Aldrich Pump LLC

1
UNITED STATES BANKRUPTCY COURT
WESTERN DISTRICT OF NORTH CAROLINA
CHARLOTTE DIVISION
)
In re 	) 	Chapter 11
)
Aldrich Pump LLC, et. al.,1 	) 	Case No. 20-30608
)
Debtors. 	) 	(Jointly Administered)
_________________________________________ )
PHASE I ORDER ESTIMATING DEBTORS’ AGGREGATE
ASBESTOS-RELATED LIABILITY
THIS MATTER is before the Court under § 502(c)(1) for estimation of the
Debtors’ aggregate asbestos-related liability based on the tort system extrapolation
method, using the Debtors’ settlement and verdict history. Pursuant to its Order
Commencing Phase I, [Dkt. No. 3011],2 the Court held a trial on this matter from
August 10, 2026, through August 14, 2026, at which the parties presented their
estimates of the Debtors’ present and future asbestos-related liabilities—each using
the tort system extrapolation method (the “Phase I Trial”). After considering all the
1 The Debtors are the following entities (the last four digits of Debtors’ taxpayer identification follow
in parentheses) Aldrich Pump LLC (2290) and Murray Boiler LLC (0679). The Debtors’ address is
800-E. Beaty Street, Davidson, North Carolina 28036.
2 Unless otherwise indicated, docket citations herein are made in reference to Case No. 20-30608.
_____________________________
Lena M James
United States Bankruptcy Judge
FILED & JUDGMENT ENTERED
Christine F. Ramsey
Western District of North Carolina
September 25 2026
Clerk, U.S. Bankruptcy Court
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 1 of 64

2
evidence admitted at the Phase I Trial, the Court estimates that the Debtors’ total
asbestos-related liability using the tort system extrapolation method is
$742,500,000. The following constitutes the Court’s findings of fact and conclusions
of law with respect to the Phase I Trial under Federal Rule of Bankruptcy
Procedure 7052, made applicable to this proceeding under Federal Rule of
Bankruptcy Procedure 9014.
I. 	JURISDICTION
The United States District Court for the Western District of North Carolina
has subject matter jurisdiction over this adversary proceeding under 28 U.S.C. §
1334(b). On April 14, 2014, the District Court entered its Amended Standing Order
of Reference, referring all proceedings arising under title 11 and arising in or
related to a case under title 11 to the bankruptcy judges for the Western District of
North Carolina. Under 28 U.S.C. § 155(a), the Honorable Albert Diaz, Chief Judge
of the United States Court of Appeals for the Fourth Circuit, assigned and
designated the above-signed judge to this Court and to the above-captioned case,
together with all associated adversary proceedings. [Dkt. No. 2363].
The Court has statutory and constitutional jurisdiction to enter a final order
estimating current and future asbestos personal injury claims. “[E]stimations of the
debtors' potential personal injury tort liabilities as an incident of the development of
a plan of reorganization are core proceedings within the bankruptcy court's
jurisdiction . . . .” A.H. Robins Co. v. Piccinin, 788 F.2d 994, 1012 (4th Cir. 1986); see
also 28 U.S.C. § 157(b)(2)(B). And this Court has constitutional authority to enter a
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 2 of 64

3
final order estimating current and pending asbestos claims “without running afoul
of the jurisdictional implications of Stern v. Marshall.” In re POC Props., LLC, 580
B.R. 504, 509 (Bankr. E.D. Wisc. 2017) (citing Stern v. Marshall, 564 U.S. 462, 499
(2011)); see also In re Rhodium Encore LLC, No. 24-90448, 2025 WL 3017436, at *3
(Bankr. S.D. Tex. 2025); In re J Publ'n Co., 660 B.R. 391, 398-99 (Bankr. N.D. Ill.
2024).
II. BACKGROUND
A. Corporate History
The Debtors in these consolidated bankruptcy cases are Aldrich Pump LLC
and Murray Boiler LLC. Their predecessors are, respectively, the former Trane
Technologies Company LLC, successor by merger to Ingersoll-Rand Company (a
former New Jersey corporation) (collectively, “Old IRNJ”), and the former Trane
U.S. Inc. (“Old Trane”). Findings of Fact and Conclusions of Law [Case No. 20-
03041, Dkt. No. 308, ¶¶ 1-2].
The Debtors’ predecessors and their subsidiary companies were, historically,
manufacturers of climate control products for buildings, homes, and transportation.
Id. Their corporate lineage stretches back to 1871, when Ingersoll Rock Drilling
Company first opened its doors. Id. ¶ 28. By the early twentieth century, they had
become global providers of industrial equipment and technology. Id. While Old
Trane and Old IRNJ did not mine or use asbestos in manufacturing products, they
did incorporate asbestos-containing components manufactured by third parties into
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 3 of 64

4
some of the industrial equipment they produced. Id. ¶ 30. These components are the
source of the Debtors’ asbestos-related liabilities.
Though both Old Trane and Old IRNJ had largely brought their use of
asbestos-containing components to an end by the mid-1980s, that time period
marked the beginning of what would become their decades-long experience
defending against asbestos-related personal injury claims in the state tort system—
an experience that continued up until the Debtors filed these case. Id. ¶¶ 31-36.
Initially, Old IRNJ and Old Trane faced relatively few claims in the context of an
asbestos litigation scene centered on larger asbestos manufacturers, commonly
known as the “Big Dusties.” Id. ¶¶ 33-34. But as those manufacturers began to exit
the state court tort system by filing for bankruptcy protection, the number of claims
brought against Old IRNJ and Old Trane in state court increased dramatically.3 Id.
Consequently, associated costs skyrocketed. Id. ¶ 35. Between 2015 and 2019, Old
IRNJ and Old Trane spent nearly $100 million annually to defend and resolve
asbestos claims. Id. In 2020, facing tens of thousands of claims with more likely to
come, Old IRNJ and Old Trane sought the protections of bankruptcy. They did so,
however, not by filing their own bankruptcy petitions, but by engaging in a series of
transactions (the “2020 Corporate Restructuring”) culminating in the creation of the
Debtors—Aldrich and Murray—and the commencement of these consolidated
bankruptcy cases. Id. ¶¶ 41-42.
3 For example, by the late 2000s, over 2,500 mesothelioma claims were asserted against Old IRNJ
and Old Trane, annually. Inclusive of claims involving lung cancer and other diseases,
approximately 5,000 asbestos-related claims were asserted annually against Old IRNJ and Old
Trane from 2015 to 2019. Id. ¶ 34.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 4 of 64

5
The 2020 Corporate Restructuring, at heart, involved the utilization of Texas
state law divisional mergers to separate the desirable assets and liabilities of Old
IRNJ and Old Trane from the unwanted asbestos-related liability each faced. Id.
¶ 42. Simply put, each company divided itself into two new entities and, as
permitted under Texas law, allocated substantially all valuable operating assets of
its predecessor to one of the new entities while allocating all of its predecessor’s
asbestos liabilities to the other. Id. ¶¶ 2, 52-64. Old IRNJ became Trane
Technologies Company LLC (“New TTC”) and Aldrich; Old Trane became Trane
U.S. Inc. (“New Trane”) and Murray. Id. ¶ 2. The Debtors’ 2020 Corporate
Restructuring was facilitated by a series of intercompany funding and support
agreements between the Debtors and related “Trane Enterprise” entities. Id. ¶¶ 47,
69-87. The divisional mergers allowed New TTC and New Trane to operate in the
ordinary course of business—free, in theory, from the burden of asbestos liability—
while Aldrich and Murray entered bankruptcy. Id. ¶ 42, 62. In this manner did Old
IRNJ and Old Trane, no longer formally extant, attempt to peel away an expensive
heritage of asbestos claims and funnel claimants seeking recovery into the cases
now pending before the Court.
B. Procedural History
The Debtors filed their respective bankruptcy petitions on June 18, 2020.
[Dkt. No. 1]; [Case No. 20-30609, Dkt. No. 1]. Following motions by the Bankruptcy
Administrator and the Debtors, respectively, the Court appointed the Official
Committee of Asbestos Claimants (“ACC”), [Dkt. No. 147], and Joseph W. Grier, III
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 5 of 64

6
as the Future Claims Representative (“FCR”). [Dkt. No. 389]. The Debtors, the
ACC, and the FCR make up the central constituencies involved in these cases and
this estimation proceeding.
In these cases, the Debtors have stated an intention to resolve their asbestos
liabilities through the consummation of a plan of reorganization that includes the
establishment of a trust under § 524(g) of title 11 of the United States Code (the
“Bankruptcy Code”). Findings of Fact and Conclusions of Law [Case No. 20-03041,
Dkt. No. 308., ¶ 24]. In furtherance of that aim, the Debtors filed a Motion for
Estimation of Prepetition Asbestos Claims, [Dkt. No. 833], alongside a Joint Plan of
Reorganization proposing such a trust with initial cash funding of approximately
$540 million in late 2021.4 [Dkt. No. 831, §§ 1.1.65, 4.7.3]. On April 18, 2022, the
Court entered an order (the “Estimation Order”) granting the Debtors’ Motion for
Estimation but extending its scope to “all current and future asbestos personal
injury claims . . . includ[ing] an estimation of mesothelioma claims plus the
application of an agreed upon calculation for non-mesothelioma claims.” Order
Authorizing Estimation of Asbestos Claims [Dkt. No. 1127, ¶ 2]. In due course, the
Court also entered a Case Management Order for the Estimation of Asbestos
Claims, which set forth an initial schedule and procedures in anticipation of a
4 The proposed initial funding amount purportedly reflects a negotiated settlement agreement
between the Debtors, the FCR, and their related professionals in which the ACC and its
professionals did not participate. [Dkt. No. 833, ¶ 10]. The Court later authorized the establishment
of a qualified settlement fund (“QSF”) in connection with the agreement contemplated by the Debtors
and the FCR. [Dkt. No. 994].
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 6 of 64

7
contested estimation proceeding to occur at an indefinite future point. [Dkt.
No. 1302].
Outside of the estimation context, the parties have litigated a swath of other
matters in these cases. For over a year following the petition date, the Debtors
pursued, and eventually obtained, a Preliminary Injunction Order from the Court
extending protection from asbestos-related litigation to a large class of non-debtor
affiliates. [Adv. Pro. 20-03041, Dkt. No. 307, 308]. In late 2021, the ACC commenced
an adversary proceeding against the Debtors and their divisional merger
counterparts under a theory of substantive consolidation. [Adv. Pro. No. 21-03029,
Dkt. 1]. Months later, the Court granted the ACC derivative standing to pursue
actions on behalf of the Debtors’ estates, [Dkt. No. 1121], and the ACC commenced
two additional adversary proceedings separately alleging fraudulent transfer and
breach of fiduciary duty against various debtor-related entities.5 Each of these ACC
adversary proceedings remains in the discovery phase, though all are presently
stayed for case management purposes. [Dkt. No. 3179; Adv. Pro. No. 22-03029, Dkt.
No. 98].
In addition to initiating adversary proceedings both against and on behalf of
the Debtors, the ACC moved to dismiss these bankruptcy cases entirely, as did a
group of asbestos claimants represented by Maune Raichle Hartley French & Mudd
(“MRHFM Claimants”). [Dkt. Nos. 1712, 1756]. The Court denied the ACC’s and
MRHFM Claimants’ motions to dismiss the cases, [Dkt. No. 2047], and their
5 See Adv. Pro. No. 22-03028, Dkt. No. 1; Adv. Pro. No. 22-03029, Dkt. No. 1.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 7 of 64

8
prolonged efforts to obtain appellate review of that decision ended with the Court of
Appeals for the Fourth Circuit’s denial of their respective petitions for permission to
appeal. [Dkt. No. 2208].6 The Debtors also successfully defended against the
MRHFM Claimants’ request for relief from the automatic stay to pursue claims in
state court. [Dkt. No. 2438]. Despite all this litigation, the status quo persisted in
these cases.
By the end of 2025, more than four years since the commencement of these
cases and nearly three years after the Court’s order initiating estimation under
§ 502(c), no date for an estimation proceeding had been established, and virtually no
further progress towards confirming a plan of reorganization had been made. The
parties had, however, assembled and exchanged initial expert reports and reliance
materials relating to estimation earlier that year.7 Sensing an opportunity to
advance these cases toward resolution, the FCR moved the Court for an order
commencing the estimation trial based on the parties’ expert reports and tort
system values. [Dkt. No. 2941]. The FCR observed that though the parties’
preferred estimation methods diverged significantly, each expert report contained a
discrete liability estimate based on a tort system extrapolation of the Debtors’
asbestos liabilities—a familiar estimation method in the context of asbestos
6 Recently, the United States Supreme Court denied certiorari on an appeal from a Fourth Circuit
decision on a related question regarding a Chapter 11 debtor’s financial solvency. See Bestwall LLC
v. Off. Comm. of Asbestos Claimants of Bestwall, LLC, 148 F.4th 233, 243 (4th Cir. 2025), cert. denied
sub nom. Comm. of Asbestos Claimants v. Bestwall LLC, 225 L. Ed. 2d. 231 (June 1, 2026).
7 The Court’s Second Amended Case Management Order for Estimation of Asbestos Claims required
the parties to exchange expert reports and reliance materials on or before September 15, 2025. [Dkt.
No. 2656, ¶ 3].
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 8 of 64

