UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA
RONDA L. DAVIS, et al., : : Plaintiffs, : Civil Action No.: 10-1564 (RC) : v. : Re Document Nos.: 169, 170 : DISTRICT OF COLUMBIA, : : Defendant. :
MEMORANDUM OPINION
DENYING DEFENDANT’S MOTION FOR SUMMARY JUDGMENT; DENYING PLAINTIFFS’ CROSS
MOTION FOR SUMMARY JUDGMENT
I. INTRODUCTION
In this putative class action suit, Plaintiffs, who are former employees of the District of
Columbia Child and Family Services Agency, allege that their terminations from the agency
during a large-scale reduction in force were unlawfully discriminatory on the basis of race. This
Court previously considered motions for summary judgment from the parties and found in favor
of Defendant District of Columbia (the “District”). See Davis v. District of Columbia, 246 F.
Supp. 3d 367 (D.D.C. 2017). Plaintiffs appealed. The D.C. Circuit largely affirmed this Court’s
decision but reversed and remanded with respect to Plaintiffs’ disparate impact claim. See Davis
v. District of Columbia, 925 F.3d 1240 (D.C. Cir. 2019). The Court of Appeals ruled that this
Court erred in holding that Plaintiffs had not identified a “particular employment practice”
susceptible to challenge for its adverse racial impact under Title VII. See 42 U.S.C. § 2000e-
2(k)(1)(A)(i). Finding that Plaintiffs did identify specific employment practices susceptible to
challenge (brought into “better focus” on appeal), the court remanded so that this Court could decide whether, given the identified employment practices, Plaintiffs have shown sufficient
statistical evidence to make out a prima facie case of disparate impact under Title VII.
The parties have fully briefed motions for summary judgment on this question. See
Def.’s Renewed Mot. Summ. J. (“Def.’s Mot.”), ECF No. 169; Pls.’ Mem. Supp. Cross Mot.
Summ. J. (“Pls.’ Mem.”), ECF No. 170; Def.’s Opp’n, ECF No. 171; Pls.’ Opp’n, ECF No. 172;
Def.’s Reply, ECF No. 174; Pls.’ Reply, ECF No. 175. For the reasons set forth below, having
considered the parties’ arguments in light of the D.C. Circuit’s opinion, the Court denies the
District’s motion for summary judgment. Because the Court bifurcated discovery, see
Scheduling Order, ECF No. 59, the parties will now have the opportunity for discovery to
address whether the reduction in force was justified by business necessity.
II. BACKGROUND 1
A. Factual Background
The District of Columbia Child and Family Services Agency (“CFSA” or the “Agency”)
exists “to ensure the safety, permanence, and well-being of abused and neglected children and to
strengthen troubled families in the District.” Def.’s Resp. to Pls.’ Statement of Undisputed
Material Facts (“Def.’s Resp. Material Facts”) ¶ 1, ECF No. 171-1. 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. Id. ¶ 3. District of Columbia law and the consent decree entered in the
1 The Court assumes familiarity with its two previous opinions and the D.C. Circuit’s opinion. See Davis v. District of Columbia, 246 F. Supp. 3d 367 (D.D.C. 2017); Davis v. District of Columbia, 949 F. Supp. 2d 1 (D.D.C. 2013); Davis v. District of Columbia, 925 F.3d 1240 (D.C. Cir. 2019). These opinions outline in detail the facts underlying this case. Nevertheless, the Court highlights the factual and procedural background relevant to the pending motions. Because the Court’s and the D.C. Circuit’s decisions have significantly narrowed the scope of this case, the Court will keep its review of the factual background brief.
2 class action LaShawn v. Bowser mandate the provision of many of these services. See Def.’s
Statement of Undisputed Material Facts (“Def.’s Material Facts”) ¶ 1; 2 see also LaShawn v.
Bowser, No. 89-1754 (D.D.C. Feb. 27, 2007), ECF No. 864 (order approving Amended
Implementation Plan).
CFSA experienced significant budgetary pressure in fiscal years 2010 and 2011. In
Fiscal Year (“FY”) 2010 (October 1, 2009 – September 30, 2010), CFSA’s local funds’ budget
was reduced by $25.3 million from the previous year. Def.’s Resp. Material Facts ¶ 8. The FY
2010 budget reduced the number of approved full-time employees and, as a result, CFSA
implemented personnel reductions to its information technology unit and public information
office. Id. ¶¶ 10–11. The D.C. Council further reduced the funds available to CFSA in FY 2011
by $12.1 million. Id. ¶ 11. To address the reduction in funding in FY 2011, CFSA used a
reduction in force (“RIF”) with an effective termination date of June 11, 2010. Id. ¶ 13. The
RIF is the subject of Plaintiffs’ lawsuit.
