garlock's reply in support of subpeona to manville trust - 533816_4646

8/21/2019 Garlock's Reply in Support of Subpeona to Manville Trust - 533816_4646

1/54

1

IN THE UNITED STATES BANKRUPTCY COURT

FOR THE WESTERN DISTRICT OF NORTH CAROLINA

Charlotte Division

IN RE:

GARLOCK SEALING TECHNOLOGIESLLC, et al.,

Debtors.1

Case No. 10-BK-31607

Chapter 11

Jointly Administered

DEBTORS’ REPLY IN SUPPORT OF MOTION FOR LEAVE

TO SERVE SUBPOENA ON MANVILLE TRUST

Debtors hereby submit this Reply in support of their Motion for Leave to Serve Subpoena

on Manville Trust (Docket No. 4599) (the “Motion”) and in opposition to the Objection of Non-

Party Manville Personal Injury Settlement Trust to the Debtors’ Motion for Leave to Serve

Subpoena (Docket No. 4638) (“Manville Objection”) and the Response and Limited Objection of

the Official Committee of Asbestos Personal Injury Claimants to Debtors’ Motion for Leave to

Serve Subpoena on Manville Trust (“Committee Objection”).

The Committee and Manville Trust do not deny Debtors’ right to issue a subpoena to the

Manville Trust seeking data regarding non-mesothelioma claimants, but challenge the scope of

the proposed subpoena (the “Subpoena”) and the confidentiality restrictions that will apply to the

information produced.2

I. The Data Requested Are Indisputably Relevant to Estimation of Non-Mesothelioma

Claims and Plan Feasibility

1 The debtors in these jointly administered cases are Garlock Sealing Technologies LLC; GarrisonLitigation Management Group, Ltd.; and The Anchor Packing Company (hereinafter “Garlock” or“Debtors”).2 The Manville Personal Injury Settlement Trust (“Manville Trust”) has agreed to submit all its objectionsto Debtors’ Subpoena to this Court for final resolution. Counsel for the Committee in a conference beforeDebtors filed their Motion stated that the Committee did not anticipate objecting to the Motion, butsubsequently filed an objection anyway.

Case 10-31607 Doc 4646 Filed 06/11/15 Entered 06/11/15 12:02:03 Desc Main Document Page 1 of 23


2/54

2

The Manville Trust first argues that the requested data are not relevant to estimation of

non-mesothelioma claims or Plan feasibility, questioning why Debtors require data pertaining to

individual claimants instead of aggregate or anonymized data, and also questioning Debtors’

need for certain kinds of data. Manville Objection ¶¶ 40-46.

A. The Court Has Long Recognized That Discovery Pertaining to Individual Asbestos

Claimants Is Legitimate and Necessary

As Debtors explained in their Motion, the Subpoena seeks information relevant to non-

mesothelioma claim estimation and Plan feasibility. For mesothelioma claims, this information

was largely collected through the personal injury questionnaire (“PIQ”). The Manville Trust

admits that it routinely provides the requested information to asbestos defendants who subpoena

it. See Manville Objection ¶ 17.

The Manville Trust’s argument that data regarding individual claimants is irrelevant to

aggregate analysis is incorrect. As he explains in the declaration attached as Exhibit A (“Bates

Declaration”), Dr. Bates’s method “relies on individual-level data from claimants that is

analyzed by using established econometric statistical methods to reach conclusions about

aggregate estimates.” Id. ¶ 5. The Court’s Estimation Opinion demonstrates Dr. Bates’s reliance

on individual claimant data. See In re Garlock Sealing Technologies LLC , 504 B.R. 71, 95

(Bankr. W.D.N.C. 2014) (describing how Dr. Bates’s estimate was based on information about

pending claimants collected through questionnaires, and criticizing Committee expert for not

relying on individual claimant data). The Manville Trust’s statement that “Debtors have no

intention of individually assessing the merits of each and every claim asserted against them,”

Manville Objection ¶ 48, is false—Dr. Bates did exactly that in rendering his “reasonable and

reliable” aggregate estimate of mesothelioma claims. See id. at 96 (noting that Dr. Bates assessed

numbers of claimants who identified exposure to a Garlock asbestos-containing product);



3/54

3

Estimation Trial Tr. 8/2/13 at 2815:6-14 (testimony of Dr. Bates) (discussing pending

mesothelioma claims, “We estimated the aggregate compensatory amount we expect each one of

them could get. We estimated the recoveries for each one of them. . . . We applied to each one of

them the likelihood—liability likelihood that we got, and essentially that gave us a valuation of

the pending claims.”).

As Dr. Bates explains in his declaration, the Manville Trust’s request for an anonymous

production would render it useless, because Dr. Bates needs identifying information in order to

match the produced data to claimants in the Garlock Analytical Database that he uses for his

expert work. Id. ¶ 6(1); see also Declaration of Garland S. Cassada (attached as Ex. B) ¶ 8

(“Cassada Declaration”) (stating that this concern was conveyed to Trust counsel). This Court

previously rejected a request by ten Trusts to anonymize data they produced to Debtors in

connection with the mesothelioma estimation trial, and no different rule should apply to this

request. See Order Granting in Part and Overruling in Part Objections to Subpoena By Delaware

Claims Processing Facility, LLC and Associated Trusts, etc. (Docket No. 2430) ¶ 7 (“The

request by DCPF, the Trusts, and the Committee for the Trust Data to be anonymized prior to

production to Debtors is denied.”).

B. The Manville Trust Routinely Produces the Requested Data to Asbestos Defendants,

Which Is All Relevant to Estimation of Non-Mesothelioma Claims and Plan Feasibility

The Manville Trust claims that Debtors did not explain in their Motion how the requested

information is relevant to estimation of non-mesothelioma claims and Plan feasibility. Manville

Objection ¶ 40. To the contrary, Debtors explained that most of the information was requested in

the Court-approved PIQ, upon which Dr. Bates relied in rendering the mesothelioma estimate

that the Court adopted. See Bates Decl. ¶ 6. The data are equally relevant to non-mesothelioma

claims estimation. Indeed, the fact that the Manville Trust “generally releases” all the data



4/54

4

Debtors have requested to asbestos defendants who subpoena it, Manville Objection ¶ 17, tacitly

admits that the evidence is all relevant to the merit of claims and not objectionable. 3

To remove any doubt about the relevance of the requested information, Dr. Bates in his

declaration explains in detail how all the data requested in the Subpoena are relevant to

estimation of non-mesothelioma claims and Plan feasibility. The data fall into five categories:

• Claimant identifying information. Dr. Bates will use fields identifying the claimant to

match the Manville Trust data to claimants in the Garlock Analytical Database, an

essential step in his analysis. Bates Decl. ¶ 7(1).

•

Disease and other information pertaining to alleged injury. The Garlock Analytical

Database currently includes only the most basic information on alleged injury (i.e., “lung

cancer,” “other cancer,” “non-malignant,” “unknown”). Id. ¶ 7(2). That information is

insufficient to determine whether non-mesothelioma claimants qualify for payment under

the Plan’s Claims Resolution Procedures (“CRP”), which include criteria regarding

diagnosis, smoking history, and diagnostic tests and exams. Id.; see also CRP at I-4 to I-

7, I-10 to I-14, II-3 to II-6 (stating medical criteria for non-malignant conditions). The

Manville Trust request will provide valuable information on these topics, Bates Decl. ¶

7(2), and is also relevant to estimation of the allowed amount of non-mesothelioma

claims, id.

• Exposure information. The Garlock Analytical Database currently contains very little

data on non-mesothelioma claimants’ occupations, industries, and exposure histories,

which are relevant both to non-mesothelioma estimation and to the payments non-

mesothelioma claimants can receive under the CRP. Id. ¶ 7(3); Estimation Opinion at 96.

3 The Trust mentions other information that it does not routinely provide to co-defendants (such ascorrespondence between the claimant and the Trust, id. ¶ 18), but Debtors have not requested that and itshould not weigh in the Court’s decision on this Motion.



5/54

5

The Manville Trust data and documents will provide valuable information regarding

these facts. Bates Decl. ¶ 7(3).

