GONZALEZ v. OWENS CORNING et al
Filing
181
OPINION re 150 MOTION to Certify Class. Signed by Chief Judge Joy Flowers Conti on 3/31/2016. (ten)
IN THE UNITED STATES DISTRICT COURT
FOR THE WESTERN DISTRICT OF PENNSYLVANIA
JAIME GONZALEZ, PATRICIA WRIGHT,
KEVIN WEST, and GERALD BOEHM,
On behalf of themselves and all others similarly
similarly situated,
)
)
)
)
)
Plaintiffs,
)
v.
) Civil Action No. 13-cv-1378
)
OWENS CORNING and OWENS
)
CORNING SALES, LLC,
)
)
Defendants.
)
______________________________________________________________________________
EDWARD MAAG and DIANE MAAG, on
behalf of themselves and all others similarly
situated,
Plaintiffs,
v.
)
)
)
)
) Civil Action No. 14-cv-0826
)
OWENS CORNING and OWENS
)
CORNING SALES, LLC,
)
Defendants.
)
__________________________________________________________________________
OPINION
CONTI, Chief District Judge
These putative class actions were filed against Owens Corning and Owens
Corning Sales, LLC (collectively, “Owens Corning” or “defendants”) and arise out of Owens
Corning’s manufacture and sale of allegedly defective Oakridge-brand fiberglass asphalt roofing
shingles. One case was filed against Owens Corning in this court, and three separate cases were
filed in federal district courts in other states. Those other federal district courts transferred the
three cases to this court. After being transferred two of those cases were consolidated for all
purposes with the case filed here at civil action 13-1378.1 (ECF No. 45.) The last case
transferred to this court (Civ. No. 14-0826) was consolidated at civil action 13-1378 only for
pretrial purposes. (14-cv-826, ECF No. 14.) The named plaintiffs in the four cases are Patricia
Wright (“Wright”), Kevin West (“West”), Jaime Gonzalez (“Gonzalez”), Gerald Boehm
(“Boehm”), and Edward and Diane Maag (the “Maags”) (collectively, “plaintiffs” or the “named
plaintiffs”).
The procedural history, facts, and legal claims pertinent to each of the cases will
be discussed in detail in the findings of fact that follow. By way of summary, plaintiffs contend
that Owens Corning acted unlawfully by manufacturing Oakridge-brand shingles in accordance
with defective design specifications, and by promising that all Oakridge-brand shingles would
last for at least 25 years, when, due to those defective design specifications all Oakridge-brand
shingles were “vulnerable” or “susceptible” to lasting no more than 20 years.
Plaintiffs filed a motion for class certification, seeking to certify classes pursuant
to Federal Rules of Civil Procedure 23(b)(1)(B), (b)(2), and (b)(3). (ECF No. 150.) Plaintiffs’
motion was fully briefed by the parties, and the court conducted a hearing on the motion on
December 17, 2015. (ECF Nos. 151-54, 161-62, 164-66, 169-70, 12/17/2015 Minute Entry.)
Thereafter, the parties filed proposed findings of fact and conclusions of law. (ECF Nos. 17677.) Several weeks later, plaintiffs volunteered a notice of supplemental authority in further
support of their motion for class certification. (ECF No. 179.) Owens Corning filed a short
response thereto, and stated its willingness to more fully respond at the court’s request. (ECF No.
180.) The court did not request further briefing.
1
Citations to the record that are not preceded by any civil action number are made to the docket
of the lead case pending at 13-cv-1378.
2
The court concludes that the proposed Rule 23(b)(1)(B) class cannot be certified
because the named plaintiffs ask this court to answer a question that the Court of Appeals for the
Third Circuit already answered. Owens Corning cannot, and has stated that it will not, relitigate
the issue. In addition, the appellate court answered the question in a way that makes it
impossible for this court to enter any classwide rulings with respect to the effect that Owens
Corning’s bankruptcy proceedings have on proposed class members’ claims.
The court also concludes that the proposed Rule 23(b)(2) and Rule 23(b)(3)
classes cannot be certified because, among other reasons, the named plaintiffs seek to pursue
relief under various state-law theories that are not the same for all members of a proposed class,
it is impossible to determine whether an owner is a member of the class, and the record
contradicts any finding that either all (or even most or many) Oakridge-brand shingles suffer
from an common defect or Owens Corning represented that its Oakridge-brand shingles would
not crack, degranulate, fragment, or deteriorate for, or would have a useful life of, at least 25
years. Plaintiffs proffer no evidence about how often Owens Corning manufactured shingles “at
or near” the allegedly defective “low-end” of its specifications. FOF 181. The only statistical
evidence in the record reflects that only one half of one percent of Oakridge-brand shingle
installations result in a warranty claim, and only half of the approximately 300 warranty claim
shingles tested by plaintiffs measured “at or near” the allegedly defective “low end” of Owens
Corning’s design specifications. The record reflects that plaintiffs acknowledge that not all
Oakridge-brand shingles will be manufactured “at or near” the allegedly defective “low end” of
Owens Corning’s design specifications, and plaintiffs never identify how near the “low end” of
Owens Corning’s design specifications a measurement must be to qualify as design defect. With
respect to plaintiffs’ claims that Owens Corning made misstatements about its Oakridge-brand
3
shingles, plaintiffs fail to establish that Owens Corning made uniform representations about the
expected useful life of Oakridge-brand shingles, or about whether the shingles would experience
any form of deterioration for a set number of years.
Under those circumstances and as more fully explained in the findings of fact and
conclusions of law, the court will deny plaintiffs’ motion for class certification.
FINDINGS OF FACT2
A. The Proposed Classes
1. The Originally-Proposed Classes
FOF 1:
In their initial briefing on the motion for class certification, plaintiffs moved to
certify the following classes:
Rule 23(b)(1)(B) Impaired Judgments Class: All individuals and entities
that own a building or structure physically located in the United States on
which Owens Corning’s Oakridge-brand shingles are or have been installed,
where those shingles were purchased on or before September 26, 2006.
Rule 23(b)(2) Injunctive Relief Class and Rule 23(b)(3) Monetary Relief
Class: All individuals and entities that own a building or structure
physically located in the states of California, Illinois, Pennsylvania, or Texas
on which Owens Corning’s Oakridge-brand shingles are or have been
installed from 1992 through 2012.
(ECF No. 154 at 23.)
2
The court makes these factual findings in furtherance of its obligations to conduct a rigorous
analysis of plaintiffs’ motion for class certification. Among those obligations is that this court
resolve all factual disputes relevant to class certification, including disputes touching on the
elements of the causes of action being asserted and the merits of a claim, and make explicit
findings that plaintiffs satisfied their burden to prove each of the requirements of Rule 23 by a
preponderance of the evidence. Reyes v. Netdeposit, LLC, 802 F.3d 469, 484 (3d Cir. 2015);
Hayes v. Wal-Mart Stores, Inc., 725 F.3d 349, 354 (3d Cir. 2013); In re Hydrogen Peroxide
Antitrust Litig., 552 F.3d 305, 307, 316 n.14, 317, 321 (3d Cir. 2008).
4
FOF 2:
With respect to the originally-proposed Rule 23(b)(1)(B) class, plaintiffs
explained in their initial briefing in support of the motion for class certification that “class
members nationwide are vulnerable to Owens Corning asserting a dischargeability
defense against them” even though the Court of Appeals for the Third Circuit ruled in
favor of Wright and West on this bankruptcy-based defense during the pendency of the
instant case. (ECF No. 154 at 34); Wright v. Owens Corning, 679 F.3d 101 (3d Cir.
2012).
a. In October 2000, Owens Corning voluntarily filed for bankruptcy relief
under Chapter 11 of the Bankruptcy Code in the United States Bankruptcy
Court for District of Delaware. Wright, 679 F.3d at 103; Wright v. Owens
Corning, 450 B.R. 541, 545 (W.D. Pa. 2011).
b. Owens Corning’s final reorganization plan was confirmed on September 26,
2006. Wright, 679 F.3d at 103; Wright, 450 B.R. at 545.
c. In Wright, the court of appeals refused to retroactively apply a newlyannounced bright-line legal test for determining whether a claim is
discharged by a bankruptcy court’s entry of a confirmation order. Wright,
679 F.3d at 107-09. The court of appeals instead engaged in a fact-specific
inquiry, as required under the previously-applicable legal test, and
concluded that Wright’s and West’s claims were not discharged by Owens
Corning’s bankruptcy proceedings. Id.; see FOF 229-31.
FOF 3:
The originally-proposed Rule 23(b)(2) and Rule 23(b)(3) classes shared the same
class definition. Membership in the proposed Rule 23(b)(2) class, therefore, was
coterminous with membership in the proposed Rule 23(b)(3) class. Owners were not
5
required to demonstrate any kind of shingle deterioration, property damage, or actual
repair or replacement costs in order to obtain relief.
a. At the hearing on plaintiffs’ class certification motion, in response to
questions from the court about the seeming redundancy of the proposed
Rule 23(b)(2) and Rule 23(b)(3) classes, plaintiffs clarified that the
proposed Rule 23(b)(2) injunctive relief class was intended to be an
alternative in the event that the court declined to certify the proposed Rule
23(b)(3) monetary relief class. (ECF No. 173 at 37-38.)
b. Plaintiffs explained at the hearing on their class certification motion that if
the court refused to certify the Rule 23(b)(3) monetary relief class, then
members of the proposed Rule 23(b)(2) class would obtain a declaration
that the shingles are defective or that Owens Corning did not provide what
was promised. (ECF No. 173 at 38-39.)
FOF 4:
Plaintiffs sought to litigate four categories of legal claims on behalf of the classes
proposed under Rules 23(b)(2) and 23(b)(3): (1) breach of express warranty (PA owners
only); (2) breach of implied warranty of merchantability (PA, CA, and TX owners); (3)
violation of state-specific consumer protection statutes (CA, TX, and IL owners); and (4)
unjust enrichment (PA, CA, TX, and IL owners). (ECF No. 154 at 40-45.)
a. The unjust enrichment claim was the only cause of action on which plaintiffs
sought class certification under the laws of all four states in which the namedplaintiffs’ structures are located. The three remaining causes of action were
being pursued on behalf of varying combinations of owners located in the states
of Pennsylvania, Illinois, California, and Texas. (Id. at 45.)
6
b. The named plaintiffs’ operative complaints assert numerous legal claims for
which class certification was not being sought.
i. For example, although Gonzalez, Boehm, and the Maags plead
breach of express warranty claims under Texas, California, and
Illinois law, respectively, plaintiffs sought to pursue only the
breach of express warranty claim asserted by Wright under
Pennsylvania law on behalf of the class. FOF 21, 46, 70, 88.
ii. Similarly, although West and the Maags plead breach of implied
warranty claims under Illinois law, the named plaintiffs sought to
certify that legal claim on behalf of those owners whose structures
are located in Pennsylvania, California, or Texas. Therefore, while
owners in those three states would have their breach of implied
warranty claims adjudicated on a classwide basis, owners in
Illinois would not, and West and the Maags would, instead,
affirmatively abandon those claims on behalf of the class. (ECF
No. 173 at 21-32); FOF 21, 46, 70, 88.
FOF 5:
In a footnote to their opening brief, plaintiffs suggested certification of an
alternative “liability-only class under Rule 23(c)(4), with damages handled in separate
proceedings.” (ECF No. 154 at 37 n.17; ECF No. 170 at 14.) Plaintiffs did not propose a
list of issues that would be decided as part of such a class, but instead argued generally
that a “liability-only class will fairly and efficiently advance this litigation” “[g]iven the
parties’ dispute regarding an intrinsic product defect as well as Owens Corning’s
7
deceptive representations about the reliability and durability of Oakridge.” (ECF No. 170
at 15.)
2. The Currently-Proposed Classes
FOF 6:
In their proposed findings of fact and conclusions of law, plaintiffs seek
certification of the same three classes under Rules 23(b)(1)(B), (b)(2), and (b)(3), with
minor modifications having been made to the originally-proposed class definitions. The
modifications made to the proposed class definitions are indicated below with
strikethrough, underlining, and highlighting, and are discussed in substance in the
findings of fact that follow.
Rule 23(b)(1)(B) Impaired Consistent-Judgments Class, defined as Aall
individuals and entities that own a building or structure physically located in
the United States on which Owens Corning’s Oakridge-brand shingles are
currently or have been installed, where those shingles were purchased on or
before September 26, 2006.
A Rule 23(b)(2) Injunctive Relief Class and Rule 23(b)(3) Monetary Relief
Class, defined as Aall individuals and entities that own a building or structure
physically located in the states of California, Illinois, Pennsylvania, or Texas on
which Owens Corning’s Oakridge-brand shingles were are or have been installed
from 1992 through 2012, and where those shingles manifested any cracking,
degranulation, fragmentation, or deterioration during the warranty coverage period.
(ECF No. 178 at 18-19 (¶ 1(a)-(b)).)
FOF 7:
The most significant difference between the originally-proposed Rule 23(b)(1)(B)
class, and the currently-proposed Rule 23(b)(1)(B) class is that a member of the
currently-proposed Rule 23(b)(1)(B) class must own a building or structure on which
Oakridge-brand shingles are currently installed. FOF 1, 6.
FOF 8:
The purpose of the proposed Rule 23(b)(1)(B) class remains the same: to address
the effect of Owens Corning’s bankruptcy discharge on owners who hold legal claims
relating to Oakridge-brand shingles. FOF 2; (ECF No. 178 at 27-30 (¶¶ 41-51).)
8
Plaintiffs contend that this class will protect “the rights of class members whose interests
are otherwise vulnerable to inconsistent dischargeability rulings.” (Id. at 30 (¶ 51).)
FOF 9:
The most significant difference between the originally-proposed Rule 23(b)(2)
and Rule 23(b)(3) classes, and the currently-proposed Rule 23(b)(2) and Rule 23(b)(3)
classes is that a member of both currently-proposed classes must demonstrate that one of
the four shingle conditions listed in the class definition manifested during the term of the
applicable warranty period. FOF 1, 3, 6.
FOF 10: Membership in the currently-proposed Rule 23(b)(2) class remains coterminous
with membership in the currently-proposed Rule 23(b)(3) class. If an owner can
demonstrate that the “shingles manifested any cracking, degranulation, fragmentation, or
deterioration during the warranty coverage period,” then that owner is a member of both
the Rule 23(b)(3) monetary damages class, and the Rule 23(b)(2) injunctive relief class.
If an owner cannot demonstrate that the “shingles manifested any cracking,
degranulation, fragmentation, or deterioration during the warranty coverage period,” then
that owner is a member of neither the Rule 23(b)(3) monetary relief class, nor the Rule
23(b)(2) injunctive relief class.
FOF 11: Despite the representations made by plaintiffs during the class certification
hearing, plaintiffs do not, in their proposed findings of fact and conclusions of law, seek
certification of the Rule 23(b)(2) injunctive relief class as an alternative to certification of
the Rule 23(b)(3) monetary damages class. FOF 3(a) & 3(b).
FOF 12: Instead, in their proposed findings of fact and conclusions of law, plaintiffs
contend that both a Rule 23(b)(3) and a Rule 23(b)(2) class should be certified because
the former will provide relief for owners who have already “suffered substantial product
9
failures” and the latter will “protect class members who have yet to suffer substantial
product failures.” (ECF No. 178 at 30 (¶ 54)).
FOF 13: Plaintiffs seek to pursue the same four categories of legal claims on behalf of the
classes proposed under Rule 23(b)(2) and Rule 23(b)(3) that were set forth in their
original briefing: (1) breach of express warranty (PA owners only); (2) breach of implied
warranty of merchantability (PA, CA, and TX); (3) violation of state-specific consumer
protection statutes (CA, TX, and IL); and (4) unjust enrichment (PA, CA, TX, and IL).
FOF 4, 21, 46, 70, 88; (ECF No. 178 at 34-43 (¶¶ 72-106).)
FOF 14: Even though plaintiffs, upon questioning by the court at the class certification
hearing, expressed their belief that the Rule 23(b)(2) and Rule 23(b)(3) classes should
have state-by-state subclasses, and stated that they “would be happy to” propose statespecific subclasses in their post-hearing filings, plaintiffs do not propose state-specific
subclasses in their proposed findings of fact and conclusions of law. (ECF No. 173 at 1719.) Plaintiffs, instead, include a proposed conclusion of law that states: “To the extent
that trial requires separate consideration of state-law claims, this Court reserves the right
to certify separate state subclasses.” (ECF No. 178 at 19 (¶ 1(d)); see also ECF No. 178 at
34 (¶ 71); but see ECF No. 178 at 34 (¶ 72) and 36 (¶ 81) (making passing reference in
proposed conclusions of law to state-specific subclasses for the express warranty and
implied warranty claims even though those classes are never defined in plaintiffs’
submission).)
FOF 15: At the hearing on plaintiffs’ motion for class certification, the court explicitly
advised plaintiffs that it was their obligation to propose state-specific subclasses at the
class certification stage, and that the matter was not a trial management issue in this case.
10
(ECF No. 173 at 18.) Plaintiffs nevertheless elected not to propose state-specific
subclasses for the proposed Rule 23(b)(2) and (b)(3) classes.
FOF 16: Plaintiffs do not include a Rule 23(c)(4) class in the list of classes that they ask
the court to certify. (ECF No. 178 at 18-19 (¶ 1).) They do, however, include five
paragraphs in their proposed findings of fact and conclusions of law which discuss a Rule
23(c)(4) liability-only class. (ECF No. 178 at 45-46 (¶¶ 115-119).)
a. Plaintiffs contend that a liability-only class is appropriate in this case
because such proceedings will be efficient, with respect to both time and
cost, and will allow for “focused damages proceedings.” (ECF No. 178 at
46 (¶¶ 118-19).)
b. Plaintiffs assert that a Rule 23(c)(4) class would “proceed to trial on the
threshold question of liability for the following claims: breach of express
warranty under Pennsylvania law; breach of implied warranty under
Illinois, Pennsylvania, and Texas law; violation of consumer protection
laws in California, Illinois, and Texas; and unjust enrichment” because,
“[n]otwithstanding some minor differences between the legal theories, all
these claims rest on a common core of facts: design defects in Oakridge;
Owens Corning’s knowledge of those defects; and deceptive
representations about the reliability and durability of Oakridge.” (Id. at 4546 (¶¶ 117, 119).)
FOF 17: The class definitions for all three currently-proposed Rule 23(b) classes include:
a. Buildings and structures that are single-family dwellings, multi-family
dwellings, commercial facilities, and government-owned properties. FOF 6.
11
b. Owners that are individuals, corporations, unincorporated associations, and
governmental or municipal entities. FOF 6.
c. Owners who purchased structures on which a previous owner installed
Oakridge-brand shingles prior to the date of purchase. FOF 6.
d. Owners who jointly own a structure. FOF 6.
e. Owners for whom another individual or entity, such as a builder or
contractor, selected Oakridge-brand shingles for installation on the structure.
FOF 6.
f. For the proposed Rule 23(b)(2) and Rule 23(b)(3) classes, owners who
purchased structures from which a previous owner removed Oakridge-brand
shingles prior to transferring ownership to the current owner. FOF 6.
g. For the proposed Rule 23(b)(2) and Rule 23(b)(3) classes , owners whose
Oakridge-brand shingles will not crack, degranulate, fragment, or deteriorate
for 30 years, or more. FOF 193(f).
h. For the proposed Rule 23(b)(2) and Rule 23(b)(3) classes, owners whose
Oakridge-brand shingles will not require repair or replacement for 30 years,
or more. FOF 193(f).
i. For the proposed Rule 23(b)(2) and Rule 23(b)(3) classes, owners whose
Oakridge-brand shingles will not cause a roofing leak, or otherwise cause
property damage for 30 years, or more. FOF 193(f).
12
FOF 25: The brochure obtained by Wright includes, on the cover, a graphic representing a
“Limited 40 Year Product Warranty.” (ECF No. 153-12 at 14.) The brochure later
explains that a purchaser “automatically get[s] limited shingle warranty coverage” and
can purchase a warranty with additional benefits if the complete Owens Corning Roofing
System is installed. (Id. at 16.) A chart is included in the brochure that lists the “shingle
warranty” for various brands of Owens Corning shingles, and the “System Warranty
Non-Prorated Period” applicable to each brand if the additional system warranty is
purchased. (Id. (listing shingle and system warranties for Oakridge-, Prominence-,
Supreme-, Supreme Shadow-, Glaslock-, and Classic-brand shingles).) The brochure
states that the “40-year warranty* covers the prorated replacement cost of new shingles
and labor.” (Id. at 17 (emphasis in original).) The asterisk directs the reader to “see
actual warranty for details.” (Id.)
FOF 26: No one at the retail store told Wright anything about the warranty coverage for the
Oakridge-brand shingles that she selected. (ECF No. 153-10 at 8 (depo. pg. 97).)
FOF 27: Wright testified that “when we bought a 40-year shingle roof, we were promised
that was supposed to have been a good roof.” (ECF No. 153-10 at 5-6 (depo. pg. 43).)
FOF 28: Although Wright has documentation of the total cost of the addition to her home
on which the shingles were installed in 1999, that documentation does not reflect, and she
does not and never did know, what portion of the $77,500 total cost was attributable to
purchasing the Oakridge-brand shingles that she picked out, as opposed to related roofing
supplies, or the labor associated with installing the roof, or other aspects of the addition
project. (ECF No. 153-10 at 7 (depo. pgs. 83-84); ECF No. 162-4 at 12-13, 20.)
15
FOF 29: No later than 2009, Wright experienced water leaking around the skylights in the
addition. (ECF No. 162-4 at 23-24; ECF No. 153-10 at 4 (depo. pgs. 35-36).) Owens
Corning’s expert witness in the field of roofing, James Johnson (“Johnson”), attributed
the leakage to improperly installed flashing around the skylights in the addition. (ECF
No. 162-7 at 6.) Plaintiffs’ expert witness, Dean Rutila (“Rutila”), agreed with Johnson,
testifying during a Daubert hearing that the leaking around Wright’s skylights was not
caused by problems with the Oakridge-brand shingles installed on the addition. (ECF No.
139 at 34.)
FOF 30: Wright contends that a roofer who inspected the roof after she experienced the
water leak in 2009 reported to her that all the shingles on her roof were cracked. (ECF
No. 153-10 at 9 (depo. pg. 121).) The roofer did not tell Wright that the cracked shingles
caused water to leak around the skylights in the addition. (Id.) There is no evidence that
the roofer advised Wright to immediately replace, or repair, the shingles on her roof.
FOF 31: Wright submitted a warranty claim to Owens Corning, but rejected its $3,400
offer because the roof would cost her $13,000 to replace. (ECF No. 153-10 at 10-12
(depo. pgs. 131, 137, 146).)
b. Facts Applicable to West
FOF 32: West had Owens Corning Oakridge Pro 30 shingles installed on his home, which
is located in Illinois, in 2005. (ECF No. 162-6 at 8; ECF No. 153-11 at 4-6; ECF No.
153-12 at 22, 24.)
FOF 33: West selected the Oakridge-brand shingle after visiting a commercial retailer
located in “Galesburg,” where he obtained a one-page flyer for “Oakridge Pro 30 Onyx
Black” shingles. (ECF No, 153-11 at 4-6; ECF No. 153-12 at 24.) West instructed the
16
contractor he had hired to replace the roof on his house to use those shingles, and the
contractor purchased the Oakridge-brand shingles on West’s behalf from the Alexander
Lumber Company. (ECF No. 162-6 at 9-10.)
FOF 34: West does not know how much his contractor paid for the shingles, and testified
that it is not readily apparent to him from the documentation that he has from the project
what portion of the total cost is attributable to the cost of the shingles, as opposed to
related supplies, and labor. (ECF No. 162-6 at 11; ECF No. 153-12 at 22.)
FOF 35: West does not recall anyone at the retail store giving him information about
Owens Corning’s shingles. (ECF No. 162-6 at 9.) He maintains that the only
representations made to him by Owens Corning are found in the one-page flyer that he
obtained at the retail store. (ECF No. 162-6 at 28-29; ECF No. 153-12 at 24.)
FOF 36: The one-page flyer that West obtained at the retail store includes a graphic
representing a “Limited 30 Year Product Warranty” and, in the text, lists a “30-Year
Limited Warranty,” as well as a “70-MPH Wind Resistance Limited Warranty.” (ECF
No. 153-12 at 24.) The flyer emphasizes the aesthetic appeal of the shingles and states
that the shingles “offer increased curb appeal and low-maintenance durability.” (Id.)
FOF 37: West testified that the shingles were not durable because they did not last for 30
years. (ECF -No. 153-11 at 13.)
FOF 38: In June 2009 water began to leak into West’s family room, which is located in the
back of his home. (ECF No. 162-6 at 14.) This is referred to as the “back roof” of West’s
home.
a. West personally viewed the area from inside his house and outside his house,
while on a ladder, and contacted the contractor who installed the roof, who
17
visited several days later and caulked some cracks in the roof. (ECF No. 1626 at 15-16.)
b. An insurance adjuster later inspected the roof and opined that the leak was
being caused by either faulty shingles or airflow problems. (Id. at 16-17.)
c. West also contacted Owens Corning to make a warranty claim, which claim
was denied because the cracks, which were vertical cracks, were deemed to
have been caused by deck movement resulting from inadequate ventilation.
(Id. at 19; ECF No. 153-11 at 9-10.)
d. Johnson confirmed the presence of both deck movement and inadequate
ventilation during his visual inspection of West’s roof. (ECF No. 162-7 at 78.) There is no indication that Rutila conducted a visual inspection of West’s
home. (ECF Nos. 151-5 to -7 and 154-2.)
FOF 39: West was unaware that the middle and main roofs of his house allegedly exhibited
“cracking and degranulation” until shortly before his deposition was taken for purposes
of this litigation in April 2010, when he was told about the condition of his shingles by an
inspector hired by plaintiffs’ counsel. (ECF No. 162-6 at 20.)
FOF 40: West’s insurance company paid him more than the actual cost of the repairs to the
ceiling in his family room, which was damaged by the water leak. (ECF No. 162-6 at 2324, 30.)
FOF 41: The contractor who replaced the roof on West’s house in 2005, reshingled the
back roof, which is above his family room, free-of-charge, in 2009. (ECF No. 162-6 at
21-22, 30.)
18
FOF 42: West is not seeking to recover for any property damage to his home, but does seek
to recover the cost to reshingle his roof. (ECF No. 162-6 at 6, 30.) West does not specify
whether this includes the middle, main, or back roofs, or some combination of the same.
There is no evidence that West, to date, has reshingled any part of his roof, other than the
back roof.
2. The Gonzalez(TX) Case
FOF 43: On February 22, 2013, the same attorneys who were representing Wright and
West before this court, filed a putative class action complaint in the United States District
Court for the Southern District of Texas on behalf of Jaime Gonzalez (“Gonzalez”) and a
nationwide class of individuals who owned structures on which Owens Corning’s
Oakridge-brand shingles were installed. (ECF No. 1 ¶ 54.) Alternatively, Gonzalez
sought to represent a class of individuals who owned structures located in Texas, where
the structure owned by Gonzalez is located. (Id. ¶ 55.)
FOF 44: Gonzalez filed a First Amended Class Action Complaint in which he abandoned
his pursuit of a nationwide class and asked the court to certify a class of “individuals and
entities that have owned, own, or acquired [structures] physically located in Texas on
which Owens Corning Oakridge shingles are or have been installed since 1986.” (ECF
No. 23 ¶ 54 (referred to as the “Gonzalez Texas Class” in the chart that follows).)
FOF 45: Gonzalez alleges that Owens Corning’s shingles are “defectively designed and
manufactured in such a way that they fail prematurely, causing damage to the underlying
structures,” which defects are manifested by “cracking, curling, degranulation,” among
other things. (ECF No. 23 ¶¶ 11, 13.)
19
FOF 49: The salesman at Leyendecker Lumber told Gonzalez that the Oakridge-brand
shingles were “one of their best sellers” and “had more warranty.” (ECF No. 162-10 at 17,
19.) Gonzalez’s contractor told him that “it’s a good shingle.” (Id. at 17.)
FOF 50: Gonzalez selected the Oakridge Pro 30 shingles based upon price and length of
warranty. (ECF No. 162-10 at 16, 20; ECF No. 153-11 at 20.)
FOF 51: Gonzalez testified that a shingle with a 20-year warranty, for example, “should last
20 years.” (ECF No. 153-11 at 17.)
FOF 52: Although Gonzalez testified that he saw Owens Corning commercials on television,
he could not recall the content of any of those advertisements. (ECF No. 162-10 at 23-24.)
FOF 53: Gonzalez did not know the price of the shingles he purchased and could only
estimate that they cost “[a] couple of thousand” dollars. (ECF No. 162-10 at 22.) He did
not know what the labor cost associated with installing the shingles was. (Id. at 24.)
FOF 54: Although the documentary evidence for Gonzalez’s project includes an invoice that
reflects the purchase of roofing supplies from Leyendecker Lumber, the exhibit provided to
the court is illegible. (ECF No. 166-2 at 20-21; ECF No. 153-11 at 27.)
FOF 55: In 2012, Gonzalez experienced water leaking in his garage, kitchen (which is
adjacent to the garage), and master bathroom. (ECF No. 162-10 at 26; ECF No. 166-2 at 511.)
FOF 56: The roof over Gonzalez’s garage is stucco, or cement, and has no shingles on it.
(ECF No. 162-10 at 10-11; ECF No. 162-17 at 4; ECF No. 162-21 at 2.)
FOF 57: Gonzalez’s insurance company estimated the cost to repair the damage to the
interior of his house, in all areas, to be approximately $6,500. (ECF No. 162-17 at 2-4; ECF
No. 166-2 at 5-11.)
21
FOF 58: The insurer paid Gonzalez for the cost to repair the interior of his home, less a
deductible. (Id.; ECF No. 166-3 at 2.)
FOF 59: Gonzalez testified that the insurance adjuster told him that the insurance company
would not pay for the repair or replacement of the roof because “it was bad shingles.” (ECF
No. 153-11 at 24-25.)
FOF 60: The insurance adjuster’s written report, however, states that “the ridge shingles had
deteriorated. The field shingles however are in perfect condition.” (ECF No. 162-21 at 2.)
Hip and ridge shingles, which fold over the peaks of a roof, are not Oakridge-brand
shingles. (ECF No. 162-7 at 8.) Field shingles are Oakridge-brand shingles.
FOF 61: Gonzalez’s insurance carrier further reported that the stucco roof over the garage
was cracked, and there were maintenance and insulation issues with the shingled roof over
the master bathroom, but there was no storm damage to the roof. (ECF No. 166-2 at 24.)
FOF 62: Johnson opined that cracks in the cement roof above the garage caused the
leaking in the garage and kitchen, and that the boot flashing for the soil stack in the
master bedroom was improperly installed, causing the leaking in the master bathroom.
(ECF No. 162-7 at 8-9.)
FOF 63: When Rutila inspected Gonzalez’s roof for purposes of preparing his April 2014
expert report, he noted that “[s]hingle tabs are cracked and the shingles have apparent
surfacing granule loss.” (ECF No. 151-5 at 12-13.) Although recognizing the fact that the
one-story attached garage has a “fluid applied membrane” roof, without any shingles,
Rutila makes no further distinction between the unshingled garage roof and the remainder
of the roof in offering his opinions, even though Gonzalez experienced leaking in the
garage and adjacent kitchen. (Id.)
