United States District Court, District of Columbia
RONDA L. DAVIS, et al., Plaintiffs,
DISTRICT OF COLUMBIA, Defendant. Re Document Nos 144, 146
MEMORANDUM OPINION GRANTING DEFENDANT'S MOTION
FOR SUMMARY JUDGMENT; DENYING AS MOOT PLAINTIFFS' MOTION
FOR CLASS CERTIFICATION
RUDOLPH CONTRERAS United States District Judge
in this case are former employees of the District of Columbia
Child and Family Services Agency who were terminated in a
broad layoff of agency staff. In this putative class action
lawsuit, Plaintiffs allege that their termination was the
result of a discriminatory downsizing of the agency's
workforce. Specifically, Plaintiffs allege that the firings,
referred to as a reduction in force, were the result of
discrimination based on both race and age. Plaintiffs also
allege that the agency's imposition of a bachelor's
degree requirement for the newly created Family Social Worker
position was discriminatory as to both race and age.
extended litigation, including the Court's resolution of
initial dispositive motions and fact and expert discovery by
the parties, the Court now considers two pending motions.
First, Plaintiffs have renewed their motion for class
certification. See Pls.' Mot. Class
Certification (“Mot Class Certification”), ECF
No. 144. Second, the District of Columbia has moved for
summary judgment. See Def.'s Mot. Summ. J.
(“Mot. Summ. J.”), ECF No. 146. For the reasons
set forth below, the Court will grant the District of
Columbia's motion for summary judgment. Because the Court
grants summary judgment for the District on all remaining
claims, the Court does not need to reach Plaintiffs'
motion for class certification. Thus, the Court denies that
motion as moot.
Court will begin with a description of the facts giving rise
to the parties' dispute and then turn to the procedural
history of this case.
Child and Family Services Agency
District of Columbia Child and Family Services Agency
(“CFSA”) is entrusted with protecting the safety,
permanence, and well-being of abused and neglected children
and with providing services to struggling families.
See Def.'s Statement Undisputed Material Facts
(“Def.'s Material Facts”) ¶ 1, ECF No.
146-2. Although the parties dispute the exact organizational
structure of CFSA during the relevant time period,
compare Id. ¶¶ 2, 5 with
Pls.' Statement Disputed Material Facts (“Pls.'
Disputed Facts”) at 2-3, ECF No. 148-1, it is
undisputed that many of CFSA's “frontline
functions” are led by the Office of Agency Programs,
including investigating reports of child abuse and neglect,
temporarily removing children from dangerous situations, and
providing direct case management. See Def.'s
Material Facts ¶ 3; Pls.' Disputed Facts ¶ 1.
District of Columbia law and the consent decree entered in
the class action LaShawn v Bowser mandate the
provision of many of these services. See Def.'s
Material Facts ¶ 4; Pls.' Disputed Facts ¶ 1;
see also LaShawn v. Bowser, No. 89-1754 (D.D.C. Feb.
27, 2007), ECF No. 864 (order approving Amended
experienced significant budgetary pressure in fiscal years
2010 and 2011. In FY 2010 (October 1, 2009 to September 30,
2010), CFSA's budget decreased by $25.3 million from the
prior year. See Def.'s Material Facts ¶ 7.
The FY 2010 budget reduced the number of fulltime employees
from 940 to 892. See Id. ¶ 8. In FY 2011,
CFSA's budget was reduced by $12.1 million and the number
of approved full-time employees was reduced further from 892
to 840. See Id. ¶ 10. The agency was required
to make the required budget and employee cuts before October
1, 2010, the beginning of FY 2011. See Id. ¶
11. CFSA implemented a reduction in force (“RIF”)
to meet the necessary reduction in personnel costs. See
Id. ¶ 12. The RIF, which had an effective
termination date of June 11, 2010, eliminated a total of 123
positions. See Id. ¶¶ 12-13. All
told, 115 employees were separated from CFSA as a
result of the RIF. See Id. ¶ 14.