• Tort claimant information. Bates White will use information regarding claimants’ tort

suits both to match claimants to the Garlock Analytical Database and to eliminate so-

called “false positives” returned by the Manville Trust request (see below for complete

discussion of this issue). Id. ¶ 7(4).

• Manville Trust claim information. Dr. Bates will use data regarding the Manville Trust

claim in analyzing the timing of Trust claims by Garlock non-mesothelioma claimants,

and in estimating the offsets that would apply to a potential plaintiff verdict. Id. ¶ 7(5).

He testified about both of these topics at the mesothelioma estimation trial.

The Manville Trust particularly objects to only a few of the items Debtors have

requested. First, it objects to producing information about the settlements claimants who sued

Garlock received from the Manville Trust. Manville Objection ¶¶ 32, 41. But this Court has

already recognized that claimants’ other settlements are relevant to Garlock’s liability, because

they are potential offsets to any verdict. See Estimation Opinion at 96; see also Bates Declaration

¶ 7(5). The Manville Trust relies on a different part of the Court’s ruling, where it held that

Garlock’s past settlements do not reflect its liability. But this does not show the irrelevance of

third parties’ settlements. Third party settlements are not relevant because they show those

parties’ liability, but because they are potential offsets to any obligation of Garlock. That is why

the Court squarely held that those settlements are relevant to estimation of the allowed amount

of asbestos claims against Garlock.

The Manville Trust also cites this Court’s discovery ruling very early in the case that

initially denied Garlock settlement-related discovery. See Manville Objection ¶ 43. It omits to



6/54

6

explain, however, that subsequent to the ruling, the Court reversed course. It permitted Debtors

to obtain discovery regarding mesothelioma claimants’ settlements, and later relied on data from

the discovery it ultimately permitted in rendering its estimation opinion. See Estimation Opinion

at 96.4

The Manville Trust next objects to the relevance of information concerning claimants’

non-asbestos-related conditions, to the extent they are contained in medical records submitted to

the Trust, and seeks to redact the information from records produced to Debtors, at Debtors’

expense. Manville Objection ¶¶ 44-45, 57(d). The Trust does not insist on this step before

producing full medical records without redaction to co-defendants in asbestos litigation, see id. ¶

17, and does not explain why it seeks to treat Debtors differently. Indeed, because these are

medical records that claimants submitted to substantiate their claims with the Manville Trust,

they are unlikely to contain information irrelevant to the conditions claimants put at issue by

making a claim against Manville and suing Garlock. Moreover, non-asbestos-conditions are

relevant because they can bear on the reliability of the diagnosis of asbestos-related disease. See,

e.g., In re Silica Products Liability Litig., 398 F. Supp. 2d 563, 628 (S.D. Tex. 2005) (finding

that reliability of silicosis diagnoses was undermined by fact that claimants had been diagnosed

with both silicosis and asbestosis, when “a golfer is more likely to hit a hole-in-one than an

occupational medicine specialist is to find a single case of both silicosis and asbestosis”). The

4 The Manville Trust’s reliance on decisions in other cases denying discovery of settlement amounts is

misplaced. One of the cited decisions expressly recognized the relevance of settlements to offsets, butdeferred production until after the liability phase—a procedure that has no application here, whereDebtors’ expert assesses aggregate liability and damages all at once. See Anderson v. Ford Motor Co.,2010 WL 4156256, at *1 (E.D. Pa. 2010). In another case, settlement amounts were not requested. SeeShepherd v. Pneumo-Abex, LLC , 2010 WL 3431633, at *1-2 (E.D. Pa. 2010). In a third, the defendantimpermissibly sought settlement amounts to prove liability of those parties in violation of Rule 408—notto prove offsets. See Dent v. Westinghouse, 2010 WL 56054, at *1-3 (E.D. Pa. 2010). The other casescited are clearly in conflict with discovery rulings made by this Court, recognizing the discoverability ofTrust claim and settlement information.



7/54

7

Manville Trust cites no authority showing this information is irrelevant, and appears to recognize

the relevance of the information because it produces the information to tort defendants. In

addition, it would be costly and wasteful to perform those redactions here because Debtors have

committed to keeping all medical information confidential pursuant to protective order. See

Motion ¶ 18.

Finally, the Committee and Manville Trust also complain that a “first name, last name”

matching procedure will result in production of irrelevant data for individuals who did not assert

claims against Garlock. Manville Objection ¶ 46. The possibility of so-called “false positives”

arises because Debtors currently have limited information in the Garlock Analytical Database

with which to perform a match—for most pending non-mesothelioma claimants, first name and

last name. But Debtors offered the Manville Trust several ways to address this issue. First, once

Bates White receives the data, it will apply advanced algorithms using the information available

to Garlock in the database to identify and remove potential false positives. Bates Declaration ¶¶

16-18. However, to ensure that Debtors obtain information for the claimants who have sued

Garlock, the discovery must initially include the broader data set, to avoid a “false negative”

problem where relevant data pertaining to Garlock claimants is mistakenly omitted from the

production because of an overly stringent matching procedure. Id.

Second, Debtors’ Motion proposed an opportunity for affected claimants (who will

receive notice through the Trust’s procedures) to object before any information is produced. This

affords persons who have never asserted a claim against Garlock the opportunity to object to the

production and use of their data in this proceeding. See Motion ¶ 17. Finally, Debtors will keep

information for non-unique claimants confidential until the bar date, after which newly gathered

information (such as SSNs) will allow further identification of “false positives” and deletion of



8/54

8

those claimants’ records. These measures represent a reasonable approach to an unavoidable

problem.5

II. Limiting Discovery Of Relevant Evidence By Requiring a Sample Would Be

Inappropriate Where The Manville Trust Has Not Even Alleged—Much Less Proved— Undue Burden

Next, the Manville Trust and Committee seek to limit Debtors’ subpoena to 10% of the

approximately 91,000 pending non-mesothelioma claimants in their database. As the basis for

this restriction, they cite vague concerns about overbreadth, largely without citing any legal

authority. See, e.g., Manville Objection ¶¶ 4, 47-51.

A. The Manville Trust Does Not Allege Undue Burden, and Therefore Cannot Justify the

Limitation It Proposes

Neither Objection, however, contains any allegation whatsoever that Debtors’ request for

data pertaining to all pending non-mesothelioma claimants will cause undue burden to the

Manville Trust (or any other allegation supporting a valid discovery objection). Discovery into

relevant evidence is generally proper, Fed. R. Civ. P. 26(b)(1), and may be limited only if the

producing party can demonstrate an undue burden or other concrete and particularized harm. See

Fed. R. Civ. P. 26(b)(2)(C) (proper to limit discovery of relevant evidence when “the burden or

expense of the proposed discovery outweighs its likely benefit”); Bank of Mongolia v. M & P

Global Fin. Servs., Inc., 258 F.R.D. 514, 518 (S.D. Fla. 2009) (“The courts have long recognized

the wide scope of discovery allowed under the Federal Rules of Civil Procedure. . . . The United

States Supreme Court has said that they are to be broadly and liberally construed.”) (punctuation

and citation omitted).

5 The only alternative to this procedure would be to seek these data from claimants themselves, but no-one denies that would be far more expensive than the Subpoena.



9/54

9

Accordingly, “[t]he party claiming that a discovery request is unduly burdensome must

allege specific facts that indicate the nature and extent of the burden, usually by affidavits or

other reliable evidence.” Tucker v. Ohtsu Tire & Rubber Co., 191 F.R.D. 495, 498 (D. Md. 2000)

(rejecting claim of undue burden where “defendant’s assertions of harassment, burden, prejudice,

and expense are generalized, non-specific objections”); see also Convertino v. U.S. Dep’t of

Justice, 565 F. Supp. 2d 10, 14 (D.D.C. 2008) (“This Court only entertains an unduly

burdensome objection when the responding party demonstrates how discovery of the document

is overly broad, burdensome, or oppressive, by submitting affidavits or offering evidence which

reveals the nature of the burden”) (punctuation and citation omitted); Bank of Mongolia, 258

F.R.D. at 519 (objecting party “must explain the specific and particular way in which a request is

vague, overly broad, or unduly burdensome. In addition, claims of undue burden should be

supported by a statement (generally an affidavit) with specific information demonstrating how

the request is overly burdensome”). A party opposing discovery into relevant information has

“an obligation to provide sufficient detail in terms of time, money and procedure required to

produce the requested documents.” Cory v. Aztec Steel Bldg., Inc., 225 F.R.D. 667, 672 (D. Kan.

