimants file successful disease claims. That argument has some initial intuitive appeal, but Dow's experts made no attempt to show what, if any, real chance there was of that happening. The Claimants' Advisory Committee's expert, on the other hand, opined that based on the remaining claimants' failure to pursue their claims for nearly twenty years despite repeated notice, and the low filing and success rates of pro se claimants (who make up the bulk of the remaining 72,000), "an unexpected huge surge of claims is extremely unlikely and virtually certain not to grow to such heights as to threaten the [$100.4 million] reserve." Although it was not Dow's burden to prove the likelihood that the cushion would be inadequate, the district court did not err by crediting the Claimants' Advisory Committee's expert to find that the risk of that happening was virtually nonexistent. On clearly erroneous review, we generally do not reweigh evidence. See United States v. Navarro-Camacho, 186 F.3d 701, 707-08 (6th Cir. 1999).
Dow's other methodological criticisms were also reasonably rejected by the district court. First, Dow argues that the Independent Assessor's methodology is flawed because it does not consider epidemiological data in estimating future liability for first-priority disease claims. But the district court heard and credited expert testimony that an epidemiological study was not only unnecessary but unfeasible given the lack of any such data. Second, Dow argues that the district court ignored its evidence of huge late claims surges in other mass-tort funds. But again, the court reasonably determined that the claims history of a similar breast-implant fund was the far more relevant precedent and that the projections already accounted for an even greater surge than occurred there. Dow attempts to aggregate its criticisms by arguing that the district court failed to consider its expert evidence at all, but in truth, the court merely weighed the competing evidence and found the Claimants' Advisory Committee's more persuasive. That Dow would have preferred the opposite is not a basis for reversal. "Where there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous." Anderson v. City of Bessemer City, N.C., 470 U.S. 564, 574 (1985).
Given the expert evidence supporting the reliability of the Independent Assessor's projections, the district court did not clearly err in concluding that the projected $100.4 million cushion was sufficiently large that all first-priority payments are virtually guaranteed. Thus, we affirm the district court's authorization of disbursal of fifty percent of second-priority payments.