Full opinion text
FINDINGS OF FACT AND CONCLUSIONS OF LAW BIFURCATED TRIAL: LIABILITY STAGE BUNTON, Chief Judge. BEFORE THIS COURT on August 15, 1988 came for trial the Plaintiff class consisting of 310 persons, present or one-time Special Agents of the Federal Bureau of Investigation, alleging that the FBI violated Title VII of the Civil Rights Act of 1964, 42 U.S.C. Sec. 2000e-2 discriminating against Hispanic agents on the basis of their national origin. On Motion of the parties, the litigation was bifurcated into a “liability” stage to determine whether the Defendant has violated Title VII and a “recovery” stage to determine the eligibility of individual class members for compensatory relief. DOES TITLE VII APPLY TO THE FBI? The Federal Bureau of Investigation is the arm of the United States Department of Justice that is responsible for investigating violations of Federal law. Within its purview are foreign counterintelligence, terrorism, organized crime, drug interdiction (together with the Drug Enforcement administration), commercial currency crimes, and abductions. Among the FBI employees are approximately 9400 Special Agents; all of whom are members of the “excepted service.” However, the Congress has made clear that employees of executive branch agencies are subject to the requirements of Title VII: “All personnel actions affecting employees or applicants for employment ... in executive agencies ... (including employees and applicants for employment who are paid from non-appropriated funds) ..., shall be made free from any discrimination based on race, color, religion, sex, or national origin.” 42 U.S.C. Sec. 2000e-16(a). Accordingly, the FBI is subject to the full reach of Title VII. Congress has demonstrated no intent to preclude judicial review of constitutional claims brought against the Bureau. Elsewhere within Title VII of the Civil Rights Act Congress gives more detail as to what actions may be considered among “all personnel actions” of Section 717. Section 703 of the Civil Rights Act of 1964, 42 U.S.C. Sec. 2000e-2, provides: “(a) It shall be an unlawful employment practice for an employer— “(1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin; or “(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.” With the exception of hiring practices, the Plaintiffs brought to this Court allegations which relate to every other element under the heading of unlawful employment practices; compensation, terms, conditions, or privileges of employment. In addition, Plaintiffs point to practices which limit, segregate, or classify employees which tend to deprive them of employment opportunities or otherwise adversely affect them. The Court exercised control over the discovery process attempting to balance the Plaintiff’s classmember’s need for access to proof in support of their claims against the significant needs of the FBI and Justice Department for confidentiality and the protection of its methods, sources, and mission. PLAINTIFF’S ALLEGATIONS The Members of the Plaintiff Class (hereafter, “Plaintiffs”) allege that practices of the FBI result in the selection of non-Hispanic agents for promotion, transfer, and other employment privileges in a manner which results in significant differences as compared to Hispanic agents. In particular, the plaintiff class alleges that impermissible considerations of ethnic heritage disfavor Hispanics for promotion, distribute temporary and permanent hardship assignments in a discriminatory manner, and apply disciplinary actions in unjustified proportions on Hispanic agents. Plaintiffs attribute the discriminatory effects to “disparate treatment” and “disparate impact” discrimination; that is, intentionally discriminatory acts and facially neutral policies and procedures of the Bureau which have adverse impact on members of a protected group. Anticipating the Defendant's defense, the Plaintiffs allege that the policies and practices of the Bureau are not justified by the business necessity defense. They further say that any proffered legitimate reason is but a pretext for impermissible considerations. The Plaintiffs point to a promotion system which contains excessive subjective elements without procedures to evaluate the decisions for compliance with EEO guidelines. Similarly, the Plaintiffs claim that other employment decisions such as transfers, temporary duty assignments and discipline are excessively subjective and thereby capricious. Plaintiffs also complain that members of the Bureau engage in “religious discrimination” against Hispanic agents for certain events not directly related to this action. In particular, the Plaintiffs allege that agents that are members of the Church of Latter Day Saints, colloquially termed “Mormons”, engage in religious discrimination. Further, at the time of trial, Plaintiffs complained of incidents of retaliation visited upon members of the class for their activities in support of the class. Named Plaintiff Bernardo Perez, as an individual, claims that he suffered discriminatory treatment on the basis of his religious affiliation and national origin. In addition, he alleges that he was subjected to retaliation by the FBI for bringing a grievance under the Bureau Equal Employment Opportunity process. DEFENDANT’S CLAIMS The Defendants deny the Plaintiff’s claims and state that all assignments, promotions, transfers and failures to promote members of the Plaintiff Class were justified by business necessity in accordance with the mission of the FBI as a law enforcement entity. Defendants deny that any job related action suffered by Plaintiffs related to this litigation constituted reprisal or retaliation; rather, that disciplinary actions taken against Plaintiffs were motivated entirely by job-related factors. Both Plaintiffs and Defendants agree that hiring practices of the FBI are not in issue in this action; that is, if the FBI violated Title VII of the civil rights act, its hiring practice would not be a direct legal cause of the violation. Thus, a person who claims that he was not hired by the FBI because of national origin discrimination could not be a member of the class in this suit. TWO PATHS CONVERGE The Supreme Court distinguished two kinds of discrimination for the purpose of analyzing the evidentiary standards which trial courts are to apply to determine whether employment practices result in prohibited distinctions which harm employees. In “disparate treatment” cases, the plaintiff must prove that the defendant had a discriminatory intent or motive. See Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 255, n. 8, 101 S.Ct. 1089, 1094-95, n. 8, 67 L.Ed.2d 207 (1981). The evidentiary burden shifts between the employee and the employer to determine whether, if discriminatory intent has been implicated, the employer can demonstrate that a legitimate nondiscriminatory reason motivated the adverse employment decision. The Supreme Court made clear at the close of its latest term that the burden shifting pattern is “only to aid courts and litigants in arranging the presentation of evidence: ‘The ultimate burden of persuading the trier of fact that the defendant intentionally discriminated against the plaintiff remains at all time with the plaintiff.’ ” Watson v. Fort Worth Bank & Trust, — U.S. -, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988), citing Texas Department of Community Affairs, and United States Postal Service Bd. of Governors v. Aikens, 460 U.S. 711, 715, 103 S.Ct. 1478, 1482, 75 L.Ed.2d 403 (1983). In a second group of cases, the Supreme Court and Circuit Appellate Courts found a violation of Title YII without proof of “intentional” discrimination. See Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). In these cases, facially neutral employment practices may be held to adversely impact on protected groups. The Court need not find that the employment practices were adopted by the employer with a discriminatory intent. These cases have been called “disparate impact” cases wherein the trial court is often treated to dueling experts who demonstrate statistical disparities rather than specific incidents of discrimination. In these cases, the distinction between “positive” and “normative” studies blurs; statisticians put on the hats of philosophers, economists, psychologists, and sociologists as they justify the underlying assumptions of their economic models, and their choice of statistical methodology, and they then offer competing explanations for those disparities. The Supreme Court recognizes and this Court accepts the principle that no matter the characterization of these factual issues, the ultimate legal issue is consistent among the two broad types of cases: Did the employer (intentionally or not) adopt practices which impermissibly discriminate against persons? Watson v. Fort Worth Bank & Trust, — U.S. at -, 108 S.Ct. at 2785 citing with approval Justice Stevens’ concurrence in Washington v. Davis, 426 U.S. 229, 253-254, 96 S.Ct. 2040, 2054, 48 L.Ed.2d 597 (1976). This principle, that the ultimate legal issue for the purposes of Title VII is not susceptible to sharp compartmentalization, maintains a certain flexibility that responds to the historical pattern of employment practices. Practices which were once overtly discriminatory were transformed after the effective date of Title VII into diverse and more insidious policies which visited disproportionate affects on protected groups. See Griggs v. Duke Power Co., 401 U.S. at 428-429, 91 S.Ct. at 852-853. Unfortunately, the setting of Griggs does not transfer well into a meaningful comparison with the case at bar. Griggs concerned the use of general aptitude tests and a high school diploma requirement which were not demonstrably related to the jobs for which they were used. The subjects of policies and practices before this Court are highly educated professionals; and unlike Griggs, the promotion system is awash in sophisticated subjective determinations. Until the Watson ruling in the closing days of the Supreme Court’s 1988 term, all of its “disparate impact” cases involved standardized employment tests or criteria. Albermarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975) (written aptitude tests); Washington v. Davis, 426 U.S. 229, 96 S.Ct. 2040, 48 L.Ed. 597 (1976) (written test of verbal skills); Dothard v. Rawlinson, 433 U.S. 321, 97 S.Ct. 2720, 53 L.Ed.2d 786 (1977) (height and weight requirements); New York City Transit Authority v. Beazer, 440 U.S. 568, 99 S.Ct. 1355, 59 L.Ed.2d 587 (1979) (rule against employing drug addicts); Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982) (written examination). Watson represents an expanded use of disparate impact theory into the area of employment policies and procedures which contain subjective evaluation elements. Heretofore, courts could presume that val-uative mechanisms which were not amenable to rigid, standardized criteria and required subjective elements must be analyzed under the disparate treatment analysis which required a finding of intentional discrimination. Over the strong objection of the employer that subjective selection practices would be impossible to defend under disparate impact analysis, the Supreme Court reasoned that employment decisions based on subjective employment criteria may have effects that violate Title VII even though they are neutral on their face and discriminatory animus cannot be easily proven. Watson — U.S. at-, 108 S.Ct. at 2786. There are two aspects of this finding which are especially relevant to the factual issues in the case at bar: (1) selection systems which combine “subjective” and “objective” criteria are generally “subjective” in nature; and (2) when discretion is delegated in an organization, there is a danger that the person designated to perform the key evaluation may act with discriminatory intent in a manner which violates Title VII. NATIONAL ORIGIN DISCRIMINATION Throughout the litigation, there seemed to be some question as to what is an “Hispanic.” Some witnesses seemed to consider it a racial distinction. One witness for the Defense, not a member of the class and carrying a surname of Northern European origin, stated that because of a Latin maternal ancestor he “liked to think that he was Hispanic.” The implication of his testimony is that he had not felt discrimination and bore Hispanic class members no ill will. No class members were in doubt that their ethnic heritage was from Spain or of Spanish origin and the Defendant does not dispute whether any individual member possesses this quality. Some class members born in Latin American Countries and naturalized citizens are covered by the prohibition against national origin discrimination. The majority of the class members are native U.S. citizens; they are covered by the prohibition against ethnic origin discrimination. For the purposes of this class, the appellation “Hispanic” will signify ethnic and national origin categories. The Bona Fide Occupation Qualification (“BFOQ”) exception of Section 703(e) applies to all forms of discrimination prohibited by Title VII except racial discrimination. The exclusion of racial discrimination from the BFOQ results from a deliberate Congressional decision to prohibit all racial classifications in employment. 42 U.S.C. Sec. 2000e-2(e)(l) (1982); 110 Cong. Rec. 2550, 7271 (1964). However, for cases where BFOQ is not in issue, the Courts seem to consider the two forms of discrimination, racial and ethnic, interchangeable and the prohibitions against each are substantially similar. See Regents of the University of California v. Bakke, 438 U.S. 265, 287-99, 98 S.Ct. 2733, 2046, 57 L.Ed.2d 750 (1978) (opinion of Powell, J); id. at 355-62 & n. 34, 98 S.Ct. at 2781-85 & n. 34 (opinion of Brennan, Marshall, White, Blackmun, JJ.); Korematsu v. United States, 323 U.S. 214, 65 S.Ct. 193, 89 L.Ed. 194 (1944); Hirabayashi v. United States, 320 U.S. 81, 63 S.Ct. 1375, 87 L.Ed. 1774 (1943). One other aspect of the question bears discussion. In Espinoza v. Farah Manufacturing Co., 414 U.S. 86, 94 S.Ct. 334, 38 L.Ed.2d 287 (1973) the Court held that the exclusion of Mexican aliens from employment at a manufacturing plant where 96 percent of the workers were Hispanic citizens of the United States did not violate Title VII. Thus, disparate treatment on the basis of alienage is not prohibited by Title VII and of course, no disparate impact upon Hispanic workers (U.S. citizens) at the plant existed. Alienage is not an issue before this Court in this cause. CLASS ACTION CLAIMS OF DISPARATE TREATMENT It is not only disparate impact cases which involve the use of statistics. When a class of persons bring a claim of disparate treatment, typical evidence in support of the claims include class-wide statistics as well as numerous anecdotal