Reduction in Force
parties view the RIF in drastically different light.
Plaintiffs posit that the RIF was a simple across-the-board
cut with discriminatory intentions and effects. Plaintiffs
point to a memorandum dated April 29, 2010 from CFSA Director
Roque Gerald to City Administrator Neil Albert that asked for
“approval to conduct a Reduction-in-Force . . . to
abolish one hundred and twenty-three . . . positions within
the Child and Family Services Agency.” Pls.' Mem.
P. & A. Opp'n Def.'s Mot. Summ. J.
(“Opp'n Summ. J.”) at 5, ECF No. 148 (quoting
Mot. Summ. J., Ex. F at 1, ECF No. 146-3). The memorandum
explained that CFSA “must conduct a realignment to
consolidate functions in accordance with the FY'2011
budget and internal re-engineering, ” and that
“[t]he deficit resulting from the realignment will
precipitate a reduction in force.” Id.
Plaintiffs note that the memorandum provided an
“agency-wide” list of positions, which
purportedly affected virtually every office of CFSA.
See Opp'n Summ. J. at 6.
Director sent a letter dated May 6, 2010 that informed the
115 CFSA employees who would be terminated by the RIF.
See Opp'n Summ. J., Ex. 2 at 1, ECF No. 148-2.
The letter did not provide a reason for the RIF, but stated
that employees would “be separated from District
government service effective 6/11/10” and that the RIF
was conducted in “accordance with Chapter 24 of the
District's Personnel Regulations and in no way reflects
adversely on your performance of your official duties.”
contrast, the District contends that the RIF was the result
of a series of separate decisions. The District claims that
“CFSA did not utilize a single uniform criteria, test
or requirement for determining which employees” would
be included in the RIF. Def.'s Mem. P. & A. Supp.
Def.'s Mot. Summ. J. (“Mem. Supp. Summ. J.”)
at 4, ECF No. 146. Instead, relying on the declaration of
Raymond Davidson, CFSA's Chief Administrative Officer at
the time of the RIF, the District contends that the decisions
to eliminate certain positions were the result of multiple
individual choices by the Director, working in consultation
with numerous managers and senior leaders in the agency.
See Def.'s Material Facts ¶ 15; see
also generally Mot. Summ. J., Ex. E (“Davidson
Decl.”), ECF No. 146-3. But see Pls.'
Disputed Facts ¶ 3. Mr. Davidson's declaration also
focused on CFSA's attempts to create new service models
for providing services in the District. See Davidson
Decl. ¶¶ 6-7, 10-12, 14.
Newly Created Family Support Worker Position
parties also apply their competing narratives to discrete
components of the RIF, particularly the elimination of the
Social Service Assistant (“SSA”) and Social
Worker Associate (“SWA”) positions and the
creation of a new Family Support Worker (“FSW”)
to Mr. Davidson, CFSA determined that it should adopt a model
where Social Workers “are ‘teamed' with a set
of skilled partners in order to more effectively serve the
needs of the Agency's children and family
clients.”' Davidson Decl. ¶ 6; see
also Def.'s Material Facts ¶ 18. To advance
that model, CFSA eliminated the SSA and SWA positions through
the RIF in favor of a newly created FSW position.
See Davidson Decl. ¶ 7; see also
Def.'s Material Facts ¶ 19. The elimination of the
SSA and SSW positions accounted for the majority of the 115
employees terminated in the RIF. Def.'s Material Facts
¶ 20; Pls.' Disputed Facts ¶ 6.
posted openings for the new FSW positions at the same time it
informed employees of the RIF. See Opp'n Summ.
J., Ex. 3 at 7, ECF No. 148-2. The FSW position required a
bachelor's degree in a social service field. Def.'s
Material Facts ¶ 32; see also Pls.'
