Opinion
Faculty association and five individual female faculty members brought class action employment discrimination suit against State University. The District Court, Benjamin F. Gibson, J., held that action was not appropriate for University-wide class action, since challenged employment practices occurred primarily on departmental level.
Order accordingly.
Mary H. Job, East Lansing, Mich., Warren J. Bennia, New York City, for plaintiffs.
Robert M. Vercruysse, Butzel, Long, Gust, Klein & Van Zile, Detroit, Mich., for defendants.
OPINION
BENJAMIN F. GIBSON, District Judge.
This is an action by the Michigan State University Faculty Association (Faculty Ass'n) and five individual female faculty members, against Michigan State University (M.S.U.), Teachers Insurance and Annuity Association and College Retirement Equities Fund (TIAA-CREF), members of the M.S.U. Board of Trustees and various M.S.U. officials. Plaintiffs allege they have been subject to employment discrimination on the basis of sex, and they are seeking declaratory and injunctive relief and back pay. Their claims are brought pursuant to Title VII of the Civil Rights Act of 1964, 42 U.S.C. s 2000e et seq.; Title IX of the Education Amendments of 1972, 20 U.S.C. s 1681 et seq.; the Equal Protection Clause of the Fourteenth Amendment, U.S.Const.Amend. XIV, s 1; 42 U.S.C. ss 1983 and 1985(3); and the Michigan Fair Employment Practices Act, as amended, M.C.L.A. s 423.301 et seq. The case is presently before the Court on the question of class certification under Fed.R.Civ.P. 23.
A previous opinion and order in this case was issued by Judge Noel P. Fox on December 30, 1977, certifying a class against M.S.U. consisting of:
All past, present, and future female faculty members who have been, are, or who will be employed by Defendant University, and who are, have been, or will be discriminated against in salaries, promotions, tenure, administrative appointments, retirement benefits, and other terms and conditions of employment as a result of Defendant University's employment practices and policies.
No evidentiary hearing was held prior to this decision. However, Judge Fox ruled from the bench that defendants could have an opportunity, after conducting discovery, to submit evidence contesting the class certification. After discovery regarding the composition of the class was completed, and pursuant to stipulations by counsel for the parties, this Court ordered on January 15, 1980 that the class represented by these plaintiffs be limited to tenure-stream faculty. The order specifically excluded from the class temporary faculty, faculty without pay, temporary staff, and continuing staff. Subsequently, nine days of evidentiary hearings were held to consider whether this case should be continued as a class action.
The prerequisites to a class action are established by Fed.R.Civ.P. 23(a):
One or more members of a class may sue or be sued as representative parties on behalf of all only if (1) the class is so numerous that joinder of all members is impracticable, (2) there are questions of law or fact common to the class, (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class, and (4) the representative parties will fairly and adequately protect the interests of the class.
The burden is on the party seeking to utilize the class action device to show that all four of these prerequisites are satisfied. Senter v. General Motors Corp., 532 F.2d 511, 522 (6th Cir. 1976), cert. denied, 429 U.S. 870, 97 S.Ct. 182, 50 L.Ed.2d 150 (1976).
Plaintiffs seek to represent a class of past, present, and future female tenure-stream faculty subject to an alleged pattern and practice of sex discrimination practiced on a university-wide basis. The record indicates that in the Fall of 1978, M.S.U. employed 319 female tenure-stream faculty members. Defendants do not dispute the claim that such a class is so numerous that joinder of all members is impracticable. Indeed, classes of this size are routinely held to satisfy the Rule 23(a)(1) prerequisite. E.g., Presseisen v. Swarthmore College, 71 F.R.D. 34 (E.D.Pa.1976) (sex discrimination in employment-joinder impracticable where size of proposed class is " at least in the hundreds" ).
This Court is aware that " Impracticability of joinder is not determined according to a strict numerical test but upon the circumstances surrounding the case." Senter v. General Motors Corp., 532 F.2d 511, 523 n.24 (6th Cir. 1976). However, when a class numbers in the hundreds or thousands the impracticability of joinder is obvious, and the issue is ordinarily treated summarily and often goes uncontested. Newberg on Class Actions s 1105b (1977). E.g., Sweet v. General Tire & Rubber Co., 74 F.R.D. 333 (N.D. Ohio 1976) (sex discrimination in employment-joinder of 612 parties found impracticable per se).
