CALIFORNIA STATE EMPLOYEES'ASS'N v. State of Cal., 682 F. Supp. 1044 (N.D. Cal. 1987)

*1045 *1046 Bernard L. Allamano, Gary P. Reynolds, Patricia L. Campbell, California State Employees' Ass'n, Sacramento, Cal., Melvin Dayley, California State Employees' Ass'n, Oakland, Cal., Winn Newman, Newman, Sobol, Trister & Owens, Washington, D.C., for plaintiffs.

Christopher W. Waddell, Sacramento, Cal., for State of California, Governor George Deukmejian, and State Dept. of Personnel Admin.

N. Eugene Hill, Asst. Atty. Gen., Janice Rogers Brown, Deputy Atty. Gen., Sacramento, Cal., for defendants.

Bion M. Gregory, Jack I. Horton, James L. Ashford, Amelia I. Budd, Office of the Legislative Counsel, Sacramento, Cal., amicus curiae.

PATEL, District Judge.

In this class action, the California State Employees' Association ("CSEA") and individual named plaintiffs are suing the State of California, Governor George Deukmejian and the State Department of Personnel Administration (collectively referred to as "the state" or "defendant"). The complaint alleges that the state has discriminated against employees in historically female jobs in violation of Title VII of the Civil Rights Act of 1964 ("Title VII"), 42 U.S.C. §§ 2000e-2000e-17, by maintaining job classifications which have been segregated on the basis of sex; compensating historically female jobs less than male jobs which require equivalent skill, effort, and responsibility; continuing the use of a system of pay and classification which perpetuates discriminatory compensation against employees in historically female jobs; and failing to correct other discriminatory conditions of employment.

The specific historically female jobs referred to are key data operator, licensed vocational nurse, office assistant, word processing technician, stenographer, registered nurse, food service worker and librarian. In the charges of the Equal Employment Opportunity Commission ("EEOC") attached to the complaint, these jobs are compared to groundskeeper, photographer, laborer, agricultural chemist, highway equipment cleaner and assistant civil engineer.

Defendant moves for summary judgment under Rule 56(e) of the Federal Rules of Civil Procedure. Defendant claims that it has (a) adhered to prevailing market rates in establishing state salaries and thus cannot be held liable for resulting discriminatory pay practices and (b) adopted a statewide affirmative action plan which proves that it does not intentionally discriminate against its female employees.

Plaintiffs have presented statistical, historical and anecdotal evidence purporting to show that (1) women and men who begin state employment with equal wages progress at different rates depending on sex; (2) class members are part of a sex-segregated workforce where wage disparities favor male-dominated classifications; (3) there are dramatic historical patterns of discrimination continuing into the present; and (4) certain employment decisions by the state are made on a subjective basis, to the disadvantage of women as a class.

The court finds that plaintiffs have presented sufficient evidence on all elements of their case to raise genuine issues of fact requiring resolution at trial. Accordingly, summary judgment must be denied.

The state developed its classification and pay plan in the 1930's. In 1934, the state's salary and classification plan specifically used sex as a criterion for setting salaries:

*1047 [t]he salary schedules ... are the result of a study of such factors as market rates [and] cost of living.... Certain supplemental factors were also taken into consideration, namely ... the age, sex, and standard of living of the employees normally recruited for a given class....

F. Becker, Administration of the Personnel Program in the State of California at 62 (1934), Pl. Ex. 301.

In addition, Fred Telford, who formulated the state salary and classification plan, id. at 44, used sex in his classification questionnaire. See The Telford Classification Manual app. 3 (1937), Pl. Ex. 302. Prior to the completion of this questionnaire, Telford wrote that salaries should be determined, in part, according to criteria including "the single man, the single woman, and the head of a family of average size." Telford, "The Principles and Technique of Preparing a Plan of Compensation for Positions and Employees in a Large Organization," 18 Public Personnel Studies 16, 18 (1925), Pl. Ex. 303. Accordingly, the state used job surveys which compiled data on the sex of individuals and on the sex composition of the jobs being surveyed. See "Employment and Wage Questionnaire," in Becker, supra, Pl. Ex. 301; see also Final Report of Findings of the State of California Employment and Wage Survey (1938), Pl. Ex. 304.

