Ross v. Buckeye Cellulose Corp.

733 F. Supp. 344 (1989)

Issiah ROSS Jr., et al., Plaintiffs,
v.
BUCKEYE CELLULOSE CORP., Defendant.

Civ. No. 86-48-ALB/AMER(DF).

United States District Court, M.D. Georgia, Albany/Americus Division.

August 11, 1989.

James Finkelstein, Albany, Ga., for plaintiffs.

John G. Skinner, Smith, Currie & Hancock, Atlanta, Ga., for defendant.

FITZPATRICK, District Judge.

These actions were tried before a jury for four months, from September 6, 1988 to January 6, 1989, on discrimination actions brought pursuant to 42 U.S.C. § 1981. On December 21, 1988, the jury returned verdicts on the liability aspect of the cases, finding liability in only two instances: Plaintiffs Ross and Plant. After additional evidence on January 6, 1989, the jury returned damage verdicts as to the two prevailing Plaintiffs. The court entered final judgment on the Plaintiffs' section 1981 claims on March 21, 1989. Now before the court are the Title VII claims brought by eleven of the thirteen Plaintiffs. Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. § 2000e, et seq.

The Title VII claims of these Plaintiffs are premised on both the disparate treatment and the disparate impact theories of recovery. The court entered judgment on the individual disparate treatment claims of the Plaintiffs on March 21, 1989, pursuant to the Eleventh Circuit's holding in Lincoln v. Board of Regents, 697 F.2d 928, 934 (11th Cir.1983).^[1] Regarding Plaintiffs' disparate impact theory of recovery, the court heard additional statistical testimony on May 1 and 2, 1989, and oral argument on June 27, 1989. Having carefully considered the entire record, the court sits prepared to rule on the remaining Title VII claims as set forth below in the court's Findings of Fact and Conclusions of Law.

Before the court embarks on this complicated history of the employment relations between Plaintiffs and Defendant, however, it must first address Defendant's Motion to Strike Portions of Plaintiffs' Proposed Findings of Fact and Conclusions of Law and Appendices. Defendant seeks to have this court strike certain portions of Plaintiffs' proposals on the grounds that: (1) Plaintiffs have identified a number of Defendant's employment practices never before questioned and not addressed by Defendant in its proposals; (2) certain of the Plaintiffs' arguments and appendices would require a reopening of the evidence which should not occur absent grounds which are not present here; and (3) Plaintiffs have attached certain documents as appendices which are not admissible as evidence, represent summaries of various exhibits, and have not been cross-examined by Defendant. The court briefly addressed these contentions at oral argument and trial, and will now rule appropriately.

As to the employment practices being attacked by Plaintiffs, the court finds that Plaintiffs are questioning Defendant's entire Pay and Progression System (the System), including all of its various elements. In order to attack any particular element of the System, Plaintiffs must assail the entire process. Thus the court will not strike any arguments regarding the several components of Defendant's Pay and Progression System preliminarily. On May 26, 1988, the court issued its Order granting summary judgment to Defendant on Plaintiff Ross' discharge claim of disparate impact. In that Order the court ruled that Plaintiff Ross' termination for "inability to perform job" was not a specific employment practice subject to disparate impact attack. The court will not vacate that ruling. Thus Plaintiff Ross' termination claim will not be analyzed in the context of this Order. Plaintiff Porter also claims that Defendant's subjective system of termination had a disparate impact on blacks. Specifically, Porter questions Defendant's `practice' of terminating employees for "absence and tardiness." The court finds that this "practice" is, like "inability to perform job," a general categorization by an employer used to discharge various employees from different types of jobs, and is not a discreet employment practice or procedure such as contemplated by the disparate impact theory of recovery. See Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 108 S. Ct. 2777, 2785, 101 L. Ed. 2d 827 (1988). The termination policies of Defendant will not be examined here.

Defendant contends that Plaintiffs are attempting to reopen the record through their proposed Findings of Fact and Conclusions of Law by presenting new evidence and theories of recovery. This contention must be viewed in conjunction with Defendant's third ground for striking portions of Plaintiffs' proposals regarding the admissibility of certain of the documents appended thereto by Plaintiffs. As stated earlier, this Order will address Plaintiffs' disparate impact theory of recovery as it relates to Defendant's Pay and Progression System; no other theory of recovery will be discussed. The court and Defendant have both long been aware of the essential subject matter of Plaintiffs' claims. Only through the presentation of evidence was Plaintiffs' theory able to take positive shape and definition. As the evidence developed, the court itself was able to grasp the particular policies and practices which contributed to the entirety of the Pay and Progression System. Because of the necessary place which each element of the System holds in relation to the overall outcome, the court is compelled to allow Plaintiffs to attack the whole System and, necessarily, each of its elements. In arguing this theory to the court, Plaintiffs are allowed to present charts, graphs, summaries, and appendices to further their arguments as well as attempt to aid the court with these documents. While the documents appended to Plaintiffs' proposed Findings of Fact and Conclusions of Law may not be admissible as evidence, the court will not strike them for that reason alone. This portion of the case is being heard by the court sitting without a jury, and this court is more that capable of separating fact from argument. Lest Defendant forgets, this court sat through the entirety of the four month jury trial, ruling on each evidentiary question. The court is well aware of what was ruled admissible evidence. The court considers those documents attached as appendices to Plaintiffs' proposed findings of Fact and Conclusions of Law to be argument aids, designed to aid Plaintiffs' attorney in presenting the case, and the court in understanding. Accordingly, the court will proceed based upon the above rulings, by first setting out its Findings of Fact. These initial findings are simply a skeletal recitation of the facts in the case and the court will supplement them as necessary in its Conclusions of Law.

I. FINDINGS OF FACT

A. BUCKEYE'S FLINT RIVER PLANT

Plaintiffs in this action are all black, present or former, employees of Defendant Buckeye Cellulose Corporation.^[2] Buckeye is an Ohio corporation and a wholly owned subsidiary of Procter & Gamble Paper Products Company (Procter & Gamble). Buckeye owns several plants in various states around the country including one in Macon County, Georgia, near Oglethorpe, known as the "Flint River Plant" (the Plant). The Plant processes raw wood into heavy sheets of paper which are eventually sold to other Procter & Gamble subsidiaries to eventually be processed into the absorbent material used in disposable diapers.

The Plant is divided into essentially five (5) main working areas: Unit 1, the Woodyard area; Unit 2, the Pulping Unit; Unit 3, the Product Unit; Unit 4, the Powerhouse Unit; and Unit 5, Plant Maintenance Services. Each of these areas serves an integral function in the processing of the raw wood products. In addition to the five general areas, the Plant also contains a technical unit (the Central Laboratory); an administrative division; and a lands and timber section. In order to better understand the operations of the Plant a brief overview of the function of each of the relevant operating Units is necessary.

Unit 1, the Woodyard, is where the wood products first enter the Plant, either on rail cars or in trucks. The wood is processed into one inch chips which are conveyed to the Pulping Unit. The Waste and Water Treatment area is also a part of Unit 1. Waste and Water Treatment include the processes by which the incoming and outgoing water for the entire Plant is chemically treated for use.

The Pulping Unit, Unit 2, is where the chips prepared in Unit 1 are put through a series of processes which separate the cellulose, the fibrous material, from the lignin and bleaches it into a white fiber.

