Stejic v. Aurora Loan Services, LLC et al

Filing 55

ORDER denying 50 Defendant Aurora Loan Services, LLC's Motion for Leave to File Sur-Reply. See document for details re Lead Case: MDL 09-2119-PHX-JAT. Signed by Senior Judge James A Teilborg on 12/21/15.(LSP)

1

WO

2
3
4
5
6

IN THE UNITED STATES DISTRICT COURT

7

FOR THE DISTRICT OF ARIZONA

8
Docket No. MDL 09-2119-PHX-JAT

9
10
11
12
13
14
15

IN RE MORTGAGE ELECTRONIC
REGISTRATION SYSTEMS (MERS)
LITIGATION

ORDER

THIS DOCUMENT RELATES TO:
CV 10-00630-PHX-JAT (Robinson)
CV 10-01550-PHX-JAT (Bilyea)
CV 10-01547-PHX-JAT (Stejic)
CV 10-01548-PHX-JAT (Molina)

16

Pending before the Court are two Motions for Class Certification pursuant to Rule

17

23 of the Federal Rules of Civil Procedure. The first motion was filed by Plaintiffs Sally

18

Robinson-Burke, Rosa Silvas, Nicholas DeBaggis, and Thomas Bilyea.1 (Doc. 1913).

19

The second motion was filed by Plaintiffs Milan Stejic and Maria Hernandez

20

(collectively, “Plaintiffs”). (Doc. 1914). The Court now rules on the Motions.

21

I.

Factual Background

22

This case originally began as a multi-district litigation (“MDL”) centralizing civil

23

actions related to the formation and operation of Mortgage Electronic Registration

24

Systems, Inc. and MERSCORP, Inc. (collectively, “MERS”). (Doc. 1). After claims in

25

twenty of the associated cases were dismissed (Doc. 1170; Doc. 1247), the remaining

26
27
28

1

On November 6, 2015, this Court dismissed with prejudice Plaintiff DeBaggis’
claims against Defendant U.S. Bank National Association. (Doc. 1978). Additionally,
Plaintiff Debaggis is not a named Plaintiff in any member case. (Doc. 1602 at 8). Thus,
Plaintiffs class certification cannot rest on allegations made by Plaintiff DeBaggis.

1

Plaintiffs filed a Consolidated Amended Complaint (“CAC”) on June 4, 2011. (Doc.

2

1424).

3

The CAC included twelve counts, including allegations of violations of Arizona

4

Revised Statute (“A.R.S.”) § 33–420 (Count I); wrongful foreclosure (Count II);

5

violations of Nevada Revised Statutes § 107.080 (Count III); violations of Oregon

6

Revised Statutes (“O.R.S.”) § 86.735 (Count IV); allegations of aiding and abetting

7

wrongful foreclosure (Count V); aiding and abetting predatory lending (Count VI); unjust

8

enrichment (Count VII); slander of title (Count VIII); violations of O.R.S. § 646.607

9

(Count IX); and violations of South Carolina Code of Laws § 39–5–10 (Count X). (Doc.

10

1424 at 7-57). Additionally, Plaintiffs sought declaratory relief (Count XI) and injunctive

11

relief (Count XII). (Doc. 1424 at 58-60).

12

The amended complaint was dismissed on October 3, 2011. (Doc. 1602). Plaintiffs

13

appealed dismissal of Counts I-VI to the Ninth Circuit Court of Appeals. The Court of

14

Appeals reversed the dismissal of Count I and affirmed the dismissal of Counts II-VI. In

15

re Mortgage Elec. Registration Sys., Inc., 754 F.3d 772, 786 (9th Cir. 2014). Specifically,

16

the Court of Appeals held that: (1) A.R.S. § 33-420 applies to Notices of Trustee Sale,

17

Notices of Substitution of Trustee, and Assignments of a Deed of Trust, documents

18

which Plaintiffs alleged to be fraudulent in the CAC; (2) Plaintiffs’ claims are not time-

19

barred; (3) Plaintiffs have standing to sue under A.R.S. § 33-420; and (4) Plaintiffs

20

pleaded their robosigning claims with sufficient particularity to satisfy Federal Rule of

21

Civil Procedure 8(a). In re Mortgage Elec. Registration Sys., Inc., 754 F.3d at 781-784.

22

On August 17, 2015, Plaintiffs filed motions seeking class certification for two

23

classes for the purpose “of pursuing a statutory damages claim on behalf of Arizona

24

property owners arising from Defendants’ violation of A.R.S. § 33-420 while conducting

25

foreclosure proceedings in the State of Arizona.” (Doc. 1913 at 1; Doc. 1914 at 1). The

26

first motion proposes a class that would be composed of property owners against whom

27

“[MERS] claimed that it was transferring the beneficial interest under the Deed of Trust

28

to a new beneficiary along with the Note and any payments due under the note, and

-2-

1

caused the assignment to be recorded.” (Doc. 1913 at 1). The second motion proposes a

2

class that would be composed “of owners of residential property in Arizona who had a

3

Substitution of Trustee and Notice of Trustee Sale recorded against their property . . .

4

whereby [MERS], purportedly acting as beneficiary, substituted the trustee, and caused

5

the substitution to be recorded.” (Doc. 1914 at 1). Plaintiffs further allege that documents

6

related to these mortgage transactions misrepresent MERS’s status and authority as a

7

beneficiary, contain false statements regarding assignments, and are robosigned. (Doc.

8

1913 at 3-6; Doc. 1914 at 3-5).

9

II.