22
FOF 64: Gonzalez submitted a warranty claim to Owens Corning, which offered a payment
of $162.63, representing the prorated replacement cost for the hip and ridge shingles on
Gonzalez’s roof. (ECF No. 162-17 at 3.) Gonzalez’s warranty claim submissions include
only pictures of hip and ridge shingles. (ECF No. 166-2.)
FOF 65: Gonzalez rejected Owens Corning’s warranty offer, stating that the damage to his
roof was quoted to be in excess of $15,000, not including the nearly $7,000 in damage to
the interior of his home, which was paid for by his insurance company. (ECF No. 166-3.)
There is no evidence that Gonzalez actually replaced the shingles on his roof.
FOF 66: Owens Corning directed Gonzalez to submit additional samples and photographs if
shingles other than the “hip and ridge” shingles were “bad/affected.” (ECF No. 166-03 at
2.) There is no evidence that Gonzalez ever supplemented his warranty claim submission
to Owens Corning.
3. The Boehm(CA) Case
FOF 67: Several days after the Gonzalez case was filed in Texas, on February 28, 2013, a
subset of the same attorneys who were representing Wright and West before this court and
Gonzalez before the Texas district court filed a putative class action complaint in the
United States District Court for the Central District of California on behalf of Gerald
Boehm (“Boehm”) and a nationwide class of individuals who owned structures on which
Owens Corning’s Oakridge-brand shingles were installed. (13-cv-936, ECF No. 35-1.)
Alternatively, Boehm sought to represent a class of individuals who owned structures
located in California. (Id.)
23
FOF 77: Although Boehm generally recalled Owens Corning commercials that included
the “pink panther,” he did not remember the content of any such advertisement, and never
saw any marketing materials, other than the brochure he obtained at Ford Wholesale,
before purchasing the shingles for his roof in 1997. (ECF No. 162-9 at 10, 17.)
FOF 78: Boehm selected the Oakridge Shadow-brand shingles based upon “the warranty,
the name [Owens Corning], and the color,” and the architectural styles (ECF No. 162-9 at
13; ECF No. 153-11 at 33.)
FOF 79: Boehm decided to purchase the Oakridge-brand shingles that were accompanied
by the warranty with the longest duration because each upgrade in the number of years
for which the warranty applied only cost a few hundred dollars more, and he considered
“it well worth the extra 4- or $500, whatever it was, to get the higher warranty.” (ECF
No. 162-9 at 12.)
FOF 80: Boehm stated “my 40-year roof that I think should probably last 40 years” and
that all he wants is for his “roof to last 40 years.” (ECF No. 162-9 at 3.)
FOF 81: In 2010, a roofer who was recoating a drain system and replacing flashing around
several solar lights on Boehm’s roof informed Boehm that the shingles on his roof
“looked like they were starting to crack.” (ECF No. 162-9 at 7-8.)
FOF 82: In February 2011, Boehm submitted a warranty claim to Owens Corning, stating
that there were top surface cracks mainly on south-facing slopes, but with evidence of
cracks in most other areas. (ECF No. 162-9 at 18-19; ECF No. 162-8 at 3.)
FOF 83: Owens Corning offered a payment of approximately $2,300, which Boehm
rejected because he estimated that it would cost more than double that amount to replace
26
his roof. (ECF No. 153-11 at 34-35; ECF No. 162-8 at 2.) Boehm never obtained an
estimate from a contractor for the cost to replace his roof. (ECF No. 162-9 at 22.)
FOF 84: The total cost to replace the roof on Boehm’s residence in 1997 was
approximately $10,000, but there is no evidence in the record about what portion of that
cost was attributable to the shingles purchased from Ford Wholesale for the project. (ECF
No. 153-11 at 39; ECF No. 162-8 at 4.) Even though the invoice indicates a $4,150
payment to Ford Wholesale, the invoice reflects that the payment included materials and
supplies other than shingles. (ECF No. 153-11 at 39.)
FOF 85: Boehm’s house has not suffered any property damage as a result of the allegedly
faulty Oakridge-brand shingles, and Boehm submits no evidence that he, to date, has
replaced the shingles on the roof of his house. (ECF No. 153-11 at 35; ECF No. 162-9 at
22-23.)
4. The Maag(IL) Case
FOF 86: The last of the four cases was filed in the United States District Court for the
Southern District of Illinois on March 25, 2014, on behalf of Edward and Diane Maag
(the “Maags” or the “Maag Case”) and a nationwide class of individuals “that have
owned, own, or acquired” structures “on which Owens Corning Oakridge shingles are or
have been installed since 1986.” (14-cv-826, ECF No. 2 ¶ 68.) Alternatively, the Maags
proposed certification of a class of individuals “that have owned, own, or acquired
[structures] physically located in the state of Illinois on which Owens Corning Oakridge
shingles are or have been installed since 1986.” (Id. ¶ 69.)
FOF 87: The Maags make the same allegations in their complaint as do the other named
plaintiffs about Owens Corning’s Oakridge-brand shingles being “plagued by design
27
samples of the shingles, which indicated that certain Owens Corning shingles were “40year warranty shingles.” (ECF No. 162-19 at 9-10.)
FOF 92: The cost of the Oakridge-brand shingles in 1999 was $3,719.40. (ECF No. 162-14
at 20.)
FOF 93: Mr. Maag stated that “a 40-year warranty seemed to me to be a warranty which
would compensate for defects for 40 years.” (ECF No. 162-14 at 27.) Mrs. Maag stated
that shingles with a 40-year warranty “were going to be the longest-life shingles” and
“would last 40 years, like they said it would.” (ECF No. 153-12 at (44, 102).)
FOF 94: Following a tornado in 2011, the Maags hired a contractor to replace a shingle
that had blown off their roof. (ECF No. 162-16 at 5-6, 11-12.) The contractor informed
the Maags that the shingles on their roof were cracked and that the roof would need to be
replaced in the next year or so. (Id.)
FOF 95:
In December 2012, a different contractor who the Maags hired to repair shingles
that had blown off the roof informed the Maags that pieces of shingles were coming off
the roof because the shingles were disintegrating. (ECF No. 153-12 at 43.)
FOF 96: The Maags’ home never suffered property damage as a result of shingle failure.
(ECF No. 162-16 at 18; ECF No. 162-14 at 24, 25.)
FOF 97: The Maags accepted a settlement payment from Owens Corning in 2013 of
$6,696.63. (ECF No. 162-15 at 2-3.) The Maags used the payment to purchase new
Owens Corning shingles to replace their roof, at a total cost of approximately $16,000.
(ECF No. 162-16 at 3, 14; ECF No. 162-14 at 25.)
29
FOF 98: By accepting the settlement payment, the Maags released Owens Corning from
any and all liability related to the complaint they made to Owens Corning’s Customer
Response Center about their shingles. (ECF No. 162-15 at 2-3.)
5. Summary of Named Plaintiffs’ Claims
FOF 99: Each named plaintiff contends that Owens Corning’s defective Oakridge-brand
shingles cause two kinds of harm: a) property damage, and b) removal and replacement
costs. FOF 19, 45, 69, 87.
FOF 100: Of the named plaintiffs, only the structures owned by Wright, West, and Gonzalez
suffered property damage. FOF 29, 38-40, 55, 58, 81-82, 85, 96.
a. Plaintiffs concur that the damage caused to Wright’s roof was not caused by
faulty Oakridge-brand shingles. FOF 29.
b. There is evidence that the damages caused to West’s and Gonzalez’s
structures were not caused by faulty Oakridge-brand shingles. FOF 38-40, 6062.
c. West’s and Gonzalez’s insurance carriers paid to repair the property damage
to their structures. FOF 40, 58.
FOF 101: Of the named plaintiffs, the record reflects that only West and the Maags replaced
the Oakridge-brand shingles on the roof of their structures. FOF 30-31, 41, 65, 85. 97.
a. West’s contractor replaced the back roof on his structure free-of-charge. FOF
41.
b. The Maags used a nearly $7,000 settlement payment that Owens Corning
made in connection with their warranty claim to purchase new Owens
Corning shingles to replace the shingles on their structure. FOF 97.
30
C.
Owens Corning’s Oakridge-Brand Shingles and Warranty Data
FOF 102: Oakridge-brand shingles are fiberglass mat shingles, which means that the mat
component of the shingle is made primarily of fiberglass. When the mat is made
primarily of paper or cloth, the shingle is referred to as an organic shingle. (ECF No. 165
¶ 11.)
FOF 103: The fiberglass mat is coated with an asphalt and limestone filler mix and then
colored ceramic granules are applied on top. (Id. ¶¶ 10-24.)
FOF 104: Each of the shingle components is supplied to Owens Corning by different
manufacturers from plants located throughout the country. (Id.) The characteristics of
each component will vary from plant to plant to account for climate differences, and raw
materials have changed in content and performance over the last two decades. (Id.)
FOF 105: At least 23 kinds of Oakridge-brand shingles were manufactured at 13 different
plants in the United States during the proposed 20-year class period. (ECF No. 151-5 at
8-9; ECF No. 165 ¶ 6.)
FOF 106: Each Owens Corning plant designs shingles to perform in the climate of the
location in which that plant is located, and correspondingly, where those shingles will
likely be installed. (ECF No. 165 ¶¶ 7-8).)
FOF 107: Owens Corning typically distributes shingles from particular manufacturing plants
to particular states. (ECF No. 165 ¶ 7.) For instance, the Owens Corning manufacturing
plant in Portland, Oregon, supplies shingles to customers located in “geo-zones” in the
Pacific Northwest, while the plant in Jacksonville, Florida, supplies shingles to customers
located in Florida and Latin America. (ECF No. 153-12 at 8-9; ECF No 165 ¶¶ 6-7.)
31
FOF 108: During the proposed 20-year class period, Owens Corning used more than 500
design specifications to manufacture Oakridge-brand shingles. (ECF No. 162-61 ¶ 34;
ECF No. 165 ¶ 43; ECF No. 165-1.)
a. Each specification provides a minimum, maximum, and target measurement for
square weight, a target and minimum measurement for asphalt weight and tear
strength, and a target measurement for mat weight. (ECF No. 162-61 ¶ 35; ECF
No. 165-1 at 2-15.)
b. An Owens Corning design specification for an Oakridge-brand shingle is,
therefore, not a list of exact measurements to which a shingle must be
manufactured, but, instead, is a collection of minimums, maximums, and target
measurements for various features of the shingle which frame the acceptable,
and preferable, parameters within which to manufacture a shingle at that
particular plant, at that particular time, using those raw materials that are
available at that location and time. (ECF No. 162-61 ¶ 35; ECF No. 165 ¶ 35;
ECF No. 165-1 at 2-15.)
c. How the various features of a shingle, such as its square weight, asphalt weight,
tear strength, and mat weight are coordinated with and related to each other
affects the quality of a shingle. (ECF No. 159 at 4, 5-6, 7, 21.) One
measurement, viewed in isolation, will not define the quality of a shingle. (Id.)
FOF 109: There is no dispute that all design specifications for Oakridge-brand shingles at
issue in this case meet the applicable industry standard, ASTM D3462. That standard is
“designed for the evaluation of products as manufactured.” (ECF No. 139 at 23; ECF No.
151-3 at 14-18.) The ASTM standard sets minimums for such measurements as tear
32
strength, net mass, mat mass, asphalt mass, and mineral matter mass for newlymanufactured shingles. (ECF No. 151-3 at 15-16.) Plaintiffs’ expert concedes that the
standard does not establish performance requirements for installed fiberglass shingles and
does not prescribe an expected service life for a shingle. (ECF No. 151-4 at 3 (¶ 11).)
FOF 110: Although other warranties could have been offered, and purchased, in connection
with Oakridge-brand shingles, a limited shingle warranty was automatically provided
with the installation of all Oakridge-brand shingles. (ECF No. 162-51 ¶¶ 4-5, 10-12 and
Table A.) During the proposed 20-year class period, the term of the limited shingle
warranty varied from 25 years, to a lifetime warranty, and provided for the replacement,
at a prorated cost, of shingles that contain manufacturing defects that would reduce the
usable life of the shingle or affect shingle performance. (Id.)
FOF 111: Owens Corning Oakridge-brand shingles were installed on more than six million
structures between 2000 and 2012. (ECF No. 166-7 at 30.)
FOF 112: Less than one half of one percent of Oakridge-brand shingle installations
produced a warranty claim between 2000 and 2012. (ECF No. 166-7 at 30; ECF No. 139
at 73.) Put another way, for every 1,000 structures on which Oakridge-brand shingles
were installed, fewer than 5 owners submitted a warranty claim to Owens Corning.
FOF 113: Data produced by Owens Corning in this litigation indicates that, between 1992
and 2012, nearly 30,000 warranty claims were made in connection with Oakridge-brand
shingles that were produced at manufacturing plants that typically distribute shingles for
installation in Pennsylvania, Illinois, California, and Texas. FOF 103-04; (ECF No. 15424 at 5; ECF No. 153-12 at 8-9; ECF No. 154-1 ¶ 44(c).)
33
FOF 114: Warranty claims are occasionally denied on the basis that the shingle submitted
with the claim was not manufactured by Owens Corning. (ECF No. 154-1 ¶ 44(d).)
FOF 115: Owens Corning administers warranty claims on a case-by-case basis. A product
quality specialist is assigned to each warranty claim that Owens Corning receives. (ECF
No. 162-51 ¶ 25.) The warranty claim review process may include reviewing
documentation and photographs, gathering information from the owner or third parties,
sending an inspector to the property, or testing the returned shingles. (Id. ¶ 26.) Claims
are sometimes paid solely to preserve customer goodwill. (Id. ¶ 29.)
FOF 116: Owens Corning does not maintain records that identify all owners who had
Oakridge-brand shingles installed on their structures from 1992 to 2012. (ECF No. 16251 ¶ 22.)
FOF 117: Owens Corning does not maintain records that indicate all owners who hold a
limited shingle warranty. (ECF No. 162-51 ¶ 22.)
D.
Owens Corning’s Representations about Oakridge-brand Shingles
FOF 118: The record reflects that purchasers receive information about Oakridge-brand
shingles while at retail locations, such as home improvement stores and lumber yards.
Specifically, purchasers obtain brochures, pamphlets, and data sheets, and view sample
boards and signage at retail outlets prior to selecting Oakridge-brand shingles. FOF 2325, 33, 35-36, 48, 72-75, 77, 90-91. Purchasers may also speak with the contractor who
will be installing the new shingles and sales associates at the retail location. FOF 49.
a. Plaintiffs submit no examples of product displays or sample boards made
available at the retail point of sale, which several named plaintiffs indicate
they relied upon in selecting Oakridge-brand shingles. FOF 48, 91.
34
b. Plaintiffs submit no examples of product packaging or inserts that
accompany Oakridge-brand shingles at the retail point of sale, or when the
product is delivered to the owner, or the owner’s contractor.
FOF 119: The named plaintiffs rely upon several different categories of evidence to support
their contention that Owens Corning, during the entire proposed 20-year class period,
“represented that Oakridge would last the duration of the warranty,” “encouraged
consumers to associate Oakridge’s warranty durations with Oakridge’s durability,” or
otherwise promised that Oakridge-brand shingles would have a useful life of at least 25
years, or for the same number of years as the length of the limited warranty. (ECF No.
178 at 10 (¶¶ 39, 41); ECF No. 162-51 at 6; ECF No. 152-1 to -12; ECF No. 153-2 to -8;
ECF No. 154-21 at 2.) These categories of evidence are: (1) product literature; (2)
internal Owens Corning documents and communications; (3) testimony of the named
plaintiffs; and (4) Owens Corning’s limited shingle warranty. The court will separately
consider the content and significance of each category of evidence in the findings of fact
that follow. The court will indicate when, instead of making findings about the evidence,
inferences are being drawn from the evidence, or conclusions are being reached about the
sufficiency of the record. None of the evidence relied upon by the named plaintiffs
supports a finding that Owens Corning promised that all Oakridge-brand shingles would
not experience cracking, degranulation, fragmentation, or deterioration for, or would have
a useful life of, at least 25 years.
35
1. Product Literature
a. The Group Exhibit of Brochures, Pamphlets, and Flyers
i. Findings of Historical Fact
FOF 120: The named plaintiffs submitted a group exhibit comprised of six Owens Corning
brochures, pamphlets, and flyers to support their assertion that Owens Corning promised
that all Oakridge-brand shingles would not experience cracking, degranulation,
fragmentation, or deterioration for, or would have a useful life of, the same number of
years as the length of the limited warranty. (ECF Nos. 152-1 to -12, 153-1 to -8.)
FOF 121: Although the record reflects the dates of these brochures, ECF No. 151-1 ¶ 27
(1995, 1999, 2002, 2004, 2007, and 2009), there is no other information about where the
brochures were distributed, in terms of both geographic market and kind of retail outlet,
how they were distributed, or for what period of time they were distributed.
FOF 122: None of the brochures include a statement that Oakridge-brand shingles will
experience no cracking, degranulation, fragmentation, or deterioration for the same
number of years as the length of the limited warranty.
FOF 123: None of the brochures include a statement that Oakridge-brand shingles will not
require repair or replacement for the same number of years as the length of the limited
warranty.
FOF 124: Although the materials refer to Oakridge-brand shingles using words such as
durable, worry-free, quality, premier, and exceptional, those words are not always used in
the same context, or to convey the same meaning. (ECF Nos. 152-1 at 3 (“enduring
quality”), 152-1 at 5 (“quality protection”), 151-2 at 2 (“top-quality construction”), 152-2
at 4 (“premium protection” and “premier shingle”), 152-4 at 3 (“outstanding
36
performance”), 152-4 at 5 (“worry-free performance”), 152-5 at 3 (“exceptional
performance”), 152-6 at 2 (“25 years of performance”), 152-7 at 4 and 153-6 at 2-3
(“lasting performance and protection”), 152-12 at 3 (“low-maintenance durability”), 1538 at 3 (“durable beauty”).)
FOF 125: The same materials equally tout, and in some examples emphasize, the aesthetic
appeal of Oakridge-brand shingles. (ECF Nos. 152-1 at 3 (“rugged appearance of added
dimension”)], 152-1 at 5 (“subtle look of dimension”), 152-2 at 2 (“dimension-like,
random look of wood”), 152-2 at 4 (“rugged wood-look beauty” “the greatest degree of
added dimension”), 152-4 at 3 (“add depth and dimension to your roof” “rich
dimensional look” “exceptional beauty”), 152-4 at 5 (“special double shadows add drama
and depth to your home”), 152-5 at 2 (“nature’s richest colors and deepest shadows”)
(“rich dimension”), 152-5 at 3 (“softly textured look”), 152-6 at 2, 4 (“subtle shadows
create interest” and “add[] depth and dimension”), 152-7 at 4 (“shadow lines create a
dramatic, three-dimensional effect”), 152-12 at 3 (“subtle shadows… offer increased curb
appeal”), 153-1 at 4 (“a more dimensional look”), 153-2 at 2 (“a dramatic, threedimensional look”), 153-8 at 2 (“a warm inviting look”).
FOF 126: Some brochures discuss the benefits of purchasing a multi-component Owens
Corning Roofing System, ECF Nos. 152-4 to -6, 152-7 to -11, 152-12, 153-1 to -6, 153-7
to -8, while others do not, ECF Nos. 152-1 to -3, see also ECF No. 162-5 at 9; ECF No.
162-5 at 12; ECF No. 162-8 at 6. Of the brochures that discuss the roofing system, some
explain that increased warranty coverage can be purchased if the roofing system is
installed, ECF Nos. 152-4, see also ECF Nos. 162-5 at 9, 162-8 at 6, while others do not,
ECF Nos. 152-7 to -11, 152-12, 153-1 to -6, 153-7 to -8.
37
FOF 127: In some materials the length of the limited shingle warranty is included in the
name of the Oakridge-brand product, e.g., Oakridge Pro 30, ECF No. 152-4 to -6, 152-7
to -11, 152-12, 153-1 to -6, see also ECF No. 162-5 at 12, while in other brochures the
name of the shingle includes no reference to the length of the limited shingle warranty,
e.g., Oakridge Shadow, ECF No. 152-1 to -3, 153-7 to -8, see also ECF No.162-5 at 9,
26-30; ECF No. 162-8 at 6.) Notably, while the majority of the brochures offered into
evidence by the named plaintiffs include the length of the limited shingle warranty in the
name of the product, e.g., ECF No. 152-1 to -12, 153-1 to -8, the majority of marketing
materials produced by Owens Corning do not include the length of the limited warranty
in the name of the product, e.g., ECF Nos. 162-5 at 9, 26-30, 162-8 at 6.
FOF 128: Each brochure includes a graphic referencing the limited shingle warranty of a set
number of years and includes information about the scope of coverage or directs the
reader to see the warranty for complete details. (ECF Nos. 152-1 at 5, 152-2 at 2 & 4,
152-4 at 4 & 6, 152-5 at 3 & 7, 152-6 at 2 & 4, 152-7 at 4, 152-12 at 3, 153-2 at 2, 153-3
at 3, 153-4 at 4, 153-5 at 5, 153-6 at 2 & 3.)
FOF 129: Phrases such as “worry-free” and “low-maintenance” are used in some, but not
all the brochures submitted into evidence by the named plaintiffs. See e.g., ECF Nos.
152-7 to -11, 153-7 to -8.
FOF 130: “Worry-free” appears in the materials as part of phrases such as “worry-free
coverage with a full 25-year warranty” and “40 years of beautiful, worry-free
performance.” (ECF Nos. 152-1 at 5, 152-4 at 5.)
38
FOF 131: The named plaintiffs identify two instances in the materials in which the phrase
“low-maintenance durability” is used. (ECF No. 152-12 at 3 and 153-5 at 5; ECF No. 178
at 10 (¶ 38).)
FOF 132: Some, but not all, of the materials include references to the amount and kind of
asphalt used. (ECF Nos. 155-1 to -3, 152-7 to -12.) None of the brochures, however,
include a statement that Oakridge-brand shingles will not require repair or replacement or
will last for the same number of years as the limited warranty because of the amount or
kind of asphalt used.
FOF 133: Plaintiffs identify only one brochure, which is dated 1999, that purportedly links
asphalt to the useful life of Oakridge-brand shingles. ECF No. 154-4 to -6. Three times
in the brochure Oakridge-brand shingles are referred to as having “[____] weatheringgrade asphalt for [__] years of [____] performance.” (Id.) For the shingles sold with a
25-year limited warranty, weathering-grade is modified by the word “quality” and the
word performance is not modified. (ECF No. 152-6 at 2.). For the shingles sold with a
30-year limited warranty, weathering-grade is modified by the word “extra” and the word
performance is modified by the word “exceptional.” (ECF No. 152-5 at 3.). For the
shingles sold with a 40-year limited warranty, weathering-grade is modified by “the
most” and the word performance is modified by the phrase “beautiful, worry-free.” (ECF
No. 152-4 at 5.)
39
ii. Inferences Drawn From, and Sufficiency of, the Evidence
FOF 134: The group exhibit of brochures, pamphlets, and flyers, in which certain words or
phrases are used in only some specimens, cannot be the basis for any reasonable
inference about the content and consistency of Owens Corning’s marketing activities,
between 1992 and 2012, in Pennsylvania, Illinois, California, and Texas.
FOF 135: The group exhibit of brochures, pamphlets, and flyers does not support any
reasonable inference that Owens Corning represented that Oakridge-brand shingles will
last for at least 25 years, or for the same number of years as the length of the limited
warranty.
FOF 136: Owens Corning does not use words such as “quality,” “premier,” “worry-free” or
“low-maintenance” consistently in the group exhibit of brochures, pamphlets, and flyers.
FOF 137: When words such as “worry-free,” “low-maintenance,” appear in the group
exhibit of brochures, pamphlets, and flyers they do not equate to a representation that all
Oakridge-brand shingles will have a useful life of at least 25 years, or for the same
number of years as the length of the limited warranty, as plaintiffs contend. (ECF No. 178
at 10 (¶ 38).)
a. In both examples identified by plaintiffs in which the phrase “worry-free” is used,
a connection is made between the absence of worry and Owens Corning’s limited
shingle warranty. A limited warranty, which according to Owens Corning’s
literature, provides for payment of replacement costs, and sometimes labor, for
faulty shingles, is something that would ease an owner’s worry or concern. Such
statements assure the owner that if the shingles need repair or replacement during
the warranty period, Owens Corning, and not the owner, will bear the cost.
40
Statements about the security provided by a warranty are not a promise that the
shingles will not need repair or replacement for the length of the warranty. That
interpretation is actually counterintuitive.
b. Regardless, even if statements about shingles being “worry-free” could arguably
be characterized as representations that the shingles will not crack, degranulate,
fragment, or deteriorate or will not need repair or replacement for the length of
the warranty, the record includes only these two isolated statements, made in
brochures dated prior to 2000, and distributed in unknown geographical locations
and through unspecified distribution methods. Plaintiffs identify no other use of
such phrases in the record, ECF No. 178 at 10 (¶ 38), and the court could locate
none in the exhibits filed in support of plaintiff’s motion for class certification.
c. Plaintiffs proffer no evidence about what the phrase “low-maintenance” means to
a purchaser of roofing shingles and provide no basis for this court to infer that
“low-maintenance durability” equates to a representation that an Oakridge-brand
shingle will last for at least 25 years, or for the same number of years as the
limited shingle warranty.
d. Like the “worry-free” statements identified by plaintiffs, these “low-maintenance”
statements are also isolated when viewed in the context of the breadth of the
proposed class definitions.
FOF 138: References in the group exhibit of brochures, pamphlets, and flyers to the amount
or quality of asphalt do not equate to a representation that all Oakridge-brand shingles
will have a useful life of at least 25 years, or for the same number of years as the length
of the limited warranty, as plaintiffs contend. (ECF No. 178 at 10 (¶ 40).)
41
a. In the single brochure relied upon by plaintiffs, Owens Corning does not make
any uniform or consistent representation about the effect that asphalt has on the
useful life of a shingle. Although the shingles sold with the 30- and 40-year
limited warranties reflect an increase in the amount of asphalt being used (from
“extra” to “the most”), the performance is described as going from “exceptional”
to “beautiful, worry-free,” not from exceptional to the best, or superior,
performance. As such, the brochure does not suggest a direct correlation between
the amount of asphalt used, and the level of performance.
b. Some brochures indicate that although different amounts of asphalt are used, the
performance stays the same. (ECF Nos. 153-5 at 5 (“durable weathering-grade
asphalt shingles bonded together with tough Fiberglas mat core for lasting
performance and protection,” ECF No. 153-6 at 2 (“extra weathering-grade
asphalt shingles bonded together with tough Fiberglas mat core for lasting
performance and protection,” 153-6 at 3 (“premium weathering-grade asphalt
shingles bonded together with tough Fiberglas mat core for lasting performance
and protection”) (emphasis added in each).)
c. As with other words and phrases identified by plaintiffs in Owens Corning’s
marketing materials, the discussion of asphalt on which plaintiffs rely appears in
one brochure, dated prior to 2000, which is an isolated example.
FOF 139: Plaintiffs offer no legal authority to support their crucial contention that phrases
such as “exceptional,” “worry-free,” “low-maintenance,: and “lasting” qualify as
actionable representations even though this legal contention is not without dispute. See
Cheatham v. ADT Corp., No. 15-2137, 2016 WL 540832, at *9 (D. Ariz. Feb. 11, 2016)
42
(claim that a security system provides “worry-free living” is puffery); Gold v. Lumber
Liquidators, Inc., No. 14-5373, 2015 WL 7888906, at *6-7 (N.D. Cal. Nov. 30, 2015)
(statements about “high quality” or “high performance” are non-actionable puffery under
California consumer protection statutes); Peruto v. TimberTech Ltd., No. 15-2166, 2015
WL 5089484, at *2, 7 & n.10 (D.N.J. Aug. 26, 2015) (finding phrase “designed to
provide years of low-maintenance use and enjoyment” to be “puffery as a matter of
law”); Elias v. Hewlett-Packard Co., 950 F.Supp.2d 1123, 1131-34 (N.D. Cal. 2013)
(statements about computer being “ultra-reliable” or “higher performance” were nonactionable puffery under California consumer protection statutes); Rochester Laborers
Pension Fund v. Monsanto Co., 883 F.Supp.2d 835, 882 (E.D. Mo. 2012) (statement
about “exceptional performance” constituted inactionable puffery under federal securities
laws); ConsulNet Computing, Inc. v. Moore, No. 04-3485, 2007 WL 2702446, at *10
(E.D. Pa. Sept. 12, 2007) (promise that a website will “alleviate ‘worry’” is “mere
puffery”); Hercules Machinery Corp. v. McElwee Bros., Inc., No. 01-3651, 2002 WL
31015598, at *5 (E.D. La. Sept. 9, 2002) (“top quality” is a “classic example of puffery”).
b. The Named Plaintiff’s Exhibits
i. Findings of Historical Facts
FOF 140: Three of the named plaintiffs produced copies of brochures or flyers that they saw
before deciding to purchase Oakridge-brand shingles in support of their claims. Plaintiffs
attached two of these items to their motion papers. (ECF Nos. 153-12 at 14-18 (Wright),
24 (West).) The court could only locate the last item in the papers filed by Owens
Corning. (ECF No. 162-8 at 6-7 (Boehm).) The three specimens are dated 1997, 1998,
and 2005.
43
FOF 141: Boehm’s brochure, which is from approximately 1997, stresses the architectural
style of the shingles, and states that they “feature the rugged look of wood, premium
protection and enduring value,” and offer maximum protection from the elements due to
“the most weathering-grade asphalt available and our tough Fiberglass mat construction.”
(ECF No. 162-8 at 6.) The brochure includes a graphic depicting a “40 year limited
product warranty.” (Id.) The text of the brochure states that the “40-year warranty*
covers the prorated replacement cost of new shingles and labor” and instructs the reader
to “see actual warranty for details.” (Id.)
FOF 142: Wright’s brochure, which is dated 1998, differs from Boehm’s brochure in that it
emphasizes the benefits of purchasing an entire Owens Corning Roofing System, which
is comprised of six parts, only one of which is the shingles. (ECF No. 153-12 at 14-18.)
With respect to the shingle component of the roofing system, the brochure includes
statements about premium protection, enduring value, the most weathering-grade asphalt,
and a tough fiberglass mat. (Id. at 17.) The brochure includes a graphic depicting a “40
year limited product warranty.” (Id.) The brochure states that the “40-year warranty*
covers the prorated replacement cost of new shingles and labor” and instructs the reader
to “see actual warranty for details.” (Id.) Wright’s brochure, however, unlike Boehm’s
brochure, offers a detailed comparison of the warranty periods, both prorated and nonprorated, for the limited shingle warranty as compared to the enhanced roofing system
warranty, and explains that the latter warranty must be separately purchased. (Id. at 16.)
FOF 143: West’s 2005 flyer, unlike Boehm’s and Wright’s brochures, is only one page. It
states that Oakridge-brand shingles offer increased curb appeal, low-maintenance
durability, and a 30-year limited warranty*. (ECF No. 153-12 at 24.) There are no
44
statements about asphalt, the fiberglass mat, premium protection, or enduring value. The
focus of the flyer is almost entirely on the aesthetic appeal of the shingles.
