Full opinion text
OPINION AND ORDER FULLAM, District Judge. This action was filed by the Attorney General against the International Union of Elevator Constructors, Local Union No. 5, alleging that the local union has engaged in a pattern or practice of resistance to the full enjoyment of the employment rights guaranteed by Title VII of the Civil Rights Act of 1964 and has interfered with the implementation of Executive Order No. 11246. The Equal Employment Opportunity Commission has been substituted for the United States as plaintiffs hérein, pursuant to the provisions of the Equal Employment Opportunity Act of 1972. The thrust of the Government’s case is that Local 5 has used its control of work opportunities in the elevator construction industry in the Philadelphia area to prevent blacks from entering or advancing in the trade, and, further, has interfered with the attempts of elevator contractors to meet the minority employment goals established by the Philadelphia Plan.a Background 1. Local 5 and its Collective Bargaining Agreements. International Union of Elevator Constructors, Local Union No. 5 (Local 5) is an unincorporated association of persons engaged in the elevator construction trade in the Philadelphia, Pennsylvania, area.b Its principal office is in Philadelphia. As of October 1973, Local 5’s membership totaled approximately 700 persons. The members of Local 5 work for approximately 30 elevator contractors, a group which includes all the major elevator contractors in the Philadelphia area and which performs most of the work in the elevator construction trade in that area. These contractors and Local 5 are, and since at least 1967 have been, bound by the terms of collectively-bargained labor agreements, negotiated by the International Union of Elevator Constructors on behalf of Local 5 and by the National Elevator Industry, Inc. (NEII) on behalf of the contractors. The first of these agreements — the “old” Standard Agreement — was in effect from January 1, 1967, to March 24, 1972; the “new” Standard Agreement has been effective since March 24, 1972. Under the terms of these agreements, Local 5 is recognized as the exclusive bargaining representative for all mechanics and helpers engaged in the elevator construction industry within Local 5’s geographic jurisdiction. 2. Worker Classification, Initiations, and Promotion. Under the terms of the Standard Agreements, members of Local 5 are classified as mechanics or helpers. Helpers are paid 70% of the mechanics’ rate. Generally, helpers are only eligible to become mechanics after they have completed two years’ employment in the industry and have passed an examination administered by a joint industry-union examining committee. The Standard Agreements provide that, to be eligible for membership in Local 5, persons must have worked in the industry as probationary helpers for at least 100 hours per month for six months within a nine-month period. Probationary helpers are paid 50% of the mechanics’ rate during their probationary period. At the end of the probationary period, their pay automatically increases to the helpers’ rate, that is, to 70% of the mechanics’ rate. This step up in pay is unrelated to initiation into membership of Local 5. It is Local 5’s normal practice to admit into union membership all helpers who have completed their probationary periods whenever an initiation is held. Initiations are scheduled by the union’s executive board, which generally bases its decision about when to enlarge membership on its view of work and employment trends in the industry. The apparent aim of the union is to avoid taking in new members when large numbers of experienced union men are unemployed. In recent years, initiations have been scheduled once every year or 15 months. 3. Work Permits and The Referral System. The Standard Agreements have required that all mechanics and helpers employed by the elevator construction contractors subject to the agreements be union members or hold union referral cards (also called “work permits”) within 30 days of their employment. Thus, to work for an employer as a probationary helper, an employee must obtain a work permit from Local 5. a. Unemployment in the Industry. It has been Local 5’s firm policy to refuse to issue work permits to new men at times when union members are out of work. Article XXII of the new Standard Agreement provides that before an employer can hire a new, inexperienced worker, the employer must first check with Local 5 to ascertain that none of its members are unemployed. Where there are a number of union members out of work the employer has the right to name the particular member it wants Local 5 to refer for employment. If the employer expresses no preference, officials at Local 5 