2005).6

The Court will search the Manville Objection in vain for an allegation that producing

information for 91,000 claimants is significantly more burdensome than producing information

for 9,100 claimants. The Trust’s overbreadth argument contains no such allegation, see Manville

Objection ¶¶ 47-51, much less an explanation by affidavit of how the “time, money, and

procedure required to produce the requested documents” will be overly burdensome. Cory, 225

F.R.D. at 672.

6 In addition, “the scope of discovery from a nonparty by means of a subpoena duces tecum under Rule 45is coextensive with that of a motion for production from a party under Rule 34.” Castle v. Jallah, 142F.R.D. 618, 620 (E.D. Va. 1992).



10/54

10

The Manville Trust could not make this claim because it is not true. Debtors crafted their

request specifically to avoid incremental burden, by limiting the request to information that is

stored electronically and linked to claimant names, and which the Manville Trust provides as a

matter of policy when it is subpoenaed by asbestos defendants. See Motion Ex. B. In conferences

with the Manville Trust’s counsel, Debtors expressed their understanding that the Trust’s

database is highly sophisticated, such that providing data for 91,000 claimants is not materially

different in terms of burden or costs than providing data and records for 9,100 claims. Cassada

Decl. ¶ 16. The Trust’s counsel stated that he “could not deny” that Debtors’ understanding was

correct. Id.

B. The “Overbreadth” Objection Is an Attempt to Dictate Debtors’ Discovery and

Undermine Their Case

Absent the ability even to allege undue burden, the objections by the Manville Trust and

Committee are nothing more than an attempt to dictate the terms of Debtors’ discovery and their

experts’ analyses, and to impose their views about disputed matters before the Court. The

Manville Trust and Committee admit as much throughout their objections. See, e.g., Manville

Objection ¶ 20 (“[H]ere, only a random, anonymized sampling of such data is likely to be

necessary to the adjudication”); ¶ 19 (claiming without citation that “mass adjudications entail

fact-finding as to representative claimants, and the extensive use of sampling and other statistical

techniques”); Committee Objection at 4 (“If, for the limited purposes of plan confirmation

proceedings in this mass-tort bankruptcy, the Debtors’ expert wishes to posit average values for

groups or types of non-mesothelioma claims using data from sources like the Manville Trust,

there is no reason why he could not do so on the basis of a statistically valid sample.”). The

Committee seems to believe that Debtors are constrained to adopt the Committee expert’s

estimation methodology, which does not rely on collecting claimant data to inform statistically



11/54

11

valid judgments. See Committee Objection at 4-5. The Court rejected that methodology,

concluding that it was unreliable because it failed to consider current data. See Estimation

Opinion at 95. In any event, it is for Debtors and their experts to decide what they need and what

constitutes an appropriate sample, not the Manville Trust or the Committee. Absent undue

burden, the Manville Trust has no reason to deny Debtors discovery of relevant evidence.7

In fact, as explained in the Bates Declaration, a 10% sample would be inappropriate here,

for several reasons. First, Bates White does not currently know the degree of overlap between

Manville Trust and Garlock claimants, and to ensure statistical validity would have to draw a

large sample that would not differ materially from Debtors’ request. Id. ¶ 11. Second, there is no

evidence that the 10% sample proposed by the Manville Trust and Committee out of thin air

would be appropriate, because “the size of a representative sample does not depend on the size of

the population but rather on the variation of the feature in the population that is going to be

analyzed and about which one will draw statistical inferences,” and Bates White has not yet had

the opportunity to perform that analysis. Id. ¶ 13. Finally, sampling makes analysis needlessly

complicated when all the information requested can be produced without any undue burden, and

a census produces analysis that is more precise than a sample. Id. ¶¶ 10, 12. The Manville Trust

provides no counter-demonstration that a 10% sample would be reasonable in this case.

The Manville Trust relies heavily on certain discovery Debtors took prior to the

mesothelioma estimation trial where they relied on a sampling approach. Manville Objection ¶¶

50-51. But it fails to note that prior to those samples of pending mesothelioma claimants,

Debtors had the opportunity to seek basic information from all mesothelioma claimants listed as

7 The Manville Trust also relies on what it has allegedly done in response to previous subpoenas seekinginformation about numerous claimants, Manville Objection ¶ 21, but those assertions do not justifylimitations on discovery where the Manville Trust has failed to allege any concrete and particularizedharm to its interests.



12/54

12

pending in the database through the original questionnaire. Later sampling approaches were

based on a variety of factors considered by Debtors and their experts, and benefited significantly

from the previous information collected from the broader group of all pending mesothelioma

claimants. See Bates Decl. ¶ 13. Those samples were crafted by Debtors’ experts—not imposed

at the behest of a discovery target attempting to dictate the terms of Debtors’ discovery in

conjunction with Debtors’ litigation adversary.

The attempt by the Manville Trust and Committee to limit Debtors’ inquiry into relevant

information, without making any allegation of undue burden, is especially problematic because

of the close association between the Manville Trust and the law firms that make up the

Committee and represent the Committee. The Committee’s claims expert, Dr. Mark Peterson, is

one of three trustees of the Manville Trust and ultimately responsible for its litigation decisions,

including presumably its decision to object to Debtors’ Subpoena. Caplin & Drysdale represents

the Selected Counsel of the Beneficiaries, a group of plaintiff law firms that is part of Trust

governance and must consent to certain Manville Trust actions. One of the three Selected

Counsel is Joseph Rice of Motley Rice, who is co-chair of the Committee in this case. The

Manville Trust and Committee disclosed none of these ties in their objections to Debtors’

Motion. These undisclosed relationships may help explain why the Manville Trust and

Committee would object to a Motion seeking information the Manville Trust normally provides

to other co-defendants as a matter of policy, and which will admittedly not cause an undue

burden on the Trust.

C. Speculating About Professional Fees Does Not Constitute a Legitimate Discovery

Objection

The Committee makes the additional argument that the Court should impose a sample to

prevent undue expense for the estates. Committee Objection at 6-7. It cites no authority



13/54

13

demonstrating that this is a legitimate objection to discovery under the Rules, and provides

nothing more than speculation about the harm it insinuates. Unreasonable professional fees are

controlled not by denying discovery that is relevant and will not cause any undue burden, but

instead by the fee objection process, in which the fee examiner now participates and over which

the Court retains complete control. No party has ever objected to the fees of Debtors or their

expert. The Court’s adoption of Dr. Bates’s mesothelioma estimate, and rejection of the

Committee’s estimate in part because its expert did not use the data Debtors collected, see

Estimation Opinion at 94-95, demonstrate that there are no grounds to question Bates White’s

professional judgment in using data.

In any event, the comparisons the Committee attempts to draw between data collection

for the mesothelioma estimation trial and Debtors’ Subpoena are crude and inapposite. As Dr.

Bates explains in his declaration, that cost came about primarily because PIQ claimants

submitted hard copy documents (many handwritten) rather than electronic data, and those

documents then had to be “scanned, classified, reviewed and analyzed for data extraction.” Bates

Decl. ¶ 19. Here, because the Manville Trust keeps their data in an organized manner, linked to

claimants, “the cost of analyzing such information would be significantly less than the one

incurred in using the PIQ information, not multiples of it.” Id.

