Willie S. Phillips Curtissene Anderson Todd Brooks, Yulander Edwards Emma Fields Richard Goudy Denise Goudy Kevin Gray Jackie McGrady Susan McNeal Carlisa Miles Linda R. Pettes Mary L. Phillips Lisa Prater Marcia Spence Jean Washington, and Lerdon Woodfolk v. William S. Cohen, Secretary of Defense

Citations: 400 F.3d 388; 2005 U.S. App. LEXIS 3917; 95 Fair Empl. Prac. Cas. (BNA) 520Docket: 03-4190

Court: Court of Appeals for the Sixth Circuit; March 8, 2005; Federal Appellate Court

SummaryOriginalCytator Case TreatmentBackward CitatorAnalysis

EnglishEspañolSimplified EnglishEspañol Fácil

This case involves a Title VII race discrimination claim by multiple plaintiffs against the Secretary of Defense, alleging that neutral promotion criteria at the Defense Finance and Accounting Service, Columbus Center (DFAS-CO) disproportionately affected African-American employees, violating 42 U.S.C. § 2000e-16. The plaintiffs appealed a magistrate judge's summary judgment in favor of the Secretary, which also found that five plaintiffs, including Curtissene Anderson and Richard Goudy, lacked standing. The court affirmed the standing determination for all but Denise Goudy, but disagreed with the summary judgment decision, citing sufficient evidence of disparate impact. Additionally, the court found the magistrate judge erred in handling the plaintiffs' motion for sanctions. Consequently, the decision was partially affirmed, partially reversed, and remanded for further proceedings. 

The DFAS-CO, established in 1991, grew rapidly and is responsible for paying contractor bills and providing accounting services within the Department of Defense. The promotion process involves decisions by department managers, collaboration with human resources to develop a recruiting plan, internal job postings, and a numerical ranking of applications based on various criteria, with oversight from the equal employment opportunity office to assess representation of underrepresented groups among finalists.

Underrepresentation is assessed by comparing the number of individuals from specific groups within the relevant grade level of a position. The hiring manager ranks finalists based on qualifications outlined in the recruiting plan and selects candidates to interview. After choosing a successful candidate, the manager must document the rationale using a "selection matrix" and must explain any decisions not to select candidates from underrepresented groups if they are in the finalist pool. The manager can choose to reject all finalists and seek external candidates. The recommendation is then submitted to human resources and the agency director for approval. Plaintiffs argue that the promotion process is complex, comprising over eighteen steps and allowing significant subjective decision-making by the manager.

The DFAS-CO addressed concerns about the low hiring rates of African-American employees in its October 1996 newsletter, noting that black males represented only 4.2% of top management positions, lower than the 6.5% suggested by statistical data. The agency is exploring strategies like rotational assignments and cross-training to improve minority representation. Employee Willie Phillips criticized the agency's policies, asserting that the newsletter implied minorities required special assistance for promotions despite their qualifications.

On December 6, 1996, a Process Action Team (PAT) was formed to investigate these concerns, conducting interviews with African-American employees and analyzing EEO data from 1995 and 1996. The PAT Report confirmed that minorities were promoted at lower rates than non-minorities, particularly at higher grades, and highlighted a pattern of upper-level officials favoring candidates from their ethnic groups. The report indicated that minorities received only 23% of employee awards and faced 49% of disciplinary actions, alongside receiving fewer high performance ratings. Recommendations included ensuring equitable distribution of temporary promotions, enforcing a zero-tolerance policy for prohibited practices, reviewing award policies, clarifying selection processes, and improving employee awareness of complaint procedures regarding promotions.

Willie Phillips filed a discrimination complaint on July 30, 1997, with the DFAS-CO equal employment opportunity office on behalf of African-American employees. The complaint alleged multiple discriminatory practices affecting promotions, including: 1) temporary assignments of white employees to higher-grade positions giving them an advantage; 2) ethnic preference for white candidates by selection officials; 3) excessive discretion for officials regarding "specialized experience" in promotions; 4) manipulation of performance reviews and job announcements favoring white candidates; and 5) promotions granted to white employees who did not meet necessary time and grade requirements.

An EEO officer's report, included with the complaint, indicated that African-Americans received approximately 22% of promotions agency-wide in 1995 and 1996, reflecting their workforce representation. However, disparities were noted in awards and disciplinary actions, with 24% of African-American employees receiving awards compared to 34.9% of white employees. In terms of disciplinary actions, African-Americans received 48.9% of total actions, despite making up 4.5% of their workforce, compared to 1.33% for white employees.