decide whom to refer. The protection of work opportunities for union men, achieved through this system of controlling the flow of new entrants into the trade, extends beyond the members of the local union. Members of other locals of the International Union of Elevator Constructors may have their names added to Local 5’s list of men out of work. If, for example, there is unemployment among members of Local 6 in Pittsburgh, the men out of work there can secure rights of first employment in the Philadelphia area. Although there are obvious economic and psychological factors which limit the free mobility of workers from one part of the country to another, it remains a fact that the re-hire rights of union members are industry-wide in nature. Probationary and 70'% helpers enjoy similar re-hire rights. With respect to the list of workers out of work, from which employers must make their selections and Local 5 must make its referrals, the new Standard Agreement provides that: “The Union shall establish, maintain and keep current an open list for the employment of workmen qualified to perform the duties required. Such list shall be established, maintained and kept current on a non-discriminatory basis and shall not be based on or in any way affected by Union membership, Union Bylaws, regulations or constitutional provisions or any other aspect or obligation of Union membership, policies or requirements.” New Standard Agreement, Art. XXII, If Ka). The Standard Agreements do not require that an employer, when exercising its right to select workers from the list, hire union members before hiring probationary and 70% helpers. The only requirement is that: “An employer shall hire experienced mechanics and helpers who permanently live in the area, are seeking employment and are qualified to perform the work required by the employer before hiring a transient employee or a new inexperienced employee.” New Standard Agreement, Art. XXII, H 1(c). However, as would be expected, employers are usually interested in hiring union members, since generally these workers have had greater experience in the elevator construction trade. The procedures which existed before the effective date of the New Standard Agreement allowed Local 5 much greater control over the employers’ exercise of their selection privileges. The Union kept only a “rough list” of men out of work, and would usually limit an employer’s choices to the more senior workers. If there were both union and nonunion men out of work, the employer generally received only the names of the union men from which to make its selection. (Under the New Standard Agreement, the employer has access to the full list of available workers.) Where the employer simply requests Local 5 to refer a worker, without naming him, the Union follows no hard and fast rules in determining whom to refer. It appears that officials of Local 5 base the order of referral sometimes on the worker’s seniority in the industry, and sometimes on the date of the worker’s last employment, with a large amount of discretion residing in the Union officials. It is fair to conclude, however, that union members are almost always referred for employment before probationary and 70% helpers. b. Full Employment and New Permits. Local 5 will issue new permits when none of its members or permit men are out of work. In such a situation, if an employer needs to expand its work force on an outside job, and if it has in mind a particular prospect for the position, such as an employee working in its shop or a friend or relative of an employee, the employer sends that person to Local 5 for a work permit. It appears that somewhere between 65 and 75% of all new probationary helpers attain their status in this way. In the remaining cases, the employer asks Local 5 to send it a new probationary helper, and Local 5 decides whom to recruit and refer. The Union keeps on file the applications for work permits submitted by persons seeking new employment in the industry. Local 5 has offered no explanation of the standards, if any, it uses to select applicants to fulfill employers’ requests for new probationary helpers. Of the applications for work permits on file as of April 27, 1973, and filed before March 15, 1972, approximately 75% were from applicants who were either related to or recommended by a union member. Less than 4% of all the applications were filed by black persons. 