III. There Is No Basis For Requiring That Non-Sensitive Information Such As Claimant

Names and Exposure Histories Be Kept Confidential

The Committee and Manville Trust also argue that all of the information produced by the

Manville Trust should be kept strictly confidential and used only for purposes of the

confirmation hearing. Debtors in their Motion and in discussions with the Trust offered to protect

truly sensitive information such as medical records, full SSNs, dates of birth, names of minors,

and financial account numbers—i.e., the claimant information this Court ruled should be



14/54

14

protected in the public mesothelioma estimation record. See Motion ¶ 18; Cassada Decl. ¶ 9. But

the Manville Trust and Committee insist on complete confidentiality, including for claimants’

identities, their occupations and industries, their histories of exposure to asbestos, the fact they

filed a claim against the Manville Trust, and many other non-sensitive facts.

A. The Manville Trust and Committee Have Not Alleged, Much Less Demonstrated, Good

Cause for Keeping Non-Sensitive Data Confidential

The blanket restrictions sought by the Committee and Manville Trust would violate the

law. The default rule in federal court is that evidence obtained in discovery may be disseminated

freely, with no restrictions on how the requesting party uses it. See San Jose Mercury News, Inc.

v. U.S. Dist. Court , 187 F.3d 1096, 1103 (9th Cir. 1999) (“It is well-established that the fruits of

pretrial discovery are, in the absence of a court order to the contrary, presumptively public.”);

Jepson, Inc. v. Makita Elec. Works, Ltd., 30 F.3d 854, 858 (7th Cir. 1994) (“Absent a protective

order, parties to a law suit may disseminate materials obtained during discovery as they see fit.”).

To impose confidentiality or use restrictions, the producing party must show “good

cause” under Federal Rule of Civil Procedure 26(c).8

“A party asserting good cause bears the

burden, for each particular document it seeks to protect, of showing that specific prejudice or

harm will result if no protective order is granted.” Foltz v. State Farm Mut. Auto. Ins. Co., 331

F.3d 1122, 1130 (9th Cir. 2003); Brittain v. Stroh Brewery, Co., 136 F.R.D. 408, 412 (M.D.N.C.

1991) (“The party must make a particular request and a specific demonstration of facts in support

of the request as opposed to conclusory or speculative statements about the need for a protective

order and the harm which would be suffered without one.”); U.S. ex rel. Davis v. Prince, 753 F.

Supp. 2d 561, 565 (E.D. Va. 2010) (“The party seeking a protective order has the burden of

8 The same “good cause” standard applies to third parties subject to subpoena. See In re Roman Catholic Archbishop of Portland in Or., 661 F.3d 417, 426 (9th Cir. 2011).



15/54

15

establishing ‘good cause’ by demonstrating that specific prejudice or harm will result if no

protective order is granted”) (punctuation and citation omitted); 8A Wright & Miller, Federal

Practice and Procedure § 2043 (3d ed.) (party requesting confidentiality “must also show good

cause for restricting dissemination on the ground that it would be harmed by its disclosure. As

with any protective-order motion, the showing should be made with appropriate specifics.”).

The “good cause” requirement is rigorous. “Broad allegations of harm, unsubstantiated

by specific examples or articulated reasoning, do not satisfy the Rule 26(c) test. . . . Moreover,

the harm must be significant, not a mere trifle.” Cipollone v. Liggett Grp., Inc., 785 F.2d 1108,

1121 (3d Cir. 1986); see also Pansy v. Borough of Stroudsburg , 23 F.3d 772, 786 (3d Cir. 1994).

The burden of proving specific harm rests with the producing party, and typically requires an

affidavit. See Brittain, 136 F.R.D. at 412-13 (“Such demonstrations are preferably made by

affidavits from knowledgeable persons . . . .”). Under these standards, courts routinely order

parties to keep confidential truly sensitive information such as medical records and trade secrets.

See, e.g., Foltz , 331 F.3d at 1131, 1137. But courts refuse to apply blanket protections that

stretch further than absolutely necessary to protect sensitive information. See Brittain, 136

F.R.D. at 412 (good cause requirement “furthers the goal that the Court only grant as narrow a

protective order as is necessary under the facts”); Foltz , 331 F.3d at 1131 (refusing to apply

blanket protection simply because some medical records may be present).

Far from making a specific and particularized showing of harm using affidavit evidence,

the Committee and Manville Trust do not even offer arguments for why information other than

SSNs, medical records, and financial account numbers should be kept confidential. The

examples the Manville Trust gives of sensitive information are all categories that Debtors have

already agreed to protect: “medical records . . . the claimant’s history of drug and alcohol abuse,



16/54

16

HIV status, sexual or emotional dysfunction, or other highly private and confidential health

information,” “descriptions of claimants’ children and other dependents’ mental and/or physical

disabilities or drug addictions,” and “Social Security numbers and dates of birth.” Manville

Objection ¶¶ 11, 12.9

The confidentiality sought by the Manville Trust and Committee would extend far

beyond these truly sensitive facts, to cover information where the “good cause” standard clearly

cannot be met. For example, it would include many non-sensitive facts about individuals who

sued Garlock—their occupational histories, their exposures to asbestos, their claims against the

Manville Trust, and even their very identities, despite that fact that those claimants have already

sued Garlock in public courts and have been publicly identified in this case in the Garlock

Analytical Database and Garrison database that are part of the estimation record, as well as

Debtors’ schedules. The Committee and Manville Trust’s approach would even protect

information about the claimant’s tort suit, a fact that is already a matter of public record in

another court.

The Manville Trust and Committee offer no argument for why any of this information

meets the “good cause” standard, instead offering only vague and non-specific concerns about

confidentiality. See Manville Objection ¶ 19 (alleging that subpoenas seeking data for multiple

claimants “implicate confidentiality concerns that subpoenas in individual actions do not”); id. ¶

20 (vague statement about the “highly sensitive nature of the claimant data”); id. ¶ 55 (alleging

without citation a claimant expectation “that personal information they submit to the Manville

Trust will be treated in a confidential manner”). These vague allegations amount to nothing more

9 The Manville Objection also mentions financial information, Manville Objection ¶ 5, but Debtors arenot aware that their request seeks such information. Depending on the nature of the financial information,it could be legitimately protected. This Court has already recognized that full financial account numbersshould be protected. See infra.



17/54

17

than “conclusory or speculative statements about the need for a protective order,” Brittain, 136

F.R.D. at 412, and do not come close to carrying the Manville Trust’s burden of demonstrating

“that specific prejudice or harm will result if no protective order is granted.” Foltz , 331 F.3d at

1130. It would also be inappropriate to apply a protective order to all of the data, merely because

some of it (SSNs, medical records) is sensitive—especially because Debtors agreed in their

Motion that the truly sensitive data should be subject to protective order. See id. at 1131.

Tellingly, when Debtors’ counsel asked counsel for the Manville Trust why non-sensitive

claimant information should be kept confidential, he received no response—only vague

allegations of sensitivity. See Cassada Decl. ¶ 13 (“Other than medical records (which the

Debtors have agreed to protect from public disclosure), Mr. Rubinstein did not during the course

of our discussions identify a single additional fact or piece of information that would be

‘private,’ but simply said his position was Manville Trust policy.”). Debtors’ counsel also

pointed out that the Manville Trust provides precisely this information to co-defendants who

subpoena it, without taking the position that it is confidential and without demanding a protective

order. Id. ¶ 12. The Manville Trust’s counsel responded that the Manville Trust reserved its right

to treat Garlock differently. Id.10

The Committee and Manville Trust also argue that the Manville Trust has an absolute

right to designate discovery as “confidential” under the Stipulated Protective Order (“SPO”),

without having to demonstrate “good cause.” Manville Objection ¶ 54; Committee Objection at

7-8. But a party has no such unilateral right. Stipulated protective orders are entered as a matter

10 The Manville Trust also relies on a licensing agreement that it requires when it provides extracts of itsdata to third parties outside of litigation. See Manville Objection ¶¶ 22-25, 55. The Manville Trust is freeto insist on any conditions it likes outside of litigation, but within litigation, it must meet the “good cause”standard, which it has failed to do with respect to non-sensitive data such as claimant identities, exposure,and occupational histories. Apparently recognizing this, the Manville Trust produces data to co-defendants pursuant to subpoena without insisting on any confidentiality restrictions at all. Cassada Decl. ¶ 12.