The white fiber, or pulp, goes through a screening series in the Product Unit, Unit 3, to remove any remaining water. The pulp is then dried and rolled into large rolls, which are either stored in the warehouse, or placed directly onto trucks or rail cars for shipping.

Unit 5, Plant Maintenance Services, provides maintenance and support services to the entire Plant. Unit 5 consists of a welding and machine shop, vibration analysis, building maintenance, electrical and instrumentation (E & I), and a storeroom. Each of the other Units, however, employs maintenance personnel and, with the exception of the Woodyard, a group of E & I employees.

In 1979, Buckeye began preparations for opening the Plant with Manager Al Eppinger. Buckeye hired a preliminary group of employees, approximately 40 persons, in December, 1979. This first hire group, known as "tech trainers," were responsible for learning the different jobs at the Plant in the various Units and developing the training program for those employees to be subsequently hired.

Buckeye began hiring the majority of its work force in the Summer of 1980. The new employees, called "technicians,"^[3] were hired in stages, with particular starting dates, either June 9, July 8, August 11, or September 9, 1980. In addition, there were two subsequent, smaller hire groups. The technicians were hired for positions as they became necessary for the operation of the Plant. Buckeye originally hired 450 employees and had a stated overall goal of hiring an employee pool consisting of 60% whites and 40% blacks, of which 20% would be female. Broken down by hire groups, only 20% of the tech trainers were black, the June hire group was approximately 30% black, 35% of the July hire group was black, and in August approximately 50% of the new hires were black. Eventually Buckeye reached its 60-40-20 goal. Upon hire, each new employee went through an orientation process whereby he or she became familiar with the various operations of the Plant, the technician system, the Pay and Progression System, and Buckeye's multicultural organization concept. After the orientation period, each employee was allowed to state three preferences for a job assignment within those positions available to the particular hire group.^[4] In making initial job assignments, Buckeye primarily considered three factors: individual preference, multicultural balance, and the company's needs. Within each Unit, the new hires went through a 3 to 6 month training period in which they learned their assigned jobs. Each Unit consisted of several teams which worked together as shift teams. Early in the Plant's operations, each Unit had 5 teams.

Initially, all of the technicians within a hire group were paid the same amount, $288.00 per week. At the 3 and 6 month periods of employment, each employee received uniform pay increases. At the end of nine months, each employee entered the career planning process, what was eventually to become the Pay and Progression System now under attack by Plaintiffs. A "career plan" set out the particular skill areas in which the employee was to be trained and the degree of proficiency to which he or she was to learn the skill. By skills, Buckeye was referring to the various jobs within each Unit necessary for its operation; for example, Unit 1 consisted of mechanical maintenance, chip unloading, long wood — short wood, waste/fuel, and waste and water treatment skill positions. Technicians were assigned "primary," "cross," and/or "secondary" skills. Buckeye claims that the intent behind the system was to allow the technicians to add to their skill base as they were able and as the Plant needs allowed. Buckeye established the Plant needs through the concept of an "ideal team skill mix." This ideal team skill mix designated the foreseeable Plant needs for 1 to 2 year periods by stating the number of particular skills needed to operate the Plant and the level of proficiency for each of those skills. Performance was supposedly a significant factor in determining the original career plans.

The career plan was a product of several elements. First, Buckeye conducted Plantwide shift team meetings to explain the system. The managers then gathered information about each technician's performance on the job. Each technician then met with his manager to discuss his performance level, and his or her interests, goals and expectations. Each technician completed an "Interest Form" to identify preferences for career plans. The Unit Management Team then met to evaluate the gathered information in conjunction with the employees' preferences and the "ideal skill mix." The Unit Management Team then assigned proposed career plans and submitted them for approval to the Pay and Progression Review Board. The Review Board either accepted or modified the proposed career plan for each employee and then assigned him or her to a pay curve level. A pay curve was, at that time, one of two pay levels on which a course of pay increases was charted in relation to the career plan and the timing intervals established by the plan. Once the career plans were approved by the Board, managers reviewed them with each technician.

Sometime in 1982, Buckeye modified the System by measuring the value of the various skill areas in relation to one another by assigning a point value to each skill. A numerical value was placed on each of the primary, secondary, and cross skills by Plant Manager Burt Richards; for example, in Unit 1, the Woodyard, a technician received 10 points for a basic assignment in mechanical maintenance and 35 for a proficient, chip unloading technicians received 4 points for basic and 5 points for proficient assignments, technicians received 4 points when assigned a basic in long wood and 8 points for proficient, short wood skills were given 6 and 11 points for basic and proficient respectively, and waste treatment was assigned 6 points for basic and 12 points for proficient. These values were not divulged to any of the technicians or managers, though testimony at trial indicated that certain employees became aware of the point values, their significance, and the skills necessary to obtain a desired goal/pay level.

In addition, Buckeye instituted a ranking system whereby, initially, the Unit Management Team force-ranked each technician's performance and contribution on a scale of 1 to 5, with 1 being the lowest possible rating and 5 being the highest. Eventually, the technicians played a more active role in the ranking system. The ranking system operated such that technicians were force-ranked, and the number of technicians which could be given any particular rank was limited, i.e., only 15% of the technicians could achieve the top rank of 5, no matter whether all technicians on a given team operated at the same level or were exceptional performers. The ranking system was almost totally a subjective program, with few, if any, standards to guide the managers.

When establishing a technician's pay curve, Buckeye primarily considered the career plans, the skills contained therein, and the points assigned to those skills. Buckeye's System considered all of the skills contained in the career plan, before they were acquired and whether or not they were ever acquired. The ranking system affected the pay of a technician only if the ranking was either above or below average, a 3 ranking, and then only, at most, by one pay curve. A variance in the pay curve was not automatic, even where the ranking varied from the average. Since the forced ranking system was based on a majority of the technicians being ranked at average, then, in practice, rankings had very little effect on the majority of the Plant's technicians. The pay curve system was further complicated by the fact that Buckeye assigned each Unit an average pay curve rate which should be maintained within the Unit. These systems of averages and curves should have resulted in a bell curve-type result, Plant-wide, as well as within each Unit and team. In reaching a particular result regarding the ideal skill mix, the ranking average, the pay curve average, the skill needs of the Plant, the Unit and the team, and the ultimate assignment of skills, the Unit Management Team had almost complete autonomy.

In 1982, Buckeye began administering examinations to gauge the progress of each technician under his career plan. These examinations were called "Qualification Review Boards." The examination was oral and was conducted by a qualification board generally consisting of between three and five members. The early qualification boards were comprised solely of management personnel until sufficient technicians attained a proficient rating, thereby qualifying to serve on a qualification board. Subsequently, the boards were made up of the technician's immediate manager, one or two individuals selected by the technician (another knowledgeable manager and a proficient technician), and up to two other management personnel. Buckeye prepared and distributed guideline questions to the qualification board members to consider asking during the examination, however, the board members were not bound by these questions and were free to ask the technician any relevant questions. The actual examination was not recorded, but the board members often made their own notes regarding the responses given by the technician. At the completion of the examination, the board members conferred with each other to reach a consensus as to the appropriate rating for the technician based upon both objective and subjective standards, i.e., the technician's performance during the examination, his observed on-the-job performance (including teamwork and leadership abilities), and the "gut feelings" of the board members. At the time in question here, the results of the Qualification Review Boards did not directly impact on the pay of a technician, however, they did impact on possible future career plan advancements. Buckeye established an appeals process available to technicians who were unhappy with their Qualification Review Board results.