In setting salaries, the state also used cost of living estimates, which it found to be different for females and males. In fact, a special committee appointed by the California Civil Service Commission found that single female clerks had a lower cost of living than single male clerks. California Civil Service Commission, Cost of Living Survey: Report to the California State Civil Service Commission at 24-25 (1923), Pl. Ex. 310.

Other alleged actions on the part of the state include discrimination against women in its recruiting practices by using, until 1972, sex-specific help wanted ads, job bulletins and job titles. See Pl. Ex. 306. Plaintiffs further claim that differential wage scales for men and women have resulted in the sex segregation of jobs, with female-dominated jobs carrying lower salaries than jobs filled predominately by men.

The evidence tends to show that salary relationships established in the 1930's were carried over and continue to prevail today. For example, the ratio of female salaries to male salaries has not changed significantly in nearly 50 years. In 1938, the median monthly income of women in state service ($119) was 73.9 percent of the same figure for men ($161). See State Personnel Board ("SPB"), Thirteenth Biennial Report at 124 (1939), Pl. Ex. 305. In 1986, the average salary of women in state employment was 74.9 percent of the average salary for men. Dep't of Personnel Admin., Report to the California Legislature and Exclusive Representatives of State Employees on Information Relevant to the Salaries for Female-Dominated Jobs at 5 (1986), Pl. Ex. 336.

In 1975, the SPB studied the role of sex discrimination in the establishment of state salaries. SPB Issue Memo, Boughton to Kurtz at 1-2 (10/31/75), Pl. Ex. 319. The study found that "[r]ecent court actions state that sex discrimination exists in wage practices in some firms which are included in our salary survey." Id. at 3. It also found that businesses employing 500 or fewer employees, which were included in the state's salary data, generally paid men more than they paid women for performing the same jobs. Id. at 6-7. Despite these findings, the state did not eliminate from its surveys the salary data from the businesses which discriminated. See Crain Deposition, 101:1-3, 102:10, 103:4, Pl. Ex. 211; Walter Deposition, 56:22-24, Pl. Ex. 216.

The state has continued to study the current impact of historical sex discrimination on salaries. Bruce Crain, the state's designated salary expert, wrote in an internal SPB memo in 1980: "We have acknowledged that discrimination has been a factor in holding the salary levels down in those occupations populated by women...." *1048 SPB Memo, Crain to Executive Staff at 1 (4/18/80), Pl. Ex. 402.

The state has admitted that women historically have been segregated into lower paying jobs. See SPB, Special Report: The Status of Women in the California State Civil Service at 4, 7 (1974), Pl. Ex. 323. Historical practices aside, plaintiffs maintain and have submitted declarations to the effect that:

(1) certain class members who have in recent years contacted state offices about job opportunities were advised to apply for clerical jobs. Teske Decl. ¶ 1, Pl. Ex. 151; Hannel Decl. ¶ 2, Pl. Ex. 119; Llata Decl. ¶¶ 1, 2 Pl. Ex. 136.

(2) women have been discouraged, and in some cases prevented, from applying for higher-paying traditionally male jobs. Scott Decl. ¶ 2, Pl. Ex. 148; Mackey Decl. ¶ 2, Pl. Ex. 137; Heischman Decl. ¶ 2, Pl. Ex. 124; Howes Decl. ¶ 2, Pl. Ex. 125; Daniels Decl. ¶ 7, Pl. Ex. 107.

(3) when male and female candidates for promotion are equally qualified, the man is promoted more often than the woman. Dodgin Decl. ¶¶ 2-5, Pl. Ex. 112.

