Haydar v. Amazon Corporate, LLC et al
Filing
88
OPINION and ORDER Granting in Part and Denying in Part Defendant's 62 MOTION for Summary Judgment and 69 Supplemental Brief; Motions terminated: 62 MOTION for Summary Judgment filed by Amazon Corporate, LLC, Garret Gaw, Peter Faricy, Joel Mosby. Signed by District Judge Laurie J. Michelson. (TMcg)
UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF MICHIGAN
SOUTHERN DIVISION
ABDULLAH HAYDAR,
Plaintiff,
v.
Case No. 2:16-cv-13662
Honorable Laurie J. Michelson
Magistrate Judge Stephanie Dawkins Davis
AMAZON CORPORATE, LLC,
GARRET GAW,
PETER FARICY, and
JOEL MOSBY,
Defendants.
OPINION AND ORDER GRANTING IN PART AND DENYING IN PART
DEFENDANT’S MOTION FOR SUMMARY JUDGMENT [62, 69]
In 2012, the retail giant Amazon Corporate, LLC hired Abdullah Haydar as a senior
technology manager. Less than three years later, Amazon fired Haydar. Haydar, who is a U.S.
citizen of Syrian descent and a practicing Muslim, maintains that his termination was motivated
at least in part by those two protected characteristics. He also maintains that when he complained
to human-resource personnel about discrimination, Amazon retaliated. So Haydar sued Amazon,
an Amazon vice president, and two of his direct supervisors. His eight-count complaint asserts
discrimination and retaliation under both state and federal law.
Amazon seeks summary judgment on all eight counts. For the reasons explained below at
great length (a byproduct of the parties filing 177 exhibits spanning 3,500 pages), the Court will
grant Amazon’s motion in part and deny it in part. In particular, Haydar’s national-origin and
religious discrimination claims will proceed to trial; Haydar’s other four counts will be
dismissed.
I.
When, as here, defendants seek summary judgment, the Court presents the facts in the
light most favorable to the plaintiff, Haydar. Matsushita Elec. Indus. Co. v. Zenith Radio Corp.,
475 U.S. 574, 587 (1986). Unless using an individual defendant’s name is necessary for clarity,
the Court simply refers to Defendants as “Amazon.”
A.
Plaintiff Abdullah Haydar is a United States citizen of Syrian descent. (R. 68-74,
PageID.8123.) He is also a practicing Muslim. (R. 68-74, PageID.8123.) Before joining Amazon,
Haydar had considerable experience as a technology manager. (R. 68-74, PageID.8123.)
B.
In November 2012, Haydar began working for Amazon as a “Senior Manager” in the
Marketplace organization. (See R. 68, PageID.6140, 6451.) At that time, Marketplace consisted
of somewhere between 800 and 7,000 employees—or, in Amazon parlance, Amazonians—and
was (and still is) headed by Defendant Peter Faricy. (R. 68-45, PageID.6743; R. 68-134,
PageID.8896.) Haydar was hired as a Level 7 (“L7”) Amazonian and reported to Ramiah
Kandasamy, an L8. (R. 68-1, PageID.6146.) Although Faricy was an L10, Amazon had no L9s
so Kandasamy reported directly to Faricy. (R. 68-1, PageID.6147.)
After about six months on the job, Kandasamy gave Haydar an unofficial performance
evaluation. (R. 68-4, PageID.6497.) To understand the evaluation, it is necessary to explain
Amazon’s rating system.
Amazonians (or managers at least) are assessed in three parts. The “performance” part
has to do with the manager’s goals, including how many projects the manager and his teams
successfully completed over the year. In terms of performance, a manager receives one of five
2
ratings: Outstanding, Exceeds, Achieves, Improvement Needed, and Unsatisfactory. (R. 68-79,
PageID.8229.) The second part—and the one most critical to this case—corresponds to the
manager’s “leadership.” (R. 68-79, PageID.8229.) The manager receives one of these three
leadership ratings: Role Model, Solid Strength, or Development Needed. (R. 68-79,
PageID.8229.) The third part of the assessment is the manager’s “growth potential”; an
Amazonian’s growth potential can be High, Medium, or Limited. (R. 68-79, PageID.8228.)
Plugging the ratings for the three parts into a matrix effectively gives the Amazonian’s
“overall value”; the overall value can be “top tier” (approximately the top 20% at Amazon),
“highly valued” (approximately the next 70%), or “least effective” (approximately the bottom
10%) (See R. 68-79, PageID.8227.) The following three examples from the performance matrix
are relevant to the ratings Haydar received:
Performance
Exceeds (second
highest rating)
Achieves (middle
rating)
Achieves (middle
rating)
The
performance
Leadership
Growth
Overall Value
Development Needed Limited (lowest rating) Highly Valued (middle
(lowest rating)
70%)
Solid Strength (middle Limited (lowest rating) Highly Valued (middle
rating)
70%)
Development Needed Limited (lowest rating) Least Effective (bottom
(lowest rating)
10%)
matrix
indicates
that
the
last
example
(Achieves/Development
Needed/Limited) is expected to be an uncommon combination. (See R. 68-79, PageID.8227.) As
will be explained in detail, this is the rating Haydar twice received.
The leadership third is itself based on several inputs: feedback from the Amazonian’s
managers, his peers, and those he supervises, i.e., “360 Feedback.” Those providing feedback
comment on up to 14 leadership principles—and, while perhaps not required by Amazon,
positive and negative remarks are the norm. (See e.g., R. 80-3; R. 80-4; R. 80-5.) Two of the 14
3
leadership principles are “earn trust of others” and “have backbone; disagree and commit.” (R.
68-79, PageID.8226.)
For Haydar’s May 2013 review, Kandasamy rated his performance as Improvement
Needed. (R. 68-4, PageID.6488.) Referring to the leadership principles, Kandasamy wrote, “I do
have serious concerns about your ability to succeed in this role and at Amazon in general, if you
do not take immediate action to improve on your effective communication, active listening,
diving deep, earning the trust of others, insisting on the highest standards[,] and being right a
lot.” (R. 68-4, PageID.6488.) Still, Kandasamy rated Haydar’s leadership at Solid Strength (the
middle rating). (R. 68-4, PageID.6497.)
In June 2013, Amazon had a 4-day “offsite” event. Faricy (the head of Marketplace) had
initially planned for Haydar to attend the event but, upon “g[etting] lots of feedback about
[Haydar’s] behavior,” Faricy suggested that Kandasamy not include him. (R. 68-75,
PageID.8129.) To Faricy’s displeasure, Kandasamy missed the memo. (R. 68-75, PageID.8129.)
At the offsite, Faricy personally “uninvited” Haydar from multiple meetings. (R. 68-75,
PageID.8129.) Moreover, Haydar recalled Faricy making inappropriate remarks: “Mr. David
Anderson had recently gone through a very rough divorce and . . . he was about to get
[re]married. Mr. Faricy suddenly . . . called Mr. Anderson to stand up and said we want to give
you some marriage advice, and then he said, and Abdullah, you get up there too, your people
need this advice as well, which was totally unsolicited and inappropriate and offensive.” (R. 681, PageID.6266 (emphasis added).) The incident was done “jokingly” (R. 68-1, PageID.6585; see
also R. 68-1, PageID.6396), and was possibly related to Haydar working in Seattle while his
wife resided in Detroit (R. 68-128, PageID.8758). Haydar recalled that, during the four days, he
4
and Anderson were “repeatedly” forced to practice lines that they would recite to their wives.
(R. 68-1, PageID.6585.) Haydar felt humiliated. (R. 68-1, PageID.6267, 6585.)
Following the offsite event, Haydar emailed Faricy to follow-up on a conversation they
had during the event. (R. 68-75, PageID.8130–8131.) Haydar told Faricy that advice “ha[d] been
lacking during my ramp-up at Amazon” and that he had not received coaching on how to
effectively converse during meetings. (R. 68-75, PageID.8130.) Faricy responded to Haydar, “[In
my opinion], any focus you have on advising others across the team is premature at this point. To
quote a famous football coach ‘keep your head down, your mouth shut and do your job’. That
advice may be a bit direct and harsh in tone, but it covers the overarching idea.” (R. 68-75,
PageID.8130.)
C.
In October 2013, Haydar’s direct supervisor changed from Kandasamy to Jim Joudrey.
(R. 68-48, PageID.6954; R. 68-78, PageID.8200.)
In December 2013, Haydar met with Joudrey about Amazon’s upcoming annual
evaluations (at Amazon, evaluations cover April 1 of one year through March 31 of the
following year). Afterwards, Haydar sent Joudrey an email summarizing their meeting: “my
current rating would be [Achieves]/[Solid Strength] and my leadership principle areas of
improvement would be ‘Earn Trust of Others’ . . . ‘Dive Deep’ and ‘Are Right, A Lot.’” (R. 687, PageID.6516.)
D.
At the beginning of 2014, Defendant Joel Mosby became Haydar’s direct supervisor. And
Haydar’s employment changed two other significant ways.
5
First, Haydar moved from Amazon’s Seattle, Washington office to its (relatively new)
Detroit, Michigan office. (R. 68-48, PageID.6954.) The “Site Leader” for the Detroit office
(since 2013) was Defendant Garret Gaw. (See R. 68-54, PageID.7406; R. 68-154, PageID.9215,
9218, 9220). Gaw started at Amazon at about the same time as Haydar, and, like Haydar, was
then an L7 Senior Manager. (R. 68-154, PageID.9211; R. 68-54, PageID.7407.)
The second significant change was that Haydar became the manager of the Seller Success
technology teams. (R. 68-1, PageID.6190–91.) As manager for the technology side of Seller
Success, Haydar had to work with Stefan Haney. (See R. 68-1, PageID.6189; R. 68-70,
PageID.8015–8016, 8022–8023, 8026.) This was because Haney, an L8 manager, headed the
business side of Seller Success. (R. 68-1, PageID.6187–6188; R. 68-70, PageID.7993–94.) A
review of some of Haney’s 360 Feedback reveals that Haney, not entirely unlike Haydar, had
issues with communication. For example, one Amazonian remarked, “[Haney] can aim to listen
better to his peers and then present his [point of view]. At times, his directive style does not lend
itself well to earning trust of others.” (R. 68-152, PageID.9072.)
The third change in early 2014 was that Mosby became Haydar’s direct supervisor. (R.
68-48, PageID.6954.) Mosby had started at Amazon about 17 months before Haydar but, at the
time he became Haydar’s direct supervisor, he was still (like Haydar) an L7 Senior Manager.
(See R. 68-88, PageID.8370.)
What occurred during Mosby’s tenure as Haydar’s direct supervisor is best described in
two parts.
1.
Sometime around February 2014, Amazon Marketplace was conducting its twice-a-year
Organizational Level Review (OLR) meetings. During an OLR meeting, Amazonians are able to
6
express agreement or disagreement with other Amazonian’s ratings. (See R. 68-78,
PageID.8202.) There is more than one level of OLR and they “roll up”; so, for example, an L8
(e.g., Joudrey) would host an OLR to go over (and possibly alter) the ratings of his L7s and then
an L10 (e.g., Faricy) would host an OLR to go over (and possibly alter) the ratings of not only
his L8s but also the L7s. (R. 68-78, PageID.8201; R. 68-81, PageID.8281.) The point of Faricy’s
OLR is to “cross-calibrat[e] across all of [Faricy’s] directs” and to assign Amazonians to a curve
(e.g., “Least Effective” for the bottom 10%). (R. 68-78, PageID.8201; R. 68-81, PageID.8262–
8265.)
It appears that coming out of Joudrey’s OLR, Haydar’s rating was Achieves (for
performance), Development Needed (for leadership), and Medium (for growth). (See R. 68-78,
PageID.8202–8203; R. 68-48, PageID.6989; R. 68-81, PageID.8284–8285.) According to the
performance matrix, this would barely equate to an overall value of “Highly Valued” (the middle
70%). (R. 68-79, PageID.8227.)