FOF 144: The promotional displays viewed by Gonzalez and the Maags at the retail outlet,
and the brochure given to Gonzalez are not available for the court’s review. FOF 48, 91.
ii. Inferences Drawn from, and Sufficiency of, the Evidence
FOF 145: Rather than being probative of Owens Corning’s allegedly uniform
representations that all Oakridge-brand shingles will have a useful life of at least 25
years, or for the same number of years as the length of the limited warranty, the named
plaintiffs’ materials include no such statements and demonstrate a general lack of
consistency in Owens Corning’s product literature.
FOF 146: There is no basis, based upon the record before the court, to infer that the
materials seen by Gonzalez and the Maags would have included any uniform
representations about the useful life of Oakridge-brand shingles. As will be discussed in
the findings of fact that follow, the only reasonable inference supported by the record is
that these items likely included references to a limited shingle warranty of a set number
of years. FOF 128.
2. Owens Corning’s Internal Documents and Communications
a. Findings of Historical Fact
FOF 147: The named plaintiffs submit various internal Owens Corning documents and
communications to support their assertion that Owens Corning promised that all
Oakridge-brand shingles would have a useful life of at least 25 years, or for the same
number of years as the length of the limited warranty. (ECF Nos. 153-9 at 3, 154-20, 154-
45
21 at 2, 154-22 at 2, 154-23 at 2, 165-2 at 5; ECF No. 178 at 9 (¶ 35) and 10 (¶¶ 39, 4143).)
FOF 148: The “Residential Roofing Technical Training Manual,” is more than 200-pages
long, is dated 1995, ECF No. 165-2 at 5, and was used in 1998, ECF No. 165 at 11.
a. Plaintiffs cite to a single page of this 200-page document. (ECF No. 178 at 10 (¶
41) (citing ECF No. 154-4 at 5); ECF No. 165-2 to -4.)
b. The page provides information about shingle warranties in general, including that
they always cover manufacturing defects only and that even without
manufacturing defects shingles eventually “wear out by losing granules to the
point of looking bad,” but is not specific to Oakridge-brand shingles. (ECF No.
165-2 at 60.)
FOF 149: The presumed television commercial script is undated and proffered without any
explanation about when it was drafted and when, if ever, it was created and broadcast.
(ECF No. 178 at 10 (¶ 39)); (ECF No. 154-21 at 2.)
a. It appears that the script submitted into evidence is not a complete copy as the
dialogue seems to end abruptly, which discounts its evidentiary significance.
(ECF No. 154-21 at 2.)
b. Plaintiffs truncate the passage they quote from the script. The actual statement in
the script is that “[e]qually important [as aesthetics], with a 30-year Limited
Warranty and a 70-miles-per-hour Wind Resistance Limited Warranty, I know
they’re durable.” (ECF No. 154-21 at 2.) Plaintiffs delete the reference to the
wind resistance warranty in their proposed findings. (ECF No. 178 at 10 (¶ 39).)
46
FOF 150: “The Architect’s Roofing Answer Book” is dated 1996. Plaintiffs proffer no
evidence that the book was used at any other time during the proposed 20-year class
period. (ECF No. 178 at 10 (¶ 39)); (ECF No. 153-9 at 3-5.)
a. The answer book discusses Owens Corning’s “full complement of roofing
products” and nowhere refers specifically to Oakridge-brand shingles.
b. The book does not appear on its face to have been created for, or distributed to,
the end purchaser of roofing shingles. It instead is directed at architects and
builders, seemingly to encourage them to recommend, or use, Owens Corning’s
shingles on their projects. Plaintiffs offer no evidence to show the contrary.
c. The statement that plaintiffs excise from the book, and rely upon in their proposed
findings of fact and conclusions of law, is taken out of context. Plaintiffs assert
that the book states: “[T]he longer the warranty, the better the shingle.” (ECF No.
178 at 10 (¶ 39).) In the book, that statement is preceded by the word “Generally”
and followed by an explanation that the warranty is prorated for both replacement
costs and labor.
FOF 151: The 2006 Owens Corning Consumer Roofing Market Structure/Segmentation
Study - upon which plaintiffs rely - reflects that warranties are not synonymous with
durability and does not prove that purchasers equate durability with warranties when
selecting shingles. (ECF No. 178 at 9 (¶ 35); ECF No. 154-20.) The study lists durability
and warranty as the top two concerns when selecting roofing materials. (ECF No. 154-20
at 4.)
47
FOF 152: The August 23 and 24, 2006 email chain indicates that “Rob” [Daenen] and Bert
Elliott “put together a first pass” at answering certain questions raised during two
presentations about residential roofing products. (ECF No. 154-22 at 2.)
a. Plaintiffs provide no information about who attended those presentations and
asked the identified questions, or under what circumstances “an Owens Corning
employee” suggested these answers. (ECF No. 178 at 10 (¶ 42).) There is no
indication that the suggested answers are Owens Corning’s “final” answers, or
that the suggestions are made with any authority to speak on behalf of Owens
Corning.
b. In any event, the suggested answers merely explain that some purchasers select an
Oakridge-brand shingle with a 50-year warranty because it is a higher number
than the 30-year warranty and that some contractors push a 50-year warranty in
higher-end neighborhoods. (ECF No. 154-22 at 3.)
c. The suggested answers also discuss the terms and benefits of Owens Corning’s
new lifetime warranty product. (Id.)
FOF 153: In a second email chain, which is dated January 26, 2011, plaintiffs contend that
Owens Corning’s “quality director” agrees with another employee’s statement that
“warranty enhancements are always due to following our competition.” (ECF No. 178 at
10 (¶ 43).)
a. As an initial matter, the employee makes various statements to which the alleged
director responds, “You got it.” (ECF No. 154-23 at 2.) It is impossible, without
further explanation, to determine with which statement, or statements, the director
is agreeing.
48
b. In any event, even accepting plaintiffs’ characterization of the email as correct, a
statement that Owens Corning changes its warranty terms in order to match what
competitors offer is innocuous and is not probative that Owens Corning promised
that all Oakridge-brand shingles would have a useful life of at least 25 years, or
for the same number of years as the length of the limited warranty.
b. Inferences Drawn from, and Sufficiency of, the Evidence
FOF 154: The single page from the training manual cannot support a reasonable inference
that Owens Corning represented that all Oakridge-brand shingles would have a useful life
of at least 25 years, or for the same number of years as the length of the limited warranty,
because it is not specific to Oakridge-brand shingles and only discusses shingle
warranties generally. FOF 148.
FOF 155: The television commercial script cannot support a reasonable inference that
Owens Corning represented that all Oakridge-brand shingles will have a useful life of at
least 25 years, or for the same number of years as the length of the limited warranty,
because it is undated and incomplete and submitted without any information about where,
if ever, it was broadcast, which are significant evidentiary deficiencies given that the
proposed class definition includes owners located in four states over a period of twenty
years. FOF 149.
FOF 156: The television commercial script cannot support a reasonable inference that
Owens Corning represented that all Oakridge-brand shingles will have a useful life of at
least 25 years, or for the same number of years as the length of the limited warranty,
because the script does not, as plaintiffs imply, associate the limited shingle warranty,
standing alone, with the durability of Oakridge-brand shingles. FOF 149.
49
FOF 157: The architect’s answer book cannot support a reasonable inference that Owens
Corning represented that all Oakridge-brand shingles will have a useful life of at least 25
years, or for the same number of years as the length of the limited warranty, because it is
not addressed specifically to Oakridge-brand shingles and is not directed to the ultimate
purchaser of shingles. FOF 150.
FOF 158: The market study cannot support a reasonable inference that Owens Corning
represented that all Oakridge-brand shingles will have a useful life of at least 25 years, or
for the same number of years as the length of the limited warranty, because the study
identifies durability and warranty as two separate features. In other words, the study
demonstrates that consumers want materials to last long (durability), and, if for some
reason they do not, they want to be compensated for any repairs or replacement
(warranties). Instead of proving that the length of a warranty and the useful life of a
product are equivalent, as plaintiffs contend, the study is evidence that the two
characteristics were separate, albeit related, features. FOF 151.
FOF 159: The two email chains cannot support a reasonable inference that Owens Corning
represented that all Oakridge-brand shingles will have a useful life of at least 25 years, or
for the same number of years as the length of the limited warranty, because they are
proffered without further explanation and contain no statements that are probative of
promises that Owens Corning made to purchasers about the useful life of Oakridge-brand
shingles. FOF 152-53.
50
3. The Named Plaintiffs’ Testimony
a. Findings of Historical Fact
FOF 160: Plaintiffs cite to the deposition testimony of the named plaintiffs in support of
their contention that “consumers consistently believed that Oakridge shingles would last
the duration of the warranty.” (ECF No. 178 at 11 (¶ 44).)
FOF 161: Only four of the six named plaintiffs, however, testified that they believed that
Oakridge-brand shingles should last for the same number of years as the length of the
limited shingle warranty. (ECF No. 153-11 at 13 (West – shingles should last 30 years “if
that’s what they’re warranted”); ECF No. 153-11 at 17 (Gonzalez – shingle “should last
20 years” because “[i]t says it’s a 20-year warranty”); ECF No. 153-11 at 31-33 (Boehm
– “my 40-year roof that I think should probably last 40 years”)); ECF No. 153-12 at 36
(Mrs. Maag - shingles with a 40-year warranty would “last 40 years, like they said it
would”).)
FOF 162: Wright testified that she thought that a “40-year shingle roof... was supposed to
have been a good roof.” (ECF No. 153-10 at 5.)
FOF 163: Mr. Maag testified that the 40-year warranty on the Oakridge-brand shingles
“seemed to me to be a warranty which would compensate for defects for 40 years.” (ECF
No. 153-12 at 50-51.) Although plaintiffs do not cite to this testimony, Mrs. Maag
testified that shingles with a 40-year warranty would “last 40 years, like they said it
would.” (ECF No. 153-12 at 36; see ECF No. 178 at 11 (¶ 45).)
51
b. Inferences Drawn from, and Sufficiency of, the Evidence
FOF 164: Even the testimony of the six named plaintiffs does not reflect that purchasers
consistently equated the length of the limited shingle warranty with the useful life of
Oakridge-brand shingles. FOF 161-63.
FOF 165: Mrs. Maag’s testimony demonstrates that two individuals who own the same
structure on which Oakridge-brand shingles are installed can have different
understandings of what Owens Corning’s limited shingle warranty provides or promises.
FOF 163.
4. The Limited Warranty on Oakridge-brand Shingles
a. Findings of Historical Fact
FOF 166: A limited warranty is automatically provided with the installation of Oakridgebrand shingles. FOF 25, 110.
FOF 167: The record reflects that the shingle warranty is always modified by the word
“limited,” and is almost always followed by either an explanation of the scope of
coverage or an asterisk that directs the reader to see the actual warranty for complete
details and limitations. FOF 25, 36, 74, 128, 141-43.
FOF 168: Contrary to plaintiffs’ assertion, Owens Corning did not always incorporate the
length of the limited warranty into the name of the product throughout the proposed 20year class period. FOF 127.
FOF 169: The record reflects that although a limited shingle warranty was always provided
in connection with Oakridge-brand shingles, the same warranty does not apply to every
Oakridge-brand shingle manufactured during the proposed 20-year class period.
52
a. Owens Corning offered at least nine different warranties between 1992 and 2012.
(ECF No. 162-51 ¶ 4.)
b. The limited shingle warranty is only one kind of warranty that Owens Corning
offered. (Id.)
c. Other warranties include the Platinum Promise Limited Warranty, the
TruProtection Preferred Limited Warranty, and the System Advantage Roofing
Limited Warranty. (Id.)
i. The prerequisites to coverage, and scope of coverage, differ with respect
to each of these warranties. (Id. ¶¶ 4-9.)
ii. Some enhanced warranties require payment of an additional fee to obtain
coverage. (Id. ¶¶ 4-17; ECF No. 162-39 at 4-5.)
d. The limited shingle warranties offered by Owens Corning between 1992 and 2012
differ in terms of, for example, length of coverage, transferability, coverage for
labor and tear-off, coverage for installation defects, proration, and coverage for
conditions such as algae and wind. (ECF No. 162-39, 162-51 to -60; ECF No.
162-51 ¶¶ 11-19.)
i. By way of example, during the first seven years of the class period,
warranties on Oakridge-brand shingles were nontransferrable. (ECF No.
162-51 ¶ 19.)
ii. After 1999, however, warranties on Oakridge-brand shingles could be
transferred one time, but only if a small payment was made and a warranty
transfer card was submitted to Owens Corning. (Id. ¶¶ 19-22.)
53
iii. During the proposed 20-year class period, the length of the limited shingle
warranties varied from 25, 30, 40, and 50 years. (ECF No. 162-51 at 5-6 (¶
10 and Table A).)
iv. In 2011, Owens Corning instituted a lifetime warranty on all Oakridgebrand shingles. (Id. at 6 (Table A); ECF No. 154-22 at 3.)
e. The limited shingle warranties offered between 1992 and 2012 contain explicit
limitations on the availability of alternative legal remedies, implied warranties,
and consequential damages. These provisions are not identical: for example,
some limited warranties purport to entirely exclude implied warranties, ECF No.
162-53 at 37, 40, while others purport to only limit the duration of any implied
warranties, ECF No. 162-53 at 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29.
Some limited shingle warranties indicate that their terms can be changed without
notice, ECF No. 162-53 at 31. All limited shingle warranties acknowledge that
limitations may be prohibited by law in some states, e.g., ECF No. 162-53 at 3,
19, 29, 31, 40.
b. Inferences Drawn From, and Sufficiency of, the Evidence
FOF 170: Plaintiffs’ fundamental proposition that the limited shingle warranties qualify as a
representation that all Oakridge-brand shingles will have a useful life of at least 25 years,
or will last for the same number of years as the length of the limited warranty is
contradicted by the record because the warranties are explicitly described as being
limited, which denotes restrictiveness, not comprehensiveness. FOF 167.
FOF 171: Plaintiffs’ fundamental proposition that the limited shingle warranties qualify as a
representation that all Oakridge-brand shingles will have a useful life of at least 25 years,
54
or will last for the same number of years as the length of the limited warranty is
contradicted by the record because the length of the warranty is not always incorporated
into the name of the Oakridge product. FOF 168.
FOF 172: Plaintiffs’ fundamental proposition that the limited shingle warranties qualify as a
representation that all Oakridge-brand shingles will have a useful life of at least 25 years,
or will last for the same number of years as the length of the limited warranty is
contradicted by the record because the terms, conditions, and limitations of the limited
warranties were not the same throughout the proposed class period. FOF 169.
FOF 173: Plaintiffs offer no legal authority to support their crucial contention that Owens
Corning’s limited shingle warranty qualifies as a representation that all Oakridge-brand
shingles will have a useful life of the same number of years as the length of the limited
warranty, even though this legal contention is not without dispute. In re IKO Roofing
Shingles Prod. Liab. Litig., No. 2:09–md–2104 (C.D. Ill. Jan. 28, 2014) (ECF No. 338 at
35-36 (citing 1/28/2014, 4/12/2013, and 4/15/2013 decisions)); see Brooks v. GAF
Materials Corp., 301 F.R.D. 229, 233 (D.S.C. 2014) (recognizing, but finding
inapplicable, the IKO MDL court’s findings); In re HardiePlank Fiber Cement Siding
Litig., 12-md-2359, 2014 WL 2987657, at *3 (D. Minn. June 30, 2014) (“An
advertisement's reference to a formal limited warranty does not, on its own, create a new
informal promise that the product will last for a certain amount of time.”); Rasmussen v.
Apple Inc., 27 F. Supp. 3d 1027, 1035 (N.D. Cal. 2014) (the purpose of a warranty is to
contractually mark the point in time when the risk of paying for repairs shifts from the
manufacturer to the consumer); Keegan v. Am. Honda Motor Co. Inc., 838 F. Supp. 2d
929, 940-41 (C.D. Cal. 2012) (same); Sears, Roebuck and Co. v. Tyco Fire Products LP,
55
833 F. Supp. 2d 892, 899 (N.D. Ill. 2011) (stating that under Texas law, an explicit
statement must be made about the useful life of a product in order to assert warranty
coverage on that basis).
FOF 174: Under all the circumstances and based upon this record, the limited warranties
referenced in Owens Corning’s product literature cannot be reasonably characterized as
affirmative representations that Oakridge-brand shingles will have a useful life of at least
25 years, or for the same number of years as the length of the limited shingle warranty.
5. Ultimate Factual Findings: Representations about Oakridge-brand Shingles
FOF 175: The record does not support a factual finding that Owens Corning “represented
that Oakridge would last the duration of the warranty,” “encouraged consumers to
associate Oakridge’s warranty durations with Oakridge’s durability,” or otherwise
promised that Oakridge-brand shingles would have a useful life of at least 25 years, or for
the same number of years as the length of the limited warranty, as plaintiffs contend.
FOF 176: The record does not support a factual finding that Owens Corning represented that
Oakridge-brand shingles would not need to be repaired or replaced for at least 25 years,
or for the same number of years as the length of the limited warranty.
FOF 177: The record does not support a factual finding that Owens Corning represented that
Oakridge-brand shingles would not experience any cracking, degranulation,
fragmentation, or deterioration for at least 25 years, or for the same number of years as
the length of the limited warranty.
FOF 178: The record does not support a factual finding that Owens Corning represented that
Oakridge-brand shingles would not experience excessive cracking, degranulation,
56
fragmentation, or deterioration, beyond ordinary wear and tear, for at least 25 years, or
for the same number of years as the length of the limited warranty.
FOF 179: The record does not support a factual finding that Owens Corning represented that
a roof on which Oakridge-brand shingles are installed would not leak for at least 25
years, or for the same number of years as the length of the limited warranty.
FOF 180: The record does not support a factual finding that Owens Corning represented that
a roof on which Oakridge-brand shingles are installed would not cause property damage
for at least 25 years, or for the same number of years as the length of the limited
warranty.
E. Design Defect
1. Findings of Historical Fact
FOF 181: According to plaintiffs, all Oakridge-brand shingles manufactured “from 1992 to
2012 (except those manufactured in Atlanta and Memphis) are defectively designed
because Owens Corning’s specifications allow Oakridge to be manufactured at or near
minimums for asphalt mass and tear strength” making them “susceptible” and
“vulnerable” to cracking, premature deterioration, and failure. (ECF No. 178 at 3 (¶¶ 6, 89).)
a. The Expert Opinion of Dean Rutila
FOF 182: Plaintiffs retained Simpson Gumpertz & Heger, an engineering firm, to
investigate alleged problems with Owens Corning’s Oakridge-brand shingles and to
provide expert opinions about the causes of the alleged failures of those shingles. (ECF
No. 151-5 at 5.)
57
FOF 183: Mr. Dean Rutila (“Rutila”), a Senior Principal with Simpson Gumpertz & Heger
and a civil engineer, was the individual responsible for the investigation. (ECF No. 151-5
at 4.)
FOF 184: Rutila “performed detailed laboratory documentation and testing on 298 shingle
samples” that were returned to Owens Corning with a warranty claim. (ECF No. 151-5 at
6, 12, 19; ECF No. 139 at 38.) These shingle samples have been referred to as the
“warranty shingles” in these proceedings.
FOF 185: It appears that Rutila also visually inspected approximately 700 shingle samples
that were returned to Owens Corning with a warranty claim, but did not conduct
laboratory testing on them. (ECF No. 151-5 at 6, 12; ECF No. 139 at 100.)
FOF 186: The warranty shingles were removed from structures after installation and
weathering and sent to Owens Corning in connection with a claim being made pursuant
to one of Owens Corning’s warranty programs. FOF 184. After processing and
administration, the warranty shingles were stored by Owens Corning in a warehouse in
Toledo, Ohio, for some time before Rutila studied them. (ECF No. 151-5 at 6.)
FOF 187: Rutila tested the warranty shingles to determine whether they met the standards
set forth in ASTM D3462, using the ASTM test methods listed in D3462, all of which are
designed to evaluate newly-manufactured products. (ECF No. 151-5 at 19-20; ECF No.
151-3 at 15-17 (referring to ASTM Test Methods D1922 and D228).)
FOF 188: Although Rutila claimed that he also tested a limited number of unused shingles
obtained from some of the named plaintiffs, ECF No. 151-5 at 19, that data was not
included in Rutila’s expert report, ECF No. 126 at 49.
58
FOF 189: Rutila reviewed Owens Corning’s warranties and warranty claim documents,
industry standards and literature, and tested the warranty shingles in order to determine
the causes of the alleged problems with Oakridge-brand shingles. (ECF No. 151-5 at 5,
9.) Rutila opined that: (1) the shingles have excessive granule loss because they are
designed to meet the minimum mass of asphalt, which is inadequate to retain the mass of
granules applied; and (2) the shingles are designed at minimum tear strengths, with
fiberglass mats designed to meet the minimum standard for mat mass, which results in
cracking. (Id. at 6-7.)
i. Owens Corning’s Daubert Motion
FOF 190: Owens Corning filed a motion challenging the admissibility of Rutila’s opinions.
(ECF No. 92.) This court has a duty to evaluate and weigh expert testimony at the class
certification stage. Marcus v. BMW of N. Am., LLC, 687 F.3d 583, 602 (3d Cir. 2012).
FOF 191: At the Daubert hearing held on Owens Corning’s motion, the court expressed
various general concerns with the admissibility of Rutila’s testimony, including that his
testing appeared to suffer from an inherent selection bias, in that all the shingles Rutila
tested were returned to Owens Corning in connection with a warranty claim, which
indicates some level of dissatisfaction with the product, and appeared to be statistically
insignificant, in that Rutila tested fewer than 300 shingles out of the millions of shingles
manufactured during the proposed class period. (ECF No. 126 at 42-43.) The court
continued the Daubert hearing, and ordered the parties to submit additional briefing about
these, and other, issues. (Id. at 48-49; 1/21/15 Minute Entry.) Plaintiffs were instructed
to produce Rutila to be questioned at the continued Daubert hearing.
59
FOF 192: At the continued Daubert hearing, plaintiffs conceded that Rutila’s testing would
become irrelevant if Owens Corning stipulated that it manufactured all Oakridge-brand
shingles within the parameters of the written design specifications it produced during
discovery in this litigation. (ECF No. 139 at 6, 12-27.) Owens Corning agreed to this
stipulation. (Id.) The court later clarified that Owens Corning did not stipulate, and the
record would not support a finding, that all Oakridge-brand shingles were manufactured
at the minimum measurements set forth in Owens Corning’s design specifications, but
only that all shingles were manufactured somewhere within the parameters established by
the various minimum, maximum, and target measurements set forth in Owens Corning’s
design specifications. FOF 108; (ECF No. 139 at 63.)
FOF 193: After Owens Corning stipulated that it manufactured all Oakridge-brand shingles
during the proposed 20-year class period within the parameters established by the various
minimum, maximum, and target measurements set forth in its design specifications, the
court ruled that Rutila could not rely upon the testing he conducted on the warranty
shingles for any reason. (ECF No. 139 at 6, 25-27; ECF No. 159 at 3.) This ruling
dispensed with the need for Rutila to demonstrate that his testing of the warranty shingles
was statistically significant, unbiased, or otherwise reliable.
a. The court did not hear testimony or argument, and made no rulings, regarding the
statistical significance or reliability of Rutila’s testing of the warranty shingles.
The court described Rutila’s testing of the warranty as irrelevant and
inadmissible. (ECF No. 139 at 26, 27; ECF No. 159 at 3.) For this reason, the
Supreme Court’s recent decision in Tyson Foods, Inc. v. Bouaphakeo, No. 14-
60
1146, 2016 WL 1092414 (S.Ct. Mar. 22, 2016), discussing the admissibility of
representative evidence in class and collective actions cases is inapposite.
b. Rutila admitted at the continued Daubert hearing that his testing of the warranty
shingles suffered from selection bias. (ECF No. 139 at 104.)
c. The court specifically explained that Rutila’s testing of the warranty shingles
could only become relevant if Owens Corning were to attack his credibility on the
ground that he had never seen an Oakridge-brand shingle, in which case he could
refer to his testing to prove otherwise. (ECF No. 139 at 24-27; ECF No. 159 at 3.)
d. Even in the rebuttal situation described above, however, the court explained that
the validity and statistical significance of Rutila’s testing would not be relevant.
(ECF No. 139 at 24-25, 95.) The testing would only demonstrate Rutila’s general
familiarity with Oakridge-brand shingles.
e. Under these circumstances, the court finds plaintiffs’ suggestion in their proposed
findings of fact and conclusions of law, ECF No. 178 at 4 n.1, that Owens
Corning has opened the door to the admission of Rutila’s testing by arguing that
Rutila cannot prove that all Oakridge-brand shingles suffer from a common
design defect to be wholly inappropriate.
i. The court unambiguously explained that Rutila’s testing of the warranty
shingles was inadmissible for this exact purpose. (ECF No. 139 at 24-27,
76-78, 80, 109.)
ii. The court explicitly held that Rutila would not be permitted to testify,
based upon his testing of the warranty shingles, that all (or even most or
many) Oakridge-brand shingles were manufactured at the allegedly-
61
defective minimum design specification measurements. (ECF No. 139 at
110.)
iii. Following the court’s Daubert ruling, plaintiffs were tasked with proving
their design defect claim by way of evidence other than Rutila’s testing of
the warranty shingles. Owens Corning’s contention that plaintiffs failed to
meet that burden does not “open the door” to the admission of Rutila’s
testing to prove plaintiffs’ design defect claim. Plaintiffs’ argument is
both circular and specifically prohibited by this court’s Daubert ruling.
ii. Rutila’s Post-Daubert Opinion
FOF 194: Rutila now contends that because Owens Corning’s design specifications allow or
permit shingles to be manufactured “at or near” industry minimums for asphalt mass, mat
mass, and tear strength, all Oakridge-brand shingles are “susceptible” and “vulnerable” to
premature failure. (Id. (¶¶ 6, 7, 9).) According to Rutila, a design specification that
allows some Oakridge-brand shingles to be manufactured at higher measurements that
will admittedly produce a longer-lasting shingle (“high-end shingles”), and some to be
manufactured at lower measurements that will produce a shingle that will not last more
than 20 years (“low-end shingles”) is a defective design specification. (ECF No. 139 at
31, 82; see ECF No. 173 at 57-58 (describing an owner’s possession of “high-end
shingles” as winning the “shingle lottery”).)
a. Rutila testified that not all Oakridge-brand shingles will be manufactured at the
low-end of Owens Corning’s design specifications. (ECF No. 139 at 81.)
b. Rutila testified that he cannot determine whether a particular shingle was
manufactured at the “high-end” or the “low-end” of a design specification
62
without examining the roof, or a shingle removed from the roof. (ECF No. 139
at 32-33.)
c. Rutila offers no opinion about where, between the lower and the higher
measurements set forth in Owens Corning’s allegedly defective design
specifications, a shingle crosses the line and becomes defective.
i. Although Rutila testified at the continued Daubert hearing that a
nondefectively designed shingle would have an asphalt mass of 20
to 23 pounds per shingle square, a net weight of 215 to 230 pounds
per shingle square, and a mat mass right around 1.5 pounds per 100
square foot, ECF No. 139 at 41-42, the only measurement that
Rutila could substantiate was the 1.5 pounds per 100 square foot
mat mass, ECF No. 139 at 46-47, ECF No. 159 at 4, 21. The court
ruled that Rutila could testify only about that measurement. (ECF
No. 159 at 21-22.)
ii. The court ruled that Rutila would be permitted to testify, as a
general proposition, but without reference to specific
measurements (other than mat mass), that the inter-relationship and
coordination between asphalt mass, net weight, and mat mass
affects the performance and reliability of a shingle. (ECF No. 159
at 4, 5-6, 7, 21.)
d. Rutila testified that even if a shingle is manufactured in accordance with what
he considered to be a defective design specification, he cannot testify about how
long any particular shingle will last or whether any particular shingle will
63
experience excessive granule loss or cracking during the term of a warranty.
(ECF No. 139 at 33.)
e. Rutila testified that some shingles manufactured within the parameters of
Owens Corning’s design specifications will last more than 20 years. (ECF No.
139 at 33, 93.)
f. Rutila testified that a shingle manufactured at the higher measurements set forth
in Owens Corning’s design specifications will last 30 years. (ECF No. 139 at
33-34; 162-3 at 14.)
g. Rutila testified that an Oakridge-brand shingle manufactured at the low-end of
Owens Corning’s design specifications will not last more than 20 years. (ECF
No. 139 at 76-77, 81, 91; ECF No. 159 at 3.)
FOF 195: Although the court ruled that Rutila would be permitted to testify that low-end
Oakridge-brand shingles would not last more than 20 years, plaintiffs offer no evidence
to establish what proportion of Oakridge-brand shingles were manufactured as low-end
shingles during the proposed 20-year class period. (ECF No. 139 at 76-77, 81, 91; ECF
No. 159 at 3.)
a. Rutila offers no admissible opinion about what proportion of Owens Corning
Oakridge-brand shingles were manufactured at or near the minimum design
specifications for tear strength, mat mass, or asphalt weight.
i. This court ruled that any opinion based upon Rutila’s testing of
the warranty shingles was not admissible. FOF 192-93.
Although the court did not need to address the myriad defects in
Rutila’s testing in reaching that ruling, some flaws include that:
64
a. Rutila’s sample size was only 300 shingles out of the
millions of shingles manufactured during the proposed
class period. FOF 111, 184, 191.
b. All shingles that were tested were sent to Owens Corning
with a warranty claim, which is indicative of some level
of dissatisfaction and product failure. FOF 184.
c. Rutila’s testing does not readily account for the differing
conditions in which the shingles were in use prior to
being sent to Owens Corning with a warranty claim.
d. Rutila tested the warranty shingles pursuant to an
industry standard “designed for the evaluation of products
as manufactured” which explicitly states that “[p]hysical
and performance requirements after application and
during in-service use of the products described herein are
beyond the scope of this material specification.” FOF187;
(ECF No. 151-3 at 14.)
ii. Even if Rutila’s testing of the warranty claim shingles was
admissible, Rutila, in his expert report, states that the “testing
demonstrates that the mass of asphalt, mass of fiberglass felt and
mass of surfacing are near is [sic] at or below minimum for
approximately half of the 286 shingles with this data.” (ECF No.
151-5 at 21 (§ 6.1).)
65
b. Plaintiffs proffer no other evidence that is probative of the quantity of Owens
Corning Oakridge-brand shingles that were manufactured as low-end shingles
between 1992 and 2012. FOF 205-16.
FOF 196: Rutila testified that granule loss and cracking on a shingle can be caused by
conditions other than a defective design specification, such as, for example, improper
installation, poor ventilation, foot traffic, poor maintenance, and ordinary wear and tear.
(ECF No. 139 at 33, 36-39; ECF No. 162-3 at 5, 8, 10, 12, 13, 15.)
a. Rutila testified that these conditions could cause a shingle to fail regardless of
the design specification to which it was manufactured. (Id.)
b. Rutila testified that all shingles experience ordinary wear and tear, which will
include granule loss, but that “excessive” granule loss is indicative of a defect.
(ECF No. 139 at 33.)
c. In the context of this testimony, plaintiffs’ reliance in their proposed findings of
fact and conclusions of law to Rutila’s conclusory statement ruling out any
other causes for degranulation or cracking of Oakridge-brand shingles for the
entire proposed class period, and an industry article about the effects of
ventilation on roofs, are not persuasive evidence to the contrary. (ECF No. 178
at 7 (¶¶ 26-27 (citing ECF No. 151-5 at 25 and ECF No. 151-9 at 2)).)