4. Layoffs and Seniority. Both the Old and New Standard Agreements require employers to lay off probationary helpers before laying off workers who have completed their probationary periods. This is the only restriction on the order of layoffs contained in the agreements. Among probationary helpers, a worker who has completed 5% months of his 6-month probationary period may be laid off before an employee with only a few days on the job. Similarly, a union member with years of experience may lose his job before a member with less seniority. In fact, nothing in either Standard Agreement requires that a senior union member not be laid off before a worker who has recently completed his probationary period but not yet been initiated into union membership (a 70% helper). However, officials of Local 5 generally do attempt to persuade employers to lay off union members with less seniority before members with more, and to lay off 70% helpers before card helpers. These efforts — a phone call or two to the employer — often succeed, but not always. Union members enjoy no collectively-bargained seniority rights, except vis-a-vis probationary helpers. Statistical Data 1. Membership of Local 5. When the instant action was filed, in March 1972, the total membership of Local 5 was 663. Of that total, only 6 members, or less than 1%, were black. In January 1973, two black probationary helpers were initiated into membership, and in August 1973, 12 new black members were added. Consequently, as of October 1973, by which time the local’s membership had grown to approximately 700, the black membership numbered 20, or about 3% of the total membership. The 14 black members initiated in January and August 1973 constitute about 10% of the 135 men admitted to Local 5’s membership between the date of filing of this suit and October 1973. 2. Work Permits Issued by Local 5. As of March 1972, Local 5 had issued only 11 work permits to black men. Two of these were issued pursuant to Orders of the Philadelphia Commission on Human Relations and a Philadelphia Common Pleas Court judge. Although the record does not clearly disclose the total number of work permits issued by Local 5 as of March 1972, it appears that between early 1967 and March 1972, approximately 215 men received permits. It follows that the nine permits issued to black men represent, at most, 4.2% of all permits issued by Local 5 as of the date of filing of this action. The number of permits issued by Local 5 to white men and black men for each year between 1969 and March 15, 1972, is shown below: 1969 53 Total 53 Whites 0 Blacks 1970 107 Total 101 Whites 6 Blacks 1971 6 Total 4 Whites 2 Blacks 1/1/72 to 0 Total 3/15/72 0 Whites 0 Blacks Between the date of filing of this suit and October 1973, 51 black men received work permits. These permits represent over 21% of the work permits issued by Local 5 over the same period. I have arrived at the following numerical conclusions regarding the percentage of white and black applicants for work permits who were granted permits : Between May 1967 and April 1973, of the 775 persons who applied to Local 5 for the work permits, 365 received them. This represents an overall issuance rate of approximately 47%. Over that same period, Local 5 issued permits to between 41 and 45 of the 59 black men who applied, representing an issuance rate to blacks of between 70 and 75%. Further, I find that between May 1967 and March 15, 1972, the date of filing of this action, 439 persons applied to Local 5 for work permits, and, of that total, 215 were issued permits. This represents an overall issuance rate of about 49%. Over that period, 9 of 17 black applicants received permits, yielding an issuance rate to blacks of 53%. 3. Referrals of New Probationary Helpers by Local 5. In the experience of the five or six elevator contractors who testified on the subject at trial, Local 5 has referred a black union member for employment only once, in October 1973. However, there is no evidence concerning how often, if at all, black union members have been out of work, or whether unemployed black union members (if at some time there were some) ever sought job referrals from Local 5. In the experience of the same elevator contractors, Local 5 has only once referred a black probationary or 70% helper. In that case, apparently, the individual had previously received a work permit in connection with employment with another contractor. Thus, there has never been an instance where Local 5 has referred a black probationary helper who was not already a per-mitholder. That is, on those occasions when the union has looked to its file of applications for work permits to fulfill an employer’s request for a worker, it has never selected a black applicant. 4. Mechanics. As of November 1973, Local 5 had only one black mechanic member, Albert Patterson. It appears from Exhibit No. 11, “List of Union Members,” that slightly more than half of Local 5’s 700 members are mechanics. 