But things changed during Faricy’s OLR. (See R. 68-81, PageID.8292–8293.) It appears
that the debate was over whether Haydar’s growth potential should be rated Medium or
Limited—if lowered to Limited, Haydar’s overall value would drop to Least Effective (the
bottom 10%). (R. 68-81, PageID.8284–8286.) Although Joudrey and another L8 in Marketplace,
Avi Saxena, were defending Haydar’s rating, one of the counterpoints was Haydar’s conduct
during Joudrey’s OLR the week before: “kind of a passionate pleading for his people, not being
as data driven as he could be.” (R. 68-78, PageID.8203.) This counterpoint “unwound a lot of”
Joudrey and Saxena’s advocacy. (R. 68-78, PageID.8203.) Joudrey further recalled: “[Faricy]
pushed on, you know, basically if there’s this much conversation maybe [Haydar’s rating] wasn’t
7
calibrated right. . . . And so I think [Faricy] made the final call to move [Haydar’s] ratings.” (R.
68-78, PageID.8203–8204.)
On April 30, 2014, Mosby provided Haydar with his annual evaluation (covering April 1,
2013 to March 31, 2014). In addition to considerable negative 360 Feedback, Mosby himself
remarked as follows: “[two weeks ago] I had to ask you to not interrupt me and let me finish my
statement at least 5 times during [a] meeting.” (R. 68-8, PageID.6524–26.) Ultimately, Haydar’s
rating was Achieves/Solid Strength/Limited for an overall value of Least Effective. (See R. 68-8,
PageID.6527; R. 68-48; PageID.6983.)
A few weeks before providing Haydar with his annual evaluation, Mosby had entered a
performance improvement plan for Haydar into Amazon’s system. (R. 68-1, PageID.6388.) But
the plan was never administered because Haydar improved his deficiencies. Indeed, in April
2014, Gaw (the Detroit site leader) sent Haydar an email stating, “Since you’ve moved to the
[Detroit] office, you’ve been a phenomenal Detroit team member and Marketplace leader.”
(R. 68-85, PageID.8363.) By the end of May 2014, Mosby updated his notes in the system: “I
have received feedback from multiple partners and peers that [Haydar’s] interaction with them
has greatly improved.” (R. 68-87, PageID.8367.)
Meanwhile, Haydar expressed concerns about his 2013–14 review to human resources. In
an email to HR, Haydar wrote, “[W]hile I am accepting of the Achieves Performance Rating
accurately reflecting my wins and misses, I am not happy about the Leadership rating being
changed from [S]olid [S]trength to Development Needed, especially if you focus on the feedback
I received at the beginning of [2014] from people I actually worked with versus some absurd
feedback Kanda[samy] had sought.” (R. 68-9, PageID.6529.) After considerable delay on the
part of HR (R. 68-83, PageID.8356), HR finally contacted Mosby. Mosby was convinced that
8
Haydar’s rating was correct. (R. 68-83, PageID.8356.) So, in July 2014, HR emailed Haydar:
“Hello sir! I have reviewed things, talked to Joel [Mosby], and after review we aren’t going to
change your rating. I am sure this is a disappointment, so I also want to reinforce that Joel sees
that you are doing very well and believes you will continue to and is invested in you.” (R. 68-10,
PageID.6532.)
Sometime around August 2014, Faricy again made remarks about Haydar’s marriage.
Haydar recalled, “we had a—maybe ten people in the room . . . and [Faricy] at the end of reading
my document said, wow, this is great work, you’re delivering so much, you’re doing such great
work, you know, if only, you know, you were a little bit better on the home front with your wife
and treated her better, etc., something along those lines.” (R. 68-1, PageID.6269.) And while
Haydar was unsure of the date (he thought maybe in October 2014), he recalled an all-hands
meeting where Faricy “made a comment along the lines of we’re so lucky to have talented
people like Garret [Gaw] and Abdullah [Haydar] leading the [Detroit] office, they’re doing a
great job and growing the office and building projects. If only Abdullah would treat his poor
wife better when he’s traveling so much and neglecting her.” (R. 68-1, PageID.6269.) Although
he did not know the timing, Haydar also testified, “Peter made references to you people, to you
people need to learn how to treat your wives better, those types of comments repeatedly, you
know, in reference to me uniquely differently than other people. He did not [even] make such
comments about [non-Syrian, non-Muslim] people . . . who went through ugly divorces and
whatnot.” (R. 68-1, PageID.6327.)
In September 2014, Mosby again updated the unissued performance improvement plan in
Amazon’s system. Mosby wrote, “Based on feedback from [Haydar’s] team (downward), other
L7’s in the org (lateral) and from L8’s ([Stefan Haney], [Avi Saxena]) the issues of Earn Trust
9
that limited Abdullah’s individual productivity have been addressed over a sustained period.
Recommend ending performance plan successfully.” (R. 68-87, PageID.8367.) Indeed, notes
from an October 2014 OLR meeting state, “Abdullah: Turing the corner, doing very well.
Responded well to feedback.” (R. 68-46, PageID.6944.)
2.
But things then went downhill for Haydar. From Amazon’s perspective, one significant
reason for the change in direction was the manner in which Haydar handled an employee
transfer. From Haydar’s perspective, Faricy had promoted Mosby and Gaw, and, after that, there
were “falsifications of performance record to justify . . . Peter Faricy’s discriminatory view” and
“double standard.” (R. 68-1, PageID.6296.) The Court recounts the transfer first, then the
promotions.
In late October 2014, a dispute over when a software developer would transfer from
Seattle to Detroit began to escalate. The Seattle manager wanted the developer to transfer in
January 2015 so there would be enough time to fill the spot “and ramp-up [the] new developer”
but Haydar, citing Amazon’s 45-day transfer policy, maintained that the developer should
transfer to his team in December 2015. (R. 68-12, PageID.6541–42.) On October 31, Mosby told
Haydar: “We need to support [the Seattle manager] and team on this, they need it! This is a
golden opportunity to earn trust with your peers. Don’t win every battle and lose the war.”
(R. 68-13, PageID.6548.) Haydar responded that the Seattle manager had made transfers difficult
in the past and that earning trust did not equate to giving the developer a bad transfer experience.
(Id.) Mosby replied, “I really don’t think this needs to be difficult. A better response would be
we’ll figure out the right plan to provide [the developer] a good experience [and] help [the
Seattle manger]. There’s a win-win here, and don’t want to hear pushback.” (Id.)
10
Haydar then attempted to resolve the issue with the Seattle manager. After considerable
back and forth with no resolution, the Seattle manager suggested raising the issue with Mosby
and Saxena (Mosby’s manager). (R. 68-11, PageID.6535.) Haydar then forwarded that email to
Mosby: “You already told me you were going to take it to Avi [Saxena] 2 weeks ago. Did you
not do so?” (R. 68-11, PageID.6535.) Mosby replied, “This is a ridiculous issue to escalate to
Avi.” (R. 68-11, PageID.6535.)
A few days later, on November 11, 2014, Mosby emailed Haydar about the transfer: “I
did speak [with the Seattle manager] today and to be quite frank disappointed that I need to get
involved [with] this. . . . As I mentioned a couple weeks ago, I would have liked to see you build
a bridge with [the Seattle manager] and find a compromise, not everything needs to be a battle.”
(R. 68-13, PageID.6545.) Haydar responded that while he had proposed many options, the
Seattle manager had been “absolutely unwilling to compromise.” (R. 68-13, PageID.6545.)
Mosby replied, “I’ve heard a lot of what [the Seattle manager] needs to do, more concerned with
self reflection. So far you’ve argued with both him and me through this issue. My feedback . . .
still holds. You need to find ways to work with people not argue every point until people stop
responding. I’m not going to argue with you about this any more than I already have” (R. 68-13,
PageID.6545.) Haydar responded to Mosby: “Normally, I would agree with you that both sides
need to compromise. And I will commit once you decide on a date. That said, I’m very
disappointed in your attitude here as I’ve successfully transitioned most of the [Seller Success]
team this year . . . . If you think this is about my inability to work with people, I’ve failed in
more ways than one.” (R. 68-13, PageID.6545.)
Two days later, Mosby and Haydar spoke on the phone. The call ended with Mosby
hanging up on Haydar. Haydar then sent Mosby an email that read, “Thanks for hanging up on
11
me.” (R. 68-13, PageID.6544.) Mosby responded: “I can only say the same thing so many times.
I was clear that I was done arguing with you 15 minutes before the call ended and clear that I had
another meeting that I was late for yet you wanted to continue to argue in circles.” (R. 68-13,
PageID.6544.) Haydar replied, “You are definitely correct that I frequently misinterpret pauses
in people speaking as the completion of their thoughts, so I then step in with my own thoughts,
which is correctly understood as an interruption even though I didn’t mean to do it, so I need to
continue to work on that. . . . When I laid out the scenario and asked you for advice on a different
approach, you kept insisting that I need to be more self-critical without actually dealing with the
merits of this issue.” (R. 68-13, PageID.6544.)
When later asked about the transfer incident, Mosby conceded that the issue was with
both the Seattle manager and Haydar. (R. 68-69, PageID.7948.) Even so, Mosby maintained that
the incident “was kind of the, you know, proverbial straw.” (R. 68-69, PageID.7927.) He
continued, “[The transfer incident] was the big one that made it very clear that he had not
materially changed his behavior based on the feedback and coaching he was getting over the
course of the year.” (R. 68-69, PageID.7939.)
Following the transfer incident, at an offsite event in November 2014, Mosby told a
human-resources employee, Kaitlin McVey, that he wanted to place Haydar on a performance
improvement plan.1 On November 15, 2014, Mosby emailed McVey: “Kaitlin, as discussed
1
Haydar claims that Mosby spoke with McVey at the offsite before the transfer incident
and thus Mosby’s handling of the transfer incident was part of Mosby’s plan to place Haydar on
a PIP. In support, Haydar relies on notes indicating that the offsite was around October 2014
(“~ October 2014”) and on McVey’s testimony (more than three years later) that she
“believe[d]” it was in October 2014. (R. 68-92, PageID.8398; R. 68-93, PageID.8421.) But
McVey clearly recalled discussing the transfer incident. (R. 68-93, PageID.8420.) And the
emails show that as of October 31, 2014, Mosby was not yet upset about the incident—at that
point he was encouraging Haydar to seize the opportunity to earn trust with a peer. (R. 68-13,
PageID.6548.) Thus, Mosby would have had no reason to approach McVey until after that.
12
yesterday, here is the current draft of the performance plan for Abdullah. . . . I’d like to move on
this fast. . . . Based on interactions with him over the last couple weeks on an internal transfer
situation, I do not believe he will be successful addressing [his leadership deficiencies].” (R. 6894, PageID.8456.) McVey recalled advising against a PIP: “From the feedback that I had
heard[,] . . . it sounded like [Haydar] was making a lot of progress and there hadn’t been a lot of
incidents related to the gaps, and from what was told to me from [Mosby], this was an isolated
incident.” (R. 68-93, PageID.8419–8420.)
The Court now turns to the promotions which underlie Haydar’s theory for why things
changed for the worse at Amazon. In October 2014, Mosby was promoted to an L8 Director.
(See R. 68-88, PageID.8370.) And in December 2014, Mosby transferred a two-person machinelearning team from Haydar’s management to Haney’s. (See R. 68-15, PageID.6555; R. 68-1,
PageID.6234–6236; R. 68-70, PageID.8028.) Third, and also in December 2014, Gaw (who was
still an L7) took over the technology side of Seller Success and became Haydar’s direct
supervisor. (R. 68-48, PageID.6955; R. 68-99, PageID.8469; R. 68-154, PageID.9212, 9214.)
