Gonzalez et al v. Abercombie & Fitch Co. et al

Filing 233

ORDER AFFIRMING SPECIAL MASTER'S DECISION (SI, COURT STAFF) (Filed on 6/23/2011)

1
2
3
4
5

IN THE UNITED STATES DISTRICT COURT

6

FOR THE NORTHERN DISTRICT OF CALIFORNIA

7
8

EDUARDO GONZALEZ, et al.,

9
United States District Court
For the Northern District of California

10
11
12

No. C 03-2817 SI
Related Cases: No. C 04-4730 SI, No. C 044731 SI

Plaintiffs,
v.

ORDER AFFIRMING SPECIAL
MASTER’S DECISION

ABERCROMBIE & FITCH CO., et al.,
Defendants.

/

13
14

On June 3, 2011, the Court held a hearing on plaintiffs’ motion to modify the consent decree and

15

partial appeal from the Special Master’s April 15, 2011 Order. For the following reasons, the Court

16

DENIES plaintiffs’ motion and AFFIRMS the Special Master’s Order.

17
BACKGROUND

18
19

On August 28, 2009, plaintiffs initiated an enforcement action before the Special Master.

20

Plaintiffs raised two claims: (1) that Abercrombie has failed to comply with its obligations under Section

21

XII.A.2 of the consent decree because the selection criteria for Abercrombie’s Model position caused

22

adverse impact in the selection rates of Latinos and African-Americans, and Abercrombie had not used

23

Best Efforts to determine whether alternative selection criteria or an alternative use of selection criteria

24

would cause less adverse impact and be equally predictive of performance; and (2) that Abercrombie

25

has failed to comply with its obligations under Section XIV of the consent decree because it failed to

26

use Best Efforts to reach the Benchmarks for hiring of African-Americans, African-American Females,

27

and Latinos to the Model position.

28

On April 1, 2011, the Special Master held a hearing, and on April 15, 2011, he issued a 42 page

decision. On the first claim, the Special Master found that plaintiffs had met their burden of showing

2

that the selection criteria caused adverse impact in the selection of African-Americans, African-

3

American women, and Latinos to the Model position. The Special Master also found that Abercrombie

4

had not met its burden to “show[] that it has used Best Efforts to determine whether alternative job-

5

related selection criteria or whether an alternative use of job-related selection criteria has less adverse

6

impact and would be equally predictive of job performance.” Consent Decree § XII.A.2. The Special

7

Master ordered Abercrombie to “make an adverse impact review and assessment of its selection criteria

8

(which includes its group interview procedures) relative to African-Americans and Latinos and

9

determine whether alternative job-related selection criteria or whether an alternative use of job-related

10
United States District Court
For the Northern District of California

1

selection criteria has less adverse impact and would be equally predictive of job performance.” Special

11

Master’s Order at 41-42.

12

However, on the first claim the Special Master denied two forms of relief that the plaintiffs

13

sought. Plaintiffs sought an order directing Abercrombie to implement the alternative job-related

14

selection criteria or alternative uses of the existing model selection procedure for a period of no less than

15

two years. The Special Master denied that relief, finding that it was not authorized by the consent

16

decree:

17
18
19
20
21
22
23
24
25
26
27
28

The Decree language at issue provides:
If the selection criteria cause such adverse impact and are challenged
under Section X of this Decree, the selection criteria shall be affirmed if
Abercrombie shows that it has used Best Efforts to determine whether
alternative job-related selection criteria or whether an alternative use of
job-related selection criteria has less adverse impact and would be
equally predictive of job performance.
This heavily negotiated Decree speaks in terms of the selection criteria being
“affirmed” if Abercrombie has carried its Best Efforts burden. As discussed above, I
have found that Abercrombie has not met its XII.A.2. Best Efforts burden. But the
Decree does not authorize the relief outlined by Plaintiffs. Contrary to Plaintiffs’
proposed relief, there is nothing in Section XII.A.2. of the Decree that requires
Abercrombie to change its original selection criteria. What the Decree states is that
Abercrombie can comply with its Best Efforts obligation by showing that it has used
Best Efforts to determine whether other selection criteria (or alternate uses of selection
criteria) have less impact and are equally predictive of job performance – i.e. do a
component analysis and using that additional information to determine whether there are
alternative job-related criteria (or selection criteria uses) having less impact that are
equally predictive. Once Abercrombie makes that determination, then it has complied
with its Best Efforts obligation under the Decree.
2

1

4

Moreover, the Plaintiffs’ proposed relief assumes that there are in fact
alternatives with lesser impact that are equally predictive. However, there is no basis in
the record to support the conclusion. In fact, the record suggests just the opposite in that
the jointly selected organizational psychologist has opined repeatedly that there are no
other such alternatives. While her opinion may change after completion of the
component analyses, there is nothing in the record that necessarily leads to that
conclusion.