18/54

18

of course, to streamline discovery and avoid repetitive “good cause” motions to the court by

according temporary protection to documents designated as confidential. See, e.g., In re Roman

Catholic Archbishop, 661 F.3d at 424 (stipulated protective orders often entered without “good

cause” showing to streamline discovery). But a party must exercise good faith in marking

documents as confidential, and even then, their designation can be challenged at any time, which

forces the producing party to demonstrate that the “good cause” standard is met. See id. at 424

(under stipulated protective order, if designation is challenged, “the party opposing disclosure

has the burden of establishing that there is good cause to continue the protection of the discovery

material”) (quotation and punctuation omitted); Foltz , 331 F.3d at 1131 (producing party had

burden of showing “good cause” with respect to each document subject to blanket protective

order when challenged). Here, to the extent the Manville Trust is attempting to avail itself of the

SPO, Debtors are challenging the designation of these documents as confidential, because neither

the Trust nor the Committee has any basis to designate non-sensitive claimant information as

such.

Finally, given that the Manville Trust cannot meet the “good cause” standard to require a

protective order limiting disclosure of non-sensitive facts about claimants, its request for

anonymized data—an even more severe restriction—clearly cannot be sustained. As described

above, Debtors explained to the Trust’s counsel that anonymizing the data would render it

useless, and this measure is also precluded by the law of this case, where a request to anonymize

Trust data was rejected. See supra.

B. The Manville Trust and Committee’s Insistence that Debtors Not Use Claimant

Information in Court Violates The Law of This Case

But the Manville Trust and Committee would go even further than a blanket protective

order applicable to all claimant information, sensitive and non-sensitive alike. They would insist



19/54

19

that Debtors redact claimants’ names before using the produced evidence in Court and not

“disclos[e] such information in any court filing,” which apparently would prevent even this

Court from seeing the evidence. Manville Objection ¶¶ 35, 57(c), Committee Objection at 2

(“Debtors have wrongly insisted that they would be entitled to make public use of any such

information . . .”); id. at 11 (claiming that Debtors should be required to anonymize data

introduced into evidence).

The request violates the law of this case. As the Court is aware, prior to the estimation

trial, the Committee and certain asbestos plaintiffs’ firms petitioned the Court (over Debtors’

objection) to restrict the use and prohibit the disclosure of all information pertaining to

claimants, including their identities, their exposure histories, and their claims histories. These

orders prohibited Debtors from sharing evidence they discovered with third parties or disclosing

the identities of asbestos claimants (or facts about their claims) in open court (or outside of

court). As a result, live testimony at the mesothelioma estimation trial about specific claimants

was offered in closed court sessions, and transcripts and documents offered into evidence or filed

with the Court were placed under seal. Cassada Decl. ¶ 10.

Members of the public objected to the Court’s orders and appealed. The District Court

reversed and remanded. Legal Newsline v. Garlock Sealing Technologies LLC , 518 B.R. 358,

367 (W.D.N.C. 2014). On remand, this Court issued a ruling that defines precisely what kinds of

information in this case can be protected from disclosure in open court and what kinds should

not.

Specifically, on October 31, 2014, this Court ruled that there was no privacy right or

other basis for protecting claimant information from the public, with the exception of (a) SSNs

(except last four digits), (b) date of birth (except year), (c) names of identifiable minors (except



20/54

20

for their initials), (d) financial account numbers (except last four digits), and (e) medical

information (except claimed disease). The Court found that “the appropriate privacy interests of

the individuals involved here are protected” by those redactions, and further found that “[a]s to

the names of adult claimants, I can’t find any extraordinary circumstances that would require

anonymity,” as “[a]ll of the adult claimants are named in state court complaints, I think also in

the schedules in this proceeding.” 10/16/14 Tr. 52-54. As a result, the Court unsealed all of the

claimant information that the Manville Trust and Committee now seek to protect, including

evidence pertaining to claimants’ occupational and exposure histories, the claims they had

asserted against Trusts and other entities, and the settlements claimants had received from Trusts

and other entities. See Order on Motions to Seal Materials in Record of Estimation Proceeding

and Protocol for Redaction of Record (Docket No. 4195) (the “Public Access Order”).11

The restriction on Debtors sought by the Manville Trust and Committee would directly

violate this Court’s ruling that claimant information used in Court should not be redacted, except

for the limited facts that truly implicate privacy concerns. Indeed, it would go even further than

the previous overruled orders did, by apparently precluding Debtors from showing this Court the

evidence. See Manville Objection ¶ 57(c) (seeking to preclude Debtors from using claimant

names in “any court filing”).

It is especially troubling that the Committee continues to press this issue, given that

litigation to establish these principles cost the estates over $1.4 million, with Caplin & Drysdale

billing fees of over $1.2 million. See Declaration of Satyra Riggins (attached as Ex. C). The

Committee has no good faith basis to continue raising obstacles to Debtors’ right to introduce

11 The Manville Trust points to a press account that used settlement information in the estimation recordas a reason why the Court should enter a protective order here, Manville Objection ¶ 53 n.6, but fails torecognize that the story was a direct consequence of this Court’s ruling refusing to keep settlementamounts confidential, and was entirely proper reliance on public judicial records.



21/54

21

non-sensitive information in open court. Indeed, when the Committee last fall was forced to

explain the basis for the blanket confidentiality protection it had long advocated for years before

this Court, it was unable even to file a motion seeking protection for more than what the Court

ultimately ordered and Debtors have offered the Manville Trust here.12

In addition, the Court’s Public Access Order further demonstrates that the Manville Trust

and Committee cannot meet the “good cause” standard with respect to non-sensitive claimant

information. The Court was applying different legal standards in the access litigation—the First

Amendment and common law rights of access to judicial records, rather than the “good cause”

standard applicable to protective orders. But at the end of the day, the Court found that claimants

who sued Debtors had no privacy interests in their identities or other facts about their claims,

apart from clearly protectable information such as SSNs, financial account numbers, and medical

records. For the same reason, the Committee and Manville Trust cannot carry their burden to

show “good cause” to justify a protective order in discovery.

Notably, the Court’s Public Access Order also limits the relevance of the protective

orders upon which the Manville Trust and Committee heavily rely, which were entered by this

Court during discovery before the Court set the boundaries of protectable information in its

Public Access Order. See Manville Objection ¶ 54 & n.8; Committee Objection at 2, 8-9. The

provisions in those orders requiring all claimant information to be introduced under seal—

including non-sensitive information such as claimant identities and exposure—clearly did not

survive the later rulings by the District Court and this Court in the public access litigation. Nor

12 The Committee argues that the Subpoena would lead to undue cost because Debtors would becompelled to redact voluminous medical records at great expense in order to introduce them intoevidence. Committee Objection at 6-7. This is not true because any medical records Debtors offer at theconfirmation hearing could be introduced under seal instead. In fact, medical records—which are easilyidentifiable—did not form a significant part of the cost of redacting the mesothelioma estimation record,which instead came primarily from reviewing non-medical documents for social security numbers andmedical information.



22/54

22

do those prior orders assist the Manville Trust in making a “good cause” showing for non-

sensitive claimant information. Those protective orders were blanket orders, and no party

demonstrated good cause for protecting particular kinds of information. The Court’s Public

Access Order demonstrates that it is not possible for any party to meet the good cause standard

with respect to non-sensitive claimant information.13

13 Debtors are also concerned that a blanket protective order could invite costly collateral litigation bythird parties seeking to eliminate the restrictions. Third parties have a right to intervene and challenge protective orders that interfere with parties’ right to disseminate information obtained in discovery. See Pansy, 23 F.3d at 777 (“We have routinely found, as have other courts, that third parties have standing tochallenge protective orders and confidentiality orders in an effort to obtain access to information or judicial proceedings.”). Debtors seek to avoid any further collateral litigation, given that public accesslitigation between September 2013 and April 2015 led to professional fees exceeding $1.4 million. See Ex. C.



23/54

23

This 11th day of June, 2015.