Haydar also testified that in this timeframe Faricy made Gaw the “manager of the entire Detroit
office” but, as noted above, Gaw was already the site leader for Detroit.
In Haydar’s view, it was not the transfer incident that was, as Mosby described it, the
“proverbial straw”; rather, it was the reorganization. In particular, Haydar believes that Faricy
had promoted Gaw and Mosby and, after that, Gaw, Mosby, and Haney began pushing Haydar
out of Amazon. (R. 68-1, PageID.6231.) Haydar testified that Faricy had a race-based “double
standard” (see R. 68-1, PageID.6303, 6306) and that Mosby and Gaw were not deserving of their
Mosby’s November 15 email strengthens the inference that he and McVey spoke on November
14. (R. 68-94, PageID.8456.) Thus, no reasonable jury could find that Mosby spoke with McVey
at the offsite before the transfer incident.
13
promotions (R. 68-1, PageID.6231, 6237, 6246). Haydar further explained, “[Mosby, Gaw, and
Haney] took Faricy’s direction . . . and made it their . . . task to . . . find excuses for judging me
negatively and pushing me out of the company despite my successes being far superior to their
performances.” (R. 68-1, PageID.6309.) As for the transfer incident, Haydar testified, “the
ultimate cause of the disagreement [between Mosby and I] was [his] ulterior motive.” (R. 68-1,
PageID.6231, 6233.) Haydar also testified, “in November 201[4] onwards, my relationship with
Stefan Haney changed dramatically 180.” (R. 68-1, PageID.6299.) Thus, in Haydar’s view,
“anything post October 2014 were . . . falsifications of [his] performance record to justify . . .
Faricy’s discriminatory view” and “double standard.” (R. 68-1, PageID.6296.)
E.
The following is what occurred during Gaw’s tenure as Haydar’s manager.
1.
In January 2015, Haney, one of Haney’s subordinates, Mat Philipsen, and McVey (HR)
provided feedback about or otherwise became involved in the issues related to Haydar’s
leadership.
That month, Haney learned of an issue about a project roadmap. (See R. 68-17,
PageID.6571–6575.) Haney emailed Mosby: “Joel—can’t begin to tell you how pissed I am at
Abdullah right now. . . . the below thread is a major trust breaker.” (R. 68-17, PageID.6570.)
When Haydar was later asked about the roadmap issue, he explained, “Ultimately the issue was
finally identified much later as being a mistake by Mr. Josh Koppelman; however, at the time it
was just another example of where Stefan Haney and Joel Mosby were, you know, taking
relatively small issues and blowing them out of proportion and claiming lost trust and whatnot all
14
starting from the October [2014] time frame when [Mosby and Haney] identified that regardless
of my performance, Mr. Faricy had a double standard for me.” (R. 68-1, PageID.6246.)
On January 9, 2015, Haydar met with Gaw and, later in the day, Gaw and Mosby. After
the meetings, Mosby and Gaw exchanged emails, including McVey (from HR) in their exchange.
Mosby noted, “I’m not sure he got the immediacy of the need to end this behavior and/or if I was
clear enough that his behavior cannot regress.” (R. 68-102, PageID.8477.) Gaw responded, “I
agree, Joel, I don’t think he got the immediacy after the meeting with the 3 of us. After the
conversation [between just him and I], he was looking for a silver lining in the conversation with
the 3 of us. I think he left that meeting feeling that he dodged a bullet.” (R. 68-102,
PageID.8477.) Haydar made personal notes of the meeting that he would end up forwarding to
HR a few months later: “Garret, Joel, and I sync and confirm [Exceeds]/[Solid Strength] for the
year, but Joel says he’s now not sure on [Solid Strength] due to [the transfer] event. He tells me I
didn’t disagree and commit and then tries to re-discuss the merits of the topic, showing that he is
stuck on this issue.” (R. 68-21, PageID.6589.)
Much transpired a few days later, on January 13, 2015.
That day, Haney, acting on “Mosby’s suggestion,” wrote a lengthy email to McVey to
“shar[e] [his] concern and feedback” regarding Haydar. (R. 68-104, PageID.8483.) Although
raising many issues, Haney noted that he had told Haydar that he “felt deceived by the variance
between our perceived commitment to process and the actions seen [with the roadmap].” (R. 68104, PageID.8484.)
Later that same day, McVey and Gaw exchanged emails regarding Haydar’s issues. (R.
68-103, PageID.8480.) Derek Oehler, a human resources lead who had recently moved to
Marketplace, was included on the email exchange. (R. 68-58, PageID.7501–7502.) McVey asked
15
Gaw to enter a coaching action into Amazon’s system. (R. 68-103, PageID.8479.) She also asked
Gaw what his “strategy” would be for “moving forward with Abdullah[.]” (R. 68-103,
PageID.8479.) Before Gaw could respond, Oehler did: “Let’s take the conversation off line.”
(R. 68-103, PageID.8479.)
Gaw complied with McVey’s request and entered a coaching action in the system. The
entry read in part, “Abdullah has begun struggling again in the same areas he’s had problems
with in the past, specifically breaking trust with others. Joel Mosby, Stefan Haney, and Mat
Philipsen have had specific issues with Abdullah not following through on commitments.”
(R.68-87, PageID.8367.)
That evening (January 13), Haydar inquired about his rating. He emailed the following to
Mosby and Gaw: “As we discussed last Friday, we are all in agreement on my performance
rating as Exceeds Expectations for all of 2014, so I wanted to follow up on the Leadership
Principles since my midyear rating was clearly Solid Strength and you expressed some doubt as
to whether it should be Solid Strength or Development Needed for the entire year. I understand
your concerns regarding the November transfer incident even though I have a different
perspective on that event.” (R. 68-14, PageID.6552.)
Haydar’s rating inquiry prompted a discussion between Gaw and Mosby (recall, Mosby
had been Haydar’s supervisor for most of 2014). Gaw asked Mosby, “are you in agreement with
Exceeds[?]”
(R.
68-30,
PageID.6622.)
Mosby
responded
to
Gaw,
“I’m
leaning
[Exceeds]/[Development Needed]/[Limited]—which is an abnormal rating.” (R. 68-30,
PageID.6622.)
Two days later (January 15, 2015), Haney forwarded to McVey several old emails that
Haney believed reflected Haydar’s leadership difficulties. (See R. 68-107, PageID.8490; R. 68-
16
108, PageID.8492; R. 68-109, PageID.8495.) Haney might have also asked his subordinate,
Philipsen, to look for similar emails. In one email that Philipsen forwarded to Haney, Philipsen
admitted to making a mistake. (R. 68-110, PageID.8497.) But when Haney forwarded the email
to McVey, he seemed to blame Haydar: “One more example of disruption along with commit
and disagree.” (R. 68-110, PageID.8497.)
The next day, Philipsen entered 360 Feedback for Haydar. Although the timing suggests
that this may have been at Haney’s urging, Philipsen testified that “Abdullah had asked for
feedback” “[a]s part of the performance review cycle.” (R. 68-153, PageID.9173.) (The 2014–15
OLRs were right around the corner.) Philipsen wrote in part, “I do not trust Abdullah, when he
makes a promise I find that he rarely keeps to the commitment. . . . I believe Abdullah can’t
remember details and rather than say ‘I don’t know’, or creating a note keeping system for
himself he will make up information on the fly or tell a self serving lie.” (R. 68-111,
PageID.8503.)
Meanwhile, Gaw had not yet answered Haydar’s inquiry about his rating. On January 20,
2015, McVey wrote to Gaw: “We wouldn’t PIP an Exceeds employee. That would send mixed
messages. If you believe that we will need to move to [PIP], then the correct [performance]
rating for him is most likely Achieves.” (R. 68-113, PageID.8542.) Gaw, using language
suggested by McVey, responded to Haydar’s rating inquiry as follows: “While some of your
work in 2014 would be considered ‘Exceeds’, the overall rating will need to take into account
peer feedback and leadership feedback. To be clear, leadership principles heavily drive the
effectiveness of an L7+ leader.” (R. 68-101, PageID.8474.)
17
2.
In late January and early February 2015, Marketplace was holding OLR meetings for the
April 2014 through March 2015 period.
In terms of performance, it appears that Haydar’s biggest success was transitioning the
Sellers Success work from Seattle to Detroit and building the development team in Detroit to
handle that work. By most, if not all accounts, Haydar had navigated the transition extremely
well and assembled a strong team of developers (many exhibited traits above their level). (See
e.g., R. 68-112, PageID.8511.) This was acknowledged by Gaw. (See R. 68-16, PageID.6562–
6563.)
As far as leadership, Haydar’s 360 Feedback was a mix of positive and negative remarks.
For example, a developer stated that Haydar had “very strong leadership skills and [stood] as a
role model.” (R. 68-112, PageID.8508.) But others commented on the issues that Gaw and
Mosby had been concerned about. One Amazonian stated, “He can easily plow through
conversations where nobody else can get a word in. This is a miss on earn trust because (myself
included) don’t always want to engage with him because they know what the communication
will be like.” (R. 68-112, PageID.8504.)
Gaw’s “initial OLR ratings” for Haydar (presumably before the Faricy OLR) were
Achieves (performance), Development Needed (leadership), and Limited (growth). (R. 68-113,
PageID.8541.) If this rating stuck, it would mean a “Least Effective” overall value rating for
Haydar—for the second year in a row.
This time, it appears Haydar’s rating did not change in Faricy’s OLR. Notes from
Faricy’s OLR state: “Was doing better, but recently ‘fell off the bus’ and is being managed out.
How do we ensure his team/leadership structure is prepared for him leaving the company? Issues
18
in Earns Trust, Disagree and Commit and Vocally Self-critical. We are working on coaching him
out.” (R. 68-116, PageID.8556.) At his deposition, Faricy explained: “I did not have sufficient
experience to judge Abdullah Haydar’s performance . . . . [Haydar] was always two to three
layers down in the organization, so those people who he worked for evaluated his performance.”
(R. 68-45, PageID.6752.)
Following Faricy’s OLR, McVey and Gaw discussed how to address Haydar’s situation.
McVey wanted to propose terminating Haydar but needed to check with “legal” first. (R. 68-117,
PageID.8563.) As it turned out, because Haydar had not been placed on a formal PIP after his
first Least Effective rating, a PIP rather than termination was the proper course. (R. 68-121,
PageID.8597.)
On February 24, 2015, Haydar was informed of his rating. (R. 68-20, PageID.6584.) At
his deposition, Haydar was asked about the negative 360 Feedback. Haydar explained, “Even on
the ones of concern such as earns trust of others, there was significantly more positive than
negative, it was maybe 150 percent to 100 percent roughly, you know, numerically, and so those
were disregarded and Mr. Gaw and others cherry-picked, you know, data to support a narrative
which supported Mr. Faricy’s biased and discriminatory view against me.” (R. 68-1,
PageID.6442.) Gaw had a different take: “there was 148 total data points, a 2.36 to 1 ratio of
strengths to areas to improve, which relative to Abdullah’s peers and others that I’ve worked
with and given reviews for, that number is low as a ratio.” (R. 68-54, PageID.7343.)
Gaw’s plan was to place Haydar on a performance improvement plan. (R. 68-16,
PageID.6568.)
19
3.
The very day Haydar learned of his rating, he sent a long email to Shelly Cerio, an HR
vice president. (See R. 68-60, PageID.7736; R. 68-125, PageID.8668.) The subject of the email
was “CONFIDENTIAL: bias issue.” Haydar wrote in part, “It has become very clear to me that I
have faced a pattern of biased treatment for the past 18+ months and I am no longer willing to
handle this within my HR[] team because I believe that this bias exists with Peter Faricy himself
as well as his directs and that our HR team are deferring to his lead rather than challenging this
bias.” (R. 68-20, PageID.6584.)