In 1982, after the first series of Qualification Review Boards had been administered to all technicians, a second career planning process took place similar to the first. Once again, Buckeye established an ideal skill mix for the Plant, each Unit, and each team. By this time the Plant operated with four instead of five shift teams, necessitating a reevaluation. Buckeye used this ideal skill mix to control the average pay curve within each Unit. The Pay and Progression Review Board, as before, had the final approval over the ideal skill mix for each Unit.

The career plans were then revised in a fashion similar to the earlier career plan system, keeping in mind the new ideal skill mix, the previous career plan, the results of each technician's Qualification Review Board, the technician's on-the-job performance, the desires of the technician, and the needs of the Plant, the Unit, and the team. As before, though, the Unit Management Team had practical autonomy in establishing the career plans, subject to final approval by the Pay and Progression Review Board and the technician's right of appeal. However, in practicality, a manager's decision was seldom reversed. If a technician had not attained a certain level of proficiency in his Qualification Review Board, he was often not allowed to receive additional skills, and was at times not allowed to retain those on his first career plan. These second career plans also contained primary, secondary, and cross skills which were to be learned by the technician, and they were the essential ingredient in establishing a technician's pay curve. Thus the pay course of each technician, once established, was the basis for his entire pay history at Buckeye, and was a chief result of the fundamental elements of the Pay and Progression System which left a great deal of responsibility with the subjective decisions of the managers. This second round of career plans was the last Plant-wide reevaluation, though modifications to certain career plans did occur on individual bases.

Buckeye froze its entire Pay and Progression System in October of 1984 because of alleged low production, waning production quality, and high costs. Apparently the secondary and cross skill system had not worked as well as expected, and valuable technicians were lost in continual training for these additional skills. Once the freeze became effective, technicians were reassigned to their primary skills and many of the problems which prompted the freeze improved. Though the pay freeze recognized the problems with the Pay and Progression System, especially the need for secondary and cross skills to move ahead, it did not attempt to erase all of these problems, especially in terms of pay and pay curves which were a result of this mad dash for additional skills to improve one's pay. As noted earlier, technicians continued to be paid for whatever additional skills were contained in their career plans, even after the freeze to remedy the additional skills problem and whether or not the skills were ever learned.

Since the 1984 freeze, Buckeye has attempted to modify and improve its Pay and Progression System with the input of technicians. This upgrade included the abandonment of the oral qualification examinations in favor of written examinations. The ranking system is now conducted by both the managers and the technicians based on a specific set of criteria, and a technician must maintain a ranking equal to his pay curve in order to remain on that curve. Thus the new, more objective ranking system had a much more profound effect on the Pay and Progression System. Career plans have been renamed "skill blocks," and while they retain the numerical valuation system for skills and contain cross skills, they do not have the same aura of secrecy nor the impact of the previous career plans. The new Pay and Progression System has implemented a more objective and measurable method of evaluating a technician's performance, which eventually effects his pay.

B. STATISTICS

Since the aspect of these cases which this court must evaluate is the theory of recovery known as disparate impact, statistical evidence is extremely important. International Bhd. of Teamsters v. United States, 431 U.S. 324, 339, 97 S. Ct. 1843, 1856, 52 L. Ed. 2d 396 (1977). Statistical evidence was presented at trial by both parties through the testimony of two statisticians, Professor Jimmy Ramsey and Professor Charles Haweworth, as well as through the use of raw numbers regarding Plant employees. Since the statistical aspect of the disparate impact analysis is relevant for the establishment of Plaintiffs' prima facie case, Wards Cove Packing Co., Inc. v. Atonio, ___ U.S. ___, 109 S. Ct. 2115, 104 L. Ed. 2d 733 (1989), the court will analyze the statistical evidence from Plaintiffs' point of view, recognizing the differences between Plaintiffs' and Defendant's evidence when relevant. However, the court will first review the raw numbers and possible disparities visible and apparent to the statistically uneducated eye of the court, mindful of the approximate 60%/40% split in the Plant's work force between whites and blacks. In making these observations, the court will look primarily to the two extremes, the lowest and the highest, because Buckeye's goal at the Plant was essential based on reaching an average, thus presumably the middle ground is where the bulk of the employees would be found. Only by analyzing the extremes can the court get a true feel for what the numbers really mean and the success of Buckeye's "average" approach, as the court is aware of the deceptive nature of that word.

The mean/average weekly salary of all blacks, Plant-wide, as of December 31, 1985, was $534.00. The mean white weekly wage was $558.00. In Unit 1, the mean white salary was $544.00, while the mean/average black salary was $514.00.

Looking at the Unit 1 technicians in relation to the Pay and Progression System in effect from 1982 to 1984,^[5] the average pay curve of white technicians was 3.1 and the average black pay curve was 2.5. The 48 white technicians averaged 36.1 points in their final career plans while the 36 black technicians averaged 29 points. Ranking affected only 29 of the 81 total technicians, raising 15 up a pay curve, and lowering 14. Of those 15 raised pay curves, 12 were white and only 3 were black, while 8 of the pay curve reductions affected blacks with only 6 whites being affected. The top 8 career plan point total positions were held by whites, as were 12 of the top 15 positions. On the lower end of the scale, 16 of the lowest 17 point total positions were occupied by blacks. As explained earlier, these point total positions were established by the skills contained in a technician's career plan, whether or not a technician ever actually learned all or any of the secondary or cross skills. These point totals were a direct result of the subjective decision-making of the managers. By being placed in the above-related positions, certain technicians at the top of the scale could reasonably be assured that at no time in their Buckeye career would they ever be paid below a pay curve of 3, while certain of the lower point-totaled technicians could reasonably anticipate being paid no higher than on a pay curve 3 basis, both of these statements were true no matter how good or bad a job the technician did in the future. In looking at these numbers, the court is mindful that a technician's later growth at the Buckeye Plant was controlled in large part by the early career plans given them by their managers and the results of their initial Qualification Review Boards, because pay was primarily based on career plan point totals and managers were reluctant to upgrade or improve a technician's career plan before a certain level of proficiency was obtained, as judged by the qualification boards. Broken down by skills, of the 23 mechanical maintenance technicians in Unit 1, all 7 of the highest paid technicians were white, while all of the blacks were paid on the lower two pay curve scales. The top 9 career plan point totals belonged to whites, though 6 of the lowest 8 point totals were given to blacks. In the waste and water treatment skills 7 of the 8 highest paid technicians were white and 4 of the 5 lowest paid were black, out of a total of 13 technicians. White waste and water treatment technicians received the 6 highest career point totals and managers gave blacks 5 of the remaining lower point totals, including the three lowest.