(4) when women are able to enter traditionally male classes, they are at risk of being treated discriminatorily. See Hull Decl. ¶ 10, Pl. Ex. 126; Grunnet Decl. ¶ 9, Pl. Ex. 118; Lara Decl. ¶¶ 5, 9, Pl. Ex. 130, 133; Carter Decl. ¶¶ 6, 10, Pl. Ex. 105; Atkins Decl. ¶¶ 12, 27, Pl. Ex. 102; Haynes Decl. ¶ 4, Pl. Ex. 123; Harling Decl. ¶ 9, Pl. Ex. 120; Tavares Decl. ¶ 12, Pl. Ex. 150.

The general policy of the state is to hire a new employee at the minimum of the salary range for the classification into which the employee is hired. If there are recruitment and retention problems for the class, or if the individual has "extraordinary qualifications," a department may offer a higher than minimum initial salary. Cal. Gov't Code § 19836. The criteria for offering an above-minimum appointment are not objectively defined. See SPB Reference Calendar (9/20-21/72) at 107, 109. Plaintiffs contend that the state commonly exercises this discretion more often in favor of men than women. Scott Decl. ¶ 3, Pl. Ex. 148.

Plaintiffs point out that defendant's exhibits reveal patterns of promotion discrimination against women. They submit evidence, for example, that supervisory-level job categories are characterized by proportionately fewer women than the corresponding non-supervising job category. See Exhibit AA to Shockley Affidavit, Pl. Ex. 331; see also Piazza Decl. ¶ 6, Pl. Ex. 142; Rosen Decl. ¶ 3, Pl. Ex. 146. Moreover, plaintiffs have presented evidence which suggests that some departments use predominately male classes as the source for promotions to higher paying positions and exclude predominately female classes. See, e.g., Elmore Decl. ¶¶ 3-4, Pl. Ex. 115; Brewer Decl. ¶¶ 3-4, Pl. Ex. 104.

Other documented complaints from plaintiff class members include numerous instances of men being promoted instead of equally or better qualified women. See, e.g., Piazza Decl. ¶¶ 2-5, Pl. Ex. 142; Daniels Decl. ¶¶ 3-5, Pl. Ex. 107.

There is also evidence that the state intentionally discriminates against women in failing to publicize promotion or transfer opportunities, which works to the disadvantage of women. See, e.g., Mackey Decl. ¶ 10, Pl. Ex. 137; Brewer Decl. ¶ 4, Pl. Ex. 104. Plaintiffs also show that women have been denied advancement because of stereotypic biases against them. See, e.g., Kagan Decl. ¶ 3, Pl. Ex. 127; Howes Decl. ¶ 2, Pl. Ex. 125; Hannel Decl. ¶ 6, Pl. Ex. 119.

Certain plaintiffs also allege that some female-dominated classes have fewer training opportunities than male-dominated classes. Plaintiffs' evidence demonstrates this and further shows that this has served to reduce women's opportunities to move into higher paying classes. Tavares Decl. *1049 ¶ 15, Pl. Ex. 150; Brewer Decl. ¶¶ 8-10, Pl. Ex. 104. Plaintiffs submit evidence that some women who are successful in securing training opportunities are set up to fail in their training assignments. Memo to Schwegal from K. Coffee (5/21/86), Pl. Ex. 335.

Defendant has argued that the market establishes wages based on the supply of and demand for labor and the value of the jobs performed. Defendant also questions the importance of discrimination as a factor in wage differentials, suggesting that other factors such as employee turnover, work performance, membership in labor unions, absenteeism and seniority may, in combination, outweigh discrimination in setting wages for female-dominated jobs. Defendant argues that the "market" is an expression of human culture which reflects inherited biases and stereotypes about women and women's work rather than a perfect impersonal machine.