Defendants do dispute plaintiffs' claim that there are questions of law or fact common to the proposed class. Plaintiffs argue that Rule 23(a)(2) is satisfied in that they present a common question as to whether M.S.U. has denied female tenure-stream faculty equal treatment with respect to salaries, promotions, tenure, administrative positions, and other terms and conditions of employment as part of a pattern and practice of unlawful sex discrimination. Defendants contend that there are no common questions of law or fact because the decisions affecting faculty members are made almost exclusively within decentralized, autonomous, academic departments on an individualized basis by different persons pursuant to nonstandardized procedures and widely differing criteria. The resolution of this dispute requires a detailed consideration of the evidence submitted to the Court regarding the administrative structure and function of M.S.U.
Michigan State University is a relatively large educational institution organized into sixteen colleges and some ninety departments or schools. It is formally governed by a Board of Trustees pursuant to the Constitution of the State of Michigan. The Board has ultimate responsibility for approving all promotions, tenure, and salary actions at the University, but faculty participation in the decision-making process is provided for in the Bylaws for Academic Governance. The principal academic officer is the Provost, who is subject to the Board and the President, and whose responsibilities include ensuring that personnel procedures produce accurate and fair evaluations.
Each of the sixteen colleges represents a general area of academic disciplines and is headed by a dean. The dean's duties include evaluating, reviewing and recommending to the Provost on faculty personnel matters, and ensuring that departmental procedures are carried out in a fair and consistent manner. The colleges are made up of one or more departments or schools which constitute the basic academic and personnel units of the University. Each of the ninety or so departments or schools is headed by a chairperson or director who is the front-line administrator for all faculty in the department.
The tenure-stream faculty to which the proposed class definition refers includes the academic ranks of professor, associate professor, assistant professor, and instructor. Tenure-stream faculty serve a probationary period of three to six years, at the end of which, if reappointed, they receive tenure and become insulated from dismissal except for cause. The general standards for promotion and tenure are established by the central administration of M.S.U. and provide that tenure-stream faculty members should be evaluated according to their performance of three basic functions: teaching, research, and service.
The procedure for promotion, tenure, and reappointment occurs in three basic steps: departmental peer review; a preliminary conference between the chairperson of the department and the dean; and the formal process of recommendation. The latter begins with the chairperson's completion of both an eight-page comprehensive form for each individual recommended and a prioritized list for all the department's faculty recommendations, leads to review by the dean and the Provost, and ends with Board approval. The procedure for salary increases also runs from the department to the Board in the case of merit increases, while across-the-board increases are established by the Provost's Office. That office also establishes a ceiling percentage amount for the merit increases, which amount is then allocated within each department by peer evaluation subject to review by the dean. A third type of procedure for salary increases occurs when studies in the Provost's Office identify certain individuals' salaries as being anomalous-the department and dean then provide justifications for either a salary adjustment or the existing salary level.
A good deal of evidence establishes that much of the important decision-making with respect to personnel matters takes place at the department level. The record is clear that teaching, research, and service are university-wide standards for the evaluation of faculty, but that the implementation of those standards occurs mainly within the individual department. The departments act as autonomous units in developing their own governance structure through bylaws, defining their program or mission as a unit, establishing standards and criteria by which performance will be judged, and establishing procedures for decision-making regarding promotions, tenure, and salary increases.
Witnesses included M.S.U.‘ s Associate Provost, Dr. Jack Kingsinger, who testified that each department has its own bylaws developed by the faculty of that department which describe the department's mission, i.e., the emphasis which that department places on its programs. He stated the University does not instruct departments as to what the process for evaluation should be, or what the bylaws should say, so long as they do not conflict with the University's policies. He further stated that the department is the primary unit to describe what constitutes the achievement necessary for merit in that department. He pointed out that some departments evaluate faculty on a national ranking and some on a departmental ranking, with the choice being made at the departmental level and not prescribed by the University administration.
Provost Clarence Leland Winder testified that the criteria for the evaluation of faculty and for determining salaries are formulated in the department by the faculty and established by the chairperson. He noted that, " (T)he faculty being fully qualified professionals in their particular area do have considerable latitude in making determinations as to standards." Isabelle Payne, Dean of the College of Nursing, and Harold David Hafs, chairperson of the Department of Dairy Science, also testified that departmental procedures are established by bylaws developed by the faculty in their departments or schools.
Departmental autonomy is further illustrated by the variations which occur between departments. Some departments provide for student representation on faculty committees, or absolute chairperson vetoes over committee recommendations, while others do not. Bylaws differ in the method for selection of the chairperson, and in the composition and functions of department advisory councils. Provost Winder explained that department evaluation methods and criteria differ because of the differing missions of the various departments-some emphasize undergraduate teaching, while others stress scholarship or clinical instruction.