9
bankruptcy history and the ACC’s preferred method. 8 The FCR reasoned that
moving forward with estimation based on the parties’ common-ground tort system
estimates, reserving alternative estimation methods for a later time, if necessary,
might both bring the parties closer to a consensual resolution of these cases and
obviate the need for years more of expensive discovery and litigation. Id. The
Debtors and the ACC agreed that a phased estimation proceeding was feasible and
carried the potential to greatly advance these cases, therefore, the Court granted
the motion. Id. The order bifurcating this estimation proceeding provides for “(a)
Phase I—where the Court [would] make a finding estimating the Debtors' liability
for current and future asbestos claims by using the tort system extrapolation
method, using the Debtors’ settlement and verdict history, and [would] not address
the Debtors’ legal liability theory, and (b) Phase II—where the Court, if necessary,
[would] hear evidence at a date to be determined concerning any alternative
estimation methodologies set forth in the Initial Expert Reports, as appropriately
supplemented, and not considered by the Court in Phase I.” Order Commencing
Phase I of the Estimation Trial with Hearings based on the Tort System
Extrapolation Method and the Parties’ Expert Reports [Dkt. No. 3011, ¶ 2].
III. 	ESTIMATION UNDER § 502(c)
Section 502(c) of the Bankruptcy Code “expressly directs bankruptcy courts to
‘estimate for purpose of allowance’ the value of claims that are contingent or
8 The Debtors prefer an estimation utilizing a “legal liability” theory as was done in In re Garlock
Sealing Techs., LLC, 504 B.R. 71, 94 (Bankr. W.D.N.C. 2014), which entails more extensive discovery
than the tort system extrapolation method. See infra note 13.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 9 of 64

10
unliquidated, to avoid undue delay in the administration of a bankruptcy.”
SummitBridge Nat'l Invs. III, LLC v. Faison, 915 F.3d 288, 293 (4th Cir. 2019)
(quoting 11 U.S.C. § 502(c)). A claim is defined in the Bankruptcy Code as a “right
to payment,” 11 U.S.C. § 101(5)(A), and “the ‘basic federal rule’ in bankruptcy is
that state law governs the substance of claims.” Travelers Cas. & Sur. Co. of Am. v.
Pac. Gas & Elec. Co., 549 U.S. 443, 450 (2007) (quoting Butner v. United States, 440
U.S. 48, 57 (1979)). This basic rule has been applied in the context of § 502(c)
estimation proceedings. Owens Corning v. Credit Suisse First Boston, 322 B.R. 719,
721 (D. Del. 2005) (citing Bittner v. Borne Chem. Co., Inc., 691 F.2d 134, 135 (3d Cir.
1982)); In re Garlock Sealing Techs., LLC, 504 B.R. 71, 94 (Bankr. W.D.N.C. 2014).
Here, within the context of a mass tort asbestos case, the Court will estimate
the pending and future asbestos claims as of the Debtors’ petition date and consider
“how a claim would have been valued in the state court system had the debtor never
entered bankruptcy.” In re Armstrong World Indus., 348 B.R. 111, 123 (Bankr.
D. Del. 2006) (citing Owens Corning, 322 B.R. at 722 (estimating the total amount,
as of the petition date, current asbestos claimants “could legitimately have claimed
as compensation” and what future claims “would have been worth in the tort system
as it existed on the petition date”)).9 While claim value is determined as of the
9 The Court’s estimation of pending and future asbestos claims under § 502(c) must look to the
claims’ value under state law, which is a fundamentally different task than “estimating the value
which claimants might take in satisfaction of their claims through some bankruptcy mechanism such
as a trust of the sort provided for at § 524(g).” In re Eagle-Picher Indus., 189 B.R. 681, 683 (Bankr.
S.D. Ohio 1995). Valuing the asbestos claims against the Debtors based on the experience of other
bankruptcy trusts “is simply not the same thing as estimating the value of future claims as of the
filing date of the bankruptcy case," id. at 692, because those trusts are “not operating in the tort
system—in which any estimation of [the Debtor's] future liability must assume that it will operate.”
In re Armstrong World Indus., 348 B.R. at 131.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 10 of 64

11
petition date under state law, the Bankruptcy Code does not—as numerous courts
have observed—explain precisely how claims are to be estimated. See, e.g., Bittner,
691 F.2d at 135; In re G-I Holdings, Inc., 323 B.R. 583, 599 (Bankr. D.N.J. 2005); In
re Garlock Sealing Techs., 504 B.R. at 87; Official Comm. of Asbestos Claimants v.
Asbestos Prop. Damage Comm. (In re Federal-Mogul Global Inc.), 330 B.R. 133, 135
(D. Del. 2005). Given that lack of explicit guidance, many courts have concluded
that “a bankruptcy court has discretion to determine the appropriate method of
estimation in light of the particular circumstances of the bankruptcy case before it.”
In re Garlock Sealing Techs., 504 B.R. at 93 (quoting In re G-I holdings, Inc. (“G-I
II”), No. 01-30135, 2006 WL 2403531, at *2 (Bankr. D.N.J. 2006)); see also Bittner,
691 F.2d at 135; In re Federal-Mogul Global, Inc., 330 B.R. at 155; In re Trident
Shipworks, Inc., 247 B.R. 513, 514 (Bankr. M.D.Fla. 2000); In re Thomson
McKinnon Sec., Inc., 143 B.R. 612, 619 (Bankr. S.D.N.Y. 1992). The Fourth Circuit
observed in In re A.H. Robins Co. that § 502(c) “contemplates a full adjudication”
that might incorporate tools such as epidemiological studies or a debtor’s claims and
settlement history to estimate aggregate liability. 880 F.2d at 720, 720 n.13.
Though estimating contingent or unliquidated mass tort claims in the
aggregate allows a bankruptcy court to avoid “unduly delay[ing] the administration
of the case,” 11 U.S.C. § 502(c), estimating pending and future asbestos claims is an
inexact science that defies “mathematical precision” or an “exact number.” In re
Armstrong World Indus., 348 B.R. at 124; see also Owens Corning, 322 B.R.at 721
(observing that claims valuation involves “predictions which are themselves based
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 11 of 64

12
upon predictions and assumptions”). One court analogized estimating mass tort
claims to determining the number of marbles in a jar—no matter the estimator’s
skill, the estimate will not be more accurate than taking the marbles out and
counting them one-by-one. In re N. Am. Health Care, Inc., 544 B.R. 684, 689 (2016)
(Bankr. C.D. Cal. 2016).
Though the Debtors, ACC, and the FCR retained numerous experts and
dedicated countless hours to researching the Debtors’ current and future liability, in
the end, “an estimation by definition, is an approximation.” In re Federal-Mogul
Global, Inc., 330 B.R. at 155. Estimation does not require “clairvoyan[ce]” or
“certainty,” but is “merely the court’s best estimate for the purpose of permitting the
case to go forward.” In re Federal Press Co., 116 B.R. 650, 653 (Bankr. N.D. Ind.
1989) (quoting In re Baldwin-United Corp, 55 B.R. 885, 898 (Bankr. S.D. Ohio
1985)).
In that spirit, estimation ideally should be accomplished quickly, efficiently,
and fairly to facilitate the timely formulation and confirmation of a chapter 11 plan
that will provide for prompt distribution to creditors. Here, while the Debtors and
the FCR are, for the most part, in agreement as to the Debtors’ estimated asbestos
liabilities, the ACC has bitterly contested virtually every issue presented in these
cases, including the estimated amount of asbestos liabilities. The hope is that an
estimate based on the ACC’s preferred method of valuation will inform the parties
and advance mediation as contemplated by the Phase I estimation order.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 12 of 64

13
IV. THE ESTIMATION EXPERTS AND RULE 702
The Debtors, the ACC, and the FCR offered experts who estimated the
Debtors’ aggregate asbestos personal injury liability, including (1) mesothelioma
claims pending but unpaid as of the petition date and (2) the present value of future
mesothelioma claims that are expected to be filed after the petition date. Each side
also offered rebuttal testimony challenging opposing experts’ methodology and
conclusions. Collectively, the Court heard from five experts with more than 100
years of asbestos-related valuation experience. Tr. 8/14/26 at 91.10
A. Proposed Experts
To provide an affirmative opinion on current and future asbestos liability, the
Debtors retained Dr. Charles H. Mullin, a Ph.D. economist and board chair at the
Bates White economic consulting firm. Dr. Mullin, who specializes in econometrics
and labor economics, has worked on mass tort forecasting since joining Bates White
in 2003 and has run the firm’s mass tort practice for more than a decade. Tr.
8/10/26 at 184-88. He has published numerous papers in peer-reviewed journals,
primarily in the field of robust estimation techniques, and has provided expert
testimony in other cases regarding mass tort liabilities. Tr. 8/10/26 at 185-86, 192.
Dr. Mullin evaluated the Debtors’ liability using a component-based tort system
extrapolation, employing a version of the Nicholson model incidence curve to
forecast future claims. He offered a preferred component-based range of $425
million to $550 million. Tr. 8/11/26 at 116-120; ALD-D5 at 75. He ultimately put
10 The transcripts for the Phase I trial are found at Dkt. Nos. 3381 (Mon. Aug. 10), 3382 (Tue. Aug.
11), 3383 (Wed. Aug. 12), 3384 (Thur. Aug. 13), and 3385 (Fri. Aug. 14).
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 13 of 64

14
forward his “best” estimate of the Debtors’ aggregate liability for present and future
mesothelioma claims as $460 million, an opinion dictated by his view on the key
assumptions integral to a component-based forecast for the Debtors. Tr. 8/10/26 at
209; Tr. 8/11/26 at 146. In addition to his affirmative estimate of the Debtors’
asbestos liability, Dr. Mullin testified in rebuttal to the ACC’s expert witness, Dr.
Andrew Sackett. No party objected to Dr. Mullin’s qualification as an expert at the
trial.
The Debtors also presented rebuttal testimony from Dr. Denise Neumann
Martin, who holds a Ph.D. in economics and specializes in microeconomics and
statistics. Dr. Martin is a senior managing director at NERA Economic Consulting,
has 35 years of experience in mass tort and product liability forecasting, has
testified regarding asbestos forecasts in both state and federal court, and is the co-
author of Estimating Future Claims: Case Studies from Mass Torts and Product
Liability (Frederick C. Dunbar et. al. 1996). Dr. Martin was retained by the Debtors
to explain her view of the accepted methodology for asbestos claims forecasting and
to assess whether the approach taken by the ACC’s expert, Dr. Sackett, was
consistent with that methodology. Tr. 8/13/26 at 197-201. No party objected to Dr.
Martin’s qualification as an expert.
For an affirmative opinion and rebuttal opinion, the FCR retained David
McKnight, a principal with the economic consulting firm Brattle Group. McKnight
is a financial economist who holds an MBA and specializes in valuation, finance,
and financial modeling. Between his 13 years at Brattle and his preceding
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 14 of 64

15
experience at NERA, McKnight has approximately 25 years of experience with mass
tort forecasting. Tr. 8/11/26 at 241-44. McKnight regularly publishes on matters
related to mass tort estimation as well as the tort system generally, and he has
testified and supported outside experts in tort-related state and federal proceedings
as well as arbitrations. Tr. 8/11/26 at 245, 247-48. McKnight evaluated the Debtors’
liability by employing a competing-risk model after adopting the LAS and Bates
White incidence curves for future claims. He corroborated that analysis with
separate evaluations using Walker model and Hazard model claim forecasts. Tr.
8/11/26 at 297-98; Tr. 8/12/26 at 156-57. McKnight estimated the Debtors’ aggregate
liability for present and future mesothelioma claims using the tort system
extrapolation method to be $535 million or $608 million, depending on whether the
Court applies a trust-return or risk-free rate in adjusting liabilities from nominal to
real dollar values. Tr. 8/12/26 at 134-35, 157-58. No party objected to McKnight’s
qualification as an expert.
For its affirmative expert testimony, the ACC put forward Dr. Andrew
Sackett, the president of Legal Analysis Systems, Incorporated (“LAS”). Dr. Sackett
is a former attorney who also holds a Ph.D. in history. He joined LAS in 2020,
where he works on asbestos claims forecasting and liability estimation, supervising
an interdisciplinary team of professionals from the fields of economics, statistics,
computer programing, and public health. Tr. 8/10/26 at 8-10. Dr. Sackett has
published a few articles on the interaction between asbestos bankruptcy trusts and
the tort system but, until this trial, had not testified in an estimation proceeding.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 15 of 64

16
Tr. 8/10/26 at 12. Dr. Sackett evaluated the Debtors’ liability using a component-
based forecast for present and future claims estimation, employing an updated
Nicholson curve to forecast future claims. He estimated the Debtors’ aggregate
liability for present and future mesothelioma claims using the tort system
extrapolation method to be approximately $1.82 billion. Tr. 8/10 at 46-48, 62-63.
The ACC presented rebuttal testimony from Yvette Austin, a senior
managing director and chair of the global finance practice group at Compass
Lexecon, an economic consulting firm. Austin holds an MBA in Finance, has over 30
years of experience in valuation, taught graduate-level courses on the subject, and
has testified in numerous mass tort valuations. Tr. 8/12/26 at 212-16. The ACC did
not retain Austin for an affirmative estimate of the Debtors’ asbestos liability but
instead tasked her with rebutting the affirmative opinions of Dr. Mullin and
McKnight. Tr. 8/12/26 at 217; Tr. 8/13/26 at 33. No party objected to Austin’s
qualification as an expert.11
B. Rule 702 Objection to Dr. Sackett’s Qualification
The FCR and the Debtors objected to the qualification of Dr. Sackett as an
expert witness at trial, arguing he lacks the academic credentials and professional
experience to be a mass tort claims forecasting expert. Dr. Sackett, they contend,
does not hold a degree in economics, finance, or statistics, has never testified as an
11 In making its findings and conclusions, the Court considered all of the testimony provided by
Austin. However, much of her testimony is of limited assistance in this proceeding because
significant portions were directed toward rebutting parts of McKnight’s initial report that had been
outdated by his March 2026 report. Separately, the Court found much of Austin’s testimony difficult
to follow and, often, flatly evasive. For these reasons, the Court’s discussion of her testimony in this
order is sparing.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 16 of 64