Mandate on Remand

10

Within the Ninth Circuit “a court will generally refuse to reconsider an issue that

11

has already been decided by the same court or a higher court in the same case.” Gonzalez

12

v. Arizona, 677 F.3d 383, 390 (9th Cir. 2012). Thus, when “the issue in question [was]

13

decided explicitly or by necessary implication in the previous disposition,” there is no

14

need to revisit the decision. United States v. Thrasher, 483 F.3d 977, 981 (9th Cir. 2007)

15

(quoting Herrington v. County of Sonoma, 12 F.3d 901, 904 (9th Cir. 1993)).

16

Additionally, “when a court is confronted with issues that the remanding court never

17

considered, the ‘mandate requires respect for what the higher court decided, not for what

18

it did not decide.’ ” Hall v. City of Los Angeles, 697 F.3d 1059, 1067 (9th Cir. 2012)

19

(quoting United States v. Kellington, 217 F.3d 1084, 1093 (9th Cir. 2000)).

20

As the dismissal of the CAC explained, Plaintiffs’ claims concerning the operation

21

of MERS have already been found not to state a claim, either by the Arizona courts or the

22

Ninth Circuit Court of Appeals. (Doc. 1602 at 5-8). In its opinion, the Court of Appeals

23

did not reverse or even address Plaintiffs’ claims pertaining to MERS’ status as a

24

beneficiary or the validity of MERS assignments. Instead, the Ninth Circuit Court of

25

Appeals addressed claims of robosigning and noted specific examples of documents

26

allegedly fraudulently signed or notarized within the CAC. In re Mortgage Elec.

27

Registration Sys., Inc., 754 F.3d at 783-84. The Court of Appeals’ opinion did not

28

-3-

1

reference any of Plaintiffs’ other claims or factual assertions from Count I. Id. at 781-84.2

2

This omission indicates the Court of Appeals’ intentions concerning the scope of its

3

mandate on remand. See Fid. Nat’l Fin., Inc. v. Friedman, 855 F. Supp. 2d 948, 956 (D.

4

Ariz. 2012) (holding that “[b]ecause the mandate is silent as to its scope, the court will

5

look at the Ninth Circuit’s opinion” to ensure that the consideration of an issue does not

6

“impermissibly exceed[] the scope of the mandate”).

7

Plaintiffs’ motions for class certification cannot be used to resurrect a dismissed

8

theory. The Court of Appeals’ reversal was limited to Plaintiffs’ claims of robosigning

9

and forgery under A.R.S. § 33-420. Thus, this Court will consider only those portions of

10

Plaintiffs’ motions which address these claims.

11

III.

Class Certification Legal Standard

12

The district courts have broad discretion to grant or deny class certification.

13

Bateman v. Am. Multi-Cinema, Inc., 623 F.3d 708, 712 (9th Cir. 2010) (citing Yamamoto

14

v. Omiya, 564 F.2d 1319, 1325 (9th Cir. 1977)). However, the United States Supreme

15

Court requires that such a determination should be made only after the district court has

16

“engage[d] in a ‘rigorous analysis’ of each Rule 23(a) factor when determining whether

17

plaintiffs seeking class certification have met the requirements of Rule 23.” Ellis v.

18

Costco Wholesale Corp., 657 F.3d 970, 980 (9th Cir. 2011) (citing Gen. Tel. Co. of the

19

Sw. v. Falcon, 457 U.S. 147, 161 (1982)). Under Rule 23, it is the party seeking class

20

certification that “bears the burden of showing that each of Rule 23(a)’s requirements and

21

at least one of Rule 23(b)’s requirements” are satisfied. Juvera v. Salcido, 294 F.R.D.

22

516, 520 (D. Ariz. 2013) (citing Dukes v. Wal–Mart, Inc., 509 F.3d 1168, 1176 (9th Cir.

23

2007), rev’d on other grounds, ––– U.S. ––––, 131 S. Ct. 2541, 180 L. Ed. 2d 374

24

(2011)).

25

Rule 23 has two implicit prerequisites that Plaintiffs must satisfy for the Court to

26

grant certification. Clay v. Am. Tobacco Co., 188 F.R.D. 483, 490 (S.D. Ill. 1999);

27
28

2

This analysis was complicated by the fact that Count I of the CAC is 18 pages
and does not break out its various theories of liability on a count by count basis.
-4-

1

Singleton v. Adick, No. CV 09–486–PHX–JAT, 2009 WL 3710717, at *2 (D. Ariz. Nov.

2

2, 2009). First, in order to maintain a class action, the class must be adequately defined

3

and clearly ascertainable. DeBremaecker v. Short, 433 F.2d 733, 734 (5th Cir. 1970); see

4

also Lozano v. AT & T Wireless Ser., Inc., 504 F.3d 718 (9th Cir. 2007) (“The district

5

court’s failure to analyze the Rule 23(a) factors in determining whether to grant class

6

certification . . . resulted in its certifying a theory with no definable class.”). The class

7

cannot be overbroad, amorphous, or vague, but must be susceptible to a precise

8

definition. Clay, 188 F.R.D. at 490. A class must be precisely defined so the Court can

9

determine who will be bound by the judgment. McHan v. Grandbouche, 99 F.R.D. 260,

10

265 (D. Kan. 1983).

11

Second, the named representative must be a member of the class. Bailey v.

12

Patterson, 369 U.S. 31, 32-33 (1962). Plaintiffs must also prove that their proposed class

13

meets all of the four requirements of Rule 23(a):

14
15
16
17
18

(1) the class is so numerous that joinder of all members is
impracticable; (2) there are questions of law and fact common to the
class; (3) the claims or defenses of the representative parties are
typical of the claims or defenses of the class; and (4) the
representative parties will fairly and adequately protect the interests
of the class.