Less than two weeks later, on March 5, 2015, Haydar had a lengthy call with Cerio and
both took detailed notes of their conversation. (See R. 68-61, 68-124.) According to Haydar’s
notes, he told Cerio about ethnic and religious bias: “[W]e woke up and came in to the breakfast
at the offsite and then [Faricy] had us stand up and asked us what were the lines we learned
yesterday. And it was all silliness, but also offensive as to why Peter was picking on me as if I
have an inherently bad relationship with my wife due to my ethnicity or religion.” (R. 68-124,
PageID.8640 (emphasis added).) But Haydar’s notes also reflect that he “ha[d] no idea who has
what religious or ethnic biases” and that he was not certain if he was “using the word bias
correctly.” (R. 68-124, PageID.8641.) On this point, Cerio’s notes agree: “Bias—it is not racial
or anything like that; Bias that nobody standing up for [Haydar]; I am the fall guy because I do
not have anyone standing up for me in the OLR.” (R. 68-61, PageID.7880.)
Following their call, Haydar sent Cerio a summary of his requested relief: (1) “[a] fairly
determined rating for my 2014–15 work, which has continuously and consistently confirmed to
be at least [Exceeds]/[Solid Strength] all year,” (2) “Mosby being severely reprimanded for his
dishonest, vindictive, and unethical actions against me in the past 3 months,” (3) “Haney being
20
reprimanded for collaborating with Joel [Mosby],” and (4) “Faricy, the rest of his directs and
their HR partners being coached on accepting severely flawed and biased performance data in an
OLR without challenging it.” (R. 68-21, PageID.6588.)
In response, Cerio asked Anne DeCleene, an attorney (R. 68-125, PageID.8660–8661), to
conduct an investigation into Haydar’s allegations. While DeCleene was human resources
personnel, she was from outside Marketplace (R. 68-60, PageID.7741–7742) and, in Cerio’s
opinion at least, “neutral” (R. 68-22, PageID.6591).
DeCleene’s investigation began toward the end of March 2015. (See R. 68-48,
PageID.6953.) She interviewed 11 people, including Haydar, Faricy, Mosby, Gaw, Haney, and
McVey. (See R. 68-48, PageID.6962.) She reviewed emails. (See R. 68-48, PageID.6963–6970.)
And 360 Feedback. (R. 68-125, PageID.8691, 8692, 8694, 8725.) Haydar’s performance
reviews, too. (R. 68-125, PageID.8718–19.) She also looked into Mosby and Haney’s
promotions. (R. 68-48, PageID.6959.) And while DeCleene was not focused on national-origin
or religious discrimination, her investigation did look into Haydar’s claims that Faricy promoted
individuals that did not deserve promotions and whether Faricy had made comments that
“illustrate ‘bias’ against Abdullah.” (R. 68-48, PageID.6959–6960.)
On April 15, 2015, DeCleene issued a report. (See R. 68-48 (report); see also R. 68-49
(supplemental report).) Regarding Haydar’s ratings, she “was unable to corroborate that there
was any bias and/or unfairness in the ratings Abdullah received in 2014 and/or 2015.” (R. 68-48,
PageID.6955.) She also could not corroborate Haydar’s claim that Mosby’s and Haney’s
promotions were due to their relationship with Faricy; she noted that both Mosby and Haney had
been twice denied promotions before getting them. (R. 68-48, PageID.6959.) And while
DeCleene found that Faricy’s comments were “not tasteful,” she found “no evidence of Abdullah
21
being subjected to any comments, decisions, treatment and/or actions that would be in violation
of any Amazon policy.” (R. 68-48, PageID.6960.)
The day after DeCleene issued her report, Cerio called Haydar to inform him of the
results of the investigation. After their phone call, Haydar sent Cerio an email. He stated that he
could not accept that his rating was accurately determined for several reasons, including that
Mosby’s behavior was “retribution” for “calling out [Mosby’s] unethical and unprofessional
behavior during the November employee transfer,” that Haney had excluded Haydar from
“critical communications regarding [Haydar’s] teams and projects,” and that Gaw had “been
overly deferential” to Mosby and Haney in assessing Haydar’s performance. (R. 68-22,
PageID.6593.) The next day, April 17, Haydar sent a similar email to McVey and Oehler in
human resources but additionally accused them of “enabl[ing]” Mosby’s behavior. (R. 68-22,
PageID.6595.)
A little over a week later (late April 2015) Haydar emailed both Gaw and Saxena (an L8
in Marketplace) about his rating. Haydar told Gaw: “If you are not vocally self-critical with HR
about this very serious mistake in determining my rating then I will be forced to call you out with
HR[.]” (R. 68-28, PageID.6617.) As for Saxena, Haydar stated: “If there is a good reason why
you allow [Mosby] and [Gaw] to drive this incorrect rating in the OLR, I would like to hear this
from you directly. Otherwise, I am forced to question your motives and will need to address this
situation through the right process.” (R. 68-29, PageID.6619.)
On April 29, 2015, Faricy emailed Michael Beary, Oehler’s supervisor in human
resources. (R. 68-132, PageID.8788.) Apparently, Haydar had heard of a tax issue affecting the
Detroit office from another Amazonian and passed that information along. (R. 68-131,
PageID.8771.) Faricy wrote to Beary, “What is the status with Abdullah? He is spinning up the
22
team in Detroit (see below), and it was surprising to see him here last week hanging outside the
leadership meeting we had in Dawson. Are we moving fast enough here?” (R. 68-131,
PageID.8770.) Beary replied in part, “[W]e will want to be buttoned up at every step with this
one. It wouldn’t be quick or easy, but if we manage it effectively—like the EB case—we’ll get to
a good outcome as quickly as possible.” (R. 68-131, PageID.8770.) “EB case” referred to an
EEOC complaint or a lawsuit filed by Elisabeth Bailey alleging that Haney had engaged in
sexual harassment. (See R. 68-70, PageID.8062–8068.)
Faricy had also learned that Haydar was attempting to transfer to another group in
Amazon and thus reached out to Ian Simpson, a Marketplace vice president directly under
Faricy: “We aligned on no interviews for Abdullah? I don’t want to pass this problem on to
another team.” (R. 68-131, PageID.8769.) Simpson replied, “I don’t intend on approving a
transfer out of our team. . . . We can’t pass this off to another team. . . . I think we need to be air
tight and manage Abdullah through the PIP process.” (R. 68-131, PageID.8769.)
4.
While DeCleene’s investigation was taking place, Haydar had a happenstance
conversation with Faricy. In March 2015, the two attended an Amazon outing and Haydar raised
the issue of his rating. Haydar recalled, “Peter made some kind of joke about the [baseball] game
going on. He said, if we all performed as poorly as the Seattle Mariners are performing right
now, . . . we would be lucky if we ever get any . . . more stock. And so I had no idea why Peter
was making that joke seemingly about performance when I was frustrated with the seeming
falsification of my performance review, and so I said to Peter, . . . if you want to discuss my
review, I’m happy to do that with you any time in detail.” (R. 68-1, PageID.6275.) According to
Haydar, “[Faricy’s] response was[,] . . . if you want to talk about your performance, talk to your
23
manager or HR. I said, Peter, for the second year in a row, you’ve reassigned me to work for a
peer L7 and they are representing me in the OLR [meetings] and I don’t think I’m getting a fair
shake and I think that . . . you . . . are the person who are making these changes so I think I’m
entitled to talk to you about it if you’re available[.]” (R. 68-1, PageID.6275.) Haydar recalled
Faricy responding with this remark: “[W]hy don’t you people ever follow the process, why don’t
you, you know, work with the manager and HR, I don’t want to talk to you about this at all.” (R.
68-1, PageID.6275.)
5.
Although it was drafted in April 2015, Haydar did not begin the 60-day performance
improvement plan until late May 2015. (R. 68-141, PageID.8936.) The plan had several
objectives. For one, Haydar was to “[r]each out to each peer/senior leader [he had] lost
significant trust with (Joel Mosby, Stefan Haney, Paul Tunney, Mat Philipsen, Doug Welzel,
Damian Poznanski) request their feedback and concern, and let them know [he was] committed
to repairing the relationship.” (R. 68-141, PageID.8936.) For another, Haydar was to improve his
relationship with his peers and senior leaders by “150%.” (R. 68-141, PageID.8936.) And
another part of the performance improvement plan required Haydar to draft a “development
plan” to rebuild trust with his “peers and senior leaders.” (R. 68-141, PageID.8936.)
As for the first objective, Haydar explained that he did reach out to those whose trust he
had lost. But Mosby and Philipsen cancelled their scheduled meetings or delayed them. (R. 68-1,
PageID.6339, 6341.) As for Haney, Haydar recalled, “we had a nice heart to heart and he said
that he was happy to work with me from that point forward and we could resolve our issues.” (R.
68-1, PageID.6339.) Poznanski said Haydar was a “role model” and Welzel “commended
[Haydar] for how [he] had transitioned the listings team to him.” (R. 68-1, PageID.6340.)
24
Haydar recalled that when he told Gaw about the meetings (or attempts to meet), Gaw
suggested that Haydar meet with still others. (R. 68-1, PageID.6340.) And when Haydar pointed
out that some of those people had no issues with him, Gaw “again changed the goals” and
suggested yet another person for Haydar to meet with. (R. 68-1, PageID.6341.)
On July 7, 2015, Haydar had a 30-day check-in on his performance improvement plan.
Gaw told Haydar that his approach to the one-on-one meetings had been flawed because Haydar
had raised what they said in “their confidential feedback” instead of “seeking guidance from
them on building the relationship with them.” (R. 68-32, PageID.6627.) After the 30-day checkin, Haydar realized that the process was flawed: “I realized that it didn’t matter that the original
goals were falsified, no matter what I did, HR was going to back Mr. Gaw’s supposed goals and
just adding random names of people to supposedly help me communicate[.]” (R. 68-1,
PageID.6365.)
On August 7, 2015, Faricy again sent an email to Beary (Oehler’s supervisor in HR): “Do
we have an update on Abdullah?” (R. 68-142, PageID.8939.) Beary in turn emailed Oehler, who,
apparently, sent Beary a response covered by attorney-client privilege. (See R. 68-142,
PageID.8939.)
After the 60-day performance improvement period (August 31 to be exact), Gaw
prepared a document commenting on the performance-improvement-plan and the developmentplan goals. (R. 68-32.) One of the development-plan goals required Haydar to “openly voice
opinions with peers and leadership, using data to thoughtfully explain counter-arguments, if any,
while actively listening to the opposing view.” (R. 68-32, PageID.6629.) Regarding this goal,
Gaw thought that Haydar had “ke[pt] [his] distance from conversations or professional debates”
and had not adequately engaged during certain meetings. (R. 68-32, PageID.6629–6630.) As for
25
improving trust by “150%,” Gaw similarly noted, “Biggest feedback [from others] was that
you’re not engaging other leaders and not participating in group discussions with other L7/L8
leadership.” (R. 68-32, PageID.6635.) In the end, Gaw gave Haydar an ultimatum: “[W]e have
decided to extend the PIP through September 25, 2015. During that time, you must immediately
conform your conduct to Amazon’s Leadership Principles at the level expected of an L7 leader.
If you fail to immediately improve, your employment with Amazon will be terminated.” (R. 6832, PageID.6628.)
In an email summarizing their meeting, Haydar explained that he had intentionally
engaged less during meetings. (R. 68-143, PageID.8946.) Haydar told Gaw and Oehler that he
“had received positive feedback in allowing others to speak more and dominating conversations
less.” (R. 68-143, PageID.8946.) Haydar continued, “Your interpretation of [that] as a failure to
meet performance expectations and successfully complete the PIP is not backed by any specific
data which you could provide[.]” (R. 68-143, PageID.8946.)