Respectfully submitted,

/s/ Garland S. CassadaGarland S. Cassada N.C. Bar No. 12352Jonathan C. Krisko N.C. Bar No. 28625Richard C. Worf, Jr. N.C. Bar No. 37143

ROBINSON BRADSHAW & HINSON, P.A.101 North Tryon Street, Suite 1900Charlotte, North Carolina 28246Telephone: (704) 377-2536Facsimile: (704) 378-4000

[email protected] [email protected]@rbh.com

Special Corporate and Litigation Counsel to the Debtors Garlock Sealing Technologies LLC,Garrison Litigation Management Group, Ltd., andThe Anchor Packing Company



24/54

EXHIBIT A

Case 10-31607 Doc 4646-1 Filed 06/11/15 Entered 06/11/15 12:02:03 Desc Ex. A-Bates Declaration Page 1 of 17


25/54

Declaration of Charles E. Bates, Ph.D.

Page 1 of 9

IN THE UNITED STATES BANKRUPTCY COURT

FOR THE WESTERN DISTRICT OF NORTH CAROLINA

CHARLOTTE DIVISION

IN RE:

GARLOCK SEALING

TECHNOLOGIES LLC, et al.,1

Debtors.

Chapter 11

Case No. 10-BK-31607

(Jointly Administered)

DECLARATION OF CHARLES E. BATES, PH.D.

I, CHARLES E. BATES, declare as follows:

(1) I am Chairman of Bates White, LLC, which is an economic consulting firm with its primary

office located in Washington, DC. The Court approved the employment of Bates White as

Asbestos Claims Consultant for the Debtors on July 21, 2010. I have been asked by

Robinson, Bradshaw & Hinson, P.A. (“RBH”) on behalf of the Debtors to submit the present

declaration (the “Declaration”) to address some of the objections stated by the Manville

Personal Injury Settlement Trust (“Manville Trust”) in its “Objections”2 and some issues

raised by the Asbestos Claimants Committee (“ACC”) in its “Response”3 in regard to the

Debtors’ Motion for Leave to Serve Subpoena on Manville Trust (Dkt. No. 4599) (the

“Motion”) in connection with the estimation of Garlock’s non-mesothelioma asbestos

liability and the feasibility of the Debtors’ Second Amended Plan of Reorganization (the

“Plan”). I have personal knowledge of the facts set forth in this Declaration and, if called asa witness, could and would testify competently to such facts under oath.

1 The Debtors are Anchor Packing Company, Garlock Sealing Technologies LLC, and Garrison Litigation ManagementGroup Ltd.

2 See Objection of Non-Party Manville Personal Injury Settlement Trust to the Debtors’ Motion for Leave to ServeSubpoena, Dkt. No. 4638.

3 See Response and Limited Objection of the Official Committee of Asbestos Personal Injury Claimants to Debtors’

Motion for Leave to Serve Subpoena on Manville Trust, Dkt. No. 4644.



26/54

Declaration of Charles E. Bates

Page 2 of 9

(2) I received my Ph.D. and M.A. in economics from the University of Rochester and my B.A. in

economics and mathematics with high honors from the University of California, San Diego.

I have taught courses in advanced statistical economic analysis and trade theory while on the

faculty of Johns Hopkins University and have published papers on advanced topics inestimation theory in peer-reviewed journals. I specialize in the application of statistics and

computer modeling to economic and financial issues. I have 20 years of experience in a wide

range of litigation and commercial consulting areas. Much of my work has involved the

design and implementation of statistical and computer modeling methods to analyze large

quantities of data and to estimate the value of product liability claims against companies.

Attached as Exhibit A is my curriculum vitae.

(3) In this matter, I am the asbestos claims estimation expert for Debtors and testified regarding

Garlock’s mesothelioma liabilities in the Estimation Trial in front of the Hon. Judge George

Hodges. Further, I am the expert for Debtors who will estimate Garlock’s non-mesotheliomaliability and will assess the Plan’s feasibility in the upcoming Confirmation Hear ing in front

of this Court.

(4) In this Declaration, I cover the following four topics:

I explain the usefulness of the data requested from the Manville Trust both for

estimating Garlock’s non-mesothelioma asbestos liability and for its usefulness in

assessing the Plan’s feasibility.

I respond to the Manville Trust and the ACC’s incorrect assertion that a sample of

claimants would be preferable to receiving information on all claimants identified in

the Motion.

I respond to the Manville Trust’s concerns regarding the matching protocol proposed

by Debtors for identifying the claimants for which the Motion seeks information.

I respond to the ACC’s incorrect assertion that using and analyzing the data requested

from the Manville Trust would be burdensome and impractical.



27/54


Page 3 of 9

The usefulness of the data requested from the Manville Trust both for estimating

Garlock’s non-mesothelioma asbestos liability and in assessing the Plan’s feasibility

(5) Contrary to the Manville Trust and the ACC’s assertions, the methodology that I employed in

the estimation of mesothelioma asbestos liability and the one that I will employ for the

estimation of non-mesothelioma asbestos liability and Plan feasibility analysis relies on

individual-level data from claimants that is analyzed by using established econometric

statistical methods to reach conclusions about aggregate estimates. Therefore, the claimant-

and claim-level information requested by Debtors in the Motion will be very useful for non-

mesothelioma asbestos liability estimation and for Plan feasibility analysis.

(6) All data categories requested by Debtors in the Motion would provide valuable information

about Garlock’s non-mesothelioma claimants that is simply not available in the Garlock

Analytical Database or in the Garrison database.

4

Most of the information requested in theMotion was requested and granted in the Personal Injury Questionnaire (“PIQ”) issued to

pending mesothelioma claimants earlier in this case. For non-mesothelioma claimants, there

is no equivalent PIQ being requested due to time and cost constraints. Additionally, the

Debtors have no information about claimants who file a Proof of Claim form (“POC”) before

the October 6, 2015 bar date and who are not listed as having a pending claim in the Garlock

Analytical Database (the “Additional Pending Claims”); such claimants will be considered in

analyses for the Confirmation Hearing, and data beyond what they will provide in the POC

will be useful, given that no PIQ will be requested from this group, either. Therefore, a time-

and cost-effective substitute to requesting a PIQ from all such claimants is requesting such

information from the Manville Trust, which already collects the data in its electronic databaseor possesses documents submitted by claimants that contain such information.

(7) The information sought in the Motion can be classified into five categories: 1) Claimant

identifying information; 2) Disease and other alleged injury-related information; 3) Exposure

information; 4) Tort claimant information; and 5) Manville Trust claim information. Below I

describe the use of these data for the analyses required for the Confirmation Hearing:

1. Claimant identifying information. As explained below, information regarding the

alleged disease, exposure histories, and claim information are important inputs in

the analyses to be performed. The claimant identifying information will be used tomatch the Manville Trust data to specific claimants in the Garlock Analytical

Database or among the Additional Pending Claims. If the information provided

does not include identifying information, such a matching exercise would not be

4 See Dr. Jorge Gallardo-Garcia’s Expert Report on this matter (Estimation Trial Exhibit GST-8004) for a detaileddescription of the construction and sources comprising the Garlock Analytical Database and for a description of the

Garrison database.



28/54


Page 4 of 9

possible. Because one of the goals of the data request to the Manville Trust is to

supplement the existing records in the Garlock Analytical Database and to provide

unavailable information regarding the Additional Pending Claims, without claimant

identifying information the Manville Trust data from the other categories requestedwould not serve that purpose. This lack of claimant-by-claimant identification

would hinder the usefulness of such Manville Trust data. Therefore, complete

claimant identifying information is essential for the usefulness of the requested

data, so an anonymity protocol as suggested by the Manville Trust would

undermine the purpose of the data request.

2. Disease and other alleged injury-related information. The Garlock Analytical

Database only records a disease field that does not provide information in enough

detail to be able to classify the pending records in that database into the disease

categories contemplated by the Claims Resolution Procedures (“CRP”) that willgovern the Settlement Facility under the Plan. The Garlock Analytical Database

provides information to consistently classify the alleged diseases into only five

categories: mesothelioma, lung cancer, other cancer, non-malignant, and unknown

disease. In turn, the POC will provide disease category information as defined in

the Plan’s CRP but will not include additional information relevant to assessment

under the CRP, such as diagnosing doctor, tests, etc.