accounts of disparate treatment. The anecdotal testimony provides the context by which to understand the statistical models. Conversely, the statistical picture provides a context by which to understand the detailed histories of class members. Similar to the axiom about the view of the “forest and the trees”, each bears scrutiny and prudent persons constantly apply the lessons of one view to interpret the other. However, the two types of evidence are not equally probative of the presence or absence of discrimination. The Supreme Court addressed the use of statistical analysis in support of allegations of disparate treatment in Hazelwood School District v. United States, 433 U.S. 299, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977). Hazelwood concerned allegations of racial discrimination in hiring teachers by a public school district in the suburbs of St. Louis, Missouri. Hazelwood stands for the proposition that appropriate statistics compare the racial (or ethnic) composition of the general population, to that of the labor market, and then to that of the those hired by the school district, for the relevant period of inquiry. Thus, the labor market must be narrowed to only those persons with undisputed qualifications for the job. Another example of applicant-flow statistics is to examine those persons in the labor market who have actually expressed an interest in the job offered by the employer. Hazelwood did not speak to the question of whether applicant-flow statistics and similar economic models are the best analytical tool in all circumstances. Though the limiting of the relevant pool appears meaningful, such an analysis does not take into account distortions in the proportion of minority applicants from such factors as deterrent effects of the very employment practice which is in dispute. Nor would such a model compensate for a combination of impermissible discriminatory policies and practices which act to narrow the applicant pool prior to the promotion decision which is alleged to be discriminatory. From Ha-zelwood to the Court’s latest pronouncement, the Supreme Court left the need for applicant-flow statistics to be determined on the facts of each case. Hazelwood, 433 U.S. 299, 308 & n. 13, 97 S.Ct. 2736, 2741-42 & n. 13; Watson, — U.S. at - -, n. 3, 108 S.Ct. at 2789, n. 3. At no time has the Supreme Court advocated abdicating the legal analysis of causation and relevance to the statistical analysis of probable outcomes of one event, given one or more underlying conditions. This Court finds statistical summaries useful but limited. How useful? “Statistical disparities must be sufficiently substantial that they raise ... an inference of causation.” Watson, — U.S. at -, 108 S.Ct. at 2789. The absence of a statistical disparity where the assumptions underlying the model are competent raises a rebuttable inference of no violation of Title VII. How limited? Precise numerical values are deceptive; they are only as descriptive as the accuracy of the assumptions which underlie the model. The proper starting point is to determine whether the parties have identified the specific employment practice or practices which are in dispute. Most economic models cannot distinguish among practices which may involve great differences in quality among the choices. Thus, when applied to subjective evaluation mechanisms for promotional opportunities, statistics may do no more than indicate a pattern of deviation. The probative weight to give the statistical relationship is a matter for the fact finder. Thus, the magnitude of the deviation which is required and the import of any deviation demonstrated in the data are both legal determinations for the Court. Though the Supreme Court has left the determination of the relevance and probative force of statistics to case-by-case analysis, it is clear that the District Court should be guided by the principle that the ultimate burden of proving that discrimination against a protected group has been caused by a specific employment practice remains with the Plaintiff at all times. Watson — U.S. at -, 108 S.Ct. at 2790. This is true of the probative weight to give to the Plaintiff’s showing in support of its prima facie case and the Defendant’s rebuttal. It is also true where Defendant attempts to demonstrate through statistics that a legitimate business reason is responsible for the disparate treatment allegation and that business necessity is responsible for the policy which has a disparate impact. CLASS ACTION CLAIMS OF DISPARATE IMPACT As stated previously, the Supreme Court formulated the theory of disparate impact in Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971) whereby a violation of Title VII could be found without proof of intentional discrimination. If the Plaintiff proves disparate impact, usually by statistical evidence, then the burden of production shifts to the defendant to prove that the employment practice is justified by “business necessity” or is “related to job performance.” Id. at 431, 91 S.Ct. at 853. In both the disparate treatment and disparate impact pattern, the burden shifts back to the plaintiff to prove that the offered justification is a pretext for discrimination. Albermarle Paper Co., v. Moody, 422 U.S. 405, 425, 95 S.Ct. 2362, 2375, 45 L.Ed.2d 280 (1975). The burden shifting pattern just described is a formulation so reasonable in appearance that one might wonder how lawyers could find sufficient differences about which to argue. However, like the directions for assembling my grandson’s model car, it is not as easy in practice as it appears on paper. Does the theory of disparate impact require a strong showing of business necessity or an effortless demonstration of relation to job performance? Though scholars can argue whether the scope of “permissible disparate impact” is narrow or broad, perhaps the more helpful inquiry with regard to Defendant’s rebuttal case is whether the showing of business necessity is a meaningful explanation which justifies distinguishing one person from another so as to slow but not derail Congress’s intent in enacting Title VII. No matter the scope of the business necessity defense, the requirement of demonstrating a business necessity is a powerful tool to tear away the layers of subjective employment practices to evaluate persons for promotion. Subjective employment practices can mask intentional discrimination whether the evaluations are made by sophisticated executive officers or delegated to numerous mid-level managers. Wat son signals the Supreme Court’s willingness to bring undisciplined systems of subjective decision making under the scrutiny of Courts applying Title VII. Watson, — U.S. at-, 108 S.Ct. at 2785. Not every subjective evaluation system is implemented in a discriminatory manner. Not every statistical showing of significant disparities between the majority and protected group is convincingly explained by a policy which is required for the best operation of the business. However, occurring in combination, a Court would do well to perform intense scrutiny to determine whether a violation of Title VII occurred. FINDINGS OF FACT AND CONCLUSIONS OF LAW Procedural Posture As stated above, this cause was brought pursuant to Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. Sec. 2000e-16. Named Plaintiff Bernardo M. Perez became a Special Agent (“SA”) with the FBI at the