In the lead up to the termination date, the CFSA Director Roque Gerald sent a
memorandum to City Administrator Neil Albert seeking “approval to conduct a Reduction-In-
Force (RIF) to abolish one hundred and twenty-three (123) positions within the Child and Family
Services Agency.” Id. ¶ 16. The parties dispute whether Mr. Gerald’s memorandum provided
an “agency-wide” list of positions subject to elimination, but they agree that the memorandum
proposed elimination of positions across multiple offices in the agency, including in the Office of
Agency Programs. See id. ¶¶ 18–19. In the end, CFSA eliminated 123 positions in the RIF,
which translated to the separation of 115 employees from the agency. Id. ¶¶ 14–15. In
2 The District’s Statement of Undisputed Material Facts is attached to its motion for summary judgment at ECF No. 169.
3 implementing the RIF, the agency reviewed its programs and determined which functions would
have the least negative impact if eliminated. Id. ¶ 21. According to the District, “CFSA did not
utilize a single uniform criteria, test or requirement for determining which employees would be
separated from the Agency in the RIF.” Def.’s Statement of Undisputed Material Facts ¶ 15,
ECF No. 146-2. 3 Instead, positions were selected for elimination after “multiple individual
decisions made by the Director working in close consultation with the Chief of Staff, the Deputy
Directors in charge of CFSA’s various divisions, and other senior level managers in the
Agency’s executive team.” Id. The District points to no objective test that CFSA used to
determine which positions would be eliminated.
Specifically, and of particular importance here, the elimination of two types of positions
in their entirety—the Social Services Assistant (“SSA”) and the Social Worker Associate
(“SWA”)—accounted for the majority of the 115 employees terminated. See Def.’s Resp.
Material Facts ¶ 22; Def.’s Material Facts ¶ 13. The District explains that the elimination of
these positions resulted from the agency’s conversion of “its workforce to the ‘team model,’
which grouped social workers with a set of skilled partners to serve client needs together.”
Def.’s Material Facts ¶ 12. According to the District, the SSA and SWA positions were “no
longer needed under the new model.” Id. ¶ 13. Though the exact percentages are disputed,
Plaintiffs object to the elimination of these positions because, according to their calculations, 98
percent of the eliminated SSAs and eleven of the thirteen eliminated SWAs were African-
American. Def.’s Resp. Material Facts ¶¶ 23–24. 4
3 The statement of undisputed material facts cited here was attached to the District’s original motion for summary judgment. 4 Plaintiffs previously pursued claims stemming from the creation of a new position that replaced these two eliminated positions. See Davis, 246 F. Supp. 3d at 374–77. The Court granted summary judgment in favor of the District on these claims, see id. at 399, and the D.C.
4 Plaintiffs hired Dr. Paige Munro, who has a Ph.D. in industrial organizational
psychology, to perform a statistical analysis of CFSA’s employment actions. See Mot. Summ. J.
Ex. H (“Munro Decl.”) ¶ 4, ECF No. 146-3. 5 In July 2012, Dr. Munro submitted her initial
expert report based on data provided by the declaration of Stan Spaght, the Human Resources
Manager for Compensation/Benefits at CFSA. Id. ¶ 2; Mot. Summ. J. Ex. I (“Munro Report”) at
1, ECF No. 146-3. Based on demographic information provided by Mr. Spaght, Dr. Munro
calculated that, of the 832 workers at CFSA before the RIF, approximately 689 were African-
American and 143 were other races. Munro Report at 1. She then calculated that 107 employees
terminated in the RIF were African-American, while only eight were other races. Id. In her
initial report, Dr. Munro ultimately 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.
She found that these results were statistically significant and violated the Equal Employment
Opportunity Commission’s 4/5ths, or 80 percent, rule. 6 Id.
Circuit affirmed, see Davis, 925 F.3d at 1254 (“The [district] court correctly concluded that plaintiffs failed to raise a triable issue” with respect to the degree requirement of the newly created position). As such, these claims do not remain pending and the Court will not discuss the facts underlying the claims further. 5 In their renewed motion for summary judgment, Plaintiffs cite back to exhibits included in the original motions for summary judgment. The Court follows suit. 6 The 4/5ths rule can be summarized as follows: “[T]he failure of a sex, race or ethnic group to have a success rate which is at least 80% of the rate of the most successful group is considered evidence of adverse impact.” E.E.O.C. v. Warchawsky & Co., 768 F. Supp. 647, 655 (N.D. Ill. 1991) (quoting United States v. Chicago, 663 F.2d 1354, 1357 n.8 (7th Cir. 1981)). This method has not been accepted by the D.C. Circuit and has been treated with caution by courts in this District. See Frazier v. Consolidated Rail Corp., 851 F.2d 1447, 1451 (D.C. Cir. 1988) (affirming district court’s rejection of four fifths rule and noting that “[t]he four fifths rule has also been used with caution by some courts”); Delgado v. Ashcroft, No. 99-cv-2311, 2003 WL 24051558, at *8–9 (D.D.C. May 29, 2003) (rejecting application of four fifths rule); Reynolds v. Sheet Metal Workers Local 102, 498 F. Supp. 952, 966 (D.D.C. 1980) (“[T]he ‘four- fifths rule’ has recently been criticized as a rather arbitrary standard, failing to account for differences in sampling size and tests results in the applicant population taken as a whole.”).