17
expert on the subject, and has limited experience mostly related to asbestos trusts
rather than estimations under the Bankruptcy Code. Tr. 8/10/26 at 35, 41. Because
the admissibility of Dr. Sackett’s testimony impacts the evidence available to it, the
Court first considers the FCR’s and Debtors’ objections to Dr. Sackett’s
qualifications before turning to the experts’ estimation opinions.
Federal Rule of Evidence 702 governs the admissibility of expert evidence,
providing that:
A witness who is qualified as an expert by knowledge, skill, experience,
training, or education may testify in the form of an opinion or
otherwise if the proponent demonstrates to the court that it is more
likely than not that:
(a) the expert’s scientific, technical, or other specialized knowledge
will help the trier of fact to understand the evidence or to determine
a fact in issue;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles and methods;
and
(d) the expert’s opinion reflects a reliable application of the
principles and methods to the facts of the case.
Fed. R. Evid. 702 (2026).
Although the Court must perform its gatekeeping role under Rule 702,
“ensur[ing] that the expert is qualified and that the expert’s testimony is both
relevant and reliable,” the Court “is not intended to serve as a replacement for the
adversary system, and consequently, the rejection of expert testimony is the
exception rather than the rule.” United States v. Smith, 919 F.3d 825, 835 (4th Cir.
2019) (internal citation omitted). The Court must “decide whether the expert has
‘sufficient specialized knowledge to assist the jurors in deciding particular issues in
the case.’” Belk, Inc. v. Meyer Corp., U.S., 679 F.3d 146, 162 (4th Cir. 2012) (quoting
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 17 of 64

18
Kumho Tire Co. v. Carmichael, 526 U.S. 137, 156 (1999)). “[A] witness’
qualifications to render an expert opinion are also liberally judged by Rule 702,”
Kopf v. Skyrm, 993 F.2d 374, 377 (4th Cir. 1993), and a witness may qualify as an
expert in a particular field through any one or more of the five bases enumerated in
Rule 702—knowledge, skill, experience, training, or education. Id. (citing
Friendship Heights Assocs. v. Vlastimil Koubek, A.I.A., 785 F.2d 1154, 1159 (4th
Cir. 1986).
Dr. Sackett testified that he works “almost exclusively [on] asbestos claims,
forecasting and liability estimation.” Tr. 8/10/26 at 8. Although he is an attorney
with a doctorate in history, Dr. Sackett maintains he has 15 years of experience in
the field of asbestos claims estimation, beginning with his work on asbestos trusts
as an attorney at the law firm of Caplin & Drysdale and then with his work at LAS
since 2020, where he has dealt with approximately 25 bankruptcy or asbestos
personal injury trusts. Tr. 8/10/26 at 9-11. In response to critiques that his academic
credentials and publication history have no connection to mass tort forecasting, Dr.
Sackett explained that he “was trained on the job, not through formal education,”
learning directly from estimation experts such as Dr. Mark Peterson, Tr. 8/10/26 at
11-13, 19, 37, a founder of LAS who has been recognized as an expert in numerous
asbestos estimation proceedings. See In re Federal-Mogul Global, Inc., 330 B.R. at
144.
Under Rule 702, the Court must “consider the proposed expert's full range of
experience and training” and not “just his professional qualifications.” Belk, Inc.,
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 18 of 64

19
679 F.3d at 162. Moreover, “an expert witness is not strictly confined to his area of
practice, but may testify concerning related applications; a lack of specialization
does not affect the admissibility of the opinion, but only its weight.” United States
v. Wen Chyu Liu, 716 F.3d 159, 168-69 (5th Cir. 2013) (quoting Wheeler v. John
Deere Co., 935 F.2d 1090, 1100 (10th Cir. 1991)); see also Martin v. Fleissner GmbH,
741 F.2d 61, 64 (4th Cir. 1984) (finding an expert’s “lack of direct experience is not a
sufficient basis to reject their testimony, but may affect the weight that testimony is
given”). The Court is satisfied that Dr. Sackett has the necessary experience and
training to qualify as an expert under Rule 702. Though Dr. Sackett is not an
economist and has undertaken claims estimation primarily through his work with
asbestos trusts rather than in the context of bankruptcy estimation proceedings,
“witnesses need not have complete knowledge about the relevant field of specialized
information to be of help to the trier of fact and thus qualified to testify as experts
under Rule 702.” 4 WEINSTEIN ’ S FEDERAL EVIDENCE § 702.04(5) (2026). Dr. Sackett
testified that he has “years of experience working on asbestos claims estimation,
which is a similar process, whether it is in a bankruptcy or dealing with a trust.” He
further explained that, although most of his estimation-related work was with
asbestos trusts, “the principles are the same . . . forecasting numbers of future
claims and valuating those claims.” Tr. 8/10/26 at 23, 36. Given the potential
overlap in claims estimation technique, and his years of experience and training in
estimating asbestos claims, the Court finds Dr. Sackett satisfies the liberal
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 19 of 64

20
standard for qualifying expert witnesses.12 Accordingly, the Court will overrule the
FCR and Debtors’ objection to Dr. Sackett’s qualifications, though his lack of
specialization relative to the other admitted experts affects the weight of his
testimony.
V. EXPERT TESTIMONY
A. Areas of Agreement on Liability Estimates
The testimony at the Phase I trial demonstrated that for purposes of
estimating liabilities using the tort system extrapolation method, the experts agree
on and employ the same standard formula, as well as several important variables
within that formula. The experts collectively referred to this as the “traditional,”
“standard,” or “accepted” framework for valuing mass torts. Tr. 8/10/26 at 40, 48,
229; Tr. 8/11/26 at 259; Tr. 8/13/26 at 204-05.
The framework involves three broad steps: Step 1 is estimating the total
number of claims, which is the sum of claims pending as of the petition date and
forecasted future claims; Step 2 provides a nominal value to those claims by
multiplying that total by a resolution rate, pay rate, and average settlement value;
Step 3 then produces the net present value of those claims by applying appropriate
inflation and discount rates to that nominal value. Tr. 8/13/26 at 204-05; Tr. 8/11/26
at 259-60. This framework aligns with how most courts have approached asbestos
12 The objecting parties’ concerns regarding the degree of Dr. Sackett’s specialized knowledge and
expertise are matters left to the trier of fact’s consideration. Both the FCR and the Debtors were able
to engage in fulsome cross-examination of Dr. Sackett at the trial to expose any perceived gaps in his
knowledge and analysis.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 20 of 64

21
estimations, “us[ing] the debtor’s pre-bankruptcy history of resolving claims
through litigation and settlements to estimate claims in the subsequent
bankruptcies.” In re Garlock Sealing Techs., 504 B.R. at 93 (citing In re Specialty
Prods. Holding Corp., No. BR 10-11779-JKF, 2013 WL 2177694, at *1 (Bankr. D.
Del. 2013) and In re Eagle-Picher Indus., 189 B.R. 681, 691 (Bankr. S.D. Ohio
1995)).13 Along with this common framework for estimating the Debtors’
mesothelioma liability, and for purposes of this Phase I estimation, the Court then
adds an additional step, applying the parties’ agreed upon formula or multiplier to
the estimate for mesothelioma claims to account for non-mesothelioma liability.
1. 	Step 1: Claim Count and Propensity to Sue
The experts agree on several key aspects of Step 1, including the incidence
curve and concomitant future claims count; the parties readily admit this level of
13 Although most have looked to the debtor’s tort system history in estimating current and future
asbestos liabilities, courts have not always agreed on the appropriate data to include and whether
“adjustments should be made to historical values.” In re Armstrong World Indus., 348 B.R. 111, 124
(D. Del. 2006) (quoting Owens Corning v. Credit Suisse First Boston, 322 B.R. 719, 723 (D. Del.
2005)). At least two courts departed from strict adherence to the historical volume of claims filings
and settlement values. In Owens Corning, the court was unwilling to extrapolate the debtor’s claims
history into the future because “some of the past results have been skewed by factors which can and
should be avoided in the future,” noting that “adjustments should be made to historical values to
account for these probable changes.” Owens Corning, 322 B.R. at 722-23. And in Garlock, the court
found the debtor’s claims resolution history “an unreliable predictor of its true liability” due to
plaintiff firms’ improper withholding of exposure evidence, endorsing instead a “legal liability”
approach that estimated the debtor’s share of liability relative to other defendants. In re Garlock
Sealing Techs., 504 B.R. at 86, 94-95. Instead of relying on the historical settlement data, the
estimate adopted in Garlock was derived from information provided in questionnaires sent to current
asbestos claimants’ law firms (PIQs), as well as data obtained from certain asbestos trusts and
bankruptcy cases. Id. at 95. The Garlock court’s reluctance to use historic claiming data, which it
considered “stale and not accurate,” necessitated a substantial, ground up approach to estimation,
including extensive, time-consuming discovery and data analysis to obtain the “freshest and most
reliable data available.” Id. at 96. In these cases presently before the Court, the multi-year discovery
process pertaining to the legal liability approach—which has expanded well beyond what was
undertaken in Garlock—has undermined the basic purpose of estimation, that is, to avoid undue
delay in the administration of a bankruptcy case.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 21 of 64

22
agreement is unusual in contested estimation proceedings. Tr. 8/14/26 at 57-58, 96-
97; 169-70. The total claims are derived from the formula:
[Pending Claims] + [Future Claims (Future Incidence X filing rate)]
= TOTAL M ESOTHELIOMA C LAIMS
The number of pending mesothelioma claims, as of the petition date, is
readily ascertainable from the Debtors’ database. The database is a record of over
380,000 asbestos-related claims against the Debtors, including those resolved and
still open, and includes demographic data—claimant name, age, residence,
occupation—as well as claimants’ medical information. Tr. 8/10/26 at 50-51. After
eliminating duplicate claims, resolved claims, and claims marked for a non-
mesothelioma disease, Tr. 8/10/26 at 50, the calculation of pending claims is
straightforward: “you can go look at the data and add them up.” Tr. 8/10/26 at 229-
30. Before adjusting for claims marked with an “unknown” disease in the database,
and certain off-complaint claims, which are points of dispute between the parties,
see infra Section V.B.2, 3, Dr. Sackett identified 6,776 open mesothelioma claims,
which generally aligns with the 6,816 claims identified by Dr. Mullin. Tr. 8/11/26 at
51-52; ALD-D5 at 49.
The formula for calculating future claims is essentially the same as that used
for pending claims but with one key difference: there is no database or list of future
claimants. Tr. 8/10/26 at 61-62. In the absence of such data, an estimate of future
claims is the product of two factors, an incidence curve—projecting the medical
cases of mesothelioma that will occur—and a filing rate, or “propensity to sue,” that
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 22 of 64

23
reflects the historical percentage of mesothelioma patients who sued the Debtors.
Tr. 8/10/26 at 62, 230.
The first factor, incidence curves, are illustrations of sophisticated methods of
modeling an exposed population that combine epidemiology, industrial hygiene,
labor economics, and statistics. Tr. 8/10/26 at 233, 238-39. This modeling is a critical
variable for courts estimating future asbestos liabilities, with the Nicholson14
approach the most frequently cited.15 Here, all three of the affirmative experts used
a variant of the Nicholson model at some point in their estimation process but with
modifications to account for census and mortality data collected since the study’s
publication date in 1982. Tr. 8/10/26 at 62-63, 237-38. Though most mesothelioma
14 Nicholson and his co-authors attempted to predict the number of people who would get asbestos
related cancer from 1980 to 2030. See Nicholson, W.J., Perkel, G., & Selikoff, I.J., Occupational
Exposure to Asbestos: Population at Risk and Projected Mortality – 1980–2030. 3 A M . J. I NDUS . M ED.
259 (1982). The Nicholson approach, and its subsequent revisions, estimate the probability of
asbestos disease by considering the population of individuals occupationally exposed to asbestos
across several industries and then taking “into account differences in intensity of exposure in various
occupations, and the resulting differential impact on incidence of disease.” In re Babcock & Wilcox
Co., 274 B.R. 230, 239 (Bankr. E.D. La. 2002). The Nicholson curve has been subsequently confirmed
by years of data, see In re Armstrong World Indus., 348 B.R. at 126, and has been periodically
updated to reflect that people are living longer and contracting mesothelioma at a higher rate than
originally forecasted. In re Specialty Prods., No. R 10-11779-JKF, 2013 WL 2177694 at *1 n.7, *15
(Bankr. D. Del. 2013).
15 See, e.g., In re Federal-Mogul Global Inc., 330 B.R. at 159 (comparing expert testimony relying
respectively on the Nicholson study curve, the National Cancer Institute’s Surveillance,
Epidemiology, and End Results (SEER) data, and an in-house model developed by one expert’s
private consulting company); In re Armstrong World Indus., 348 B.R. at 130 (noting that all three
expert predictions of future asbestos personal injury claims start with the Nicholson study); In re
Eagle-Picher Indus., 189 B.R. at 683 (observing that all but one of the experts recognized the
Nicholson report as authoritative in estimating the number of future claims); In re Garlock Sealing
Techs., 504 B.R. at 96-97 (adopting expert estimate using Bates White incidence model, an updated
version of the Nicholson model); In re Specialty Prods., No. BR 10-11779-JKF, 2013 WL 2177694, at
*1 (Bankr. D. Del. 2013) (observing that two experts used the Nicholson approach while the third
averaged the Nicholson and Peto approaches); Owens Corning, 322 B.R. at 721 (“Epidemiological
studies can give us some information about the likely occurrence of asbestos-related diseases in the
population in the future . . . .”).
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 23 of 64