Both for liability estimation and for Plan feasibility analysis, additional diagnosis

information is required. Lung cancers, other cancers, and non-malignant cases need

to be classified based on their diagnosis (both in terms of the type of disease and thereliability of the diagnosis), smoking histories, and diagnostic tests and exams as

stated in the CRP. Further, the Garlock Analytical Database shows almost 21,000

pending records with no disease information at all (unknown disease). The

Manville Trust database would have such information if the corresponding

claimants filed a claim against both Garlock and the Manville Trust.

Detailed disease information is present and readily available, in part, in the

Manville Trust electronic database and, in part, in the medical records attached by

claimants to their Manville Trust claims. The detailed classification of claimants

into disease categories is fundamental in order to estimate the number and amount

of non-mesothelioma claims given that the liability determination and potential

compensatory award depend on the injured party’s disease. Additionally, this

detailed disease classification will allow estimating the number of claims to be filed

with the Settlement Facility and that potentially would qualify for payment under

the CRP.



29/54


Page 5 of 9

3. Exposure information. As in the case of mesothelioma claims, job and exposure

histories are very important to determine the likely contact of claimants with

Garlock asbestos-containing products and alternative sources of exposure to

asbestos, which are directly relevant in terms of liability estimation and are alsoamong the information requested by the Plan’s CRP for settlement determination.

5

The Garlock Analytical Database has almost no data on occupations and industries

for pending non-mesothelioma claims records, and, when available, the information

is sparse. For the Additional Pending Claims, Debtors will have no exposure and

job history information other than the assertion of Garlock product contact provided

in the POC. The Manville Trust database includes useful occupation and industry

information for most claims captured for the purposes of the Manville Trust claim.

That data can be further supplemented with the exposure documents submitted by

claimants as attachments to their Manville Trust claims.

Therefore, the job and exposure information from the Manville Trust is a readily

available source of information that can be of great aid in the classification of

Garlock non-mesothelioma pending claims and Additional Pending Claims into

Contact Groups both for liability estimation and for Plan feasibility analysis.

4. Tort claimant information. The information requested in the Motion from the

Manville Trust in this regard — information that includes data fields such as filing

date, jurisdiction, and representing law firm — are data fields that Bates White uses

as inputs to its claim/claimant matching algorithms and are also used to confirm

matches across claims databases. In this case, the information from the ManvilleTrust will be used in two ways: 1) as an input in the matching between the Manville

Trust database and both the Garlock Analytical Database and the Additional

Pending Claims to aid in the identification and eventual elimination of “false

positives” (see below for further discussion); and 2) like the other data categories

requested by the Motion, to supplement the Garlock Analytical Database when

some of that information is missing and to provide otherwise unavailable

information for the Additional Pending Claims.

5. Manville Trust claim information. Information regarding the non-mesothelioma

claimant’s Manville Trust claim is useful for liability estimation for two main

reasons: 1) filing date, fact of settlement, settlement/payment date, and other status

fields are used in analyzing the timing of filing claims against Garlock versus the

progression of the Manville Trust claim; and 2) payments from tort parties and

5 In the estimation of Garlock’s mesothelioma liability, Mr. John Henshaw (the Industrial Hygiene expert for Debtors)established a classification of industries and occupations into five Contact Groups that were characterized by the

frequency, regularity, and proximity of contact with asbestos-containing gaskets and packing.



30/54


Page 6 of 9

trusts are used to estimate the amount of offsets that would apply to a potential

plaintiff verdict in joint-and-several and hybrid jurisdictions.6

(8) In summary, each of the data categories requested by Debtors in the Motion is an important

input for the estimation of Garlock’s non-mesothelioma liability and for the Plan feasibility

assessment that I will perform for the Confirmation Hearing. Currently, there are no

alternative sources of discovery or publicly available information that would provide these

data in a more time- and cost-effective manner than having the Manville Trust produce them

to Debtors.

The Manville Trust and the ACC are incorrect in asserting that a sample of claimants

would be preferable to receiving information on all claimants identified in the Motion

(9) Both the Manville Trust and the ACC erroneously assert in their objections that providing

information for the claimants identified in the Motion is excessive and that providing the

information for a limited sample of such claimants would be less burdensome and would

suffice for the Confirmation Hearing analyses. This is simply not true.

(10) First, the Manville Trust maintains its records in electronic form readily available for

electronic transfer, so the burden of providing the data and documents for the claimants

identified in the Motion will be essentially the same as the burden of providing the same

information for a subset of those claimants.7 When the data and documents are stored in

electronic form in an organized manner, querying a smaller or a larger subset of the records

only changes the computer processing time (likely only by minutes at the most) but not the

effort that an individual has to put into the task. Therefore, the Manville Trust and the ACC’s

assertions that providing data on a sample of claims would be less burdensome are simply

not true.

(11) Second, although I expect that the pending non-mesothelioma records and the Additional

Pending Claims will significantly overlap with the Manville Trust claims, the claimants who

would overlap are not currently known and cannot be identified without actually performing

a full matching exercise between the Garlock records and the Manville Trust database.

Given an unknown non-match rate, drawing a statistical representative sample would require

a large sample that, as I explained before, would require the same amount of effort as

producing information for all claims identified in the Motion.

6 As I covered in my Expert Report, section V.3.7 (Estimation Trial Exhibit GST-0996), hybrid jurisdictions are those thatapply a combination of several and joint-and-several liability sharing rules when apportioning compensatory damages.

7 See the Declaration of Garland S. Cassada.



31/54


Page 7 of 9

(12) Third, although statistical sampling is useful when the cost of gathering the information for

the whole population is significantly higher than the cost of gathering information for a

reduced group, drawing inferences about the whole population from a sample is less precise

than actually observing a given characteristic from the population itself. Because a sample(even if representative) is not the same as the population, all analyses have to be modified to

account for the statistical uncertainty that a sample introduces. This increases the technical

difficulty of the analysis and increases the amount of work that has to be performed to test

and ensure that the statistics being used to draw conclusions about the population are truly

representative of such population. Therefore, any time that a census is feasible in terms of

time and cost, it is preferable to sampling.

(13) Fourth, some of the Manville Trust’s assertions regarding sampling and statistical analysis

are incorrect or misleading. The Manville Trust asserts that a sample of 10% of the

population would be sufficient for the analyses to be performed. This is simply incorrect. A basic statistical concept in sampling is that the size of a representative sample does not

depend on the size of the population but rather on the variation of the feature in the

population that is going to be analyzed and about which one will draw statistical inferences.

Therefore, the sample size cannot be defined as 10% of the population without knowing

which features in the data are going to be analyzed or the analytical methodology that is

going to be applied. In this sense, the Manville Trust shows a complete lack of

understanding of the most basic statistical concepts.

(14) A misleading assertion by the Manville Trust is that, because some of the analyses used in the

estimation of Garlock’s liability for mesothelioma claims were based on sampling, it must bethat a sample for requesting data from the Manville Trust will suffice for analysis. In those

cases, samples were used for a variety of reasons (including considerations of cost) and were

feasible because the PIQ had already provided broad data regarding pending mesothelioma

claimants that could guide the sampling exercise. These factors are not present here because,

as I explained above, the Manville Trust maintains its records in electronic form so no

information or documents need to be gathered by the Manville Trust to respond to the

Motion’s request; further, there has been no PIQ that could guide the construction of an

appropriate sample.

(15) In summary, although sampling is a useful tool when the time and cost of gathering data is

significant, it is always preferable in terms of statistical precision and technical difficulty to

have access to data from the whole population. In this case, because the Manville Trust has

all data requested in the Motion in an organized electronic form, a sample would not be better

for the Confirmation Hearing analyses than having access to all data requested by Debtors.



32/54


Page 8 of 9

Response to the Manville Trust’s concerns regarding the matching protocol proposed

by Debtors

(16) The Motion requests Manville Trust data for all claimants in the Manville Trust records who

match the first and last name of one of the pending non-mesothelioma records or who will

match a claimant with an Additional Pending Claim. Although this process will generate

matches to Manville Trust claimants who are not Garlock claimants (referred to as “false

positives” by the Manville Trust) in an initial step, Bates White will analyze the data received

from the Manville Trust to confirm or reject matches by using Bates White’s matching

algorithms, which utilize all available identifying claimant and claim information to evaluate

a match.