FOF 197: Rutila testified that he could not determine the cause of a roofing leak without
seeing the roof. (ECF No. 139 at 33-34.)
FOF 198: Owens Corning’s expert witness James S. Johnson (“Johnson”) testified that
different kinds of shingle cracks are caused by different factors. For example, a spider-
66
web-like crack is distinguishable from a vertical crack, the latter of which is typically
caused by deck movement or improper ventilation. (ECF No. 162-7 at 10.)
FOF 199: Johnson testified that not all shingle cracks are an indicator of imminent shingle
failure. In this circumstance, a surface crack is distinguishable from a crack that is deep
enough to compromise the shingle’s ability to adequately shed water. (Id.)
FOF 200: In addition to the insufficient asphalt mass and insufficient mat mass/tear strength
design defect theories proffered by Rutila in his original expert reports and testimony,
Rutila advances a new design defect theory in plaintiffs’ proposed findings of fact and
conclusions of law, i.e., that Owens Corning’s design specifications are defective because
they allow an excessive amount of filler to be added to the asphalt. FOF 203. Before
proceeding to analyze the record evidence in support of each of these three design defect
theories, however, the court makes the following general findings.
a. The reference in plaintiffs’ proposed findings of fact and conclusions of law to
the Atlanta and Memphis plants, ECF No. 178 at 3 (¶ 6), indicates that plaintiffs
persist in their attempts to rely upon Rutila’s testing of the warranty shingles,
despite the court’s Daubert rulings precluding that reliance.
i. Rutila excluded these two plants from his written expert opinions
because warranty shingles from those two manufacturing plants
were not available to him for testing. (ECF No. 151-4 at 6 (¶
21(d)); ECF No. 151-5 at 11 (chart), 15 (§ 5.3.1.).)
ii. This court ruled during Daubert proceedings that Rutila cannot
base his opinions on his testing of the warranty shingles. FOF 19293.
67
b. Rutila offers his design defect opinions, which purport to apply to all Oakridgebrand shingles produced for the proposed 20-year class period, even though he
“only [saw] Owens Corning specification ‘targets’ for 2000 through 2012.”
(ECF No. 178 at 2 (¶ 5); ECF No. 151-4 at 5 (¶ 21(b)); ECF No. 165-1 (design
specifications from 1993 to 2012).)
c. Rutila never defines the point at which the measurements in Owens Corning’s
design specifications go from being nondefective to defective for any of his
three design defect theories. FOF 194(c).
d. Rutila never quantifies what proportion of Oakridge-brand shingles were
manufactured as low-end shingles, under any of his three design defect theories,
during the proposed 20-year class period, and, in fact, his testing of the warranty
shingles, although inadmissible, indicates that allegedly defective shingles were
produced only about half of the time. FOF 195(a).
FOF 201: Despite pointed and extended questions from the court at oral argument, ECF No.
173 at 91-99, plaintiffs offer no legal authority to support their theory that admittedly
nondefective products can, nevertheless, be considered defectively designed if a design
specification establishes a range of measurements, some of which will produce defective
products and some of which will produce nondefective products, especially where, as
here, plaintiffs failed to identify where within the range of measurements the design
crosses the line from producing nondefective products to producing defective products or
to quantify how often defective products, versus nondefective products, were produced,
even though this legal contention is seemingly novel, illogical, and contrary to the weight
68
of authority. Reyes v. Netdeposit, LLC, 802 F.3d 469, 485 (3d Cir. 2015) (the standard is
not whether it is mathematically or scientifically possible that one of the telemarketing
firms used by defendants did not engage in the allegedly wrongful conduct, but whether
plaintiff established that it is more likely than not that the telemarketing firms used by
defendants engaged in the allegedly wrongful conduct); In re Whirlpool Corp. FrontLoading Washer Products Liab. Litig., 722 F.3d 838, 847 (6th Cir. 2013) (plaintiffs
established that all Duet-brand washing machines were built to designs with “nearly
identical engineering” that differed only in the machines’ size and aesthetics and expert
witness opined that a common defect, failure to self-clean the tub, was present in all
machines regardless of slight design differences); Marcus, 687 F.3d at 602-03 (expert
witness testified, after reviewing “thousands of pages of specifications” that all tires,
regardless of model or size, are substantially similar in construction and all tires manifest
the same defective characteristic, i.e., extra stiffness); Wolin v. Jaguar Land Rover N.
Am., LLC, 617 F.3d 1168, 1172 (9th Cir. 2010) (same design defect, i.e., a geometry
defect in vehicle’s alignment, was present in each class member’s car); Pella Corp. v.
Saltzman, 606 F.3d 391, 391 (7th Cir. 2010) (all ProLine casement windows were
designed to allow water to seep behind aluminum casing, which accelerated wood rot);
Martin v. Ford Motor Co., 292 F.R.D. 252, 256-59, 267 (E.D. Pa. 2013) (class action
involved only “second generation” “rear twist beam axle,” which was manufactured by
one company for fewer than five years); Brunson v. Louisiana-Pacific Corp., 266 F.R.D.
112, 114, 119 (D.S.C. 2010) (all trimboard suffered from same manufacturing defects
that made it rot, warp, and crack prematurely); Payne v. Goodyear Tire & Rubber Co.,
216 F.R.D. 21, 23, 28 (D. Mass. 2003) (hosing used in radiant floor heating system was
69
defectively designed causing oxidation, hardening, cracks, and eventually, leaks; testing
and other evidence confirmed presence of defect in all hosing).
(a) Design Defect #1 – Insufficient Quantity of Asphalt
FOF 202: The first design defect identified by plaintiffs is that Owens Corning used
insufficient quantities of asphalt for the amount of granules being applied to the shingles,
causing the excessive granules to fall off, which degrades shingle performance. (ECF No.
151-4 ¶ 22(d); ECF No. 151-5 at 6, 21; ECF No. 151-6 at 4-5, 7-8; ECF No. 178 at 3 (¶
6) and 5 (¶ 12).)
a. Granules, which are usually ceramic, are applied to the surface of a shingle after
the fiberglass mat has been coated with an asphalt/filler mixture. (ECF No. 165
at 7 (¶19).) The granules are what provide color to the shingle. (Id. at (¶ 20).)
b. Rutila’s original expert report and supplemental declaration both identify this
“insufficient quantity of asphalt” design defect as the cause of degranulation of
Oakridge shingles. (ECF No. 151-5 at 6; ECF No. 151-4 at 6 (¶ 22).)
c. Plaintiffs specifically contend that “Owens Corning’s design specifications for
Oakridge are defective because they allowed asphalt mass to be at or close to
the 15 pounds-per-square foot minimum, but that [such an] asphalt mass is not
sufficient to adhere mineral mass [granules] to the shingle.” (ECF No. 178 at 5
(¶ 12 (citing Rutila’s report, deposition, and Daubert hearing testimony)).)
i. Although testifying that the asphalt mass can only be deemed
insufficient in relationship to the granule mass and other
measurements, FOF 194(c)(ii), Rutila does not identify what
amount of granule mass is appropriate if asphalt mass is at or close
70
to 15 pounds-per-square foot. (ECF No. 139 at 30-31; ECF No.
151-8 at 5-6.)
ii. Although the court prohibited Rutila from relying upon the specific
measurements to which he testified at the continued Daubert
hearing because he produced no documentation or support for
them, FOF 194(c)(i), plaintiffs, nevertheless, cite to Rutila’s
testimony about these numbers to support this design defect theory.
(ECF No. 178 at 5 (¶ 12) (citing ECF No. 139 at 29-33).) This is
improper and such evidence will not be considered.
iii. The record includes no evidence establishing how much granule
mass is appropriate if asphalt mass is at or close to 15 pounds-persquare foot.
iv. The record includes no evidence establishing what shingle mass
and mat mass is appropriate if asphalt mass is at, or close to, 15
pounds-per-square foot, so that granule mass can be calculated by
subtracting asphalt mass and mat mass from shingle mass to arrive
at granule mass.
(b) Design Defect #2 – Insufficient Quality of Asphalt
FOF 203: At oral argument and in plaintiffs’ proposed findings of fact and conclusions of
law, plaintiffs advanced a new asphalt-based design defect theory: degranulation occurs
because Owens Corning uses design specifications that allow the filler content of the
asphalt to be at or above 66%. (ECF No. 173 at 62-63; ECF No. 178 at 5-7 (¶¶ 13-24).)
71
In other words, plaintiffs now identify “insufficient quality of asphalt” as a design defect
that causes Oakridge-brand shingles to degranulate. (ECF No. 178 at 9 (¶ 34).)
a. Filler is a stabilizer that is mixed with the asphalt before being applied to the
fiberglass mat. (ECF No. 165 at 6-7 (¶¶ 16-17, 19).)
b. Under ASTM D3462 the maximum filler permitted is 70% of the asphalt
mixture. (ECF No. 151-3 at 16.) Under ASTM D3462, a shingle manufactured
with a filler content of 66% falls within acceptable industry standards.
i. The Owens Corning design specification data offered into
evidence reflects that the maximum filler percentage is always
below 70%, the industry standard. (ECF No. 170-2 at 2-17.)
ii. The Owens Corning design specification data offered into
evidence reflects that target filler percentage is typically at 67%
or less. (Id.) The historic averages for all Owens Corning design
specifications offered into evidence reflect target filler of 65.9%
and a maximum filler of 67.1%. (Id. at 17.)
iii. None of the evidence propounded by plaintiffs contradicts these
facts. (ECF No. 178 at 5 (¶¶ 14-16).)
c. There is no evidence in the record, expert or otherwise, that a design
specification that sets filler content at 66% or higher is a defective design.
i. Rutila never proffered an opinion in a written expert report, during
his deposition, or at the Daubert hearing that a specification that
sets filler content above any particular percentage, including 66%,
is a design defect.
72
a. The only reference to “filler percents [sic]” in Rutila’s
submissions is a single paragraph in the declaration he
filed in opposition to Owens Corning’s Daubert motion.
(ECF No. 154-2 at 7 (¶ 23) (citing ECF No. 154-3).) The
declaration refers to an internal Owens Corning report,
dated January 5, 1995, that summarizes testing conducted
“at the Houston roofing plant” in 1994 “primarily aimed
at the Oakridge II product line.” (ECF No. 154-2 at 7 (¶
23) (citing ECF No. 154-3 (“the Houston Report”)).)
(i) According to Rutila, the Houston Report
concluded that filler percentages were too
high and asphalt content too low for the
Oakridge II product to last 25 years. (ECF No.
154-2 at 7 (¶ 23); see ECF No. 139 at 55
(testifying, at continued Daubert hearing,
about this report).)
(ii) Rutila does not opine in any written report,
based upon the Houston Report, or any other
evidence, that a design specification with filler
content set at 66%, or any particular
percentage, is a defective design.
(iii)The court never ruled that Rutila could proffer
an expert opinion that a design specification
73
with filler content set at or above 66% was
defective. FOF 182-200.
ii. The Houston Report, to which Rutila cites, does not state that a
design specification that allows filler content to be 66% is
defective. (ECF No. 178 at 6-7 (¶¶ 19-20, 24).)
a. The Houston Report recognizes, generally, that a
correlation can exist between filler percentage and
shingle performance. (ECF No. 154-2 at 4.)
b. The Houston Report recommends a specific reduction in
filler percentage for one line of lightweight Oakridgebrand shingles, but explains that filler percentages could
not be reduced on “heavy weight” Oakridge-brand
shingles because face-to-face sticking increased as a
result. (ECF No. 154-3 at 7.)
c. Owens Corning’s Materials Engineer Leader for the
Roofing and Asphalt Division explains that “technology
and raw materials for the manufacture of Oakridge
shingles have changed significantly since this report was
written in 1995.” (ECF No. 165 ¶ 38.)
d. The Houston Report evidences an isolated issue,
experienced at one point in time more than twenty years
ago and in one manufacturing plant in connection with
one kind of Oakridge-brand shingle.
74
iii. The internal and industry documents to which plaintiffs cite in
their proposed findings of fact and conclusions of law do not state
that a design specification setting filler content at or above any
particular percentage, including 66%, is defective.
a. As stated above, the Houston Report does not state that a
filler percentage at, or in excess of, 66% constitutes a
design defect. FOF 203(c)(i) – (iii)(a); (ECF No. 178 at
5-7 (¶¶ 17, 19-20, 24).)
b. The 1993 article from the 10th Conference on Roofing
Technology, states that “increasing filler percentages
does not materially improve shingle performance,” but
does not state that a filler percentage at any specific level
is a design defect. (ECF No. 178 at 5 (¶ 13).) The article
does not identify any particular filler percentage as being
acceptable or unacceptable. The article does not discuss
Owens Corning or Oakridge-brand shingles specifically.
In addition, the article is outdated given the advances in
material composition and technology in the shingle
industry. (ECF No. 165 at 12 (¶ 38).)
c. The internal Owens Corning communications and
presentations about the cost of asphalt do not indicate that
a filler percentage at, or in excess of, 66% constitutes a
design defect. (ECF No. 178 at 5-6 (¶¶ 17-18).)
75
(i) The internal discussions are sporadic.
(ii) The internal discussions reject certain production
options on the basis of cost, but do not indicate
that the selected, less-expensive, options
constitute design defects.
b) The internal emails from 2006 and 2009, which consider
adjustments to filler percentages at particular plants to
address specific quality control issues, do not state that a
filler percentage at, or in excess of, 66% constitutes a
design defect. (ECF No. 178 at 6 (¶¶ 21, 23).)
(i) The emails are sporadic.
(ii) The emails are directed to specificallyidentified production concerns. They are not
comprehensive discussions about proper filler
percentages for all Oakridge-brand shingles
produced over the proposed 20-year class
period.
(iii) For at least one quality control issue the “root
cause” of the problem was not filler
percentage, but “coater scrapers.” (ECF No.
151-8 at 22-23.)
(iv) The emails never state that filler percentages
at 66%, or higher, constitute a design defect.
76
c) A single, general reference in a 1998 Owens Corning
training manual, in an introductory section entitled
“Shingle Raw Materials,” to the negative effects of
“adding too much filler” does not indicate that a filler
percentage at, or in excess of, 66% constitutes a design
defect. (ECF No. 178 at 6 (¶ 22) (citing ECF No. 154-4 at
4 (bates ending ‘637 [incorrectly cited as ECF No. 154-4
at 3 (bates ending ‘614)]).)
iv. The record reflects that the kind of filler being used, the product
being made, the location of the manufacturing plant, and the kind of
asphalt being used all affect the appropriate filler percentage. (ECF
No. 165 at 6 (¶¶ 16-18); ECF No. 154-7 at 16.)
(c) Design Defect #3 – Insufficient Mat Mass and Tear Strength
FOF 204: The last design defect identified by plaintiffs is that the mat mass set forth in
Owens Corning’s design specifications marginally satisfies the ASTM D3462 standard
while the tear strength is at or close to the 1700-gram minimum. (ECF No. 151-5 at 1718; ECF No. 178 at 3-4 (¶ 6-9).)
a. ASTM D3462 sets the minimum mat mass at 1.35 pounds-per-100-square feet
and the minimum tear strength at 1,700 grams. (ECF No. 151-3 at 15-16; ECF
No. 139 at 101-02.)
b. The court permitted Rutila to testify that a mat mass of 1.5 pounds, or more, is
appropriate in order to produce a “longer-lasting shingle.” (ECF No. 139 at 4243; ECF No. 159 at 21.)
77
c. There is no evidence in the record establishing the appropriate tear strength for
any particular corresponding mat mass measurement, even though Rutila
acknowledges that the appropriateness of design specification measurements
can only be judged in relationship to each other. FOF 194(c)(ii).
d. The record reflects that Owens Corning’s design specification measurements
met or exceeded industry standards at all pertinent times.
i. Owens Corning’s design specifications almost always set the
minimum tear strength at or above the 1,700-gram industry
standard. (ECF No. 165-1 at 1-15.) The target tear strength is
always 1,800 grams or higher, with the target at or exceeding 2,000
grams the vast majority of the time. (Id.) In those four instances,
out of hundreds, where the minimum tear strength is below 1,700
grams, the mat mass is 1.69 pounds, or more, which Rutila admits
is an increase in mat mass that would materially improve shingle
performance and would produce a “longer-lasting shingle.” (Id. at
2, 4; ECF No. 139 at 42-43, 101-02; ECF No. 159 at 21.)
ii. Owens Corning’s design specifications include only a target
measurement for mat mass. (ECF No. 165-1 at 1-15.) That target
is always at least .27 pounds above the industry minimum of 1.35
pounds-per-100-square foot. (Id.) Rutila admits that such an
increase in mat mass would materially improve shingle
performance. (ECF No. 139 at 101-02.)
78
e. The record does not reflect, as plaintiffs contend, that Oakridge-brand shingles
“often missed” or “narrowly met” tear-strength targets. (ECF No. 178 at 4 (¶
10).)
i. One category of evidence on which plaintiffs rely for this assertion
relates to failures of Oakridge-brand shingles to pass Underwriters
Laboratories (“UL”) testing for minimum tear strengths. (ECF No.
178 at 4 (¶ 10) (citing ECF No. 154-13 to -15).) This category
consists of a total of two UL incidents, one in 2010 and one in
2012. The Summit, Illinois plant was issued a variation notice in
2010 because tear strengths were 24 grams below the 1,700-gram
tear strength minimum when tested. (ECF No. 154-13 to -15.) A
second sample had to be tested by the UL representative at the
Memphis, Tennessee plant in 2012, because the first sample tested
below the 1,700-gram tear strength minimum. (Id.) The record
reflects that both incidents were promptly resolved to the
satisfaction of UL, without the need for further corrective action.
ii. The other category of evidence on which plaintiffs rely for this
assertion is email communications between Owens Corning
employees. (ECF No. 178 at 4 (¶ 10) (citing ECF No. 154-6 and 12).) This category consists of a total of two emails, one in 2005
and one in 2009. (ECF No. 154-6 and -12.) The emails reflect
isolated incidents, limited in time and location, in which certain
products tested at low tear strength. (ECF No. 154-6 and -12.)
79
iii. Plaintiffs proffer no other admissible evidence that Oakridge-brand
shingles often missed or narrowly met tear-strength targets.
b. Alleged Prevalence of Defects: Systematic Flaws and Suppression of Information
FOF 205: Plaintiffs endeavor to establish that the design defects identified by Rutila are
prevalent in Oakridge-brand shingles through evidence that the shingles suffered from
“systematic flaws” and Owens Corning suppressed negative information about those
flaws. (ECF No. 178 at 8-9 (¶¶ 28-34).)
FOF 206: As set forth previously in these findings of fact, plaintiffs’ expert witness, Rutila,
can offer no opinion about how often Oakridge-brand shingles are manufactured at
design specification measurements that he claims are defective. FOF 195(a).
FOF 207: The named plaintiffs contend that Owens Corning’s warranty program and
internal quality control practices and communications are proof that Oakridge-brand
shingles suffer from systematic flaws that cause granule loss, regardless of the particular
design defect theory at issue. (ECF No. 178 at 8-9 (¶¶ 28-33).)
a. The evidence establishes that fewer than half of one percent of Oakridge-brand
shingles result in a warranty claim. FOF 195(a)(ii). This level of warranty
activity is not indicative of systematic problems with a product.
b. Plaintiffs contend that Owens Corning’s formation of “granule loss” teams is
evidence of “systematic flaws.”
i. The record reflects, however, that it was routine practice for Owens
Corning to form “focused quality teams,” such as granule loss teams, to
address myriad issues, including color, brittleness, and deformation,
80
among other issues. (ECF No. 154-18 at 4; ECF No. 154-8 at 10-12;
ECF No. 154-18 at 5.)
ii. The record reflects that a second granule loss team was formed in 2012
not because the first granule loss team failed to adequately address the
issue, resulting in a recurrence, but because the focus shifted to different
manufacturing plants. (ECF No. 154-17 at 4-5 (first team (2009-2011) in
Jacksonville, Houston, and Denver and second team (2012) in Irving,
Portland, Savannah, and Compton); ECF No. 154-18 at 5 (same).)
c. The UL letter to which plaintiffs cite is dated 2012 and reflects that a single
consumer complaint about granule loss was received by UL. (ECF No. 154-16.)
A single consumer complaint is not indicative of systematic problems with a
product.
d. The R&A Quality Overview cited by plaintiffs, which is dated 2011, addresses
claims for all Owens Corning shingles, not only Oakridge-brand shingles. (ECF
No. 154-18; ECF No. 165 at 12 (¶ 39).)
e. The email from Mark Zeorlin, dated 2009, in which filler percentages are
discussed is a) an isolated communication, that b) discusses products other than
Oakridge, and c) involves a situation in which it was later determined that
“coater scrapers” were the root cause of granule loss. (ECF No. 154-10; ECF
No. 151-8 at 23.)
FOF 208: The named plaintiffs contend that two emails are proof that Owens Corning
suppressed negative information about Oakridge-brand shingles, and support an inference
that design defects were prevalent in Oakridge-brand shingles. (ECF No. 178 at 9 (¶ 34).)
81
a. In the 2008 email, an Owens Corning employee states that he “cannot believe
some of the data.” (ECF No. 170-4.) The email attaches various data. Without
explanation or context, the reader does not know what the author “cannot
believe” and why he “cannot believe” it. (ECF No. 170-4.)
b. The 2009 email concerns test results that involve an ongoing lawsuit filed by a
homeowner against Owens Corning, which likely explains why the author did
not want to “state what is highlighted below.” (ECF No. 170-3.)
2. Inferences Drawn from, and Sufficiency of, the Evidence
FOF 209: The record contains no evidence of any set of specific design specification
measurements that will produce defective shingles, or identify how “near” to the
minimums of Owens Corning’s design specifications a shingle must be manufactured in
order to be defective. FOF 194(c).
FOF 210: Not all Oakridge-brand shingles will be manufactured at allegedly defective
design specification measurements. FOF 194(c).
FOF 211: Some Oakridge-brand shingles will last for 30 years, or more. FOF 194(f).
FOF 212: Some Oakridge-brand shingles will fail for reasons unrelated to the design
specifications to which they were manufactured. FOF 196.
FOF 213: The record contains no evidence, expert or otherwise, to support a finding that
Oakridge-brand shingles were often, usually, or typically manufactured at or near the
minimums of Owens Corning’s design specifications. FOF 195.
FOF 214: The record contains no evidence, expert or otherwise, to support a finding that
Oakridge-brand shingles were often, usually, or typically manufactured below target
specifications. FOF 195.
82
FOF 215: The record does not support an inference that Oakridge-brand shingles often
missed, or narrowly met, tear-strength targets. FOF 204.
FOF 216: The evidence relied upon by plaintiffs to establish systematic flaws and
suppression of negative information is innocuous and sporadic and not probative that
any defects were prevalent in Oakridge-brand shingles. FOF 205-08.
3. Ultimate Factual Findings: Design Defects
FOF 217: The record does not support a finding that Owens Corning’s design specifications
are defective because they call for an insufficient quantity of asphalt mass for the amount
of granules being affixed to the shingle.
FOF 218: The record does not support a finding that setting filler percentage at or above
66% constitutes a design defect.
FOF 219: The record reflects that the design specifications to which Oakridge-brand
shingles were manufactured between 1992 and 2012 do not contain the purported
insufficient mat mass and tear-strength defect identified by plaintiffs.
FOF 220: Even if the record established the existence of any of the three design defects
identified by plaintiffs, there is no evidence that the defects were prevalent in Oakridgebrand shingles.
F. Identification of Owens Corning Shingles
FOF 221: All Oakridge-brand shingles have release tape on the back, to keep the shingles
from sticking together during shipping. (ECF No. 165 at 10 (¶ 31).)
FOF 222: Before 1995, the release tape did not include any identifying information. (Id.)
FOF 223: After 1995, the release tape identifies only the plant at which a shingle was
produced, but does not indicate the brand of Owens Corning shingle. (ECF No. 173 at 68-
83
69.) Owens Corning’s plants will produce more than one brand of shingle at the facility.
(Id.) Identification of the manufacturing plant, therefore, is not dispositive of whether the
shingle is an Oakridge-brand shingle.
FOF 224: Oakridge-brand shingles that do not meet Owens Corning specifications may be
sold on the seconds market, without any warranty. (ECF No. 165 at 10 (¶ 32).) Although
a mark is to be applied to such product, it may not be readily apparent to a consumer that
the shingle was sold on the seconds market. (Id.)
FOF 225: There are two or three experts at Owens Corning who can determine whether a
shingle is an Oakridge-brand shingle by looking at it. (ECF No. 173 at 70.)
FOF 226: During Owens Corning’s administration of warranty claims, it sometimes, but not
often, denies a claim on the basis that the product is not an Owens Corning product. FOF
114.
G. Owens Corning’s Bankruptcy Proceedings and the Court of Appeals’ Decision
FOF 227: In 2000, Owens Corning and several related entities filed for bankruptcy relief
under Chapter 11 of the Bankruptcy Code in the United States Bankruptcy Court for the
District of Delaware. Wright, 450 B.R. at 545.
FOF 228: On September 26, 2006, the bankruptcy court confirmed Owens Corning’s
reorganization plan (the “Bankruptcy Order”). (ECF No. 162-49.) Under the terms of the
Bankruptcy Order and pursuant to 11 U.S.C. § 1141 all claims that existed against Owens
Corning as of September 26, 2006 were discharged.
FOF 229: Owens Corning moved for summary judgment in this case on July 19, 2010,
arguing that because the claims of named plaintiffs Wright and West existed before
September 26, 2006, they were discharged when its reorganization plan was approved.
84
This court granted summary judgment in favor of Owens Corning based upon the Court
of Appeals for the Third Circuit’s en banc decision in Jeld-Wen, Inc. v. Van Brunt (In re
Grossman’s Inc.), 607 F.3d 114 (3d Cir. 2010) (“Grossman’s”).
FOF 230: In Grossman’s, the court of appeals ruled that a claim arises for purposes of
determining its dischargeability in bankruptcy when the claimant is exposed to the
debtor’s product or conduct, even if the injury does not manifest itself, and is not
discovered, until after the reorganization. Grossman’s, 607 F.3d at 121. In other words, a
claim arises on the date that a product is purchased, not on the date that it fails, or the
purchaser discovers that it failed. In reaching this holding, the court overruled Avellino
& Bienes v. M. Frenville Co. (In re M. Frenville Co.), 744 F.2d 332 (3d Cir.1984). Id. at
117.
FOF 231: On appeal, this court’s summary judgment decision was affirmed, in part, and
reversed, in part. Wright, 679 F.3d at 105-06. The appellate court recognized that, under
Grossman’s, the claims of Wright and West were discharged by virtue of the Bankruptcy
Order because they purchased their Oakridge shingles prior to September 26, 2006. Id. at
106. The court of appeals, however, declined to apply the rule established in Grossman’s
retroactively and held that the “Frenville test” applies to “bankruptcy cases in which
reorganization plans are proposed and confirmed prior to June 2, 2010, when Grossman’s
was decided.” Id. at 109. Because Owens Corning’s reorganization plan was confirmed
before June 2, 2010, the court of appeals held that the Frenville test, and not the
Grossman’s test, applied.
FOF 232: After deciding the proper test to apply to Wright’s and West’s claims, the court of
appeals engaged in the case-specific, and fact-based assessment required by Frenville to
85
determine when their claims arose, and whether the notice that they received about
Owens Corning’s bankruptcy satisfied the requirements of due process. Wright, 679 F.3d
at 107-09.
FOF 233: Owens Corning does not dispute that the Frenville test must be applied to
determine whether claims were discharged by the Bankruptcy Order. (ECF No. 173 at
14-15, 28-33.)
FOF 234: The Bankruptcy Order vests jurisdiction over any and all matters arising out of
Owens Corning’s bankruptcy case in the bankruptcy and federal courts located in
Delaware, which is located within the jurisdiction of the Court of Appeals for the Third
Circuit. (ECF No. 162-49 at 64-65.)
FOF 235: The Court of Appeals for the Third Circuit, in Wright applied third-circuit law to
the claims of not only Wright, who is a Pennsylvania owner, but also West, who is an
Illinois owner, to determine the effect of Owens Corning’s Bankruptcy Order. Wright,
679 F.3d at 105-09.
FOF 236: The proposed class period begins fourteen years before the Bankruptcy Order was
entered. FOF 6, 227.
CONCLUSIONS OF LAW
A. General Legal Principles
COL 1:
To be certified, a class must satisfy the four prerequisites of Federal Rule of Civil
Procedure 23(a): (1) numerosity; (2) commonality; (3) typicality; and (4) adequacy of
representation. FED. R. CIV. P. 23(a).
86
COL 2:
Additional requirements must be satisfied depending on whether a plaintiff seeks
certification of a class under Rule 23(b)(1), (2), or (3). Reyes, 802 F.3d at 482 (citing
Marcus, 687 F.3d at 590).
COL 3:
To determine whether to certify a class, the court must be “satisfied after a
rigorous analysis, that the prerequisites of Rule 23 are met.” Reyes, 802 F.3d at 484
(citing In re Hydrogen Peroxide Antitrust Litig., 552 F.3d 305, 309 (3d Cir. 2008)). The
rigorous analysis requires the court to make explicit findings that the requirements of
Rule 23 are met, not that they are supported by some evidence or could be met later in the
case. In re Hydrogen Peroxide, 552 F.3d at 321.
COL 4:
Plaintiffs have the obligation, at the class certification stage, to establish how the
evidence and the law support classwide treatment of their claims, including through the
production of expert testimony and opinions. In the Matter of IKO Roofing Shingle
Products Liability Litig., 757 F.3d 599, 603-04 (7th Cir. 2014) (noting that “it may
require sophisticated analysis” to determine what failure data and testing in accordance
with manufacturing standards reveal about the defectiveness of shingles.)
COL 5:
The proponent of class certification has the burden of proving each of the
prerequisites under Rule 23(a) and that the class fits within the desired categories of class
actions set forth in Rule 23(b) by a preponderance of the evidence. In re Hydrogen
Peroxide, 552 F.3d at 307, 316 n.14, 317 (citation omitted); see Hayes v. Wal–Mart
Stores, Inc., 725 F.3d 349, 354 (3d Cir. 2013) (“It is plaintiff's burden to show that a class
action is a proper vehicle for this lawsuit”).
COL 6:
In assessing whether a plaintiff has satisfied its burden, the district court “cannot
be bashful” and must resolve all factual and legal disputes relevant to class certification,
87
including disputes touching on the elements of the causes of action, and the merits of a
claim. Reyes, 802 F.3d at 484.
COL 7:
The Court of Appeals for the Third Circuit recently set forth the district court’s
duty in ruling on a motion to certify a class by stating that “it is now clear that the District
Court must: (1) conduct rigorous analysis, (2) review all avenues of inquiry in which it
may have doubts (even if it requires reviewing the merits), (3) be satisfied and (4) make a
definitive determination on the requirements of Rule 23, or even (5) require that a
plaintiff demonstrate actual, not presumed conformance with Rule 23 requirements.”
Reyes, 802 F.3d at 485. Plaintiffs, however, need not establish the validity of their claims
at the class certification stage. Id.
B. Rule 23(a) Requirements
COL 8:
In order to be certified, a class must first satisfy the four prerequisites of Federal
Rule of Civil Procedure 23(a): (1) numerosity; (2) commonality; (3) typicality; and (4)
adequacy of representation. FED. R. CIV. P. 23(a); Reyes, 802 F.3d at 482.