One of the key features of departmental decision-making is peer review. Associate Provost Kingsinger testified that the significance of faculty input into questions of promotion, tenure, and reappointment is that:
(T)he individuals who are doing the evaluation ... (of) the persons under consideration (are) peers. They are the individuals that are closely identified in a professional way with the accomplishments of the specific individual who is coming up for evaluation or judgment and since their professional careers are similar and they understand the operations of the department and the kind of research and kind of public service that is done within that unit, they, as a set of peers, can make the substantive judgments on their colleagues for promotion and tenure.
Since a faculty member's peers are found within the same academic discipline, the locus of a peer review method of general personnel decision-making inevitably tends to be the department. Provost Winder testified as follows:
Q. All right now ... let's take for example instructors. (T)he means of evaluating whether or not someone is a good instructor rests with the department?
A. Yes.
Q. Does the final determination as to whether or not someone (is) a good instructor pardon me, is good at performing the art of instruction, does the final determination as to that individual's performance rest within the department?
A. The evaluation of that individual with regard to that function is made primarily by the department.
Q. You say the evaluation. (N)ow I asked you the final decision as to whether or not that individual is possessed of sufficient quality, by that I mean the quality of being an instructor, quality research ... and administrative service, the final decision whether that individual possesses those qualities to warrant either promotion or tenure, does that rest in the department?
A. As a general matter, yes.
Q. Well, could you clarify that a little bit, as a general matter yes?
A. In most instances the evaluation of the department is the determinative evaluation. That's the normal event.
Q. When you say it's determinative ... do you mean that there are no subsequent evaluations that take place of the performance of individual faculty members?
A. There may be an additional assessment evaluation in an unusual instance. As a general matter, the process beyond that is one of looking at procedure, of being as certain as one can be that the criteria employed by the department are appropriate to its mission and that they have been applied in each individual case fairly, and that the standard of performance is a high quality standard.
The Dean of the College of Agriculture succinctly stated:
Q. Based on your experience, Dean Anderson, in the College of Agriculture since May of 1977 where in your opinion does decisionmaking reside regarding the employment decisions that affect tenure stream faculty members?
A. There is absolutely no question that it resides at the department level or the school level.
The record also presents evidence that each specific type of challenged employment decision occurs mainly at the departmental level. As to tenure, several witnesses at various levels of the administrative hierarchy testified that they had never rejected a departmental tenure recommendation or had one rejected by the next highest level. Provost Winder could not recall ever overturning such a recommendation, and an " extremely high" percentage of such recommendations are accepted by the deans.
As to promotions, the deans and the Provost have a somewhat greater effect, but the decisions are still essentially departmental ones. The Provost's Office exercises some control over the number of promotions which occur in a given year, and monitors procedures and criteria, but it does not usually become involved in substantive judgments regarding which individuals will receive promotion. Deans occasionally get involved, but this is most likely when the issue is either early promotion or greatly delayed promotion. Testimony indicated that deans rejected or influenced departmental recommendations for promotion " in a few instances; " twenty percent of the time; and in " one out of seven" cases. One former college dean testified that he focused on procedures and standards rather than individual candidates for promotion. The fact that promotion decisions are generally considered to be a departmental function can be seen in the testimony of plaintiffs' witness, Dr. Patrick Larrowe, an economics professor:
Q. As matter of fact, if the dean told your department that four people had to be promoted tomorrow, wouldn't the department be irate?
A. Yes.
....
Q. The recommendation with respect to promotion is very important department prerogative, wouldn't you agree?
A. Oh, yes it is.
Dr. Harold Hafs, Chairperson of the Department of Dairy Science, responded to a similar question by stating, " I think our faculty would be very offended by such a suggestion. They value their autonomy in these kinds of decisions highly."
As to salaries, plaintiff Dorothy McMeekin testified that the statements on faculty salary made by her department chairperson are usually final, but can be appealed. It is quite rare for the Provost's Office to get involved with the pay raise of any individual; the Dean of the College of Agriculture testified that he had submitted some 1100 salary recommendations, none of which were questioned or changed by the Provost. Other testimony indicated that a dean raises questions in cases of extremes; in ten to twenty out of 275 recommendations; in four out of eighty-four recommendations; and ten to twenty-five per cent of the time; and that deans have overturned recommendations in one out of 1100 instances, and in one school, not once since 1961.
As to anomaly increases, although the University identifies the individuals to be reviewed, it is the departments which set the specific dollar adjustment in every instance.