Of a total 81 technicians in Unit 4, 22 of the 28 highest paid technicians were white while 10 of the 15 lowest paid were black. The average pay curve for whites was 3.5 and for blacks was 2.9. Technician rankings affected 15 whites positively and only 3 whites negatively, 4 blacks were affected positively and 6 negatively. The average career plan point total for whites was 44.0 and 36.6 was the black average. Of the 62 process technicians, the average pay curve was 3.47 for whites and 3.0 for blacks; the rankings affected eight black technicians, four positively and four negatively, and whites were affected positively +7; the average career plan point totals were 43.1 for whites and 36.6 for blacks. In regard to the 1982 Qualification Review Boards, the top 8 mechanical maintenance results belonged to whites, and 8 of the 10 lowest results belonged to blacks. Seven of the 8 highest paid mechanical maintenance technicians were white and 4 of the 6 lowest paid were black.

As far as Unit 2 is concerned, the data before the court is less extensive, but does reveal some information. Of the 22 mechanical maintenance technicians, 8 of the top 9 1982 Qualification Review Board results went to whites, and 6 of the 10 lowest results went to blacks. Whites were paid 7 of the top 9 salaries on curves 5 and 4^[6], while blacks were paid 8 of the 13 lowest salaries. Insofar as the career plan points were concerned, 8 of the highest point totals belonged to whites and blacks were given 6 of the 7 lowest totaled career plans by their managers. Within the oxygenation skill, of 9 technicians, the top 4 1982 Qualification Review Board scores went to whites and 3 of the bottom 5 scores went to blacks. The 4 highest totaling career plans belonged to whites and, of the remaining career plans, blacks received the two lowest.

The court wants to reiterate that these numbers are based on bare numerical observations made by the court. However, they do aid the court immensely in now deciphering what can only be described as technically confusing statistical evidence provided by the parties' expert witnesses. The court heard extensive testimony from these witnesses during the actual jury trial of the section 1981 claims and again on May 1, 1989, and will now attempt to capsulize their conclusions. Plaintiffs rely upon the expertise of Professor Jimmy Ramsey of Albany State University. Professor Charles Haweworth testified in rebuttal for Defendant.

Professor Ramsey used a standard deviation test to discern the degree to which chance might have contributed to the numerical differences between the salaries of whites and blacks at the Plant. The standard deviation is a measuring device for calculating "the likelihood that a given disparity occurred solely as a result of chance." Maddox v. Claytor, 764 F.2d 1539, 1551 n. 13 (11th Cir.1985). A statistician measures the deviation or disparity by comparing the actual salaries of the black employees to their "expected" salaries. For example, if ten technicians within a given skill of a given Unit are on pay curve 5 (the highest pay curve) then the statistician will assume that 4 of those technicians would be black because the Plant's population was approximately 40% black. Correlatively, if 10 technicians were being paid on the lowest pay curve, the statistician would assume that 6 of those technicians would be white. If nine of the 10 top paid technicians were white and only 1 black, then the difference between the expected number of blacks and the actual number of blacks on pay curve 5 would be -3, with the negative symbol signifying an underrepresentation of blacks. This numerical disparity is then divided by the calculated standard deviation to arrive at the "Z-value" for the numerical disparity. "This Z-value is an index of random probability." Id. A Z-value of greater than 1.96 or less than -1.96 denotes that something other than chance probably caused the differential. The Z Standard Deviation Test (Z Test) is based on a bell curve shaped theory of probability. The shape of the curve is established by the intervals of frequency distribution, in our case the pay scales. The bell curve analysis is based on a normal or symmetrical distribution, i.e., the variable distribution consisted of 50% above and 50% below the mean. Professor Ramsey used the Z Test to analyze the differences between the salaries of white and black technicians at the Buckeye Plant who were on the payroll as of December 31, 1985.^[7] In his analysis, the Professor only sought to discover whether or not a significant difference existed between the given salaries of whites and blacks at the Plant, and did not attempt to discover or consider any possible specific reasons for discerned disparities. Professor Ramsey first analyzed the salaries of technicians in Unit 1, the Woodyard. In 1980 and 1981 the Z-value of any disparity did not indicate any real significance between the salaries of whites and blacks, -1.48 in 1980 and -1.549 in 1981. In 1982, the year of the first Qualification Review Boards and the second career planning stage, the deviation became significant, reaching -3.019 in 1982 and remaining below -3.0 until 1986. Using a 1985 Plant-wide comparison of black and white technician salaries, Professor Ramsey calculated the Z-value to be -5.74, indicating a significant disparity between the salaries of all white and all black technicians.^[8]

Professor Ramsey also calculated the Z-value of the technicians subdivided by hire groups.^[9] Addressing the December 3, 1979, hire group, Professor Ramsey discovered a Z-value disparity of -2.307 as of the salary date December 31, 1985. The Z-value differential for the July 7, 1980, hire group was calculated to be -3.019. The July 7 hire group consisted of 85 technicians, 40 black and 45 white. The September 8, 1980, hire group, consisting of 55 total hires (30 blacks, 25 whites), had a Z-value differential of -2.199. All 3 of these Z-values, being below -1.96, allegedly represent a significant inequality in salary between blacks and whites.

Attempting to rebut Plaintiffs' statistical evidence as presented by Professor Ramsey, Defendant relies upon the testimony of Dr. Charles Haweworth. Dr. Haweworth first explained that 3 methods for determining statistical probability exist: (1) binomial approximation; (2) Fisher's Exact; (3) conventional probability theory. He then explained the probability equation for a standard deviation. Standard deviation is equal to the square root of n times p times q, where n equals the pool of events, q equals the probability of a particular event occurring and p is equal to 1-q, the probability of the event not occurring. Standard deviation represents an amount over or under probability. In addressing Professor Ramsey's analyses, Professor Haweworth indicated 4 problems with the calculations, 2 of which were technical and 2 of which were omissions. The technical problems were: (1) the data was not "normal" data; and (2) the statistical test administered by Dr. Ramsey was not valid. The 2 necessities omitted from Dr. Ramsey's calculations were: (1) hire date (length of service); and (2) prior training/experience.

Normal distribution is what many phenomena are, i.e. approximately the same distance above and below the average (a bell curve). Professor Haweworth then explained that the salary distribution at the Buckeye Plant was not symmetrical, in other words, the data was skewed. Specifically, while the white average at the Plant was $558.00, the 1985 salary figures showed that 118 whites were being paid below the average and only 78 above the average. As to blacks in 1985, only 46 were being paid below the black average while 95 black technicians had salaries above the black average.^[10] The 1984 salary figures exhibited a similar pattern. Dr. Haweworth testified that these numbers represented a skewed or abnormal distribution of variables and that accordingly a Z-value Standard Deviation Test was inappropriate under such circumstances.

Because of the allegedly skewed data, Professor Haweworth concluded that the appropriate means for calculating any disparity between white and black salaries at the Plant would be 1 of 3 methods: (1) mean or average; (2) median, the middle; or (3) mode, the most frequent value. Because of the skewedness of the relevant salary data, Professor Haweworth felt that the median was the most appropriate method, and showed the following:^[11]

              MEDIAN
             BLACK         WHITE
      1980   $288.00       $288.00
      1981   $394.00       $394.00
      1982   $452.00       $456.00
      1983   $504.00       $512.00
      1984   $542.00       $542.00
      1985   $542.00       $542.00

This testimony shows that the medians of black and white salaries at the Buckeye Plant were approximately equal, though in certain years the white median exceeded the black median of hundreds of technicians in a system based on average and which placed the bulk of its technicians in the middle salary ranges. Professor Haweworth's above-related explanation of abnormal or skewed data simultaneously explains his stated technical problem number 2, that the statistical tests used by Professor Ramsey were not valid.