19

Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d 979, 985 (9th Cir. 2015) (quoting

20

Fed. R. Civ. P. 23(a)).

21
22
23
24
25
26
27
28

Finally, Plaintiffs must also prove that at least one of the following Rule 23(b)
requirements is met:
(1) the prosecution of separate actions would create a risk of : (a)
inconsistent or varying adjudications or (b) individual adjudications
dispositive of the interests of other members not a party to those
adjudications; (2) the party opposing the class has acted or refused to
act on grounds generally applicable to the class; or (3) the questions
of law or fact common to the members of the class predominate over
any questions affecting only individual members, and a class action
is superior to other available methods for the fair and efficient
adjudication of the controversy.
-5-

1

Fed. R. Civ. P. 23(b).

2

IV.

3

Discussion
A.

Adequately Defined and Clearly Ascertainable

4

Plaintiffs claim the proposed classes for each motion would number in excess of

5

50,000 members, based on a sample of documents from three counties in Arizona. (Doc.

6

1913 at 7; Doc. 1914 at 7). Plaintiffs Sally Robinson-Burke, Rosa Silvas and Thomas

7

Bilyea propose certification of a class defined as:

8
9
10
11
12

owners of residential property in Arizona who had an Assignment of
Deed of Trust recorded against their property between February 21,
2008 and the present whereby [MERS] claimed that it was
transferring the beneficial interest under the Deed of Trust to a new
beneficiary along with the Note and any payments due under the
note, and caused the assignment to be recorded.

13

(Doc. 1913 at 1) (footnote omitted). In the proposed order, Plaintiffs Sally Robinson-

14

Burke, Rosa Silvas and Thomas Bilyea add that the purpose of certification is the

15

recovery of statutory damages arising

16
17
18
19
20
21

from the recording by Defendants . . . of any of the following
documents against any of the class members’ property, in violation
of A.R.S. § 33-420: (1) Assignment of Deed of Trust . . . ; (2)
Substitution of Trustee; (3) Notice of Trustee’s Sale; and/or (4)
Trustee’s Deed Upon Sale, which listed one of the Defendants
named herein as a beneficiary.
(Doc. 1916 at 1-2).

22

Plaintiffs Milan Stejic and Maria Hernandez propose a class consisting of “owners

23

of residential property in Arizona who had a Substitution of Trustee and Notice of

24

Trustee Sale recorded against their property between January 27, 2009 and the present

25

whereby [MERS], purportedly acting as beneficiary, substituted the trustee, and caused

26

the substitution to be recorded.” (Doc. 1914 at 1) (footnote omitted). In their proposed

27

order, Plaintiffs Milan Stejic and Maria Hernandez add that the purpose of class

28

certification is the recovery of statutory damages arising

-6-

1
2
3
4
5

from the recording by Defendants . . . either directly or through an
agent, of any of the following documents against any of the class
members’ property, in violation of A.R.S. § 33-420: (1) Substitution
of Trustee; (2) Notice of Trustee’s Sale; and/or (3) Trustee’s Deed
Upon Sale, which listed [Defendants] as a beneficiary.
(Doc. 1915 at 1-2).

6

In both motions, Plaintiffs’ also claim that documents related to these mortgage

7

transactions misrepresent MERS’s status and authority as a beneficiary, contain false

8

statements regarding assignments, and are robosigned. (Doc. 1913 at 3-6; Doc. 1914 at 3-

9

5). However, none of the proposed class definitions specifically incorporate or require

10

these additional allegations.

11

As discussed above, the Ninth Circuit Court of Appeals reversed only the claims

12

of robosigning contained within Count I of the CAC. In re Mortgage Elec. Registration

13

Sys., Inc., 754 F.3d at 781-84. All other claims or theories within Count I were dismissed

14

by this Court, and that dismissal was not reversed by the Court of Appeals. (Doc. 1602 at

15

5-6). This Court will not read into the Court of Appeals’ decision a reversal as to every

16

factual predicate of Count I on which the Court of Appeals was silent. Thus, only

17

allegations of robosigning survive as a potential basis for Plaintiffs’ proposed classes. In

18

each motion seeking class certification, Plaintiffs assert that many documents were

19

robosigned. (Doc. 1913 at 5; Doc. 1914 at 5). This is the sole remaining claim before the

20

Court. However, the arguments and evidence Plaintiffs present to satisfy the Rule 23(a)

21

requirements largely ignore this claim. (Doc. 1913; Doc. 1914). This abandonment of the

22

remaining claim is further demonstrated by Plaintiffs Robinson and Bilyea’s statement

23

that robosigning “does not form the basis of the class claims.” (Doc. 1941 at 1).

24

Additionally, the research performed by Dr. Wilson does not pertain to robosigning, the

25

sole claim remaining before the Court.

26

Defendants oppose Plaintiffs’ motion, claiming their class definitions identify tens

27

of thousands of potential plaintiffs, but fail to provide how they would identify who the

28

actual class members are. (Doc. 1928 at 7). Defendants claim that determining the actual

-7-

1

members of the class, as well as the applicable documents would be next to impossible,

2

requiring the Court to conduct “individualized ‘mini-trials’ just to determine who is a

3

member of the class.” Id. at 9. Additionally, Defendants claim that proving Plaintiffs’

4

robosigning claims would require a jury “to examine separately each document to

5

determine who signed it.” Id. at 14. In so arguing, Defendants seem to presume that a

6

robosigning requirement would be read into each class definition, even though the actual

7

definitions merely require that MERS signed the relevant documents.3

8

Plaintiffs reply by asserting that class members can be determined using “public

9

records and/or . . . electronic information within the possession of [] Defendants.” (Doc.