Thus, it is clear to this Court that the decision with respect to each of the challenged employment practices is made primarily on the department level. The testimony of Dr. Kingsinger is illustrative:
Q. Okay. Now, as a member of the faculty, a former chairman, as the assistant vice president for research and now as associate Provost can you tell us where the decision is made with respect to a recommendation for tenure in the process, a significant decision?
A. That decision emanates from the department.
Q. Where is the decision made with respect to promotion?
A. In the department.
Q. Where is the evaluation of an individual carried on?
A. In the department.
Q. Where is an individual's pay determined?
A. In the department.
The point of all this evidence is that departmental decision-making negates the existence of any common questions of law or fact to satisfy Rule 23(a)(2) with a University-wide class. Plaintiffs argue that informal personnel procedures, through which higher level administrators review and affect departmental decisions, demonstrate that higher level administrators do exert sufficient influence to raise common questions about a pattern and practice of discrimination. However, plaintiffs' claims in this case are essentially allegations of discriminatory treatment. As such, they require evidence of discriminatory intent on the part of the defendants. International Brotherhood of Teamsters v. United States, 431 U.S. 324, 335-36, n.15, 97 S.Ct. 1843, 1854-55, n.15, 52 L.Ed.2d 396 (1977). Plaintiffs have not demonstrated that defendants have willfully discriminated or condoned discrimination on the basis of sex. The evidence of informal input by deans and the Provost's Office does not persuade this Court that the critical decisions affecting tenure-stream faculty are made anywhere other than in the departments. Nor does the formal authority of the Board of Trustees establish discriminatory intent where the Board does not inquire into the specifics of individual personnel decisions. Therefore, there is insufficient evidence of any intent to discriminate in the upper administrative levels of the University to satisfy the commonality criterion for a class action.
Defendants admit that plaintiffs' claim against the Teachers Insurance Annuity Association and the College Retirement Equity Fund (TIAA-CREF) is one of disparate impact. By virtue of Judge Fox's Opinion of December 30, 1977, certification of a class against TIAA-CREF was ordered held in abeyance, for reasons of convenience and efficiency, pending the resolution of the identical issue in other cases. That order continues in effect until such time as this Court is made aware that the reasons for the order no longer apply.
Although discriminatory intent can be inferred from statistics, the statistical evidence presented in this case does not provide a basis to find the asserted common question as to intent. Plaintiffs' mathematician, Dr. Stewart, testified that his study of a judgment sample of 1976 faculty showed that only 53 percent of the differences between salary in male and female faculty could be explained by the factors of hire year, birth year, department, and degree. Beyond the problems with sampling technique identified by defendants, this study ignores important factors like the relevant labor market, merit, and years of teaching experience, and thus gives rise to no reasonable inference of discrimination.
Plaintiffs' exhibits showed that, as of 1978, female faculty in the tenure system earned less than men on the average, and constituted a proportion of tenured and upper-rank faculty that was lower than their overall representation within the tenure-stream. Such gross statistics, which fail to consider critical factors such as departmental requirements and the relevant labor market, provide no basis for a finding of gender discrimination where there are legitimate special qualifications for advancement and no evidence is introduced as to the number and availability of qualified women. Grano v. City of Columbus, 637 F.2d 1073, 1078 (6th Cir. 1980). Plaintiffs' statistics are at least as consistent with women having only recently entered the M.S.U. tenure stream in significant numbers as they are with intentional discrimination.
Defendants' statistician, by taking account of additional factors such as department, rank, and ten-month salary, purported to explain over 95 percent of the salary variance at the associate professor and assistant professor ranks. He concluded that sex was a variable that was not statistically significant. Although defendants' statistical analysis is far from determinative, since it includes variables which could be gender-related and it excludes professors, instructors, and all faculty hired before 1971, it does further dispel any vestige of inferential discriminatory intent.
Plaintiffs argue, as an alternative to a disparate treatment analysis, that this case presents a class-action common question as to whether the current personnel procedures of M.S.U. have a disparate impact on women. They have failed, however, to point out any facially neutral policy which has an adverse impact on female faculty. The evidence indicates essentially that a three-part standard for faculty evaluation is employed university-wide, and that upper-level administrators review departmental personnel decisions for fairness and consistency. Such review procedures would seem to have a beneficial impact on women who might have been treated unfairly at the departmental level. At any rate, plaintiffs have failed to show that such procedures have an adverse impact upon women in the M.S.U. tenure stream as required by Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). Thus, there is no common question of disparate impact upon the proposed class.