In explaining the alleged omissions, Professor Haweworth stated that his analysis showed length of service to have a definite effect on technician salaries. In other words, he found a statistical relationship between salary and length of service at the Plant. As of 1985, the average length of service for whites was 65.9 months, and for blacks, 64.7 months.^[12] In concluding that prior training was a necessary element in comparing the black and white technicians' salaries, Professor Haweworth analyzed the initial employment applications of all particular curve members in relation to the particular job being performed (skill area). For example, of the 34 technicians on pay curve 5 (29 whites, 5 blacks), 12 were performing mechanically related jobs.^[13] Of these 12 mechanical technicians, 6 had extensive mechanical experience prior to beginning work at Buckeye, 3 had related experience, and 2 had nothing to tie them to mechanics.^[14]

The thrust of Professor Haweworth's testimony was an attempted rebuttal or invalidation of the statistical evidence of Professor Ramsey. At this point the court will not discuss its conclusions as to the conflict between these expert opinions. Having set forth the relevant facts, the court now sits prepared to discuss and reach its conclusions of law. Since these facts, as stated earlier, are simply skeletal, the court will supplement them as necessary below.

II. CONCLUSIONS OF LAW

A. JURISDICTION

Plaintiffs brought their claims of disparate impact under the provisions of Title VII of the Civil Rights Act of 1964, 78 Stat. 253, as amended, 42 U.S.C. § 2000e et seq. Defendant contends that these claims are time barred pursuant to the Supreme Court's recent holding in Lorance v. AT & T Technologies, Inc., ___ U.S. ___, 109 S. Ct. 2261, 104 L. Ed. 2d 961 (1989). Defendant argues that Plaintiffs are only complaining of the 1982 development of the Pay and Progressions System and not its implementation, and that accordingly Plaintiffs' charges were not timely filed with the Equal Employment Opportunity Commission (EEOC). Plaintiffs, for some reason, have not seen fit to respond to this contention; however, for several reasons the court cannot agree with Defendant.

Section 706(e) of Title VII requires that a "charge ... shall be filed [with the EEOC] within [the prescribed time period] after the alleged unlawful employment practice occurred." 42 U.S.C. § 2000e-5(e). The prescribed time period for Plaintiffs' claims was 180 days. Defendant claims that since most of the Plaintiffs did not file any charges with the EEOC until 1985, two or three years after the development of the Pay and Progression System, they were not filed within the prescribed time period, are not cognizable by this court, and must be dismissed. Defendant points to the Lorance decision for the proposition that a plaintiff cannot maintain a continuing violation theory of disparate impact resulting from the nondiscriminatory implementation of a system which supposedly had a discriminatory origin. The problem with this argument is that Lorance concerns an attack on a seniority system which is afforded special treatment by Title VII and the Supreme Court. See 42 U.S.C. § 2000e-2(h); and Trans World Airlines v. Hardison, 432 U.S. 63, 81, 97 S. Ct. 2264, 2275, 53 L. Ed. 2d 113 (1977). In order to maintain an action contesting a seniority system, a plaintiff must show intentional discrimination, which would naturally require reference to specific acts and events in time. The entire Lorance opinion is premised on the significance of the seniority system and cannot be applied in the situation before this court. The Lorance Court recognized the distinction between the seniority system there at issue and a normal claim of disparate impact in dismissing the contention that by not allowing a continuing violation challenge to seniority systems, the court must still allow plaintiffs to attack the system under a disparate impact theory. The Court apparently recognized that the very nature of a claim of disparate impact contemplates continuing violations. Justice Scalia stated that for the purposes of disparate impact "the statute of limitations ... [will] run from the time that impact is felt." Lorance, ___ U.S. at ___, 109 S.Ct. at 2267. The impact of the Pay and Progression System, if the court accepts Plaintiffs' contentions, is being felt even today at the Buckeye Plant. The Lorance Court's rejection of a continuing violation theory for seniority systems does not affect disparate impact claims against other types of practices and policies.

Even disregarding the seniority system distinction, the Lorance case holding does not require a dismissal of the claims now before this court. Justice Scalia cites the case of United Air Lines v. Evans, 431 U.S. 553, 97 S. Ct. 1885, 52 L. Ed. 2d 571 (1977), for holding that "[a] challenge to a neutral system may not be predicated on the mere fact that a past event which has no present legal significance has affected the calculation of seniority credit, even if the past event might at one time have justified a valid claim against the employer." Id. at 560, 97 S.Ct. at 1890 (emphasis added). This court cannot say that the Pay and Progression System "has no present legal significance." As a result of the System-wide freeze in 1984, the effects of the system were locked into place and any possibility of remedying the situation was stalled. The 1984 freeze gave the System "a present legal significance." Thus, even under the Lorance seniority system rational, the court must hear Plaintiffs' claims, as they were timely filed within the prescribed time limitations.

Assuming arguendo that Lorance could bar these claims, Plaintiffs would still have one final argument. Certain extraordinary circumstances can justify a court's equitable tolling of Title VII timing requirements: (1) when a state court action is pending; (2) when the defendant conceals an act supporting the Title VII cause of action; or (3) when the defendant misled the complainant regarding the nature of his rights under Title VII. Manning v. Carlin, 786 F.2d 1108, 1109 (11th Cir.1986). The court finds that the second extraordinary circumstance is applicable under the present facts and requires the court to equitably toll any timing deficiencies. Defendant admits that certain necessary information was not disseminated to technicians, which information, if known, would have supported this Title VII cause of action, e.g., the various assigned skill points, the significance of the Qualification Review Boards and the rankings, the interrelationship of the different parts of the Pay and Progression System, etc. Accordingly, the court will equitably toll any timing inadequacies as to Plaintiffs' disparate impact claims. For the foregoing alternative grounds, the court finds that it has jurisdiction to address Plaintiffs' remaining Title VII claims based on a disparate impact theory of liability.

B. DISPARATE IMPACT

1. BURDENS OF PROOF AND PERSUASION

In the case of Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971), the United States Supreme Court recognized the Title VII theory of recovery known as disparate impact, whereby an employee can challenge employment "practices that are fair in form but discriminatory in practice." Id. at 431, 91 S.Ct. at 853. Under the disparate impact theory, a facially neutral employment practice can be found violative of Title VII without a showing of an employer's subjective intent to discriminate which is a necessary element of an individual "disparate treatment" case. The Supreme Court revisited the proof structure which had developed through case law regarding the disparate impact theory and redefined its parameters in the recent case of Wards Cove Packing Co. v. Atonio, ___ U.S. ___, 109 S. Ct. 2115, 104 L. Ed. 2d 733 (1989).

Wards Cove involved Alaskan salmon canneries which hired unskilled cannery workers, most of whom were nonwhites, and noncannery workers, most of whom were higher paid whites. The cannery employees brought suit challenging the canneries' hiring/promotion practices under a disparate impact theory as causing a racial stratification of the canneries' work force because they denied nonwhites employment opportunities in noncannery positions. The Court of Appeals held that the plaintiff class had made out a prima facie case of discrimination and remanded the action for further findings regarding whether or not the employers could satisfy their burden of proving the challenged practices' business necessity. The Supreme Court first evaluated the Court of Appeals' definition of a prima facie case and analyzed whether such had been established by the plaintiffs at the district court level. Specifically the Supreme Court addressed the statistical evidence and its significance to a prima facie case. The Court, after deciding that the statistical evidence was not sufficient, then announced the subsequent proof and evidentiary scheme to be used upon a finding of a prima facie case.