10

1947 at 3; Doc. 1948 at 3). Plaintiffs further claim that whether documents contain

11

language assigning the note “can be made using ‘objective criteria’ and a simple count of

12

the assignments (and related documents) at issue make it ‘administratively feasible’ to

13

determine the members of the class.” (Doc. 1947 at 3; Doc. 1948 at 3). To support their

14

argument, Plaintiffs submitted a statement of work (Doc. 1947-1 at 2; Doc. 1948-1 at 2)

15

from a research plan used to generate a database of MERS assignments to Defendants.

16

(Doc. 1947 at 3; Doc. 1948 at 3). However, Plaintiffs do not satisfy the prerequisite that

17

the proposed classes be adequately defined and clearly ascertainable.

18

1.

The Proposed Classes Are Not Adequately Defined

19

Both motions claim the purpose of class certification is to recover statutory

20

damages “on behalf of Arizona property owners arising from Defendants’ violation of

21
22
3

23
24
25
26
27
28

“Robosigning” as used by the Court of Appeals, and this Court, is inartful
because it has never been exactly defined. However, this Court finds that what
“robosigning” does not encompass is every assignment ever made by MERS. In other
words, the Court of Appeals clearly intended robosigning to include a problem with the
execution of the documents, not that by merely executing a document MERS was
robosigning.
The Court reaches this definitional conclusion in part based on the plain language
of the Court of Appeals’ decision. Further, the Court of Appeals found that
“robosigning” is a cause of action. However, the Arizona Courts have held that MERS
assigning a note and deed of trust is not a cause of action. See Sitton v. Deutsche Bank
Nat. Trust Co., 31 P.3d 237, 243 ¶28 (Ariz. Ct. App. 2013). Thus, the definition of
“robosigning” as used by the Ninth Circuit Court of Appeals must be something more
than MERS assigning a note or deed of trust.
-8-

1

A.R.S. § 33-420 while conducting foreclosure proceedings in the State of Arizona.”4

2

(Doc. 1913 at 1; Doc. 1914 at 1). However, neither proposed class definition expressly

3

limits class membership to those who have been foreclosed or noticed for trustee sale.

4

Thus, the proposed class definitions would expand the potential class plaintiffs to include

5

members who do not fall within the scope of Count I of the CAC. (Doc. 1424 at 7-25).

6

The Ninth Circuit has not addressed the issue as to whether a class can be expanded

7

beyond those harmed as identified in the complaint or the full scope of a district court’s

8

discretion to modify a class definition. See, e.g., Bueche v. Fid. Nat. Mgmt. Servs., LLC,

9

No. 2:12-CV-01114 JAM, 2014 WL 2468601, at *3 (E.D. Cal. June 2, 2014).

10

Nonetheless, this Court will not accept such expanded definitions as proposed by

11

Plaintiffs because the definitions are inconsistent with their own arguments regarding

12

who is impacted by Defendants’ alleged wrong. Further, the Court will not exercise its

13

discretion by redefining the classes in this case, because as discussed below, the even

14

redefined classes (consistent with both the complaint and the mandate) would still suffer

15

from deficiencies that are fatal to class certification.

16

Additionally, Plaintiffs seek to establish classes in which any person who meets

17

the criterion, from either February 21, 2008 (Doc. 1913 at 1) or January 27, 2009 (Doc.

18

1914 at 1) to the present, would be a class member. This timeframe could potentially add

19

new members daily. Even if the Court of Appeals had reversed the entirety of the

20

dismissal of Count I, an open ended time period would frustrate any attempt for a precise

21

definition of either proposed class. See Clay, 188 F.R.D. at 490 (finding a proposed class

22

of “all persons in the United States who, as children, purchased and smoked cigarettes

23
24
25
26
27
28

4

Other statements appearing within the motions support the limitation to only
those members foreclosed using MERS. For example, “Plaintiffs’, and similarly situated
class members’, properties were foreclosed upon pursuant to Arizona’s nonjudicial
foreclosure, or Deed of Trust, statute, A.R.S. § 33-801, et. seq.” (Doc. 1913 at 3, n. 2;
Doc. 1914 at 3, n. 2). Further, at oral argument, Plaintiffs’ counsel specified that the
Robinson class was targeted at circumstances in which the note and deed of trust were
assigned to a new person or entity prior to foreclosure and the Stejic class was targeted at
circumstances where MERS directly forecloses. Thus, even though neither class
definition specifically requires a foreclosure, it is clear that Plaintiffs intend such a
limitation.
-9-

1

designed, manufactured, promoted, or sold by the defendants” to be overly broad and

2

amorphous). Thus, Plaintiffs’ proposed classes are not adequately defined.5

3

2.

The Proposed Classes Are Not Clearly Ascertainable

4

The proposed classes are not ascertainable because there is no reasonable way to

5

determine which documents are relevant to Plaintiffs’ allegations. See, e.g., Martin v.

6

Pac. Parking Sys. Inc., 583 F. App’x 803, 804 (9th Cir. 2014) cert. denied, 135 S. Ct.

7

962, 190 L. Ed. 2d 833 (2015) (holding the district court did not abuse its discretion by

8

finding plaintiff’s “proposed class was not ascertainable because there was no reasonably

9

efficient way to determine which of the hundreds of thousands of individuals who used

10

parking lots” fell within the plaintiff’s proposed class).