Plaintiffs argue that this case is not distinguishable from others in which a class was certified. In particular, they point to the decision in Senter v. General Motors Corp., 532 F.2d 511 (6th Cir. 1976), that an action alleging discrimination in promotions was appropriately certified by the District Court as a class action despite the employer's argument that the individualized nature of promotion decisions negated any common questions of law or fact. The opinion makes it clear that Senter was a disparate impact case. The facially neutral policy being challenged was the promotion procedure in a division of the company that accorded great weight to the subjective evaluation of an hourly worker's foreman. The procedure did not have objective criteria, an established means of publicizing openings, or a publicized means of applying for promotion other than by direct recommendation of the foreman. The adverse impact was demonstrated by statistics which showed a lower proportion of blacks in the supervisory ranks than in the total work force, where no special qualifications for supervision had been established. In this case, M.S.U.‘ s systematic, periodic peer review of tenure-stream faculty according to clearly established criteria does not present the same sort of suspect procedure, and the statistics fail to demonstrate any adverse impact which could support a common question of disparate impact.
Plaintiffs also cite several cases involving allegations of employment discrimination against colleges and universities in which class actions were upheld. Defendants respond with similar cases denying class certification. Many of plaintiffs' cases are distinguishable with respect to the facts in that they demonstrated a stronger degree of control by the university's central administration over personnel decision-making than is present here. For example, in Rajender v. University of Minnesota, 20 Empl. Prac. Dec. P 30,21 4 (D.Minn.1978), the court determined that answers to interrogatories " made it apparent that the central university administration exercises considerable control over each employment decision. This is evidence that factual questions arise which are common to the class of plaintiffs." 20 Empl. Prac. Dec. at 12,154.
This Court is aware of the admonition contained in the Supreme Court's unanimous decision in East Texas Motor Freight System Inc. v. Rodriguez, 431 U.S. 395, 97 S.Ct. 1891, 52 L.Ed.2d 453 (1977). The Court there reversed the Fifth Circuit's certification of a class, stating that, " we are not unaware that suits alleging racial or ethnic discrimination are often by their very nature class suits, involving classwide wrongs. Common questions of law or fact are typically present. But careful attention to the requirements of Fed.Rule Civ.Proc. 23 remains nonetheless indispensable." 431 U.S. at 405-06, 97 S.Ct. at 1897-98. Thus Lamphere v. Brown University, 71 F.R.D. 641 (D.R.I.1976), which certified a class of female faculty allegedly subject to discrimination, has lost persuasive authority because it cited the overturned Rodriguez opinion of the Fifth Circuit for the proposition that the requirements of Rule 23 must be applied liberally in Title VII cases.
The Court is of the opinion that this case is much like Sanday v. Cernegie-Mellon, 15 Empl. Prac. Dec. P 8088 (W.D.Pa.1976), in which a university-wide class of faculty was denied based upon quite similar facts-a decentralized method of decision-making involving peer group evaluation in autonomous departments-which failed to show any common questions or typicality of claims. See also, Stastny v. Southern Bell Tel. & Tel. Co., 628 F.2d 267 (4th Cir. 1980) (reversal of certification of multi-facility class of female employees alleging discrimination in promotions and pay, where evidence showed unit autonomy and unit-based decision-making).
Defendants argue that careful analysis of the individual plaintiffs' claims shows them to be unique and related to only one department, thus preventing them from being typical of the claims of the class under Rule 23(a)(3). Although some of the same considerations that applied in the common question issue would apply here, the Court's finding that a common question is lacking makes this decision unnecessary.
Defendants further contend that Rule 23(a)(4) is not satisfied and plaintiffs cannot be adequate representatives of the class. The Court notes that there is some question on this issue, given the evidence that the Michigan Education Association, although not a party, is financing and directing this lawsuit, employing one of plaintiffs' attorneys and retaining the other. However, it is unnecessary to reach a decision on whether the Rule 23(a)(4) prerequisite is satisfied, and similarly, whether Rule 23(b) is met.
The Court rejects the possibility of certifying classes as to each department, because there is insufficient evidence of the requisite numerosity in such an approach. It is the opinion of the Court that this action shall no longer be maintained as a class action, because there are no questions of law or fact common to the class as required by Fed.R.Civ.P. 23(a)(2). The earlier class certification is hereby withdrawn. The Court notes that this opinion is in no way a decision on the merits of the individual plaintiffs' claims which remain to be tried. This Court merely holds that the record before it at this time does not support class certification. The Court, however, reserves the right to certify an appropriate class in the future if facts should develop to support such action.