5

Special Master’s Order at 40-41. The Special Master also denied plaintiffs’ request that Abercrombie

6

be required to conduct “inter-rater reliability analyses” as part of the adverse impact review. The

7

Special Master held that Ҥ XII.A.2 speaks to selection criteria and their uses. Inter-rater reliability

8

studies do not fall within the scope of this Decree provision and, in any event, I find that the evidence

9

does not support a finding that inter-rater reliability issues are a potential cause of adverse impact.” Id.

10

at 36-37. The Special Master noted that “[n]either of the parties cited any research or other evidence

11

to suggest that there is a relationship between adverse impact and inter-rater reliability.” Id. at 15.

2

United States District Court
For the Northern District of California

3

12

On plaintiffs’ second claim under Section XIV of the consent decree, the Special Master rejected

13

plaintiffs’ contention that Abercrombie failed to use Best Efforts to reach the Benchmarks for hiring of

14

African-Americans, African-American Females, and Latinos to the Model position.

15
16
17
18
19
20
21
22
23
24
25
26
27
28

Plaintiffs’ position in essence is that Abercrombie’s obligations to use Best
Efforts under XII.A.2 coincides with its Best Efforts obligations under XIV. Thus
plaintiffs contend that once it has established that the Benchmark obligations were not
satisfied, then Abercrombie was required under the Uniform Guidelines to reexamine the
PSP on a component analysis. In Plaintiffs’ view, given Abercrombie’s decision not to
undertake the component analysis and instead to continue to rely on efforts that had not
been wholly successful in achieving Benchmarks, its plan could not be found to be
reasonably designed and implemented to achieve the Benchmarks. I disagree with
Plaintiff’s position.
I find that Abercrombie did use Best Efforts to meet its Benchmark obligations
under Section XIV. Abercrombie’s Best Efforts obligations under Section XII.A.2. and
Section XIV are not, as argued by Plaintiffs, fungible. The former is aimed at a very
narrow issue of concern centering around the hiring selection criteria adopted by
Abercrombie as part of its new hiring procedures. Section XII.A.2. tracks in many
respects the obligations of parties under Title VII and the Uniform Guidelines under the
adverse impact model. While changing the respective burdens of the parties in some
respects, the adverse impact, causation, and alternative criteria concepts of Section
XII.A.2. track those found in Title VII and used in adverse impact cases under Title VII.
Thus, interpreting section XII.A.2. consistent with Title VII selection concepts and
principles is logical and consist[ent] with the intent of the parties [as] evidenced by the
structure and language of the Decree.
But the Benchmark provision is different. The Benchmarks are not quotas. They
are simply targets established to afford guidance as to whether Abercrombie is making
selection decisions in a way as to afford equal employment opportunity. Section XIV
is structured differently and serves a different purpose than the Section XII provision.
3

1
2
3
4

It does not adopt Title VII standards but simply looks to whether Abercrombie had a
plan that was properly funded and implemented that was reasonably designed to meet
the Benchmarks. Abercrombie clearly had such a plan (which it updated from time to
time) to reach the Benchmarks and as outlined above it and ATP successfully
implemented that plan. That point is clearly made by among other things:
•

Abercrombie’s achievement of most of the Model Hiring Benchmarks for
females and Asian-Americans throughout the life of the Decree.

•

Abercrombie’s increase in minority representation in the Model position
during the life of the Decree. During the life of the Decree, Abercrombie
has more than quadrupled the incumbency rate of African-Americans,
Asian-Americans, and Latinos in the Model position, and more than
quintupled the incumbency rate of people who self-identify as non-white.
More than half of those holding the Model position on October 31, 2010
were nonwhite.

•

Abercrombie’s hire rates of Latinos, African-Americans, AsianAmericans, and African-American Females into the Model position have
steadily risen since the inception of the Decree in April 2005.

•

Applicant rates of African-Americans and Latinos have also steadily
risen during the life of the Decree, reflecting Abercrombie’s successful
efforts to recruit and attract minority applicants.