Disputed Facts ¶ 8 (acknowledging the degree
requirement). Debra Porchia-Usher, CFSA Deputy Director for
Agency Programs, stated that FSWs would be “able to
perform casework activities that support investigative social
workers to obtain the information necessary to complete
investigations of abuse and neglect, including interviewing
family members, children, and caretakers[;] assessing the
safety of a home and the well-being of children[;] conducting
home visits; documenting information into the case record;
coordinating with other team members for meetings; and
coordinating needed services for families and
children.” See Mot. Summ. J., Ex. B
(“Porchia-Usher Decl.”) ¶ 12, ECF No. 146-3.
contend that the SSA and FSW positions involved the same
responsibilities, and that the only difference between the
positions was the FSW's degree requirement. Opp'n
Summ. J. at 8-11. Plaintiffs present CFSA's job
descriptions of the relevant positions, and suggest that the
descriptions are largely identical. See Opp'n
Summ. J. at 8-10; Porchia-Usher Decl., Exs. A-B. For,
instance, the SSA responsibilities included:
. “[P]roviding direct support to
social work staff and the social work function.”
Porchia-Usher Decl., Ex. A.
. “Conducting] non-clinical home
visits accompanying a Social Worker or a Social Service
Assistant, as needed, for reasons of safety . . . .”
. “Providing] transportation
assistance for clients . . . .” Id.
. “Supporting] social workers
and supervisory social workers in implementing service plans
by supervising/facilitating visits, making referrals or
scheduling service with providers . . . .” Id.
. “Participating] in case and
supervisory conferences as needed; bring[ing] problem issues
to the attention of the worker and/or supervisor for
discussion in these conferences.” Id.
. “Assisting] the social
worker in completing specified paper searches to locate
hard-to-find families by searching and clarifying data,
checking files, and contacting other agencies.”
job posting included a number of similar responsibilities,
. “Performing] casework, group work,
and community organization work under the supervision of a
Licensed Independent Social Worker . . . ”
Porchia-Usher Decl, Ex. B.
. “Accompanying] a Social Worker as
needed, for reasons of safety or participating] in home
visits and/or investigations to determine the needs of
clients . . . .” Id.
. “Providing] transportation
assistance for clients . . . .” Id.
. “Supporting] social workers
and supervisory social workers in implementing service plans
by supervising/facilitating visits . . . .”
. “With guidance, developing] plans
for and providing] appropriate assistance and services on a
continuing basis to children and family members ”
. “Assisting] the social
worker in completing specified paper and record searches to
locate hard-to-find families by engaging diligent search
services or by searching and clarifying existing data,
checking files and contacting other agencies as
also present evidence that SSAs were already completing some
of the additional responsibilities-specifically casework and
assessments during home visits-that were included in the FSW
job description. Plaintiff Darius Morris, for instance,
testified during a deposition that SSAs routinely made home
visits and assessments, but that CFSA did not enter those
visits and assessments into its database system. See
Opp'n Summ. J., Ex. 11 at 21-26, ECF No. 148-3. In her
response to an interrogatory, Plaintiff Stephanie Alston
stated that, despite her lack of a bachelor's degree, the
Georgia Avenue Family Support Collaborative, a CFSA
contractor, hired her as a Family Support Worker.
See Opp'n Summ. J., Ex. 14 at 2, 4, ECF No.
148-6.Plaintiffs contend that CFSA's decision
to contract the FSW function to an outside vendor, who in
turn hired Ms. Alston, shows that a bachelor's degree was
not required to perform the duties of a FSW. See
Opp'n Summ. J. at 10-11.
Plaintiffs present an email sent by CFSA employee Jenna Beebe
on May 26, 2010. See Opp'n Summ. J., Ex. 6 at 1,
ECF No. 148-2. Ms. Beebe states that “one of the other
supervisory interviewers mentioned to me that they felt that
the FSW would be used in the exact fashion SSA's were
prior to RIF.” Id. Ms. Beebe went on to say
that, in her understanding, “FSW's will have much
higher requirements and acquired skills.” Id.