entry level of GS-10 on September 16, 1963. At the time this cause came to trial, Mr. Perez was an Assistant Special Agent in Charge (“ASAC”), at the level of GS16, at the FBI’s El Paso, Texas, Field Office. On October 26, 1983, while acting as an ASAC at the FBI’s Los Angeles, California, Field Office, Mr. Perez filed a formal Equal Employment Opportunity Commission (“EEOC”) complaint of discrimination based on national origin. Later the discrimination claim was amended to add religious discrimination and reprisal for filing the grievance. On March 26, 1984, the FBI, through its designated Equal Employment Opportunity (“EEO”) officer, notified Perez of a proposed finding of “no discrimination.” On April 17, 1984, Plaintiff requested a hearing before an EEOC Attorney-Examiner. On January 15, 1984, Perez filed a second EEO complaint stemming from alleged reprisals. By September of 1984, a proposed disposition of “no discrimination” was issued and Perez again requested a hearing before an EEOC Attorney-Examiner. The two claims were consolidated for hearing. The final decision by the Department of Justice was returned on December 19, 1986 and rejected Perez’s claims of reprisal and discrimination based on national origin and religion. This suit was filed on January 16, 1987. This Court has proper jurisdiction of the parties and the subject matter of this cause. The Plaintiff has exhausted all administrative remedies as to both the individual and class complaints. This Court certified the class as to all past and present employees of the FBI. The evidence at trial supported this Court’s findings in support of class certification with regard to numerousity, typicality, commonality and adequacy of representation. The Federal Bureau of Investigation The Federal Bureau of Investigation is an agency organized under the Executive Branch of the Federal Government. The present Director is William S. Sessions. At the time of trial of this cause, the Bureau numbered approximately 23,300 employees. Of that number, about 9,400 employees are “Special Agents,” the professional staff who perform the investigative functions and staff nearly all of the management policy setting positions within the Bureau. The 9400 Special Agents staff 58 field offices, 400 Resident Agencies , 14 Legal Attaches (“Legates”) in foreign countries, and training and administrative facilities located in Washington D.C. The FBI is known as the “investigative arm” of the United States Department of Justice (hereafter “DOJ”). Though the FBI is autonomous with regard to most of the practices at issue in this cause, the Department of Justice, through its Office of Professional Responsibility, reviews administrative disciplinary investigations conducted by the Bureau of some high level Special Agents and makes decisions binding on the Bureau with respect to the administrative actions. At FBI Headquarters, there are three broad divisions, each headed by an Executive Assistant Director: Law Enforcement Services, Investigations and Administration. The top level professional staff within these three divisions are Special Agents who have spent their entire professional career within the Bureau. With the exception of clerical support staff, most “headquarter” positions are supervisory or “management level” positions at the Bureau. Typically, the assignment of a Special Agent to Washington represents career advancement under the Bureau Career Development Program (CDP), the promotional ladder within the Bureau. Though there are supervisory positions at each field office and Legate, “brick agents” or street agents do not get the nod to go to FBI Headquarters unless they are promoted to supervisor as part of their transfer or comparatively soon after arriving. After entering the bureau, Special Agents are normally promoted from Grade 10 to Grade 11 after two years of competent service; from Grade 11 to Grade 12 after two more years; and from Grade 12 to Grade 13 after completion of three successful years at Grade 12. Management ranks begin at Grade 14. At trial and beyond, the Defendant did not concede the existence of an “accelerated promotion” program whereby agents who are selected for Headquarter assignments receive promotions at a faster rate into supervisory levels and thereafter, though some witnesses called by Defendant admitted to the practice. The preponderance of the evidence demonstrated a policy and practice of promoting agents assigned to Headquarters ahead of peers in field offices. At Headquarters, from the second most top level working down, positions along the hierarchy are named Assistant Directors, Deputy Assistant Directors, Section Chiefs, Assistant Section Chiefs, Unit Chiefs, and Supervisory Special Agents. At the field offices, the highest ranking Special Agent is the Special Agent in Charge (“SAC”). The SAC is assisted by one or more Assistant Special Agents in Charge (“ASAC”), Field Supervisors, Supervisory Senior Resident Agents, Relief Supervisors, and Investigative Special Agents. Utilization of Spanish Language Skills A starting point to understanding the interplay of duty assignments, conditions of employment, and the promotional practices in the Bureau is an examination of the Bureau practice for utilization of the Spanish language skills of its agents. It is significant to the Court that there developed at trial a significant difference between policy and practice at the Bureau with regard to the treatment of persons who possess language skills. These differences are at the heart of Plaintiffs’ claim. In theory, Special Agents enter the Bureau through one of five programs or categories which they designate by election. Those categories are Law, Language, Science, Accounting or Diversified. The Diversified category, also known as the “Modified” program, admits candidates who otherwise meet Bureau education requirements and have some prior professional or experiential background which the Bureau can utilize or develop through training to assist the Bureau. Still at the level of theory, the Bureau policy is that an agent’s skills, ability and expertise for which he or she was hired may be used by the Bureau at any and all times during a Bureau career. Candidates are informed of the Bureau’s prerogative in this regard at the time of their entrance into the Bureau. In practice, with respect to some abilities and skills, particularly ability in the Spanish language, the utilization of these skills is particularly burdensome on agents and significantly affect in an adverse manner the conditions of employment and promotional opportunities. Some of the related allegations of the Plaintiff concern: the manner in which these skills are identified for use by the Bureau; the presumption that the skill will be utilized by agents who entered the Bureau under a skill program other than language; and the implications for career advancement for the agent whose Spanish ability is utilized on a regular basis. The Personnel Director of the Bureau testified that Special Agents who qualify as a lawyer but who have a language ability need not take a language exam. This view was echoed by other top officials. The consensus among management is that Bureau Special Agents cannot be forced to take a language test. Under that principle, an agent who enters under the accountant or law qualification, or under the modified program could not be forced to