Opinion
No. 85-6907 DWW (Tx).
November 6, 1987.
Marc Coleman, Middleton Coleman, San Pedro, Cal., Janis Galef, Oceanside, Cal., for plaintiffs.
John H. Stephens, Donald Conway, Cox, Castle Nicholson, Los Angeles, Cal., for defendant, Plant Field Service Corp.
ORDER GRANTING and DENYING SUMMARY JUDGMENT
This is a class action suit for sex discrimination in employment brought under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. Four of the plaintiffs also allege individual claims stemming from their unsuccessful attempts to obtain employment with defendant Plant Field.
On October 22, 1986, the court granted plaintiffs' request for class certification and certified the class of plaintiffs as follows:
The request was granted pursuant to Rule 23(a) of the Federal Rules of Civil Procedure, in that the plaintiffs satisfy the four articulated prerequisites for class status: numerosity, commonality, typicality, and adequacy of representation.
All women who are past (since July 15, 1983) or present applicants for hire or who were since July 15, 1983 and up to the date of this order deterred from applying for hire into laborer, helper or crafts positions with Plant Field Service Corporation.
As to their Title VII claims, plaintiffs allege in their complaint that they are now suffering and will continue to suffer irreparable injury from defendant's policies, practices, customs, and usages. They therefore pray this Court to:
(a) permanently enjoin the defendant, its agents, successors, officers, employees, attorneys and those acting in concert with it or them from engaging in each of the unlawful practices, policies, customs and usages set forth in the Complaint
(b) to order modification or elimination of practices, policies, customs and usages set forth in the Complainy and all other such practices shown to be in violation of applicable law so that they do not discriminate on the basis of sex;
(c) immediately assign plaintiffs and the class they represent to those jobs that they would now be occupying but for the discriminatory practices of defendant;
(d) to adjust the wage rates, salaries, bonuses, and benefits for plaintiffs and the class they represent to that level which they would be enjoying but for the discriminatory practices of defendant;
(e) to compensate and make whole plaintiffs and the class they represent for all earnings, wages, and other benefits they would have received but for the discriminatory practices of the defendant;
(f) and award plaintiffs and the class they represent the costs and disbursements of this action, and reasonable attorney's fees.
As to their individual claims for violation of Government Code Section 12940, plaintiffs pray that this Court award them:
(a) actual damages in an amount subject to proof at trial;
(b) compensatory damages in the amount of $200,000.00; and punitive damages in the amount of $200,000.00.
Both parties assert that there are no material factual issues and that this matter is ripe for summary judgment. The proof for the class claim consists of statistical computations compiled by counsel and an expert retained by each party. What plaintiffs intend to show by their expert's calculations is that the percentage of female employees working at Plant Field is well below what it would normally be, and that this result is due to the improper hiring practices employed by the defendant. In contrast, Plant Field's analysis of the statistics are intended to show that its hiring practices do not discriminate and that its balance of male to female employees is within an expected statistical range.
FACTS
Plant Field provides a cleaning service to oil refineries in the Los Angeles and Orange County areas. Much of this work is termed "shutdowns" in which the refineries are shutdown to be cleaned. Independent companies such as defendant are hired to do this work. Plant Field must then hire a large number of workers to perform the cleaning operation, including crafts persons (pipefitters, millwrights and carpenters), helpers and laborers (usually unskilled assistants).
Plant Field is not a union facility. It hires through its personnel coordinator who has unfettered discretion in hiring. There are no minimum job qualifications for laborers; helpers are semi-skilled, and crafts persons are skilled.
Employees are hired in two ways. First, someone may be referred to Plant Field by the refinery that is being cleaned or a former employee may be rehired. The second way is that applications are accepted from potential new employees by word-of-mouth referral and, if needed, advertisements are placed in the classified ads of local newspapers. New applications remain active for a period of six months during which time applicants are instructed to remain in contact with Plant Field by calling regularly to inquire about openings. When hiring is to be done, the personnel coordinator reviews pending applications and selects the person best suited for the job classification.
The individual plaintiffs are four females who applied for employment with Plant Field on or about February 5, 1984, three of whom had previously applied for positions. They all applied for helper positions and two of them also applied for crafts positions as pipefitters. Plaintiffs had experience in the positions they sought, either working in similar clean-up operations or for refineries. They had heard about Plant Field's hiring for an upcoming shutdown only a couple of days before submitting their applications and were instructed to call regularly thereafter to see if work was available. The application form completed by each plaintiff states that it would remain active for a period of six months. Plaintiffs, however, were never hired. The plaintiffs have evidence that positions were filled by Plant Field with other new hires during the six month period following the date of their application. A large majority of these new hires were male.
NATURE OF THE MOTION
Both parties move for summary judgment on the contention that there are no genuine issues of material fact which require a trial. Each attempts to prove its claims with an analysis of the statistics and supporting case law. Defendant attempts to prove that its process of recruiting new employees and its actual hiring practices are not unexpected but are within range as far as the balance of male to female employees is concerned. Plaintiffs, on the other hand, claim the statistics show the converse and attempt to make out a prima facie case of discrimination.
DEFENDANT'S CONTENTIONS
I. Plaintiffs have failed to make out a prima facie case under a disparate impact theory.
Defendant claims that the theory of disparate impact is designed to focus on the effects of a specific employment practice, neutral on its face, but alleged to be discriminatory in operation, a criterion that is absent in the instant case. Defendants charge that it is plaintiffs' inability to cite a specific, identifiable employment evaluation mechanism that is offensive in operation which precludes application of disparate impact analysis in the instant case. Defendants claim that plaintiffs cannot bring the facts in the instant case within the cases that hold that subjectivity in hiring is suspect because the facts are not comparable.