5 The District hired Dr. Stephen G. Bronars, who holds a Ph.D. in economics, to serve as a
defense expert. See Mot. Summ. J. Ex. J (“Bronars Report”), ECF No. 146-3. 7 Dr. Bronars
contended that Dr. Munro’s analysis was “based on the assumption that all CFSA employees
faced the same risk of termination during the RIF.” Id. ¶ 11. In contrast to this assumption, and
based on the declaration of CFSA’s Chief Administrative Officer at the time, Raymond
Davidson, see Mot. Summ. J. Ex. E (“Davidson Decl.”), ECF No. 146-3, Dr. Bronars’s analysis
“assume[d] that there [we]re 7 different sets of layoff decisions made by CFSA during the RIF”
corresponding to 7 different positions or divisions affected by the layoffs. Id. ¶ 21. Dr. Bronars
also began his analysis with a different total number of CFSA employees at the time of the RIF,
based on a declaration from the Director of Human Resources that stated thirty employees in the
office of Fiscal Operations Administration were not subject to the RIF. See Mot. Summ. J. Ex.
C, ECF No. 146-3. Analyzing the various divisions of CFSA separately, Dr. Bronars compared
the actual results of the RIF to the expected outcomes from a random neutral process and
determined 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.” Bronars Report ¶
22. He determined that this result was not statistically significant. Id. ¶¶ 22–23.
Dr. Munro submitted a rebuttal report in May 2014. See Mot. Summ. J. Ex. K, ECF No.
146-4. She first replicated her original calculations using the updated total number of employees
and concluded that the termination rate for African Americans was 444% the rate of Caucasians.
7 The District claims in its briefing that Dr. Bronars’s report is irrelevant to the present question of whether Plaintiffs have made out a prima facie case of disparate impact. See Def.’s Opp’n at 10 n.5 (“The District . . . does not rely on Dr. Bronars in support of its motion for summary judgment . . . .”). However, because Plaintiffs compare the statistical analysis of Dr. Munro to that of Dr. Bronars, see Pls.’ Mem. at 15–16, the Court briefly describes his findings.
6 Id. at 3. She defended her assumption that all positions were at risk of termination, stating that
“at the time of [her] original 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. at 8–12. She stated that she “would always choose
agency-wide adverse impact analysis unless there was uncontroverted evidence that certain
positions were in fact the only positions at risk.” Id. at 12. Finally, she explained that analyzing
the RIF on the subgroup level, where some subgroups are very racially homogenous, makes
predicting differentials between expected and observed outcomes “completely meaningless.” Id.
at 14. And by breaking the larger sample size (all employees) into smaller subgroups (specific
positions or divisions), Dr. Munro argued that Dr. Bronars made it less likely that any
statistically significant result would be found. Id. at 14.
B. Procedural History
Plaintiffs filed their initial complaint on September 16, 2010. See Compl., ECF No. 1.
The current operative complaint is Plaintiffs’ Third Amended Complaint, filed on May 31, 2013.
See 3d Am. Compl., ECF No. 66. The Third Amended Complaint brings claims of age
discrimination pursuant to the District of Columbia Human Rights Act (“DCHRA”) and race
discrimination pursuant to Title VII and the DCHRA. See id. ¶¶ 78–105. After consideration of
two rounds of dispositive motions, Plaintiffs’ racial discrimination claim based on a theory of
disparate impact caused by the manner in which the RIF was conducted is the sole remaining
claim.
Most relevant to the present motions is this Court’s opinion granting summary judgment
to the District. With respect to Plaintiffs’ disparate impact claim, the Court determined that
Plaintiffs had “failed to identify a specific employment practice” susceptible to challenge under
7 Title VII. Davis, 246 F. Supp. 3d at 394. The Court found that “simply pointing to a RIF
generally is not sufficient” to support a disparate impact claim. Id. at 395 (citing Leichihman v.
Pickwick Int’l, 814 F.2d 1263, 1269 n.5 (8th Cir. 1987)). The Court noted that “Dr. Munro
analyzed the agency-wide termination rates for African Americans” and that she “analyzed the
RIF as if every employee stood an equal chance of termination.” Id. at 396 (emphasis in
original). Because Plaintiffs did not more specifically identify the employment practice for the
observed statistical disparities, the Court granted summary judgment for the District. Id. at 397.
The D.C. Circuit reversed this Court’s decision on this issue. Davis, 925 F.3d at 1254.
The court explained that “[d]isparate impact analysis is ‘no less applicable to subjective
employment criteria than to objective or standardized tests.’” Id. at 1249 (quoting Watson v.
Fort Worth Bank and Trust, 487 U.S. 977, 990 (1988)). For this reason, “[a]n employer’s
undisciplined system of subjective decisionmaking [can have] precisely the same effects as a
system pervaded by impermissible intentional discrimination.” Id. (quoting Wal-Mart Stores,
Inc. v. Dukes, 564 U.S. 338, 355 (2011)). The court stated that “[t]here is no mystery in this case
as to the layoff practices plaintiffs challenge: the Agency’s choices to (a) target the SWA and
SSA job categories for elimination; and (b) allow managers to make putatively individualized,
discretionary and subjective choices of which positions to winnow from other units.” Id. at
1250. In other words, “[w]hat is at issue here is not a RIF in the abstract . . . but the means by
which the Agency implemented it.” Id. at 1243. Noting that the District explained that CFSA
did not use uniform criteria to select positions for termination, the court stated “[a]s the Agency
itself describes it, the procedures for culling jobs fit Watson’s description of ‘an employer’s
undisciplined system of subjective decisionmaking’ as to which ‘it is difficult to see why Title
VII’s proscription against discriminatory actions should not apply.’” Id. at 1250 (quoting
8 Watson, 487 U.S. at 990–91). Accordingly, the court remanded to determine whether “plaintiffs
clear the statistical hurdle” of making out a prima facie case in light of the identified
employment practices. Id. at 1253. The parties’ motions for summary judgment currently before
the Court speak to this issue.