take a language exam. At the time of trial, officers of the Bureau with supervisory responsibility over these areas testified that they did not know of any person who was forced to take the language exam. Their belief was that all persons are given the language exam; anyone may refuse to take it. The evidence heard at trial does not support this view as to Hispanic agents. The Court finds that a significant number of Hispanic agents were ordered to take the exam on threat of insubordination. No person outside the protected group faced similar requirements. Other Hispanic agents were simply given the exam; no Hispanic agent testified to knowledge of the opportunity to refuse the exam. The Defendant did not rebut this claim. Similarly, though management expressed its view that the Bureau cannot force persons who enter under an alternate qualification program to monitor Spanish language wiretaps, the preponderance of the evidence received at trial established that Hispanic agents from any classification have been called to monitor Spanish language wiretaps. With respect to distinctions within the linguistic qualification category, the stated policy of the Bureau is to make no distinction between “native speakers” and “learned speakers.” All linguists must sign a “contract” or affirmation that the Bureau can rely on their language skill and make assignments throughout their career based on their skill. The Bureau maintains that the “needs of the Bureau” require that the Bureau can call upon agents at any time in their career to perform the special skill for which they were hired by the Bureau and that this requirement is a legitimate, nondiscriminatory reason for employment decisions made exclusively on the basis of an agent’s language skill and that this is a “business necessity” within the meaning of Title VII case law. The evidence at trial demonstrated that Spanish speaking Hispanic agents who enter the Bureau under another specialty who are ostensibly not subject to the skills use requirement as to language, nonetheless are used for the same purposes. The Defendant, through its officers, tendered the argument that there is a distinction between a superior informally calling on the language skills of an agent and the Bureau “officially” using that skill, meaning that the superior may be permitted to do that which the employer cannot do. However, from the standpoint of Title VII law and the findings of this Court, it is a distinction without a difference. The practice of informally selecting the skills of Hispanic Spanish speakers who entered under an alternate program other than language is so widespread a practice that it is a policy of the Bureau. As to those persons, Anglo and Hispanic, who enter as language specialists and sign the contract, whether the Bureau will equally rely on the language skills of “native speakers” (predominantly Hispanic) and “learned speakers” (both Hispanic and Anglo) is hotly contested. It appears to the Court that with regard to Hispanics, the Bureau does not distinguish between learned and native speakers; their language skills are used in far greater proportion than Anglo counterparts in every category; undercover assignments, witness interviews, and Title III wiretaps. The Bureau makes no formal distinction between learned and native speakers; however, in practice, Anglos who acquire Spanish language proficiency voluntarily “opt out” and are relieved from assignments made in reliance on the language skill. No similar opportunity has been given Hispanic agents. The Court finds that Hispanic Special Agents, be they learned or native speakers, are assigned to Spanish language wiretaps in proportion significantly greater than Anglo Spanish speaking counterparts. The practice of the Bureau of utilization of Spanish language skills of its agents evolved over time. The systematic discrimination identified in these findings probably did not begin as intentional bias. It appears that during the 1960’s and early 70’s, when Hispanic agents entered under an alternate qualification program than language, it was presumed that (1) they spoke Spanish and (2) they were amenable to using their abilities for the Bureau. One example of a common pattern is one agent member of the Plaintiff class, after becoming a Special Agent, who was asked informally whether he knew Spanish. He entered under the “modified program” in the late 1960’s. After responding that he had some minimal Spanish speaking ability, he was called upon regularly to do translations and Title III wiretap assignments over many years. In 1984, the agent was asked to take a Spanish exam. He was tested over the phone and determined to be proficient. There was no notice to him that he could refuse to take the exam. The same agent stated that he is not proficient in Spanish, cannot conduct hostile interviews, and cannot write the language. Some Special Agents within the class testified that it was not made known to them the category assigned to them upon their entry into the Bureau. Many agents in Plaintiff class were under a misapprehension as to their classification. A common pattern is that some believed that they entered under a “diversified” or “modified” program reflecting a liberal arts education and some professional experience in military service, police, or business. However, during the process of discovery for this civil litigation, some agents who examined their personnel files learned that they were classified under the language program along with, or in the place of, their expected classification. It was not uncommon for agents to discover they had qualified under two programs — one for their profession and the other based on their language skill. One class member who graduated high in his class at a prominent law school and is a member of the California bar found upon review of his file that he had entered under law, modified, and minority programs. He took the Spanish exam by order of his superiors and has served primarily on Spanish language special assignments and wiretaps since taking the language test. He testified that he could not translate written Spanish to English and possesses only a rudimentary knowledge of simple Spanish conversation. The agent testified that he has applied for nine (9) in-service training programs and received none in (6) years with the Bureau; “[t]his reflects a Bureau attitude that Hispanics are not qualified for those experiences which can make them upwardly mobile.” The same agent, by way of illustration that he had not been considered for promotion on the basis of his legal abilities, noted that one of his classmates who entered the Bureau was sitting at the Defense table assisting the Defense of the case in her capacity as Special Agent assigned to legal division at headquarters. The illustration is not dis-positive of an individual claim of discrimination; however, together with other accounts, it supports this Court’s decision to discount the significance of entry classifications and focus rather on treatment of employees within and without the Bureau’s entry level program classifications. While class members who qualified under an alternate plan than language were selected for testing in the Spanish language, Anglo counterparts were not tested for language skills. The testing was not voluntary; failure to submit to the testing would be sanctioned formally or