11

Here, Plaintiffs hired Real Estate Services and Technology (“REST”), a research

12

company run by Fred Popke, to locate documents consistent with their claims concerning

13

MERS assignments. (Doc. 1913-8 at 2). REST determined there were 83,000 primary

14

and 418,000 secondary documents which pertained to assignments from MERS to one of

15

the named Defendants within Maricopa, Pima, and Pinal counties from February 15,

16

2008 through July 13, 2015.6 (Doc. 1913-8 at 2, 4; Doc. 1914-5 at 2, 4). From the 83,000

17

primary documents, REST acquired copies of 1,200 assignments of a Deed of Trust.

18

(Doc. 1913-8 at 3; Doc. 1914-5 at 3). Researchers gave a “higher priority to

19
20
21
22
23
24
25
26
27
28

5

The Court has already concluded that the proposed classes are not adequately
defined for the reasons stated. Moreover, the Court notes that at oral argument Defendant
Aurora argued that Plaintiffs made no effort to identify which Defendants would be in
which classes. Aurora further argued that as to all the proposed class members, Plaintiffs
made no effort to show Aurora was involved in each transaction. Plaintiffs did not
dispute that they treated all Defendants as a single entity under an agency or acting in
concert theory. This Court disagrees with Plaintiffs’ theory that if Aurora was an actor in
one MERS’s foreclosure in Arizona, Aurora is properly swept into all MERS
transactions. Thus, for this additional reason the Court finds the class is not adequately
defined and is overbroad because it would sweep in various Defendants who were totally
uninvolved in certain transactions.
6

Mr. Popke stated in his declaration that the initial search of indexed information
included transactions between MERS and America’s Servicing Company, Bank of
America, Central Mortgage Company, Quality Loan Service, US Bank, and Wells Fargo.
(Doc. 1913-8 at 3; Doc. 1914-5 at 3). Additionally, the figures presented by Mr. Popke
were “rough/rounded counts,” and included information obtained from third parties.
(Doc. 1913-8 at 3; Doc. 1914-5 at 3).
- 10 -

1

[a]ssignments that were associated with a subsequent Re-Conveyance, Notice of Trustee

2

Sale and/or Trustee Sale instrument as a supplemental (related) instrument.”7 (Doc. 1913-

3

8 at 3; Doc. 1914-5 at 3). The number of documents selected from each county depended

4

upon the number of MERS assignments within that county. (Doc. 1913-8 at 3; Doc.

5

1914-5 at 3). REST then individually inspected each document to extract any information

6

not supplied by public records or third party data providers. (Doc. 1913-8 at 4; Doc.

7

1914-5 at 4).

8

The compiled data was then applied to a “statistical sampling scheme” developed

9

by Dr. Jeffrey R. Wilson, a statistician hired by Plaintiffs to develop a method to

10

determine what “sample was a good representative of the entire population.” (Doc. 1913-

11

7 at 4). Dr. Wilson applied his sampling scheme to the data provided by REST to

12

“determine what population percentages and population totals of the MERS-related

13

assignments in the 83,000 recorded primary documents included an assignment of the

14

note.” (Doc. 1913-7 at 5). Dr. Wilson concluded that between 50,038 and 56,234 of the

15

83,000 primary documents contained assignments. (Doc. 1913-7 at 38).

16

Statistical sampling is widely accepted among the circuits to prove liability,

17

including this one. See Jimenez v. Allstate Ins. Co., 765 F.3d 1161, 1167 (9th Cir. 2014)

18

cert. denied, 135 S. Ct. 2835 (2015). However, use of statistical sampling for class

19

certification would be inappropriate in this case for the following reasons.

20

First, the scope of REST’s research appears to be centered entirely on assignments

21

by MERS to one of the named Defendants. (Doc. 1928-2 at 64). As discussed above, this

22

Court has already ruled on the allegations concerning the validity of MERS assignments.

23

(Doc. 1609 at 5-6). What is more, Plaintiffs claim the proposed classes are composed of

24

Arizona property owners who have been subject to foreclosure proceedings. (Doc. 1913

25

at 1; Doc. 1914 at 1). However, the research conducted by REST focused on locating and

26

identifying assignments of Deeds of Trust, without identifying which assignments

27
28

7

There is no indication how many of the 1,200 sample documents related to a
subsequent foreclosure.
- 11 -

1

subsequently led to foreclosure.

2

Plaintiffs do not represent that the same method used by REST to identify

3

assignment documents would be applicable to identifying robosigned documents. Nor do

4

Plaintiffs represent that Dr. Wilson’s statistical sampling scheme would be equally useful

5

if applied to allegedly robosigned documents. Thus, the relevance of Dr. Wilson’s results

6

to claims of robosigning serve only to demonstrate the difficulty in locating potentially

7

fraudulent documents. Further, as indicated above, Plaintiffs’ proposed class definitions

8

do not have robosigned documents as a requirement for class membership.

9

Second, Plaintiffs seek class certification using a statistical study based on a

10

sample of a sample. Essentially, Plaintiffs seek to represent that 1,200 documents, chosen

11

from 83,000 documents from three counties is representative of all documents in all

12

fifteen counties in Arizona. Dr. Wilson declared this sample size was sufficient to

13

“ensure that the sample was a good representative of the entire population.” (Doc. 1913-8

14

at 4). However, Dr. Wilson admitted in his deposition that the sample size was intended

15

to represent a population of 83,000 documents. (Doc. 1928-2 at 75). Had Dr. Wilson’s

16

study focused on robosigned documents, the population of documents could be

17

significantly different, because Section 33-420 applies to more than assignment

18

documents. For example, the group could be much larger if Plaintiffs could allege that all

19

such documents were robosigned. The larger population would require a larger sample

20

size to adequately represent it. Additionally, a potential plaintiff could have several

21

documents that would fall under Section 33-420; which would further complicate any

22

attempt to use statistical sampling to identify potential class members. Conversely, if not

23

all such documents are alleged to be robosigned, this group sweeps non-class members

24

into the statistical sample.