III. LEGAL STANDARD
Under Rule 56 of the Federal Rules of Civil Procedure, a court must 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(a). A “material” fact is
one capable of affecting the substantive outcome of the litigation. See Anderson v. Liberty
Lobby, Inc., 477 U.S. 242, 248 (1986). A dispute is “genuine” if there is enough evidence for a
reasonable jury to return a verdict for the non-movant. See Scott v. Harris, 550 U.S. 372, 380
(2007). The inquiry under Rule 56 is essentially “whether the evidence presents a sufficient
disagreement to require submission to a jury or whether it is so one-sided that one party must
prevail as a matter of law.” Anderson, 477 U.S. at 251–52.
The principal purpose of summary judgment is to determine whether there is a genuine
need for trial by disposing of factually unsupported claims or defenses. See Celotex Corp. v.
Catrett, 477 U.S. 317, 323–24 (1986). The movant bears the initial burden of identifying
portions of the record that demonstrate the absence of any genuine issue of material fact. See
Fed. R. Civ. P. 56(c)(1); Celotex, 477 U.S. at 323. In response, the non-movant must point to
specific facts in the record that reveal a genuine issue that is suitable for trial. See Fed. R. Civ. P.
56(c)(1); Celotex, 477 U.S. at 324. “[T]he non-movant ‘may not rest upon mere allegation or
denials . . .’ but must [instead] present ‘affirmative evidence.’” See Laningham v. U.S. Navy,
813 F.2d 1236, 1241 (D.C. Cir. 1987) (quoting Anderson, 477 U.S. at 256–57). In considering a
9 motion for summary judgment, a court must “eschew making credibility determinations or
weighing the evidence.” Czekalski v. Peters, 475 F.3d 360, 363 (D.C. Cir. 2007). Conclusory
assertions offered without any evidentiary support do not establish a genuine issue for trial. See
Greene v. Dalton, 164 F.3d 671, 675 (D.C. Cir. 1999).
IV. ANALYSIS
Title VII of the Civil Rights Act makes it unlawful for an employer “to fail or refuse to
hire or to discharge any individual . . . because of such individual’s race.” 42 U.S.C. § 2000e-
2(a)(1). A plaintiff bringing a claim under Title VII may establish racial discrimination “by
proving either that the employer acted with a discriminatory motive (a ‘disparate treatment’
claim), or that its action was the result of a process that, while apparently ‘fair in form,’ was
‘discriminatory in operation’ (a ‘disparate impact’ claim).” Davis, 925 F.3d at 1248 (quoting
Griggs v. Duke Power Co., 401 U.S. 424, 431 (1971)). As the D.C. Circuit acknowledged, the
DCHRA “tracks Title VII in all respects relevant to this case” and, therefore, those claims are
appropriately treated “as coextensive with the[] federal claims.” Id.
Plaintiffs alleging a disparate impact claim must “isolat[e] and identify[] the specific
employment practices that are allegedly responsible for any observed statistical disparities.”
Wards Cove Packing Co., Inc. v. Atonio, 490 U.S. 642, 656 (1989) (quoting Watson, 487 U.S. at
994). Forcing a plaintiff to identify the specific employment practice at issue ensures that “the
myriad of innocent causes that may lead to statistical imbalances” do not create liability against
employers that have no control over those imbalances. Wards Cove, 490 U.S. at 657. In this
way, Title VII requires evidence that “goes beyond . . . show[ing] that there are statistical
disparities in the employer’s work force.” Watson, 487 U.S. at 994; Wal-mart Stores, Inc. v.
10 Dukes, 564 U.S. 338, 357 (2011) (pointing to “overall sex-based disparity” in workforce is not
enough).
Once a plaintiff identifies a specific employment practice, Title VII requires that the
plaintiff show “that each particular challenged employment practice causes a disparate impact.”
42 U.S.C. § 2000e-2(k)(1)(B)(i). However, “if the complaining party can demonstrate to the
court that the elements of a respondent's decisionmaking process are not capable of separation
for analysis, the decisionmaking process may be analyzed as one employment practice.” Id.
This exception to the requirement that a plaintiff offer proof of causation of each particular
employment practice “authorizes employees to demonstrate causation by showing the
discriminatory effect of an overall [] process, rather than the discriminatory effects of specific,
isolated practices playing a role in the process.” Onyewuchi v. Mayorkas, 766 F. Supp. 2d 115,
131 (D.D.C. 2011). Courts have found practices incapable of separation for analysis where the
decisionmaking process is entirely subjective. See McReynolds v. Sodexho Marriott Servs., Inc.,
349 F. Supp. 2d 1, 27–29 (D.D.C. 2004). The Supreme Court has explained that a “robust
causality requirement . . . protects defendants from being held liable for racial disparities they did
not create.” Texas Dep’t of Hous. & Cmty. Affairs v. Inclusive Communities Project, Inc., 576
U.S. 519, 542 (2015).