24
claims against the Debtors arise from occupational exposure to asbestos-containing
products, the experts also used at least one curve that factored in varying degrees of
background mesothelioma incidence16 to account for past claims against the Debtors
from people who did not have occupational exposure. Tr. 8/10/26 at 240.
Dr. Mullin employed three different forecasts that vary in their incorporated
rates of background mesothelioma: a low-end curve centered on occupational
exposure that excludes any background mesothelioma diagnoses with no link to an
asbestos-containing product (“Non-ACP” diagnoses); a preferred curve that adds an
additional 10% from Non-ACP diagnoses, which he believes aligns better with the
tort system data; and a higher curve that adds 20% from Non-ACP diagnoses. Tr.
8/10/26 at 240.17 Dr. Sackett’s LAS incidence curve aligns closely with Dr. Mullin’s
high-end 20% Non-ACP curve. Tr. 8/10/26 at 98, 244; Tr. 8/11/26 at 100. McKnight
originally used the KPMG-Nicholson model in his September 2025 report, which is
strictly based on occupational exposure;18 however, in the interest of providing the
Court with a more direct comparison for the Phase I trial, McKnight produced an
16 The background rate of mesothelioma includes idiopathic mesothelioma—spontaneous
development without exposure to asbestos—as well as mesothelioma arising from other sources of
exposure to asbestos outside of direct employment in asbestos-related industries. Tr. 8/12/26 at 12-
14.
17 Use of Dr. Mullin’s preferred 10% Non-ACP and his high-end 20% Non-ACP—and Dr. Sackett’s
curve—results in roughly a 6% difference to future claims. Tr. 8/10/26 at 245.
18 The KPMG/Nicholson model is a forecast based on the population of occupationally exposed
individuals in certain industries and does not include any background rate of mesothelioma from
people who worked outside of those occupations. The KPMG/Nicholson curve has a much faster
decay than the curves produced by Dr. Sackett or Dr. Mullin, generating a significantly lower
number of potential claims. Tr. 8/11/26 at 265-66, 287-88.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 24 of 64

25
additional forecast using Dr. Sackett’s LAS curve but otherwise retaining his
original methodology. Tr. 8/12/26 at 20, 30-31.19
The incidence curves are remarkably similar given the characteristics of the
exposed population. As most of the claims against the Debtors derive from
occupational exposure, and the Debtors ended the use of asbestos-containing
products in the 1980s, nearly the entirety of the exposed population is near or past
retirement age at this point. The shape or overall decline in the incidence curve,
therefore, is driven primarily by general mortality rates of people 70 years and
older. Tr. 8/10/26 at 234. As a result, the experts’ incidence curves—particularly Dr.
Sackett’s LAS curve, which McKnight later adopted, and Mullin’s high-end 20%
non-ACP curve—follow similar tracks after controlling for the percentage of
background mesothelioma diagnoses. The Court favors the LAS curve and Dr.
Mullin’s high-end curve given their similarities, the adoption of the LAS curve by
McKnight, and the experts’ agreement that individuals with background
mesothelioma diagnoses were filing claims against the Debtors prepetition. Tr.
8/10/26 at 240; Tr. 8/12/26 at 134, 151-52.
The second variable, filing rate or propensity to sue, represents the
percentage of forecasted mesothelioma diagnoses that lead to claims against the
Debtors in the tort system. The rate is the historical number of claims filed against
the Debtors relative to the total number of mesothelioma cases. Tr. 8/10/26 at 64-65.
19 As part of her rebuttal of McKnight’s KPMG/Nicholson curve, Austin calculated a curve that was
“pretty close” to Dr. Sackett’s and Dr. Mullin’s high-end curve; it was closer still to Dr. Mullin’s
preferred 10% Non-ACP curve. Tr. 8/13/26 at 16, 109-10.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 25 of 64

26
Given the time that tends to elapse between a claimant’s mesothelioma diagnosis
and their initiating a lawsuit, Dr. Sackett and Dr. Mullin employed a one-year gap
when comparing the diagnosis dates and filing dates. Tr. 8/10/26 at 65, 232-33.
Despite some distinctions in the treatment of “off-complaint” claims and unknown
claims, see infra Section V.B.2, 3, the experts agreed on the basic approach to
determining propensity to sue and the baseline filing rate. Tr. 8/10/26 at 233.
These points of agreement are ultimately reflected in the experts’ future
claims count. Dr. Mullin’s high-end estimate for future claims using a 20% Non-
ACP curve is around 38,258 while Dr. Sackett projects 38,276 using his Nicholson
model—a difference of only 18 claims. Tr. 8/10/26 at 230, 234; ALD-D4 at 18. In
using the LAS curve for his March 2026 report, McKnight arrived at a similar
future claims count of around 38,000, noting, “when you look at these three
forecasts, there’s really quite a lot of agreement amongst the three experts in terms
of what the number of claims – future claims is for the Debtors here.” Tr. 8/12/26 at
17-18, 154-155.
2. 	Step 2: Calibration Period and Average Settlement Values
The second step of the experts’ common framework involves the valuation of
the pending and future claims, essentially asking how many of those filed claims
are going to get paid and what amount they will receive. Tr. 8/13/26 at 205. To do
so, the total claims generated in Step 1 are multiplied by three factors, as reflected
in the formula:
[Total Claims] X [Resolution Rate] X [Pay Rate] X [Average Settlement Value] =
TOTAL M ESOTHELIOMA LIABILITY
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 26 of 64

27
The parties agree on the basic formula but disagree on the proper
implementation or calculation of the individual factors. Before considering those
factors, however, it is important to highlight the parties’ consensus on the
appropriate calibration period, an unseen but critical aspect of calculating some of
the factors in Step 2 as well as the propensity to sue or filing rate from Step 1. The
calibration window is the historical period used to estimate the factors inputted into
the formula or, in other words, “the period of history you are assuming is
representative of what the future will be like.” Tr. 8/11/26 at 11. The filing rate, pay
rate, and average settlement value factors are then determined using data collected
from the chosen calibration period.
Dr. Sackett utilized a three-year window from 2017 through 2019. Tr. 8/10/26
at 54-55. Given asbestos is a mature tort with a relatively stable set of average
settlement values, McKnight chose a five-year calibration period to incorporate
more data, including the only verdict against the Debtors. Tr. 8/11/26 at 295-96. Dr.
Mullin considered and tested a three, five, and ten-year calibration period to assess
the sensitivity of his claims estimates to that variable; after observing the same
stability, Dr. Mullin noted his indifference between a three and five-year window.
Tr. 8/10/26 at 252; Tr. 8/11/26 at 12.
The consensus among the three experts was that a more recent calibration
period was preferable to an older window. McKnight noted five years “is about as
long a calibration period [as] I would typically use.” Tr. 8/11/26 at 295-96. While
testing the sensitivity to a ten-year calibration period, Dr. Mullin noted that it is “a
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 27 of 64

28
long time ago, so you probably want to use the three or five-year window in this
context.” 8/11/26 at 11-12. Dr. Sackett selected a three-year period, believing that
the Debtors’ most recent history is the best guide to the future. Tr. 8/10/26 at 54-55.
This common approach between the experts also aligns with guidance from caselaw
that estimation “should be based upon settlement values for claims close to the
filing date of the bankruptcy case.” In re Eagle-Picher Indus., 189 B.R. at 691; In re
Federal-Mogul Global, Inc., 330 B.R. at 157. For a mature asbestos tort with a
stable history, the use of the three or five-year calibration period “almost makes no
difference in the final calculation.” Tr. 8/11/26 at 12. The three-year calibration
period produces a slightly higher forecast than the five year-window, but with little
overall effect on the final number, a difference of “less than five percent” in
McKnight’s view and “about 1 percent” according to Dr. Mullin. Tr. 8/10/26 at 252;
Tr. 8/11/26 at 296-97.
Once the calibration period is determined, the total number of estimated
mesothelioma claims from Step 1 is multiplied by the resolution rate, which reflects
the percentage of total claims that are resolved, either through payment or
dismissal, against the number of claims that never resolve. Tr. 8/13/26 at 219. The
three affirmative experts share an understanding of the meaning of a resolution
rate and its place in the common tort valuation framework but diverge in how that
rate is implemented and reflected in their respective models20 or, in Dr. Sackett’s
20 Dr. Mullin and Dr. Sackett used a component-based method, which separately estimates a
resolution rate, whereas McKnight used a competing risk model that already incorporates the
concept into calculating the probability of a filed claim getting paid. Tr. 8/12/26 at 39, 80-81.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 28 of 64

29
case, whether the use of a resolution rate in this Phase I estimation is even helpful
or statistically significant. Tr. 8/10/26 at 53-54, 246-47; Tr. 8/12/26 at 39, 80-81. See
infra Section V.B.1.
The next factor is the payment rate, which addresses how many of the claims
resolved during the calibration period received payment. Tr. 8/10/26 at 46, 52; Tr.
8/11/26 at 9. The parties agree on the importance of the payment rate to the overall
formula and validity of a calibration period but diverge in the specifics of how the
rate is calculated21 and, more significantly, how best to account for “off-complaint”
firms and whether excluding older claims is appropriate. Tr. 8/11/26 at 11-15; Tr.
8/12/26 at 131; see infra Section V.B.3, 4.
The final factor in Step 2 is the historical average mesothelioma settlement
value, which is the average amount that paid claims received during the calibration
period. Tr. 8/10/26 at 57. While the prior factors determine how many individuals
are going to receive payment, this factor addresses “what size the check is going to
be [to a claimant].” Tr. 8/10/26 at 251. This is another area in which the experts are
in near-complete agreement on both the approach and the result. Dr. Sackett and
Dr. Mullin both arrived at $56,000 for Aldrich claims and were within $1,000 for
Murray ($93,000 for Dr. Sackett and $92,000 for Dr. Mullin). Tr. 8/11/26 at 16-17.
McKnight’s average settlement values, after accounting for off-complaint firms, are
similar, with around $59,465 for Aldrich and $92,339 for Murray. Tr. 8/14/26 at 67;
21 Dr. Mullin and Dr. Sackett’s component-based analysis looks at payment rates as a percentage of
claims resolved, while McKnight’s competing risk model computes payments as a percentage of
claims filed. Tr. 8/11/26 at 297-99.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 29 of 64

30
Tr. 8/12/26 at 118. After multiplying the total mesothelioma claims from Step 1 by
each of the computed three factors: resolution rate, pay rate, and average
settlement value, the result is the nominal total mesothelioma liability, or the
liability in dollars as of the time the forecast is being prepared.
3. 	Step 3: Net Present Value Formula
The third and final step of the experts’ common framework takes the nominal
total mesothelioma liability amount and applies two factors, the projected inflation
rate and the discount rate, to determine net present value:
[Total Mesothelioma Liability] X [Inflation] ÷ [Discount Rate]
= NET PRESENT VALUE TOTAL M ESOTHELIOMA L IABILITY
The first factor, inflation, looks to the future and considers whether value that is
assigned to mesothelioma claims in today’s dollars will increase over time and at
what rate. The second factor then takes that amount and asks how much money
would need to be invested at an appropriate discount rate to allow for the payment
of those claims as they come due. Tr. 8/13/26 at 205, 227-28.
The experts agree on the concept of adjusting nominal payment streams to
net present value using inflation and discount rates. Tr. 8/10/26 at 146-47, 159-60,
205; Tr. 8/11/26 at 259-60. This also accords with prior decisions estimating current
and future asbestos liability. See, e.g., In re Eagle-Picher Indus., 189 B.R. at 691; In
re Federal-Mogul Global Inc., 330 B.R. at 157; In re Armstrong World Indus., 348
B.R. at 126 n.20; In re Garlock Sealing Techs., 504 B.R at 88. There is an enormous
divide between the experts, however, on the appropriate inputs for those two
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 30 of 64

31
factors, which in turn creates the greatest discrepancy in the overall liability
estimates. See infra Section V.B.7.
4. 	Gross-up for Non-Mesothelioma Liability
The final additional step in determining the Debtors’ total asbestos liability is
to account for and value non-mesothelioma claims. The Estimation Order provided
for the use of a simple formula or multiplier that would be applied to the Court’s
estimate of the Debtors’ mesothelioma liability, [Dkt No. 1127], which is a technique
that has been employed in multiple asbestos bankruptcy estimation proceedings to
expedite and simplify the estimation process. See, e.g., In re Specialty Prods., No.
BR 10-11779-JFK, 2013 WL 2177694, at *1 n.2 (Bankr. D. Del. 2013); In re Bestwall
LLC, No. 17-31795, [Dkt. 1577] at 6 (Bankr. W.D.N.C. Jan. 19, 2021); In re DBMP
LLC, No. 20-30080, [Dkt. 1239] at ¶ 4 (Bankr. W.D.N.C. Nov. 29, 2021).
Here, the parties have stipulated for purposes of Phase I only, that based on
Debtors’ historical ratio of mesothelioma versus non-mesothelioma claim payments
from 2015-2020, approximately 82.7% of Aldrich’s total liability and 80.6% of
Murray’s total liability represented mesothelioma claims. Therefore, the parties
agreed on multipliers of 1.21 for Aldrich and 1.24 for Murray to be applied to the
Court’s estimate for mesothelioma claims to derive an estimate of the Debtors’ total
liability for all asbestos claims. [Dkt. No. 3349]. The Debtors and the FCR—but not
the ACC—also agreed on the use of a single multiplier of 1.22 that could be applied
to the combined mesothelioma liability of the Debtors.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 31 of 64