25

Third, Mr. Popke admitted in his deposition that counties provide varying levels of

26

information for download to the public and use different computer programs, all of which

27

depends upon the accuracy of the data entered by the counties or third parties. (Doc.

28

1928-2 at 65-66). Thus, the document population was limited to documents accessible by

- 12 -

1

researchers and consequently the sample size necessary to represent that pool was based

2

upon this limitation. Id. Even if Plaintiffs’ research had focused on identifying

3

robosigned documents, allowing the use of statistical sampling here would be

4

inappropriate, because the information relied on is inconsistent or unreliable. See Pierce

5

v. Cty. of Orange, 526 F.3d 1190, 1200 (9th Cir. 2008) (holding that using statistical

6

sampling to calculate damages is inappropriate when the information used is unreliable).

7

Fourth, and possibly most significant, if this Court redefined the class to be limited

8

to the claim remaining after remand in Count I, any documents obtained from public

9

records or Defendants would require examination of each individual document to

10

determine which documents were forgeries or robosigned. Defendants correctly argue

11

that this would require an individual analysis of hundreds of thousands of documents to

12

determine potential class members.8 (Doc. 1920 at 10, 14). Even if Plaintiffs could obtain

13

the records of every foreclosure related real estate transaction in Arizona, they would still

14

be left with the task of identifying which ones are fraudulent under Section 33-420.

15

Because of the foregoing, Plaintiffs’ proposed classes are not adequately defined

16

or clearly ascertainable. Additionally, the Court will not utilize its discretionary powers to

17

redefine the membership of the proposed classes because Plaintiffs have not met the

18

prerequisites for Rule 23 even using a hypothetically redefined class. Thus, Plaintiffs’

19

motions for class certification will be denied because these prerequisites are not met.

20

Alternatively, the Court will consider the other factors required for class certification

21

under Federal Rule of Civil Procedure 23 to determine whether class certification would

22

be appropriate under those tests.

23

B.

24

Named Representatives Fall Within the Proposed Class
1.

Numerosity

25

The focus of the numerosity requirement involves an inquiry into whether joinder

26

of all potential plaintiffs would be impracticable. Fed. R. Civ. P. 23(a)(1). Numerosity

27
8

28

Dr. Wilson admitted in his deposition that individual inspection of each the
83,000 primary documents identified by Mr. Popke would be “beyond comprehension.”
(1928-2 at 75).
- 13 -

1

requires examination of the facts of each case and does not impose any absolute

2

limitation. Gen. Tel. Co. of the Northwest, Inc. v. Equal Employment Opportunity

3

Comm’n, 446 U.S. 318, 330 (1980). While no absolute limits exist, the Supreme Court

4

has suggested that a class of fifteen members is too small to meet the numerosity

5

requirement. Harik v. California Teachers Ass’n, 326 F.3d 1042, 1051 (9th Cir. 2003)

6

(citing Gen. Tel., 446 U.S. at 330).

7

Plaintiffs assert that the numerosity requirement of Rule 23(a)(1) is satisfied

8

because there are “over 50,000 potential plaintiffs” in each of the proposed classes. (Doc.

9

1913 at 7; Doc. 1914 at 7). However, the only evidence offered by Plaintiffs is the

10

statistical study of Dr. Wilson. As discussed above, the study has no application in the

11

context of robosigning and forgery. A trial court conducting Rule 23(a) analysis “may

12

rely on common sense assumptions to support findings of numerosity.” Alberto v. GMRI,

13

Inc., 252 F.R.D. 652, 660 (E.D. Cal. 2008) (citing Manual for Complex Litigation

14

(Fourth) § 23.22(3) (2008)). However, nothing in the facts before the Court would permit

15

such an assumption. Plaintiffs’ assertion that the Rule 23(a)(1) requirement is satisfied

16

appears to be based upon the research of REST and Dr. Wilson concerning MERS

17

assignments. (Doc. 1913 at 7; Doc. 1914 at 7). In other words, Plaintiffs have offered no

18

evidence of robosigning or forgery beyond the named Plaintiffs. Thus, Plaintiffs have

19

failed to satisfy the numerosity requirement of Rule 23(a)(1).

20

2.

Commonality

21

To satisfy the commonality requirement, there must be questions of law and fact

22

common to the members of the class. Fed. R. Civ. P. 23(a)(2). The answer to this shared

23

question must “resolve an issue that is central to the validity of each one of the [class

24

members’ claims] in one stroke.” Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541, 2551

25

(2011); Edwards v. First Am. Corp., 798 F.3d 1172 (9th Cir. 2015). In other words,

26

commonality requires that a plaintiff must “have suffered the same injury” as the

27

members of the proposed class. Falcon, 457 U.S. at 157.

28

Defendants argue that Plaintiffs cannot satisfy the commonality requirement

- 14 -

1

because robosigning claims “cannot be tried for the class as a whole with common

2

evidence.” (Doc. 1920 at 14; Doc. 1928 at 14). Defendants argue that each document

3

would have to be looked at individually for robosigning. (Doc. 1920 at 14; Doc. 1928 at

4

14). As discussed above, the Court agrees. Nonetheless, the Court need not consider this

5

argument in the context of commonality because Plaintiffs’ proposed class definitions do

6

not require robosigned or forged documents – only documents assigned by MERS.