Defendant cites Domingo v. New England Fish Co., 727 F.2d 1429 (9th Cir. 1984), in which "channel hiring" was held suspect when the employer hired for each job department through totally separate channels, which is not the same as the case at issue. Similarly, the court in Atonio v. Wards Cove Packing Co., 810 F.2d 1477 (9th Cir. 1987), included the identification of "specific employment practices" in a list of requirements for a prima facie case of discrimination. Even if plaintiff asserts the word-of-mouth theory as a specific act, defendant claims that such a theory, standing alone, will not be sufficient.
II. Plaintiffs' class action claim must fail because their statistical "proof" is inadequate to show discriminatory motive.
Plant Field contends that plaintiffs cannot make a prima facie case of discrimination because their statistics do not reflect meaningful results of discrimination. Defendant argues that they instead show that an overwhelming majority of those persons interested in engaging in this form of employment are male. Any use of general population statistics by plaintiffs is meaningless because of this. Instead, the use of qualified labor market or applicant flow statistics is called for; the statistics used by defendants to successfully demonstrate the absence of discrimination.
a. Plaintiffs' statistics are inaccurate and unreliable.
1. The proper standard for statistical comparison is actual applicant flow. It cannot be disregarded based on the unfounded suggestion that the industry "grapevine" did not reach women.
Defendants claim that the only appropriate method of asserting discrimination is to show that an employment practice selects members of a protected class in a proportion smaller than their percentage in the pool of actual applicants. Moore v. Hughes Helicopters, Inc., 708 F.2d 475, 482 (9th Cir. 1983). By suggesting a proxy pool or any other group (other than union members), plaintiff is misjudging the use of statistics in this application. Defendants claim that the use of applicant flow data disproves discrimination.
2. Comparison with the actual applicant pool shows no gross disparity.
Defendant claims that because plaintiff cannot make out a statistical showing of discrimination from the actual applicant pool, they fail to prove their case. According to the defendant, the use of a proxy pool as suggested by plaintiff is inappropriate and misleading.
b. Comparison of defendant's work force with the relevant labor market shows an absence of discriminatory results.
Defendant did a study of the four individual plaintiffs, who are members of Oil Chemical and Atomic Workers Union ("OCAW"), comparing them with other women in the union. The study showed a statistical disparity of 2-3%. Defendant charges that while these statistics are not conclusive, they are more convincing than plaintiffs' analysis which uses the incorrect pool, and they are not convincing support of a charge of discrimination.
III. The individual plaintiffs cannot state prima facie claims under the four part McDonnell Douglas standard because two elements are missing, job opening and continuing search.
Defendant claims that plaintiffs cannot show a prima facie case for their individual claims because two of the necessary elements, job availability and the employer's continuing search to fill positions, required by McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), are absent here. Plant Field claims that on the date plaintiffs applied for employment, defendant had already filled all positions for its Champlin Oil shutdown and was no longer hiring.
IV. Plaintiffs fail to identify a specific employment practice causing a discriminatory impact.
a. Evidence of "sexual stereotyping" does not exist.
Defendant claims that it has not presented itself as having preconceived notions about women or their mental or physical abilities. It maintains that the notions outlined in its Motion for Summary Judgment were simply an expression of its desire not to be held liable for historical disparities endemic to the labor pool from which it hires. Penk v. Oregon State Board of Higher Education, 816 F.2d 458 (9th Cir. 1987).
b. "Word of Mouth Recruitment" is not a challengeable practice of Plant Field's; It is an uncontrolled and nondiscriminatory operation of the industry-wide labor market.
The defendant maintains that the fact that word of employment is spread by mouth is not a challengeable "practice" and the cases presented by plaintiff are distinguishable in that they present practices that kept potential applicants ignorant or discouraged, which is not the situation in the instant case. Also, defendant charges plaintiff with producing evidence that the channels of communication involved are not an informal, industry-wide network which helps women as well as men.
c. Plaintiffs fail to present a less restrictive alternative to the use of word of mouth.
Defendants contend that plaintiffs cannot show a less restrictive alternative to word-of-mouth recruitment because they cannot show a causal link between the practice and the women who allegedly do not receive information, making it impossible to replace that communication with another. As well, defendant points out that the industry-wide grapevine continues to work regardless of what Plant Field does. How defendant could continue to compete in a industry in which it could not participate is questionable, and not addressed by the plaintiffs.
PLAINTIFFS' CONTENTIONS
I. Plant Field's hiring and recruitment practices are illegal.
a. Subjective hiring system.
Plaintiffs claim that Plant Field uses no objective criteria in determining who is hired. Although there is at least one test given (used to determine the qualifications of applicants for the "pipefitter" positions), plaintiffs maintain that the tests are not graded in such a way that they independently determine whether an applicant is qualified. Plaintiffs contend further that questioning conducted during the brief interview by the Personnel Coordinator occurs only sporadically, as time permits, and that applications are accepted without regard to the response or the absence thereof.
b. Sexual stereotyping.
Plaintiffs charge that the defendant has a stereotyped view of the physical and mental abilities of women, and refuses to hire them (based on this preconceived prejudice) because the jobs it offers are dirty and dangerous. Plaintiffs charge that Plant Field's misconceptions of women's abilities and its failure to consider the obvious qualifications of named plaintiffs is "the kind of invidious discrimination that violates Title VII." Plaintiffs allege that the defendant intentionally refuses to hire qualified female applicants, and has acquired such a reputation accordingly.
See Plaintiffs' Memoranda of Points and Authorities, p. 13, quoting Thorne v. City of El Segundo, 726 F.2d 459, 468 (9th Cir. 1983) (refusal to hire woman because of sex-stereotyped view of her physical abilities is the kind of invidious discrimination that violates Title VII.)
c. Word of mouth recruitment.
Plaintiffs contend that Plant Field's hiring practices are illegal because word-of-mouth recruiting acts to perpetuate characteristics of the existing work force, thereby limiting opportunities for prospective female applicants. As well, plaintiffs claim that a consistently enforced discriminatory employment policy will deter applications from those who are aware of the jobs but are unwilling to subject themselves to certain rejection.
Domingo v. New England Fish Co., 727 F.2d 1429 (9th Cir. 1984).