If a plaintiff offers proof of causation, the burden then shifts to the employer to
“demonstrate that the challenged practice is job related for the position in question and consistent
with business necessity.” 42 U.S.C. § 2000e-2(k)(1)(A)(i). And if the employer can present
evidence of business necessity, a plaintiff may rebut by showing that there is a less-
discriminatory “alternative employment practice” that would serve the employer’s legitimate
business interests. Id. § 2000e-2(k)(1)(A)(ii).
11 The Court understands the District to raise three primary arguments to support the
position that Plaintiffs have failed to establish a prima facie case of disparate impact. First, the
District argues that Plaintiffs’ statistical analysis does not demonstrate that each of the two
specific employment practices they challenge had a disparate impact on African Americans. For
this reason, the District claims the analysis is not relevant. Second, and relatedly, the District
argues that Plaintiffs have not shown that the identified employment practices are incapable of
separation for analysis, rendering their agency-wide analysis improper. Third, the District
attacks Dr. Munro’s analysis because of her alleged failure to consider important variables. The
Court considers each argument below. After considering the District’s objections, the Court
considers whether Plaintiffs have offered sufficient statistical evidence to make out a prima facie
case. 8
A. Relevance of Plaintiffs’ Statistical Analysis
The District argues that Plaintiffs cannot establish that either of the two identified
employment practices had a disparate impact on African Americans. Def.’s Mot. at 13–22. The
District claims that Dr. Munro’s analysis amounts to “unreliable bottom-line reasoning” because
8 Plaintiffs also argued in their opening brief that Defendants cannot demonstrate a business necessity sufficient to overcome their presentation of disparate impact and that, as such, they are entitled to summary judgment en toto. See Pls.’ Mot. at 26–37. They apparently abandon this argument in subsequent briefing, recognizing that such issues will be the subject of the second phase of the case. See Pls.’ Opp’n at 7. To the extent Plaintiffs continue to seek summary judgment with respect to the entire case, that relief is denied as premature. To the extent Plaintiffs seek summary judgment with respect to the prima facie element of their claim, that relief is also denied. See Carson v. Sim, No. 04-cv-1641, 2009 WL 3151335, at *1 (D.D.C. Sept. 24, 2009) (“[A] party may not file a motion for partial summary judgment on a fact or an element of a claim.” (citations omitted)); Franklin-Mason v. Penn, 259 F.R.D. 9, 11 (D.D.C. 2009) (denying summary judgment where movant sought “judgment on less than the entire surviving claim”); Beard v. District of Columbia Hous. Auth., 584 F. Supp. 2d 139, 140 n.1 (D.D.C. 2008); LaPrade v. Abramson, No. 97-cv-10, 2006 WL 3469532, at *8 (D.D.C. Nov. 29, 2006) (“Rule 56 does not contemplate a motion for partial summary judgment . . . [to] be entered as to a fact or an element of a claim.”).
12 it started with the assumption that all jobs were equally at risk for termination and because she
only considered two variables: race and retention status. Id. at 16. The District maintains that
Dr. Munro should have conducted a separate analysis for each identified employment practice.
Id. at 18. The District argues that Dr. Munro could have conducted a more nuanced statistical
analysis, utilizing advanced modeling and multiple regression to adjust for additional variables,
but she did not. Id. As such, it is the District’s position that “the statistical analysis [Plaintiffs]
offer is irrelevant to whether either of the employment practices they identify had a disparate
impact on African Americans. Id. at 2.
Plaintiffs argue that “Dr. Munro’s reports conclusively establish[] that the RIF resulted in
a substantial disparate impact on African-American employees of CFSA.” Pls.’ Mot. at 14.
They highlight that Dr. Munro’s analysis showed that “African Americans comprised 82.8% of
the total workforce prior to the RIF, yet they comprised 93% of the 115 workers terminated in
the RIF.” Id. They explain that “[c]omparing the 15.5% termination rate for African-Americans
with the 5.6% termination rate for other races, the chi-square analysis showed a p-value of 0.001,
which means the likelihood that the disparity was a result of chance is only one in one thousand.”
Id. at 15. Plaintiffs say that Dr. Munro’s analysis easily clears the initial statistical hurdle to
show a prima facie case of disparate impact because the statistical disparity is too great to have
occurred by chance, is based on a large sample size, and meets the EEOC’s 80% rule. See id. at
16–22.
In response to the District’s arguments, Plaintiffs say that “the Circuit Court explicitly
rejected [the District’s] attempt to analyze the implementation of the RIF as multiple separate
employment practices.” Pls.’ Opp’n at 2. Plaintiffs argue that, given the D.C. Circuit’s opinion,
“[a]t this stage, [they] must simply present statistically valid evidence showing that ‘the practices
13 of the Agency in selecting for elimination jobs and job categories’ in executing the RIF
disproportionately impacted African-American employees.” Id. at 17 (citing Davis, 925 F.3d at
1243). According to Plaintiffs, “[t]he range of actions which must be reflected in the statistical
inquiry is the entire re-organization scheme the Agency employed . . . including each decision as
to which job categories and specific positions to eliminate or reduce in size.” Id. For this
reason, Plaintiffs argue that the specific processes CFSA “‘used in order to cut positions to meet
its budget shortfall’ must be considered holistically and relative to each other.” Id. at 18 (quoting
Davis, 925 F.3d at 1252).