To establish a disparate impact prima facie case, a plaintiff has the initial "burden of showing a causal link between challenged employment practices and racial imbalance in the work force ..." Id. at ___, 109 S.Ct. at 2125. Consequently, a plaintiff's disparate impact prima facie case must contain three elements: (1) articulation of a specific employment practice under attack, Watson, 487 U.S. at 994, 108 S.Ct. at 2788 ("[e]specially in cases where an employer combines subjective criteria with the use of more rigid standardized rules or tests," as is this case in the present action); (2) a racial disparity, usually shown through statistical evidence, International Bhd. of Teamsters, supra; (3) a specific causal link between the challenged employment practice and the statistical disparity. In discussing the statistical evidence necessary for the prima facie case, the Wards Cove Court made it clear that the comparisons must be relevant to similarly situated employees and the articulated practice, they must clearly indicate a racial stratification which is then shown to have been caused by a specific employment practice.

Once a prima facie case of discrimination is shown, the case then moves into what the Court called the "Business Justification" Stage, which

contains two components: first, a consideration of the justifications an employer offers for his use of these practices; and second, the availability of alternate practices to achieve the same business ends, with less racial impact.

Wards Cove, ___ U.S. at ___, 109 S.Ct. at 2125. The ultimate issue in the "Business Justification" Stage is "whether a challenged practice serves, in a significant way, the legitimate employment goals of the employer." Id. at ___, 109 S.Ct. 2125-26. In this Stage, the employer only bears a burden of production. The plaintiff has the ultimate burden of persuasion on either of the alternative theories of recovery: (1) the employer's business justification; or (2) less racially significant available alternative practices. When analyzing the evidence in the Business Justification Stage, district courts are cautioned to remember that "[c]ourts are generally less competent than employers to restructure business," Furnco Constr. Corp. v. Waters, 438 U.S. 567, 578, 98 S. Ct. 2943, 2950, 57 L. Ed. 2d 957 (1978). With this recently enunciated evidentiary scheme in mind, the court will now proceed to analyze Plaintiffs' disparate impact claims.

2. PRIMA FACIE CASE

Under Wards Cove, Plaintiffs must first establish a prima facie case by: (1) articulating a specific employment practice; (2) showing a racial disparity; and (3) demonstrating a causal connection between the specified employment practice and the racial disparity. In the present action, Plaintiffs point to Defendant's Pay and Progression System as the challenged employment practice. In attacking the System, Plaintiffs necessarily dispute its several components, i.e., the career plans, skill points, the ranking system, Qualification Review Boards, the pay curves, and the 1984 freeze. The court finds that Plaintiffs have satisfied step 1 of their prima facie case. Wards Cove, ___ U.S. at ___, 109 S.Ct. at 2124. The Pay and Progression System in operation at the Buckeye Plant until the 1984 freeze was an integral part of the Plant's function and was a definitive policy/practice of the company. The practically autonomous subjective authority given lower level management in carrying out the System makes it even more suspect. See Watson, supra. Having found that step 1 of the prima facie case has been satisfied, the court must now consider the statistical evidence presented by both parties.

Previously, the court meticulously described the relevant statistical evidence presented by the Parties' expert witnesses, Professors Ramsey and Haweworth. See infra., Part I.B. Plaintiffs rely not only upon the testimony of Dr. Ramsey, but on the raw number comparisons of the results of the Pay and Progression System, i.e., the high percentage of blacks in the lower realms of the career plan point totals, Qualification Review Board results, rankings, and pay curves/salaries. Plaintiffs, while recognizing the impact of small variable pools on Z and T Test scores, also show that Defendant's expert recognized that if the Z or T Tests are appropriate, then the test results signify meaningful disparities between white and black salaries at the Plant. Plaintiffs argue that the test scores, though questionable, support the numerical disparities shown through percentages and raw numbers.

Defendant counters by first arguing that Plaintiffs' statistical evidence is not probative of a disparity because the comparison must be between similarly situated individuals, which Defendant contends Professor Ramsey did not consider. Under Defendant's theory, a Plant-wide analysis is not valid because the comparisons must be done on a Unit and skill area basis, i.e., statistical comparisons must be very narrowly drawn. Defendant relies upon the testimony of Professor Haweworth to show: (1) that Professor Ramsey's analysis was wrong; (2) the variable distribution was skewed; (3) another test should have been used; (4) that the standard deviation tests were inappropriate under circumstances of this case; and (5) the proper test is to look at the medians, and upon comparing the medians one readily finds no discrimination. Finally, Defendant argues that Plaintiffs' statistics are useless because they do not consider any variables such as length of service, prior experience, training, etc. For the reasons set forth below, the court is convinced that a statistical disparity exists at the Buckeye Plant between black and white technicians in terms of pay.

The court will not regurgitate the statistical evidence as set forth previously, but will address Defendant's attacks on Plaintiffs expert witness and other statistical evidence. Regarding Defendant's first contention as to the scope of the comparisons, the court is not convinced that under the present situation narrowly drawn comparisons are absolutely necessary. All technicians, Plant-wide, were hired on an equal basis, were allegedly given equal opportunities and were given equal pay. The Buckeye Plant operated under a technician system, whereby employees learned primary, secondary and cross skills in order to facilitate a more efficiently run operation. Because of the technician system, its function and operation, the court is convinced that Plant-wide analyses are relevant in a limited sense. Under Buckeye's system, all technicians are similarly situated individuals. Moreover, the statistical evidence is not limited to Plant-wide comparisons. As shown above, Plaintiffs and their expert presented some Unit comparisons as well as skill area comparisons. The statistical comparisons utilized by Plaintiffs were not only narrowly drawn for specific comparisons, but also provided a broader perspective of the entire Plant.

Professor Ramsey used the Z-value standard deviation statistical analysis. Though the variable pools which he used were at times small, the test results convince the court that they are at least partially reliable. Professor Haweworth contends that the Z test is inappropriate because of the skewedness of the data. The court cannot agree. As stated earlier, the court finds that only through an analysis of extremes can we truly discern a meaning from numerical comparisons because of Buckeye's "average" approach and its deceptive nature.^[15] Thus, the skewedness of the data, rather than being a detriment to Plaintiffs evidence, further supports a statistical difference. Professor Ramsey's analysis is supportive of step 2 in Plaintiffs' prima facie case.

As to Professor Haweworth's testimony that other tests would have been more appropriate, the court, not being an expert in statistical analysis, will not second guess Professor Ramsey's choice of testing methods. The court recognizes that experts within any given field may differ in their opinions. While not choosing Professor Ramsey over Professor Haweworth, the court will give some deference to Plaintiffs' expert since Plaintiff does bear the burden of proof. Additionally, the median test suggested by Professor Haweworth contravenes the court's earlier finding that the more relevant information is to be found at the outer regions or extremes of a system based on averages.