COL 9:
For purposes of the Rule 23(a) analysis, plaintiffs propose two different classes: a
class whose members own structures in any state (under Rule 23(b)(1)(B)) and a class
whose members own structures in Pennsylvania, Illinois, California, or Texas (under
Rule 23(b)(2) and (b)(3)). (ECF No. 154 at 23.) Where relevant, the court will
differentiate between these two categories of classes in the following discussion of the
Rule 23(a) requirements. If no distinction is made, the analysis is the same. For ease of
reference, the first class will be referred to as the “nationwide class” and the second class
will be referred to as the “four-state class.”
88
COL 10: Even where Owens Corning does not challenge plaintiffs’ ability to meet their
burden of proving each of the Rule 23(a) requirements, this court has an independent
duty to ensure that plaintiffs prove, by a preponderance of the evidence, that each of the
Rule 23(a) requirements is met.
1.
Numerosity
COL 11: Under Rule 23(a), a plaintiff bears the burden of establishing numerosity by a
preponderance of the evidence. Marcus, 687 F.3d at 594-95. Plaintiff must prove that the
putative class is “so numerous that joinder of all members is impracticable.” FED. R. CIV.
P. 23(a)(1).
COL 12: The Court of Appeals for the Third Circuit has said that, “[n]o minimum number
of plaintiffs is required to maintain a suit as a class action, but generally if the named
plaintiff demonstrates that the potential number of plaintiffs exceeds 40, the first prong of
Rule 23(a) has been met.” Stewart v. Abraham, 275 F.3d 220, 226-27 (3d Cir. 2001)
(citing 5 James Wm. Moore et al., Moore's Federal Practice § 23.22[3][a] (Matthew
Bender 3d ed. 1999)).
COL 13: A court cannot “assume,” “speculate,” or defer to “common sense” with respect to
how many class members exist. Marcus, 687 F.3d at 595-97. Instead, a plaintiff must
produce evidence, direct or circumstantial, specific to the products, problems, parties, and
geographic areas actually covered by the proposed class definitions to allow a district
court to make a factual finding on this requirement. Id. at 596.
COL 14: Owens Corning does not dispute that plaintiffs can satisfy the numerosity
requirement for both the nationwide class and the four-state class. (ECF No. 176 ¶ 90;
ECF No. 173 at 9.)
89
COL 15: The Nationwide Class: Data produced by Owens Corning in this litigation reflects
that it manufactured millions of square feet of Oakridge-brand shingles between 2000 and
2006. FOF 111; (ECF No. 154-24.) The court can reasonably infer from this
manufacturing data, which covers only a portion of the proposed class period, that there
are significantly more than 40 people throughout the country who purchased Oakridgebrand shingles before September 26, 2006 and have those shingles currently installed on
structures that they own. The numerosity requirement of Rule 23(a) is met with respect
to the proposed nationwide class.
COL 16: The Four-State Class: Data produced by Owens Corning in this litigation
indicates that, between 1992 and 2012, nearly 30,000 warranty claims were made in
connection with Oakridge-brand shingles that were produced at manufacturing plants that
typically distribute shingles for use in Pennsylvania, Illinois, California, and Texas. FOF
106-07, 113. The court can reasonably infer from this volume of warranty claims that
substantially more than 40 claims were made by owners located in Pennsylvania, Illinois,
California, and Texas whose shingles manifested any cracking, degranulation,
fragmentation, or deterioration during the warranty period. FOF 6. The numerosity
requirement of Rule 23(a) is met with respect to the proposed four-state class.
COL 17: Plaintiffs satisfied their burden of proving numerosity by a preponderance of the
evidence for the proposed nationwide class and the proposed four-state class.
2.
Commonality
COL 18: “A putative class satisfies Rule 23(a)'s commonality requirement if ‘the named
plaintiffs share at least one question of fact or law with the grievances of the prospective
class.’” Reyes, 802 F.3d at 486 (citing Rodriguez v. Nat'l City Bank, 726 F.3d 372, 382
90
(3d Cir. 2013)). “Commonality does not require perfect identity of questions of law or
fact among all class members. Rather, even a single common question will do.” Reyes,
802 F.3d at 486 (citing Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011)).
COL 19: A district court must ask whether determining the truth or falsity of a common
contention will resolve an issue that is central to the validity of each one of the claims in
one stroke. Id. at 487. “What matters to class certification ... is not the raising of
common questions—even in droves—but, rather the capacity of a classwide proceeding
to generate common answers apt to drive the resolution of the litigation.’” Dukes, 564
U.S. 338, 131 S.Ct. 2541, 2551 (emphasis and ellipsis in the original).
COL 20: The bar for establishing commonality is “not high” and is “easily met.” In re
Cmty. Bank of N. Va. Mortgage Lending Practices Litig., 795 F.3d 380, 397 (3d Cir.
2015) (Community Bank III); Reyes, 802 F.3d at 486 (citing Baby Neal v. Casey, 43 F.3d
48, 56 (3d Cir. 1994)).
COL 21: Where an action is to proceed under Rule 23(b)(3), district courts typically
analyze Rule 23(a)’s commonality requirement together with the more stringent
predominance requirement of Rule 23(b)(3). Reyes, 802 F.3d at 486; see Sullivan v. DB
Investments, Inc., 667 F.3d 273, 297 (3d Cir. 2011). The Court of Appeals for the Third
Circuit, however, recently cautioned that commonality must be established before
predominance can be considered. Reyes, 802 F.3d at 486 (emphasis in original). While
not faulting the district court in Reyes for considering the two requirements together, the
court of appeals, “in the interest of clarity,” separately addressed each requirement. Id.
Especially where, as here, plaintiffs propose classes under Rules 23(b)(1)(B) and (b)(2),
which do not include a predominance requirement, as well as Rule 23(b)(3), which does
91
include a predominance requirement, the court will not collapse the commonality
requirement of Rule 23(a) into the predominance requirement of Rule 23(b)(3) and will,
as the court of appeals suggests in Reyes, separately consider them.
COL 22: The Nationwide Class: Plaintiffs contend that commonality is satisfied for the
proposed nationwide class because there is “a common question on whether Owens
Corning’s bankruptcy discharged claims of homeowners who purchased Oakridge on or
before September 26, 2006.” (ECF No. 178 at 22 (¶ 17).) Plaintiffs state that “resolution
of this question is central to whether class members can proceed with claims against
Owens Corning.” (Id.)
a. At the hearing on plaintiffs’ motion for class certification, plaintiffs claimed that
the purpose of the proposed nationwide class was to obtain a ruling with respect
to whether the Frenville test or the Grossman’s test would apply to any
challenge made by Owens Corning to a class member’s claim on the ground that
the claim was discharged by the Bankruptcy Order. (ECF No. 173 at 10.) This
question is not justiciable and cannot satisfy the commonality requirement.
Neale v. Volvo Cars of N. Am., LLC, 794 F.3d 353, 366 (3d Cir. 2015)
(“[b]efore even getting to the point of class certification, however, class
representatives need to present a justiciable claim”).
i. The Court of Appeals for the Third Circuit has already ruled that
the Frenville test applies to any dischargeability defense asserted by
Owens Corning. FOF 230-31.
ii. Owens Corning concedes that it cannot relitigate this issue. FOF
233; (ECF No. 176 at 48 (¶ 232).)
92
iii. This question is not justiciable as it presents no live case or
controversy between the proposed nationwide class and Owens
Corning sufficient to support this court’s exercise of jurisdiction.
Valley Forge Christian Coll. v. Americans United for Separation of
Church and State, 454 U.S. 464, 471 (1982) (the “judicial power” of
the federal courts is limited to the resolution of certain “cases” and
“controversies”). A case can become moot when the issues
presented are no longer ‘live’ or the parties lack a legally
cognizable interest in the outcome. Powell v. McCormack, 395 U.S.
486, 496 (1969); Toll Bros., Inc. v. Twp. of Readington, 555 F.3d
131, 137 (3d Cir. 2009).
iv. The commonality requirement is not met with respect to the
proposed nationwide class to the extent that plaintiffs seek a ruling
about whether the Frenville test or the Grossman’s test applies to
Owens Corning’s Bankruptcy Order because the question that the
named plaintiffs ask this court to answer has already been answered
by the Court of Appeals for the Third Circuit. There is no present
dispute for this court to resolve.
v. Because this question was actually litigated and determined by the
appellate court, and was essential to its ruling, Owens Corning
would be barred from relitigating it in any court by principles of
collateral estoppel. In re Docteroff, 133 F.3d 210, 214 (3d Cir.
1997); Arlington Indus., Inc. v. Bridgeport Fittings, Inc., 106 F.
93
Supp. 3d 506, 513-14 (M.D. Pa. 2015) (citing Peloro v. United
States, 488 F.3d 163, 175 (3d Cir. 2007), and Burlington N. R.R. v.
Hyundai Merch. Marine Co., 63 F.3d 1227, 1231–32 (3d Cir.
1995)).
b. To the extent that what plaintiffs actually seek on behalf of the proposed
nationwide class is a ruling from this court that the claims of any owner who
purchased Oakridge-brand shingles before September 26, 2006, are not
discharged by Owens Corning’s Bankruptcy Order, the court of appeals’
decision in Wright specifically precludes this relief.
i. In Wright, the Court of Appeals for the Third Circuit held that the
Frenville test must be applied to determine the effect of Owens Corning’s
Bankruptcy Order on an owner’s claim.
a) Under the Frenville test, a court must look to the
underlying substantive state limitations law to determine
when a cause of action arises. Wright, 679 F.3d at 10506. For example, if the discovery rule applies under the
pertinent state law, then the claim does not arise until the
claimant discovered the injury.
b) The Frenville test is, by its nature, a case-specific, factbased assessment that cannot be applied on a classwide
basis to the claims of all owners throughout the country
who are asserting different legal claims under different
state laws and based upon different pertinent facts,
94
including, for example, when ownership of the structure
began, when the shingles exhibited failure, when a
warranty claim was submitted to Owens Corning, when
property damage was suffered, and when roofing repairs
were made.
c) Although classwide relief based upon the purchase date
of the shingles could be available under the bright-line
Grossman’s test, the court of appeals specifically rejected
application of that test to Owens Corning’s Bankruptcy
Order, even for those owners who are located in
jurisdictions outside the boundaries of the Court of
Appeals for the Third Circuit. FOF 230-31, 235.
ii. It does not follow from the court of appeals’ holding in Wright that
Wright’s and West’s claims were not discharged that all potential class
members’ claims are also not discharged by the Bankruptcy Order. In
Wright, the bankruptcy court applied the fact-specific Frenville test to the
circumstances surrounding Wright’s and West’s ownership of Oakridgebrand shingles and knowledge about Owens Corning’s bankruptcy
proceedings, and held that, under those facts, the Bankruptcy Order did
not operate to discharge their claims.
iii. Plaintiffs did not propose any method by which the court could
determine, on a classwide basis, when class members’ claims arose under
the laws of all fifty states, for the variety of substantive legal claims being
95
asserted by the named plaintiffs, and what circumstances surround each
class members’ ownership of Oakridge-brand shingles and knowledge
about Owens Corning’s bankruptcy proceedings.
iv. The commonality requirement is not met with respect to the
proposed nationwide class to the extent that plaintiffs seek a ruling
that the claims of any owner who purchased Oakridge-brand
shingles before September 26, 2006, are not discharged by Owens
Corning’s Bankruptcy Order because the Frenville test does not
permit this court to answer that question on a classwide basis.
COL 23: The Four-State Class: Plaintiffs contend that commonality is satisfied for the
proposed four-state class because in class actions relating to products liability, if there is
a classwide question about whether a product is intrinsically faulty, or whether a
manufacturer knew about or concealed such faults, commonality is satisfied. (ECF No.
178 at 22 (¶ 15) (citing Wolin, 617 F.3d at 1173, and Zeno v. Ford Motor Co., Inc., 238
F.R.D. 173, 185 (W.D. Pa. 2006))). Plaintiffs further explain that the common questions
in this case include whether Oakridge has intrinsic design defects, whether Owens
Corning knew about those design defects, whether Owens Corning misled consumers
about Oakridge’s usable life, and whether Oakridge’s defects caused it to fail before the
warranty period lapsed, which allow for classwide liability to be resolved in a single
stroke. (ECF No. 178 at 22 (¶ 16).)
COL 24: Plaintiffs satisfy the commonality requirement with respect to those legal claims
that are based upon allegedly false representations about Oakridge-brand shingles, but
only to the extent that those claims are grounded in the theory that a limited warranty of a
96
set number of years qualifies as an affirmative representation about the useful life of a
shingle.
a. There is no dispute that Owens Corning consistently offered limited
warranties of a set number of years in connection with the sale of Oakridgebrand shingles. FOF 128, 141-42, 145, 166-69.
b. Whether a limited warranty of a set number of years can, as a matter of law,
constitute a representation about the useful life of a product is a common
question with a common answer. But see COL 74(b), 173 (discussing
plaintiffs’ failure to establish that Owens Corning’s limited shingle
warranties qualify as affirmative representations about the useful life of
Oakridge-brand shingles). Although the length of the limited warranties
varies from 25 to 50 years, and now includes a lifetime term, this distinction
does not destroy commonality with respect to this legal theory.
COL 25: A classwide proceeding cannot generate common answers with respect to any
legal claims based upon the theory that Owens Corning made specific misrepresentations
in its product literature about Oakridge-brand shingles.
a. The record reflects that Owens Corning made no consistent or uniform
representations about the useful life of Oakridge-brand shingles in its product
literature. FOF 134, 146, 175-80.
b. The record reflects that Owens Corning made no consistent or uniform
representations about the qualities or characteristics of Oakridge-brand
shingles in its product literature. FOF 134, 146, 175-80.
97
c. The record reflects that even the named plaintiffs relied upon different
marketing materials and product literature, and that some also considered
oral representations made by third parties. FOF 140-46.
COL 26: Plaintiffs cannot satisfy the commonality requirement with respect to any legal
claims that are based upon an alleged design defect.
a. The record reflects that there is not a single design specification that applies
to all Oakridge-brand shingles manufactured in Pennsylvania, Illinois,
California, and Texas between 1992 and 2012, and there is no evidence that
all (or even most or many) Oakridge-brand shingles contain a particular
design specification measurement that renders the shingle defective. FOF
108, 195, 200(c). This fact alone distinguishes the instant case from the
decisions relied upon by plaintiffs, and other design defect precedent. Neale,
794 F.3d at 357 (all legal claims would be based upon whether Volvo’s
sunroof drainage systems were defectively designed); In re Whirlpool, 722
F.3d at 847 (plaintiffs established that all Duet-brand washing machines
were built to designs with “nearly identical engineering” that differed only
in the machines’ size and aesthetics and expert witness opined that a
common defect, failure to self-clean the tub, was present in all machines
regardless of slight design differences); Marcus, 687 F.3d at 602-03 (expert
witness testified, after reviewing “thousands of pages of specifications” that
all tires, regardless of model or size, are substantially similar in construction
and all tires manifest the same defective characteristic, i.e., extra stiffness);
Wolin, 617 F.3d at 1172 (same design defect, i.e., a geometry defect in
98
vehicle’s alignment, was present in each class member’s car); Pella, 606
F.3d at 391 (all ProLine casement windows were designed to allow water to
seep behind aluminum casing, which accelerated wood rot); Martin, 292
F.R.D. at 256-59, 267 (class action involved only “second generation” “rear
twist beam axle,” which was manufactured by one company for fewer than
five years); Brunson, 266 F.R.D. at 114, 119 (all trimboard suffered from
same manufacturing defects that made it rot, warp, and crack prematurely);
Payne, 216 F.R.D. at 23, 28 (hosing used in radiant floor heating system was
defectively designed causing oxidation, hardening, cracks, and eventually,
leaks; testing and other evidence confirmed presence of defect in all hosing).
b. There is no evidence identifying, for purposes of comparison, what design
specification measurements are defective. FOF 108, 194(c), 200(c).
c. The record reflects that Owens Corning’s allegedly defective design
specifications will produce both shingles that will last less than 20 years,
and shingles that will last more than 30 years, with no way to determine,
without inspecting a sample shingle, which shingles a particular owner has.
FOF 194(b), (f) & (g).
d. The record includes no evidence establishing the frequency at which
Oakridge-brand shingles were manufactured at the low-end of Owens
Corning’s design specifications, such that inferences could be drawn about
the prevalence of the alleged design defect. FOF 195; Reyes, 802 F.3d at
485 (the standard is not whether it is mathematically or scientifically
possible that one of the telemarketing firms used by defendants did not
99
engage in the allegedly wrongful conduct, but whether plaintiff established
that it is more likely than not that the telemarketing firms used by
defendants engaged in the allegedly wrongful conduct). Instead, the
statistical evidence in the record reflects that only around half of the
warranty shingles measured at, near, or below industry minimums. FOF
195(a)(ii). This sample, which was small and inherently biased, could not
support any inference that all (or even most or many) Oakridge-brand
shingles were manufactured at the low-end of Owens Corning’s design
specifications, even if the sample was admissible. FOF 193(b).
e. Plaintiffs submit no legal authority to support their proposition that a design
specification that sets a range of measurements, some of which will produce
defective products and some of which will not, can establish a design defect
claim. FOF 201; COL 73(b).
COL 27: Plaintiffs satisfied their burden of proving commonality by a preponderance of the
evidence only for the proposed four-state class, and only with respect to those legal
claims based upon the theory that a limited warranty qualifies as a representation about
the useful life of a shingle.
3. Typicality
COL 28: In Marcus, the Court of Appeals for the Third Circuit explained that the typicality
requirement aids a court in determining whether maintenance of a class action is
economical and whether the named plaintiff's claim and the class claims are so
interrelated that the interests of the class members will be fairly and adequately protected
in their absence. Marcus, 687 F.3d at 597-98 (citing General Telephone Co. of the Sw. v.
100
Falcon, 457 U.S. 147, 158 n.13 (1982)). Typicality “screen[s] out class actions in which
the legal or factual position of the representatives is markedly different from that of other
members of the class even though common issues of law or fact are present.” Id. at 598.
COL 29: To determine whether a plaintiff’s position is markedly different from the class as
a whole, a court compares three distinct, though related, concerns: (1) the claims of the
class representative must be generally the same as those of the class in terms of both (a)
the legal theory advanced and (b) the factual circumstances underlying that theory; (2)
the class representative must not be subject to a defense that is both inapplicable to many
members of the class and likely to become a major focus of the litigation; and (3) the
interests and incentives of the representative must be sufficiently aligned with those of
the class. Marcus, 687 F.3d at 599.
COL 30: The Nationwide Class: Plaintiffs contend that the typicality requirement is met
for the nationwide class because the bankruptcy discharge defense that Owens Corning
asserted against Wright and West “has ramifications for all other homeowners who
purchased Oakridge before Owens Corning’s bankruptcy was finalized,” and “supplies a
common course of conduct that has equal ramifications for Plaintiffs and the class.” (ECF
No. 178 at 24 (¶ 25).)
a. To the extent that the purpose of the proposed nationwide class is to obtain a
ruling with respect to whether the Frenville test or the Grossman’s test would
apply to any bankruptcy discharge defense mounted by Owens Corning to a
class member’s claim, the court finds that the typicality requirement of Rule
23(a) is readily met. In doing so, the court puts aside the conclusions reached
101
with respect to the ability of this court to issue any rulings on this question in
light of the court of appeals’ decision in Wright. COL 22(b).
b. To the extent that what plaintiffs actually seek on behalf of the proposed
nationwide class is a ruling from this court that the claims of any owner who
purchased Oakridge-brand shingles before September 26, 2006, are not
discharged by Owens Corning’s Bankruptcy Order, the court concludes that
plaintiffs cannot satisfy their burden to prove typicality by a preponderance of
the evidence.
i. The claims of the named plaintiffs are generally the same as those
of the class in terms of the legal theory advanced. Both groups
would be seeking an adjudication with respect to whether Owens
Corning’s Bankruptcy Order discharges their claims.
ii. The claims of the named plaintiffs are not generally the same as
those of the class in terms of the factual circumstances underlying
that legal theory.
a) As explained in previous findings and conclusions,
Wright directs a court to assess the factual circumstances
surrounding each individual claim in light of the legal
cause of action being asserted to determine when that
claim arose under state law and whether notice of the
bankruptcy proceedings was sufficient. FOF 229-32;
COL 22(b); Wright, 679 F.3d at 103-08. Because the
Frenville test is, by its nature, fact-specific the factual
102
circumstances pertinent to how Owens Corning’s
bankruptcy discharge defense applies to the claims held
by the named plaintiffs and the class members cannot be
“generally the same.”
(i) By way of example, the record reflects that
none of the named plaintiffs suffered property
damage or otherwise knew that the Oakridgebrand shingles on their roofs failed prior to
September 26, 2006, the date on which Owens
Corning’s Bankruptcy Order was signed. FOF
29, 38, 55, 81, 94. This fact was critical to
determining whether or not Owens Corning’s
Bankruptcy Order discharged Wright’s and
West’s claims. Wright, 679 F.3d at 107-09.
The proposed nationwide class, however,
includes owners who purchased structures on
which Oakridge-brand shingles were installed
at any time prior to September 26, 2006. It is
therefore entirely plausible, and in fact likely,
that class members who own structures on
which Oakridge-brand shingles were installed
in the early 1990s would have experienced
property damage or had other actual notice of
103
the alleged failure of their roofing shingles
before September 2006. See FOF 22, 29, 32,
38, 47, 55, 71, 81, 89, 94 (evidencing that the
named plaintiffs obtained actual knowledge
about the alleged failure of their Oakridgebrand shingles no more than 13 years after
installation). Such factual circumstances
affect the accrual of an owner’s claim under
state law, and, in turn, the sufficiency of the
notice provided about Owens Corning’s
bankruptcy case, both of which are critical
issues under Wright. Wright, 679 F.3d at 10709.
(ii) The factual circumstances of the named
plaintiffs and the members of the proposed
nationwide class will also differ due to the
breadth of the proposed class definition, which
focuses on current ownership of a structure.
Although the named plaintiffs all directed that
Oakridge-brand shingles be installed on the
roof of their structures, it is likely that some
members of the proposed nationwide class
will have purchased a structure with
104
Oakridge-brand shingles already installed on
it. The date of that purchase, the information
and inspections exchanged in connection with
the transfer of ownership, and related facts
will affect how the Frenville test applies.
(iii) Lastly, the named plaintiffs are all individuals
who own single-family dwellings, whereas the
proposed nationwide class will include
entities, associations, and governmental
bodies, as well as owners of multi-family
dwellings and commercial properties. These
circumstances will inevitably cause divergent
facts to emerge between the named plaintiffs
and the class members with respect to how the
Frenville test applies.
b) Under these circumstances, the named plaintiffs cannot
satisfy their burden of showing that their claims are
generally the same as those of the class in terms of the
factual circumstances underlying the legal theories.
iii.
Wright and West would be subject to a unique defense that is both
inapplicable to many members of the class and likely to become a
major focus of the litigation with respect to the proposed
nationwide class, and they would have differing interests and
105
incentives than the other named plaintiffs and class members. The
court of appeals issued a final ruling in Wright that Wright’s and
West’s claims were not discharged by Owens Corning’s
Bankruptcy Order. FOF 232; COL 22(b). These two named
plaintiffs have no interest or incentive in relitigating that issue, and
would be prohibited from doing so if they tried. Wright’s and
West’s claims cannot be typical of the claims held by members of
the proposed nationwide class for this additional reason.
iv. Gonzalez, Boehm, and the Maags will not be subject to any unique
defenses and will have the same interests and incentives as those
members of the proposed nationwide class, and in this respect, their
claims are typical of the claims held by members of the proposed
nationwide class.
a) Although the court of appeals’ holding about the
applicability of the Frenville test will apply to these four
named plaintiffs, as well as to all members of the
proposed nationwide class, the court of appeals did not
apply the Frenville test to their claims in Wright. That
ultimate issue is still an open question for Gonzalez,
Boehm, and the Maags, just as it will be an open question
for all (or even most or many) members of the proposed
nationwide class.
106
b) Gonzalez, Boehm, and the Maags will have the same
interest and incentive to ensure that their claims, like the
claims of the proposed nationwide class, are not deemed
discharged by Owens Corning’s Bankruptcy Order.
c. In summary, although the interests and incentives of Gonzalez, Boehm, and the
Maags are sufficiently aligned, and they are not subject to any unique defenses,
plaintiffs nevertheless failed to satisfy the typicality requirement because their
factual circumstances cannot be generally the same as those of the other
members of the proposed nationwide class because the Frenville test, by its
nature, is fact- and case-specific.
COL 31: The Four-State Class: Plaintiffs contend that the typicality requirement is met
with respect to the proposed four-state class because this litigation arises out of a
common course of wrongdoing by Owens Corning with respect to defective design,
knowledge of defect, and misrepresentations about Oakridge-brand shingles. (ECF No.
178 at 24 (¶ 24).) For the reasons set forth in the following conclusions of law, the court
finds that plaintiffs failed to demonstrate typicality by a preponderance of the evidence
for the proposed four-state class.
a. The legal theories advanced by the named plaintiffs are not generally the same as
those of the class.
i. The named plaintiffs are from different states, and propose to pursue
different combinations of legal claims against Owens Corning on a
classwide basis. FOF 13.
107
a) For example, Wright, who is from Pennsylvania, is the only
named plaintiff who seeks class certification for an express
warranty claim. She also pursues claims for breach of the
implied warranty of merchantability and unjust enrichment on
behalf of the class, but does not pursue any consumer
protection claims on behalf of the class, even though such
claims are pled in her complaint. FOF 4(b), 13.
b) In contrast, West and the Maags, who are Illinois owners, are
not pursuing any kind of warranty claim, but are seeking class
certification of consumer protection and unjust enrichment
claims. FOF 4(b), 13.
c) These examples demonstrate that the named plaintiffs, who
purport to represent owners located in all four states, do not
assert “generally the same” legal theories as all members of the
proposed four-state class.
ii. State-specific subclasses could possibly remedy this concern.
a) “[I]t is not the District Court that is to bear the burden of
constructing subclasses. That burden is upon the [plaintiff]
who is required to submit proposals to the court.” Reyes,
802 F.3d at 494 (citing U.S. Parole Comm’n v. Geraghty,
445 U.S. 388, 408 (1980)).
b) Despite being specifically informed of the need to propose
state-specific subclasses at the class certification phase, and
108
despite stating their intention to propose state-specific
subclasses in the post-hearing briefing, plaintiffs
nevertheless failed to propose state-specific subclasses in
any filing made in support of their motion for class
certification. FOF 14-15; COL 31(a)(ii). Plaintiffs, instead,
continue to assert that the court should be responsible for
constructing state-specific subclasses as a matter of trial
management, if necessary. FOF 14.
c) Under the circumstances, plaintiffs’ failure to propose
state-specific subclasses cannot be characterized as an
inadvertent oversight.
d) In any event, due to other deficiencies in the named
plaintiffs’ satisfaction of the Rule 23(a) and Rule 23(b)
requirements, the crafting of state-specific subclasses
would not ultimately alter the court’s decision with respect
to the propriety of certifying the class.
b. The factual circumstances underlying the legal theories of the named
plaintiffs are generally the same as those of the members of the proposed
four-state class, but only with respect to any legal claims that are based
exclusively on a design defect and not with respect to any legal claims that
are based upon Owens Corning’s alleged misrepresentations.
i. Plaintiffs argue that typicality is met because Owens Corning
promised all class members that Oakridge-brand shingles would last
109
for 25 years or more, but due to a common and intrinsic design
defect, no Oakridge-brand shingle will last that long.
ii. The named plaintiffs’ factual circumstances with respect to Owens
Corning’s alleged misrepresentations about the useful life of
Oakridge-brand shingles are not generally the same as those of the
members of the proposed four-state class, but are generally the
same as those of the members of the proposed four-state class with
respect to the alleged design defect for the following reasons:
a) Factual Circumstances – Specific Misrepresentations: The
record before the court establishes that Owens Corning did
not engage in any uniform marketing scheme with respect
to Oakridge-brand shingles such that this court could find
that the named plaintiffs’ factual circumstances are
generally the same as those of members of the proposed
four-state class. FOF 134-39, 145-46, 154-59, 164-65, 17580; COL 25. In fact, the record reflects that even the
named plaintiffs were exposed to different marketing
materials with different statements and representations
about Oakridge-brand shingles. FOF 140-46. It would defy
logic for this court to find that the factual circumstances of
the named plaintiffs and the proposed four-state class are
generally the same when the factual circumstances among
the named plaintiffs are not even generally the same.
110
b) Factual Circumstances – Limited Warranties as
Misrepresentations: The only factual circumstance that is
generally the same among the named plaintiffs and between
the named plaintiffs and the members of the proposed fourstate class is that all owners will hold, at least, the limited
shingle warranty of at least 25 years that is automatically
provided with all Oakridge-brand shingles. FOF 128, 16774. The record reflects, however, that there are significant
factual differences among the named plaintiffs and there
will be significant factual differences between the named
plaintiffs and the members of the proposed four-state class
with respect to the terms of those warranties.
(i) The limited shingle warranties include differing
limitations, disclaimers, terms, and scope of
coverage. FOF 169. For example, some warranties
disclaim implied warranties, while others do not.
FOF 169(e). Although all owners will hold a
limited shingle warranty, the record reflects that the
terms of those warranties will not be the same.
(ii) In addition, the limited shingle warranties are only
one kind of warranty held by purchasers of
Oakridge-brand shingles. FOF 169(c). Some class
members will have purchase enhanced warranties,
111
which include still different terms and limitations.
Id. There is no evidence that any named plaintiff
purchased such enhanced warranty protection.
c) In summary, the named plaintiffs failed to prove, by a
preponderance of the evidence, that the named
plaintiffs’ factual circumstances with respect to Owens
Corning’s alleged representations about the useful life
of Oakridge-brand shingles are generally the same as
those of the members of the proposed four-state class.
Although the named plaintiffs and all class members
will hold a limited shingle warranty of at least 25
years, that common factual circumstance is dwarfed by
the diverse terms, limitations, and kinds of warranties
held by the class members, which can be critical to the
validity of their legal claims.
d) Factual Circumstances - Design Defect: In contrast, the
named plaintiffs’ factual circumstances with respect to
the alleged design defect are generally the same as those
of the members of the proposed four-state class.
(i) Regardless of the particular design defect
theory that the named plaintiffs pursue, the
factual circumstances of the named plaintiffs
and the members of the proposed four-state
112
class are generally the same. FOF 200, 202-04
(identifying three possible design defects:
insufficient quantity of asphalt, insufficient
quality of asphalt, and low mat mass/tear
strength). Whether or not Oakridge-brand
shingles suffer from any of these purported
design defects will be proven by the same
evidence. There are no facts unique to any
named plaintiff or class member that will affect
how Owens Corning designed its shingles.
(ii) For purposes of analyzing the typicality factor,
the court uncritically accepts the named
plaintiffs’ theory that all Oakridge-brand
shingles are defectively designed because
Owens Corning’s design specifications “allow”
both the manufacture of shingles that will last
only 20 years (low-end shingles) and shingles
that will last more than 30 years (high-end
shingles). But see COL 73 (discussing
plaintiffs’ failure to establish that a common
design defect exists where a portion of
manufactured products will admittedly not be
defective and no evidence with respect to the
113
frequency at which the manufactured product
includes the defect is proffered).