informally. Defendant did not refute the clear implication of the evidence that only Hispanic surname agents were selected for the testing. The presumption of the Bureau that not enough non-Hispanic agents have a second language skill to justify testing every agent, in the context of the Bureau’s insistence that the vital interests of the Bureau’s mission require the skill, is inconsistent and undercuts Defendant’s argument that disparities can be justified by reasons of business necessity. For the tests given to Plaintiff class members who did not enter the Bureau under the language category, the test results are highly suspect. Many of these tests were given by oral conversation exams conducted over the phone and lasting fifteen minutes or less. Some agents who failed a written exam were ordered on threat of insubordination to retake the written exam until a passing grade was achieved. The Hispanic Special Agents who were selected for testing were already qualified into the Bureau under another classification. Once tested, Hispanic agents were assigned regularly to special duties such as wiretaps, undercover assignments, special duties, and duty station transfers based on their language skills. Some class members testified to their low functioning in Spanish language, and that the testing methods (oral telephonic exams) incorrectly evaluated their skills at a higher level. These agents’ objections to the accuracy of their language rating went unheeded. The Court finds that on the basis of Hispanic surname or other selection based on the agent’s national origin, the Bureau singled out agents to take the language skills test and made assignments significantly affecting their conditions of employment in an adverse manner. The Court further finds that no similar presumption attached to non-Hispanic agents. Though the Defendant attaches significance to the fact that Special Agents who possess special skills should be expected to use them to further the mission of the Bureau, the burden of that requirement falls disproportionately on Hispanic agents. Plaintiff class members testified to their personal knowledge that non-Hispanic Spanish linguists had stated their preference to be relieved of their requirement to use the Spanish language in the course of their duties after a period of three to five years utilizing the language. Defendant’s pointing out that most Hispanic agents had not made such a request is not convincing. The evidence demonstrated that the practice of the Bureau is to place the weight of deplored assignments on Hispanic Special Agents. Whereas Anglos are relieved from the requirement of using their language skills and focus on other skills which would enhance promotional prospects, the same opportunity is denied Hispanic agents. The preponderance of the evidence demonstrated that Anglo linguists, whether trained by the Bureau or entering the Bureau with their language skill, are not utilized for unpopular assignments in significantly disparate amount, and Anglos may be relieved of the burden of using the language skill in the regular course of their employment — an opportunity not equally available to the Plaintiff class members. “Minority” Program Hiring Defendants did not offer evidence in support of a theme they interjected on cross examination: that some Hispanic agents who enter under a “minority” entry program are required to meet lower standards. The theme invites this Court to find that “minorities” who substantially meet the same rigorous requirements as Anglo applicants are not full peers for the purpose of employee benefits, conditions of employment, and promotion. It follows, under this argument, that the Bureau could impose different conditions of employment between minority applicants and non-minority because in fact there are two classes of agents with a different employment relationship to the Bureau. However, no officer of the FBI testified that there exists two classes of Special Agents or a policy distinguishing promotional opportunities or conditions of employment between minority program Special Agents and others. Witnesses at the top level of the promotional pyramid denied that there is any two tier program, that linguists are not eligible for promotion, or that Hispanics are promoted less often. There was no direct evidence of the presence of a “minority” hiring program, or that Hispanic agents are judged by different standards. Further, there was no evidence that lower standards would result in hiring of Hispanic Special Agents who could not match their Anglo peers in every category necessary to the Bureau’s mission. This Court does not find credible the implication that some members of the Plaintiff class entered the Bureau as “second class” agents and that disparate conditions of employment, expectations for their duties, or promotional decisions could be made in some function of the manner of their entry. The evidence does not support it. The proposition is inconsistent with other positions taken by the Defendant through its Assistant Directors. Spanish Speaking Agent’s Duty Assignments The discussion of language skill qualification becomes significant in relation to the duty assignments made to Spanish speaking Special Agents and the result of these assignments on promotional opportunities. Hispanic agents who speak Spanish, whether they be language specialists or qualified under another program, such as law or accounting have been used in disproportionate numbers to perform Title III wiretaps. The wiretaps involve long periods from 8-12 hours during which an agent is confined or secreted in a vehicle or room listening to conversations made over a telephone. When a judgment is made in the course of an investigation that the wiretap may involve some conversations in the Spanish language, then FBI practice is to assign a Spanish speaking agent to monitor the wiretap. A wiretap is extremely difficult for the persons involved. Among the reasons the duty is difficult are these factors: 1. Shifts are long, often 12 hours in duration and movement is restricted. 2. A common assignment involves rotations of “12 hours on and 12 hours off”, 30 to 90 day assignments. 3. The duty is not voluntary. Agents are assigned to the duty without regard to their present assignments or other duty. 4. Agents assigned to Title III wiretaps are taken off of the active cases they have developed. Their own cases are reassigned to other agents and not returned to the agent who monitored the wiretap at the conclusion of the assignment. 