Atonio v. Wards Packing Co., Inc., 810 F.2d 1477 (9th Cir. 1987).
International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977).
Plaintiffs claim that there is evidence there exists a reputation for not hiring females which caused at least one prospective applicant not to apply. Plaintiffs take issue with the defendant's hiring process and its practice of seeking new applicants by word-of-mouth referrals. Plaintiffs allege that this procedure maintains a work force of more than 95% male employees because it limits dissemination of job availabilities to potential female applicants.
II. Impact of Plant Field's practices.
a. Disparate impact case.
Plaintiffs disagree with the idea that only objective practices can be analyzed for disparate impact. Plaintiff cites Atonio, supra, which involved word-of-mouth recruitment, rehiring of members of past work forces, and channel hiring. In Atonio, the Ninth Circuit scrutinized the lack of objective qualifications and the resultant use of subjective criteria in hiring (and promoting). Similarly, plaintiffs claim that hiring decisions are made by the personnel coordinator without objective standards. According to plaintiffs, the coordinator simply picks out the person he considers to be best for the job. Although plaintiffs assert that a lack of objective criteria would be suspect by itself, there are specific employment practices that can be challenged directly, and those practices have lead to disparate impact on a protected class.
b. Disparate treatment case.
Plaintiffs claim that they meet the McDonnell Douglas test for disparate treatment for their individual claims in that they (1) are members of a protected of a protected class; (2) applied and qualified for jobs that defendant was seeking to fill; (3) were rejected; and (4) defendant continued to seek applicants after rejecting them. Plaintiffs assert that they further show disparate treatment through their use of statistical analysis and testimony.
411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973).
III. Statistical proof.
Plaintiffs claim that they have used the correct statistical pool for analysis of Plant Field's hiring practices. Plaintiffs suggest the comparison of the employer's work force to the general population since the jobs involved are unskilled or involve skills which many people possess or can easily learn. For the more skilled jobs, plaintiffs suggest that a proxy pool of industry-wide figures or non-union people who signed up at the OCAW Union Hall to work as laborers, helpers, and craftpersons.
Plaintiffs contend that the use of defendant's actual applicant data is not appropriate because of the underrepresentation of female applicants. It is also not appropriate to use union data, since Plant Field is a non-union employer.
Plaintiffs offer a statistical expert, Dr. Kirkpatrick, who found disparities between Plant Field's hiring of females and the percentage of females in the labor pool, as well as between Plant Field's hiring and the percentage of females found in OCAW's records. The plaintiffs determined the percentage of female applicants who sought employment through the OCAW for employment in petroleum refining and related industries to be 10.09%. Plaintiffs next determined the percentage of females employed by Plant Field for each of the years 1983 (from July 15th), 1984, and 1985. These numbers are (after correction) 1.17%, 1.81%, and 2.17%, respectively. Plaintiffs then prepared a statistical analysis comparing the OCAW percentage to each of the Plant Field percentages, resulting in the following standard deviations for each of the years 1983 through 1985: 5.44, 6.08 and 6.39. All three represent probabilities exceeding one-in-a-million chances.
Plaintiffs also prepared an analysis using general population and found even more outrageous standard deviations. According to plaintiffs, they thus satisfied their prima facie case, which the defendant has not rebutted.
DISCUSSION
I. Traditional Proof.
Traditional proof of discrimination begins with the determination of a disparity between the minority group's representation in a relevant population and that group's representation in the particular position under scrutiny. The demonstration of a percentage difference between the two is sufficient to constitute a violation of Title VII, even absent an explanation in terms of differing job-related abilities. The law initially presumes that no such disparity exists.
"Statistics and the Law" 32 Hastings Law Journal 67 (September 1980).
II. Griggs v. Duke Power
In the landmark case of Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), the Court stated that employment procedures that result in unequal representation are not saved by an employer's "good intentions or absence of discriminatory intent." Id. at 432, 91 S.Ct. at 854. It is the inequality itself that raises an inference of discrimination. Plaintiff need not establish causation for relief; plaintiff's case may be shown by percentages alone.
Under Griggs, the employer must show job-relatedness, which the plaintiff may counter by showing that other, less discriminatory means are available to serve the same purpose. When actions are brought privately or on behalf of a class, some courts formerly added an additional component to the plaintiff's prima facie case: the demonstration of at least one specific act of discrimination against plaintiff. McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668. (1973). Subsequent cases have acknowledged the likelihood of futility of such applications and have not required them. Currently, it appears that demonstration of significant percentage differences that infer bias are sufficient, even absent direct evidence.
Griggs established the following:
1. neutral employment qualifications will be judged by their impact, and not by the good faith in which they were instituted.
2. once a discriminatory impact is established the employer carries a burden of establishing that the qualifications are justified by objective proof of "business necessity."
3. "Business necessity," in turn, will be evaluated not in general or conclusory terms, but by specific reference to the employee's ability to perform a particular job.
Rodriquez v. East Tex. Motor Freight, 505 F.2d 40 (5th Cir. 1974) (statistics and statistics alone may be enough); Babrocky v. Jewel Food, 773 F.2d 857 (7th Cir. 1985) (a plaintiff may establish a prima facie case with statistics alone).
III. Disproportionate Impact.
Courts have recognized the evidence of disproportionate impact of employment practices as a signal to look for its causes; the application of legitimate neutral factors, the application of an illegitimate factor or policy, or just chance.
Factors that bear on the inference of discriminatory purpose are the nature and degree of disproportionate impact; the foreseeability and desirability of the disproportionate impact; timing; and the inherent plausibility of the alleged basis of the defendant's action. The disparate impact model prohibits the use of selection criteria and procedures which substantially burden protected group members unless they can be justified in a manner that would not be required in the absence of the unequal results.
David C. Baldus and James W.L. Cole, Statistical Proof of Discrimination, McGraw-Hill Company: New York, 1980.