(iii)Under plaintiffs’ theory of the case, this court
concludes that the named plaintiffs’ factual
circumstances with respect to the design
defects allegedly affecting Oakridge-brand
shingles are generally the same as those of the
members of the proposed four-state class.
iii. Two of the named plaintiffs, however, are subject to a defense that
is both inapplicable to many members of the class and likely to
become a major focus of the litigation, regardless of the legal
theory being advanced. Marcus, 687 F.3d at 599.
a. The Maags accepted a substantial payment, by way of
credit, from Owens Corning, agreeing in the process to
release their claims against Owens Corning. FOF 97-98.
The Maags are subject to a defense based upon this release
that will be inapplicable to many members of the proposed
four-state class, and that will likely become a major focus
of the litigation. It follows that the Maags’ claims cannot
be typical of class members’ claims under these
circumstances.
b. There is no indication that the other named plaintiffs are
subject to a unique defense that is likely to become a
114
major focus of the litigation. Although West and
Gonzalez received payment from their insurers for the
property damage that they suffered, and West’s
contractor replaced his back roof free-of-charge, such
circumstances and set-offs will likely be common to the
class and would have to be administered in arriving at a
final claim value for each class member.
iv. The interests and incentives of the named plaintiffs are not sufficiently
aligned with those of the members of the proposed four-state class.
a. Some named plaintiffs experienced property damage (Wright,
West, and Gonzalez), while others suffered no property
damage (Boehm and Maags). FOF 100. Of the three named
plaintiffs who suffered property damage, two were fully
compensated for the damage by an insurance company. FOF
100(c). There is no dispute that the property damage suffered
by the other named plaintiff was not caused by a failure of her
Oakridge-brand shingles. FOF 100(a). All these
circumstances will affect what relief the named plaintiffs seek
to obtain in order to resolve their claims. Yet, the named
plaintiffs propose to represent a class comprised of members
who have suffered out-of-pocket costs associated with
property damage, and members who have not. There is a lack
of alignment in this respect.
115
b. The named plaintiffs also purport to represent a class that will
undoubtedly include owners who replaced the shingles on
their structure as a precaution to prevent property damage.
Although two named plaintiffs replaced the shingles on their
roofs, one incurred no cost to do so, and the other received
partial reimbursement from Owens Corning under the terms of
the limited shingle warranty and pursuant to a waiver of future
claims. FOF 101. There is a lack of alignment in this regard
with respect to obtaining remuneration for costs incurred by
class members who preventatively replaced their shingles.
c. The interests and incentives of the named plaintiffs will
also be dissimilar because the named plaintiffs and the
members of the proposed four-state class obtained
warranties of differing numbers of years and purchased
shingles at different times. By way of demonstration,
Gonzalez, who holds a 20-year limited warranty on
shingles purchased in 2002, has a different interest than
Boehm, who holds a 40-year limited warranty on shingles
purchased in 1997. Both named plaintiffs hold different
interests from an owner who purchased a structure in
2010 on which Oakridge-brand shingles with a 50-year
limited warranty were already installed, or an owner who
installed Oakridge-brand shingles with a lifetime limited
116
warranty on a structure in 2012. Although it may be
conceivable that a method could be created to standardize
such disparate values, interests, and incentives, plaintiffs
offer no method for doing so in support of their motion
for class certification, and, in fact, fail to even recognize
that this situation exists. Therefore, there is a lack of
alignment in this regard.
COL 32: Plaintiffs failed to establish typicality by a preponderance of the evidence for
either the proposed nationwide class or the proposed four-state class. Upon consideration
of whether the claims of the named plaintiffs and the class members are generally the
same, whether the named plaintiffs are subject to a defense that is inapplicable to the
class and central to the litigation, and the alignment of interests and incentives, the court
finds that the named plaintiffs’ position is markedly different from that of the class as a
whole. This conclusion means that the court need not proceed further with the class
certification analysis for either proposed class under any subsection of Rule 23(b). In the
interest of completeness and to facilitate review by an appellate court, the court will,
nevertheless, proceed with the analysis of plaintiffs’ motion for class certification as
though plaintiffs met their burden to prove typicality by a preponderance of the evidence
for all proposed classes.
4. Adequacy
COL 33: Rule 23(a)’s fourth requirement is that the representative plaintiffs must “fairly
and adequately protect the interests of the class.” FED. R. CIV. P. 23(a)(4). The adequacy
requirement concerns both “the experience and performance of class counsel” and “the
117
interests and incentives of the representative plaintiffs.” Dewey v. Volkswagen
Aktiengesellschaft, 681 F.3d 170, 181 (3d Cir. 2012) (citing In re Cmty. Bank of N. Va.,
418 F.3d 277, 303 (3d Cir. 2005) (Community Bank I)).
COL 34: “ʻThe principal purpose of the adequacy requirement is to determine whether the
named plaintiffs have the ability and the incentive to vigorously represent the claims of
the class.ʼ” Community Bank III, 795 F.3d at 393 (quoting In re Community Bank of
Northern Virginia, 622 F.3d 275, 291 (3d Cir. 2010) (Community Bank II)). In fact,
“‘the linchpin of the adequacy requirement is the alignment of interests and incentives
between the representative plaintiffs and the rest of the class.’” Community Bank III, 795
F.3d at 393 (quoting Dewey, 681 F.3d at 183).
COL 35: This inquiry is closely tethered to the typicality inquiry. Danvers Motor Co., Inc.
v. FordMotor Co., 543 F.3d 141, 149 (3d Cir. 2008). The goal is to ensure that the
named plaintiff's claims “are not antagonistic to the class.” Id. at 150 (citing Beck v.
Maximus, Inc., 457 F.3d 291, 296 (3d Cir. 2006)).
COL 36: Owens Corning makes one challenge to the adequacy of both the named plaintiffs
and class counsel, i.e., they improperly engaged in claim splitting by inconsistently
choosing to seek class certification for only certain legal claims in each state. (ECF No.
176 ¶¶ 109-19.) Although not explicitly stated, Owens Corning’s argument applies only
to the proposed four-state class, and not to the proposed nationwide class. Plaintiffs, in
response, contend that the doctrine of claim splitting only applies when class
representatives prosecute claims for economic losses, but not claims for personal injuries
arising out of the same facts. (ECF No. 178 ¶¶ 38-40.)
118
a. The court rejects Owens Corning’s contention that plaintiffs and class counsel
are inadequate representatives of the proposed four-state class under the claimsplitting doctrine.
i. The Court of Appeals for the Third Circuit has not decided how the
issue of claim splitting applies in the context of class action
proceedings. In re Paulsboro Derailment Cases, No. 13-784, 2014
WL 4162790, at *15 (D.N.J. Aug. 20, 2014). The appellate court’s
only statement on the issue concerned a class representative’s
pursuit of separate claims for economic harm, while abandoning
claims for harm resulting from physical injuries. Nafar v.
Hollywood Tanning Sys., Inc., 339 F.App’x 216, 224 (3d Cir.
2009).
ii. The decisions relied upon by Owens Corning apply the doctrine of
claim splitting to class action proceedings in precisely this same
context of pursuing claims for economic harm, while abandoning
claims for harm resulting from physical injuries. (ECF No. 164 at
60-61 & n.61 (citing Pearl v. Allied Corp., 102 F.R.D. 921, 923
(E.D. Pa. 1984); Thompson v. Am. Tobacco Co. Inc., 189 F.R.D.
544, 551 (D. Minn. 1999); Martin v. Home Depot U.S.A., Inc., 225
F.R.D. 198, 203-04 (W.D. Tex. 2004) (splitting property damage
claims as well as personal injury claims); and In re MTBE Products
Liab. Litig., 209 F.R.D. 323, 339 (S.D.N.Y. 2002).)
119
iii. Owens Corning does not accuse the named plaintiffs or class
counsel of engaging in claim splitting by abandoning relief for
physical injury or property damage.
iv. There is no evidence in the record that the named plaintiffs or class
counsel are engaging in claim splitting by abandoning relief for
physical injury or property damage.
COL 37: Class counsel is qualified consistent with the requirements of Rules 23(a)(4),
(g)(1), and (g)(4). Class counsel have sufficient experience and have adequately
performed in these proceedings to date. In reaching this conclusion, the court considers
(i) the work counsel has done in identifying or investigating potential claims; (ii)
counsel’s experience in handling class actions, other complex litigation, and the kinds of
claims asserted in the action; (iii) counsel’s knowledge of the applicable law; and (iv) the
resources counsel will commit to representing the class.
a. This matter has been litigated before this court for more than five years and the
court is familiar with class counsel’s performance and experience.
b. Class counsel demonstrated experience in handling complex litigation,
knowledge of relevant legal principles, and the ability and willingness to
commit sufficient resources to this class litigation.
c. Class counsel will fairly and adequately represent the interests of the class.
COL 38: Gonzalez, Boehm, and the Maags are adequate representatives of the proposed
nationwide class, but no named plaintiff is an adequate representative of the proposed
four-state class.
120
a. The Proposed Nationwide Class: As discussed in the context of the typicality
requirement of Rule 23(a), the interests and incentives of Wright and West with
respect to the effect of Owens Corning’s Bankruptcy Order on their claims
differ from those of the proposed class. COL 30(b)(3). The remaining named
plaintiffs, however, share common interests and incentives with the members of
the proposed nationwide class with respect to ensuring that Owens Corning is
not able to assert a bankruptcy discharge defense against their claims.
b. The Proposed Four-State Class: Plaintiffs failed to establish the adequacy of the
named plaintiffs to represent the proposed four-state class by a preponderance
of the evidence for several reasons:
i. Even though the election and pursuit of differing legal causes of
action does not constitute claim splitting, as Owens Corning asserts,
the court finds that for this reason the named plaintiffs cannot be
deemed adequate representatives of all members of the proposed
four-state class.
a) Each named plaintiff is advancing a different bundle of
legal claims under a different body of state law. FOF 13.
The only legal cause of action being pursued by all six
named plaintiffs is unjust enrichment. Id. There are any
number of conflicts between the interests and incentives
of the named plaintiffs and class members in each of the
four states being represented by the proposed four-state
class.
121
(i) West, an Illinois owner, cannot be deemed an
adequate representative of owners whose
structures are located in Pennsylvania, who,
unlike him, are pursuing breach of express and
implied warranty claims. FOF 13. In turn,
West, who is pursuing consumer protection
claims on a class wide basis, is not an
adequate representative of Pennsylvania
owners who are not pursuing such claims.
FOF 13.
(ii) Gonzalez will have no incentive to pursue
breach of express warranty claims, which are
only being pursued by Wright on behalf of
Pennsylvania residents. FOF 13. In fact,
Gonzalez may be disincentivized to pursue
such claims in favor of the equitable unjust
enrichment claim and the implied warranty
claim that he is pursuing under Texas law.
(iii)The Maags and West have no interest in
pursuing any warranty claims, as they are
pressing only consumer protection and unjust
enrichment claims on behalf of the Illinois
owners. FOF 13. They, in fact, might be
122
incentivized to renounce any warranty claims
so that they can obtain equitable relief instead.
(iv) Courts have deemed this lack of alignment to
prevent the adequacy requirement from being
satisfied. Martin, 292 F.R.D. at 270.
(v) Although state-specific subclasses could
possibly remedy this lack of alignment, due to
the named plaintiffs’ failure to satisfy other
Rule 23(a) and Rule 23(b) requirements, the
crafting of state-specific subclasses would not
ultimately alter the court’s decision with
respect to the propriety of certifying the class.
ii. The court additionally concludes that the Maags cannot be adequate
representatives of any member of the proposed four-state class
because they accepted an offer of settlement from Owens Corning
with respect to their claims. FOF 97-98; COL 31(b)(3). This
circumstance creates inherent conflicts, instead of alignment,
between the Maags and the members of the proposed four-state
class who did not accept an offer of settlement and, in the course of
doing so, agree to release his or her claims against Owens Corning.
COL 39: Even though class counsel are adequate, the named plaintiffs failed to establish
that they are adequate representatives of the proposed four-state class. Gonzalez, Boehm,
and the Maags are, however, adequate representatives of the proposed nationwide class.
123
In the interest of completeness and to facilitate review by an appellate court, the court
will, nevertheless, proceed with the analysis of plaintiffs’ motion for class certification as
though plaintiffs met their burden to prove adequacy by a preponderance of the evidence
for all proposed classes.
C. Rule 23(b) Requirements
1. Ascertainability
COL 40: “It is the Plaintiffs’ burden to show by a preponderance of the evidence that the
class is currently and readily ascertainable.” Community Bank III, 795 F.3d at 392 (citing
Carrera v. Bayer Corp., 727 F.3d 300, 305 (3d Cir. 2013)).
COL 41: The ascertainability requirement does not apply to a Rule 23(b)(2) class, but does
apply to a Rule 23(b)(1)(B) class and a Rule 23(b)(3) class. Shelton v. Bledsoe, 775 F.3d
554, 559-62 (3d Cir. 2015).
COL 42: The four requirements of Rule 23(a) are separate and distinct from the
ascertainability inquiry. Byrd v. Aaron’s, Inc., 784 F.3d 154, 172 (3d Cir. 2015).
COL 43: Ascertainability ensures that a proposed class will function as a class, and focuses
on whether individuals fitting the class definition may be identified without resort to
mini-trials. Byrd, 784 F.3d at 162, 164-65. “If class members are impossible to identify
without extensive and individualized fact-finding or ‘mini-trials,’ then a class action is
inappropriate.” Marcus, 687 F.3d at 593; see also William B. Rubenstein, Newberg on
Class Actions § 3:3 (5th ed. 2011) (class members should be ascertainable through a
manageable process that does not require much, if any, individual factual inquiry).
COL 44: In order to demonstrate ascertainability, a plaintiff is required to show that: (1) the
class is defined with reference to objective criteria; and (2) there is a reliable and
124
administratively feasible mechanism for determining whether putative class members fall
within the class definition. Byrd, 784 F.3d at 163 (citing Carrera, 727 F.3d at 355); see
also Hayes, 725 F.3d at 355; Marcus, 687 F.3d at 598.
COL 45: Ascertainability is an essential requirement for class certification because it
“allow[s] potential class members to identify themselves for purposes of opting out of a
class,” and “ensures that a defendant’s rights are protected by the class action
mechanism.” Carrera, 727 F.3d at 307; Marcus, 687 F.3d at 591-92 (“Clearly delineating
the contours of the class . . . serves several important purposes, such as providing the
parties with clarity and assisting class members in understanding their rights and making
informed opt-out decisions.”).
COL 46: Before proceeding with the analysis, the court notes that plaintiffs offer no
briefing and submit no proposed findings of fact or conclusions of law directed to the
ascertainability requirement for the proposed nationwide class. (ECF No. 154 at 24-25;
ECF No. 178 at 19-21 (¶¶ 5-8); ECF No. 170 at 4-5.) Given that it is plaintiffs’ burden to
establish that the nationwide class is ascertainable, the court could find on this basis alone
that the proposed nationwide class cannot be certified. The court will, nevertheless,
proceed with the analysis, applying the arguments that plaintiffs make in connection with
the proposed four-state class where possible, and otherwise extrapolating from the record
the pertinent information.
a. Objective Criteria
COL 47: The Proposed Nationwide Class: The proposed nationwide class is defined by the
following criteria: (a) ownership of a structure located in the United States (b) on which
125
Oakridge-brand shingles are currently installed, (c) that were purchased before
September 26, 2006. FOF 6. These are all objective criteria.
COL 48: The Proposed Four-State Class: The proposed four-state class is defined by the
following criteria: (a) ownership of a structure in Pennsylvania, Illinois, California, or
Texas (b) on which Oakridge-brand shingles were installed between 1992 and 2012 (c)
where those shingles manifested any cracking, degranulation, fragmentation, or
deterioration (d) during the warranty coverage period. FOF 6. The court finds, for the
following reasons, that the degranulation and deterioration elements of the class
definition are not objective criteria. With those elements excised, the court would find
that the proposed four-state class definition is based upon objective criteria.
a. Ownership of a structure located in one of four states, the date of installation,
and the length of the warranty period are all objective criteria.
b. Degranulation and deterioration, which are included in the proposed class
definition “manifestation of any cracking, degranulation, fragmentation, or
deterioration,” are not based upon objective criteria. Neither party addresses
this issue in its papers.
i. A crack is an objective criteria; either the shingle has a crack in it,
or it does not. Although the record reflects that there are different
kinds of cracks - surface cracks, deep cracks, vertical cracks, spiderweb-like cracks - and that not all cracks indicate imminent shingle
failure and not all cracks can be attributed to faulty shingles, the
proposed four-state class definition requires only that “any crack”
126
be manifest. FOF 198-99. A crack is an observable, objective
criteria.
ii. To the extent that fragmentation is understood to mean that a piece
of a shingle has been torn away or otherwise separated from the
remainder of the shingle, fragmentation is an objective criteria;
either the entire shingle is intact, or part of it is gone.
iii. Degranulation and deterioration, on the other hand, are subjective
terms of degree which, by virtue of the prefix de-, indicate a change
in condition and require a comparison to some other specimen.
This subjectiveness is not a matter of semantics.
a) Consider an owner, who after receiving notice of this
class action, climbs on his roof (or engages another
person to do so for him) to view his shingles in an effort
to decide whether he falls within the class definition. No
shingles are cracked. All the shingles are intact. How is
the owner to determine if his shingles are “degranulated”
or “deteriorated?”
(i) To what are the shingles on the roof to be
compared? A new shingle? Other shingles of
the same age that have been in use in the same
geographic area? Photographs of shingles
manufactured in the same year and from the
same plant as the owner’s shingles? The
127
record reflects that shingles manufactured in
1992 will have different qualities and use
different materials than shingles manufactured
in 2012. FOF 104, 106, 108(b), 203(c)(ii)(c).
The record reflects that a shingle
manufactured for use in California will have
different qualities and use different materials
than shingles manufactured for use in
Chicago, Illinois. FOF 104, 106, 108(b),
203(c)(ii)(c). Assessing whether a shingle
manifests “degranulation” or “deterioration” is
not a one-size-fits-all exercise.
(ii) How is the owner to determine what amount
of “degranulation” and “deterioration”
qualifies him to become a member of the
proposed four-state class? Even plaintiffs’
expert acknowledges that all shingles
experience ordinary and acceptable wear and
tear. FOF 196. How much degranulation and
deterioration represents acceptable wear and
tear and how much represents excessive wear
and tear? This determination will depend on
various factors such as the age of the shingle,
128
the exact product installed on the roof, and the
environment in which the shingle has been
weathering. Acceptable wear and tear for a
shingle sold with a 25-year limited warranty
that was installed on a roof 23 years ago will
undoubtedly be different than acceptable wear
and tear for a shingle sold with a lifetime
limited warranty that was installed on a roof 5
years ago. None of these are objective criteria
and plaintiffs propose no way for class
members to make those assessments in order
to evaluate their qualifications for class
membership.
(iii)Must the owner remove samples from the roof
and submit them for testing to determine
whether the shingles have suffered sufficiently
excessive degranulation and deterioration to
qualify as a member of the proposed four-state
class? Apart from this approach raising
serious concerns with respect to administrative
feasibility, the record does not contain a single
set of measurements, for all shingles of all
ages installed in all four states over a period of
129
twenty years that will indicate excessive
degranulation or deterioration. FOF 209; COL
26(b).
b) For all these reasons, the degranulation and deterioration
criteria are not objective. Unless those two criteria are
excised from the class definition, the proposed four-state
class is not ascertainable. In the interest of completeness
and to facilitate review by an appellate court, the court will,
nevertheless, proceed with the analysis of plaintiffs’ motion
for class certification as though the degranulation and
deterioration elements were excised from the proposed
four-state class definition and the class was defined entirely
by reference to objective criteria.
b. Administrative Feasibility
COL 49: The proposed nationwide and four-state classes share several criteria, such as
ownership of a structure with Oakridge-brand shingles installed on it, and either the
purchase or installation date of those shingles. The court will assess whether there is a
reliable and administratively feasible mechanism for determining whether putative class
members fall within the proposed nationwide class and the four-state class definitions
together, where possible, in the conclusions of law that follow. Where necessary, the
administrative feasibility of determining whether members fall within the two kinds of
proposed classes will be separately addressed.
130
COL 50: Ownership of a Structure (both classes): There is a reliable and administratively
feasible mechanism for determining whether a putative class member owns a structure.
Tax records, property records, deeds, and mortgages are just a few examples of the kinds
of documentation that class members could produce, if deemed necessary, to prove
ownership of a structure.
COL 51: Identifying Oakridge-brand Shingles (both classes): Plaintiffs contend that
Oakridge-brand shingles can be identified by (1) their mass, color, and dimensions; (2)
release tape; and (3) receipts or invoices. (ECF No. 178 at 20 (¶ 8).) According to
plaintiffs, Owens Corning’s administration of a warranty program is evidence that
identification of Oakridge-brand shingles is administratively feasible. (Id.) Identification
of Oakridge-brand shingles is not administratively feasible on a classwide basis for either
the proposed nationwide class or the proposed four-state class for the following reasons:
a. Class members whose structures have Oakridge-brand shingles installed on
them cannot be determined from Owens Corning’s corporate records. FOF
117.
b. Class members whose structures have Oakridge-brand shingles installed on
them cannot be determined by visual inspection, except by only a few Owens
Corning employees. FOF 225.
c. Class members whose structures have Oakridge-brand shingles installed on
them cannot be determined by release tape. FOF 221-23. At most, the release
tape will indicate that the shingle was manufactured at a plant that produces
Oakridge-brand shingles. It does not necessarily follow that the shingle is an
Oakridge-brand shingle.
131
d. Class members whose structures have Oakridge-brand shingles installed on
them cannot be determined by reference to receipts or invoices.
i. The record reflects that even the named plaintiffs could not produce
receipts or invoices establishing their purchase of Oakridge-brand
shingles. FOF 28, 54, 84, 92.
ii. Given that the proposed nationwide class spans nearly fifteen years
and the proposed four-state class spans twenty years, it is likely that
ownership of the structure will have changed, making
documentation for roofing projects completed prior to transfer less
likely to be available.
e. Owens Corning’s maintenance and management of a warranty program do not
establish that structures on which Oakridge-brand shingles are installed can be
identified in an administratively feasible way for either the proposed
nationwide class or the proposed four-state class.
i. Owens Corning adjudicates warranty claims on a fact-specific,
case-by-case basis. FOF 115. The warranty claims process
includes an exchange of information between the owner and
Owens Corning, and could include submission of sample shingles
and on-site inspections. Id.
ii. The warranty program accounts for a miniscule percentage of
Owens Corning’s production of Oakridge-brand shingles. FOF
112. Owens Corning’s administration of 5 claims does not mean
132
that 1,000 claims can be feasibly administered on a classwide
basis. Id.
iii. Owens Corning occasionally denies warranty claims because the
claimant does not own Owens Corning shingles. FOF 114.
COL 52: Establishing Purchase or Installation Date (both classes): The ability to establish
the purchase date, for the proposed nationwide class, and the installation date, for the
proposed four-state class, is not administratively feasible for the same reasons set forth
immediately above with respect to “identifying Oakridge-brand shingles.” COL 51(d). If
an owner cannot identify the shingles as Oakridge-brand shingles, it follows that it is
unlikely that the owner can establish the purchase or installation date with documentation
or other evidence. Again, transfer in the ownership of the structure further complicates
administration of this criterion. Plaintiffs proffer no reliable or administratively feasible
mechanism for determining a putative class member’s purchase or installation date of
Oakridge-brand shingles.
COL 53: Establishing Manifestation of Shingle Conditions (the four-state class): Plaintiffs
contend that it is administratively feasible to determine the manifestation of any cracking,
degranulation, fragmentation, or deterioration because “[i]f an objective inspection
indicates even a minimal manifestation of any of these defects” the owner will qualify as
a member of the class. (ECF N. 178 at 20 (¶ 8(c)).) Establishing manifestation of one of
the four enumerated shingle conditions, however, is not administratively feasible on a
classwide basis for the proposed four-state class for the following reasons:
133
a. Plaintiffs, in their argument, recognize that every roof of every potential class
member will have to be visually examined in order to determine whether “any
cracking, degranulation, fragmentation, or deterioration” has occurred.
i. Putting aside for the moment the issues raised earlier about how an
owner is to assess whether a shingle is sufficiently “degranulated”
or “deteriorated” to qualify for class membership, COL 48, the need
for an individual examination and subjective assessment of each
potential class member’s shingles signifies a lack of administrative
feasibility for the proposed four-state class.
ii. Plaintiffs propose no standard procedure pursuant to which an
owner can determine whether one of the enumerated shingle
conditions exist without personally inspecting, or engaging
someone else to inspect, the shingles.
b. Individual factual inquiries will be required to determine whether each
potential class member’s Oakridge-brand shingles manifest any cracking,
degranulation, fragmentation, or deterioration. In fact, if this factual inquiry is
not made, the class definition would be effectively reduced to include any
owner of a structure on which Oakridge-brand shingles were installed during
the proposed 20-year class period, regardless of the physical condition of the
shingles. That kind of class definition although eliminating the present
concerns about administrative feasibility would be impermissibly overinclusive and would not align with plaintiffs’ legal theories and record
evidence. There is no evidence to support certification of a class comprised of
134
all owners who own a structure on which Oakridge-brand shingles were
installed between 1992 and 2012; plaintiffs do not seek certification of such a
class.
COL 54: In summary, both the proposed nationwide class and the proposed four-state class
are defined by reference to objective criteria, if degranulation and deterioration are
removed from the four-state class definition. Plaintiffs fail to demonstrate, however, that
membership in either proposed class can be determined by a reliable and administratively
feasible mechanism.
2.
Rule 23(b)(1)(B) – The Nationwide Bankruptcy Discharge Class
COL 55: Federal Rule of Civil Procedure 23(b)(1)(B) provides that an action may be
maintained as a class action if the prerequisites of subdivision 23(a) are satisfied and
“[t]he prosecution of separate actions by or against individual members of the class
would create a risk of … adjudications with respect to individual members of the class
which would as a practical matter be dispositive of the interests of the other members not
parties to the adjudications or substantially impair or impede their ability to protect their
interests.” FED. R. CIV. P. 23(b)(1)(B).
COL 56: Because class members are not provided notice or an opportunity to opt out of a
class certified under Rule 23(b)(1)(B), the rule is narrowly interpreted. Ortiz v.
Fibreboard Corp., 527 U.S. 815, 844 (1999); Patton v. Topps Meat Co., LLC, No. 0700654, 2010 WL 9432381, at *6 (W.D.N.Y. May 27, 2010).
COL 57: Although Rule 23(b)(1)(B) has been used in labor relations cases, certain ERISA
suits, and suits in which inmates seek injunctive relief, the traditional and most common
use of the rule is in "limited fund" cases. 1 McLaughlin on Class Actions § 5:8 (12th ed.)
135
In those cases, the rule protects plaintiffs where separate lawsuits might exhaust a
defendant's resources, such that earlier plaintiffs might recover to the prejudice or
exclusion of later plaintiffs. Id. In a limited fund class action, the court's adjudicative
authority derives from its jurisdiction over the fund, and the concomitant power to decide
the claims of all entities asserting rights to the fund. Id.
COL 58: Rule 23(b)(1) does not expressly require a significant degree of similarity among
the factual circumstances of putative class members, but courts have interpreted the Rule
to include that requirement. Id.
COL 59: Although plaintiffs contend that Rule 23(b)(1)(B) applies outside the limited fund
context and where “a statutory defense will impair the rights of all class members,” the
decisions cited by plaintiffs in support of these points are limited fund or ERISA breach
of fiduciary duty cases. (ECF No. 178 at 27 (¶¶ 43-44).) Plaintiffs cite to no legal
authority in which a court has certified a class pursuant to Rule 23(b)(1)(B) in order to
adjudicate the effect of a bankruptcy court’s order of discharge on claims for damages
asserted by claimants after a reorganization plan was confirmed. The court found no
such legal authority.
COL 60: Putting aside this lack of legal authority in connection with plaintiffs’ proposed
Rule 23(b)(1)(B) class, the court examines whether certification of the proposed
nationwide class could nevertheless be appropriate.
COL 61: It remains unclear, even after oral argument and the submission of proposed
findings of fact and conclusions of law, what relief plaintiffs seek on behalf of the
proposed Rule 23(b)(1)(B) class members. COL 22. Because, however, there is no relief
136
that plaintiffs could obtain on behalf of the proposed nationwide class under any possible
theory, this deficiency in plaintiffs’ submissions is ultimately without consequence.
a. At oral argument, the named plaintiffs claimed that they sought a ruling, on
behalf of the proposed nationwide class, with respect to whether the Frenville
test or the Grossman’s test applies to any challenge made by Owens Corning to
class members’ claims on the ground that those claims were discharged by the
Bankruptcy Order. FOF 22(a).
i. The Court of Appeals for the Third Circuit issued a precedential
opinion in Wright deciding this precise issue.
a) The appellate court, applying Third Circuit law to the
claims of owners located in both Pennsylvania and
Illinois, held that the Frenville test applies to any
dischargeability defense raised by Owens Corning
because the Bankruptcy Order was entered in 2006,
which is before Grossman’s was decided. FOF 230-36.
b) Owens Corning concedes that it cannot relitigate this
issue. FOF 233; COL 22(a)(ii). The doctrine of collateral
estoppel will prevent it from doing so. COL 22(a)(v).
c) Plaintiffs recognize that the effect of Owens Corning’s
Bankruptcy Order will be determined in accordance with
the law of the Court of Appeals for the Third Circuit
because the United States Bankruptcy Court for the
District of Delaware issued the Bankruptcy Order and
137
that court, and the federal district court in Delaware,
retain jurisdiction under the Bankruptcy Order over all
matters concerning Owens Corning’s bankruptcy
proceedings. FOF 234-35; (ECF No. 178 at 28 (¶ 47).)
ii. The question that plaintiffs ask this court to answer on behalf of the
proposed nationwide class is moot in light of the court of appeals’
decision in Wright and cannot be adjudicated by this court. Neale,
794 F.3d at 366 (“[b]efore even getting to the point of class
certification, however, class representatives need to present a
justiciable claim”).
a) Article III, Section 2 of the Constitution limits the
“judicial power” of the United States to the resolution of
certain “cases” and “controversies.” Valley Forge, 454
U.S. at 471. The “case” or “controversy” requirement is
enforced through a number of justiciability doctrines,
which include standing, ripeness, mootness, and the
prohibition on advisory opinions. Toll Bros., 555 F.3d at
137.
b) A case can become moot when the issues presented are
no longer ‘live’ or the parties lack a legally cognizable
interest in the outcome. Powell, 395 U.S. at 496.
c) The supplemental authority submitted by plaintiffs is
inapposite. (ECF No. 179 (citing Campbell-Ewald Co. v.