5. Title III wiretap assignments often take an agent away from his or her home office. Agents travel away from their families for long periods. The Court finds by the great weight and preponderance of the evidence that: (1) Hispanic agents suffer disparate treatment in the conditions of their employment; and (2) these conditions affect their promotional opportunities in an adverse manner. The Defendant introduced evidence that current assignment rosters reflect that Spanish speaking agents, both Anglo and Hispanic, are similarly available and assigned to these wiretaps. However, not one Anglo agent appeared and testified that his Spanish language skills have been used in a similar manner on Title III wiretap duty. On the contrary, Hispanic agent after agent testified they have never seen an Anglo Spanish speaker on a Title III wiretap assignment in their career. Class members testified to the existence of a widely recognized “Taco Circuit” or “Tortilla Circuit” whereby Hispanic Spanish speaking agents were regularly chosen for 30 to 90 day assignments doing wiretap duty. They shared this duty with other Hispanic Spanish speaking agents. They saw the same faces and conspicuously did not see Anglo Spanish speakers. Some witnesses had performed 20 to 25 wiretaps lasting a minimum of 30 days each never observing one non-Hispanic sharing similar duty. Hispanic agents who qualified under the accountant and lawyer programs are not immune from assignment to wiretaps for significant periods at significant frequency. Agents testified that a widely used aphorism in the Bureau is “[other than Hispanic language specialist Special Agents,] only Hispanic accountants and only Hispanic lawyers sit on wiretaps.” Implications For Promotion The Spanish language wiretap duty has a marked and significant effect on the promotional opportunities of Hispanic agents. An agent whose cases are reassigned cannot “collect the ticket” when the investigation bears fruit. Other agents compile a long record over years of successful operations as the Hispanic linguist performing frequent temporary duty assignments receives letters of appreciation but fewer successes on the important indices which lead to promotion, closed cases and informant development. The fact that an individual is given positive recommendations by his superior alone does not effectively refute an allegation that a person has suffered an adverse employment condition or that he has not suffered discrimination in employment. Testimony throughout trial demonstrated that small differences and nuances in subjective evaluations have great significance for the true import of an evaluation. Discriminatory animus is elusive; Title VII analysis focuses on the effects of prohibited discrimination and permits this Court, upon consideration of the entire record, to infer the cause. An agent who travels the “Taco Circuit” is often unavailable when special training opportunities arise. These valuable Hispanic linguists are passed over for in-service training or career development opportunities because their skills are needed more in Spanish language related assignments. As a result, Hispanic agents are not exposed to the managers who make the subjective evaluations and determinations for career advancement. Hispanic agents do not gain similar professional experience of Anglo peers, experiences which are necessary to advancement. A frequent complaint supported by the preponderance of the evidence is that an Hispanic agent with five (5) years of Bureau tenure who has ridden the “Taco Circuit” may not have the experience of an Anglo on duty for two years. This is a distinction which is significant to promotional opportunities in the manner in which the FBI determines career advancement. The Court finds that Title III wiretaps and other language related temporary duty assignments are not rewarded by promotion or other conditions of employment in a manner which reflects the importance of their contribution to the mission of the FBI. Business Necessity The Bureau argues that requiring agents who have Spanish language skill to utilize the skill at any time is job-related and necessary to the mission of the FBI; the national drug problem has caught the Bureau with a severe shortage of agents with skill in the Spanish language. However, this defense does not permit the Bureau to visit the burdens of Spanish language duty in a disparate manner on Hispanics over non-Hispanics. Nor does business necessity justify benefit and promotion systems which do not reward Hispanic agents in a manner commensurate with their admitted contribution to the Bureau. The protection of the public safety and welfare does not justify the discriminatory practices demonstrated at trial. Corrections in these systems should not seriously jeopardize the effectiveness and success of the FBI in its law enforcement effort. There was evidence which supported the view that some temporary duty assignments for which an Hispanic is specifically requested are not substantially related to the work that is required. The Bureau has no review mechanism by which one supervisor’s request for an Hispanic is examined to determine the use of the language skill; interview, undercover, or wiretap monitoring. It is never asked whether Spanish language is to be used at all, on occasion, or throughout the duty. Evidence demonstrated that class members were assigned in significant numbers to temporary duty assignments. This takes an agent away from his own case load, home office, and family environment, and Spanish speaking ability is not always required by the nature of the assignment. To justify the effects of the disparate employment conditions, the Bureau must develop a valuative and review mechanism whereby a substantial relationship exists between the special duty assignment which is necessary to the mission of the Bureau and the hardship on the officer. Such a review mechanism does not exist. Hispanic class members suffered disparate conditions of employment in this regard without a showing of justification of business necessity in violation of Title VII. Undercover Assignments Plaintiffs allege that Hispanics perform undercover assignments in disproportionate amounts and that undercover operations are not voluntary but a requirement of their employment. That Hispanics are better in undercover work by virtue of their appearance appears to be a self-evident proposition. However, the Defendant has not demonstrated by study or testimony the truth of this proposition. A competent study comparing professional investigative techniques in the area of drug interdiction may demonstrate a significant relationship between assignment of Hispanic agents to undercover work in disparate proportions to the important goals of the Bureau. It is equally possible that the hidden presumption is that non-Hispanic agents are not expected to alter their appearance and manners to take on the persona of Anglo criminal elements engaging in the same undercover drug purchases and contacts. Presumably, some non-Hispanic Americans are dealing with Latin American drug suppliers in significant numbers and thus, there is no basis to presume, without examination, that the bulk of undercover drug investigative assignments should be borne by Hispanic agents. Though the FBI’s assignment system contains some degree of coercion, Title VII does not require that assignments be made on a voluntary basis as the system is advertised. The appropriate inquiry is whether or not the undercover assignment system creates a condition of employment that is borne more heavily by Plaintiff class members. Does this system have effects on promotion or benefit assignments which distinguish between persons on impermissible grounds? One could imagine a system whereby tall persons or strong persons are requested to shoulder heavier burdens. This does not unfairly discriminate against tall and strong persons over short and weak persons unless tall and strong persons are not compensated in some manner for their greater contribution and the very reason they are not compensated is that the employer has an irrational bias against these persons. Title VII is implicated when the greater burden is shouldered by members of a protected group and unfair promotion