In response, Ms. Porchia-Usher stated that CFSA would need
“to be conscious that all FSW know the
expectations” and that “performance evaluations
[clearly] outline the current position expectations.”
Hiring for the FSW Position
posting for the newly created FSW position stated that the
position would be open as of May 5, 2010 and that CFSA
planned to fill 35 vacancies. See Opp'n Summ.
J., Ex. 4 at 1, ECF No. 148-2. On June 7, 2010, CFSA Director
Gerald sent an email stating that “[t]he first of our
new Family Support Workers (FSWs) came on board today”
and that each of the 17 new FSWs “are former CFSA
employees we're re-hiring after the reduction in force
last month.” Opp'n Summ. J., Ex. 5 at 1, ECF No.
148-2. Director Gerald also stated that “Human
Resources is prepared to make offers to an additional group
shortly. All those candidates are from outside CFSA.”
to the District's data, 44 employees who were separated
through the RIF applied for an FSW position. See
Def.'s Material Facts ¶ 36. Of the 44 applicants, 30
held a bachelor's degree and therefore satisfied the
minimal education requirements of the position. Id.
¶ 37. Two declined to be interviewed and two others took
other positions in the agency. Id. ¶ 38.
Following an interview of all qualified applicants, CFSA
hired 18 employees who had been terminated in the RIF.
Id. ¶ 41. The District contends that all 18
re-hires were African-American and 7 were age 40 or older.
Id. ¶¶ 42-43. Plaintiff Mr. Ajakaiye was
one of the 18 initial FSW hires, and is the only plaintiff in
this case who was hired for that position. Opp'n Summ. J.
the process of discovery, the parties retained experts to
provide statistical analyses of the RIF and related decisions
by CFSA. Plaintiffs hired Dr. Paige Munro, who has a Ph.D. in
industrial organizational psychology. See Mot. Summ.
J., Ex. H (“Munro Decl.”) ¶ 1, ECF No.
146-3. The District hired Dr. Stephen Bronars, who has a
Ph.D. in economics. See Mot. Summ. J., Ex. J ¶
2. Both of the experts produced two reports using statistical
analyses to examine the purported impact of CFSA's
Dr. Munro's July 2012 Report
Munro submitted her initial report for the Plaintiffs on July
28, 2012. See Munro Decl. at 4; see also
Mot. Summ. J., Ex. I, ECF No. 146-3 (Dr. Munro's report).
Dr. Munro based her report on data provided by the District
in the declaration of Stan Spaght, the Human Resources
Manager for Compensation/Benefits at CFSA. Id.
¶ 2; see also Mot. Summ. J., Ex. I at 1. Mr.
Spaght stated that, before the RIF, CFSA had a total
workforce of 832. See Opp'n Summ. J., Ex. 10
(“Spaght Decl.”) ¶ 5, ECF No.
148-2. According to Mr. Spaght, prior to the RIF, 82.8% of
employees were African-American and 62.7% were age
40 and over. Id. Finally, Mr. Spaght's
declaration stated that, of the 115 workers terminated in the
RIF, 93% were African-American, and 74.8% were workers 40 and
on that data, Dr. Munro calculated that, of the 832 workers
at CFSA before the RIF, approximately 689 were
African-American and 143 were other races. See Mot.
Summ. J., Ex. I at 1. Similarly, approximately 522 employees
were 40 and over and 310 were younger than 40. Id.
Dr. Munro also calculated that 107 employees terminated in
the RIF were African-American (8 were other races), and 86
were 40 and over (29 were younger than 40). Id. Dr.
Munro then conducted a statistical analysis to determine
whether the rate of termination for African-Americans and
employees 40 and over differed in a statistically significant
way from the termination rates for non-African-American
employees and employees under 40. Id. Dr.