32
B. Points of Disagreement on Liability Estimates
While each expert followed the general methodology as discussed in section
A, their implementation of that methodology varied. On one hand Dr. Mullin and
Dr. Sackett employed component-based forecasts, estimating total liability by first
evaluating individual claims groups and then adding those groups together. On the
other hand, McKnight combined those individual pieces of liability into a
multifactor analysis, making it more difficult to separate and compare his
treatment of individual claims groups. Tr 8/10/26 at 212. Notwithstanding these
implementation differences, the experts’ testimony made clear that Dr. Mullin and
McKnight are closely aligned in their view on most impactful estimation factors
while Dr. Sackett is the outlier.
1. 	Resolution Rate and Long Pendency Mesothelioma Claims
As described above, the parties agree that there are approximately 6,800
pending mesothelioma claims. Of the approximately 6,800 pending mesothelioma
claims, between 300 and 400 have been pending since before 2010, and another 425
to 450 were filed between 2010 and 2014. Tr. 8/11/26 at 53; ALD-D5 at 50.
McKnight, Dr. Mullin, and Dr. Martin all consider the age of a claim critically
informative and provided extensive testimony explaining their reasoning. Dr.
Sackett viewed all of these pending claims as the same, regardless of age.
Within the context of asbestos litigation, a resolution rate addresses the
theory that not every filed claim is resolved. When using a component-based model,
unresolved claims may be accounted for by multiplying the total claims number by
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 32 of 64

33
the percentage of claims that are resolved. Tr. 8/11/26 at 297. In their respective
component-based models, Dr. Mullin included a resolution rate while Dr. Sackett
did not.
Dr. Mullin explained that based on his experience, across asbestos
defendants, the percentage of unresolved claims is typically between 1% and 5%.
Dr. Mullin reviewed the Debtors’ database and found that, going back to the year
2003, 2% of claims remain unresolved and that percentage of unresolved claims
remains constant through 2012. In his estimation, these are claims that will never
resolve. Murray’s data is not quite as consistently low, in part because it had such a
small number of claims early on, but Murray also falls within the 1% and 5% range
after 2003. Based on this data from the Debtors’ database, Dr. Mullin concluded
that the Debtors’ experience a 2% abandonment rate, or 98% resolution rate. Tr.
8/10/26 at 246-47.
Dr. Mullin also reviewed the Debtors’ payments on mesothelioma claims over
the five-year period prior to the petition date and found that 99% of the Debtors’
payments are for claims that had been filed within five years of the payment. Dr.
Mullin indicated this finding was consistent with “almost every tort defendant.” As
a result, of the approximately 6,800 total pending mesothelioma claims, Dr. Mullin
calculated the total payment on pending mesothelioma claims in the database filed
within five years of the petition date and then added a 1% “gross up” to account for
the historical average of the few long pendency claims that would receive payment.
Tr. 8/11/26 at 54-56.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 33 of 64

34
McKnight also stressed the importance of the date of the claim based on his
review of the Debtors’ database. He also found that claims were much more likely to
be paid in the first five years from filing, and then they were “exceedingly unlikely
to be paid after that.” Tr. 8/11/26 at 298. Not including off-complaint claims, see
infra Section V.B.3, his analysis of the claims showed the probability of payment of
a claim after four years is less than 1%. Tr. 8/12/26 at 24. His competing risk model
specifically considers the length of time a claim has been pending when predicting
the probability of the claim being paid and, if paid, when it is most likely to be paid.
Tr. 8/12/26 at 82-83. McKnight did not use a resolution rate because the competing
risk model “essentially takes account of that” by computing payments as a
percentage of claims filed rather than a percentage of claims resolved. Tr. 8/11/26 at
299.
In contrast, Dr. Sackett did not consider the age of a claim in his methodology
and did not distinguish the approximately 850 claims filed before 2015 from those
claims filed within five years of the petition date. Tr. 8/10/26 at 51-56; Tr. 8/11/26 at
56-58. Though Dr. Sackett did not use a resolution rate, he agreed that the tort
system extrapolation methodology can consider that not all filed claims are
resolved. Tr. 8/10/26 at 113. He explained that he did not include a resolution rate
in his calculations because he had seen, in other contexts, that old claims do get
resolved. He reasoned that the resolution rate can vary by year, as well as the mix
of claims filed that year, so he did not consider it helpful. Dr. Sackett reached that
conclusion not based on anything in the Debtors’ claims database, but on his
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 34 of 64

35
“experience forecasting claims,” primarily in the trust system. Tr. 8/10/26 at 120-21.
He further argued that if he had included a resolution rate, it would only result in
the exclusion of a “couple percent of claims,” an amount he found not significant. Tr.
8/10/26 at 53-54.
Dr. Martin offered rebuttal testimony on Dr. Sackett’s failure to use a
resolution rate, describing his failure to do so as a foundational deviation from
accepted methodology. Dr. Martin described the known phenomenon of stale
pending claims, that is, those claims that have been pending for some time with no
activity. She explained that one can determine the period of time it takes for a claim
to become stale by looking at a specific defendant’s database and calculating the
time from filing a claim to resolution, year by year, until there is little resolution
activity. Tr. 8/13/26 at 218-20. Here, instead of looking at the Debtors’ database to
determine the time from filing to resolution and identifying the number of years
after which nothing happened, Dr. Sackett assumed that every claim would come to
a resolution and “assigned zero claims to the abandoned bucket.” In Dr. Martin’s
opinion, by not considering the possibility that some claims are abandoned and
instead assuming every single claim would reach a resolution, Dr. Sackett assigned
substantial value to stale, otherwise valueless claims and erroneously inflated his
estimate. Tr. 8/13/26 at 222-23.
The Court finds Dr. Mullin and McKnight appropriately accounted for both
stale claims and the failure of a small percentage of claims to resolve and agrees
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 35 of 64

36
with Dr. Martin that Dr. Sackett’s failure to acknowledge that some claims are
abandoned inflated his estimate.22
2. 	Pending unknown disease claims
For a significant number of claims in the Debtors’ database, the disease of the
claim is listed as unknown. Approximately 3,700 pending claims against Aldrich
and 3,500 pending claims against Murray in the database do not identify a disease.
The parties dispute how these unknown claims should be accounted for in
estimation.
In his estimation, Dr. Sackett first identified the percentage of pending
known disease claims that are designated as mesothelioma claims and then
assumed that a similar percentage of pending unknown disease claims would
likewise be mesothelioma. Doing so led him to recategorize an additional 1,108
pending unknown claims against Aldrich as mesothelioma claims and an additional
678 pending unknown claims against Murray as mesothelioma claims. Tr. 8/10/26
at 79. As with pending known mesothelioma claims, Dr. Sackett did not consider
the filing date of any unknown claim in making his calculations. Nevertheless, Dr.
Sackett conceded that the vast majority of claims filed in the early 2000s were
nonmalignant cases (non-mesothelioma). TR. 8/10/26 at 128-30.
Both Dr. Mullin and Dr. Martin found fault with Dr. Sackett’s reclassification
of unknown disease claims to mesothelioma claims. As he did with pending
22 Austin did not address Dr. Mullin’s resolution rate or his treatment of long pendency claims in her
rebuttal testimony, nor did she critique McKnight’s consideration of the age of a claim in his
analysis.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 36 of 64

37
mesothelioma claims, Dr. Mullin reviewed unknown disease claims in the Debtors’
database to consider the date each claim was filed. Of the roughly 7,000 pending
unknown disease claims, he found that only 87 were filed within five years of the
petition date while the rest were long pendency claims that would only account for
1% of the Debtors’ payments. Tr. 8/11/26 at 58-60.
Dr. Mullin also considered the pending unknown disease claims in the
context of the evolution of the claims mix in the Debtors’ data. From the 1990s
through 2006, nonmalignant claims far exceeded mesothelioma claims; by 2007, the
Debtors’ data tracks the larger asbestos tort system transition to the modern
mesothelioma era, where the opposite is true. The vast majority of the unknown
disease claims in the Debtors’ database were filed during the nonmalignant era
between 1993 and 2005. Tr. 8/11/26 at 60-61; Tr. 8/10/26 at 220-21.
Dr. Mullin provided further support for his contention that pending unknown
disease claims are of minimal value by detailing the matching process he performed
between the Aldrich and Murray claims databases and the public Garlock claims
database.23 Tr. 8/11 at 66-70. By comparing the disease categorization of the same
23 The ACC objected to certain testimony of Dr. Mullin regarding his matching of the Debtors’
unknown disease claims on the basis that it was not disclosed as required by Fed. R. Civ. P.
26(a)(2)(B), which requires timely disclosure via written report of a testifying expert’s opinions and
the facts or data considered in forming them. Tr. 8/11 at 62-64. Corresponding Fed. R. Civ. P. 37(c)(1)
provides that information not disclosed as required by Rule 26(a) cannot be used at trial unless the
lack of disclosure was justified or harmless. Having considered the testimony offered by Dr. Mullin
and the parties’ post-trial submissions on Rule 26 issues [Dkt. Nos. 3347, 3350], the Court will
overrule the ACC’s objection. The disclosure of Dr. Mullin’s reliance materials, including his
deduplicated claims database, the “Deduplicated claims data with reconciled group settlement
info.dta” dataset, and a script demonstrating his matching process across the databases relevant
here, is sufficient under Rule 26 to allow his testimony regarding the matching process he
performed, including the basic arithmetical conclusions he derived from the resulting data (e.g., the
number of claims matched and the percentage of those claims identified as mesothelioma claims),
which was provided by Bates White. Dr. Mullin’s testimony (in the same line of questioning) is also
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 37 of 64

38
claims across multiple databases, one can identify instances where a claim
designated as an unknown disease claim in one database is designated a
mesothelioma claim (or otherwise) in another. Matching pending unknown disease
claims across databases in this way offers insight into the likelihood that any such
claim from a given database is a mesothelioma claim, making it easier, in turn, to
reliably estimate the value of that set of claims. Dr. Mullin testified—and Dr.
Martin’s testimony supports the notion—that attempting to match unknown
disease claims across databases is part of the standard approach to estimating
liability in this context. Tr. 8/11/26 at 68; Tr. 8/13/26 at 209.
By matching the Debtors’ unknown disease claims across their respective
databases and the public Garlock claims database (in addition to filed PIQs), Dr.
Mullin was able to identify about half of the Debtors’ roughly 140,000 total
unknown disease claims by disease type, including about half of the approximately
7,000 pending unknown disease claims. He observed that about 95% of the claims
that became identifiable by disease type after this matching process were non-
malignant and that only about 2.4% were mesothelioma claims, both as to the total
number of matched claims and the much smaller set of matched pending claims.
These observations increased Dr. Mullin’s confidence that the Debtors’ pending
permissible, of course, to the extent he simply describes what Dr. Sackett did to account for unknown
disease claims in his estimation. The critique Dr. Mullin makes of Dr. Sackett’s conclusions in the
portions of testimony subject to the objection reflects his assessment that Dr. Sackett overstated
unknown disease claim values based on an imputation error and does not explicitly source any
criticism to a matching process (or failure to implement a matching process), which might have
caused his testimony to fall afoul of Rule 26. Finally, the inclusion of Dr. Mullin’s matching
testimony, even were it undisclosed, is harmless because it is only one of several significant reasons
to doubt Dr. Sackett’s evaluation of unknown disease claims. For these reasons, the ACC’s objection
to Dr. Mullin’s testimony about matching unknown disease claims is overruled.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 38 of 64

39
unknown disease claims are of minimal value because his matching process
essentially reduced the total number of pending unknown disease claims for which
estimation was necessary while simultaneously showing that those claims are very
unlikely to be mesothelioma. Tr. 8/11/26 at 68-69.
While Dr. Sackett was aware that matching claims across databases to
account for unknown disease claims has been done in previous instances of
estimation—including by his own firm, LAS—he was not certain whether the data
he relied on to evaluate unknown claims against the Debtors reflected a matching
process. Tr. 8/10 at 133-34. In addressing Dr. Sackett’s opinion on this issue, Dr.
Martin explained that matching unknown disease claims across databases in the
context of estimation is one way that experts have determined, generally, that such
claims are unlikely to be mesothelioma claims. She observed that in every publicly
available report she could find from LAS, it had used a matching method to
reclassify unknown disease claims as known disease claims and found that only a
very small percentage were mesothelioma claims. Her observation on that point is
consistent with Dr. Mullin’s testimony that less than 3% of the Debtors’ pending
unknown disease claims are likely to become mesothelioma claims. It strongly
suggests that Dr. Sackett’s decision to treat nearly 30% of pending unknown disease
claims against Aldrich and 19% against Murray as mesothelioma claims resulted in
a gross overestimation of the value of that group of claims. Tr. 8/13 at 208-17.
Dr. Martin also described the theory that claimants with mesothelioma are
very likely to specify that disease in a complaint because it is the disease most
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 39 of 64

40
closely connected to asbestos, is highly fatal, and carries a higher dollar value than
other diseases. There is evidentiary support for that theory in the form of claims-
matching between defendants or asbestos trusts, as well as long-term monitoring of
claim databases, which illustrate the gradual transition of claims from unknown to
known diseases. Based on that evidence and her own experience working with
asbestos defendants’ databases, Dr. Martin concluded that claims initially labeled
as an unknown disease almost never transition to, or later become identified as,
mesothelioma claims. Tr. 8/13 at 207-09.
McKnight’s view of unknown disease claims was consistent with Dr. Mullin
and Dr. Martin. He reviewed those claims and observed that they were very old. In
his experience, long-pending unknown disease claims will not be mesothelioma
claims. Therefore, McKnight treated the unknown disease claims as non-
mesothelioma claims, which would be accounted for by the multiplier. Tr. 8/11/26 at
302-03.
The Court finds Dr. Mullin’s, Dr. Martin’s, and McKnight’s testimony on
pending unknown disease claims highly persuasive. By overlooking key features of
that claim set, such as the high proportion of long-pending claims and their
relationship with historical trends in the claims mix, Dr. Sackett significantly
overstates their probable value.
3. 	Tolled/Off-Complaint/Administrative Claims
Prior to bankruptcy, the Debtors entered into administrative settlement
agreements with three law firms that provided for the resolution of pending and
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 40 of 64