In light of the D.C. Circuit’s opinion, Plaintiffs have the better argument; the District’s
position is untenable given the language of the opinion. The court pointed to Dr. Munro’s
analysis as demonstrating “the clear racial effect of the Agency’s method of implementing its
reduction in force,” stating that her statistics showed a “dramatic over-representation of African
American employees in the positions chosen for elimination.” Davis, 925 F.3d at 1251. The
court specifically highlighted that the targeted positions in the RIF “were disproportionately
occupied by African Americans” and that, based on Dr. Munro’s report, “the termination rate
was 444% higher for the African American employees than Caucasians.” Id. Moreover, the
court stated that “challenging the processes by which [the RIF] was implemented . . . does not
render irrelevant the agency-wide statistics.” Id. at 1253. The court explained that “the agency-
wide statistics speak to the threshold issue [of] [w]hether, in shrinking an agency of 892 people
by at least 52 employees, the identified mechanisms by which defendants did so had a
statistically significant racial impact.” Id. The court also commented on Dr. Bronars’s analysis,
stating that he “presumed operational justification for (and therefore excluded from analysis) the
Agency’s selection of certain offices for downsizing and its wholesale elimination of the SSA
14 and SWA jobs.” 9 Id. But this premise “put the ‘cart before the horse’ by assuming the very
facts that a successful statistical showing by plaintiffs would next require the Agency to show:
that the reason the Agency targeted certain positions for elimination was justified by business
necessity.” Id.
The court’s analysis makes clear that the agency-wide statistics presented by Dr. Munro
are relevant to both challenged employment practices. Both challenged employment practices
represent “the process the Agency used to select positions for the chopping block.” Id. at 1252.
The Agency selected the SSA and SWA positions for elimination. See Def.’s Resp. Material
Facts ¶ 22; Def.’s Material Facts ¶ 13. The only meaningful way the impact of this decision can
be measured is relative to the positions that were not eliminated. See Davis, 925 F.3d at 1250
(citing cases analyzing RIFs that targeted demographically disproportionate departments for
layoffs). Likewise, the only way to make sense of the subjective decisionmaking behind the
elimination of other positions—the second identified employment practice—is to compare the
eliminated positions to the positions that survived the RIF. The selection of positions for
elimination “is what plaintiffs here identify, and is the kind of practice the disparate impact
theory of discrimination exists to scrutinize.” Id. at 1250–51. As such, Dr. Munro’s analysis
represents probative evidence related to both challenged employment practices. That it to say
that if either identified employment practice were challenged by itself, the relevant statistical
inquiry would focus on the agency-wide statistics. This conclusion is fortified by the cases cited
9 The District appears to suggest that, in order to more accurately measure the impact of the procedures CFSA utilized to implement the RIF, Dr. Munro should have removed the SSA and SWA positions from her analysis because both positions were entirely eliminated. See Def.’s Mot. at 17. This is exactly the type of analysis the D.C. Circuit indicated was more appropriate at the next stage of the litigation. Once discovery is had and the facts are established as to why and how the RIF was conducted, the Court will expect more sophisticated analysis that more closely analyzes how the RIF was actually conducted.
15 approvingly by the D.C. Circuit. See Shollenbarger v. Planes Moving & Storage, 297 Fed.
App’x 483, 486 (6th Cir. 2008) (concluding the plaintiffs made out a prima facie case based on
“a rudimentary statistical analysis” that considered elimination of departments relative to all
departments); Council 31, Am. Fed’n of State, Cty. Mun. Emps. v. Ward, 978 F.2d 373, 379 n.3
(7th Cir. 1992) (“The important questions would seem to be: (1) what percentage of employees
laid off in a random selection would be black; and (2) what are the confidence intervals
surrounding that outcome?”); Sengupta v. Morrison-Knudsen Co., Inc., 804 F.2d 1072, 1077 (9th
Cir. 1986 (finding that the plaintiff’s evidence of disparate impact on the department level was
insufficient because the relevant statistical pool was the entire business unit subject to layoffs).
Whether the two employment practices are considered together or separately, the relevant
statistical inquiry remains the same: an analysis of eliminated positions relative to preserved
positions. Accordingly, even if Plaintiffs failed to demonstrate that the two practices are not
capable of separation for analysis (the Court concludes below Plaintiffs do so demonstrate), the
Court finds that Dr. Munro’s reports provide probative statistical evidence of the racially
disparate impact of the two identified employment practices. 10
B. Capability of Separation for Analysis
The District argues that Plaintiffs have failed to show that the two identified employment
practices are not capable of separation for analysis, rendering the agency-wide statistics, as
opposed to a more targeted analysis of each practice, insufficient. Def.’s Mot. at 20–22. The
District notes it is Plaintiffs’ burden to show that “CFSA’s decision to eliminate two entire job
categories is incapable of being separately analyzed from CFSA’s other decisions regarding
10 As noted above, the District offers no contrary statistical analysis in support of the current motion for summary judgment. See Def.’s Opp’n at 10 n.5 (“The District . . . does not rely on Dr. Bronars in support of its motion for summary judgment . . . .”).