Defendant's final argument concerns the consideration of variables. While the consideration of variables is important, the court is not convinced that Professor Ramsey's failure to specifically consider those factors delineated by Defendant defeats his statistical data. In many ways, the noted variables are considered because they are an integral part of Buckeye's formula. Moreover, the purpose of step 2 of the disparate impact prima facie case is to simply show a statistical disparity, not to prove a causal connection. Some of the variables suggested by Defendant should not have made any significant difference. For example, all of the technicians were hired within an approximate 9 month time period, were paid the same salary, and were subsequently placed on career plan courses not particularly based on length of service as much as supposed ability. Training and prior experience may have contributed to salary and progression, but according to Defendant's own evidence, technicians were not always given their first choices of skill areas, the preferable jobs were filled early, and not all hires with prior experience in a particular area were placed in their areas of prior expertise. While other variables may be somewhat important, they do not defeat Plaintiffs' evidence, as oftentimes they were factored into the System, were not extremely important, or were not factors at all.

In analyzing Plaintiffs' statistical evidence, the court is mindful that "statistics are not irrefutable; they come in infinite variety and, like other evidence, they may be rebutted. In short, their usefulness depends on all of the surrounding facts and circumstances." International Bhd. of Teamsters, 431 U.S. at 340, 97 S.Ct. at 1856-57. Though Defendant did attempt to rebut Plaintiffs' evidence, it succeeded only partially. As stated earlier, the court is only relying on Professor Ramsey's testimony as supportive of the raw numbers, averages, and other relevant comparisons set forth infra at Part I.B. Viewing Plaintiffs' expert's testimony together with the court's observations above and the anecdotal evidence at trial regarding the individual claims, see Kilgo v. Bowman Transportation, Inc., 789 F.2d 859 (11th Cir. 1986), the court finds that Plaintiffs have shown that a statistical disparity exists at the Buckeye Plant in regard to black and white salaries.

Step 3 of the prima facie case requires Plaintiffs to establish a causal connection between the contested employment practice and the apparent statistical disparity. Defendant can hardly dispute that the Pay and Progression System was the primary if not sole method of determining a technician's salary. The career plans and point totals determined a technician's pay curve, in conjunction with the ranking system. For the purposes of Plaintiffs' prima facie case, the court finds that a causal connection existed between the exhibited racial disparity and the Pay and Progression System. Consequently, the court must now examine the evidence regarding the second stage of the Wards Cove disparate impact analysis, the Justification Stage.

3. BUSINESS JUSTIFICATION STAGE

Upon a prima facie showing of discrimination based on a disparate impact theory of recovery, the case moves into the second evidentiary stage known as the Business Justification Stage. Wards Cove, ___ U.S. ___, 109 S.Ct. at 2125. The eventual question in this examination concerns the extent that the challenged practice furthers the employers legitimate employment goals. Id. This Stage contains two alternative aspects: (1) the proffered business justification; and (2) available alternative practices. Id. If a plaintiff can persuade the court on either of these aspects of this Stage of the case, then he can recover.

a. Employer's Justification

In the first part of the Business Justification Stage,

[t]he touchstone of this inquiry is a reasoned review of the employer's justification for his use of the challenged practice.

Wards Cove, ___ U.S. ___, 109 S.Ct. at 2126. The practice need not be "essential" or "indispensable." Id. Thus the court must look at Defendant's propounded business justification, and Plaintiffs' counterarguments and evidence regarding the pretext of such justification.

Defendant contends that the Pay and Progression System and its various elements serve the legitimate goals of the Buckeye Plant and its technician system. Defendant's intent and design in utilizing the technician system was to develop a multi-skilled, practically autonomous work-force which would ultimately benefit the company through increased productivity and reduced production costs. The system was also designed to benefit the technicians through increased flexibility and higher pay. Buckeye's use of the technician system nation-wide had been shown to be successful. In setting up the Plant, Buckeye sought to incorporate the successful features of previous technician systems and avoid the less productive features. The ideal skill mix constituted that mix of skills ideally required to operate the Plant and support the technician principles. The ranking system ensured that contribution was a factor which impacted on a technician's pay and progression. The Qualification Review Boards measured a technician's progress in his assigned skill areas. The skill point values were assigned to value particular skills and control costs. The career plan process broke job classification barriers and offered technicians broad job responsibilities in numerous skill areas and allowed them to contribute Plantwide in a variety of ways. The career plans, in particular, contributed to the autonomous goals of the Buckeye Plant.

Plaintiffs' sole argument with regard to the particular business justifications specified by Defendant is that Plaintiffs somehow feel that Defendant has completely failed to advance any probative business interest such that would justify its use of the Pay and Progression System as it operated prior to the 1984 freeze. In addition Plaintiffs contend that Defendant has not articulated a business justification for the 1984 freeze. Plaintiffs rely upon the Eleventh Circuit decision in Craig v. Alabama State University, 804 F.2d 682 (11th Cir. 1986). Craig stands for the proposition that in the absence of evidence advancing an employer's legitimate business justification, a court should not engage in a "common sense" drawing of inferences from the evidence in order to satisfy the defendant's burden. Id. at 690. Plaintiffs contend that for this court to find that Defendant has shown a legitimate business justification, it would necessarily have to engage in speculation to create the justification in the stead of hard evidence and articulated argument. The court cannot agree. Defendant presented ample evidence at trial regarding the technician system, its use at the Plant, its purposes and the reasons behind its use. Defendant's goal in using the Pay and Progression System was an ultimately autonomous Plant where any number of technicians were interchangeable without any variation in ability to run the Plant. Plaintiffs have the eventual burden of persuasion that they were denied employment opportunities because of their race. Plaintiffs must persuade this court that Defendant had no legitimate business justification for the System. Wards Cove, ___ U.S. at ___, 109 S.Ct. at 2126. Plaintiffs have failed in this burden. Their only argument, apparently, is that Defendant has failed to articulate and show a business interest. The court finds that Defendant has satisfied its burden of producing evidence, and that Plaintiffs have failed to persuade the court that Defendant's articulated reasons do not justify its use of the System. The court reaches this conclusion based upon the evidence and without speculation. Consequently, Defendant has prevailed on Part One of the Business Justification Stage of this case and the court must now move on to Part Two.

b. Available Alternative Practices

Though Plaintiffs have failed to persuade the court on Part One, they may still prevail by showing that "other tests or selection devices, without a similarly undesirable racial effect, would also serve the employer's legitimate interest[s] ..." Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S. Ct. 2362, 2375, 45 L. Ed. 2d 280 (1975). By showing available alternative practices, the plaintiff would satisfy the core issue of the Business Justification Stage: whether the challenged practice was merely a "pretext" for discrimination. Id. In evaluating the evidence at this Stage of the action, the court, as noted above, is reminded that it, in many instances, must bow to the business acumen of the employer. Furnco, supra. The present circumstances offer a very unique situation which the court must evaluate in terms of disparate impact, Wards Cove, and the effect of any possible remedy.