41
future asbestos claims from claimants represented by those firms through an
administrative procedure rather than a complaint filed in the tort system (taking
them “off-complaint”). These agreements provided for an “annual settlement
payment limitation,” in other words, a yearly cap, of the amount to be paid to the
firms to satisfy claims held by each law firm’s clients. Payments under the
agreements were made as each firm submitted individual claims through the
administrative process. As part of these agreements, the statute of limitations for
the individual claims was tolled for their duration, without the need to file a
complaint. ALD-D6; ALD-D7. Claims presented to the Debtors for payment under
these agreements, 80% of which came from “law firm #3,”24 appear in the Debtors’
database as paid claims. However, claims held by clients of off-complaint firms that
were not presented to the Debtors for payment under the agreements are not
reflected in the database. Tr. 8/10/26 at 84; Tr. 8/11/26 at 78-79. The experts
diverge in their opinions on the proper treatment of claims connected to these
administrative agreements, especially those for which a PIQ was filed but which
were not presented to the Debtors for payment and are therefore not in the
database.
In his testimony, McKnight described the impact of these administrative
agreements on the Debtors’ historical data. When a firm enters into such an
agreement, the number of its clients’ claims the Debtor ultimately pays stays about
24 The Court entered a stipulated order prior to the Phase I Trial that required the parties to use
anonymous identifiers for certain identifying information during the trial. [Dkt. No. 3265].
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 41 of 64

42
the same, but since the firm stops filing all its claims as it had previously done in
the traditional tort system, the filing rate, payment rate, and propensity to sue from
those firms will not be indicative of what would have occurred without the
agreement. Tr. 8/11/26 at 291-93. Wary of the data distortions the off-complaint
claims cause, McKnight excluded them from his claim forecast analysis. Instead, he
calculated a forecast only for “on-complaint” claims and then applied a multiplier of
1.6 to account for the dollars being paid to off-complaint firms based on the Debtors’
historical ratio between payments to on-complaint and off-complaint firms. Tr.
8/12/26 at 89, 167-68.
Austin took issue with McKnight’s use of a multiplier, which imputes the
behavior of the on-complaint claims onto the off-complaint claims rather than
modeling the off-complaint claims on their own trend. Tr. 8/12/26 at 171; Tr. 8/13/26
at 68-69. She believed the use of a fixed 1.6 multiplier was inappropriate given
McKnight’s calculation that on-complaint claim values were decreasing while,
under her own assessment, off-complaint claim values were “fairly steady” or
“slightly increasing.” Tr. 8/13/26 at 62-64. Austin did not address, however, why the
off-complaint and on-complaint settlements would not be similarly impacted by
their shared claimant pool’s advancing age and declining exposure levels.
For his part, Dr. Sackett theorized that the administrative agreements left a
pool of tolled, unfiled claims against the Debtors that were viable for future
compensation. Specifically looking at law firm #3, Dr. Sackett suspected, given that
it came very close to hitting the payment cap under its administrative agreement
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 42 of 64

43
each year, there were additional mesothelioma claims it did not file. In other words,
had law firm #3 not reached its annual limit, it would have filed more
administrative claims for payment. Instead, it simply held those claims. As further
evidence in support of his theory, he noted that prior to entering the agreement, law
firm #3 filed about 12% of the mesothelioma claims against the Debtors, but once
the agreement was in effect, it filed only 4% (through the administrative, off-
complaint process), an 8% decline. Tr. 8/10/26 at 83-84.
To confirm his theory that law firm #3 did not submit all claims it believed
should be paid to the Debtors for payment, Dr. Sackett reviewed the PIQs
submitted by law firm #3 for pending claims. He found that it had submitted 806
PIQs for pending Aldrich claims and 944 PIQs for pending Murray claims that were
not in the Debtors’ database because they were off-complaint, or “tolled claims.” Dr.
Sackett counted these PIQs, as well as others submitted by individuals not
appearing in the Debtors’ database, as additional pending claims. He then
estimated their value by applying the same payment rate he used for on-complaint
database claims. Tr. 8/10/26 at 140-43; Tr. 8/11/26 at 82. Accounting for off-
complaint claims in this way increased Dr. Sackett’s pending and future claim
counts by approximately 10% and his estimate of the Debtors’ liability by $128.5
million. Tr. 8/10/26 at 84-86; Tr. 8/11/26 at 93-94.
Again, consistent with most of his calculations, Dr. Sackett did not consider
the age of the PIQ claims in his calculations, though he agreed that some of them
were from before 2010, which was prior to any administrative settlement
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 43 of 64

44
agreements and not covered by any tolling provision. Tr. 8/10/26 at 139-41. Dr.
Sackett also agreed that prior to the agreements, a percentage of claims filed by law
firm #3 were dismissed, though he did not know what percentage of claims. He
nevertheless maintained that firms submitted PIQs only for claims they think have
value, perhaps to support his treatment of unsubmitted off-complaint claims as
compensable at the same rate and amount as on-complaint claims. Tr. 8/10/26 at
144-45. When this view was challenged with the suggestion that law firm #3’s
submission of PIQs for settled claims showed it did “not do a lot of filtering,” Dr.
Sackett merely stated, “you know, people make mistakes.” Tr. 8/10/26 at 145.
Dr. Mullin prepared both an affirmative opinion on the off-complaint claims
and rebuttal to Dr. Sackett’s treatment of off-complaint claims. In doing so, he
reviewed the dismissal rate for claims filed by law firm #3 over a three-year period
prior to its administrative agreement with the Debtors and found that in the tort
system, 36.2% of its claims resulted in settlement and 63.8% in dismissal. Diverging
slightly from Dr. Sackett, he calculated that law firm #3 filed 13.8% of the
mesothelioma claims against the Debtors prior to its administrative agreement.
This led Dr. Mullin to observe that, prior to its agreement, claims filed by law firm
#3 and later settled accounted for about 5% of total claims against the Debtors,
while those later dismissed accounted for roughly 8%. Dr. Mullin then concluded
that the 8% decline in filed claims following the administrative agreement between
the Debtors and law firm #3 simply reflects the historical tort system dismissal rate
for law firm #3’s claims—claims that would have been dismissed in the tort system
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 44 of 64

45
were now simply missing from the database by function of the agreement. Tr.
8/11/26 at 80-83. In other words, the numbers reflect the administrative agreement
working as planned: “[t]he compensable claims get presented and paid each year,
the noncompensable claims go unfiled, and the percentage of unfiled claims
corresponds almost perfectly to the percentage of claims that before the agreement
resulted in dismissals.” Tr. 8/11/26 at 83. Like Dr. Sackett, Dr. Mullin also reviewed
the PIQs filed in these cases, observing that law firm #3 filed almost half of them
despite filing just 14% of total claims prior to the agreements.25 He noted that while
some law firms appeared to filter their inventory of pending claims and file PIQs
only for those they believed were compensable, law firm #3 appeared to file PIQs for
virtually every claim they had handled, including dismissed and already-paid
claims. Tr. 8/11/26 at 88-89.
25 The ACC objected to certain testimony of Dr. Mullin regarding his analysis of Law Firm #3’s PIQ
filtering on the basis that it was not disclosed as required by Fed. R. Civ. P. 26(a)(2)(B), which
requires timely disclosure via written report of a testifying expert’s opinions and the facts or data
considered in forming them. Tr. 8/11/26 at 86-88. Having considered the testimony offered by Dr.
Mullin and the parties’ post-trial submissions on Rule 26 issues, [Dkt. No. 3347], the Court will
overrule the ACC’s objection. Dr. Mullin’s evaluation of PIQs in relation to off-complaint claims and
the opinions he develops therefrom are thoroughly discussed in his corrected initial report, including
his observations regarding the PIQ “filtering” differences between various firms. See Corrected
Initial Expert Report of Charles H. Mullin, Ph.D. (Sept. 15, 2025) at Appendix E. They were also
discussed during his deposition. See Mullin Dep. at 257:15–258:13. Even if this disclosure were
somehow technically insufficient under Rule 26 as rebuttal testimony because of its absence from Dr.
Mullin’s rebuttal report, nearly all of it would still stand as affirmative opinion testimony based on
his corrected initial report. This is, of course, because Dr. Mullin was retained by the Debtors as both
an affirmative and rebuttal expert witness in this proceeding. Finally, to the degree that some small
portion of Dr. Mullin’s discussion of PIQs to rebut Dr. Sackett’s testimony might not have been
adequately disclosed, the Court finds that lack of disclosure to be harmless. The gist of Dr. Mullin’s
criticism of Dr. Sackett’s treatment of unknown claims is clear from the reports and was known by
the ACC months in advance of trial. Dr. Mullin’s deposition testimony provided further clarity on his
analysis of PIQ filtering. And, importantly, excluding Dr. Mullin’s limited testimony about law firm
#3’s purported failure to filter claims would not impact the Court’s general assessment of the
inadequacy of Dr. Sackett’s analysis of off-complaint claims. Therefore, the ACC’s objection is
overruled.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 45 of 64

46
The Court has concerns about Dr. Sackett’s treatment of PIQs for off-
complaint claims in his calculations. His theory that law firm #3 would hold back
hundreds of valid mesothelioma claims entitled to payment over years because it
had reached the annual cap in its administrative agreement with the Debtors defies
common sense and suggests law firm #3 entered into an agreement that conflicted
with its clients’ interests. Further, his failure to even consider law firm #3’s
dismissal rate prior to its agreement or the age of the firm’s PIQ claims in the
context of valuing the off-complaint claims reflects his more rudimentary approach
to estimation. Thus, the Court finds the very impetus for his theory that the unfiled
claims are of equal value to filed claims, that is, the fact that firms spent up to their
annual cap each year and the reduction in claims by law firm #3 from 12% of the
total to 4% of the total after it entered into the administrative settlement
agreement, to be flawed. The Court instead finds that the approaches of McKnight
and Dr. Mullin more accurately assess the current and future value of the off-
complaint claims for purposes of estimation. Though distinct, each approach
accounts for the distortions in the data created by the administrative agreements by
making reasonable assumptions and drawing appropriate inferences based on
information available in the Debtors’ claims database and PIQs.
4. 	Payment Rate/Data Truncation
Despite the parties’ general agreement regarding the reasonableness of a
three-year calibration period, Dr. Sackett made a related analytical choice that
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 46 of 64

47
noticeably boosted his payment rate. According to McKnight and Dr. Mullin, that
approach was both erroneous and statistically significant.
Dr. Mullin calculated pay rates, not including the off-complaint claims, for
ten, five, and three-year calibration periods on a law firm by law firm basis, as the
rates vary significantly. Tr. 8/11/26 at 13. His pay rate is calculated as the
percentage of claims resolved in that calendar year that received payment. Tr.
8/11/26 at 10. Using a three-year calibration period, which aligned with Dr. Sackett,
Dr. Mullin found a 38% pay rate for Aldrich and a 16% pay rate for Murray. He
found a 1% difference for Murray using a five-year calibration period, but Aldrich
remained the same. Tr. 8/11/26 at 12-13.
Dr. Sackett calculated a cumulative payment rate of 49.3% for Aldrich during
his same three-year calibration period from 2017-2019 and a 21.4% rate for Murray.
Like Dr. Mullin, Dr. Sackett first calculated the percentage of claims resolved
during the three-year calibration period, but, critically, he omitted any claim that
was filed before 2013. Tr. 8/10/26 at 54-56. This decision, which in statistical terms
is known as data truncation, was significant; had Dr. Sackett used same calibration
period but included claims filed prior to 2013, the payment rate would have been
45.6% for Aldrich and 19.5% for Murray. Dr. Sackett’s use of a truncated data set
has a significant impact; his payment rate is approximately 7.5% more than it
should have been for Aldrich and approximately 20% more for Murray. Tr. 8/11/26
at 74-75. Dr. Sackett ’s stated rationale for excluding the claims filed prior to 2013
was that the “tort system changes over time” and he wanted to ensure “the
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 47 of 64

48
settlements that [he] was using were the . . . closest in time to the pending claims
and the future claims.” He could not identify any data in the Debtors’ history
supporting this methodology, instead pointing to a vaguely described
“understanding of potential changes in the tort system.” Tr. 8/10/26 at 55-56; 123-
24.
The difference in percentages is even greater in comparison to McKnight’s
payment rate. Like Dr. Sackett, McKnight calculated the payment rate for claims
filed from 2013 to 2020 and found that the overall payment rate, excluding off-
complaint claims, was 35.4%. But because he used a competing risk model that
looks at the probability of a filed claim getting paid, rather than a component-based
forecast, the truncation effect is not at issue.26 As part of his rebuttal of Dr.
Sackett’s opinion, McKnight noted the truncation error or bias in Dr. Sackett’s
payment rate, the result of excluding pre-2013 filings in component-based forecast.
Tr. 8/12/26 at 81-82.
Initially, the Court observes that all settlements the Debtors reached with
claimants from 2017 to 2019 are equivalently occurring within the calibration
period, regardless of when the claim was initially filed. Dr. Sackett was unable to
articulate a reason why two settlements occurring in 2019—with one claim filed in
2012 and another in 2013—should be treated in such a strikingly different fashion.
Dr. Sackett’s approach also lacks consistency with his other analytical choices as
26 McKnight used a period from 2014 to 2020 in his September 2025 report but, in his March 2026
report, extended the period to 2013 to better conform with Dr. Sackett’s calibration period for his
payment rate. Tr. 8/12/26 at 22. Nevertheless, McKnight noted that there are some attributes that
are not comparable to other experts because of the implementation differences. Tr. 8/12/26 at 26.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 48 of 64