5. Local 5’s Geographic Jurisdiction: Racial Composition and Distribution of Union Membership. The following information may be derived from demographic data contained in a 1970 United States Bureau of the Census report entitled “General Social and Economic Characteristics — New Jersey” : City of City and Pa. Suburbs of Local 5’s Phila. Phila. Jurisdiction Percent of blacks in the total population 33.5% 19.4% 15.9% Percent of blacks in the civilian labor force of males over 16 28.5%' 15.9% 13.0% Percent of blacks in the unemployed civilian labor force of males over 16 43.3% 31.6% 26.1% Percent of blacks in the category of unemployed males over 16 whose last occupation was “Craftsman, Foreman, or Kindred Worker” 36.5% 24.0% 20.1% More than half of Local 5’s members work in the City of Philadelphia. It appears that the proportion of the Union's membership working in the City or in the four surrounding Pennsylvania counties is between 75 and 80 percent. Most of the remaining members are employed on jobs in New Jersey. Only a relatively few members, representing approximately 5% of Local 5’s membership, work in Delaware. Instances of Alleged Overt Discrimination 1. Issuance of Work Permits. A. Kenneth Reid. In December 1969, Kenneth Reid, a black man, called Local 5 to inquire about obtaining a work permit. He was told that there were union men out of work, and, apparently, was effectively discouraged from applying for a permit. Whether he was told that applications were not being accepted is not clear. Mr. Reid was hired on January 23, 1970, by Security Elevator Company (Security), a company with several federal contracts and, as such, subject to the obligations of the Philadelphia Plan. On March 18, 1970, he did apply to Local 5 for a permit, but none was issued. When, some two months later, he asked about the status of his application, he was told that he could not be granted a permit because there were still union men out of work. From January 1970 to June 1971, officials of Security made at least seven telephone calls to Local 5 to request that a work permit be issued for Mr. Reid so that Security could increase its “minority time.” Each time the employer was told that no permit could be issued because there were union men out of work. Unable to secure a work permit for Mr. Reid, Security was reluctant to send him to a construction site, and, instead, assigned him to work in its shop. However, after 18 months in the shop, and still without a permit, Mr. Reid was told by his employer to report to a federally funded job site at 4th and Arch Streets, Philadelphia. For several weeks thereafter, Mr. Reid reported to the construction site every day but was given no work to do. The president of Security explained that the company was willing to featherbed in this way to increase the amount of “minority time” on the project to satisfy Philadelphia Plan requirements. On July 7, 1971, Mr. Reid was finally issued a work permit. It appears that Local 5 had agreed to do this in return for Security’s commitment to hire three union workers, currently out of work, at a time when Security did not need additional men. Between December 1969, when Mr. Reid first called Local 5 about obtaining a work permit, and July 7, 1971, when he finally received one, 108 white men were issued permits, as follows: No. of Work Permits Month Issued to White Men December 1969 4 January 1970 1 February 1970 5 March 1970 34 April 1970 9 May 1970 4 June 1970 5 July 1970 17 August 1970 17 September 1970 9 October 1970 1 November 1970 1 December 1970 0 January 1971 1 February 1971 0 March 1971 0 April 1971 0 May 1971 0 June 1971 0 July 1971 0 Over the same period, six blacks other than Mr. Reid received work permits, all issued in 1970. Two permits were issued to blacks within a week of Mr. Reid’s application, one on March 16, 1970, the other on March 24,1970. B. Jessie Oliver. On September 16 or 17, 1970, Jessie Oliver, a black man, applied for a job with Westinghouse Electric Corporation (Westinghouse), and, one week later, he was hired. On September 23, 1970, Mr. Oliver applied to Local 5 for a work permit. However, the business manager for Local 5, Robert Williams, told him that no permit could then be issued because there were union men out of work. When a Westinghouse supervisor called Local 5 to request that Mr. Oliver be issued a permit, he received the same response. Mr. Oliver was promptly sent by Westinghouse to work at the Veterans Stadium construction site, notwithstanding his inability to obtain a work permit. When the Local 5 mechanic in charge at that work site learned that Jessie Oliver had no permit, he told Mr. Oliver that he could not work without a permit, and called Business Manager Williams. The mechanic was instructed to have the Local 5 members employed by Westinghouse at that site walk off the job. Those instructions apparently were carried