To satisfy their burden of proof as to available alternatives, the Plaintiffs primarily rely upon the fact that Defendant froze the old Pay and Progression System in 1984 and has subsequently instituted a new, less racially impacting, system. Defendant concedes that a freeze took place in 1984 and recognizes that the system now being utilized is not the old System, but Defendant contends the new system is simply a modification of that old System, which was never intended to be permanent. In addition Defendant argues that the modifications made to the old System were minor alterations and that the System and its components, to a large extent, remain intact. Plaintiffs counter that the new system, though similar to the old, advances more objective policies with less racially disparaging effects. The court agrees with Plaintiffs.

In 1984, as a result of unstable Plantwide operating costs and reliability, Buckeye froze the Pay and Progression System in order to maintain the status quo and restabilize. Once operations improved and at the urging of the technicians, a task force began investigating alternatives to the old System. As the task force worked, it conducted surveys to discern ideas and approval from management and technicians. Rather than implement an entirely new system, the task force decided that a modification of the old System would better serve the needs of the company. The new system contained many of the general elements of the old System, but provided for more technician input, and more objective standards and criteria. While technicians are still required to go through qualification procedures, the tests are now objective and written, technicians also have access to the tests. Both managers and technicians rank employees against a specific set of criteria. Career plans are now known as "skill blocks." As before, skill areas are assigned a numerical value and the point total of a skill block determines the technician's pay curve. Upon implementation, however, no adjustments were made as to individual technicians regarding any discrepancies or disparities which resulted from the old System.

As to Defendant's contention that the new system is not an alternative system but simply the old System slightly modified, the court finds that the significance of the modifications in relation to this case make the present system a less racially impacting alternative practice. The modifications implemented by Defendant go to the very heart of Plaintiffs' complaints with the old System. Managers no longer have practical subjective autonomy, technicians have more input to the system, the standards and criteria are now more objective, and technicians now know and understand the system and how it works. The new system removes the secrecy of the old System and the almost sovereign managerial power of the old System, both of which allowed much room for personal biases and prejudices.

The court is convinced that the new system contains the very type alternative practices contemplated by the Wards Cove Court. The problem now facing the court is that Wards Cove appears to contemplate a situation where suggested alternatives are rejected by the employer. ___ U.S. at ___, 109 S.Ct. at 2126. Here, the proposed alternative has been implemented by Defendant. To now impose a remedy against Defendant would, in effect, punish Defendant for taking remedial measures. However, in instituting the new policies, Defendant did not adjust the pay and progression of the affected blacks to answer the racial disparities evidenced in Plaintiffs' prima facie case. Because the disparate effects of the old Pay and Progression System still linger at the Plant, the court finds that some type of remedy must be had. Plaintiffs have carried their burden of showing available alternative practices which have a less racially-disparate impact on blacks and the court must address the question of remedy.

c. Remedy

Since the new policy is now operating at the Plant, the court need not address the implementation of any alternative practices. The sole remedy now available is some method of erasing or equalizing the racial disparities caused by the old System. Because of the expertise within the industry required to address a remedy and because of the lack of sufficient evidence, the court will not, nor can it, decide an appropriate remedy at this time. A subsequent hearing will be held to discuss the remedy stage of this case.

III. CONCLUSION

As set forth above the court finds that Plaintiffs have established their claims of disparate impact against Defendant Buckeye. The issue of a remedy is reserved, pending further proceedings. Because of the implications of this Order and the extensiveness of any subsequent remedial stage of the case, the court also finds that "there is no just reason for delay," Fed.R. Civ.P. 54(b) and directs the Clerk of court to enter judgment as to the liability aspects of these actions.

NOTES

[1] The Lincoln case stands for the proposition that a district court is bound by a jury's conclusions in individual claims of discrimination.

[2] Defendant's corporate name changed in the midst of trial to Proctor & Gamble Cellulose Corporation, but for the purposes of this Order, Defendant will be referred to as "Buckeye."

[3] The term "technician" has no practical significance, nor does it relate to either the difficulty of the job to be performed or the education of the employee.

[4] As Buckeye began to hire in groups, certain jobs were either necessary for the start-up of the Plant or were preferred position. Thus these position were filled by the early hire groups and not available to the later hires.

[5] The court notes that during the period of time under scrutiny here the work force at Buckeye was fluid, i.e., the technicians often moved from one Unit to another because of transfer or cross-training. As a result of this fluidity, the number of technicians within the Units varied. The court, however, will attempt to use total numbers when possible. Moreover, the court finds that any slight variation in numbers, about which Defendant might argue, is not relevant to the overall thrust of the numerical findings.

[6] At this juncture of the Plant's history, technicians were paid on one of 5 pay curves, with pay curve 1 containing the lowest salaries and pay curve 5 paying the highest.

[7] Professor Ramsey analyzed the salaries of all hourly wage earners at the Plant, excluding administrative and clerical personnel. He did not subdivide the technicians in job classifications on either a Plant-wide or Unit-by-Unit basis.

[8] The court finds this disparity of particular interest because, as both experts testified, the larger the sample size, the more accurate the calculation/analysis, and concomitantly, the lower the number of variables, the less accurate the results. Professor Ramsey considers 10 to be a minimal number of sample variables for a reliable Z-value, but recognized that some statisticians might consider 30 to be the minimal number. A Plant-wide sample of all technicians is the largest sample under these circumstances. Defendant argues, however, that in order for any of these statistics to have meaning here, they must compare technicians within the same Units and the same skill area. Defendant further argues that because the Unit and skill area sample sizes would be small, then the Z Test is inappropriate for the purposes of this case. The court is troubled by these arguments for several reasons, and will consider all of the statistical evidence before reaching a final decision and discussing Defendant's contentions, infra, Part II, B, 2.

[9] Length of service is one factor which Defendant contends contributed to any pay disparity between blacks and whites, because, as stated earlier, the first hire groups were heavily white-populated and Buckeye only achieved its desired 60%/40% white-to-black ratio by having its last hire group composed of 50% blacks.

[10] Professor Haweworth further testified that the average salary of Plaintiffs was $27.00 below the black average.

[11] As noted earlier, an analysis of the average or mean in regard to technician salaries at the Plant concerns the court because of Buckeye's quest for attaining averages both Plant-wide and by Units in regard to pay curves, career plan point totals, and rankings.

[12] Professor Haweworth prepared a series of documents to apparently show that length of service partially explains any disparities between black and white salaries. While the court understands this argument, it also notes that these documents consistently bear out the Plaintiffs' contentions regarding the top-heaviness of white salaries and the bottom-heaviness of black salaries, though, as pointed out by Professor Haweworth, the bulk of all salaries were in the middle of the salary range. Upon cross-examination, however, Professor Haweworth recognized Plaintiffs' contention. Moreover, Professor Ramsey's hire group Z-values exhibit a significant disparity between blacks and whites within the same hire group.

[13] Though Professor Haweworth analyzed the job applications in relation to related prior experience, he did not look to see whether any technicians had prior experience/expertise in a particular area but had not been placed in a related skill position.

[14] Professor Haweworth also conducted T-value Standard Deviation Tests on the differences between black and white salaries. The T Test is similar to the Z Test, but is allegedly more accurate on smaller pools of variables. The Professor analyzed particular skill areas within each Unit; however, he also continued to contend that both the T and Z Tests were inappropriate because of the skewedness of the data. These Tests assume a normal distribution. Because Professor Haweworth relied so heavily on his contention that these Tests were unreliable, the court will not discuss his findings.

[15] See infra p. 351 and n. 9.