To allow the state to shield itself behind a defense of "market realities," however, would be to ignore the historical role of the State of California in establishing those market rates. Further, with the exception of a small number of trade classes, plaintiffs correctly note that the state has never been required to set its civil service salaries in accordance with the prevailing market rates. Leighton Deposition 78:5-21, Pl. Ex. 212. Under the superseded California Government Code § 18850, the SPB had no mandate to pay prevailing rates, but was only to consider them. Id. at 78:5-10, Pl. Ex. 212. Section 18850 did require, however, that "like salaries shall be paid for comparable duties and responsibilities." See State Trial Attorneys' Ass'n v. State, 63 Cal. App. 3d 298, 304, 133 Cal. Rptr. 712 (1976). Plaintiffs contend that the SPB interpreted the holding in that case to mean that "paramount concern shall be given to internal relationships" in state civil service positions. See id. at 302, 133 Cal. Rptr. 712.

In addition to internal relationships, the state has apparently adhered to a number of other factors in setting salaries, including but not limited to: cost of living, supply and demand of personnel, relative importance of classes, job entry qualifications, working conditions, past and present pay relationships, and recruitment and retention problems. Leighton Deposition, 14:11-15:22, 11:20-12:6, Pl. Ex. 212. Thus, the plaintiffs attempt to show the state's actual use of prevailing market rates has been limited or superimposed upon discriminatory considerations. Gleed Decl. ¶ 8, Pl. Ex. 116.

Since the inception of collective bargaining in 1981, moreover, the state has set salaries without surveying prevailing rates. Crain Deposition 91:18-92:12, Pl. Ex. 211. Thus, the state cannot rely on a market rate defense to insulate it completely from a charge of intentional discrimination. It appears from the most favorable view of plaintiffs' evidence that the state continues to maintain many of the internal relationships that were established as early as the 1930's; that these were based on consciously discriminatory factors; and, as a result, that the state continues knowingly to discriminate against women. Plaintiffs have mustered evidence which, if believed and if susceptible to the reasonable inferences plaintiffs suggest, may be sufficient to establish intentional discrimination.

Defendant secondarily argues that the existence of a "model" affirmative action plan proves that the state has no intention to discriminate. The court does not find this argument grounds for granting defendant's motion for summary judgment.

The state's Affirmative Action Program was established by statute in 1977. Cal. Gov't Code §§ 19790-19798. The program leaves the establishment and implementation of affirmative action goals and timetables to individual departments. Cal. Gov't Code §§ 19790, 19794. The SPB is responsible for providing statewide enforcement, monitoring and coordination of the program. Cal. Gov't Code § 19790.

*1050 Despite its well-intentioned goals, the materials submitted to this court indicate that the Affirmative Action Program has been ineffective in eliminating sex segregation in state civil service. See "Sex Segregation by Job Category," Pl. Ex. 502. In 1984 a Select Committee of the California State Assembly found that a principal factor contributing to this failure has been the state's inability to assume a leadership role in promulgating and enforcing effective policies. Assembly Select Comm. on Fair Employment Practices, Effectiveness of Affirmative Action in the Public Sector at 30 (1984), Pl. Ex. 505. The committee also found that the SPB failed to commit adequate resources to affirmative action. Id. at 82. The state's lofty goals for its affirmative action plan lack credibility in light of its failure to integrate men into historically female-dominated classifications. Walter Deposition, 37:2-6, Pl. Ex. 217.

Another demonstration of the state's efforts to achieve equality between the sexes is its 1977 enactment of the Upward Mobility Act, which was designed to create an "effective upward mobility program for state employees concentrated in low-paying occupations." Cal. Gov't Code §§ 19400-19406. But individual class members state that upward mobility opportunities for women in clerical classifications remain limited. Allen Decl. ¶¶ 2-5, Pl. Ex. 101; Hannel Decl. ¶¶ 4, 15, Pl. Ex. 119.