16 which positions to abolish.” Id. at 20. According to the District, Plaintiffs have proffered no
evidence to meet this burden. Id. at 21. The District maintains that the practices are capable of
separation for analysis because the SSA and SWA positions “were abolished to make room for
the new ‘teaming’ model implemented in Agency Programs” and that the record includes “the
data the employees would have needed to separately analyze the effect of this discrete decision.”
Id. at 21–22. The District also notes that Dr. Munro herself stated that the fifty-seven terminated
SSA positions would be an adequate sample size to conduct an analysis. Id. at 22.
Plaintiffs argue that “it has been [their] position throughout the entire case that the
[CFSA’s] entire process is incapable of separation for analysis” and that they “have not
claimed—and this court has never found—that there were discrete, identifiable steps to CFSA’s
overall process that has distinguishable isolated effects.” Pls.’ Opp’n at 12. Because the Circuit
emphasized that “it is the processes the Defendants’ used to select positions to eliminate that is
being challenged here . . . the decision to abolish the SSA and SWA job categories, as well as
other ‘individualized, discretionary and subjective’ choices made by individual managers are all
subject to review here.” Id. at 13. For Plaintiffs, “these two actions function as one inseparable
and discretionary decisionmaking process, where the individual effects of each action cannot be
identified or quantified in any meaningful way for statistical analysis.” Id.
The Plaintiffs again have the better argument (at least at this stage of the litigation). The
D.C. Circuit stated that “[w]hat calls for identification and scrutiny, and what plaintiffs challenge
here, is . . . the process the Agency used to select positions for the chopping block.” Davis, 925
F.3d at 1252. For purposes of the current motion, that process involves both challenged
17 employment practices. Moreover, the Court does not now assess the bases for selecting those
positions; that will come at the next stage of the litigation. 11
C. Consideration of Other Variables
Finally, the District argues that Dr. Munro failed to consider a number of variables. First,
the District maintains that Dr. Munro should have specifically accounted for CFSA’s adoption of
the “new teaming model” that resulted in the elimination of the SSA and SWA positions. See
Def.’s Mot. at 9, 21–22. Second, the District claims that Dr. Munro should have accounted for
the “nondiscretionary” strictures imposed by CFSA’s compliance plan in LaShawn v. Bowser.
Id. at 23. According to the District, CFSA “could not abolish any of its case-carrying social
worker positions” because of the compliance plan—this, the District argues, should have been
factored into Dr. Munro’s analysis. Id. Finally, the District argues that Dr. Munro should have
considered the “strictly regulated process” used to determine retention. Id. at 24. Because
employees who held abolished positions could compete to remain employed based on a number
of factors, the District argues that CFSA “did not necessarily know which employees would lose
their jobs.” Id. As such, “[a] higher concentration of African Americans among the less-senior
employees could lead to a higher percentage of African Americans among the separated
employees, without raising any inference of a disparate impact based on race.” Id. Plaintiffs
counter that none of these variables undermine their statistical showing because they “did not
limit the Agency’s range of available options in implementing the RIF.” Pls.’ Opp’n at 21.
11 At this stage, the agency-wide statistics offered by Dr. Munro speak to the process identified by the D.C. Circuit. However, a more sophisticated and distinct analysis may be required at the next stage of the litigation, where the District will be permitted to present evidence of the business justifications for selecting the SWA and SSA positions for elimination, as well as other eliminated positions, and more fully explain how positions were selected in light of the agency’s obligations under LaShawn v. Bowser and District retention policies.
18 The Court finds that the omission of these variables from Dr. Munro’s analysis does not
render it unreliable or irrelevant for the purpose of establishing a prima facie case at this stage of
the litigation. The CFSA’s adoption of a “team model” might be a reason the District can offer
as a justification for its actions, but it does not change whether or not the challenged employment
actions resulted in a racial disparity. See Shollenbarger, 297 Fed. App’x at 486 (“At this step in
the analysis—the prima facie step—[the defendant’s] reasons for selecting certain departments
[for elimination] is immaterial . . . .”). For similar reason, the requirements outlined in the
LaShawn compliance plan do not affect Plaintiffs’ prima facie statistical showing. The LaShawn
compliance plan does not mandate CFSA’s adoption of the “team model.” It does not mandate
the elimination of the SSA and SWA positions. It does not mandate a particular decisionmaking
methodology for a RIF. See LaShawn v. Bowser, No. 89-1754 (D.D.C. Feb. 27, 2007), ECF No.
863-1 (Amended Implementation Plan). While compliance with LaShawn will likely play a role
in the next stage of this case, 12 this variable did not limit CFSA’s discretion such that it had to be
accounted for by Dr. Munro at this stage. See Carpenter v. Boeing Co., 456 F.3d 1183, 1198–
202 (10th Cir. 2006) (explaining that failure to account for nondiscretionary overtime eligibility
variables rendered analysis insufficient to establish prima facie case).