Proof of a substantial adverse impact triggers a demand for a justification, and liability follows if a sufficient justification is not forthcoming. There have been cases where the proof of disparate impact failed for a lack of proof that the disproportionate selection rates were caused by the scores of the test administered. As the Supreme Court noted in Dothard v. Rawlinson, 433 U.S. 321, 97 S.Ct. 2720, 53 L.Ed.2d 786 (1977):
See Cooper and Sobol, "Seniority and Testing Under Fair Employment Laws: A General Approach to Objective Criteria of Hiring and Pro. motion," 82 Harv.L.Rev. 1598 (1969) and Shober, "Probing the Discriminatory Effects of Employee Selection Procedures With Disparate Impact Analysis Under Title VII," 56 Tex.L.Rev. 1 (1977).
Lee v. City of Richmond, 456 F. Supp. 756 (E.D.Va. 1978); Hester v. Southern Railway Co., 497 F.2d 1374 (5th Cir. 1974).
There is no requirement that a statistical showing of disproportionate impact must always be based on analysis of the characteristics of actual applicants. The application process might itself not adequately reflect the actual potential applicant pool, since otherwise qualified people might be discouraged from applying because of a self-recognized inability to meet the very standards challenged as being discriminatory. Id. at 330, 97 S.Ct. at 2727, citing Griggs, 401 U.S. at 430, 91 S.Ct. at 853.
"To establish a prima facie case of discrimination, a plaintiff need only show that the facially neutral standards in question select applicants for hire in a discriminatory pattern. Once it is shown that the employment standards are discriminatory in effect, the employer must meet the burden of showing that any given requirement has a manifest relationship to the employment in question." Dothard v. Rawlinson, 433 U.S. 321, 329, 97 S.Ct. 2720, 2726-27, 53 L.Ed.2d 786 (1977) (emphasis added).
Griggs and its progeny have established a three-prong analysis for disparate impact claims. To establish a prima facie case of discrimination, a plaintiff must show that the facially neutral employment practice had a significantly discriminatory impact. If that showing is made, the employer must then demonstrate that "any given requirement has a manifest relationship to the employment in question." The plaintiff still may prevail, if it can be shown that the employer was using the practice as a pretext or if less discriminatory methods for achieving the same objective can be shown. There is no "inflexible formula" of what constitutes a prima facie case; it varies with respect to differing factual situations. Id. 431 U.S. at 358, 97 S.Ct. at 1866.
See also McDonnell Douglas v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973) and Williams v. Anderson, 562 F.2d 1081 (8th Cir. 1977).
IV. The Uniform Guidelines.
The Code of Federal Regulations generally endorses a broad version of the theory of disparate impact, indeed in some respects one broader than that endorsed by the Supreme Court in Griggs and Albermarle Paper. On the issue of pretext, Albemarle Paper clearly assigns the burden of production and persuasion to the plaintiff, for instance, by requiring proof that a validated employment practice has a greater adverse impact than some equally valid alternative. 422 U.S. at 436, 95 S.Ct. at 2380-81. The uniform guidelines, however, impose upon the employer the burden of proving that a validated employment practice has the least adverse impact among available alternatives. 29 C.F.R. See. 1607.3(B). This provision of the uniform guidelines is indicative of the general disapproval of any employment practice that has an adverse impact.
422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975).
In Pouncy v. Prudential Insurance Company, 668 F.2d 795, (5th Cir. 1982), the Fifth Circuit held that the theory of disparate impact does not apply (a) to subjective employment practices or (b) to a broad attack on the cumulative effect of several employment practices. The first holding in Pouncy has given rise to conflicting decisions. Several circuits have agreed with the Fifth Circuit, yet others have reserved decision, are internally divided, or have followed previous cases that reached the opposite result.
Talley v. United States Postal Service, 720 F.2d 505, 507 (8th Cir. 1983); cert. den., 466 U.S. 952, 104 S.Ct. 2155, 80 L.Ed.2d 541 (1984); EEOC v. Federal Reserve Bank, 698 F.2d 633, 639 (4th Cir. 1983), rev'd on other grounds sub. nom Cooper v. Federal Reserve Bank, 467 U.S. 867, 104 S.Ct. 2794, 81 L.Ed.2d 718 (1984).
Robinson v. Polaroid Corp., 732 F.2d 1010, 1015 (1st Cir. 1984); Eastland v. Tennessee Valley Authority, 704 F.2d 613, 619-20 (11th Cir. 1983), cert. denied, 465 U.S. 1066, 104 S.Ct. 1415, 79 L.Ed.2d 741 (1984).
Moore v. Hughes Helicopters, Inc., 708 F.2d 475, 481-82 (9th Cir. 1983).
Segar v. Smith, 738 F.2d 1249, 1270-72 (D.C. Cir. 1984); Rowe v. Cleveland Pneumatic Co., 690 F.2d 88, 95 (6th Cir. 1982); Bauer v. Bailar, 647 F.2d 1037, 1042-43 (10th Cir. 1981); Williams v. Colorado Springs, Colorado School District No. 11, 641 F.2d 835, 839-40 (10th Cir. 1981); Whack v. Peabody Wind Eng'g Co., 595 F.2d 190, 194 (3d Cir. 1979).
The argument against application of the theory of disparate impact is that subjective employment practices are difficult to validate, so that proof of adverse impact would, as practical matter, establish a violation of Title VII, forcing employers to abandon subjective employment practices or to establish preferences. The argument for application of the theory, however, is equally powerful. Subjective employment practices are likely to hide discrimination that can be eliminated by requiring proof of business necessity or job relationship. Moreover, the Supreme Court itself has stated, although only in a footnote, that "[e]ither theory may, of course, be applied to a particular set of facts."
International Board of Teamsters v. United States, 431 U.S. 324, 335 n. 15, 97 S.Ct. 1843, 1854-55 n. 15, 52 L.Ed.2d 396 (1977).
The second holding in Pouncy, that the theory of disparate impact does not apply to the cumulative effect of several employment practices, has been more widely accepted. It is also more easily justified. The plaintiff must give the defendant notice of the employment practices that require validation. Otherwise, the defendant cannot concentrate its resources on defending particular employment practices and, in particular, on showing that they are justified by job relationship or business necessity.