The SPB has yet to establish a policy as to what constitutes a "low-paying occupation," see Walter Deposition, 51:4-20, Pl. Ex. 215, and it does not have a consistent definition of upward mobility. Connell Deposition, 41:10-42:28, Pl. Ex. 209. Similarly, the use of "bridging classes" has had almost no success in moving women into higher-paid, male-dominated categories. Instead, bridging classes appear to have moved women into female-dominated and mixed job categories. Silva Report, Final Report: Upward Mobility in California State Service at 16-17 (1985), Pl. Ex. 508.

Confidence in the Affirmative Action Program is further weakened by the SPB's failure to enforce California Government Code § 19402, which requires departments that do not foresee meeting their goals to request a reduction in hiring requirements. Plaintiffs present evidence that departments rarely make these requests; instead, they simply fail to meet their goals. Aguilera Deposition, 55:7-16, Pl. Ex. 202.

Moreover, plaintiffs allege that despite its responsibility under California Government Code § 19790 to coordinate and monitor affirmative action efforts, the SPB does not analyze on an ongoing basis the effectiveness of the Upward Mobility Program. In fact, plaintiffs found that in 12 years the SPB Women's Program has not investigated the effectiveness of its program, except for one report which concluded that the program was not assisting women into higher-paid, traditionally male occupations. Stausboll Decl. ¶ 6, Pl. Ex. 501.

In its submissions, the state has detailed many of the good things it has done to try to eradicate sex discrimination. Defendant's Reply Memorandum at 10-12. Conceding the partial success of these efforts, they still cannot stave off a denial of summary judgment. The existence of an affirmative action program does not by itself negate the possibility that plaintiffs may be able to prove intentional discrimination in other aspects of the state's employment practices. The program's existence and implementation merely raise issues of fact. For these reasons, summary judgment is denied on both grounds raised by defendant.

Under Federal Rule of Civil Procedure 56, summary judgment shall be granted "against a party who fails to make a showing sufficient to establish the existence of an element essential to that party's case, and on which that party will bear the burden of proof at trial.... since a complete failure of proof concerning an essential element of the nonmoving party's case necessarily renders all other facts immaterial." Celotex Corp. v. Catrett, 477 U.S. 317, *1051 322-23, 106 S. Ct. 2548, 2553, 91 L. Ed. 2d 265 (1986). See also Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 106 S. Ct. 2505, 2510 (1986) (a dispute about a material fact is genuine "if the evidence is such that a reasonable jury could return a verdict for the nonmoving party.").

The court's function, however, is not to make credibility determinations. Anderson, 477 U.S. at 249-50, 106 S. Ct. at 2511. The inferences to be drawn from the facts must be viewed in a light most favorable to the party opposing the motion. T.W. Elec. Serv. v. Pacific Elec. Contractors Ass'n, 809 F.2d 626, 630-31 (9th Cir. 1987).

The court finds that a fair resolution of the issues raised in this case requires a trial on the merits, as it is not appropriate for the court to "resolve existing factual issues through a `trial by affidavits.'" Proctor v. Consol. Freightways Corp. of Delaware, 795 F.2d 1472, 1477 (9th Cir. 1986). Plaintiffs have produced facts which are material to their case and which raise genuine issues requiring resolution at trial. Summary judgment is therefore denied.

Title VII makes it unlawful for employers to "discriminate against any individual with respect to his compensation terms, conditions, or privileges of employment" or 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 adversely affect his status as an employee" because of the individual's sex. 42 U.S.C. § 2000e-2. Plaintiffs may be able to proceed under either of the two traditional methods of establishing discrimination under Title VIIthe disparate treatment model or the disparate impact model.