On September 14, 2015, Haydar emailed Gaw and Oehler ahead of a scheduled
performance-improvement-plan meeting. (R. 68-143, PageID.8941.) Haydar wrote in part, “this
topic of engagement in certain meetings was not a substantial topic at the 30 day review, so your
claim in that regard isn’t backed by facts. This seems to be a repeat of your actions on my annual
performance review, where the goals of my PIP are being changed mid-flight as happened at the
beginning of the year.” (R. 68-143, PageID.8942.)
Two days later (September 16), Haydar sent an email to Amazon’s CEO, Jeff Bezos. (R.
68-145, PageID.8969.) A month before, the New York Times had published “Inside Amazon:
Wrestling Big Ideas in a Bruising Workplace.” (R. 68-144.) As the title of the article suggests,
the piece criticized management practices at Amazon. (See id.) In response to the article, Bezos
26
sent a “Dear Amazonians” email asking Amazonians to “escalate to HR” if they knew of “any
stories like those reported.” (R. 68-145, PageID.8970.) Bezos also stated, “You can email me
directly.” (R. 68-145, PageID.8970.) On September 16 (a month after Bezos’ email), Haydar did
just that. Haydar wrote, “I am following up on your [email about the New York Times article] as
well as my previous email to you in May . . . . It is now irrefutably clear that I have been the
subject of illegal management abuse for the past 10 months, including a fraudulent review
process and PIP process. . . . HR is interested in making my situation ‘go away’ and apparently
nothing else, least of all challenging abusive managers who lie and cheat to advance their careers
and harm others.” (R. 68-145, PageID.8969.)
Haydar believes that his email to Bezos contributed to his termination. (See R. 72,
PageID.9640.) The same day it was received (September 16), a copy of the email was forwarded
to Cerio (HR). (R. 68-145, PageID.8969.) And two minutes after that, Oehler created a ticket in
Amazon’s system for Haydar to be terminated on September 23, 2015 (two days before the
planned end of the PIP). (R. 68-146, PageID.8972.) Although initially unsure, Oehler ultimately
testified that he did not learn of Haydar’s email to Bezos until after Haydar was terminated. (R.
68-58, PageID.7558, 7636.) Gaw said the same. (R. 68-54, PageID.7420.)
September 22, 2015 marked the end of Haydar’s almost-three-year tenure at Amazon.
(See R. 68-148, PageID.8983.) Although Oehler and Beary approved the decision, it was Gaw
who terminated Haydar. (R. 68-54, PageID.7424; R. 68-58, PageID.7635.) Gaw explained, “we
had worked with Abdullah on finding—providing him more and more opportunities to show that
he can earn trust with folks and be vocally self-critical, and . . . at that point it had turned into
more sulking and just not participating in meetings, and so at the end of the day, it wasn’t just
being able to stop the erosion of trust, but . . . we were looking for him to start building it back.”
27
(R. 68-54, PageID.7446.) Gaw testified that Faricy had given him no input regarding Haydar’s
termination. (R. 68-54, PageID.7425.)
F.
In November 2015, Anas Fattahi, took over one of the two teams that Haydar had been
managing. (R. 68-54, PageID.7366.) Fattahi took over the other team about four months later (in
the interim Gaw ran the team). (R. 68-54, PageID.7366.) Fattahi, like Haydar, is male, married,
Syrian-American, and Muslim. (R. 68-54, PageID.7366; R. 68-68, PageID.7910.)
G.
In February 2016, Haydar filed a charge with the Equal Opportunity Employment
Commission asserting retaliation and religious and national-origin discrimination. (See R. 68156, PageID.9241.) The EEOC issued Haydar a right-to-sue letter in July 2016. (R. 1, PageID.4.)
Haydar filed this lawsuit in October 2016. His complaint contains eight counts. (R. 1.)
Four of them correspond to Haydar’s assertion that Amazon discriminated on the basis of his
national origin and religion in violation of Title VII of the Civil Rights Act and Michigan’s
analog, the Elliott-Larsen Civil Rights Act. A fifth count asserts that Amazon discriminated
based on marital status in violation of ELCRA. Two more counts assert that Amazon retaliated
against him after he complained of discrimination in violation of Title VII and ELCRA. And
Haydar’s eighth count claims that Amazon violated Michigan public policy.
Following considerable discovery, Amazon moved for summary judgment on all eight
counts. (R. 68.)
28
II.
“The court shall grant summary judgment if the movant shows that there is no genuine
dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R.
Civ. P. 56.
III.
A.
Although most of Haydar’s claims under Michigan law parallel his federal claims, two
lack a federal counterpart: that Amazon discriminated against him because of his marital status
and that Amazon violated Michigan public policy. As Amazon is entitled to summary-judgment
on these two claims, addressing them first permits the remaining six claims to be addressed in
pairs.
Michigan’s Elliott-Larsen Civil Rights Act prohibits an employer from treating an
employee differently based on his marital status. Mich. Comp. Laws § 37.2202(1)(a); Miller v.
C.A. Muer Corp., 362 N.W.2d 650, 653–54 (Mich. 1984). Thus, the marital-status provision of
ELCRA does not concern itself with discrimination based on the characteristics of an employee’s
spouse. See id.; Ortman v. Gordon Food Serv., Inc., 570 N.W.2d 152, 153 (Mich. Ct. App.
1997).
Haydar says that Faricy would not have made remarks about how he was treating his wife
if he “were a single Muslim man.” (R. 72, PageID.9637.) Haydar even goes so far to say that
Faricy’s comments are “direct evidence” of animus toward the status of being married. (R. 72,
PageID.9637.)
The Court disagrees. As an initial matter, Faricy’s reference to “you people” cannot
reasonably be construed to mean “you [married] people.” It appears that even Haydar does not
29
believe that (see R. 68-1, PageID.6327) and the heart of his discrimination theory relies on
stereotypes about married Syrian or Muslim men, not married men generally. So in terms of
marital-status discrimination, what Haydar is left with is that Faricy (1) made Haydar stand up in
front of a group at an offsite event to receive marriage advice and forced Haydar to recite lines
that would make his wife feel loved and (2) made comments during two meetings (and other
times) to the effect of, “If only Abdullah would treat his poor wife better when he’s traveling so
much and neglecting her.” Haydar is correct that had he been single, Faricy would not likely
have made these remarks about Haydar’s non-existent spouse. The problem for Haydar, though,
is that these comments at most reflect a weak marital-status animus, i.e., that Faricy did not like
Haydar because Haydar was married as opposed to single. The comments reflect that Faricy
wanted Haydar to treat his spouse better—not that he didn’t want Haydar to have a spouse.
Indeed, Haydar claims that Faricy viewed Gaw, Mosby, and Haney through rose-colored
glasses—but each were (and, apparently, are) married. (R. 68-54, PageID.7321; R. 68-69,
PageID.7921; R. 68-70, PageID.7977.) The record does not permit a reasonable jury to find that
an animus toward married employees “made a difference” in Haydar’s ratings, performance
improvement plan, or termination. See Hazle v. Ford Motor Co., 628 N.W.2d 515, 522 (Mich.
2001).
As for Haydar’s public-policy claim, Haydar has made no attempt to respond to
Amazon’s arguments. (See generally R. 72; R. 70, PageID.9331 n.15.) So he has abandoned his
public-policy claim. See Brown v. VHS of Michigan, Inc., 545 F. App’x 368, 372 (6th Cir. 2013);
Clark v. City of Dublin, 178 F. App’x 522, 524–25 (6th Cir. 2006).
30
B.
The Court thus turns to Haydar’s claims of national-origin and religious discrimination
under Title VII and ELCRA.
1.
Although Haydar has countered Amazon’s arguments for summary judgment, his
affirmative “theory of the case” is less clear. As this Court understands it, the following is
Haydar’s theory of national-origin and religious discrimination.
Regarding animus, Haydar points to Faricy specifically and Amazon’s culture generally.
He says, “As a Muslim man, Plaintiff has had to endure the stereotype of members of his religion
as backwards, misogynistic, and abusers of women. Faricy’s comments about Plaintiff being a
‘poor husband’ were unprompted, unexplained and never levelled at any similar non-Arab, nonMuslim individual.” (R. 72, PageID.9636.) And Haydar describes Amazon’s corporate culture as
“ruthless, unsympathetic, and a hotbed for discrimination.” (R. 72, PageID.9637 n.9.) In support
of this assertion, Haydar cites the New York Times exposé on Amazon, that Cerio (from HR)
had never substantiated a claim of discrimination during her tenure, and that not one of the
people who responded to Bezos’ email about the NY Times article had their claims
substantiated. (R. 72, PageID.9637 n.9.)
Notably, Haydar does not have evidence that the two people most responsible for his
evaluations and termination—Mosby and Gaw—held animus toward Syrians or Muslims.
Instead, Haydar’s theory seems to be that Faricy held the unlawful animus, Amazon corporate
culture tolerated it, Mosby, Gaw, and Haney understood that Faricy did not like Haydar (even if
they did not know the reasons), and, taking a cue from their leader, Mosby, Gaw, and Haney
overstated Haydar’s leadership deficiencies and unfairly criticized him for those deficiencies.
31
(See R. 68-1, PageID.6231, 6296, 6303, 6306.) Alternatively, or perhaps additionally, Haydar
says that Faricy “plotted behind the scenes to secure [his] termination.” (R. 72, PageID.9605.)
2.
If the Court has accurately portrayed Haydar’s theory of discrimination, it is not a
particularly strong one.
Start at the beginning. As Amazon points out (R. 70, PageID.9325), before Faricy had
made any remarks about Haydar’s relationship with his wife, Kandasamy had given Haydar a
negative six-month review. Indeed, Faricy had invited Haydar to attend the June 2013 offsite
event but then changed his mind after receiving negative feedback about Haydar. The order of
events suggests that Haydar’s performance, rather than his national origin or religion, is what
shaded Faricy’s initial assessment of Haydar.
Moving forward on the timeline to Haydar’s first Least Effective assessment (February
2014), it is true that Joudrey (Haydar’s then-direct supervisor) and Saxena (a Marketplace L8),
thought that Haydar’s rating should be Achieves/Development Needed/Medium; it was only
during Faricy’s OLR that Haydar’s growth potential was lowered from Medium to Limited.
But, as Amazon notes (R. 70, PageID.9326 n.6), there is a possible non-discriminatory
reason for the change. During Joudrey’s OLR, which preceded Faricy’s by a week or so, Haydar
apparently engaged in “passionate pleading” for those he supervised and had not been “as data
driven as he could be.” (R. 68-78, PageID.8203.) Haydar’s conduct was brought up during
Faricy’s OLR as a significant counterpoint to Joudrey’s and Saxena’s defense of Haydar’s rating.
Again moving forward on the timeline, from May 2014 (or so) until October 2014, both
Gaw and Haney provided positive feedback about Haydar’s leadership. (R. 68-85, PageID.8363,
8365; R. 68-87, PageID.8367.) Indeed, it appears that this feedback played a large role in Mosby
32
cancelling a performance plan and concluding that Haydar had corrected his deficiencies. (R. 6887, PageID.8367.) If Gaw, Haney, and Mosby were motivated to push Haydar out of Amazon
(because they got the impression that Faricy did not like Haydar), then why would each of them
have effectively supported Haydar for many months?
Haydar’s answer to this question seems to be that things changed in October or
November 2014 after Faricy reorganized Marketplace, including by promoting Mosby and Gaw.
(R. 68-1, PageID.6231, 6233, 6237, 6299, 6296, 6309.)