On May 5, 1998, an administrative law judge recommended dismissal of the complaint due to untimeliness and inapplicability of the "continuing violation" doctrine. Subsequently, on August 7, 1998, Phillips filed a federal complaint against Secretary of Defense William S. Cohen, reiterating the discriminatory practices. The district court dismissed this complaint on July 30, 1999, for lack of timely administrative filing, but a panel later reversed this decision, applying equitable tolling under the "continuing violation" doctrine.

On remand, the district court appointed a magistrate judge for further proceedings, including trial, on May 2, 2002. On October 15, 2002, Phillips and other plaintiffs filed an amended complaint, adding Secretary of Defense Donald Rumsfeld as a defendant and withdrawing their class action status. They alleged discrimination under a disparate impact theory, highlighting that during the 1991-95 hiring expansion at DFAS-CO, many new hires were white women with no federal experience, while most African-American employees had transferred from other positions. A group of white supervisors allegedly held exclusive social events for white employees, leading to morale issues. A 1997 report indicated that promotion policies adversely affected African-Americans, suggesting that DFAS-CO engaged in practices favoring white employees in promotions, discipline, awards, and performance evaluations.

Both plaintiffs and the Secretary presented expert opinions regarding African-American representation and promotion rates at DFAS-CO. Plaintiffs' expert, Ramona Paetzold, found that African-Americans were under-represented in job grades and promoted at lower rates than whites. She based her analysis on census data and employee records from 1992-1996, concluding that African-American promotion rates (42.23%) were lower than those of white employees (45.87%), with a significant disparity for African-American women (42.34%) compared to white women (56.07%). However, Paetzold did not analyze the qualifications of the employees applying for promotions.

The Secretary's expert, John G. Claudy, also assessed representation and promotion rates, concluding that African-Americans were actually over-represented in every grade level at DFAS-CO compared to the expected figures based on the African-American working population with relevant skills.

Claudy and Paetzold approached the issue of promotion rates among African-Americans differently amid a lack of data on promotion applications. Claudy utilized "constructed pools of applicants" based on annual employee data snapshots, analyzing promotions over pairs of years and assessing promotions against pools of employees at the same pre-promotion grade and job series. His findings indicated that African-American employees experienced higher-than-expected promotion rates in half of the fourteen grade levels, while lower-than-expected rates were observed in the other half. Only one grade level (5) exhibited a significant downward departure from expected numbers. 

The Secretary moved for summary judgment, arguing that plaintiffs did not demonstrate a genuine issue of material fact regarding disparate treatment in promotions and that several plaintiffs lacked standing. The magistrate judge granted the motion on August 4, 2003, determining that four plaintiffs lacked standing due to insufficient evidence of applying for and being denied promotions. However, several remaining plaintiffs were found to have standing. The magistrate judge also concluded that the plaintiffs failed to establish a prima facie case of impact disparity, favoring Claudy's statistics over Paetzold's due to the latter's omission of merit factors, which undermined the reliability of her data. Claudy's finding of a significant disparity at only one grade level was deemed insufficient to prove disparate impact, and the PAT Report did not provide adequate evidence to shift the burden of proof to the Secretary.

Regarding document discovery, a protective order was granted to the Secretary pending a decision on summary judgment, requiring the production of relevant documents. Following the dismissal of the complaint for failure to exhaust administrative remedies, which was later reversed, it was revealed during discovery that DFAS-CO did not possess the documents used for the PAT Report or the promotion records requested. DFAS-CO claimed that the PAT documents were lost during interactions with a consulting firm, and that promotion information was destroyed after two years per agency regulations.

Plaintiffs filed a motion for sanctions on April 24, 2003, arguing that a 1998 protective order obligated DFAS-CO to locate and preserve discovery documents. The Secretary contended that this obligation was conditional upon a denial of summary judgment. The magistrate judge granted the plaintiffs' motion but postponed the determination of sanctions until trial. On appeal, the plaintiffs argued that the magistrate judge abused his discretion by not imposing sanctions immediately. 

The plaintiffs appealed the magistrate judge’s decisions regarding standing for specific plaintiffs and the grant of summary judgment to the Secretary, as well as the failure to impose sanctions at the time. The district court's grant of summary judgment is reviewed de novo, while the decision on sanctions is reviewed for abuse of discretion.

To establish standing under Title VII, a plaintiff must demonstrate personal injury from alleged discrimination and that the injury can be remedied through the requested relief. At the summary judgment stage, presenting some proof of injury is sufficient. The magistrate judge cited the requirement that a plaintiff must be qualified for the position sought to establish standing in disparate impact cases under Title VII. He found that four plaintiffs lacked standing due to not submitting affidavits, while one, Denise Goudy, was deemed not to have established standing due to insufficient specifics regarding her promotions. 