Spaulding v. University of Washington, 740 F.2d 686, 707 (9th Cir. 1984), cert. denied, 469 U.S. 1036, 105 S.Ct. 511, 83 L.Ed.2d 401; Robinson v. Polaroid Corp., 732 F.2d 1010, 1014 (1st Cir. 1984); Zahorik v. Cornell Univ., 729 F.2d 85, 95 (2d Cir. 1984); Mortenson v. Callaway, 672 F.2d 822, 824 (10th Cir. 1982).
V. The Use of Statistics.
In Barnett v. W.T. Grant Company, 518 F.2d 543 (4th Cir. 1975), the Court of Appeals held that the District Court erred in requiring proof of actual discrimination in addition to the statistical data implying discrimination. The Court held that statistics can, in appropriate cases, establish a prima facie case of discrimination without the showing of specific instances of overt discrimination. The Court relied on statistics that it admitted were not overwhelming, but simply suggestive. The Court also stated that "in addition to statistics, a court must also examine `patterns, practices and general policies' to ascertain whether racial discrimination exists."
See Long v. Sapp, 502 F.2d 34 (5th Cir. 1974); United States v. Chesapeake Ohio Ry., 471 F.2d 582 (4th Cir. 1972).
The use of statistical analysis has been approved in various cases to show disparate impact. The purpose of statistical analysis is to determine whether the effects of the defendant's actions were random and within the range of expectation or probability, or whether defendant's acts produced results, the probability of which was unlikely absent intentional conduct. The test used compares one survey result with another to determine their relationship in terms of probability or standard deviations. If the difference between the expected value and the observed result is greater than two or three standard deviations, it is not likely the result of chance.
See Dothard v. Rawlinson, supra; and Atonio v. Wards Cove Packing Co., supra.
The general procedure for calculating statistical disparities between results for minority and majority groups (the "2 or 3 standard deviation" rule, which permits an assessment of statistical significance in terms of a ratio of the estimated disparity to its standard deviation) was first presented in Castenada v. Partida or Hazelwood School District v. United States (a discrimination disparate treatment case), and further developed in Hazelwood where the court stated that in cases involving large samples, "if the difference between the expected value and the observed number is greater than 2 or 3 standard deviations," the hypothesis (that teachers were hired without regard to race) would be suspect. The Supreme Court gave the rule its imprimatur in Whitus v. Georgia, 385 U.S. 545, 87 S.Ct. 643, 17 L.Ed.2d 599 (1966), and it has been applied frequently since.
430 U.S. 482, 97 S.Ct. 1272, 51 L.Ed.2d 498 (1977).
433 U.S. 299, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977).
The court has left the probative force of statistical evidence to be established in each case. This depends not only on the accuracy of empirical data available and the validity of statistical methods used, but also on the soundness of the economic model presupposed by the statistical methods and the relevance of that model to the claims before the court. A corollary of the court's refusal to impose hard-and-fast requirements upon statistical proof is its refusal to specify the elements of a plaintiff's prima facie case in class claims of disparate treatment.
Similarly, the Supreme Court has steadily discounted the significance of its early decision in McDonnell Douglas for individual individual claims of disparate treatment. This development has called into question the scope and effect of the structure of burdens of production established in McDonnell Douglas. The Court has also remained ambivalent between two interpretations of the theory of disparate impact — a narrow interpretation that substitutes the objective issues of adverse impact and job relationship for the subjective issue of intentional discrimination, and a broad interpretation that requires equal employment of different groups in the absence of proof of business necessity. Adoption of the broader interpretation would, in effect, require affirmative action in the absence of business necessity and, in most cases, it would make statistical proof decisive on the issue of discrimination.
Kaye, D.H. and Mikel Aikin, Statistical Methods in Discrimination Litigation, Marcel Dekker, Inc.: New York, 1983.
In order to carry out a thorough statistical analysis, the labor market for the jobs at issue must be defined. After that, the proportion of a particular group among those in the labor market and the proportion among those who possess the disputed qualification must be established. Finally, the two proportions must be compared by statistical means to determine the probability that any difference between them resulted solely by chance.
VI. Title VII.
In an employment discrimination action under Title VII, gross statistical disparities between the composition of the employer's work force and that of the general population may, in a proper case, alone constitute prima facie proof of a pattern or practice of discrimination. Hazelwood, supra. In Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396, the Supreme Court, in considering the role of statistics in "pattern-or-practice" suits under Title VII, noted that
The situation in which a particular "pattern" adhered to or a "practice" engaged in is the basis of the action.
absent explanation, it is ordinarily to be expected that nondiscriminatory hiring practices will in time result in a work force more or less representative of the racial and ethnic composition of the population in the community from which employees are hired. Evidence of long-lasting and gross disparities between the composition of a work force and that of the general population thus may be significant even though Sec. 703(j) makes clear that Title VII imposes no requirement that a work force mirror the general population. Id. at 340, 97 S.Ct. at 1857.
The obligation imposed by Title VII of the Civil Rights Act is to provide an equal opportunity for each applicant regardless of class membership, without regard to whether members of the applicant's class are already proportionately represented in the work force. Connecticut v. Teal, 457 U.S. 440, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982).
Discriminatory intent is not prerequisite to success of all Title VII claims. In order to succeed on a disparate impact claim, plaintiff need only demonstrate the lack of objective criteria and a disparity in hiring or promotions. Plaintiff need not prove that he would have been the most qualified for the job. Hung Ping Wang v. Hoffman, 694 F.2d 1146 (9th Cir. 1982).
In Gay v. Waiters' and Dairy Lunchmen's Union, Local No. 30, 694 F.2d 531 (9th Cir. 1982), the court held that a disparate impact case under Title VII may be established without any proof of intentional discrimination; where business practice, neutral on its face, is shown to have a substantial, adverse impact on some group protected by Title VII. Good intent or absence of intent is no defense to disparate impact Title VII claims. A prima facie case of disparate treatment under Title VII is established by the proof of facts supporting an inference of intentional discrimination.