7

Further, Plaintiffs’ experts have not analyzed commonality in terms of robosigning. As

8

the Court has held above, a class of only MERS documents is beyond the mandate as to

9

what states a claim. Accordingly, the Court cannot find commonality on robosigning

10

when the only evidence presented and classes proposed deal with MERS assignments.

11

Thus, the commonality requirement of Rule 23(a)(2) has not been satisfied.

12

3.

Typicality

13

The typicality requirement of Rule 23(a)(3) requires that the claims or defenses of

14

the representative parties be typical of those of the class. Fed. R. Civ. P. 23(a)(3).

15

Typicality does not require class members and named plaintiffs to suffer identical

16

injuries, only that “their claims be ‘typical’ of the class, not that they be identically

17

positioned to each other or to every class member.” Parsons v. Ryan, 754 F.3d 657, 686

18

(9th Cir. 2014); see also Ellis, 657 F.3d at 985 n. 9 (“Differing factual scenarios resulting

19

in a claim of the same nature as other class members does not defeat typicality.”).

20

Additionally, the United States Supreme Court has recognized that the requirements of

21

commonality and typicality “tend to merge.” Falcon, 457 U.S. at 157-158, n. 13. Both

22

requirements are used in “determining whether under the particular circumstances

23

maintenance of a class action is economical and whether the named plaintiff’s claim and

24

the class claims are so interrelated that the interests of the class members will be fairly

25

and adequately protected in their absence.” Id.

26

Defendants present various arguments that Plaintiffs’ claims lack typicality.

27

However, many of Defendants’ arguments are based upon MERS assignments and trustee

28

designations, not the allegations of robosigning remanded to this Court by the Ninth

- 15 -

1

Circuit Court of Appeals. Defendants claim that Plaintiffs’ lack typicality because

2

“whether a plaintiff can satisfy the elements of A.R.S. § 33-420 can only be determined

3

‘on a case-by-case basis.’ ” (Doc. 1928 at 5; 1929 at 5) (quoting Coventry Homes, Inc. v.

4

Scottscom P’ship, 745 P.2d 962, 966 (Ariz. Ct. App. 1987)). Plaintiffs reply by stating

5

that Defendants “attempt to distinguish each Plaintiff’s situation based on whether the

6

underlying deeds of trust named MERS, whether the plaintiffs challenge(d) their trustee

7

sales, and the number of documents recorded against the plaintiffs’ and the class

8

members’ properties does not defeat typicality” (Doc. 1941 at 4).

9

Plaintiffs’ proposed classes do not satisfy the typicality requirement of Rule

10

23(a)(3). The proposed classes do not require robosigning. As noted above, Plaintiffs’

11

experts did not analyze robosigning or present evidence to suggest robosigning had

12

occurred to proposed class members. The only link between Plaintiffs and proposed class

13

members are MERS assignments, which fall outside the mandate to this Court. Thus,

14

Plaintiffs have failed to satisfy the typicality requirement of Rule 23(a)(3).

15

4.

Adequacy

16

Rule 23(a)(4) permits certification of a class if “the representative parties will

17

fairly and adequately protect the interests of the class.” Fed. R. Civ. P. 23. Under Rule

18

23(a)(4), the adequacy of a class representative to represent the members of the class

19

requires a district court to consider: “(1) do the proposed representatives have any

20

conflicts of interest with the other class members and (2) will the named plaintiffs and

21

their counsel prosecute the action vigorously on behalf of the class?” Evon v. Law Offices

22

of Sidney Mickell, 688 F.3d 1015, 1031 (9th Cir. 2012) (quoting Hanlon v. Chrysler

23

Corp., 150 F.3d 1011, 1020 (9th Cir. 1998)).

24

a.

Conflicts of Interest

25

Defendants claim Plaintiffs are inadequate class representatives because they

26

“unilaterally decid[ed] that class members will only be allowed to recover statutory, not

27

actual, damages.” (Doc. 1918 at 13-14; Doc. 1920 at 4-5; Doc. 1928 at 9; Doc. 1929 at

28

9). Plaintiffs reply that any class member seeking other than statutory damages may opt

- 16 -

1

out of the class. (Doc. 1946 at 2-3; Doc. 1947 at 3-4; Doc. 1948 at 4). Plaintiffs note that

2

Section 33-420 permits that an action for quiet title may be joined with an action for

3

damages. (Doc. 1947 at 4; Doc. 1948 at 4).

4

The Court finds that Plaintiffs’ pursuit of statutory damages does not render

5

Plaintiffs inadequate to act as class representatives. Class members seeking actual

6

damages or to quiet the title of their property may opt out of the class or fall within a

7

subclass.

8

Defendants also claim Plaintiffs are inadequate because Plaintiffs’ agreements

9

with their attorneys provide that if they lose or their attorneys withdraw, Plaintiffs will be

10

responsible for all costs and expenses. (Doc. 1928 at 10; Doc. 1929 at 10). Defendants

11

claim this provides little incentive for Plaintiffs to question the decisions of their

12

attorneys or prosecute claims on behalf of class members. (Doc. 1928 at 10; Doc. 1929 at

13

10). Plaintiffs dispute Defendants’ claim, pointing out that Section 33-420 provides for

14

the recovery of attorney’s fees, which can be spread over the class. (Doc. 1947 at 1-2;

15

Doc. 1948 at 1-2).