Munro's analysis concluded that the RIF termination rate
for African-Americans was 15.5%, while the same rate for
non-African-Americans was 5.6%. Id. at 3-4. A
parallel analysis concluded that the RIF termination rate for
employees 40 and over was 16.5%, while the same rate for
those under 40 was 9.4%. Id. at 4-5.
Munro's analysis relies on the 4/5ths rule, which she
refers to as a “rule of thumb to evaluate adverse
impact in a hiring or termination scenario.”
Id. at 3. Simply put, “if 100% of a majority
group pass a test then the minority group should not pass at
a rate lower than 80% or the test result is said to be
adversely impacting the minority group.” Id.
Dr. Munro concluded that the RIF violated the 4/5ths rule on
the basis of both race and age. Id. 4-5. Dr. Munro
also concluded that both results are strongly statistically
significant. Id. Finally, Dr. Munro found that
re-hiring 18 African-American to the FSW position “does
not change the net results notably.” Id. at 4.
Dr. Bronars's January 2014 Report
District's expert, Dr. Bronars, produced his initial
report on January 8, 2014. See Mot. Summ. J., Ex. J.
Dr. Bronars begins his analysis with a different total number
of CFSA employees at the time of the RIF. Dexter Starkes,
CFSA Director of Human Resources, stated in his declaration
that 30 employees in the office of the Fiscal Operations
Administration “contribute to the agency's overall
headcount, [but] they report to the Office of the Chief
Financial Officer and are not technically CFSA employees. As
a result, these positions were not subject to inclusion in
the RIF.” Mot. Summ. J., Ex. C ¶ 7 (“Starkes
Decl.”), ECF No. 146-3. Therefore, Dr. Bronars began
his analysis with 802 employees at the time of the RIF,
instead of 832. See Mot. Summ. J., Ex. J.
Bronars took a very different statistical approach than Dr.
Munro. Dr. Bronars explained that Dr. Munro's approach
“is based on the assumption that all CFSA employees
faced the same risk of termination during the RIF.”
Id. ¶ 11. Instead, Dr. Bronars stated that his
“analysis allows the risk of termination to vary
between divisions within the CFSA and among certain jobs
within the CFSA.” Id. ¶ 12. Dr. Bronars
based his assumption of disparate risk on the declaration of
Raymond Davidson, CFSA's Chief Administrative Officer at
the time of the RIF. Id.; see also Davidson
Decl. Therefore, instead of one analysis of all CFSA
employees, Dr. Bronars “assume[d] that there are 7
different sets of layoff decisions made by CFSA during the
RIF” corresponding to 7 different positions or
divisions affected by the downsizing. See Mot. Summ.
J., Ex. J. ¶ 21.
Bronars also used a different standard for statistical
significance. Relying on Supreme Court precedent, he would
“conclude that there is statistical evidence of
discrimination . . . if the difference between the actual
outcomes of the termination process and the outcomes that
would have been expected from a random neutral process is so
large that it would have occurred 5% of the time.”
Id. ¶ 18. This difference corresponds to two
standard deviations. Id. ¶ 19.
Bronars's analysis concluded that, “while 107
African American employees were laid off during the RIF, had
employees within each of the seven position and division
groups been selected in a race-neutral manner, 105.21 African
American employees would have been terminated.”
Id. ¶ 22. Similarly, “while 81 employees
age 40 and above were laid off during the RIF, had employees
within each of the seven position and division groups been
selected in an age-neutral manner, 79.09 employees . . .
would have been terminated.” Id. ¶ 23.
Dr. Bronars found that neither difference was statistically
significant. Id. ¶¶ 22-23.
Dr. Bronars turned to the hiring for the FSW position. Dr.
Bronars contends that Dr. Munro “effectively assumes
that all 115 individuals who were initially terminated during
the RIF were equally qualified and equally interested in an
FSW position.” Id. ¶ 24. Based on
information provided by the Attorney General's Office,
Dr. Bronars states that 44 former employees applied for the
FSW position, 2 withdrew or declined to be interviewed, and
14 were determined not to be qualified for the position.