138
Gomez, 136 S.Ct. 663 (2016)).) In Gomez, the Supreme
Court held that the parties remained adverse, and the case
and controversy remained live, after a settlement offer
was rejected. Gomez, 136 S.Ct. at 669-72. The facts of
the instant case are distinguishable. Gomez has no
bearing on this case.
b. To the extent that the named plaintiffs actually seek a ruling, on behalf of the
proposed nationwide class, that the claims of any owner who purchased
Oakridge-brand shingles before September 26, 2006, are not discharged by
Owens Corning’s Bankruptcy Order, the court of appeals’ decision in Wright
specifically precludes this relief. COL 22(b). This court cannot enter any order
that would contradict the appellate court’s rulings.
i. Under the Frenville test, a court must make a fact-specific inquiry
to determine how Owens Corning’s Bankruptcy Order affects an
owner’s claim. FOF 2, 232; COL 22(b)(i)(b), 30(b). The legal
claim being asserted, the state law being applied, and facts such as
when the shingles exhibited failure, whether a warranty claim was
made, when property damage was suffered, and when roofing
repairs were made all affect how the test applies to a particular
claim. The Frenville test, by its very nature, cannot be applied
uniformly on a classwide basis to the claims of owners throughout
the country.
139
ii. The court of appeals, in Wright, held that the claims of Wright and
West were not discharged by virtue of the Bankruptcy Order, but
did so only after assessing the facts surrounding Wright’s claim and
West’s claim to determine when those claims accrued under state
law, and what notice Wright and West received about Owens
Corning’s bankruptcy proceedings and what effect that notice had
on their due process rights. FOF 232; Wright, 679 F.3d at 105-09.
This kind of assessment cannot be made on a classwide basis for
claims held by owners throughout the country.
iii. The court of appeals’ decision in Wright precludes this court from
making any classwide rulings that claims are, or are not, discharged
by Owens Corning’s Bankruptcy Order. The nature of the Frenville
test precludes such relief.
COL 62: This court will not certify a nationwide class pursuant to Rule 23(b)(1)(B)
because the relief being sought is either moot or must be obtained on an individual, rather
than a classwide, basis.
2. Rule 23(b)(3) – The Monetary Damages Class
COL 63: A class action can be certified under Rule 23(b)(3) where the court finds that the
questions of law or fact common to class members predominate over any questions
affecting only individual members, and that a class action is superior to other available
methods for fairly and efficiently adjudicating the controversy. FED. R. CIV. P. 23(b)(3).
COL 64: The rule provides that the following matters are pertinent to these findings: (1) the
class members' interests in individually controlling the prosecution or defense of separate
140
actions; (2) the extent and nature of any litigation concerning the controversy already
begun by or against class members; (3) the desirability or undesirability of concentrating
the litigation of the claims in the particular forum; and (4) the likely difficulties in
managing a class action. Id.
a. Predominance
COL 65: The predominance requirement of Rule 23(b)(3) imposes a “far more demanding”
standard than the commonality requirement of Rule 23(a). Community Bank III, 795 F.3d
at 399 (citing Amchem Products, Inc. v. Windsor, 521 U.S. 591, 623-24 (1997)).
COL 66: The predominance criterion “tests whether proposed classes are sufficiently
cohesive to warrant adjudication by representation.” Sullivan, 667 F.3d at 296 (citing In
re Ins. Broker Antitrust Litig., 579 F.3d 241, 266 (3d Cir. 2009)).
COL 67: In assessing predominance, a court “must examine each element of a legal claim
‘through the prism’ of Rule 23(b).” Marcus, 687 F.3d at 600; see Community Bank III,
795 F.3d at 399-400.
COL 68: It is the plaintiff’s burden to demonstrate that the elements of each legal claim are
capable of proof at trial through evidence that is common to the class rather than
individual to its members. Marcus, 687 F.3d at 600 (citing In re Hydrogen Peroxide, 552
F.3d at 311).
COL 69: “‘Because the nature of the evidence that will suffice to resolve a question
determines whether the question is common or individual, a district court must formulate
some prediction as to how specific issues will play out in order to determine whether
common or individual issues predominate in a given case.’” Marcus, 687 F.3d at 600
(quoting In re Hydrogen Peroxide, 552 F.3d at 311).
141
COL 70: Rule 23(b)(3) does not require a plaintiff seeking class certification to prove that
each and every element of her claim is susceptible to classwide proof. Amgen Inc. v.
Conn. Ret. Plans and Trust Funds, 133 S. Ct. 1184, 1196 (2013). “What the rule does
require is that common questions ‘predominate over any questions affecting only
individual [class] members.’” Id. (quoting FED. R. CIV. P. 23(b)(3) (alteration and
emphasis in the original)).
COL 71: The named plaintiffs seek to pursue four categories of legal claims on behalf of
the proposed four-state class: (1) breach of express warranty (PA); (2) breach of implied
warranty of merchantability (PA, CA, TX); (3) violation of state consumer protection
statutes (IL, CA, TX); and (4) unjust enrichment (PA, IL, CA, TX). FOF 13. For
purposes of the following analysis, and in order to facilitate appellate review, the court
puts aside the inherent, and dispositive, problems created by the named plaintiffs’
attempts to represent a class that will include members owning structures in four different
states, who are each advancing different combinations of legal claims under different
state laws than the named plaintiffs, which problems have been discussed previously in
these findings and conclusions. FOF 13; COL 31(a), 38(b).
COL 72: According to plaintiffs, Owens Corning engaged in a single course of wrongdoing
by manufacturing Oakridge-brand shingles in accordance with design specifications that
were defective because they made all Oakridge-brand shingles susceptible and vulnerable
to having a useful life of no more than 20 years, even though Owens Corning promised
that the shingles would have a useful life of 25 years, or more. (ECF No. 178 at 2-11 (¶¶
5-46), 46 (¶ 117).) Based upon the legal claims on which the named plaintiffs seek class
certification, it is plaintiffs’ burden to demonstrate that it can prove these accusations at
142
trial through evidence that is predominately common to the class rather than individual to
its members.
a. Under plaintiffs’ theory of the case, each and every legal cause of action that
plaintiffs have elected to pursue on behalf of the proposed four-state class
requires proof that Oakridge-brand shingles were defectively designed. (ECF
No. 178 at 35, 38, 40, 41, 43 (¶¶ 77, 87, 97, 101, 106).); Cleary v. Philip
Morris Inc., 656 F.3d 511, 516, 519 (7th Cir. 2011) (under Illinois unjust
enrichment law, “a plaintiff must allege that the defendant has unjustly retained
a benefit to the plaintiff’s detriment, and that defendant’s retention of the
benefit violates the fundamental principles of justice, equity, and good
conscience”); Altronics of Bethlehem, Inc. v. Repco, Inc., 957 F. 2d 1102,
1105 (3d Cir. 1992) (to prove breach of implied warranty of merchantability
under Pennsylvania law, “plaintiffs were required to show: (1) that the product
malfunctioned . . .”); Tietsworth v. Sears, 720 F. Supp. 2d 1123, 1142 (N.D.
Cal. 2010) (to state a claim for breach of implied warranty of merchantability
under California law, “[t]here must be a fundamental defect that renders the
product unfit for its ordinary purpose”); Webb v. UnumProvident Corp., 507
F.Supp.2d 668, 680 (W.D. Tex. 2005) (violation of Texas Deceptive Trade
Practices Act can be based upon a failure to disclose defect); Mary E. Bivins
Found. v. Highland Capital Mgmt., L.P., 451 S.W.3d 104, 112 (Tex. App.
2014) (under Texas unjust enrichment law, “[t]o prevail, a plaintiff must show
that the defendant obtained a benefit from the plaintiff by fraud, duress, or the
taking of an undue advantage. . . . To prevail, a plaintiff must show that a
143
defendant holds money which in equity and good conscience belongs to the
plaintiff.”); Peterson v. Cellco P’Ship, 164 Cal. App. 4th 1583, 1593 (2008)
(“The elements of an unjust enrichment claim are the receipt of a benefit and
[the] unjust retention of the benefit at the expense of another.”); Lackner v.
Glosser, 892 A.2d 21, 34 (Pa. Super. Ct. 2006) (under Pennsylvania law, the
elements of unjust enrichment involve “benefits conferred on defendant by
plaintiff, appreciation of such benefits by defendant, and acceptance and
retention of such benefits under such circumstances that it would be inequitable
for defendant to retain the benefit without payment of value”); Everett v. TKTaito, LLC, 178 S.W.3d 844, 853 (Tex. App. 2005) (under Texas law, a
product is unmerchantable if a product is defective when it leaves the
defendant’s possession).
b. Some of the legal causes of action require proof that Owens Corning made
affirmative misrepresentations about the useful life of Oakridge-brand shingles.
(Id. at 40 (¶ 97)); Durell v. Sharp Healthcare, 183 Cal. App. 4th 1350, 1366-67
(2010) (plaintiff in an action under California Consumer Legal Remedies Act
must show that defendant’s conduct was deceptive); Sevidal v. Target Corp.,
189 Cal. App. 4th 905, 924 (2010) (California’s Unfair Competition Law and
False Advertising Law require showing of improper conduct by defendant);
Clark v. Experian Info. Solutions, Inc., 256 F. App’x 818, 821-22 (7th Cir.
2007) (“A claim for consumer fraud under the [Illinois Consumer Fraud Act]
contains five elements: ‘(1) a deceptive act or practice by the defendant . . .’”);
Goodman v. PPG Indus., Inc., 849 A.2d 1239, 1245 (Pa. Super. Ct. 2004)
144
(express warranties are bargained, “dickered,” individualized promises that the
goods will perform up to the specific standards set forth in the warranty).
i. Under plaintiffs’ theory of this case, and based upon the record
evidence, the existence of a design defect is fundamental to even
plaintiffs’ misrepresentation-based claims. The only
misrepresentations or omissions that Owens Corning is accused of
making are that Oakridge-brand shingles will last for at least 25
years, or for the same number of years as the limited shingle
warranty, when, because of a defective design, the shingles will last
20 years, at the most. According to plaintiffs, Owens Corning’s
alleged representations are only false if its design specifications for
Oakridge-brand shingles cause them to fail before year 25.
ii. As will be detailed in the conclusions that follow, the record before
the court and the factual findings previously made by this court
render it impossible for plaintiffs to meet their burden to prove a
design defect by evidence common to the class in this case. It
therefore follows that it is likewise impossible for plaintiffs to meet
their burden to prove any misrepresentation claims by evidence
common to the class.
iii. It follows that predominance cannot be established for any of the
legal claims for which the named plaintiffs seek class certification
under Rule 23(b)(3). Under these factual circumstances, it is
unnecessary for the court to digest the myriad individual issues that
145
would arise in the context of each legal claim being asserted under
each of the four state’s laws. The analysis need not reach this level
of detail because the factual record makes it impossible for the
named plaintiffs to establish that the two essential elements upon
which all their legal claims are based, i.e., design defect and
misrepresentation, can be proven by evidence common to the class.
This inability precludes the named plaintiffs from meeting the
predominance requirement for any legal claims they seek to assert
on behalf of the proposed four-state class, regardless of the specific
elements of any particular state-law claim.
COL 73: Plaintiffs fail to demonstrate that they could prove the existence of an alleged
design defect plaguing Oakridge-brand shingles through evidence that is common to the
proposed four-state class for the entire 20-year class period.
a. As an initial matter, this court recognizes that design defect cases can be
properly certified as class actions where plaintiffs establish that a product was
manufactured according to uniform design specifications that result in a
common, classwide defect. Neale, 794 F.3d at 357 (court deemed
predominance satisfied because all legal claims would be based upon whether
Volvo’s sunroof drainage systems were defectively designed); In re Whirlpool
Corp., 722 F.3d at 847 (plaintiffs established that all Duet-brand washing
machines were built to designs with “nearly identical engineering” that differed
only in the machines’ size and aesthetics and expert witness opined that a
common defect, failure to self-clean the tub, was present in all machines
146
regardless of slight design differences); Marcus, 687 F.3d at 602-03 (expert
witness testified, after reviewing “thousands of pages of specifications” that all
tires, regardless of model or size, are substantially similar in construction and
all tires manifest the same defective characteristic, i.e., extra stiffness); Wolin,
617 F.3d at 1172 (same design defect, i.e., a geometry defect in vehicle’s
alignment, was present in each class member’s car); Pella, 606 F.3d at 391 (all
ProLine casement windows were designed to allow water to seep behind
aluminum casing, which accelerated wood rot); Martin, 292 F.R.D. at 256-59,
267 (class action involved only “second generation” “rear twist beam axle,”
which was manufactured by one company for fewer than five years); Brunson,
266 F.R.D. at 114, 119 (all trimboard suffered from same manufacturing
defects that made it rot, warp, and crack prematurely); Payne, 216 F.R.D. at 23,
28 (hosing used in radiant floor heating system was defectively designed
causing oxidation, hardening, cracks, and eventually, leaks; testing and other
evidence confirmed presence of defect in all hosing).
b. Despite pointed and extended questions from the court at oral argument, ECF
No. 173 at 91-99, plaintiffs offer no legal authority to support their seemingly
novel and illogical theory that admittedly nondefective products can,
nevertheless, be considered defectively designed if a design specification
establishes a range of measurements, some of which will produce defective
products and some of which will produce nondefective products, especially
where, as here, plaintiffs fail to identify where within the range of
measurements the design crosses the line from producing nondefective
147
products to producing defective products or to quantify how often defective
products, versus nondefective products, were produced.
c. In this case, plaintiffs offered no evidence that Owens Corning manufactured
Oakridge-brand shingles for use in Pennsylvania, Illinois, California, and
Texas for the proposed 20-year class period in accordance with a single,
defective design specification, or substantially similar or nearly identical
defective design specifications. FOF 104-05. There is likewise no evidence
that Oakridge-brand shingles contain a particular design specification
measurement that renders all (or even most or many) shingles defective. FOF
209; COL 26(b).
i. The record reflects, instead, that Owens Corning used more than
500 design specifications during only a portion of the proposed 20year class period, which specifications differed based upon the
location of the manufacturing plant and the technology and raw
materials available at the time. FOF 104, 106-08. Plaintiffs proffer
no evidence that all these specifications share a common
measurement or characteristic that renders them defective,
regardless of any differences in the specifications.
ii. The record reflects that each design specification includes not one
specific set of measurements for the individual elements of a
shingle, such as mat mass, asphalt mass, and tear strength, but
various minimum, maximum, and target measurements for the
different features of a shingle, which establish the allowable and
148
preferred parameters for manufacturing a shingle according to that
design specification. FOF 108. Every shingle will not be
manufactured to the allegedly defective low-end measurements,
which alone is fatal to class certification. Neale v. Volvo Cars of N.
Am., LLC, No. 10-4407, 2013 WL 785056, at *4 (D.N.J. Mar. 1,
2013) (at class certification stage, plaintiffs need not prove that all
products are actually defective, but must present proof that a
common, allegedly defective, design element is present in all
products, e.g., sound plugs with a plus-shaped opening at the
bottom).
a) Rutila acknowledges that the same design specification
can produce both shingles that will last the same number
of years as the length of some limited shingle warranties
(nondefective shingles) and shingles that will not last for
the length of the shortest limited shingle warranty, i.e., 25
years (defective shingles). FOF 194(e) – (g).
b) The record is devoid of any evidence quantifying how
often defective, as opposed to nondefective, shingles are
produced according to Owens Corning’s design
specifications. FOF 195. Rutila’s own testing of the
nearly 300 warranty shingles, which is inherently flawed
and biased in favor of identifying defective shingles and
has been deemed inadmissible, reveals that defective
149
shingles were produced only about half of the time. FOF
195(a)(ii); Reyes, 802 F.3d at 485 (the standard is not
whether it is mathematically or scientifically possible that
one of the telemarketing firms used by defendants did not
engage in the allegedly wrongful conduct, but whether
plaintiff established that it is more likely than not that the
telemarketing firms used by defendants engaged in the
allegedly wrongful conduct).
c) Rutila acknowledges that a shingle would have to be
individually inspected to determine whether it is a
nondefective or defective shingle. FOF 194(b).
iii. There is no evidence in the record of any particular measurement or
set of measurements for the individual elements of a shingle that
constitute a design defect.
a) The only evidence of this nature is Rutila’s testimony that
a shingle with a mat mass of at least 1.5 pounds-per-100square foot will produce a longer-lasting shingle. FOF
194(c)(i), 204(b). Even this evidence, however, must be
viewed in light of Rutila’s other testimony that it is the
way in which the measurements for each element of a
shingle correspond to each other that affects a shingle’s
quality. FOF 194(c)(ii). According to Rutila, a single
150
measurement, in isolation, cannot be probative of a
design defect.
b) Although Rutila opines that shingles manufactured at or
near the minimums of Owens Corning’s design
specifications will be defective because they will not last
more than 20 years, Rutila does not articulate how “near”
to the minimums a shingle must be in order to be
considered defective. FOF 194(c), 209.
d. In this case, although all the legal claims that the named plaintiffs seek to
pursue on behalf of the proposed four-state class require proof of a design
defect, the evidence of record contradicts the conclusion that all (or even most
or many) Oakridge-brand shingles manufactured for use in Pennsylvania,
Illinois, California, and Texas between 1992 and 2012 contain any of the three
design defects identified.
i. The evidence proffered by plaintiffs fails to substantiate their theory
that Oakridge-brand shingles are defectively designed because
Owens Corning’s design specifications call for an insufficient
quantity of asphalt. FOF 202, 217.
ii. The evidence proffered by plaintiffs fails to substantiate their theory
that a design specification that sets filler percentage at, or above,
66% is defective. FOF 203, 218.
iii. The evidence proffered by plaintiffs fails to substantiate their theory
that Oakridge-brand shingles are defectively designed because
151
Owens Corning’s design specifications call for insufficient mat
mass and tear strength. FOF 204, 219-20.
COL 74: Although the inability to prove the existence of a design defect by evidence that is
predominantly common to the class is fatal to plaintiffs’ misrepresentation-based legal
claims, COL 72, the court nevertheless, for the sake of completeness, will engage in a
predominance analysis of these claims. Plaintiffs fail to demonstrate that Owens
Corning’s alleged misrepresentations about Oakridge-brand shingles are capable of proof
at trial through evidence that is common to the proposed four-state class for the entire 20year class period.
a. The record reflects that Owens Corning did not engage in a uniform and
consistent marketing campaign that included representations that Oakridgebrand shingles would have a useful life of at least 25 years, or would not crack,
degranulate, fragment, or deteriorate for at least 25 years. COL 25, 31(b). The
record reflects, instead, that even the named plaintiffs were exposed to
different promotional materials and oral representations from third parties.
FOF 140-46; COL 25.
b. Owens Corning’s limited shingle warranties do not qualify as representations
that Oakridge-brand shingles would have a useful life of at least 25 years, or
for the same number of years as the length of the limited warranty. FOF 17074.
i. At the hearing on plaintiffs’ motion for class certification, the court
questioned plaintiffs about their failure to proffer legal authority to
152
support their theory that a warranty is a representation about the
expected lifespan of a product. (ECF No. 173 at 39-41.)
ii. This court noted that the court presiding over an MDL proceeding
involving IKO-brand roofing shingles rejected the precise legal
argument being made by plaintiffs in this case about a limited warranty
equating to a representation about longevity or durability. The court
presiding over that MDL commented on more than one occasion that a
“limited warranty itself is not an affirmation of fact about the useful life
of [] shingles” but instead “simply means that if the product fails within
the warranty period, there may be recourse under the warranty, but does
not lead consumers to reasonably believe that the shingles will last for
the duration of the warranty.” In re IKO Roofing Shingles Prod. Liab.
Litig., No. 2:09–md–2104 (C.D. Ill. Jan. 28, 2014) (ECF No. 338 at 3536 (citing 1/28/2014, 4/12/2013, and 4/15/2013 decisions)); see Brooks,
301 F.R.D. at 233 (recognizing, but finding inapplicable, the IKO MDL
court’s findings); In re HardiePlank Fiber Cement Siding, 2014 WL
2987657, at *3 (“An advertisement's reference to a formal limited
warranty does not, on its own, create a new informal promise that the
product will last for a certain amount of time.”)
iii. Although plaintiffs represented at the hearing on their motion for class
certification that the IKO MDL court’s rulings on this point were
criticized and reversed by the Court of Appeals for the Seventh Circuit,
they were not. (ECF No. 173 at 39-40, 45-46, 100.) While one court of
153
appeals judge made remarks during oral argument, without supporting
legal precedent, that questioned the IKO MDL court’s conclusion about
the relationship between a warranty and the useful life of a product,
(ECF No. 173 at 100), that issue was not the basis for the appellate
court’s reversal. The court of appeals reversed the IKO MDL court’s
order refusing to certify the class because that the court mistakenly
imposed a “commonality of damages” requirement at the class
certification stage. In re IKO Roofing Shingle, 757 F.3d at 603.
Notably, in the IKO MDL cases, the central theory advanced by the
named plaintiffs was that IKO falsely told customers that its shingles
met an ASTM industry standard, when they did not. Id. at 599, 603
(summarizing damages theories as “the difference in market price
between a [shingle] as represented and a [shingle] that does not satisfy
the [ASTM] standard” and “damages, if nonconformity to the [ASTM]
standard caused [shingle] failure”). In this case, plaintiffs’ theories are
significantly different than those raised in the IKO MDL cases,
especially because the parties agree that all Oakridge-brand shingles
meet relevant ASTM industry standards. FOF 109, 203(b)(i), 204(d);
COL 77(b)(ii)(b)(iv).
iv. Despite this line of inquiry from this court at the class certification
hearing, and plaintiffs’ contention that the IKO MDL court’s findings
were rejected by the court of appeals, plaintiffs’ proposed findings of
fact and conclusions of law still include no citation to any legal authority
154
to support their foundational contention that a limited warranty of a set
number of years is a representation about the useful life of a product.
(ECF No. 173 at 39-40, 45-46, 100.)
v. Plaintiffs’ theory is not only novel and unsupported, but also is contrary
to law. Rasmussen, 27 F. Supp. 3d at 1035 (the purpose of a warranty is
to contractually mark the point in time when the risk of paying for
repairs shifts from the manufacturer to the consumer); Keegan, 838 F.
Supp. 2d at 940-41 (same); Sears, Roebuck, 833 F. Supp. 2d at 899
(stating that under Texas law, an explicit statement must be made about
the useful life of a product in order to assert warranty coverage on that
basis).
c. Owens Corning’s limited shingle warranties are not consistently understood as
representations that Oakridge-brand shingles would have a useful life of at
least 25 years, or for the same number of years as the length of the limited
warranty. FOF 160-65.
d. There is no other evidence in the record probative of plaintiffs’ assertion that
Owens Corning made consistent representations to all (or even most or many)
members of the proposed four-state class that Oakridge-brand shingles would
have a useful life of at least 25 years.
e. The record contains no evidence with respect to what proportion of members
of the proposed four-state class were exposed to Owens Corning’s
representations about Oakridge-brand shingles, which is a relevant deficiency
155
given that the proposed four-state class includes owners who did not select
Oakridge-brand shingles for the structure. FOF 17.
COL 75: Even if plaintiffs could establish the two essential elements underlying each of
their legal claims. i.e., design defect and misrepresentations, by evidence common to the
class, the court would nevertheless conclude that plaintiffs could not satisfy the
predominance requirement because individual questions with respect to injury, causation,
and damages will predominate over these two common issues.
COL 76: Plaintiffs must prove, for each kind of legal claim being asserted on behalf of the
class, that injury was suffered. Reese v. Ford Motor Co., 499 F. App’x 163, 167 (3d Cir.
2012) (Under Pennsylvania law, plaintiff's injury must be caused by defect in product in
order to recover for breach of implied warranty of merchantability); Cleary, 656 F.3d at
516, 519 (Illinois unjust enrichment plaintiff “must allege that the defendant has unjustly
retained a benefit to the plaintiff's detriment”); Clark, 256 F. App’x at 821-22 (Illinois
Consumer Fraud Act plaintiffs must establish injury); Mary E. Bivins, 451 S.W.3d at 112
(“To prevail, a plaintiff must show that a defendant holds money which in equity and
good conscience belongs to the plaintiff.”); Durell, 183 Cal. App. 4th at 1366-67 (relief
for violation of California Legal Rights Act limited to those that are injured); Pfizer Inc.
v. Superior Ct., 182 Cal. App. 4th 622, 631-32 (2010) (“one who . . . could not possibly
have lost money or property as a result of the unfair competition is not entitled to
restitution” under California’s Unfair Competition Law or False Advertising Law);
Cardinal Health 301, Inc. v. Tyco Elecs. Corp., 169 Cal. App. 4th 116, 145 (2008) (“To
recover on a breach of warranty cause of action, the plaintiff must show the breach
caused the plaintiff to suffer injury, damage, loss[,] or harm.”) (internal quotation
156
omitted); Peterson, 164 Cal. App. 4th at 1594 (a plaintiff must establish “actual injury to
bring an unjust enrichment claim”); Gen. Motors Corp. v. Garza, 179 S.W.3d 76, 81
(Tex. App. 2005) (“If a product is shown to be unmerchantable, a plaintiff must then
establish that the defect caused him to suffer injury.”) (internal quotation omitted); Wall
v. Parkway Chevrolet, Inc., 176 S.W.3d 98, 105 (Tex. App. 2004) (Texas Deceptive
Trade Practices Act plaintiffs must prove “economic or mental-anguish damages”);
Lewis v. Bayer AG, 70 Pa. D. & C. 4th 52, 86-87 (Pa. Ct. Com. Pl. 2004) (“If an
individual . . . suffers no injury, no equitable claim for unjust enrichment can lie.”).
COL 77: Although plaintiffs acknowledge that injury is a prerequisite to relief for at least
some legal claims, plaintiffs fail to specifically address how they intend to prove injury
on a classwide basis. (ECF No. 178 at 39 (¶ 91) and 40 (¶ 96) (consumer protection
statutes require showing of loss or harm to consumer) and at 42 (¶ 104) (unjust
enrichment requires showing that retention of the benefit is unjust).) Under the
circumstances, plaintiffs failed to meet their burden to establish that injury can be proven
by evidence that is common to the proposed four-state class.
a. Plaintiffs generally contend that certification is proper because “observable
injury” is included in the definition of the class. (ECF No. 178 at 33 (¶ 69).)
This is presumably a reference to the requirement that class members establish
the presence of “any cracking, degranulation, fragmentation, or deterioration”
in order to qualify as a member of the proposed four-state class. The record,
however, fails to establish that the presence of one of those conditions in a
shingle constitutes injury, harm, or economic loss to the owner. To the
contrary, in their complaints, the named plaintiffs contended that injury was
157
suffered when property damage occurred or when repair or replacement costs
were incurred. FOF 99.
i. Plaintiffs submit no evidence that a shingle manifesting “any
cracking, degranulation, fragmentation, or deterioration” will fail.
ii. Plaintiffs submit no evidence that a shingle manifesting “any
cracking, degranulation, fragmentation, or deterioration” is likely to
fail.
iii. Plaintiffs submit no evidence that a shingle manifesting “any
cracking, degranulation, fragmentation, or deterioration” will cause
property damage.
iv. Plaintiffs submit no evidence that a shingle manifesting “any
cracking, degranulation, fragmentation, or deterioration” is likely to
cause property damage.
v. Plaintiffs submit no evidence that a shingle manifesting “any
cracking, degranulation, fragmentation, or deterioration” requires
immediate or imminent shingle repair or replacement. To the
contrary, Owens Corning submitted evidence indicating that not all
“cracks” are indicative of imminent shingle failure. COL 48(b)(i).
vi. The record is devoid of any evidence that possession of a shingle
that is cracked, degranulated, fragmented, or deteriorated results,
without more, in legally compensable injury to the owner.
b. To the extent that plaintiffs’ injury theory is that all class members were
injured because they did not “get what they were promised,” the record reflects
158
that this theory cannot be established with evidence that is common to the
class.
i. For purposes of this analysis, although actually dispositive, the
court puts aside the fact that plaintiffs failed to establish that: (1)
Oakridge-brand shingles share a common defect; and (2) Owens
Corning made uniform representations to all (or even most or
many) members of the proposed four-state class about the useful
life of Oakridge-brand shingles.
ii. Plaintiffs cannot prove by evidence that is common to the class that
all (or even most or many) class members were injured because
Owens Corning promised that Oakridge-brand shingles would last
25 years, or more, but delivered shingles that will not last more than
20 years.
a) The factual circumstances relevant to what was promised
and what was delivered are individualized, not common.
(i) The record reflects that Owens Corning sold
Oakridge-brand shingles with limited
warranties ranging from 25 years to a lifetime
term. FOF 110, 169(d)(iv); COL 24(b).
(ii) The record reflects that some owners also
purchased enhanced warranties. FOF 169.
(iii)The record reflects that some owners will have
installed Oakridge-brand shingles on their
159
structure 24 years ago, while other owners
will have installed Oakridge-brand shingles on
their structure 4 years ago. FOF 6.
(iv) The record reflects that some owners’
Oakridge-brand shingles will last for more
than 30 years. FOF 194(F), 211; COL 26(c).
(v) The record reflects that a determination cannot
be made about how long a shingle is likely to
last without examining the shingle. FOF
194(b) & (d).
(vi) In support of their motion for class
certification, plaintiffs proffered no method by
which a comparison could be made, on a
collective basis, between what was promised
and what was delivered to members of the
proposed four-state class such that injury
could be established on a classwide basis.
b) Rutila offers no admissible expert opinion from which this
court could find that all (or even most or many) members
of the proposed four-state class were injured because they
failed to receive what they were promised.
(i) Rutila admits that he cannot determine
whether a shingle was manufactured to the
160
low-end of Owens Corning’s design
specifications without examining the shingle.
FOF 194(b).
(ii) Rutila admits that some Oakridge-brand
shingles manufactured within Owens
Corning’s design specification parameters will
last 30 years. FOF 194(f).
(iii)Rutila offers no admissible opinion about how
frequently Owens Corning manufactured
Oakridge-brand shingles at the low-end of its
design specifications. FOF 195(a).
(iv) Rutila does not dispute that all Owens Corning
design specifications exceed industry
standards. FOF 109.
c) Evidence about the warranty claims submitted to Owens
Corning with respect to Oakridge-brand shingles is not
probative that all (or even most or many) members of the
proposed four-state class were injured because they failed
to receive what they were promised.
(i) The warranty claims represent less than half
of one percent of all Oakridge-brand
installations and, therefore, are not
statistically significant for the purpose of
161
proving that all (or even most or many)
members of the proposed four-state class did
not get what they were promised. FOF 112.
(ii) Owens Corning pays some warranty claims
to preserve customer goodwill, not because
the shingles are defective. FOF 116.
(iii)Owens Corning’s warranty claims process
itself contradicts any contention that injury
can be proven with classwide evidence in
this case. The warranty program requires a
case-by-case review and assessment of each
owner’s situation before a determination is
made about what financial remuneration, if
any, is appropriate. FOF 115. The fact that
such determinations are made on an
individual basis, indicates that it is
impossible to establish that all (or even most
or many) of the proposed four-state class did
not get what they were promised on a
classwide basis.
d) The quality control issues and internal emails and
documents cannot establish injury on a classwide basis
because that evidence reflects only sporadic and
162
inconsistent instances in which concerns about Oakridgebrand shingles were addressed, and is not probative that
all (or even most or many) members of the proposed
four-state class did not get what they were promised. FOF
147-59, 205-09.
COL 78: Plaintiffs must prove causation for each category of legal claims being asserted.