49
well his own testimony. By truncating the data, he calculated a pay rate based only
on more recent claims, but he then applied it to any claims filed prior to 2013. He
also refused to concede that older claims are less likely to be paid, while
simultaneously excluding pre-2013 claims from his payment rate calculation due to
ill-defined “changes over time.” Tr. 8/10/26 at 55, 120-22. Lastly, the Court notes
that Dr. Sackett’s data truncation bias is particularly troubling given that he makes
efforts to include a large number of claims from pre-2013 in his total number of
pending mesothelioma claims, including (1) long pendency claims, (2) unknown
disease claims, and (3) “tolled” or off-complaint claims.
5. 	Verdicts
Over the course of decades of asbestos litigation including approximately
50,000 filed claims, all claims against the Debtors have either settled, been
dismissed, or remain unresolved save for one resulting in a $7 million jury verdict
against Aldrich in 2016, which, according to the Debtors’ database, was ultimately
resolved and paid in late 2017. Tr. 8/10/26 at 71. Dr. Mullin, McKnight, and Dr.
Martin explained that, because the payment on the verdict is in the Debtors’
database, along with any subsequent impact on settlements values, the possibility
of future verdicts is accounted for in the post-verdict settlement data. Tr. 8/11/26 at
43; Tr. 8/12/26 at 85-86; Tr. 8/13/26 at 225. In fact, McKnight specifically used a
five-year calibration period to capture the 2016 verdict, opining that was an
appropriate way to estimate the likelihood and value of future verdicts based on the
Debtors’ database. 8/12/26 at 85-86. The Court agrees.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 49 of 64

50
Dr. Sackett testified that, in his opinion, the existence of the verdict in the
database suggested a change in the Debtors’ tort system history. He considered his
experience at asbestos litigation conferences and opined that once there is a verdict,
other firms see it and are aware of it. Though recognizing that four years had since
passed without a subsequent verdict, he nevertheless assumed that the Debtors
would be liable for six future verdicts, three for Aldrich and three for Murray. As for
the amounts of the forecasted verdicts, Dr. Sackett relied on the average
mesothelioma verdict against other defendants over the last three postpetition
years, 2023 through 2025, which he calculated to be $15.3 million. Tr. 8/10/26 at 71-
73, 106-08.
Dr. Mullin rebutted Dr. Sackett’s opinion, pointing out that one verdict out of
the 50,000 claims in the past forecasts a 75% chance of one verdict in the remaining
38,000 future claims. He also took issue with the dollar amount of the projected
verdicts, noting Dr. Sackett’s source publication is based almost entirely upon high-
end verdicts and does not account for downward adjustments—such as offsets and
apportionment between multiple defendants—that markedly reduce the amount
ultimately paid to claimants. Tr. 8/11/26 at 44-46.
Dr. Martin similarly disagreed with Dr. Sackett’s opinion, pointing out that it
departed from accepted methodology and the tort system experience by layering on
six hypothetical verdicts, and explaining that settlement amounts embed verdict
risk because the parties determine the settlement amount by a cost benefit analysis.
As part of her analysis, she also reviewed all publicly available LAS reports and
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 50 of 64

51
found they all used average settlement values, not hypothetical future verdicts, to
estimate liability for pending and future claims, unlike Dr. Sackett. Tr. 8/13/26 224-
26. Dr. Sackett was also not aware of any reports in which LAS calculated verdicts
separately in a tort system extrapolation. Tr. 8/10/26 at 111-112.
During Dr. Sackett’s testimony, the Debtors and FCR objected on the basis
that his testimony regarding jury verdicts was not a Phase I issue because it was
not based on tort system extrapolation. Tr. 8/10/26 at 69-70. The Court overruled
that objection, interpreting the scope of Phase I broadly and finding that Dr.
Sackett did tether his opinion on future verdicts to the payment of the one verdict in
the database. Tr. 8/11/26 at 6-7. However, upon review of the record, the Court will
exercise its gatekeeping role under Rule 702 and exclude Dr. Sackett’s opinion
about future verdicts against the Debtors on the basis that the testimony is not the
product of the application of reliable principles and methods to the facts. Fed. R.
Evid. 702.27 “An expert’s opinion should be excluded when it is based on
assumptions which are speculative and are not supported by the record.” Tyger
Constr. Co. v. Pensacola Constr. Co., 29 F.3d 137, 142 (4th Cir. 1994) (citation
27 Though the FCR joined the Debtors in arguing that Dr. Sackett’s estimate of six additional
verdicts was beyond the scope of Phase I, the FCR’s objection—particularly counsel’s remark that
“we can’t imagine verdicts”—can also be framed as a reliability challenge under Rule 702(c). Tr.
8/10/26 at 70. Regardless, the Court can consider the admissibility of expert testimony sua sponte “so
long as the court has an adequate record on which to base its ruling,” Miller v. Baker Implement Co.,
439 F.3d 473 (8th Cir. 2006), and “if the materials before the court make it clear that the proffered
testimony is unreliable, exclusion of the testimony is appropriate even in the absence of an
evidentiary hearing.” 4 W EINSTEIN’S FEDERAL E VIDENCE § 702.02(6)(c) (2026); see also Koehler v.
Infosys Techs. Ltd., 181 F.4th 769, 778 (7th Cir. 2026). The ACC presented Dr. Sackett’s full
testimony on his six projected verdicts, in the face of objections and strenuous cross-examination,
ensuring the Court has more than an adequate record on which to assess the reliability of Dr.
Sackett’s opinion.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 51 of 64

52
omitted). As described above, Murray does not have a single verdict in its tort
system history, and Aldrich has just one, with their combined history of 50,000
claims, yet Dr. Sackett projected six verdicts among the 38,000 forecasted future
claims. While the basis for this forecast is technically the verdict in the Debtors’
database, the Court can discern no reliable statistical and economic principles and
methods applied to the facts here.
Dr. Sackett described the 2016 verdict as “a change in the tort system” that
would make it “more likely that [plaintiffs’ firms] would try for that in the future.”
Tr. 8/10/26 at 71, 105. The Court has carefully reviewed the record, and in forming
his opinion, Dr. Sackett did not rely on any facts or data in the Debtors’ tort system
history after the 2016 verdict to support his assertion that the verdict caused a
change in the tort system, nor did he point to any specific experience he has had
with another defendant, or even observed in the course of his work, that exemplifies
the trend he predicts. And to the extent that Dr. Sackett’s conclusion was based on
his prior experience as an asbestos litigation attorney, he failed to “explain how his
experience leads to the conclusion reached, why his experience is a sufficient basis
for the opinion, and how his experience is reliably applied to the facts.” United
States v. Wilson, 484 F.3d 267, 274 (4th Cir. 2007) (quoting Fed. R. Evid. 702
advisory committee’s note).
Although “Rule 702 is flexible . . . [that] flexibility does not eliminate the
requirement that the expert reliably connect her principles and methods to the facts
of the case.” Near v. Enerco Grp., Inc., 182 F.4th 341, 362 (4th Cir. 2026) (citing
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 52 of 64

53
Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 594 (1993)). The ACC failed to
show by a preponderance of the evidence that Dr. Sackett’s estimate of six
additional verdicts is the product of reliable principles and methods; therefore, the
Court will exclude his opinion as to the verdicts. Alternatively, even if the testimony
were not excluded, the Court would give it no weight, as the number of hypothetical
verdicts was purely speculative, the forecasted amounts were not based on the
Debtors’ settlement and verdict history, and any heightened risk of an adverse
verdict would be accounted for in the average settlement value.
6. 	Average Settlement Value Trend (Future Projection)
The parties dispute whether average settlement values are declining or
remaining flat over time. Based on his three-year calibration period, Dr. Sackett
concluded that the Debtors’ average settlement values are stable. Tr. 8/10/26 at 147.
Dr. Mullin gave extensive testimony on factors that affect average settlement
rates over time, which, while very helpful in gaining a deeper understanding of the
subject matter, the Court will only describe briefly here. He described, as it is
generally known, prices rise over time due to inflation. At the same time, the
average asbestos claimant is getting older, which reduces the value of their claim,
as the average settlement value is largely compensation for pain and suffering and
lost life. Lastly, as time goes by, the average asbestos claimant has had fewer years
of occupational exposure to asbestos-containing gaskets, further reducing the value
of the claim against the defendant. Tr. 8/11/26 at 20. Dr. Mullin reviewed the
Debtors’ data to ascertain whether these factors applied specifically to them and
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 53 of 64

54
found that the average age of a paid claimant has risen over time from being in
their late 60s in the early 2000s, to nearly 75 years old presently. Dr. Mullin
reviewed each Debtor’s average settlement value over the last 14 years and
observed that for Murray, the trend line shows a decline over time averaging about
$10,000 every four years. Tr. 8/11/26 at 35. As for Aldrich, the average settlement
value either declined slightly or remained flat, reflecting the impact of the verdict.
Dr. Mullin calculated alternate total estimates using either a declining trend line or
a flat trend line, conceding that the decline is small for Aldrich. Tr. 8/11/26 at 35-41.
McKnight’s regression analyses further confirmed the prevailing impact of
claimant age on settlement values, what he characterized as “the well-known fact in
these mass tort litigations that age – as claimants get older, they tend to get paid
less.” Tr. 8/12/26 at 119-20. Given the fixed, aging population of individuals exposed
to the Debtors’ asbestos-containing products and finding that other variables such
as a claimant’s state or law firm were unlikely to change with a mature tort like
asbestos, McKnight found that “for every year older a claim[ant] gets . . . their
average settlement values go down by about 1 percentage point.” Tr. 8/12/26 at 122.
Austin critiqued McKnight’s opinion because he only controlled for age when
he performed his regression analysis looking at claim value when there were
multiple variables in the data that could affect claim value. Tr. 8/12/26 at 248-49.
The Court found McKnight’s explanation as to why he purposefully controlled for
age—because he knew age would change over time while there was no reason to
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 54 of 64

55
believe the other variables (such as venue and law firm) would change—to be
helpful and persuasive.
7. 	Discount Rate
All experts agree that a discount rate must be applied to the forecasted
liability to account for the fact that claims will be paid out over 40 years into the
future. Dr. Mullin and McKnight agree that the discount rate with the appropriate
trust rate of return should be 6.11%28 or, if using a risk-free rate of return, 4.5%.
There was extensive testimony by all experts on whether to use a trust rate of
return or a risk-free rate, with compelling arguments on both sides. But, given the
context and circumstances presented here, as Dr. Mullin recognized when he
alluded to public policy choices, Tr. 8/11/26 at 112, and Austin effectively discussed
in some detail during her testimony,29 the Court agrees with Judge Hodges’ decision
in Garlock on this point and finds the long-term risk-free rates appropriate. In re
Garlock Sealing Techs, 504 B.R. at 97; see also In re Federal-Mogul Global, Inc., 330
B.R. at 163-64; In re Eagle-Picher Indus., 189 B.R. at 692.
28 Dr. Mullin adjusted the trust rate of return he used in his affirmative report down slightly to
match McKnight’s proposed trust rate of return, testifying that McKnight’s opinion was based on a
“full-fledged” portfolio analysis for purposes of simplicity and because it was “arguably preferable.”
Tr. 8/11/26 at 111.
29 Austin questioned the methodology used by McKnight and Dr. Mullin in deriving the trust return
rate, noting it was unclear how the sampled trusts were selected, why other trusts were excluded,
and why McKnight looked at the expected return for certain asset classes held by the trusts, rather
than the actual returns on investment. More significantly, Austin questioned whether prospective
trust returns are a relevant metric for determining the appropriate discount rate in this Phase I
estimation, where future asbestos claims are valued as if they remained in the tort system and
where no hypothetical trust yet exists. Tr. 8/12/26 at 234-238.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 55 of 64

56
In calculating his proposed risk-free rate, McKnight took the yield curve as of
June 2025, when the discount rate on 10-year Treasuries was 4.5%, which was
equal to the 100-year average rate of return and indicative of what he considered a
normal interest rate environment. He used the full yield curve to account for a
lower discount rate for liabilities coming due in a short period of time and an
increased rate for liabilities into the future. For an inflation rate, McKnight used
the 2% inflation rate targeted by the Federal Reserve. Tr. 8/11/26 284-86; Tr.
8/12/26 at 70. In his range of estimates, Dr. Mullin also includes projections using
the same 4.5% risk-free rate of return. Tr. 8/11/26 at 107-112, 169-70. McKnight
clarified the distinction between discounting something to the petition date,
meaning the date to which one discounts, and choosing the appropriate rate for that
discounting. He noted that, over any long period of time in the U.S. economy,
interest rates are higher than inflation rates.
Dr. Sackett calculated his proposed discount rate for a period through 2059
by using a blended rate of inflation of 3.25%—derived from the average of the
Consumer Price Index (CPI), the Medical Care CPI, and the Employment Cost
Index from 2020-2025—and a 1% risk-free rate of return, which is a blend of the
rates as of June 2020 on spot 3-month, 10-year, and 30-year United States Treasury
yields. The combination of these two variables left Dr. Sackett with a net discount
rate of negative 2.2%. Tr. 8/10/26 at 89-90, 163.
Dr. Sackett is not an economist and did not choose the rate in consultation
with an economist at LAS. He explained that he used the spot rates because he was
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 56 of 64