Illicit motive may, however, be inferred in a classwide employment discrimination claim from a sufficient showing of disparity between class members and comparably qualified members of the majority group. If the employer defends an employment discrimination action by explaining that there is a reason for the disparity in treatment of whites and nonwhites, he must prove the necessity. Potential applicants may be discouraged by an employer's prior announcement of qualifications such as height, education, or experience requirements, which may not be valid requirements. Moore v. Hughes Helicopter, 708 F.2d 475 (9th Cir. 1983). Similarly, potential applicants may be discouraged from applying because of the employer's reputation for discrimination or because of discouragement upon initial inquiry before actually filling out an application. International Brotherhood of Teamsters, 431 U.S. 324, 365-366, 97 S.Ct. 1843, 1869-70, 52 L.Ed.2d 396 (1977).
VII. Word-of-Mouth Recruitment.
The use of referrals from current employees (as a source of new hires) which tends to perpetuate a low percentage of a disadvantaged group may violate Title VII if it is the employer's primary means for recruiting applicants. Relying on word-of mouth or walk in applicants when the only applicants likely to walk are members of the majority group amounts to unlawful discrimination.
EEOC Decision No. 70-62 (1969) CCH Employment Practices Guide par. 6048.
In Markey v. Tenneco Oil Co., 707 F.2d 172 (5th Cir. 1983), a class action race discrimination case alleging discriminatory exclusion of blacks in an entry level position which did not require special skills, the plaintiffs argued that Tenneco's word-of-mouth recruitment method was inherently discriminatory with the predominantly white overall work force. This argument failed in part because the court found that 40% of the recent applicants were black. Additionally, the court reasoned that other recruitment methods must have been operating because the record suggested that half of those recently hired knew no one at the refinery before employment, and of those who did, half were black. Therefore is was appropriate to examine the applicant flow, and the applicant flow showed no discrimination in recent hiring.
VII. Applicant Flow Data.
In the absence of reasonable applicant flow data, courts tend to turn to information on the general population of workers. The approach is that if the pool of applicants does not reveal the composition of those who would want the job, it should be taken to be the same as the composition of the general labor force. This population is then compared to the employer's work force, or to recent hires. This approach may be too simplistic. Not all individuals are interested in or qualified for all jobs. There must be an adjustment at least for inherently "obvious" requirements, and for some element of self-selection.
The court has observed that in some cases it is the defendant's burden to establish that general population statistics are inappropriate because of special qualifications required for the job. If the defendant succeeds in showing that the plaintiff should not have used general population statistics, the plaintiff then needs an opportunity to adjust the statistical proof to reflect the change. EEOC v. Radiator Specialty Co., 610 F.2d 178 (4th Cir. 1979).
The court identified three categories in which the allocation of the appropriate labor pool is uncomplicated:
(1) those cases where it is manifest as a matter of law that there are no special skills (making general population statistics appropriate);
(2) special qualifications are manifest as a matter of law (requiring the plaintiff to use the qualified labor market);
(3) those cases where it is not clear whether a job requires special qualifications (requiring the defendant to establish that the general population statistics are inappropriate).
The Supreme Court addressed the problem of limiting the population figures to those who possess special skills in Hazelwood, supra. The court held that a general population comparison is appropriate for a job that requires skills generally possessed or easily acquired by members of the general population. For a job that requires special skills, the relevant population is those who possess the special skills. Unfortunately, it is often difficult to define which skills are so patently prerequisite for a job such that a specialized labor market should be defined.
433 U.S. at 308 n. 13, 97 S.Ct. at 2742 n. 13.
IX. Self Selection.
Self selection of individuals away from particular jobs poses another difficulty with general population figures. Although employers may not refuse to consider individual applicants because of a stereotyped assumption about groups, many qualified members of those groups may prefer to avoid some jobs. When applicant flow is unreliable, it is particularly difficult to account for such self-selection. [One common method is to use census figures. Such figures give the representation of women, for example, in a particular job category in a limited geographic region. Low representation may reflect historical discrimination against women in that job group, but it may also reflect self-selection.]
X. Rule 23(b)(2) of the Federal Rules.
Rule 23(b)(2) of the Federal Rules of Civil Procedure states in pertinent part that:
An action may be maintained as a class action if the prerequisites of subdivision (a) are satisfied, and in addition:
(1) the prosecution of separate actions by or against individual members of the class would create a risk of
(a) inconsistent or varying adjudications with respect to individual members of the class which would establish incompatible standards of conduct for the party opposing the class, or
(b) adjudications with respect to individual members of the class which would as a practical matter be dispositive of the interests of the other members not parties to the adjudications or substantially impair or impede their ability to protect their interests; or
(2) the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole; or
(3) the court finds the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy.
Although the order granting class certification does not expressly state so, it is completely obvious that class certification was granted pursuant to Rule 23(b)(2). Title VII cases have generally been categorized as being of the subdivision (b)(2) type. 15 Am.Jur.2d Sec. 393. The Advisory Committee on the 1966 Amendments to the Rules specifically used the example of actions in the civil rights field, where a party is charged with discriminating unlawfully against a class, as illustrative of the use to which subdivision (b)(2) should be put. Most class actions in employment discrimination litigation seek injunctive or declaratory relief. The aptness of designating employment discrimination suits as a class actions under Rule 23(b)(2) has been similarly recognized by the Ninth Circuit. Gibson v. Local 40, Supercargoes Checkers, 543 F.2d 1259 (9th Cir. 1976); Penk v. Oregon State Board of Higher Education 93 F.R.D. 45 (D.Or. 1981).
See Order For Class Certification:
"Plaintiffs' primary objective is injunctive relief" (p. 1, line 26, 27).
The court certified the class as a whole by definition without requiring individual notice to the class members (p. 3, line 15-24).