or benefit distribution, by the preponderance of the evidence, is shown to be significant, pervasive, and not justified by business necessity. The Court finds that the contribution of both mono-lingual Hispanic agents and bilingual Hispanic agents in undercover work is not rewarded by benefits and promotion. Although the Bureau has articulated a business necessity supporting the assignment of Hispanic agents to undercover work in statistically disproportionate manner, the Plaintiffs have demonstrated that the contribution of Hispanic agents does not translate into employment benefits. On the entire record introduced at trial, the Court finds that this disparity — in benefits and promotional opportunities — is not justified by reasons of business necessity. An appropriate remedy which satisfies Title VII may not require that certain agents be promoted into positions where their skills are not used by the Bureau. Recognition can be given in other terms. A promotional system may be devised which keeps agents uniquely qualified to further the FBI mission in the field yet at higher grade or benefit level. Though Title VII may not prohibit the Bureau from assigning Hispanic Special Agents to undercover work in disproportionate numbers, Title VII does prohibit the Bureau from failing to credit adequately the contribution of the undercover agent to the mission of the Bureau in terms of promotions and benefits. The record is replete with stories of successful short term and long term undercover assignments for which the Hispanic agent performing the duty perceives that his work did not translate into promotion while non-Hispanic agents who “have the ticket” on the case receive career benefits for their work. Defendants, by way of rebuttal, offered evidence of letters of commendation given to undercover Hispanic Special Agents. Plaintiffs allege that no credit is given on their evaluations in a meaningful manner which translates into promotions. Upon review of the entire record, the Court finds credible the view of Plaintiffs that the Bureau’s evaluation of the contribution of Hispanic agents on undercover assignments, not the fact of making undercover assignments, violates Title VII’s prohibition against discrimination of members of protected groups for promotion and benefits. Defendants maintained at all times that undercover assignments, special duty assignments, and routine assistance for fellow agents on an informal basis are all “voluntary” decisions; the person to whom an inquiry is made may refuse to take the assignment with no adverse result. Plaintiffs offered evidence that while the nature of the FBI mission exerts pressure on all agents to accept assignments, they are not strictly voluntary; thus the burdens fall most heavily on Hispanic agents. Plaintiffs argue that refusing an assignment, or refusing to assist fellow agents when asked, will reflect badly on periodic evaluations of the agent. Refusal transfers into evaluations whereby the agent is judged “not aggressive”, or “not dedicated.” Such an. evaluation has a significant negative impact on an agent’s career. Ad hoc Investigatory Assistance Class members testified to numerous incidents where they were asked to assist supervisors or fellow agents by performing translations, conducting Spanish language interviews, or accompanying non-Spanish speaking agents on field assignments. On these occasions, though the Spanish speaking agent assists and performs necessary paperwork involving translations, the “ticket” on the case is held by the mono-lingual agent who requested the assistance. The class member’s contribution is not recognized in a manner which leads to promotion; the administrative statistical profile. Gratitude of a fellow agent does not make up for the disruption in the helping agent’s case work. From this pattern came the term “Anglo helper” which testimony demonstrated is in wide, though informal, use throughout the Bureau. Class members from every qualifying classification, attorney, accountant, modified, and linguist testified that they were expected to perform this assistance in significant amounts. Defendants did not introduce evidence that the burden of performing translation assistance was borne equally by Anglo Spanish speakers. The Court finds that because of the manner in which contribution is measured in the Bureau, the evaluation mechanisms do not significantly account for the contribution of members of Plaintiff class who perform the ad hoc translation assistance. The Needs of the Bureau The phrase “needs of the bureau” was often intoned by Defendant witnesses to explain policies which require personal sacrifice on the part of Special Agents for the convenience of Bureau activities. No document of the Bureau memorializes the meaning of “needs of the Bureau.” This lack of precision is not fatal to the concept from the perspective of Title VII. However, the phrase is not prophylactic of all policies of the Bureau, or of decisions made by supervisors in the Bureau exercising their discretionary authority. The witnesses for the Defendant seemed to use the term as a self-justifying proposition. This Court will define the “needs of the Bureau” inversely; those things which the law forbids cannot be within those policies for which “needs of the bureau” is sufficient justification. Though the findings of the Court with respect to Plaintiffs central allegations, taken together, is a limitation as a matter of law on the scope of the needs of the Bureau, the following specific findings are made: The “needs of the Bureau” could not include keeping files on agent’s family members engaging in first amendment protected activity. The “needs of the Bureau” is often recited to justify transfers of agents, one of the most vitriolic issues for Plaintiff class members. Though as a matter of policy all transfers are made strictly on the basis of the needs of the Bureau, in fact, personal needs of agents are taken into consideration routinely and the transfer system is highly susceptible to bargaining with superiors and subjective determinations. Also, to the extent that the requirement to go where one is assigned is a condition of employment, there is the possibility that the weight of certain “hardship” transfers may be visited in an unfair manner. Thus, whether transfers are applied unevenly is subject to scrutiny to determine whether there is a pattern and practice of discrimination. The Bureau has devoted significant administrative resources to easing the burdens associated with field offices where the cost of living is high or conditions are dangerous for agents and their families. The Court is of the opinion that the “needs of the Bureau”, that is, business necessity for the purposes of Title VII, permits a disproportionately high transfer rate of Hispanic Spanish language Special Agents to places where they are needed so long as the transfer does not result in uncompensated burdens in terms of conditions of employment or causes one to forego promotional opportunities. The transfer system is a subjective system which is susceptible to invidious distinctions or favoritism of persons who “have the right connections.” In the second part of this bifurcated proceeding, the Court will revisit the issue of transfers and whether the Bureau can devise a system of transfers which is subject to systematic review within the Bureau. As a mat