The District’s argument that Dr. Munro’s analysis failed to account for retention
processes also misses the mark. The District explains that “[i]f the abolished positions are
occupied, employees are selected for separation through a strictly regulated process of lateral
competition” and that employees compete with other employees with similar salary grades and
12 The LawShawn compliance plan may have, as a practical matter, limited the agency’s discretion to cut certain positions, thus, calling into question Dr. Munro’s assumption that every position had an equal chance of being cut in the RIF. But that question will be explored at the next stage.
19 titles. Def.’s Mot. at 4. But whether employees selected for termination in the RIF had an
opportunity to compete for retention in some other position does not change the fact that
“[t]erminating a large group of employees in a compressed timeframe is clearly an adverse
employment action within the meaning of Title VII.” Davis, 925 F.3d at 1252. In Carpenter, the
variables not accounted for objectively limited which employees would be eligible for overtime.
See 456 F.3d at 1198–202. Here, while the retention processes may have changed the ultimate
outcome for particular individuals, that all came after implementation of the RIF. The retention
processes did not objectively limit which positions CFSA selected for elimination—and because
“Plaintiffs challenge the practices of the Agency in selecting for elimination jobs and job
categories disproportionately held by African American employees,” Davis, 925 F.3d at 1243,
the retention processes have no bearing at this stage. While the District argues it did not
ultimately know who would be separated because of the retention processes, that does not
change the underlying statistical finding that the positions selected were overwhelmingly held by
African Americans. For that reason, Carpenter is distinguishable, and Dr. Munro’s analysis is
not deficient for having not considered District retention processes at the prima facie stage.
The District wants Plaintiffs to account for the business pressures that led to their plan for
implementing the RIF—these arguments put the cart before the horse “by assuming the very
facts that a successful statistical showing by plaintiffs would next require the Agency to show:
that the reason the Agency targeted certain positions for elimination was justified by business
necessity.” Id. at 1253; see also Shollenbarger, 297 Fed. App’x at 486. At that point, a more
sophisticated statistical analysis may very well be required. Accordingly, the Court finds that, at
this stage of the litigation, the District’s arguments do not undermine the probative value of Dr.
Munro’s analysis for the purposes of establishing a prima facie case.
20 D. Sufficiency of Statistical Evidence
Having considered the District’s objections, the Court turns to Plaintiffs’ statistical
evidence of disparate impact. The Court concludes that Plaintiffs clear the statistical hurdle
required to make out a prima facie case. Dr. Munro calculated that, of the 832 workers at CFSA
before the RIF, approximately 689 were African-American and 143 were other races. Munro
Report at 1. She determined that 107 employees terminated in the RIF were African-American,
while only eight were other races. Id. In her initial report, Dr. Munro ultimately 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. She found that these results were statistically significant, with
a p-value of .001, and violated the Equal Employment Opportunity Commission’s 4/5ths, or 80
percent, rule. 13 Id.; see Ricci v. DeStefano, 557 U.S. 557, 587 (2009) (“[A] prima facie case of
disparate-impact liability [is] essentially a threshold showing of a significant statistical disparity .
. . .”) (internal punctuation and citations omitted). In her rebuttal report, see Mot. Summ. J. Ex.
K, she replicated her original calculations using the updated total number of employees and
concluded that the termination rate for African Americans was a staggering 444% the rate of
Caucasians. Id. at 3; cf. Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 307–08 (1977)
(“Where gross statistical disparities can be shown, they alone may in a proper case constitute
prima facie proof of a pattern or practice of discrimination.”). To make out a prima facie case
Plaintiffs “need only offer statistical evidence of a kind and degree sufficient to show the
employment decision disproportionately impacts [the protected class].” Aliotta v. Bair, 614 F.3d
13 As noted above, the D.C. Circuit has not endorsed this methodology and courts in this District have treated it with caution. See supra note 6; Frazier, 851 F.2d at 1451; Delgado, 2003 WL 24051558, at *8–9; Reynolds, 498 F. Supp. at 966. Although it is unclear whether this analysis is appropriate in this Circuit, the District did not challenge it. See Def.’s Opp’n at 10 (noting Dr. Munro’s application of the 80 percent rule but not challenging it).
21 556, 565 (D.C. Cir. 2010). Plaintiffs have cleared that initial bar. Dr. Munro’s analysis shows
that the processes by which CFSA implemented the RIF—the challenged employment practices
identified by Plaintiffs and the D.C. Circuit—disproportionately impacted African Americans.
Although discovery and further factual development may indicate that Dr. Munro’s analysis
bears no relation to how the RIF was actually conducted as a result of the legitimate business
necessities, that is a question for another day. Accordingly, the Court finds that Plaintiffs have
made out a prima facie case of disparate impact.
V. CONCLUSION
For the foregoing reasons, Defendant’s motion for summary judgment (ECF No. 169) is
DENIED and Plaintiffs’ cross motion for summary judgment (ECF No. 170) is DENIED. An
order consistent with this Memorandum Opinion is separately and contemporaneously issued.
Dated: October 19, 2020 RUDOLPH CONTRERAS United States District Judge