Unlike what would have occurred in a (b)(3) action, class members were not notified and given an opportunity to "opt out" of the suit. Rule 23(c)(2), Federal Rules of Civil Procedure; Paddison v. Fidelity Bank, 60 F.R.D. 695 (D.C. Pa.).
Rule 23, Federal Rules of Civil Procedure, "Notes of Advisory Committee on 1966 Amendments to Rules."
When a Title VII suit is eligible for treatment under subdivision (b)(3) of Rule 23 it is because a plaintiff's allegations of discriminatory treatment may present questions of law or fact common to the members of the class which predominate over any questions affecting only individual members. Branham v. General Electric Company, 63 F.R.D. 667 (D.C.Tenn.); Gerstle v. Continental Airlines, Inc., 50 F.R.D. 213 (D.C.Colo.). The standard for establishing a (b)(3) class is much higher than for a (b)(2) class: (1) class actions under (b)(2) have a wider res judicata effect; (2) under (b)(3), the court must direct individual notice to as many members of the class as is possible; and (3) class members in a (b)(3) action must be given the option to "opt out," or exclude themselves if they do not want the action maintained on their behalf. Paddison v. Fidelity Bank, 60 F.D.R. 695 (D.C.Pa.).
XI. Back Pay.
Back pay has generally been accepted as an appropriate remedy in class actions brought pursuant to Rule 23(b)(3) of the Federal Rules of Civil Procedure. Most Title VII class actions, however, are brought pursuant to Rule 23(b)(2), and the entitlement of the unnamed class members is governed by that rule. Rule 23(b)(2) speaks only of injunctive relief and declaratory relief, not monetary relief, and some courts have held that classwide back pay relief is not available in Title VII class actions brought under Rule 23(b)(2). One concern is the computation of the award of back pay.
Title VII specifies only that back pay liability accrue from a date not more than 2 years prior to the filing of a charge with the Commission, and that interim earnings or amounts earnable with reasonable diligence by the person or persons discriminated against must operate to reduce the back pay otherwise allowable. 42 U.S.C. § 2000e-5(g).
The major element of most back pay awards is salary. However, back pay has been held to include other elements such as overtime, shift differentials, sick pay, an allowance for vacation time which would have been earned, allowances for holidays, travel expenses, bonuses, raises, and interest.
Approximations of what the discriminated group might have earned can be based on the earnings of a group of employees, not injured by the discrimination, comparable in size, ability, and length of employment to the plaintiff or class member. Ordinarily, the starting date of the back pay period will be the date on which the unlawfully discriminatory conduct occurred, if the conduct consisted of a single definite act.
In 1972, Title VII was amended to provide that back pay liability shall not accrue from a date more than 2 years prior to the filing of a charge with the EEOC. No back pay award under Title VII may cover a period beginning earlier than the effective date of Title VII, July 2, 1965.
The end of the back pay period is ordinarily held to be the date of the court's order authorizing back pay, although sometimes a variation of this date is used. Termination dates have included the date on which a discriminatory practice was ended by the employer, the date on which a Title VII court complaint was filed, and the date of trial. A claimant's actions may further limit the back pay period. It has been held that the award of an individual who was discriminated against should end with his acceptance of interim employment. However, it has been held that the refusal of back pay claimants to accept a defendant's offer to rehire them did not operate to cut off their entitlement to back pay because the offer did not include back pay.
There is a "mitigation requirement" in Title VII that dictates that interim earnings or amounts earnable with reasonable diligence by the person or persons discriminated against shall operate to reduce the back pay otherwise allowable. Awards of back pay made in civil rights cases to applicants who were denied employment because of their race must be reduced by their actual earnings after the time of rejection, or amounts earnable with reasonable diligence. Civil Rights Act of 1964 Sec. 706(g) as amended 42 U.S.C. § 2000e-5(g). Employment Op. Com'n v. Detroit Edison Co., 515 F.2d 301 (6th Cir. 1975).
XII. Punitive or Compensatory Damages.
Civil Rights Act of 1964 provisions relating to discrimination in employment do not authorize the awarding of punitive or compensatory damages as such "other equitable relief" as the court deems appropriate. 42 U.S.C. § 2000e-5(g). Therefore, punitive damages have been awarded only rarely in Title VII cases. Since the relief provisions of Title VII do not specifically authorize an award of either compensatory or punitive damages for discrimination in employment practices, back pay in such cases is considered a form of restitution, not an award of damages.
XIII. Attorney's Fees.
Reasonable attorney's fees are expressly authorized for the prevailing parties in employment discrimination suits under Title VII of the Civil Rights Act of 1964. 42 U.S.C. § 2000e-5(k). An award of attorney's fees should fairly compensate the attorney for the reasonable value of the services beneficially rendered, based on the circumstances of the particular case.
Lindy Bros. Builders, Inc. v. American Radiator Standard Sanitary Corp., 487 F.2d 161 (3d Cir. 1973); 7A C. Wright and A. Miller, Federal Practice Procedure Sec. 1803.
The starting point for the determination of a fee should be a reasonable hourly rate multiplied by the number of hours reasonably expended on the case. Both the hourly rate and the number of hours expended should be shown by supporting evidence. A fee may be adjusted upward or downward based on additional considerations.
New York State Ass'n for Retarded Children, Inc. v. Carey, 711 F.2d 1136 (2d Cir. 1983) and National Ass'n of Concerned Veterans v. Secretary of Defense, 675 F.2d 1319 (D.C. Cir. 1982) (absent contemporaneous records, fee based on minimum time necessary).
Institutionalized Juveniles v. Secretary of Public Welfare, 758 F.2d 897 (3d Cir. 1985).
The United States Supreme Court has said that the public interest in having injunctive actions brought to eradicate discriminatory employment practices can be vindicated by awarding in "all but unusual circumstances" attorney's fees. The Court noted that if successful civil rights plaintiffs were routinely forced to bear their own attorneys' fees, few aggrieved parties would be in a position to advance the public interest by invoking the injunctive powers of the federal courts, and that Congress had therefore enacted provisions for counsel fees to encourage individuals injured by discrimination to seek judicial relief.