5
6
7
8
9
United States District Court
For the Northern District of California

10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28

Special Master’s Order at 37-39.
In addition to ordering Abercrombie to make an adverse impact review and assessment of its
selection criteria, the Special Master recommended to the Court that the Court consider modification
of the consent decree to permit a limited extension of the decree for the purpose of making the review
and assessment. Id. at 42. The Special Master also ordered that at the completion of the review and
assessment, Abercrombie shall submit a report to plaintiffs and the Court concerning the review and
assessment. Id.
Plaintiffs filed a timely appeal of the Special Master’s order. Plaintiffs request that the Court
reverse the Special Master’s finding with respect to inter-rater reliability, as well as the Special Master’s
finding with respect to their claim under Section XIV. Plaintiffs also request that the Court modify the
consent decree by ordering Abercrombie to implement alternative Model selection criteria or alternative
use of the existing criteria, and extend the consent decree until the alternative criteria are fully
implemented and evaluated for a period of 18 months. Abercrombie did not file an appeal, opposes
plaintiffs’ motion, and states that it has begun conducting the review and assessment ordered by the
Special Master.
4

1

LEGAL STANDARD

2

Section X.B.2 of the consent decree states that “The Special Master shall have authority to

3

resolve all disputes arising under the Decree, subject to limitations and standards set forth in the

4

Decree.” Section X.B.4 contains a detailed process by which disputes are submitted to and decided by

5

the Special Master. Section X.B.7 states that “The decision rendered by the Special Master shall be

6

affirmed unless the Court determines that the Special Master made clearly erroneous findings of fact

7

or wrongly interpreted or applied the Consent Decree.”

8
9
United States District Court
For the Northern District of California

10

DISCUSSION
I.

Modification of consent decree

11

Plaintiffs request that the Court modify the consent decree by ordering Abercrombie to identify,

12

develop and implement alternative Model selection criteria or alternative use of the existing criteria, and

13

extend the decree until the alternative criteria are fully implemented and evaluated for a period of 18

14

months. Plaintiffs contend that the Court has the authority to modify the consent decree pursuant to

15

Section IX.B.2. of the decree, which provides that the decree may be modified “[i]n the event that

16

changed or other circumstances make modification of the Decree necessary to ensure its purposes are

17

fully effectuated.” Consent Decree § IX.B.2. The Court may modify the consent decree only upon a

18

showing by clear and convincing evidence that changed or other circumstances make a modification

19

necessary. Id. Plaintiffs argue that the Special Master’s finding that the selection criteria causes adverse

20

impact, and that Abercrombie has failed to use Best Efforts to consider less discriminatory alternatives,

21

constitute “changed circumstances” that modification is necessary to effectuate the purposes of the

22

consent decree.

23

Plaintiffs also argue that the Court has the inherent equitable authority to enforce the consent

24

decree such that it can go beyond the relief ordered by the Special Master. Plaintiffs contend that

25

regardless of the terms of the decree agreed upon by the parties, the Court has inherent “continuing

26

authority and discretion” to ensure compliance until Abercrombie has fulfilled its “express obligations.”

27

EEOC v. Local 580, Int’l Assoc. of Bridge, Structural & Ornamental Ironworkers, 925 F.2d 588, 593

28

(2d Cir. 1991). Plaintiffs argue that they have met the test that the Supreme Court has established for
5

1

modifying a consent decree, which requires plaintiffs to “satisfy the initial burden of showing a

2

significant change either in factual conditions or in the law warranting modification of the decree. . .

3

The district court must then determine whether the proposed modification is suitably tailored to resolve

4

the problems created by the changed factual or legal condition.” United States v. Asarco, Inc., 430 F.3d

5

972, 979 (9th Cir. 2005) (citing Rufo v. Inmates of Suffolk County Jail, 502 U.S. 367, 384, 391 (1992)).

6

Plaintiffs assert that the Special Master’s findings constitute a “significant change in circumstances”

7

sufficient to warrant modification of the decree. Thus, both of plaintiffs’ arguments in support of

8

modifying the consent decree rest on the contention that the Special Master’s findings of adverse impact

9

and Abercrombie’s failure to use Best Efforts under Section XII.A.2. are “changed circumstances” that

United States District Court
For the Northern District of California

10

warrant modification.