However, the court clarified that Denise Goudy met the standing requirements by stating she was denied promotions in favor of less qualified individuals. Consequently, the court reversed the magistrate judge’s ruling regarding Denise Goudy’s standing but upheld the ruling for the other four plaintiffs.

Title VII prohibits employment discrimination based on race, color, religion, sex, or national origin, including practices that limit or classify employees in a way that deprives individuals of employment opportunities. The statute applies to federal employers, including DFAS-CO, which is not disputed. Discrimination under Title VII can be proven through disparate treatment or disparate impact. Disparate impact addresses employment practices that appear neutral but disproportionately affect protected groups, ensuring that employers do not inadvertently eliminate qualified members of these groups.

In cases involving summary judgment, plaintiffs must identify the employment practices in question and demonstrate their disparate impact on the protected group. The 1991 amendments to Title VII altered the requirements for proving disparate impact, indicating that if a decision-making process cannot be separated for analysis, it can be considered as a singular practice. Under these amendments, once a plaintiff establishes a prima facie case, the burden shifts to the employer to demonstrate a business necessity for the practices causing the disparate impact.

The magistrate judge agreed to assess DFAS-CO's entire promotion decision process to determine disparate impact, a finding not contested on appeal. The plaintiffs' complaint identified five employment practices at DFAS-CO that allegedly had a discriminatory effect: detailing, promoting unqualified white employees, manipulating performance reviews to disadvantage African-Americans, vague selection criteria allowing bias, and an overwhelmingly white selection committee. These subjective practices can still be considered within the framework of disparate impact analysis, similar to previous cases involving favoritism. Conversely, practices exhibiting explicit bias must be addressed through disparate treatment claims.

The magistrate judge determined that it was unnecessary to evaluate whether the DFAS-CO's promotion practices could be separated for analysis since the plaintiffs did not demonstrate a disparate impact. Although plaintiffs claimed the promotion process was inseparable, they acknowledged it comprised eighteen distinct steps, a point the Secretary did not contest. Therefore, it is assumed that these practices cannot be separated.

To establish a disparate impact, plaintiffs are guided by Supreme Court precedents which suggest comparing the percentage of protected group members (e.g., African-Americans) in promotions to those in a qualified local labor force or applicant pool. An EEOC regulation indicates a discriminatory effect is suggested if the percentage of selected members from the protected group is 80% or less than that of the non-protected group. In promotion cases, the focus is on the ratio of protected group members receiving promotions to those seeking them.

The DFAS-CO's practice of destroying promotion data biennially hindered analysis. Two approaches were available: Claudy’s method, which created pools of similarly situated workers, and Paetzold’s global rate comparison. The magistrate judge found that plaintiffs failed to show a negative impact on African-Americans through Paetzold's analysis, citing issues of multiple promotions for individuals and a lack of qualification consideration.

The magistrate judge favored Claudy's "constructed pools" approach, despite acknowledging it was based on assumptions about employee qualifications. However, the judge's evaluation of expert credibility during a summary judgment motion was inappropriate, as such determinations should be reserved for a jury. The magistrate’s examination of the evidence was deemed erroneous or incomplete.

The Court critiques the magistrate judge's evaluation of expert opinions, noting two main flaws. First, skepticism towards the "constructed pools" methodology is warranted, as highlighted by the Seventh Circuit, which cautioned against the retrospective creation of qualifications for positions. The plaintiffs' claims that managers altered rank and grade requirements to favor white employees do not adequately consider scenarios where lower-graded white employees were compared with higher-graded African-American employees. 

Second, the Court observes that promotion cases are hesitant to apply the Wards Cove standard regarding the use of qualified individuals for disparate impact analysis. Relevant case law, such as Teal and Scales, indicates that disparate impact analysis should focus on the composition of candidates and those promoted, without needing to adjust for qualifications. The Court emphasizes that while both experts' analyses have flaws, these do not negate their findings. Paetzold’s report indicates that African-Americans may have been less frequently promoted, providing sufficient grounds to contest summary judgment.

The Court further stipulates that expert statistical evidence should not be viewed in isolation; it must be assessed alongside the entire body of evidence. The PAT Report offers additional insights into DFAS-CO's promotion practices, revealing that promotions to higher grade levels often occur through an appointment process that bypasses competitive applications, potentially obscuring the discriminatory impact on African-Americans' upward mobility.