Cleary, 656 F.3d at 516, 519 (unjust enrichment plaintiff “must show a detriment – and,
significantly, a connection between the detriment and the defendant's retention of the
benefit”); Clark, 256 F. App’x at 821-22 (In a case alleging deception under the Illinois
Consumer Fraud Act, it is not possible for a plaintiff to establish proximate causation
unless the plaintiff can show that he or she was, in some manner, deceived’ by the
misrepresentation); Altronics, 957 F.2d at 1105 (under Pennsylvania law, to prove breach
of implied warranty of merchantability, “plaintiffs were required to show: (1) that the
product malfunctioned; (2) that plaintiffs used the product as intended or reasonably
expected by the manufacturer; and (3) the absence of other reasonable secondary
causes”); Sevidal, 189 Cal. App. 4th at 929 (“both the named plaintiff and unnamed class
members must have suffered some damage caused by a practice deemed unlawful under
the California Consumers Legal Remedies Act); Uzyel v. Kadisha, 188 Cal. App. 4th
866, 892 (2010) (Under California law, “[a] plaintiff seeking a money judgment . . . as a
remedy for unjust enrichment need [ ] establish a causal connection between the
wrongful conduct and the profits to be disgorged”); Cardinal Health 301, 169 Cal. App.
4th at 145 (“To recover on a breach of warranty cause of action, the plaintiff must show
the breach caused the plaintiff to suffer injury, damage, loss or harm.”); Garza, 179
163
S.W.3d at 81 (under Texas law, “[if] a product is shown to be unmerchantable, a plaintiff
must then establish that the defect caused him to suffer injury”); Wall, 176 S.W.3d at 105
(to prevail under the Texas Deceptive Trade Practices Act, a plaintiff must show that the
wrongful act was a producing cause of the plaintiff's economic or mental-anguish
damages).
COL 79: Plaintiffs failed to meet their burden to establish that causation can be proven by
evidence that is common to the proposed four-state class.
a. Based upon this court’s prior findings and conclusions, plaintiffs cannot
establish that Owens Corning’s defective design specifications were the cause
of the harm suffered by all (or even most or many) members of the proposed
four-state class.
i. For purposes of this analysis, although dispositive, the court puts aside
plaintiffs’ failure to establish that Owens Corning’s Oakridge-brand
shingles share a common design defect. COL 73.
ii. Even if the named plaintiffs proved that Oakridge-brand shingles had the
same design defect, individual inquiries about why a shingle cracked,
degranulated, fragmented, or deteriorated, and what caused property
damage, if any, are incompatible with the predominance requirement.
Marcus, 687 F.3d at 604 (noting that “any tire can go flat for myriad
reasons” and “even defective tires can go flat for reasons completely
unrelated to their defects”).
a) The record reflects that there are many reasons that a
shingle might crack, degranulate, fragment, or deteriorate
164
that are unrelated to the design specification used to
manufacture the shingle, including ordinary and
acceptable wear and tear. FOF 196.
b) The record reflects that individual inspection is required
to establish the cause of shingle failure, and resultant
property damage, if there is any. FOF 197.
iii.
Even if the named plaintiffs proved that Oakridge-brand shingles
had the same design defect, they propose no method of
quantifying, on a classwide basis, the harm that is caused simply
by virtue of the fact that a shingle manifests “any cracking,
degranulation, fragmentation, or deterioration during the warranty
coverage period.”
iv. Even if the named plaintiffs proved that Oakridge-brand shingles
had the same design defect, they propose no method of
quantifying, on a classwide basis, the harm that is caused to all
class members given that the harm caused by a shingle lasting for
24 years instead of 25.2 years is different in character and value
than the harm caused by a shingle lasting 5 years instead of 50
years.
b. Based upon this court’s prior findings and conclusions, plaintiffs cannot
establish that Owens Corning’s representations about Oakridge-brand shingles
were the cause of the harm suffered by all (or even most or many) members of
the proposed four-state class.
165
i. It is plaintiffs’ burden to prove, by common evidence, that the
proposed four-state class was exposed to a misrepresentation about
Oakridge-brand shingles. Mazza v. Am. Honda Motor Co., 666 F.3d
581, 596 (9th Cir. 2012) (California Unfair Competition Law “does
not allow a consumer who was never exposed to an alleged false or
misleading advertising . . . campaign to recover damages”); Clark,
256 F. App’x at 821-22 (for Illinois Consumer Fraud Act claim, “it
is not possible for a plaintiff to establish proximate causation unless
the plaintiff can show that he or she was, in some manner, deceived
by the misrepresentation”) (internal quotation omitted); Yurcic v.
Purdue Pharma, LP, 343 F. Supp. 2d 386, 394 (M.D. Pa. 2004) (for
breach of express warranty under Pennsylvania law, “[i]t is
important [ ] that the buyer at least be aware of the seller’s
representation”); Sevidal, 189 Cal. App. 4th at 929 (“Because the
majority of unnamed class members did not view the alleged
misrepresentation, they could not satisfy this causation element of
the [California Legal Remedies Act] claim”); Pfizer, 182 Cal. App.
4th at 631-32 (“[O]ne who was not exposed to the alleged
misrepresentations and therefore could not possibly have lost money
or property as a result of the unfair competition is not entitled to
restitution” under California’s False Advertising Law or Unfair
Competition Law).
166
ii. For the reasons set forth elsewhere in these findings and conclusions,
the record contradicts a finding that plaintiffs could prove with
common evidence that all (or even most or many) members of the
proposed four-state class were exposed to the same, or even
substantially similar, representations about the useful life of
Oakridge-brand shingles. FOF 120-59, 175-80; COL 25, 31(b)(ii)(a).
iii. Even if plaintiffs could prove that all (or even most or many)
members of the proposed four-state class were exposed to the same
representations, the record contradicts a finding that plaintiffs could
prove with common evidence that all (or even most or many)
members of the proposed four-state class decided to purchase
Oakridge-brand shingles based upon them.
a) The record reflects that even the named plaintiffs selected
Oakridge-brand shingles based upon different factors,
including brand name, color, aesthetics, and warranty
length. FOF 50, 78, 90.
b) Plaintiffs propose no method of proving that both current
owners who installed Oakridge-brand shingles on the
structure, and current owners who purchased a structure
with Oakridge-brand shingles previously installed on it
relied upon Owens Corning’s representations or engaged
in the same decision-making process with respect to
selecting Oakridge-brand shingles.
167
c) Where named plaintiffs proffer no evidence that the class
members’ decision-making process was uniform, the
predominance requirement cannot be met. Marcus, 687
F.3d at 607-08.
COL 80: Even if the named plaintiffs could prove injury and causation with common proof,
the court finds that plaintiffs failed to proffer any method or theory by which damages
could be determined on a classwide basis so that this court can evaluate how to weigh this
factor in the predominance balance.
a. Although plaintiffs contend, generally, that there are several “effective
methods for calculating damages” such as informal claims procedures,
hearings before a special master, or summary jury trials, and using “industry
resources such as R.S. Means,” they present no damages models or methods
for this court to assess at the class certification stage. (ECF No. 178 at 44-45
(¶¶ 112-13).) Plaintiffs’ contention that they have no obligation to do so
because damages discovery has not yet taken place misses the mark. (Id. at 45
(¶ 114).) Plaintiffs need not have access to specific damages discovery from
Owens Corning in order to formulate possible damages models that comport
with their theory of the case and the facts disclosed to date in these
proceedings.
b. This court acknowledges that “an inability to calculate damages on a classwide
basis will not, on its own, bar certification,” because the predominance
requirement directs a court to weigh those issues of law and fact that can be
proven by common evidence, against those issues that cannot. Reyes, 802 F.3d
168
at 485; Neale, 794 F.3d at 374-75. There is no requirement that the named
plaintiffs prove that damages are calculable on a classwide basis for each
member of the proposed four-state class in order to secure certification. City of
Sterling Heights Gen’l Employees’ Ret. Sys. v. Prudential Fin., Inc., No. 125275, 2015 WL 5097883, at *13-14 (D.N.J. Aug. 31, 2015) (citing Neale, 794
F.3d at 374-75); In re Wilmington Trust Securities Litig., 310 F.R.D. 243, 246
(D. Del. Sept. 3, 2015) (finding predominance met where plaintiffs presented
an expert’s “event study methodology” to calculate damages on a classwide
basis). In this case, however, many other issues of law and fact are not
susceptible to common proof, and the record indicates not only an inability to
measure the exact amount of damages owed to each class member, but also a
lack of any method by which entitlement to damages could be determined on a
classwide basis based upon the facts, legal theories, and expert testimony
presented in this case. In these circumstances, plaintiffs failure to demonstrate
that damages can be proven by common evidence further demonstrates that
certification of the proposed four-state class is improper.
c. Because the named plaintiffs did not advance any damages theory that they
will pursue on behalf of the class, the court will examine the most likely
damages theories based upon the facts of this case and the legal causes of
action being asserted on behalf of the class.
i. To the extent the named plaintiffs will seek to recover, on behalf of
the class, any costs incurred by class members who replaced their
Oakridge-brand shingles, individual questions will predominate. In
169
order to recover such damages, the class member would have to
demonstrate, among other things, that shingle replacement was
required in order to prevent property damage or ensure consistent
shingle performance, the cost of the new shingles, the cost of the
old shingles, and the age of the Oakridge-brand shingles at the time
of replacement for purposes of depreciation. Because there is no
basis on this record to assume that every Oakridge-brand shingle
replaced had to be replaced because it was defectively designed,
class members would have to prove that that shingle failure was not
caused by some other condition. FOF 196. This is an individual
issue. The record reflects that determining the price actually paid
for the Oakridge-brand shingles will not be readily provable on a
classwide basis. COL 51(d). This is an individual issue. The
named plaintiffs proffer no method by which depreciation of the
existing Oakridge-brand shingles could be determined on a
classwide basis. This could be an individual issue.
ii. To the extent the named plaintiffs will seek compensation, on
behalf of the class, for any property damage suffered when a roof
with Oakridge-brand shingles on it leaked, individual questions will
predominate. Based upon the record before the court, an individual
assessment of each property is required to determine the cause of a
leak and any resultant property damage. FOF 197. The record
reflects that some of the named plaintiffs who suffered property
170
damage could not show that Oakridge-brand shingle failure was the
cause. FOF 100; Gates v. Rohm and Haas Co., 265 F.R.D. 208, 233
(E.D. Pa. 2010) (“common issues do not predominate as to
Plaintiffs’ property damage claims”); Martin v. Home Depot
U.S.A., Inc., 225 F.R.D. 198, 201-02 (W.D. Tex. 2004) (“[T]here
are too many property-specific factors influencing the issues of
whether, and to what extent, treated wood leaches, and whether a
given property would be contaminated if the wood did leach, to
permit certification. . . . [A]djudicating each claim would require
intense scrutiny of each potential plaintiff's property to a degree that
would not be feasible.”). The record further reflects that owners
may have received payment for those damages, from an insurer for
example, further complicating any damages model that will seek to
reimburse class members for property damages. FOF 100(a).
iii. To the extent the named plaintiffs seek to recover, on behalf of the
class, on a diminution-in-value theory, innumerable individual
issues will predominate.
a) The record reflects that not all Oakridge-brand shingles
will be defective, and that individual inspection is
required to identify those shingles that were
manufactured at the allegedly defective low end of
Owens Corning’s design specifications. FOF 194(a)-(b).
Plaintiffs proffered no evidence from which this court
171
could determine that it is more likely than not that
Oakridge-brand shingles will be defective. FOF 195;
COL 73(c)(ii)(b) (citing Reyes, 802 F.3d at 485). There
is no evidence that every, or even most or many,
Oakridge-brand shingles will be manufactured using
allegedly defective design specification measurements.
FOF 195, 213-15; COL 73(c)(ii) (citing Neale, 2013 WL
785056, at *4). The record reflects that it would be
difficult, if not impossible, to make this assessment
because plaintiffs do not identify what combination of
measurements define the point at which a shingle crosses
the line from being nondefective to defective. FOF
194(c), 209.
b) Based upon the proposed class definition, some members
of the proposed four-state class will not have purchased
Oakridge-brand shingles, but will have purchased a
structure with those shingles already installed on it. In
these circumstances, a diminution-in-value theory does
not fit.
c) Even for those class members who did install Oakridgebrand shingles on the structure, calculating the
diminution-of-value will depend on various individual
factors, and could not be accomplished on a classwide
172
basis. By way of example, the calculation would depend
upon the purchase price of the Oakridge-brand shingles,
adjusting for inflation and market differences, the age of
the Oakridge-brand shingles, and the condition and
expected future lifespan of the Oakridge-brand shingles.
These are individual questions, some of which the named
plaintiffs cannot even answer. COL 51(c).
COL 81: Even if the named plaintiffs could prove injury, causation, and damages with
common proof, the court finds that individual defenses would additionally bar
certification of the class. Any defense based upon the timeliness of the claims brought by
members of a proposed four-state class in which membership commenced 24 years ago
will raise predominantly individual fact issues.
a. Determination of statute of limitations defenses can prevent a finding of
predominance. Thorn v. Jefferson-Pilot Life Ins. Co., 445 F.3d 311, 326 (4th
Cir. 2006) (affirming holding that plaintiffs “failed to show that the
predominance requirement of Rule 23(b)(3) has been satisfied because
individualized proof is patently required to litigate the defendants’ statute of
limitations defense”) (quotations omitted); Barnes v. Am. Tobacco Co., 161
F.3d 127, 149 (3d Cir. 1998) (“we believe that determining whether each class
member’s claim is barred by the statute of limitations raises individual issues
that prevent class certification”); In re Actiq Sales and Marketing Practices
Litig., 307 F.R.D. 150, 164 (E.D. Pa. 2015) (statute of limitation issues
rendered putative class unmanageable); Anderson Living Trust v. WPX Energy
173
Prod., LLC, 306 F.R.D. 312, 450 (D.N.M. 2014) (“Defendants’ statute-oflimitations defense also cuts against certification”); In re Ford Motor Co. E350 Van Prods. Liab. Litig. (No. II), No. 03-4558, 2012 WL 379944, at *36
(D.N.J. Feb. 6, 2012) (denying class certification where defendant’s “statuteof-limitations defenses and any applicable discovery rule and/or other equitable
tolling doctrine would require inquiries into the individual circumstances of
each Plaintiff”).
b. In this case different statutes of limitations, ranging from two to five years,
apply to the claims being asserted on behalf of the proposed four-state class. 2
years: Texas DTPA, Tex. Bus. & Com. Code. Ann. § 17.565; TX unjust
enrichment, Elledge v. Friberg-Cooper Water Supply Corp., 240 S.W.3d 869,
871 (Tex. 2007); 3 years: CLRA, Cal. Civ. Code § 1783(a); FAL, Cal. Code of
Civil Procedure Section 338(a); CA unjust enrichment, Allen v. Similasan
Corp., No. 12-cv-0376, 2013 WL 2120825, at *6 (S.D. Cal. May 14, 2013);
ICFA, 815 ILCS 505/10a(e); 4 years: PA breach of express and implied
warranty, Westfield Ins. v. Detroit Diesel Corp., No. 10-cv-100, 2012 WL
1611311, at *7 (W.D. Pa. May 8, 2012); PA unjust enrichment, 42 Pa. Cons.
Stat. § 5525(a)(4); CA breach of implied warranty, Mexia v. Rinker Boat Co.,
Inc., 174 Cal. App. 4th 1297, 1307 (2009); UCL, Cal. Bus. & Prof. Code §
17208; TX breach of implied warranty, Tex. Bus. & Com. Code. Ann. §
2.725(b); 5 years: IL unjust enrichment, 735 ILCS 5/13-205 (2010). Individual
factual circumstances with respect to knowledge and notice may affect how
those statutes of limitations apply.
174
c. In addition, equitable doctrines, such as laches, estoppel, and waiver may bar
claims of some class members, who could have been aware, as early as 1992,
of the alleged defective condition of their Oakridge-brand shingles.
d. Plaintiffs proffer no way for this court to adjudicate those issues on a classwide
basis.
b. Superiority
COL 82: “ʻThe superiority requirement asks a district court to balance, in terms of fairness
and efficiency, the merits of a class action against those of alternative available methods
of adjudication.ʼ” Community Bank III, 795 F.3d at 409 (quoting Community Bank I,
418 F.3d at 309).
COL 83: “[F]requently, a finding that individual issues do not predominate is accompanied
by a finding that a class action is not superior,” because where “a number of ‘mini-trials’
would be imperative, the costs associated with a class action would be very high; due to
the necessary individual litigation of a number of issues, the cost savings [of class
litigation] would not be marked.” Sanneman v. Chrysler Corp., 191 F.R.D. 441, 454-55
(E.D. Pa. 2000); see Newton v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 259 F.3d
154, 191 (3d Cir. 2001) (court considering superiority requirement “must address ‘the
difficulties likely to be encountered in the management of a class action.’”) (citing FED.
R. CIV. P. 23(b)(3)(D)).
COL 84: Plaintiffs’ superiority argument is that owners will have no financial incentive to
pursue their claims individually because “the cost of litigating a federal lawsuit” will
overtake the amount of damages. (ECF No. 178 at 43, 44 (¶ 109, 111).)
175
a. There, however, is no requirement that owners file suit in federal court in order
to pursue the legal claims on which class certification is sought, as plaintiffs
contend. (ECF No. 178 at 44 (¶ 111).) In fact, it is unlikely that the amount in
controversy requirement could be met on an individual basis to establish
diversity jurisdiction in a federal district court in order to pursue the state-law
claims being asserted in this case. Owners have the ability to seek recovery for
damages in state court, where claims of several thousand dollars are not
uncommon and where procedures to administer claims of that size are
available.
b. Plaintiffs’ lament that class members will be effectively deprived of
representation by counsel if they are forced to litigate their claims individually
is not well-founded. Before proceeding, it is important to note that the correct
inquiry is whether owners will be left with virtually no access to counsel, not
whether class counsel appearing in this litigation would agree to litigate
individual cases.
i. The claims held by the named plaintiffs, and by extension, the
claims held by members of the proposed four-state class, are not
financially insignificant and range from $5,000 to $22,000. (ECF
No. 178 at 44 (¶ 111); FOF 31, 57, 65, 83-84, 97. This is not a case
in which an individual has suffered harm in connection with a
minor or fleeting consumer transaction or service, such as
purchasing yogurt or vitamins, paying at a restaurant with a credit
card, or using an automated teller machine. See e.g., Reyes, 802
176
F.3d at 491 (noting that no rational lawyer would represent a party
who holds a $30.22 claim).
ii. State consumer protection statutes typically provide for fee-shifting,
which could incentivize attorneys to represent owners. See e.g.,
Boehm v. Rivercsource Life Ins. Co., 117 A.3d 308, 336 (Pa. Super.
Ct. 2015); Graciano v. Robinson Ford Sales, Inc., 144 Cal. App. 4th
140, 159-60 (2006).
iii. There is no basis on this record to find that failure to certify the
class will effectively deny owners access to counsel.
c. Elsewhere in plaintiffs’ proposed findings and conclusions, plaintiffs intimate
that owners will be unable to pursue litigation individually because it will be too
expensive to obtain expert opinions establishing the “modes of failure in their
shingles.” (ECF No. 178 at 46 (¶ 118).) The expert opinion that the named
plaintiffs presented in this case to obtain class certification was for the purpose
of establishing that all Oakridge-brand shingles manufactured in four states over
a 20-year time period were defective. Although the court will not dispute that
the opinion, and the discovery and legal proceedings relating to it, were likely
expensive, individual owners could be able to establish this fact by experts who
developed expertise by experience, i.e., a roofer, building inspector, etc., and
whose fees would likely be less than those charged by plaintiffs’ expert.
COL 85: In summary, individual issues predominate over common issues and class
treatment is not superior to individual litigation of claims under the circumstances of this
case.
177
a. Owners have an interest in individually controlling litigation that affects whether
financial remuneration will be provided to repair or prevent property damage.
The record reflects that owners’ circumstances and experiences with Oakridgebrand shingles are not uniform: some have suffered property damage, while others
have not; some have accepted reimbursement from Owens Corning, while others
have not; some have likely replaced the shingles on their roofs as a precaution,
while others have not; some owners selected Oakridge-brand shingles, while
others purchased structures on which they were already installed; some owners
have had Oakridge-brand shingles installed on their roofs for almost 25 years,
while others had them installed less than 5 years ago. These varying factual
circumstances are indicative of an interest in individually controlling litigation.
b. There is no benefit to using the class action mechanism based upon pending
litigation because no liability issues have yet been decided in this matter. The
record reflects that some owners reached settlement of their claims against Owens
Corning without the need to resort to litigation. FOF 97.
c. It is not more desirable to concentrate the litigation in one particular forum
because the record reflects that differences in geography affect how shingles are
manufactured, and how they are expected to weather in the ordinary course.
d. Finally, as set forth elsewhere in these findings and conclusions, based upon this
record, there are innumerable difficulties in managing these claims as a class
action.
178
3. Rule 23(b)(2) – The Injunctive Relief Class
COL 86: Rule 23(b)(2) provides that “[a] class action may be maintained if Rule 23(a) is
satisfied and if ... the party opposing the class has acted or refused to act on grounds that
apply generally to the class, so that final injunctive relief or corresponding declaratory
relief is appropriate respecting the class as a whole.” In re Processed Egg Products
Antitrust Litig., 312 F.R.D. 124, 165 (E.D. Pa. 2015).
COL 87: “Subsection (b)(2) class actions are ‘limited to those class actions seeking
primarily injunctive or corresponding declaratory relief.’” Processed Egg Products, 312
F.R.D. at 165 (citing Barnes, 161 F.3d at 142). If monetary damages are appropriate,
certification is proper under the standards and requirements applicable to Rule 23(b)(3),
not Rule 23(b)(2). Dukes, 131 S.Ct. at 2557-58; In re Ford, 2012 WL 379944 at *38-39
(“In light of the guidance provided by Wal-Mart [v. Dukes], this Court concludes that it
would be inappropriate to permit Plaintiffs to sidestep the (b)(3) requirements under the
guise of a (b)(2) class.”).
COL 88: Rule 23(b)(2) classes must be cohesive. Shelton, 775 F.3d at 561 (citing Barnes,
161 F.3d at 143). A Rule 23(b)(2) class is not cohesive if its members raise “disparate
factual circumstances” or “significant individual issues.” In re Ford, 2012 WL 379944 at
*38-39 (citing Barnes, 161 F.3d at 143).
COL 89: The Court of Appeals for the Third Circuit has recognized that Rule 23(b)(2)’s
cohesiveness requirement may be more stringent than the predominance and superiority
requirements of Rule (b)(3) because all class members will be bound by a single
judgment and (b)(2) requires no notice and provides no opportunity to opt out. Gates, 655
F.3d at 264 & n.12; see also Phillips Petroleum Co. v. Shutts, 472 U.S. 797, 811-12 & n.3
179
(1985) (requiring notice and opt-out rights, which are provided in (b)(3) but not (b)(2)
class actions, when action is “predominately” for money damages).
COL 90: A Rule 23(b)(2) injunctive relief class is typically used in consumer product class
actions to obtain a declaration that a product is defective so that relief can be easily
pursued once that defect manifests or causes damages. See e.g., Pella, 606 F.3d at 392
(affirming certification of a Rule 23(b)(2) class of owners whose Pella windows had not
yet manifested the alleged design defect (which caused accelerated wood rot) or been
replaced). In such cases, Rule 23(b)(3) class members are entitled to damages because
the defect already manifested itself (assuming all other elements of a particular legal
claim are established), while Rule 23(b)(2) class members are guaranteed damages if the
defect ever manifests and causes harm (again, assuming all other elements of a particular
legal claim are established).
COL 91: At the hearing on plaintiffs’ class certification motion, plaintiffs explained that, in
this case, the proposed Rule 23(b)(2) injunctive relief class was intended to be an
alternative in the event that the court declined to certify the proposed Rule 23(b)(3)
monetary relief class. In that event, the Rule 23(b)(2) class would provide class members
with a declaration that the shingles are defective or that Owens Corning did not provide
what was promised. FOF 3(a) and (b). Because, however, plaintiffs have not satisfied the
requirements of Rule 23(b)(3) and rely upon the same evidence and legal theories in
seeking certification under Rule 23(b)(2), they cannot satisfy the requirements of Rule
23(b)(2) in this case.
COL 92: Despite plaintiffs’ statements at the class certification hearing, in their proposed
findings of fact and conclusions of law, plaintiffs contend that both a Rule 23(b)(3) and a
180
Rule 23(b)(2) class should be certified because the former will provide relief for owners
who have already “suffered substantial product failures” and the latter will “protect class
members who have yet to suffer substantial product failures.” (ECF No. 178 at 30 (¶ 54)).
COL 93: Plaintiffs specifically seek declarations, on behalf of the proposed four-state
injunctive relief class, that all Oakridge-brand shingles suffer from a design defect, and
that Owens Corning cannot deny warranty claims or make further false representations
about Oakridge-brand shingles. (ECF No. 178 at 30 (¶54) and 31 (¶ 56).)
COL 94: Despite claiming that the proposed Rule 23(b)(2) class and the proposed Rule
23(b)(3) class will consist of different members and afford its members different relief,
plaintiffs continue to assign the same definition to both classes. (ECF No. 178 at 18-19 (¶
1).) This alone makes certification of the Rule 23(b)(2) injunctive relief class improper.
a. In order to qualify as a member of both classes, an owner need only show that
the Oakridge-brand shingles installed on the structure “manifested any
cracking, degranulation, fragmentation, or deterioration during the warranty
coverage period.” FOF 6. There is no distinction between the classes with
respect to the condition of the shingles, whether the shingles have been
replaced, or whether the structure has suffered property damage. Members of
both classes have suffered the same level of “product failure.” The proposed
Rule 23(b)(2) class, therefore, will not, as plaintiffs contend, protect owners
who have not yet suffered “substantial product failure.”
b. According to plaintiffs’ theory of the case, owners were harmed by virtue of
the fact that Owens Corning manufactured Oakridge-brand shingles in
accordance with specifications that were defective and caused the shingles not
181
to last as long as Owens Corning promised. Membership in the Rule 23(b)(3)
damages class is not based upon an owner having been forced to replace the
shingles on the structure. Membership in the Rule 23(b)(3) damages class is
not based upon an owner having suffered property damage as a result of
shingle failure. Membership in the Rule 23(b)(3) damages class is based upon
the mere possession of Oakridge-brand shingles because, according to
plaintiffs, they all are “susceptible” to premature failure due to defective design
specifications, which, standing alone, constitutes harm. There is no need,
under plaintiffs’ theory of the case, for an injunctive relief class to be certified
in order to await “substantial product failure” because members of both classes
have suffered the same “product failure,” i.e., “manifestation of any cracking,
degranulation, fragmentation, or deterioration,” and the same harm, i.e.,
possession of shingles that are susceptible to not lasting as long as the length of
the limited shingle warranty.
COL 95: To the extent that plaintiffs intended to define the proposed Rule 23(b)(2)
injunctive relief class as those owners whose shingles have not yet “manifested any
cracking, degranulation, fragmentation, or deterioration,” certification of the class
nevertheless remains improper because, for all the reasons set forth in this court’s
analysis of the Rule 23(a) and Rule 23(b)(3) requirements, such a class would not be
cohesive. The court need not reiterate here all the “disparate factual circumstances” and
“significant individual issues” identified elsewhere herein, but incorporates the following
findings and conclusions, and any other pertinent findings or conclusions, by reference:
FOF 100-01, 104-11, 115, 118-80; COL 23-27, 31, 38(b), 72-81.
182
4. Rule 23(c)(4) – Certification of “Particular Issues” Class
COL 96: Federal Rule of Civil Procedure 23(c)(4) provides that “[w]hen appropriate, an
action may be brought or maintained as a class action with respect to particular issues.”
COL 97: Plaintiffs do not ask this court to certify a class pursuant to Rule 23(c)(4), ECF
No. 178 at 18-19 (¶ 1), but their last five proposed conclusions of law address this kind of
class, Id. at 45-46 (¶¶ 115-19), and, therefore, in the interest of completeness, the court
will address the propriety of certifying such a class.
COL 98: Plaintiffs’ proposed conclusions of law do not define the scope of the liabilityonly trial nor propose what common proof would be presented, making certification
improper on this basis alone. Gates, 655 F.3d at 274.
COL 99: Plaintiffs instead argue, generally and circularly, that “the parties will proceed to
trial on the threshold question of liability for the following claims: breach of express
warranty under Pennsylvania law; breach of implied warranty under Illinois,
Pennsylvania, and Texas law; violation of consumer protection laws in California,
Illinois, and Texas; and unjust enrichment” and “that all these claims rest on a common
core of facts: design defects in Oakridge; Owens Corning’s knowledge of those defects;
and deceptive representations about the reliability and durability of Oakridge.” (ECF No.
178 at 46 (¶¶ 117, 119).)
COL 100: In Gates, the Court of Appeals for the Third Circuit enumerated certain factors
that district courts should consider when deciding whether or not to certify an “issue
class.” Those factors include:
the type of claim(s) and issue(s) in question; the overall complexity of
the case; the efficiencies to be gained by granting partial certification in
light of realistic procedural alternatives; the substantive law underlying
the claim(s), including any choice-of-law questions it may present and
183
whether the substantive law separates the issue(s) from other issues
concerning liability or remedy; the impact partial certification will have
on the constitutional and statutory rights of both the class members and
the defendant(s); the potential preclusive effect or lack thereof that
resolution of the proposed issue class will have; the repercussions
certification of an issue(s) class will have on the effectiveness and
fairness of resolution of remaining issues; the impact individual
proceedings may have upon one another, including whether remedies
are indivisible such that granting or not granting relief to any claimant
as a practical matter determines the claims of others; and the kind of
evidence presented on the issue(s) certified and potentially presented on
the remaining issues, including the risk subsequent triers of fact will
need to reexamine evidence and findings from resolution of the
common issue(s).
Gates, 655 F.3d at 273.
COL 101: Although plaintiffs recognize that some of these factors must be considered, they
contend simply that certification of a Rule 23(c)(4) class is appropriate because all their
legal “claims rest on a common core of facts.” (ECF No. 178 at 46 (¶ 117).) This
contention is insufficient to meet plaintiffs’ burdens and obligations to justify
certification of a particular issues class.
COL 102: The court finds that the Gates factors do no warrant certifying a particular issues
class. The claims being pursued on behalf of the proposed four-state class arise under the
laws of four different states. FOF 13; COL 22(b)(i)(b), 31(a)(i), 38(b)(i)(a), 71. The
claims involve individual issues, and this court has already set forth the reasons why they
are not susceptible to proof by common evidence. COL 23-27, 31, 38(b), 72-81. No
efficiencies are gained by litigating specific issues pertinent to these legal claims on a
classwide basis.
COL 103: In particular, several courts have found unjust enrichment claims to be unsuitable
for class treatment where the claim required a highly individualized inquiry in order to
determine whether a defendant had been unjustly enriched in a particular circumstance.
184
Walney v. SWEPI LP, No. 13-102, 2015 WL 5333541, at *15 (W.D. Pa. Sept. 14, 2015)
(citing decisions).
COL 104: In particular, the record reflects that the terms of the express warranties held by
Pennsylvania owners will differ, making this claim unsuitable for class treatment. FOF
169.
For the foregoing reasons, plaintiffs’ motion for class certification (ECF No. 150)
is denied. An appropriate order will be entered contemporaneously with this opinion.
Dated: March 31, 2016
BY THE COURT:
/s/ Joy Flowers Conti
Joy Flowers Conti
Chief United States District Judge
185
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?