11

The Court is not persuaded by plaintiffs’ arguments. “A court should not ordinarily modify a

12

decree, . . . ‘where a party relies upon events that actually were anticipated at the time it entered into a

13

decree.’” Id. (quoting Rufo, 502 U.S. at 385). Here, the consent decree explicitly anticipates that

14

plaintiffs may challenge selection criteria if they cause adverse impact. Consent Decree § XII.A.2. The

15

consent decree also states that selection criteria “shall be affirmed if Abercrombie shows that it has used

16

Best Efforts to determine whether alternative job-related selection criteria or whether an alternative use

17

of job-related selection criteria has less adverse impact and would be equally predictive of job

18

performance.” Consent Decree § XII.A.2. The consent decree does not state what will happen if

19

Abercrombie does not show that it has used Best Efforts, and as the Special Master noted, the consent

20

decree does not require Abercrombie to change its selection criteria. By providing that plaintiffs could

21

challenge the selection criteria if they cause adverse impact, and stating the circumstances under which

22

the selection criteria would be affirmed, the parties effectively anticipated the situation presented here,

23

where the selection criteria are challenged and the Special Master finds that Abercrombie has not used

24

its Best Efforts to determine whether alternative job-related selection criteria or whether an alternative

25

use of job-related selection criteria has less adverse impact and would be equally predictive of job

26

performance. Thus, the Special Master’s findings do not constitute “changed circumstances” warranting

27

modification of the consent decree.

28

The Court also finds that plaintiffs’ proposed modification is not appropriate because it (1)
6

assumes that there is alternative selection criteria that causes less adverse impact, and (2) ignores the

2

fact that the consent decree requires that the determination of alternative selection criteria include a

3

determination that any such alternative selection criteria be “equally predictive of job performance.”

4

Consent Decree § XII.A.2. Plaintiffs request an order directing Abercrombie to “identify, develop, and

5

implement less discriminatory alternative model selection criteria or alternative use of the existing

6

criteria, which Abercrombie shall implement,” and that the decree be extended until the alternative

7

criteria have been fully implemented and evaluated for 18 months. Proposed Order (Docket No. 228-1).

8

The Special Master’s order requires Abercrombie “to make an adverse impact review and assessment

9

of its selection criteria (which includes its group interview procedures) relative to African-Americans

10
United States District Court
For the Northern District of California

1

and Latinos and determine whether alternative job-related selection criteria or whether an alternative

11

use of job-related selection criteria has less adverse impact and would be equally predictive of job

12

performance.” Special Master’s Order at 41-42. Unlike the broader relief sought by plaintiffs, the

13

Special Master’s more limited relief is tailored to and consistent with the consent decree.

14

The Court is sympathetic to plaintiffs’ argument that simply requiring Abercrombie to do the

15

component analysis of the selection procedure and consider less discriminatory alternatives, without

16

requiring that Abercrombie actually implement any such alternatives, does not ensure employment on

17

a non-discriminatory basis. At the hearing, Abercrombie’s counsel stated that if the results of the review

18

ordered by the Special Master show that there are alternative selection criteria that cause less adverse

19

impact and that are equally predictive of job performance, he believed that Abercrombie would

20

implement such alternative selection criteria. Abercrombie’s counsel also acknowledged that if

21

Abercrombie did not implement such alternative selection criteria, Abercrombie would be exposed to

22

liability in a future lawsuit.

23

At the hearing, plaintiffs’ counsel expressed concern that plaintiffs’ ability to participate in the

24

component study would be limited, and he suggested that Dr. Lundquist was not jointly selected by the

25

parties. However, as Abercrombie notes, notwithstanding plaintiffs’ assertion that they did not jointly

26

select Dr. Lundquist, the consent decree states that “[t]he parties stipulate and agree upon

27

Abercrombie’s retention of Kathleen Lundquist, a professional industrial organizational psychologist,

28

to develop a written job analysis and job-related criteria for each in-store store position.” Consent
7

1

Decree § XII.A.1. In addition, although plaintiffs want the ability to “evaluate” whatever alternative

2

selection criteria are created, the consent decree provides that plaintiffs are “provided an opportunity

3

for comment.” Consent Decree § XII.A.4, § XII.B.9.

4
5

II.

Inter-rater reliability analysis
The Special Master found that Abercrombie is not required to undertake an empirical analysis

7

of inter-rater reliability because (1) inter-rater reliability is outside the scope of Section XII.A.2, and (2)

8

because neither party had cited any research or other evidence to suggest that there is a relationship

9

between adverse impact and inter-rater reliability. Plaintiffs challenge this finding, arguing that the

10
United States District Court
For the Northern District of California

6

Special Master’s interpretation of the consent decree is incorrect, and that his factual findings are clearly

11

erroneous.

12

The Court affirms the Special Master’s decision. As the Special Master found, Section XII.A.2

13

addresses selection criteria and their uses, and does not mention inter-rater reliability. Although

14

plaintiffs cited literature to the Special Master and this Court regarding the connection between

15

reliability and validity, neither party cited any literature or research suggesting a connection between

16

inter-rater reliability and adverse impact.1 In addition, as Abercrombie notes, the evidence before the

17

Special Master and this Court shows that (1) Dr. Lundquist validated the PSP, (2) plaintiffs’ expert Dr.