out. Westinghouse continued to employ Mr. Oliver, but assigned him to work in a stock room rather than send him to outside construction jobs. Mr. Oliver thereafter filed a complaint with the Philadelphia Commission on Human Relations, which then brought charges of discrimination against Local 5. On November 20, 1970, following a hearing, the Commission ordered Local 5 to issue Mr. Oliver a work permit and to accord him seniority dated back to September 23, 1970, when Mr. Oliver first applied for a permit. The permit was issued to Mr. Oliver on December 7, 1970. Westinghouse then sent him back to the Veterans Stadium job site, where he worked until the job was completed. From the beginning of September 1970 through September 23, 1970, when Mr. Oliver first applied for a work permit, eight white men were issued permits. Between September 23, 1970, and December 7, 1970, three white men were issued permits. In fact, permits were issued to white men on September 22, 1970, and September 24, 1970, the days before and after Mr. Oliver’s application was rejected. However, the table of the number of permits issued to white men in 1970 and early 1971, above at p. 1247, suggests that October 1970 marked the beginning of an extended period of unemployment in the elevator industry. C. Robert Scott. In July 1971, the Corbett Elevator Company (Corbett) met Robert Scott, a black man, through the Joint Apprentice Outreach Program, and directed him to obtain a work permit from Local 5. Mr. Scott applied for a permit on July 15, 1971, and the president of Corbett called a Local 5 official to request that the application be granted. Although Local 5 did not issue Mr. Scott a work permit when he first applied, three weeks later the union notified Corbett that a permit would be issued. By that time, however, Mr. Scott had found another job. Accordingly, no permit was issued. By reference again to the table at p. 1247, it is apparent that, with the exception of the permit issued to Mr. Reid on July 7, 1971, no work permits were issued by Local 5 to anyone, white or black, from February through July 1971. In late August 1971, Mr. Scott left his new job and was hired by Corbett to work in its shop. After working there a year, Mr. Scott again sought to obtain a work permit so that he could become a union helper. Although Mr. Scott’s application of July 26, 1972 was at first denied, it appears that, soon thereafter, Local 5 was again prepared to issue Scott a permit. Before this could be accomplished, however, Scott was injured in an automobile accident and was unable to work for six months. Scott has never received a work permit from Local 5. In May and June 1972, no permits at all were issued. In July 1972, Local 5 issued two permits to white men. It appears that the elevator construction trade then experienced an upswing in activity, for in August and September 1972, the union issued 45 and 33 permits, respectively. In each of those months, three of the workers receiving permits were black men. D. Harold Johnson. On September 10, 1971, Harold Johnson, a black man, was hired by Energy Elevator Company (Energy). On the same day, at Energy’s instructions, Mr. Johnson applied to Local 5 for a work permit, but was informed by Business Manager Williams that no permits could be issued because there were 40 union men out of work. On September 13, 1971, Mr. Johnson was sent by Energy to work at a construction site at 22nd and Venango Streets, Philadelphia. However, when the mechanic in charge of the Local 5 workers at that site discovered that Johnson had no work permit, he told Johnson that he was not permitted to work there. Mr. Johnson’s employer nevertheless directed him to report to the construction site daily, and, accordingly, between September 13, and October 5, 1971, Johnson spent eight-hour days at the job site but did no work. During this period, he was kept on Energy’s payroll. Representatives of Energy made several calls to Local 5, explaining Energy’s Philadelphia Plan obligations and seeking to obtain a permit for Mr. Johnson. On October 5, 1971, Mr. Johnson filed a complaint against Local 5 with the Philadelphia Commission on Human Relations, and on October 6, 1971, in an action brought by the Commission on Mr. Johnson’s behalf, Common Pleas Court Judge Ned Hirsh ordered Local 5 to issue Johnson a work permit. Mr. Johnson received a permit later that day. From October 7, 1971, to February 1, 1972, Mr. Johnson was employed by Energy as a probationary helper at the 22nd and Venango location and several other job sites. His employer laid him off on February 1,1972. For the year 1971, a total of six work permits were issued. Four permits were issued to white men and two to black men, Kenneth Reid (July 7, 1971) and Harold