16

The Court disagrees with Defendants’ assertion that Plaintiffs have a financial

17

incentive to follow the decisions of their attorneys. (Doc. 1928 at 10; Doc. 1929 at 10).

18

As Plaintiffs point out, any costs and attorney fees are recoverable under the statute.

19

(Doc. 1947 at 2; Doc. 1948 at 2). Because any recovery of costs and attorney fees

20

depends upon the success of an action under Section 33-420, Plaintiffs have a financial

21

incentive to do everything possible to ensure the success of the class, including exerting

22

control over their attorneys.

23

Defendants further claim that Plaintiffs are inadequate representatives because

24

they lack basic knowledge of their claims and unduly rely on their counsel. (Doc. 1920 at

25

5-6; Doc. 1928 at 10; Doc. 1929 at 9-10). Plaintiffs deny these claims, and point to

26

Plaintiffs’ depositions as evidence they possess sufficient knowledge to act as class

27

representatives. (Doc. 1946 at 2-3; Doc. 1947 at 5; Doc. 1948 at 4).

28

The Court recognizes that complex litigation can be confusing to the lay person.

- 17 -

1

The present case concerns complex real estate transactions that would likely be confusing

2

even to persons with considerable expertise. Additionally, this case has continued for

3

over five years in state and federal court, and at one time included numerous member

4

cases. See In re Conseco Life Ins. Co. LifeTrend Ins. Sales & Mktg. Litig., 270 F.R.D.

5

521, 531 (N. D. Cal. 2010) (“Plaintiffs are laypersons and cannot be expected to define

6

the scope of the class or name all of the causes of action in more precise terms. The fact

7

that they are familiar with the basis for the suit and their responsibilities as lead plaintiffs

8

is sufficient to establish their adequacy.”). The Court finds that Plaintiffs have

9

demonstrated sufficient knowledge to perform as class representatives.

10

Plaintiffs have sufficiently shown that they will fairly and adequately protect the

11

interests of the class and that there are no conflicts of interests. Thus, Plaintiffs are

12

adequate class representatives for the proposed class.

13

b.

Represented by Competent and Qualified Counsel

14

Plaintiffs are represented by Valerie R. Edwards and William A. Nebeker, both of

15

whom are attorneys at the law firm of Koeller, Nebeker, Carlson & Haluck, L.L.P. (Doc.

16

1913-5 at 2; Doc. 1913-6 at 2; Doc. 1914-1 at 79-84). Plaintiffs assert these attorneys

17

have “substantial experience in large complex litigation matters, including class actions,

18

and the firm has the necessary resources to prosecute this action on behalf of [P]laintiffs

19

and the class members.” (Doc. 1913 at 11-12; Doc. 1914 at 11-12). Ms. Edwards has

20

been a member of the Arizona bar since 1996, and has served as counsel on a number of

21

class actions, including several dealing with predatory lending. (Doc. 1913-6 at 2-3). Mr.

22

Nebeker has been a member of the Arizona bar since 1977, and has litigated a number of

23

class actions, many of which were in the federal courts. (Doc. 1913-5 at 2-3). The Court

24

finds Ms. Edwards and Mr. Nebeker to be competent and qualified counsel to represent

25

the interests of the class.

26

5.

Rule 23(b) Requirements

27

Next, the Court turns to the Rule 23(b) requirements. The circumstances of this

28

case weigh heavily against a finding that “a class action is superior to other available

- 18 -

1

methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23(b)(3).

2

As discussed above, determining class membership would require an individual

3

examination of potentially hundreds of thousands of documents. Calculating damages

4

would also be difficult, since each class member could be entitled to multiple awards of

5

statutory damages based upon the number of robosigned documents and statistics could

6

not be used to determine the correct amount to be awarded. See Jimenez, 765 F.3d at

7

1167 (“[S]tatistical sampling and representative testimony are acceptable ways to

8

determine liability so long as the use of these techniques is not expanded into the realm of

9

damages.”). Additionally, A.R.S. § 33-420 also provides for recovery of attorney fees and

10

costs. Because of the unique difficulties in determining whether a document falls under

11

Section 33-420 as well as determining damages, a class action is not superior to other

12

methods for fairly and efficiently adjudicating this case. See Fed. R. Civ. P. 23(b)(3).

13

V.

Conclusion

14

Plaintiffs’ proposed classes do not satisfy the requirements of Rule 23(a),

15

particularly the requirements of “numerosity,” “commonality” and “typicality.”

16

Additionally, the proposed classes do not satisfy the prerequisites that the classes be

17

adequately defined and clearly ascertainable. Finally, the proposed classes do not satisfy

18

Rule 23(b). Thus, for the foregoing reasons, Plaintiffs’ motions for class certification are

19

denied.

20

Accordingly,

21

IT IS ORDERED that Plaintiffs Sally Robinson-Burke, Rosa Silvas and Thomas

22
23
24

Bilyea’s Motion to Certify Class (Doc. 1913) is DENIED.
IT IS FURTHER ORDERED that Plaintiffs Milan Stejic and Maria Hernandez’s
Motion to Certify Class (Doc. 1914) is DENIED.

25

IT IS FURTHER ORDERED that Defendant Aurora Loan Services, LLC’s

26

Motion for Leave to File Sur-Reply (Doc. 1952; Doc. 50 in CV 10-1547-PHX-JAT) is

27

DENIED as unnecessary because of this Court’s denial of class certification.

28

- 19 -

1
2
3

IT IS FINALLY ORDERED that the Clerk of the Court shall file a copy of this
Order in all member cases listed above.
Dated this 21st day of December, 2015.

4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

- 20 -