To establish a case of disparate treatment, a plaintiff must present direct or circumstantial evidence of the employer's discriminatory intent. American Fed'n of State, County, and Mun. Employees, AFL-CIO v. Washington [hereinafter AFSCME] 770 F.2d 1401, 1406 (9th Cir. 1985); Penk v. Oregon State Bd. of Higher Educ., 816 F.2d 458, 463 (9th Cir.), cert. denied, ___ U.S. ___, 108 S. Ct. 158, 98 L. Ed. 2d 113 (1987). While wage disparities between comparable but unequal job classifications "may be relevant to a determination of discriminatory animus, job evaluation studies and comparable worth statistics alone are insufficient to establish the requisite inference of discriminatory motive critical to the disparate treatment theory." AFSCME, 770 F.2d at 1407. Statistical evidence must be supplemented by independent evidence of discrimination in order to establish discriminatory intent. Id. In addition, evidence that the employer established a wage structure based on prevailing market rates does not by itself support an inference of intent to discriminate. Id.

The lower court decision in AFSCME suggested some types of independent evidence that can serve to corroborate statistical evidence of wage disparities:

[T]he historical context out of which the challenged practices arise; obstacles confronting applicants and/or employees; subjective employment practices utilized by the Defendant resulting in a pattern disfavoring females; the foreseeable adverse impact of those practices; the increase in pay to the Plaintiffs since filing of the instant suit; discriminatory treatment in other areas of employment; and ... recognition of disparate treatment by responsible State officials.

578 F. Supp. 846, 858 (W.D.Wash.1983) (judgment reversed in 770 F.2d 1401 because the corroborative evidence was found insufficient to support an inference of intent); see also Carpenter v. Stephen F. Austin State Univ., 706 F.2d 608, 623 (5th Cir.1983) (plaintiffs may buttress statistical evidence with evidence of a history of discrimination, individual instances of discrimination and opportunities to discriminate).

Plaintiffs in this case have presented statistical evidence of wage disparities. They have supplemented statistics with evidence that for decades the state overtly discriminated *1052 on the basis of sex and that many wage-setting practices and policies have remained fundamentally unaltered since overt discrimination was practiced. In addition, plaintiffs have detailed specific incidents of intentional discrimination.

Much of the plaintiffs' non-statistical evidence predates both the Equal Employment Opportunity Act of 1972, 86 Stat. 103, which expanded the coverage of Title VII to include public employees, and the 1977 opening date of the plaintiff class in this case. Plaintiffs argue that this evidence is relevant (1) to demonstrate present effects of past discriminatory acts, United Airlines v. Evans, 431 U.S. 553, 558, 97 S. Ct. 1885, 1889, 52 L. Ed. 2d 571 (1977); (2) because current practices may arguably be continuations of the pre-1977 illegal practices, Bazemore v. Friday, 478 U.S. 385, 106 S. Ct. 3000, 3010, 92 L. Ed. 2d 315 (1986); and (3) as background for the pattern and practice of discrimination developed by plaintiffs. Evans, 431 U.S. at 558, 97 S. Ct. at 1889. The court agrees that this evidence, though alone insufficient to prove present motive, may help support an inference of post-1977 intentional discrimination.

The statistical evidence presented by plaintiffs shows a general discriminatory pattern which implies motive and intent. See Diaz v. American Tel. & Tel., 752 F.2d 1356, 1363 (9th Cir.1985). The statistical studies performed by plaintiffs' expert, Dr. Richard Drogin, show that women in state service are subjected to discrimination in compensation. Dr. Drogin plotted the career development of women and men who began state service in the same job categories, with the same salaries. After eight years of service, the salaries of the male employees had outpaced the salaries of the female employees to an extraordinary degree in state service as a whole and particularly in the clerical and administrative job categories. Drogin Decl., Pl. Ex. 111.

The clerical job category showed the largest single difference between female and male salaries after eight years, with a z-value (size of the disparity in terms of standard deviations) of 15.745. This figure may support the argument that the state's Upward Mobility Program is ineffective. Drogin's analysis further shows a z-value of 9.169 for disparities in starting salaries between women and men entering state service in the same classifications.