Johnson (October 6, 1971). It appears that, except for the permit issued to Mr. Johnson on October 6, no permits at all were issued during the months of August, September or October 1971. E. James Foster, Ronald Baldwin, and Arthur Keitt. Between January and March 1972, James Foster, Ronald Baldwin, and Arthur Keitt, all black men, were hired by Otis Elevator Company (Otis). On March 7, 1972, accompanied by representatives from the Urban Coalition and the Journeyman Outreach Program, the three men went to the offices of Local 5 and filed applications for work permits. Robert Williams, Local 5’s Business Manager, explained that no permits could be issued because there were union men out of work. An official at Otis called Local 5 to request issuance of permits to its new black employees, citing the need to comply with requirements of the Philadelphia Plan in connection with its work at the new federal courthouse at 6th and Market Streets. Otis offered to hire three white union men, currently out of work, in exchange for Local 5’s promise to issue the desired work permit. Robert Williams said that permits could only be issued if Otis would hire 12 unemployed Local 5 men, but apparently did not specify that the 12 men be white. Otis would not agree to this proposal. On March 7, 1972, following their unsuccessful efforts to obtain work permits from Local 5, Messrs. Foster, Baldwin and Keitt filed a complaint against Local 5 with the Philadelphia Commission on Human Relations. One week later, on March 15, 1972, the United States Department of Justice filed the instant action. On March 17, 1972, Foster, Baldwin and Keitt received work permits from Local 5, and were immediately sent by Otis to work at the 6th and Market Streets site. In late March 1972, the International Union of Elevator Constructors, including Local 5, began a strike which lasted until July 1972. From January 1, 1972, to March 15, 1972, no permits were issued by Local 5 to any men, black or white. 2. Referral for Employment. A. Harold Johnson. Harold Johnson was laid off by Energy on February 1, 1972. Believing that Mr. Johnson was not a promising worker, Energy never exercised its selection rights to have him referred for re-employment. In March and April 1972, Mr. Johnson called officers of Local 5 and asked if they knew of any opportunities for employment. He was told that “things were slow,” and was not referred to any employers. When he called Energy from time to time following his layoff, again he was told that business was slow. In August 1972, Johnson again sought to be referred by Local 5. Art Rodgers, Secretary-Treasurer of Local 5, remarked that he had recently referred a few men to Otis Elevator Company, apparently at the company’s request, and suggested that Johnson apply for work there. Mr. Johnson did apply, but was never offered a job by Otis. The facts relating to the issuance of work permits between May and September 1972, discussed above in connection with Robert Scott’s case, suggest that although there may have been unemployment in the trade in the first half of 1972, by July 1972 Local 5 was beginning to issue work permits to new men. B. Jessie Oliver. Westinghouse laid off Jessie Oliver in March 1971, when the work at Veterans Stadium was completed. Mr. Oliver was never told of the Union’s hiring hall system, and he never asked to be referred for employment. However, there is no evidence that Local 5 informed any worker, white or black of the Union’s referral practices. Westinghouse recalled Mr. Oliver about three months after his layoff. 3. Initiation into Membership. Although Union membership does not affect step-ups in wage levels (from 50% to 70% to 100% of the full mechanic’s rate), it does make a difference in two ways which have been mentioned earlier. First, Local 5 officials tend to refer card helpers (union members) before 70% helpers (non-union workers) when employment opportunities become available. Second, Local 5 officials often exert pressure on employers to lay off 70% helpers before card helpers. A. Albert Patterson. Albert Patterson, a black man, was employed by Westinghouse on October 2, 1967, and received a work permit from Local 5 the next day. He completed his probationary period on April 3, 1968. However, although then eligible for Local 5 membership, Mr. Patterson was not initiated into membership until July 8, 1969. Between April 3, 1968, and July 8, 1969, 46 white men were admitted to membership, most of them at regular initiations held in April 1968 and May 1969, and a few at “special initiations.” A month before the April 1968 initiation, when it was scheduled, Albert Patterson was not yet eligible for membership. The May 1969 initiation was scheduled at a time when Mr. Patterson was