(17) The broad first matching step based on first and last name is performed in order to avoid

“false negatives” and ensure that Garlock receives data on as many Garlock claimants as possible at the end of the process. When matching two different data sets (especially when

the data sets are large, as in this case) the more fields one uses in a strict comparison across

databases, the more likely it is that records belonging to the same individual would not be

matched due to differences in the data fields used in the matching exercise. For instance, one

may think that matching on Social Security Number (“SSN”) and claimant name would be

sufficient to efficiently and uniquely identify claimants across databases. However, in this

case, such a strict match rule would result in far too few matches because the Garlock

Analytical Database does not have SSN information for almost two-thirds of the pending

non-mesothelioma claims. Therefore, if the SSN field were used in matching the Garlock

non-mesothelioma claims to the Manville Trust, only one-third of the pending non-mesothelioma claims would have a chance of matching. Such a matching exercise would

introduce both known and unknown biases into the data provided by the Manville Trust for

the matched records because, in my experience, the fact that a data field is missing for a

record in an asbestos database is not random.

(18) In summary, the first matching step to be performed by the Manville Trust on first and last

name would ensure the integrity of the data provided by the Manville Trust, whereas the

second matching step to be performed by Bates White will endeavor to minimize the number

of false positives that remain in the data for Confirmation Hearing analyses.



33/54


Page 9 of 9

The ACC is incorrect in asserting that using and analyzing the data requested from the

Manville Trust would be burdensome and impractical

(19) The ACC misleadingly argues that using and analyzing the data requested from the Manville

Trust would come at a cost and effort multiples of the one incurred in processing and

analyzing the mesothelioma PIQ information. This is false and highly misleading. The cost

of processing and analyzing the PIQ data was directly associated with the way in which

claimants and their representing law firms submitted their information in response to the

questionnaire.8 Claimants did not respond to the PIQ questions in the electronic format

provided but rather submitted thousands of paper documents and handwritten PIQ formats

that needed to be scanned, classified, reviewed and analyzed for data extraction. With

respect to the Motion, that is not an issue because the Manville Trust already has an

electronic database with a large portion of the data requested already captured, and it also has

an electronic document repository where the documents submitted by claimants are alreadylinked to claimants and classified. Given that all data and documents would be provided by

the Manville Trust in a clean and organized manner, the cost of analyzing such information

would be significantly less than the one incurred in using the PIQ information, not multiples

of it. Therefore, the comparison made by the ACC regarding the cost of analysis between the

PIQ and the Manville Trust request is simply inappropriate and false.

(20) In the event that the Court seeks a more detailed understanding of my opinions, I am

available to appear before the Court.

I declare under penalty of perjury that to the best of my knowledge the foregoing is true andcorrect.

Dated on this 11th day of June, 2015.

_____________________________________

Charles E. Bates, Ph.D.

8 See Dr. Gallardo-Garcia’s Expert Report, section II.3 (Estimation Trial Exhibit GST-8004) for a detailed description ofthe information submitted by claimants in response to the mesothelioma PIQ and the steps performed for processing the

data for analysis.



34/54

EXHIBIT A



35/54

1300 Eye Street NW, Suite 600, Washington, DC 20005 main 202.408.6110 fax 202.408.7838

BATESWHITE.CO

SUMMARY OF EXPERIENCE

Charles E. Bates has extensive experience in statistics, econometric modeling, and economic analysis.

He specializes in the application of statistics and computer modeling to economic and financial issues.

Dr. Bates has more than 20 years of experience and provides clients with a wide range of litigation and

commercial consulting services, including expert testimony and guidance on economic and statistical

issues.

Dr. Bates is a recognized expert in asbestos-related matters. He speaks in national and international

forums on the asbestos litigation environment and estimation issues. Dr. Bates is frequently retained to

serve as an expert on such matters in large litigations and has testified before the United States Senate

Judiciary Committee and Federal Bankruptcy Court.

SELECTED ASBESTOS AND PRODUCT LIABILITY EXPERIENCE

Served as an asbestos liability valuation expert on behalf of Garlock Sealing Technologies in its

bankruptcy proceedings. Testified before the US Bankruptcy Court for the Western District of North

Carolina both in preliminary case hearings and at trial.

Served as an asbestos liability valuation expert on behalf of Specialty Products Holding Corp./Bondex

International in its bankruptcy proceedings.

Retained as an asbestos liability valuation expert on behalf of the Official Committee of Unsecured

Creditors of Motors Liquidation Company (f/k/a General Motors Corporation) in its bankruptcy

proceedings.

Authored expert report and provided deposition testimony regarding the value of diacetyl claims on

behalf of the Official Committee of Equity Security Holders in the Chemtura Corporation bankruptcy

proceedings.

Testified in deposition on behalf of the ASARCO Unsecured Creditors Committee in the ASARCO

bankruptcy proceedings regarding the valuation of past and future asbestos-related personal injury

claims.

CHARLES E. BATES, PHD

ChairmanAREAS OF EXPERTISE

Asbestos liability estimation

Economic analysis

Econometrics

Statistical analysis

Microsimulation modeling

Database design/applications



36/54

E.COM

CHARLES E. BATES, PHDPage 2 of 6

Authored expert report and provided deposition testimony on behalf of the policyholder in the matter

of Imo Industries, Inc. v. Transamerica Corp.

Currently retained as an expert by Fortune 500 companies to produce asbestos expenditure

estimates for annual and quarterly financial statements. Estimations aid clients with Sarbanes-Oxley

compliance.

Currently retained as an expert in asbestos estimation and insurance valuation, for numerous

asbestos litigation matters, on behalf of insurance companies, corporations, and financial creditors’

committees of federal bankruptcy proceedings.

Testified before the Senate Judiciary Committee on the economic viability of the Trust Fund proposed

under S.852, the Fairness in Asbestos Injury Resolution (FAIR) Act of 2005. Testimony clarified Bates

White's independent analysis on the estimate of potential entitlements created by the administrative

no-fault trust fund that uses medical criteria for claims-filing eligibility.

Testified in deposition on behalf of Liberty Mutual Insurance Company in the Plibrico bankruptcy

proceedings regarding the valuation of past and future asbestos personal injury claims and exposure

criteria in plan proponents proposed trust distribution procedures.

Testified at deposition on behalf of the joint insurers defense committee to address the fraction of

expenditures associated with the company’s asbestos installation operations in Owens Corning v.

Birmingham Fire Insurance Company of Pennsylvania.

Testified in the Babcock & Wilcox bankruptcy confirmation hearing on behalf of the Insurers Joint

Defense Group to address asbestos liability. Developed claims criteria evaluation framework to

assess asbestos liability forecasts and trust distribution procedures.

Testified at deposition on behalf of Sealed Air in the fraudulent conveyance matter regarding the 1998

acquisition of Cryovac from W.R. Grace. Directed estimation of foreseeable asbestos liability for

fraudulent conveyance matter to advise the debtor in the bankruptcy of a defendant with over $200

million in annual asbestos payments. Developed asbestos liability forecasting model and software.

Directed industry research and interviewed industry experts.

Testified at deposition on behalf of Hartford Financial Services Group to address the asbestos liability

of MacArthur Company and Western MacArthur Company. Estimated asbestos liability in the context

of bankruptcy proceedings.

Testified at deposition on behalf of the Center for Claims Resolution in arbitration proceedings of GAF

v. Center for Claims Resolution.

Served as testifying expert on behalf of CSX Transportation on the suitability of asbestos claim

settlements for arbitration proceedings of CSX Transportation, Inc. v. Lloyd’s, London.

Developed an econometric model of property damage lawsuits for estimating the future liability of a

former asbestos manufacturer arising from the presence of its asbestos products in buildings.



37/54

E.COM

CHARLES E. BATES, PHD

garlock's reply in support of subpeona to manville trust - 533816_4646

Documents