But it is not clear why the promotions would have caused Gaw, Mosby, and Haney to
became more beholden to Faricy. Gaw was already site leader of the Detroit office before
October 2014. (R. 68-59, PageID.7668; R. 68-88, PageID.8370.) As for Gaw becoming Haydar’s
supervisor, it is not clear that was Faricy’s decision and, in any event, the move may have been
necessitated by Mosby’s promotion to an L8 Director. (See R. 68-54, PageID.7409.) As for
Mosby’s promotion, Faricy could not unilaterally promote people to L8 under Amazon’s system.
Saxena (Mosby’s direct supervisor) believed Mosby deserved a promotion and, over many
months, prepared a detailed document for that promotion. (R. 68-59, PageID.7678.) That
document included supporting remarks from many L10s and L8s—none of whom were Faricy.
(R. 68-88, PageID.8376–83.) And promotions were also discussed during OLRs (R. 68-59,
PageID.7677) and went through several levels of approval, only one of which was from Faricy
(R. 68-59, PageID.7674, 7676).
And there was another explanation for why things changed for Haydar in October 2014:
the employee transfer incident. Indeed, Mosby testified that the transfer incident was the
“proverbial straw.”
33
As for Haney’s change of heart, in January 2015, he felt that Haydar’s handling of the
roadmap issue was “a major trust buster.” (R. 68-17, PageID.6570.) Further, while Haydar
asserts that Haney was a “protégé of Faricy who had a close personal and professional
relationship with him” (R. 72, PageID.9611), the record does not support a close “personal”
relationship. The two attended work-related social events, including lunches and happy hours
(R. 68-45, PageID.6779–6780), but Faricy testified that he “never had outside-of-work social
interactions” with Haney (R. 68-45, PageID.6779).
So the record provides some non-discriminatory, leadership-related reasons for Haydar’s
second Least Effective rating. And two such ratings all but guaranteed that Haydar would be
either terminated or placed on a performance-improvement plan. Amazon ultimately chose the
latter, but the former might have been completely warranted.
That leaves the end of the timeline. When Haydar did not succeed during the first 60 days
of his performance improvement plan, Amazon extended the PIP, giving Haydar yet another
chance to demonstrate change. But, according to Gaw at least, “it had turned into more sulking
and just not participating in meetings, and so at the end of the day, it wasn’t just being able to
stop the erosion of trust, but . . . we were looking for him to start building it back.” (R. 68-54,
PageID.7446.)
Finally, very-shortly after Haydar was terminated, Anas Fattahi, a male, SyrianAmerican, Muslim, took over one of Haydar’s teams and, four months later, took over the other.
And it appears that feedback for Fattahi’s promotion to L7 was being collected before Haydar
filed his EEOC complaint. (Compare R. 68-155, PageID.9235, with R. 68-156, PageID.9241.)
And if Amazon did not then know that Haydar intended to take legal action, its intent to place
34
Fattahi in Haydar’s position cuts against Haydar’s theory that he was terminated on account of
his national origin or religion.
3.
All of that said, two things cut strongly in Haydar’s favor. For one, this is summary
judgment and so the record, and all reasonable inferences drawn from the record, are viewed in
the light most favorable to Haydar. See Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475
U.S. 574, 587 (1986). For another, Haydar has elected a mixed-motive route to the jury. (R. 72,
PageID.9627–9628 & n.7.) And the law governing mixed-motive claims is more friendly than
single-motive claims to those in Haydar’s position.
A mixed-motive theory is viable when the evidence shows that while legitimate reasons
motivated the adverse action, the employee’s protected characteristic also motivated the action.
Wright v. Murray Guard, Inc., 455 F.3d 702, 711 (6th Cir. 2006); see also Griffin v. Finkbeiner,
689 F.3d 584, 595 & n.7 (6th Cir. 2012) (suggesting that court may infer a mixed-motive theory
based on the evidence). Under this theory, an employee need only show that his protected
characteristic was “a motivating factor” for the action, “even though other factors also motivated
the practice.” 42 U.S.C. § 2000e-2(m).
More important for present purposes is that an employee’s summary-judgment burden is
“not onerous” under a mixed-motive theory. To start, the McDonnell-Douglas/Burdine
framework does not apply. White v. Baxter Healthcare Corp., 533 F.3d 381, 400 (6th Cir. 2008).
Indeed, in some cases an employee might not be able to prove a prima facie case under that
tripartite framework but might still be able to reach a jury on a mixed-motive theory. See Wright,
455 F.3d at 717 (Moore, J., concurring). And in all cases, to survive summary judgment on a
mixed-motive theory, the evidence need only permit a reasonable jury to find “(1) the defendant
35
took an adverse employment action against the plaintiff; and (2) ‘race, color, religion, sex, or
national origin was a motivating factor’ for the defendant’s adverse employment action.” White,
533 F.3d at 400 (quoting 42 U.S.C. § 2000e-2(m)). “This burden of producing some evidence in
support of a mixed-motive claim is not onerous and should preclude sending the case to the jury
only where the record is devoid of evidence that could reasonably be construed to support the
plaintiff’s claim.” Id. Further, “as inquiries regarding what actually motivated an employer’s
decision are very fact intensive, such issues will generally be difficult to determine at the
summary judgment stage and thus will typically require sending the case to the jury.” Id. at 402
(internal quotation marks omitted).
Here, “the record is [not] devoid of evidence that could reasonably be construed to
support,” White, 533 F.3d at 400, Haydar’s claim of national-origin and religious discrimination.
To start, a jury could find that Faricy harbored discriminatory animus toward Syrians or
Muslims (or, at least, Syrian men or Muslim men). Cf. Reed v. Procter & Gamble Mfg. Co., 556
F. App’x 421, 429 (6th Cir. 2014) (finding mixed-motive claim failed where there was “no
evidence” that decision-maker harbored racial animus). Amazon claims that “Haydar has no
evidence that the supposed comments about ‘you people’ and leaving his wife behind while he
commuted to Seattle were in any way tied to his religion or national origin.” (R. 70, PageID.9331
n.13.) But a reasonable jury can make reasonable inferences. Haydar testified as follows, “So
Peter [Faricy] made references to you people, to you people need to learn how to treat your
wives better, those types of comments repeatedly, you know, in reference to me uniquely
differently than other people. . . . [S]omehow his view of me was influenced by my inherent
characteristics as someone with a name Abdullah coming from a Muslim or an Arab background
that I somehow mistreat my wife.” (R. 68-1, PageID.6327.) From that, a jury could reasonably
36
infer that Faricy’s reference to “you people” was a reference to “you [Syrians]” or “you
[Muslims].” And because that is a reasonable inference, at this stage of the case, the Court must
draw it.
So Haydar clears the animus hurdle and the question becomes whether animus was “a
motivating factor” in his termination (or loss of promotion opportunities) even if legitimate
reasons, such as Haydar’s inability to “earn trust” and “disagree and commit,” primarily
motivated his termination. As the narrative set out above indicates, the connection between
animus and adverse action is not strong. But, for several reasons, the Court believes that the
record permits a reasonable jury to find that Haydar’s status as a Muslim man or SyrianAmerican man was a motivating factor in his termination.
First, the animus came from the top. Faricy was not just any Amazonian—he was the
Vice President of Marketplace (R. 68-48, PageID.6954), a sizable organization within Amazon.
And it is reasonable to think that those lower in the corporate hierarchy take cues from the top.
See Griffin v. Finkbeiner, 689 F.3d 584, 595 (6th Cir. 2012) (“As the mayor . . . Finkbeiner was
in a position to shape the attitudes, policies, and decisions of the division’s managers, . . . just as
when a major company executive speaks, everybody listens in the corporate hierarchy.” (internal
quotation marks and alterations omitted)). Indeed, Haydar testified, “I believe that Mr. Faricy is a
strong leader with a strong personality and inflicts that personality on to his organization and
people don’t say no to Mr. Faricy.” (R. 68-1, PageID.6435.) And Haydar’s expert stated, “If
someone in a leadership position speaks poorly about my peer it is likely that I’m going to echo
those concerns. It could be groupthink or it could be political astuteness. Why would you voice a
different opinion than your boss?” (R. 68-158, PageID.9262.) And Faricy made his inappropriate
remarks in front of other Amazonians.
37
Amazon attempts to make Faricy’s animus irrelevant by arguing that “Haydar’s
conspiracy theory depends on Faricy holding a supposed discriminatory animus, and then
directing his subordinates to take adverse action.” (R. 70, PageID.9325 (emphasis added).) But
that need not have been the case. If Faricy held discriminatory animus and his subordinates came
to understand that Faricy thought Haydar was a poor fit for Amazon—even if they did not know
why Faricy thought that—then Faricy’s animus would have colored how his subordinates viewed
Haydar. And if Mosby, Gaw, and Haney came to the understanding that Haydar was not a fit for
Amazon, they might have been more critical of Haydar for his deficiencies than they otherwise
would have been. This, in turn, would mean that the feedback they gave, their reactions to
Haydar’s miscues, and in the case of Gaw and Mosby, the ratings they assigned, were worse than
they would have been had Faricy thought well of Haydar.
Second, part of Amazon’s ethos is criticism. This is reflected in the 360 Feedback the
Court has reviewed at length. And the New York Times reported that “[e]very aspect of the
Amazon system amplifies the others to motivate and discipline the company’s marketers,
engineers and finance specialists” and that, in terms of feedback, Amazonians made “quiet pacts
with colleagues to bury the same person at once.” (R. 68-144, PageID.8957, 8960.) It is
reasonable to infer that a critical environment (as compared to a non-judgmental one) is more
amenable to mountains being made out of molehills. Or, in Amazon parlance, more amenable to
interpreting smaller issues with earning trust or disagreeing-and-committing as larger ones.
Cf. Rachells v. Cingular Wireless Employee Servs., LLC, 732 F.3d 652, 665 (6th Cir. 2013)
(“While evidence of a discriminatory atmosphere may not be conclusive proof of discrimination
against an individual plaintiff, such evidence does tend to add ‘color’ to the employer’s
38
decisionmaking processes and to the influences behind the actions taken with respect to the
individual plaintiff.”).
Third, a jury could find that Faricy himself was directly involved in some of the events
related to Haydar’s termination. From Joudrey’s testimony, a reasonable jury could infer that it
was Faricy who made the ultimate decision to lower Haydar’s 2013–14 growth rating from
Medium to Limited (thus knocking Haydar into the bottom 10%): “I think [Faricy] made the
final call to move [Haydar’s] ratings.” (R. 68-78, PageID.8203–8204.) And while Haydar’s first
Least Effective rating did not directly lead to a PIP or termination, it may have contributed to the
PIP a year later. (See R. 68-45, PageID.6790 (providing that two Least Effective ratings result in
an “automatic[]” PIP).) Additionally, both just before and during the PIP, Faricy took an interest
in Haydar’s employment status. (R. 68-131, PageID.8770; 68-142, PageID.8939.)
Fourth, a jury could find that the PIP was “mission impossible” for Haydar, which would
further corroborate Haydar’s theory that Faricy wanted Haydar out of Amazon and that Gaw
understood this. In addition to Haydar’s testimony that his PIP goals were a moving target
(R. 68-1, PageID.6340, 6365), there is evidence suggesting that Amazon was not interested in
seeing Haydar complete the PIP successfully. Notes from Faricy’s 2014–15 OLR, which took
place before the PIP, state, “Was doing better, but recently ‘fell off the bus’ and is being
managed out. . . . We are working on coaching him out.” (R. 68-116, PageID.8556 (emphasis
added).) Not long after that, Faricy asked Beary in HR, “Are we moving fast enough here?” To
this, Beary suspiciously replied, “[W]e will want to be buttoned up at every step with this one. It
wouldn’t be quick or easy, but if we manage it effectively—like the EB case—we’ll get to a
good outcome as quickly as possible.” (R. 68-131, PageID.8770.) Although Beary testified that a
“good outcome” included performance improvement (as opposed to just termination), “EB” was
39
in fact terminated and, apparently, sued Amazon (or Haney) for sexual harassment. (R. 68-132,
PageID.8822, 8824.) And then in August 2015—before the first 60 days of the PIP were up—
Faricy emailed Beary again: “Do we have an update on Abdullah?” (R. 68-142, PageID.8939.)