Albemarle Paper Co. v. Moody, 422 U.S. 405, 95 S.Ct. 2362, 45 L.Ed.2d 280.
XIV. Statistical Analysis.
The defendant's expert statistician, Dr. Anne S. Davis, examined the data presented by Dr. Kirkpatrick, who appeared as an expert for the plaintiffs. Dr. Davis recorded errors in Dr. Kirkpatrick's calculations and discrepancies in his analysis. For instance, Dr. Kirkpatrick reported that 9 women were hired out of 497 people hired in 1984, or 2.35%. He reported that in 1983 13 women were hired out of 600 people hired, or 2.9%. Dr. Davis corrects these percentages to 2.17% and 1.81% respectively.
The table for new hires is as follows:
1983 1984 1985 ---- ---- ----
Males 338 488 587 Females 4 9 13 Total 342 496 600 %Females 1.17 1.81 2.17
Similarly, Dr. Davis contests Dr. Kirkpatrick's use of the figure 10.09% as the percentage of women in the relevant labor pool. This figure was supposed to represent the percent of women on the OCAW sign-up sheets for 1983-1984, but close inspection of the correct counts cannot produce that number:
1983 1984 1985 ---- ---- ----
Males 101 22 130 Females 10 1 9 Total 111 23 139 %Females 9.01 4.35 6.4
The total OCAW referrals from the sign-up sheets and the OCAW referral cards are 6.42% (1983), 3.25% (1984), and 5.82% (1985). The general figure for female applicants presented by Dr. Davis is 3.9%, a figure that the plaintiffs do not dispute. Comparing the applicant flow with the relevant female labor market using the total OCAW referrals (which includes the referral cards) yields: Total OCAW Referrals
Percentage of Females in Total OCAW Referrals Compared to 3.9% Female Applicants
1983 1984 1985 ---- ---- ----
Standard Deviations 1.95 -.63 1.54
A negative standard deviation indicates over-representation of females in the applicant pool.
OCAW Referrals Compared to 2.96% Female Applicants
1983 1984 1985 ---- ---- ----- Standard Deviations 2.53 .43 1.35
Dr. Davis concluded that with respect to the years 1984 and 1985, there was no relationship between the sex of an applicant and whether the applicant was hired by Plant Field. With respect to the years 1984 and 1985. Dr. Davis concluded that there was no relationship between the sex of an individual in the relevant labor pool and whether or not they applied for a position. With respect to the years 1984 and 1985, Dr. Davis concluded that there was no relationship between the sex of an individual in the relevant labor market and whether or not they were hired by Plant Field.
With respect to the year 1983, Dr. Davis was unable to conclude whether there was a relationship between the sex of an individual in the relevant labor market and whether the individual applied to Plant Field and whether the individual was hired. Dr. Davis points out that plaintiff's data may not be limited to the certified class period for 1983, and the results yield inconsistent conclusions depending on the exact data for the period.
CONCLUSION
Individual Claims.
The plaintiffs have satisfied their burden of proof in establishing a prima facie case of disparate treatment. They have shown that they are members of a group protected by Title VII. They have shown that they applied for positions and that they could have successfully carried out the duties of the positions. Defendants do not dispute that plaintiffs were never offered the positions for which they applied and that males whose applications were received after plaintiffs' applications were pending received the jobs. This fact is particularly germane when considering that defendant has expressed a desire to hire individuals who are experienced in the work for which it seeks employees.
Plaintiffs have made and supported their prima facie case under the McDonnell Douglas framework. Therefore, their cross-motion for summary judgment is hereby GRANTED.
Class Claim.
The plaintiffs' class claim is somewhat more difficult. First, plaintiffs have misused the applicable statistical analysis in that they chose the incorrect groups for comparisons and then used faulty analysis with respect to those groups. It is impossible to get the standard deviation figures (deviations from the mean) that plaintiff presents when using the correct figures as stated above. Similarly, using the applicant flow pool as suggested in the Discussion section, we find additional disparities from plaintiffs computations and analysis. There appears to be no relationship between the sex of an individual in the relevant labor force and whether or not he was hired; there appears to be no relationship between the sex of an individual and whether or not the applicant applied for a position with Plant Field (except for the obvious difference between the absolute number of women to men).
With regard to the substantial flaws in plaintiffs' statistical analysis and plaintiffs' inability to substantiate the figures presented or to make a convincing argument on their own behalf, this court will rely on defendant's statistical computations and analysis which leave doubtful whether discrimination occurred in the years 1984 and 1985. For the year of 1983, however, defendant has not disproved plaintiff's case and this court is not moved to assume it could. Therefore, plaintiffs' class cross-motion for summary judgment is hereby GRANTED.
Based on the foregoing DISCUSSION defendant's motion for summary judgment with respect to the individual plaintiffs and the plaintiff class is hereby DENIED.
With respect to the individual plaintiffs and the plaintiff class, this court hereby enjoins defendant from any practices, policies, customs, and usages which have herein been identified as discriminatory. To this end, the court advises that the defendant respond to this declaratory and injunctive relief in at least the following ways:
— Contact community and female organizations and educational institutions for employment referrals;
— Advertise in communications media (such as newspapers, radio, and TV) which especially appeal to women;
EEOC Compliance Manual Secs. 1112.4.
— Establish a file of female applicants;
— Discontinue contacts that refer applicants on a discriminatory basis:
— Eliminate any discriminatory word-of-mouth recruiting;
— Eliminate specific factors which brought about discriminatory recruitment, and correct any factors which may result in future discrimination;
— Immediately employ the charging party.
The court having considered the memoranda and due deliberation having been had thereon, it is
ORDERED, that the defendant's motion for summary judgment be and the same hereby is DENIED, and it is further
ORDERED, that the plaintiffs' cross-motion for summary judgment be and the same hereby is GRANTED.
IT IS SO ORDERED.