Disparities smaller than those arrived at in plaintiffs' analysis have been held to support an inference of intentional discrimination. The Ninth Circuit has observed that z-values of 5 to 6 and 12 have been held sufficient to establish an inference of intentional discrimination. Gay v. Waiters' Union, Local No. 30, 694 F.2d 531, 551 (9th Cir.1982) (citing Castaneda v. Partida, 430 U.S. 482, 496 n. 17, 97 S. Ct. 1272, 1281 n. 17, 51 L. Ed. 2d 498 (1977) and Hazelwood School Dist. v. United States, 433 U.S. 299, 308-09 n. 14, 97 S. Ct. 2736, 2742 n. 14, 53 L. Ed. 2d 768 (1977)).

Defendant argues that Drogin's analyses suffer from exaggeration, oversimplification, distortion, and missing variables which seriously undermine an inference of discrimination. Defendant's Reply Memorandum at 26-7. Although Dr. Drogin's analysis may be flawed, the court may strike it upon a summary judgment motion only if the proposed testimony is so deficient in methodology and analysis that the court would be without a proper basis to evaluate plaintiffs' claims. See Dolphin Tours, Inc. v. Pacifico Creative Serv., Inc., 773 F.2d 1506, 1512 (9th Cir.1985).

In addition to their statistics, plaintiffs have presented evidence that defendant's classification and pay system has, in its most important respects, remained unchanged since the 1930's. Plaintiffs allege historical sex discrimination, supporting this contention with evidence that other factors in addition to prevailing market rates were used in establishing salaries. The evidence includes a 1923 survey of family budgets, a 1925 article by Fred Telford, a 1937 questionnaire which asks for the sex of the respondent, a 1938 Final Report of Findings of the California Wage Survey, the 1939 Thirteenth Biennial Report *1053 of the State Personnel Board and a 1983 SPB document which indicates the age, sex and standard of living of employees normally recruited for a given class.

In Bazemore v. Friday, the Supreme Court identified the continued use of internal relationships and a salary structure that carry the indicia of earlier discriminatory practices as a violation of Title VII. 478 U.S. 385, 106 S. Ct. 3000, 92 L. Ed. 2d 315. The Court held that an employer is obligated to correct ongoing salary disparities disfavoring protected groups even when such disparities were once legal. Id. 106 S. Ct. at 3006-07. Thus, defendant may be liable if it has continued to operate a discriminatory wage and classification system developed before 1977. Id. at 3006. The Supreme Court observed:

To hold otherwise would have the effect of exempting from liability those employers who were historically the greatest offenders.... A pattern or practice that would have constituted a violation of Title VII, but for the fact that the statute had not yet become effective, became a violation upon Title VII's effective date, and to the extent an employer continued to engage in that act or practice, he is liable under that statute.

Id. at 3006.

Defendant did not change its practice of including in its salary survey wage data from firms known to discriminate against women. With the institution of collective bargaining, defendant adopted the pre-existing pay and classification system and used internal relationships as its starting point without rectifying known past discriminatory practices. In EEOC v. Inland Marine Indus., 729 F.2d 1229 (9th Cir.), cert. denied, 469 U.S. 855, 105 S. Ct. 180, 83 L. Ed. 2d 114 (1984), the Ninth Circuit held:

By refusing to change his subjective wage-setting policies or to bring black wages in line with those of whites, [the defendant's agent] ratified the existing disparities. His ratification constituted all the intent the court needed to find [the defendant] guilty on a disparate treatment theory.

Id. at 1235 (citations omitted). Plaintiffs have presented evidence that defendant recognized sex discrimination in wage-setting policies and considered less discriminatory alternatives, but did not make any significant changes to its system.

From plaintiffs' evidence it thus appears that in the establishment of compensation plans for some jobs, factors other than prevailing market rates were considered. Specifically, plaintiffs have presented evidence that the state was aware, from at least the 1930's into the 1970's and 80's, that pay ratesincluding those adopted from the private sectorcarried the indicia of sex discrimination. To further bolster their statistical evidence, plaintiffs have presented evidence of particular instances of discrimination. Such anecdotal evidence may be combined with statistical studies to support an inference of intentional discrimination. AFSCME, 770 F.2d at 1407.