Finally, even though Gaw and Oehler extended Haydar’s PIP until September 25, 2015, Gaw’s
testimony indicates that the two had concluded that Haydar should be terminated before that
date. (See R. 68-54, PageID.7445.) Indeed, Oehler created a ticket for Haydar’s termination on
September 16, 2015. (R. 68-146, PageID.8972.) Taking all this evidence in the light most
favorable to Haydar, a jury could reasonably infer that the PIP was setup to give Haydar almost
no chance of success. And if a jury found that, it would be more likely to find that Haydar’s
supervisors understood that Faricy did not think Haydar belonged at Amazon.
Fifth, in both annual ratings, Haydar was not too far from the Highly Valued range. In
2013–14, he would have received a Highly Valued rating if he been given a Medium growth
rating instead of a Limited one—a rating which was the subject of considerable debate during
Faricy’s OLR. As for 2014–15, Mosby was initially leaning towards an Exceeds/Development
Needed rating (R. 68-30, PageID.6622), and Exceeds employees are usually given Highly
Valued ratings even if they have the lowest leadership and growth ratings (see R. 68-79,
PageID.8227). Indeed, only after McVey told Gaw that an Exceeds Amazonian would not be
placed on a PIP, did Gaw adjust Haydar’s performance rating downward to Achieves. All of this
is to say that if just some of the feedback that Haydar received had been a little better (say from
Haney and Philipsen), or if Gaw and Mosby had been a little more forgiving in how they
assessed Haydar’s feedback, Haydar might have avoided a PIP.
40
And that last point ties back to the first point: that perhaps those under Faricy treated
Haydar just a little more harshly than the average Amazonian because they understood that, in
Faricy’s view, Haydar did not belong at the company.
In short, the “burden of producing some evidence in support of a mixed-motive claim is
not onerous and [this Court] should preclude sending the case to the jury only where the record is
devoid of evidence that could reasonably be construed to support [of Haydar’s] claim.” White,
533 F.3d at 400; cf. Wallner v. Hilliard, 590 F. App’x 546, 556 n.3 (6th Cir. 2014) (“In a mixedmotive case, even if the employer’s FMLA-neutral reason did have a basis in fact, did motivate
the employer’s action, and was sufficient on its own to motivate the employer’s action, a mixedmotive FMLA plaintiff may still withstand summary judgment so long as her exercise of her
FMLA rights also contributed to the employer’s action.” (internal citation omitted)). All things
having been considered, Haydar may present a mixed-motive claim to the jury.2
2
It has not gone unnoticed by the Court that Haydar’s theory of the case seems to be
based on a cat’s-paw theory of liability. See Staub v. Proctor Hosp., 562 U.S. 411, 416 n.1
(2011). This is because Haydar does not have any evidence that the primary decisionmakers,
Gaw and Oehler (and more indirectly, Mosby), held animus toward Syrians or Muslims. And
while a reasonable jury could infer that Faricy had animus, Faricy was not the one who
ultimately decided that Haydar should be terminated. (See R. 68-54, PageID.7424; R. 68-58,
PageID.7635.) That said, the Court has not cast its mixed-motive analysis in the cat’s paw mold
because neither party mentions “cat’s paw” or cites cases that deal with that theory.
If the cat’s-paw framework controls here, Haydar might still reach a jury under that
theory. Taking the record in the light most favorable to Haydar, it seems as though a reasonable
jury could infer that Faricy had anti-Syrian or anti-Muslim animus, that Faricy’s actions (his
emails, statements in OLRs, and statements in meetings) indicated to Gaw, Mosby, Beary, and
others that he did not want Haydar to remain at Amazon, that Gaw, Mosby, Beary, and others
were influenced by Faricy’s actions and thus assessed Haydar’s leadership deficiencies more
harshly than they otherwise would have (or, similarly, were not inclined to tolerate his
deficiencies), and that the leadership deficiencies were the reason for Haydar’s termination. See
Staub, 562 U.S. at 419, 422 (holding that “if a supervisor performs an act motivated by
[discriminatory] animus that is intended by the supervisor to cause an adverse employment
action, and if that act is a proximate cause of the ultimate employment action, then the employer
is liable” and further providing that “[p]roximate cause . . . excludes only those links that are too
remote, purely contingent, or indirect” (internal quotation marks and alterations omitted));
41
4.
Although his brief not clear on this point, it appears that Haydar also pursues a singlemotive theory, i.e., that Amazon’s asserted reason for his termination is false, and that the real
reason was discrimination. (See R. 72, PageID.9628–9637.)
Assuming there is a tenable distinction between single- and mixed-motive cases, see Hon.
Timothy M. Tymkovich, The Problem with Pretext, 85 Denv. U. L. Rev. 503, 522–24, 529
(2008) (discussing the “artificial distinction between mixed motive and single motive Title VII
cases”); Griffith v. City of Des Moines, 387 F.3d 733, 744–46 (8th Cir. 2004) (Magnuson, J.,
concurring) (reasoning that following 1991 amendments to Title VII and Desert Palace, singleand mixed-motive claims rest on a “fictional dichotomy” and that the “only rational conclusion is
that no distinction between single and mixed motives exists”), it seems that success on a singlemotive theory would require an employee to prove that his protected characteristics alone were
sufficient to motivate the employer to take the adverse action, see Price Waterhouse, 490 U.S. at
247 (plurality) (explaining that McDonnell-Douglas/Burdine framework is premised on “either a
legitimate or an illegitimate set of considerations led to the challenged decision”); id. at 260
(O’Connor, J.) (“In pretext cases, ‘the issue is whether either illegal or legal motives, but not
both, were the true motives behind the decision.’ In mixed-motives cases, however, there is no
one ‘true’ motive behind the decision.” (quoting NLRB v. Transportation Management Corp.,
462 U.S. 393, 400, n.5 (1983)); Spees v. James Marine, Inc., 617 F.3d 380, 390 (6th Cir. 2010)
(explaining that in a mixed-motive case “both legitimate and illegitimate reasons motivated the
employer’s decision” but in a single-motive case “an illegitimate reason motivated an
employment decision”); White v. Baxter Healthcare Corp., 533 F.3d 381, 400 (6th Cir. 2008)
Chattman v. Toho Tenax Am., Inc., 686 F.3d 339, 351 (6th Cir. 2012) (applying Staub in Title
VII context).
42
(providing that in single-motive cases, the McDonnell-Douglas/Burdine framework serves to
“smok[e] out the single, ultimate reason for the adverse employment decision” (internal
quotation marks omitted)); Costa v. Desert Palace, Inc., 299 F.3d 838, 856 (9th Cir. 2002) (en
banc) (“If, based on the evidence, the trial court determines that the only reasonable conclusion a
jury could reach is that discriminatory animus is the sole cause for the challenged employment
action or that discrimination played no role at all in the employer’s decisionmaking, then the jury
should be instructed to determine whether the challenged action was taken ‘because of’ the
prohibited reason.” (citing 42 U.S.C. § 2000e 2(a)). Making this the dividing line between
single- and mixed-motive cases means that for Haydar to succeed on a single-motive theory he
must show that his status as a Syrian-American and as a Muslim were alone enough of a reason
for Amazon to terminate his employment (or deny him a promotion).
That is a considerable showing and the Court doubts that Haydar can make it. What has
already been said goes a long way to explaining the Court’s doubt. Additionally, the Court notes
that 360 Feedback for Haydar’s 2014–15 review reveals that eight Amazonians—none of whom
were Haney, Philipsen, Gaw, or Mosby—all cited Haydar’s communication problems. (R. 68112, PageID.8500 (Degwekar); R. 68-112, PageID.8501 (Tollini); R. 68-112, PageID.8503
(Welzel); R. 68-112, PageID.8504 (Poznanski); R. 68-112, PageID.8505 (Marx); R. 68-112,
PageID.8510 (Westlake); R. 68-112, PageID.8515 (Keller); R. 68-112, PageID.8518 (Yavno).)
All of that said, it strikes the Court that whether this case should be cast as a singlemotive or mixed-motive case is more a jury-charge question than a summary-judgment question.
The Court has already found that Haydar may proceed to trial on a mixed-motive theory. And, at
trial, whether he pursues a single- or mixed-motive theory (or both), the parties’ strategies will
likely be the same: Amazon will stress legitimate reasons for its actions and Haydar will stress
43
discriminatory ones. See Griffin v. Finkbeiner, 689 F.3d 584, 594 n.7 (6th Cir. 2012) (“We note
that single-motive and mixed-motive theories are not distinct claims, but rather different ways of
analyzing the same claim.”). As a practical matter then, there is no need to decide whether a
reasonable jury could find for Haydar on a single-motive theory. See Wallner v. Hilliard, 590 F.
App’x 546, 552 (6th Cir. 2014) (“Although Wallner has pursued an FMLA retaliation claim on
both single-motive and mixed-motive theories—arguing them in the alternative—we see no need
to inquire whether Wallner has marshaled enough evidence to withstand summary judgment on a
single-motive theory. . . . Because the operative statutory language permits recovery on a mixedmotive theory, we see no reason to produce dicta about whether Wallner possesses even more
evidence than is required by the statute.”); Griffith, 387 F.3d at 744 (Magnuson, J., concurring)
(“There is no need for a plaintiff to prove the more onerous single-motive case, when all that
Title VII requires a plaintiff to prove is that discrimination was a motivating factor in the
employment decision.”). The Court will thus wait to hear the evidence at trial to decide whether
single- or mixed-motive jury instructions (or both) are proper. See Costa v. Desert Palace, Inc.,
299 F.3d 838, 856 (9th Cir. 2002) (en banc); Ninth Circuit Model Jury Instructions § 10.1 & cmt.
(2017), available at http://www3.ce9.uscourts.gov/jury-instructions/node/167.3
3
For what it is worth, the Court did examine the fourth element of McDonnell-Douglas’
prima facie case: whether Haydar (1) was “replaced by someone outside the protected class” or
(2) was “treated differently than similarly-situated, non-protected employees.” Peeples v. City of
Detroit, 891 F.3d 622, 634 (6th Cir. 2018). At best, Haydar barely scratches out this fourth
element.
Regarding the first possibility, shortly after Haydar was terminated, Anas Fattahi—who,
like Haydar, is male, married, Syrian-American, and Muslim—took over one of Haydar’s teams
and, four months later, took over the other. Although Haydar is correct that Fattahi was not
promoted to L7 until after Haydar filed his EEOC complaint, Haydar overlooks the fact that
feedback for Fattahi’s promotion was solicited earlier, in January 2016. (Compare R. 68-155,
PageID.9235 with R. 68-156, PageID.9241.) That is, Gaw had, apparently, been considering
Fattahi for a promotion before Amazon learned of the EEOC complaint.
44
5.
Thus far, the Court has applied only Title VII law in addressing Haydar’s claims of
national-origin and religious discrimination. There may be differences between a mixed-motive
claim under Title VII and one under ELCRA. But if there are, Amazon has not raised them. It
states, “Because ELCRA claims are analyzed using the same standards as those brought under
Title VII, the analysis of Haydar’s Title VII national origin and religious discrimination claims
applies equally to his ELCRA claims.” (R. 69, PageID.9302.)
Given that Amazon has asserted that the same result is required under Title VII and
ELCRA, and given this Court’s Title VII analysis, the Court finds that Amazon has not
discharged its summary-judgment burden on Haydar’s national-origin and religious
discrimination claims under ELCRA. Those too will go to trial.