18

Zedeck acknowledged that the work for Dr. Lundquist was within professional standards, and (3) at the

19

April 1, 2011 hearing, plaintiffs agreed that the PSP was properly validated by Dr. Lundquist. On this

20

record, the Court concludes that the Special Master did not err in denying this aspect of plaintiffs’

21

enforcement action.

22
23

III.

Best efforts under Section XIV

24

The Special Master found that Abercrombie made Best Efforts to meet the Model hiring

25

Benchmarks for African-American and Latino applicants, and that the Best Efforts obligations under

26
1

27
28

At the hearing, Abercrombie’s counsel also stated that an inter-rater reliability analysis could
not simply be folded into the empirical study that is underway because currently there is one “rater” per
group interview, whereas an inter-rater reliability analysis would require that there be multiple raters
per group interview.
8

1

Sections XII.A.2 and XIV were distinct. Plaintiffs contend that the Special Master erred because the

2

Special Master “failed to defer to the informed findings of Court-appointed Monitor Alvarez in his most

3

recent report to the Court declining to find that Abercrombie engaged in Best Efforts as to these groups

4

for the Model position.” Motion at 23:1-4. See Lee Decl. Ex. 4 (Monitor’s Report).
Abercrombie responds that the consent decree does not state, or even imply, that the Special

6

Master is required to defer to the Monitor. The consent decree provides that the Special Master is the

7

trier of fact and has the authority to resolve disputes, whereas the Monitor is not a trier of fact, does not

8

take or weigh evidence, and has no authority to make findings. Compare Consent Decree § XV (role

9

of Monitor) with § X.B.2, 4.g., 4.h., 7 (role of Special Master). Abercrombie also asserts that the

10
United States District Court
For the Northern District of California

5

Monitor did not state that Abercrombie failed to use Best Efforts, but instead that he was “reserv[ing]

11

judgment on the question of best efforts in Model hiring until Abercrombie has a final opportunity to

12

thoroughly examine all critical assumptions in the Model hiring process and address them directly.”

13

Lee Decl. Ex. 4 at 53. Abercrombie argues that the Special Master reviewed and weighed the Monitor’s

14

Reports, in addition to the extensive evidence submitted by the parties, and correctly held that

15

Abercrombie met its Best Efforts obligations under Section XIV.

16

The Court concludes that the Special Master did not err. The Court agrees with Abercrombie

17

that the consent decree does not require the Special Master to defer to the Monitor, and in any event,

18

the Monitor reserved judgment on the question of whether Abercrombie satisfied its Best Efforts

19

obligations with regard to Benchmarks. The Special Master found that while the Best Efforts obligation

20

under Section XII.A.2 tracks in many respects the obligations of parties under Title VII, Section XIV

21

does not adopt Title VII standards but simply looks to whether Abercrombie had a plan that was

22

properly funded and implemented that was reasonably designed to meet the Benchmarks.2 Plaintiffs do

23

not challenge any of the factual findings underpinning the Special Master’s conclusion that Abercrombie

24

had a plan that was reasonably designed to meet the Benchmarks, and that Abercrombie successfully

25

implemented that plan. As the Special Master noted, throughout the life of the consent decree,

26
27
28

2

The consent decree defines “Best Efforts” as “implementing and adequately funding a plan
reasonably designed to comply with all the specific objectives to which the best efforts are directed.”
Consent Decree § III.H.
9

1

Abercrombie achieved most of the Model Hiring Benchmarks for females and Asian-Americans

2

throughout the life of the consent decree, and more than quadrupled the incumbency rate of African-

3

Americans, Asian-Americans, and Latinos in the Model position.

4
5

CONCLUSION
For the foregoing reasons, the Court DENIES plaintiffs’ motion to modify the consent decree

7

and AFFIRMS the Special Master’s Order. Docket No. 228 in C 03-2817 SI; Docket No. 71 in C 04-

8

4731 SI. The Court modifies the consent decree to permit a limited extension of the decree for the

9

purpose of making the review and determinations ordered by the Special Master. At the completion of

10
United States District Court
For the Northern District of California

6

the review and required determinations, Abercrombie shall submit a report to plaintiffs and the Court

11

concerning the subject review and determinations.

12
13

IT IS SO ORDERED.

14
15

Dated: June 23, 2011
SUSAN ILLSTON
United States District Judge

16
17
18
19
20
21
22
23
24
25
26
27
28
10