57
valuing the claims from the perspective of June 2020, the petition date. Dr. Sackett
did not provide a meaningful explanation for why he felt it appropriate to use
postpetition data to calculate an inflation rate but not to calculate a discount rate.
Dr. Sackett’s testimony also reflects that he did not consider the changing
characteristics of the claims over time, though he agreed that as time passes, the
claimants are both getting older and there are “more years” between the claimants
and the time when asbestos left the marketplace. Tr. 8/10/26 at 149. And though he
was unable to identify any economic period where negative real interest rates
persisted for over thirty years, Dr. Sackett applied a negative 2.2% discount rate to
his projection of future asbestos claim values, believing that a negative rate was
“not unprecedented” and that this specific situation should not be compared to “the
entire American economy.” Tr. 8/10/26 at 90, 149, 161-163.
McKnight critiqued Dr. Sackett for using the spot rate at the petition date,
which was during the COVID-19 crisis and a historically low interest rate
environment, explaining it not a useful metric to find the discount rate for a series
of liabilities extending many years into the future. He took issue with Dr. Sackett’s
inflation rate as well, which was based on the actual inflation rate in the U.S. using
post-petition data (as opposed to the prepetition data for the discount rate). Lastly,
he challenged Dr. Sackett’s use of a historically low-interest rate prepetition and
high inflation caused by supply chain disruptions postpetition, to calculate a
negative discount rate that he then assumes will persist for the next 50 years. Tr.
8/12/26 at 70-73.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 57 of 64

58
Dr. Martin offered extensive rebuttal testimony addressing Dr. Sackett’s
calculation of a discount rate. Echoing McKnight, she distinguished between the
date to which a liability is discounted, i.e., the petition date, and the rate for that
discounting. 8/13/26 at 240. In her view, because these are long-tailed asbestos
liabilities, Dr. Sackett should have looked at two factors to determine the
appropriate inflation: (1) what is the correct average rate of inflation over the life of
the liability and (2) whether there would be changes in the characteristics of the
claims over time. 8/13/26 at 230. As to the correct average rate of inflation, Dr.
Martin took issue with Dr. Sackett blending CPI with Medical Care CPI and
Employment Cost Index, both of which are higher than CPI. She pointed out that
medical expenses are embedded in CPI, and medical costs are not a significant
driver of claim value. Tr. 8/13/26 at 236. If Dr. Sackett had used CPI, his inflation
rate would have been 2.3% and in line with the Federal Reserve’s target rate.
Dr. Martin also found that Dr. Sackett failed to analyze historical trends or
the impact of the changing claims mix. Dr. Martin reasoned that the dispute
between the parties regarding average settlement value is whether the claims are
flat or declining. In either case, the fact that claim values are not increasing is
evidence that something is tempering the effects of inflation. And as both Dr. Mullin
and McKnight discussed at length, the average age of a claimant is increasing over
time, and claim values decrease with age. 8/13/26 at 234-38.30
30 The ACC objected to the admission of Dr. Martin’s testimony that the value of claims decreases
with claimant age on the basis that it was not disclosed in her rebuttal report as required by Fed. R.
Civ. P. 26(a). Tr. 8/13 at 232. Having considered the testimony offered by Dr. Martin and the parties’
post-trial submissions on Rule 26 issues [Dkt. Nos. 3347, 3350], the Court will overrule the ACC’s
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 58 of 64

59
As with Dr. Sackett’s choice of inflation rate, Dr. Martin found Dr. Sackett
similarly deviated from the accepted method of selecting a discount rate that
reflects economic expectations over the 40-year life of these long-tailed liabilities.
She emphasized the discount rate should embed similar expectations and be
consistent with the chosen inflation rate. In Dr. Martin’s opinion, Dr. Sackett’s
decision to base his discount rate on the daily Treasury yields on the petition date
in June 2020 was not appropriate, as rates were extraordinarily low and not
expected to persist. Tr. 8/13/26 at 242. Like McKnight, Dr. Martin took issue with
Dr. Sackett’s assumption that aberrant rates would persist for 40 years and noted
that Congressional Budget Office projections from 2020, both before and after June
2020, rates to be up to 2% by 2025.31 Tr. 8/13/26 at 247.
McKnight’s and Dr. Martin’s familiarity and depth of understanding of the
subject matter were apparent during their testimony, and the Court found their
critiques of Dr. Sackett both credible and reliable. In pairing an exceptionally low
interest rate from the COVID-19 pandemic—what McKnight described as the
objection. While the ACC accurately observes that Dr. Martin’s rebuttal report does not contain her
opinion that an aging claimant population entails decreasing average claim values over time, the
Court declines to exclude her testimony on the matter, finding the nondisclosure harmless. Dr.
Martin’s opinion about the effect of an aging claimant population on average claim values over time
was part of her deposition testimony months before trial. See Deposition of Denise Neumann Martin,
Ph.D. (May 5, 2026) at 136:10–137:8. Additionally, Dr. Martin was just one of three credible expert
economists to identify this relationship between claimant population age and claim values over time.
Both Dr. Mullin and McKnight provided robust testimony on the impact of claimant age on inflation
rates and opined, as Dr. Martin did, that the average age of claimants against the Debtors increases
over time, which has the effect of decreasing average claim value. Tr. 8/11 at 21-24; Tr. 8/12 at 119-
23. Dr. Sackett did not account for this relationship in choosing an inflation rate. Consequently, the
exclusion of Dr. Martin’s testimony on the issue would have minimal impact on the Court’s overall
assessment of an appropriate discount rate. Therefore, the ACC’s objection is overruled.
31 Rates recovered more quickly than projected in 2020 and were up to 4% by 2023.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 59 of 64

60
lowest rate in the U.S. economy over the last 100 years—with the relatively high
inflation rates in recent years, Dr. Sackett derives and then forecasts a long-term
negative real discount rate, which Dr. Martin characterizes as “unrealistic” and
“impossible.” Tr. 8/11/26 at 285; Tr. 8/13/26 at 255.
Dr. Sackett’s calculation, as he himself concedes, runs counter to the basic
economic principle of the time value of money, the idea that “a dollar today is worth
more than a dollar in the future.” Tr. 8/10/26 at 164. Given the analytical flaws in
Dr. Sackett’s methodology and results, the Court declines to apply his aberrational
negative real discount rate. The Court is left then with the 4.5% risk-free rate
proposed by McKnight and Dr. Mullin, which it finds to be a reliable discount rate
in calculating the present value of the Debtors’ asbestos liabilities for purposes of
estimation.
VI. CONCLUSION
As discussed above, there is, for the most part, a consensus among the
experts on the standard framework for asbestos claims estimation. Many common
areas of disagreement in this context—the appropriate incidence curve, future claim
counts, recent average settlement values, and the calibration period—are either
uncontested or do not result in a statistically significant distinction. Despite this
rare agreement, the experts stridently disagree at times on the proper inputs for the
standard framework as well as specific modeling decisions, echoing prior courts’
observations that “relatively minor variations in underlying assumptions can skew
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 60 of 64

61
the end result enormously.” In re Armstrong World Indus., 348 B.R. at 133 (quoting
Owens Corning, 322 B.R. at 721).
After reviewing the trial testimony and evidence, the Court finds Dr. Sackett
made several dubious methodological choices, which predictably move his overall
and individual component estimates beyond where the Debtors’ history and
accepted economic principles would suggest. It is difficult to discern whether Dr.
Sackett’s analytical choices represent a genuine failure to understand the nuances
of claims estimation or, alternatively, are the product of some ulterior objective. Dr.
Sackett made assumptions that were not supported by the Debtors’ claim history,
presuming that every claim would resolve, for instance, and disregarded trends
apparent in the Debtors’ database, such as the increasing age of the average
claimant. Dr. Sackett also engaged frequently in what could be characterized as
temporal disregard: taking a disease mix from the modern era and imposing it on
unknown disease claims from decades ago; giving the same value to a claim that
has been pending since 2009 as a claim filed in 2018; and generating a negative real
discount rate by combining a historically anomalous, prepetition spot rate from
2020 with actual postpetition inflation through 2025. These issues, along with his
frequent inability either to fully explain the basis for his opinion or to defend his
conclusions, lead the Court to conclude that his opinion should be given little, if any,
weight.
Conversely, the Court finds the estimates of Dr. Mullin and McKnight using
a higher 20% non-ACP curve and a 4.5% risk-free discount rate are both highly
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 61 of 64

62
persuasive and firmly supported by the Debtors’ claim and payment history.
McKnight offers a recommendation of $608 million net present value, using his
model but inputting the forecasted claims number from Dr. Sackett’s LAS incidence
curve. The high end of Dr. Mullin’s preferred estimated range of net present
liability, also using the 20% non-ACP curve that corresponds with that of LAS and
assuming average settlement values for Aldrich remain flat into the future, is $550
million. Tr. 8/11/26 at 117-21; ALD-D5 at 75. And despite employing different
methodologies that make direct comparisons challenging, Dr. Mullin’s forecasts
generate similar ranges to McKnight’s competing risk model. Tr. 8/10/26 at 205-09,
211-13; ALD-D3 at 7-8. The two experts arriving in the same general place, despite
operating independently and with different tools and approaches, confirms the
overall reasonableness of their respective estimates. It was apparent during the
trial that each considered the other’s opinion both competent and well-founded, and
the Debtors did not offer rebuttal testimony to contradict or call into doubt the
FCR’s proposed estimate. Though the Court finds, for the reasons articulated above,
that the risk-free rate is the more appropriate discount rate given the purpose of
this estimation and the current stage of the bankruptcy proceeding, it otherwise
finds McKnight’s and Dr. Mullin’s conclusions persuasive, well-reasoned, and
supported by statistical and econometric principles.
The Court is mindful that “each interested party can be expected to have
selected an expert whose opinions will advance the cause of that party.” See Owens
Corning, 322 B.R. at 721; In re Federal-Mogul Global, Inc., 330 B.R. at 156. But the
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 62 of 64

63
experts’ potential partiality can provide some assistance here in choosing between
the comparable estimates of Dr. Mullin and McKnight. In light of the trust funding
already agreed to by the Debtors, there is no question sufficient funds will be
available to cover the current claimants;32 the critical question concerns the
asbestos liability for future claimants years from now. A future claimants’
representative “serves the interests of his constituents, the future claimants,” In re
Imerys Talc. Am., Inc., 38 F.4th 361, 377-78 (3d Cir. 2022), and “must perform
fiduciary-like duties in his or her very special role of negotiating for individuals who
will be required to participate in a claim-resolution procedure (the trust via the
channeling injunction) that they had no say about.” In re Fairbanks Co., 601 B.R.
831, 838 (Bankr. N.D. Ga. 2019). With that in mind, though the estimates provided
by McKnight and Dr. Mullin are generally comparable and, in the abstract, equally
sound, the Court will give more weight to the slightly higher estimate offered by the
FCR’s expert, McKnight. The FCR in these cases has decades of relevant experience
and is the only party specifically charged with representing the interests of future
asbestos claimants, who stand to lose the most if the eventual trust is insufficiently
funded. Further, given the uncertainty inherent to estimation, resolving that
32 Although the FCR remains free to assert whatever position he believes is appropriate in
estimation, see Amendment No. 1 To Plan Support Agreement [Dkt. No. 2721, Ex. B], he agreed with
the Debtors on a proposed plan years ago with a section 524(g) trust funded in the amount of $545
million, with no less than $125 million available to pay current asbestos claimants. [Dkt. No. 833, at
¶¶ 10-12]. The QSF approved by this Court—and already funded by the Debtors—provided $270
million toward that proposed funding amount. [Dkt. Nos. 994; 2378 at 38]. By comparison, under the
high end of Dr. Mullin’s preferred range, current claims are valued at no more than $115 million. Tr.
8/11/26 at 113-15; ALD-D5 at 74-75.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 63 of 64

64
uncertainty in favor of the higher amount is appropriate under the circumstances of
these cases.
Therefore, under 11 U.S.C. § 502(c), after carefully considering the evidence,
as well as the purpose and limitations of the estimation process, the Court finds the
best estimate of the Debtors’ liabilities for mesothelioma claims using the tort
system extrapolation method as provided by the Court’s Phase I order to be $608
million net present value. The Court then applies the stipulated multiplier of 1.21
for Aldrich and 1.24 for Murray to value the non-mesothelioma asbestos personal
injury claims, resulting in a total estimated liability for purposes of Phase I of this
estimation proceeding of $742,500,000.33
SO ORDERED.
This Order has been signed 	United States Bankruptcy Court
electronically. The Judge’s
signature and Court’s seal
appear at the top of the Order.
33 McKnight’s final recommendation of $608 million did not include a breakdown of the total liability
between Aldrich and Murray, and the parties could not agree on a single multiplier to use across
both Debtors. [Dkt. No. 3349]. Therefore, the Court has compared the experts’ calculations for the
relative share of liability held by Aldrich and Murray—particularly Dr. Sackett’s LAS curve-derived
proportion, which was the same curve used by McKnight—and estimates the ratio to be 63% for
Aldrich ($463,478,400) to 37% for Murray ($278,950,400). Applying the 1.21 multiplier for Aldrich
and 1.24 for Murray to those respective amounts yields a total liability figure of $742,428,800, which
the Court will round up to $742,500,000.
Case 20-30608 Doc 3390 Filed 09/25/26 Entered 09/25/26 15:43:00 Desc Main
Document Page 64 of 64