Id. ¶ 25. Ultimately, 18 of the 28 qualified
and interested former CFSA employees were re-hired as FSWs.
Id. Because all 18 re-hires were African American,
Dr. Bronars concludes “[t]here is, of course, no
possibility of racial disparity in the re-hiring
to the age of applicants, Dr. Bronars states that “11
of the former CFSA employees re-hired into the FSW position
were under age 40 and 7 were age 40 and above.”
Id. ¶ 26. Dr. Bronars notes that half of the
interested and qualified applicants for the FSW positon were
age 40 and above. Id. Thus, in an “age-neutral
process one would expect that half of the 18 re-hired former
CFSA employees would be under age 40 and half would be age 40
and above.” Id. Dr. Bronars concludes that
“[t]he difference between 7 [actual] and 9 [expected]
re-hired employees is equivalent to a difference of 1.58
standard deviations and is therefore statistically
Dr. Munro's May 2014 Report
Munro submitted a rebuttal to Dr. Bronars's report on
dated May 7, 2014. See Mot. Summ. J., Ex. K, ECF No.
146-4. To start, Dr. Munro raises issues with the updated
data provided by the District and relied upon by Dr. Bronars.
Dr. Munro concludes that 30 employees are not accounted for
in the updated data and that racial identification is not
provided in 84 cases. Id. at 2. Nevertheless, Dr.
Munro replicated her original calculations using the new
data, and found what she describes as “more significant
results.” Id. at 2-4. While Dr. Munro's
initial analysis concluded that the RIF's “African
American termination rate was 277% the rate for non-African
Americans, ” her updated analysis concluded that
“the termination rate of African Americans was 444% the
rate of Caucasians, which is much, much more
notable.” Id. at 3 (emphasis added).
Similarly, Dr. Munro's initial analysis found that
employees 40 and over were terminated “at 176% the rate
of those under 40, ” but after considering the new
data, she found that employees 40 and over were terminated
“at 179% the rate of those under 40.”
Id. at 4. Dr. Munro concludes that the termination
rates for both African-Americans and employees 40 and over
violated the 4/5ths rule and that both results are
statistically significant. Id. at 3-4.
Munro also identified several purported methodological and
statistical issues with Dr. Bronars's previous report.
Id. at 5-7 (summarizing concerns). First, Dr. Munro
questions Dr. Bronars's assumption that different
positions faced unequal risks of termination in the RIF.
Id. at 5, 8-11. Dr. Munro contends that Dr. Bronars
bases this assumption on the relatively recent declaration of
Mr. Davidson, but that, “at the time of [her] analysis,
the most practical assumption to make was that aside
from the SSAs (that were 100% terminated), every other
position and employee . . . was at equal risk for
termination.” Id. at 5; see also Id.
Dr. Munro raises two statistical issues with Dr.
Bronars's analysis. Id. at 6, 12- 16. For one,
sample size is an important factor in finding a statistically
significant result, and by breaking the larger sample size
(all employees) into smaller subgroups (specific positions or
divisions), Dr. Bronars made it less likely that any
statistically significant result would be found. Id.
at 14. For another, the subgroups analyzed by Dr. Bronars
were very homogenous-at least one subgroup was entirely
African-American, for example-and Dr. Munro contends that
homogenous subsamples can make an effort to predict
differentials between expected and observed outcomes
“completely meaningless.” Id. Third, Dr.
Munro suggests that Dr. Bronars's “conclusions are
not dispositive of adverse impact.” Id. at 16.
According to Dr. Munro, the homogeneity of Dr. Bronars's
subsamples raises another possibility-“that the
‘targeted' positions and divisions were
disproportionately occupied by African Americans and those
over 40.” Id. Finally, Dr. Munro concludes her
report with a lengthy analysis of the financial and agency
efficiency implications of the RIF. Id. at 22-31.
Dr. Bronars's ...