As for the second possibility, Haydar initially identified Gaw, Mosby, and Haney as
Amazonians who were neither Syrian-American nor Muslim and were treated better. (R. 72,
PageID.9635.) In supplemental briefing, Haydar identified six more Amazonians that are
allegedly comparable to him. (R. 81, PageID.10111–10112.) The Court has painstakingly
reviewed the evidence of record regarding all of these individuals’ ratings—including the
lengthy performance evaluations and 360 Feedback for all nine asserted comparables. That
review revealed that only MK and JT even arguably had similar leadership deficiencies as
Haydar (yet had better ratings). (See R. 80, PageID.9848, 9855, 9856, 9928, 9935, 9936.) But
these two Amazonians had different supervisors than Haydar; and their supervisors may have,
for reasons entirely unrelated to national origin or religion, assessed 360 Feedback differently
than Gaw or Mosby. And while the Court acknowledges that Mosby was promoted to L8 despite
negative 360 Feedback (R. 80-13), it is also true that Mosby was given consecutive
“Outstanding” performance ratings prior to promotion (R. 68-88, PageID.8375). It may be that
Mosby’s accomplishments significantly outweighed his leadership deficiencies.
45
C.
What remains of Amazon’s motion is whether it is entitled to summary judgment on
Haydar’s claims of retaliation under Title VII and ELCRA.
Before turning to the parties’ arguments, the Court notes that there may be differences in
the legal standards governing Title VII and ELCRA retaliation claims. But again, if there are
differences, neither Haydar nor Amazon has raised them. Haydar asserts: “Courts follow the
same analysis when determining whether an employee has been subjected to discrimination or
retaliation under Title VII and the Elliott-Larsen Civil Rights Act.” (R. 72, PageID.9627 n.6.)
And Amazon says, “The Title VII [retaliation] analysis applies equally to Haydar’s ELCRA
claim.” (R. 69, PageID.9311 n.19.) So the Court will analyze Haydar’s retaliation claims under
Title VII standards.
To “establish a prima facie case of retaliation [under Title VII,] a plaintiff must establish
that: (1) she engaged in a protected activity; (2) her exercise of such protected activity was
known by the defendant; (3) thereafter, the defendant took an action that was ‘materially
adverse’ to the plaintiff; and (4) a causal connection existed between the protected activity and
the materially adverse action.” Rogers v. Henry Ford Health Sys., 897 F.3d 763, 775 (6th Cir.
2018) (internal quotation marks omitted). And “the final element requires proof of so-called ‘butfor’ causation,” i.e., that the unlawful retaliation would not have occurred but for the employee’s
protected conduct. See Mys v. Michigan Dep’t of State Police, 886 F.3d 591, 600 (6th Cir. 2018);
MacEachern v. Quicken Loans, Inc., No. 17-1005, 2017 WL 5466656, at *5 (6th Cir. Oct. 17,
2017).
46
Amazon asserts that “the but-for causation standard obliterates Haydar’s [retaliation]
claim.” (R. 69, PageID.9315.) Although the Court would use less colorful language, it agrees
with Amazon’s bottom line.
A good place to start the analysis is to identify the two ends of the causal chain: the
protected conduct and the adverse action. Haydar asserts that the protected conduct was his
March 5, 2015 phone call with Cerio. (R. 72, PageID.9638.) According to Haydar’s notes (but
not Cerio’s), Haydar told Cerio the following on March 5th: “Peter was picking on me as if I
have an inherently bad relationship with my wife due to my ethnicity or religion.” (R. 68-124,
PageID.8640.) Haydar also asserts—in vague and conclusory fashion—that even if the March 5
call was not protected conduct, his “written complaints to [Cerio], Bezos, and others constitute
protected activity.” (R. 72, PageID.9639.) But it appears that Haydar’s written complaints to
Cerio were all around March 5 (R. 68-20, PageID.6584 (February 24, 2015); R. 68-21,
PageID.6588 (March 5, 2015); R. 68-22, PageID.6593 (April 16, 2015)), and so the causal
analysis as to those emails do not materially differ from the March 5 call. The Court, however,
will address Haydar’s email to Bezos separately. As for the “materially adverse” action,
Burlington N. & Santa Fe Ry. Co. v. White, 548 U.S. 53, 57 (2006), Haydar has identified the
PIP and termination. (R. 72, PageID.9639–9640.)
With the end points established, the Court can examine the connection between them. As
for being placed on a PIP, McVey advised that a PIP was the proper course in February 2015.
(R. 68-121, PageID.8597.) And Faricy testified that following two Least Effective ratings, a PIP
would be “automatic.” (R. 68-45, PageID.6790.) That is relevant because it was decided that
Haydar would receive a second Least Effective rating in February 2015. (R. 68-20,
PageID.6584.) In other words, a PIP was the plan by February 2015. (R. 68-45, PageID.6790.)
47
Yet the protected conduct—Haydar’s call to Cerio—was on March 5, 2015 (or, at earliest, the
same day he learned of his second Least Effective rating). So no reasonable jury could find that a
complaint to Cerio was a but-for cause of the PIP.
The Court likewise finds that no reasonable jury could find that Haydar’s complaints to
Cerio were a but-for cause of his termination.
Timing again is not in Haydar’s favor. He was terminated more than six months after his
March 5 call and about five months after his April 16 email to Cerio. See Nguyen v. City of
Cleveland, 229 F.3d 559, 567 (6th Cir. 2000) (noting that the “cases that have permitted a prima
facie case to be made based on the proximity of time have all been short periods of time, usually
less than six months” (internal quotation marks omitted)); Hafford v. Seidner, 183 F.3d 506, 515
(6th Cir. 1999) (finding, where five months separated protected conduct and adverse action, that
without something in addition to timing, the required casual connection was lacking).
Second, the record does not suggest that anyone at Amazon had much of a motive to
retaliate. When Haydar told Cerio about Faricy’s comments, her reaction was to conduct a
thorough investigation and to bring in a lawyer who she viewed as “neutral” to conduct it. And
the people primarily involved in terminating Haydar—Gaw and Oehler—were not the subject of
Haydar’s complaints of discrimination. As for Faricy, DeCleene cleared him of violating “any
Amazon policy.” (R. 68-48, PageID.6960.) So it is even a stretch to infer that Faricy had much of
a motive to retaliate for Haydar’s complaint of bias. Indeed, Faricy apparently did not inquire
into Haydar’s status until a month after his interview with DeCleene—and that inquiry was
prompted by running into Haydar and an email that Haydar had sent about a tax issue. (R. 68131, PageID.8770.) All said, Haydar points to nothing in the vast summary-judgment record
48
indicating that Gaw, Oehler, or Faricy ever mentioned Haydar’s complaint once DeCleene’s
investigation was complete.
In short, there was a five- or six-month gap between the protected conduct and adverse
action, the protected conduct involved complaints against those who were not the primary
decision-makers, and the protected conduct triggered a thorough investigation into the
complained-of remarks. In all, on this record, a reasonable jury could not find that had Haydar
not made his complaints to Cerio, he would have kept his job. See Univ. of Texas Sw. Med. Ctr.
v. Nassar, 570 U.S. 338, 360 (2013) (“Title VII retaliation claims must be proved according to
traditional principles of but-for causation, not the lessened causation test stated in § 2000e–
2(m).”).
Haydar makes several arguments for a different result. None persuade.
Haydar claims he has “direct evidence” that he was terminated because of his complaints
to Cerio. In support, Haydar points to the portion of DeCleene’s report where DeCleene faulted
Haydar for a pattern of “push[ing] back and/or escalat[ing] concerns.” (R. 68-48, PageID.6957.)
But, read objectively, DeCleene’s reference was to Haydar’s propensity to complain about his
ratings and other work-related criticism to human-resources; she was not referencing Haydar’s
complaints of ethnic or religious bias. (See R. 68-48, PageID.6957.)
Haydar also attempts to establish a causal connection by arguing that “individuals who
did not engage in protected activity were not terminated” and so, Haydar implies, his protected
conduct must have been a significant reason for his termination. (See R. 72, PageID.9639.) But
for this argument to hold sway, it must be that the primary difference between Haydar and an
Amazonian who was not terminated was the protected conduct. Otherwise, another difference
among the two—aside from protected conduct—could very well explain why one was terminated
49
and was not. Yet Haydar has identified no Amazonian (1) with his track record, (2) who kept
their job, and (3) did not make a complaint of discrimination.
Haydar also seeks to strengthen the connection between his protected conduct and
termination by arguing that the reasons for his termination were false. (R. 72, PageID.9639.) But
even if Amazon’s reason for terminating Haydar was false, Haydar points to nothing in the
record suggesting that it was to cover up retaliation for his reports to Cerio. See Mulvey v.
Hugler, No. 17-5633, 2018 WL 2771346, at *4 (6th Cir. Apr. 3, 2018) (providing that at the third
step of McDonnell Douglas the plaintiff must show that the employer’s explanation was a pretext
“to hide unlawful retaliation”); Tingle v. Arbors at Hilliard, 692 F.3d 523, 530 (6th Cir. 2012)
(“[T]he employee also must offer some evidence that not only were the employer’s reasons false,
but that retaliation was the real reason for the adverse action.”); cf. Reeves v. Sanderson
Plumbing Prod., Inc., 530 U.S. 133, 148 (2000) (noting that “there will be instances where,
although the plaintiff has established a prima facie case and set forth sufficient evidence to reject
the defendant’s explanation, no rational factfinder could conclude that the action was
discriminatory”). Indeed, Haydar offers another reason for the fabrication: discrimination.
Finally, and perhaps as an alternative theory of retaliation, Haydar points to his
September 2015 email to Bezos and the fact that the very same day—indeed, two minutes after
Cerio was sent a copy of the email—Oehler created a ticket for Haydar’s termination. But
Haydar has no evidence that Cerio, let alone Oehler, saw the email in the two minutes before
Oehler created the ticket. Moreover, both Oehler and Gaw testified that they did not see
Haydar’s email to Bezos until after Haydar was terminated. And all Haydar has to rebut that
testimony is an implication that Oehler is not credible because his deposition testimony was
initially—but not ultimately—equivocal. That is not enough to survive summary judgment. See
50
Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 256 (1986) (explaining that, to survive summary
judgment, a plaintiff cannot “merely assert[] that the jury might, and legally could, disbelieve the
defendant’s denial”).
In short, Haydar’s evidence of retaliation amounts to not much more than termination
five or six months after protected conduct. So Amazon is entitled to summary judgment on
Haydar’s retaliation claim under Title VII. See Vereecke v. Huron Valley Sch. Dist., 609 F.3d
392, 401 (6th Cir. 2010) (“At this point, our case law can fairly be characterized as recognizing
the possibility that, on a particular set of facts, extremely close temporal proximity could permit
an inference of retaliatory motive, but also recognizing that often evidence in addition to
temporal proximity is required to permit the inference.”). And given Haydar’s concession that
his Title VII and ELCRA claims are governed by the same legal standards, Amazon is also
entitled to summary judgment on Haydar’s retaliation claim under ELCRA.
IV.
For the reasons given, the Court GRANTS IN PART and DENIES IN PART Amazon’s
motion for summary judgment. (R. 62, 69.) Amazon is entitled to summary judgment on
Haydar’s claims that Amazon discriminated on the basis of marital status (Count VI), violated
Michigan public policy (Count VIII), and retaliated in violation of Title VII and ELCRA (Counts
III, VII). But Haydar may present his claims of national-origin discrimination (Counts I, IV) and
religious discrimination (Counts II, V) to a jury.
SO ORDERED.
Dated: September 7, 2018
s/Laurie J. Michelson
U. S. DISTRICT JUDGE
51
CERTIFICATE OF SERVICE
The undersigned certifies that a copy of the foregoing document was served on the
attorneys and/or parties of record by electronic means or U.S. Mail on September 7, 2018.
+
s/Teresa McGovern
Case Manager Generalist
52
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?