Plaintiffs have presented sufficient statistical, historical and anecdotal evidence in support of their disparate treatment claim to survive defendant's motion for summary judgment. Though defendant has raised serious challenges to much of plaintiffs' evidence, it is not appropriate for the court to assess the credibility of evidence put forward by the non-moving party at the summary judgment stage. T.W. Elec. Serv. v. Pacific Elec. Contractors Ass'n, 809 F.2d 626, 630-31 (9th Cir.1987). Plaintiffs should note, however, that the court will apply a strict standard in reviewing the evidence at the time of motions in limine.

Unlike disparate treatment, the disparate impact model does not require proof of discriminatory intent. Instead, a plaintiff must prove the existence of facially neutral employment practices that "in fact fall more harshly on one group than another and cannot be justified by business necessity." Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 335-36 n. 15, 97 *1054 S.Ct. 1843, 1854 n. 15, 52 L. Ed. 2d 396 (1977).

Disparate impact analysis was originally applied to objective employment criteria such as test score, diploma, and height and weight requirements. See Griggs v. Duke Power Co., 401 U.S. 424, 430, 91 S. Ct. 849, 853, 28 L. Ed. 2d 158 (1971); Dothard v. Rawlinson, 433 U.S. 321, 97 S. Ct. 2720, 53 L. Ed. 2d 786 (1977). The Ninth Circuit has recently expanded the scope of the disparate impact theory, however:

We now hold that disparate impact analysis may be applied to challenge subjective employment practices or criteria provided the plaintiffs have proved a causal connection between those practices and the demonstrated impact on members of a protected class. The three elements of the plaintiffs' prima facie case are that they must (1) show a significant disparate impact on a protected class, (2) identify specific employment practices or selection criteria and (3) show the causal relationship between the identified practices and the impact.

Atonio v. Wards Cove Packing Co., 810 F.2d 1477, 1482 (9th Cir.1987). Atonio allows for the application of disparate impact analysis to hiring and promotion practices lacking well-defined criteria. Id. at 1481-82.

Plaintiffs have presented statistical evidence of systematic wage disparities between men and women in state employment sufficient to allow them to survive summary judgment on the first element of the Atonio test. Plaintiffs have also identified the following specific employment practices as potentially discriminatory: exercising discretion to hire new employees at salaries above the minimum of the salary range for the position more often in favor of men than women; using predominately male classes as the source of promotions to higher paying positions to the exclusion of predominately female classes; failing to publicize promotion and transfer opportunities; and providing more training opportunities to male-dominated than to female-dominated classifications.

Plaintiffs' evidence thus far is insufficient to permit an inference that the challenged practices contributed to wage disparities between particular groups of male and female employees. Plaintiffs have failed to trace precisely how the identified practices as applied to specific categories of women result in the women being paid less than men who were similarly situated before the practice was applied. Thus, at this stage plaintiffs have not met Atonio's third requirement of proof of causation.

Plaintiffs have, however, raised factual issues regarding causation through declarations stating that the challenged practices have restricted women's promotion opportunities. Plaintiffs' showing on the first two Atonio elements, combined with those declarations, is sufficient to allow plaintiffs to survive a summary judgment motion. The issue of causation will require a highly fact-based inquiry and is therefore inappropriate for summary adjudication. Whether plaintiffs will be able to link more precisely particular employment practices to discriminatory salary disparities between well-defined groups of employees is a question more appropriate for resolution at trial.

Plaintiffs' evidence raises significant issues regarding facts material to their claims under Title VII on the theories of both disparate treatment and disparate impact. Accordingly, defendant's motion for summary judgment is denied.

IT IS SO ORDERED.