Eighty-two percent of employees in a study indicated that they view career advancement opportunities unfavorably due to factors such as favoritism, nepotism, racism, and downsizing. Evidence suggests racial inequity in the distribution of awards, performance reviews, and disciplinary actions, which are critical for merit promotions, particularly disadvantaging African-Americans. Although plaintiffs could not conclusively demonstrate lower promotion rates compared to white employees, the findings from the DFAS-CO-commissioned report and the statistical data raise genuine issues of fact that render summary judgment inappropriate regarding disparate impact claims. Key evidence includes: (1) indications of less frequent promotions for African-American employees, (2) mandatory promotion factors that disadvantage these employees due to racial disparities, and (3) widespread perceptions of favoritism and lack of transparency in the promotions process.

In a related matter, the magistrate judge found the Secretary had violated a discovery order by failing to secure and produce responsive documents, notably the PAT Report and promotion records, which were lost or destroyed. The judge deemed the agency negligent but not grossly negligent or guilty of spoliation. Plaintiffs sought severe sanctions, including judgment in their favor or exclusion of the Secretary's expert report, but the magistrate considered these too extreme, opting to defer sanction decisions until trial. However, since summary judgment was ultimately granted to the Secretary, no sanctions were imposed. The court's discretion in sanction decisions involves assessing factors like willfulness, prejudice to the opposing party, notice of potential sanctions, and whether previous lesser sanctions were applied. The magistrate judge noted the inability to determine the prejudice caused by the lost documents without evaluating the case merits, leaving the sanctions issue unresolved by the time summary judgment was issued.

The magistrate judge determined that both parties were adversely affected by lost evidence that revealed deficiencies in their expert reports. Although the judge had previously ruled that the plaintiffs were entitled to sanctions, the issue of sanctions was not addressed following the summary judgment ruling, resulting in no sanctions being imposed. This omission constituted an abuse of discretion. The court affirmed in part and reversed in part the magistrate judge's decision, remanding for further proceedings. The plaintiffs' Title VII claim of race discrimination under a disparate impact theory continues, involving Willie S. Phillips and several others. 

The analysis included statistical disparities in promotions and disciplinary actions, with references to the EEOC's "four-fifths rule" which was met by the presented data. One plaintiff, Lori Dillard, had withdrawn from the case. The expert Paetzold noted limitations in the employment records available, as the Department of Defense destroys data every two years. The magistrate judge's opinion cited precedent that outlines the three elements of a prima facie disparate impact case: identification, disparate impact, and causation. In this context, the disparate impact and causation elements merge since the plaintiffs challenge the overall promotion process.

Causation regarding promotion rates for African-American employees versus white employees at DFAS-CO was addressed separately from disparate impact, despite both sections targeting the same issue. Plaintiffs noted that many African-American employees began in 1991 through transfers, while many white employees were newly hired during a rapid recruitment period. Evidence from DFAS-CO's EEO officer indicated that the promotion process, including awards distribution and disciplinary actions, negatively impacted African-American employees, who had a lower likelihood of receiving awards compared to their white counterparts and faced less disciplinary action. The dissenting opinion argued that the lower court's summary judgment in favor of DFAS-CO should stand, as plaintiffs failed to establish that African-Americans were promoted at a lower rate. Although some record information suggested discriminatory practices in performance reviews and discipline, these factors did not constitute barriers to promotion eligibility. The court has not determined if plaintiffs must prove that these factors lead to a lower overall promotion rate to substantiate their claims, but plaintiffs asserted that the entire promotional process should be considered as a single employment practice for disparate impact analysis, a point not contested by DFAS-CO. Even when considering plaintiffs' expert's findings, the evidence indicated that African-Americans were not promoted at significantly lower rates than white employees, with a slight difference in promotion rates during 1992-1996 being statistically insignificant.

The promotion rate for African-American employees significantly surpasses the promotion rate for white employees, which typically would not indicate a disparate impact under federal guidelines (29 C.F.R. 1607.4(D)). In cases where data is limited, the assessment of disparate impact should consider all available evidence (Bazemore v. Friday). Although a report indicated that DFAS-CO's subjective promotion practices fostered favoritism and were perceived as racist, this evidence does not sufficiently demonstrate an overall disparate impact due to the absence of supporting statistical data. The plaintiffs did not successfully prove that the promotions process negatively affected African-Americans, leading to the affirmation of summary judgment. Additionally, there is a disagreement regarding the magistrate judge's decision to delay sanctions against DFAS-CO for discovery violations until after the case's merits were heard. The magistrate, despite acknowledging the violation, acted within his discretion by not imposing immediate sanctions, especially since summary judgment was ultimately granted. Notably, disparities in promotion rates were highlighted, such as white women having a 56.07% promotion rate compared to African-American women’s 42.34% from 1992-1996.