out of work due to a serious car accident. However, Mr. Patterson had previously written to Local 5 requesting admission to membership. And a worker’s eligibility for membership is not affected by whether or not he is employed at the time of an initiation. The initiation of Mr. Patterson on July 8, 1969 was, however, a special initiation scheduled solely for his admission to membership. B. Jessie Oliver and Kenneth Reid. Messrs. Oliver and Reid completed their probationary periods in March 1971 and January 1972, respectively. They and six other men were admitted into membership at an initiation held in January 1973. This was the first regular initiation after the dates on which Oliver and Reid attained eligibility for membership. In the interim, two white men were made members at special initiations, following their return from military service. 4. Layoffs. A. Kenneth Reid. After the construction work at 4th and Arch Streets was finished, Security laid off several helpers while keeping Mr. Reid on the payroll and sending him to another job site. The decision to keep Mr. Reid rather than the other employees was based on Security’s need to fulfill its Philadelphia Plan obligations. The Local 5 employees at the new job site refused to work with Mr. Reid because, under the provisions of the collective bargaining agreement, he (as a probationary helper) should have been laid off before any 70% or card helpers were laid off. A two-week strike ensued. When the labor dispute was referred to arbitration, the arbitrator found the layoffs of the helpers improper and awarded them back pay. 5. Promotion to Mechanic. Albert Patterson became the only black mechanic member of Local 5 after passing the most recent mechanic’s examination, administered in February 1972. At the time, Mr. Patterson and Levi Miller were the only black members with two years’ experience in the trade (the condition for eligibility for the mechanic’s exam). Apparently, Mr. Miller did not choose to take the exam. A mechanics’ exam was scheduled for September 1973, at which time between 80 and 100 helpers, including 6 blacks, had completed the two years’ experience in the trade required to be eligible for the exam. It appears that nearly all these men intended to take the exam. The exam was not given, however, because the men at the testing site all walked out after listening to remarks made by Robert Williams, Local 5’s business manager. Mr. Williams told the men that it was against their best interests to take the exam at that time, since it would produce an over-supply of mechanics which management could exploit in future negotiations with the Union. Although it appears that some of the men who felt constrained to walk out would have preferred to take the exam, it is unknown whether these men were white or black. The National Elevator Industry Educational Program, a joint management-union program designed to train helpers and prepare them for the mechanics’ exam, was discontinued in 1973 when Mr. Williams, co-chairman of the program’s governing committee, refused to sign the necessary agreements. The basis for his refusal does not appear from the record. Discussion The facts have been set forth in considerable detail to illustrate the difficulty in distinguishing legitimate union practices, aimed at protecting union membership, from practices which are impermissible because of their discriminatory effects. Under one interpretation, many of the instances cited by the Government as examples of overt discrimination may be viewed as the unavoidable result of color-blind provisions of the collective bargaining agreement regarding new hiring, layoffs, and other employment matters. Practices neutral on their face, however, seldom operate in perfectly neutral ways. And, where their effects are discriminatory, shades of motivation may cover the spectrum. Discriminatory effects, disguised by a surface neutrality, may be sometimes the product of conscious design, sometimes the result of subconscious desire. The consequences may never be recognized for what they are, or, though unintended, they may be acknowledged and allowed to continue. 1. The Government’s Statistical Case. One fact remains basic and unaltered, irrespective of questions of how and why, of operation and intent: When this action was commenced, in March 1972, less than 1% of Local 5’s members were black. Compared with any of the several base percentages suggested by the Government to be indicative of the availability of black workers in Local 5’s area, this black membership percentage is low, and significantly low. A showing that the Union’s black membership is grossly underrepresentative of the black population available for work in a demographic area establishes a prima facie case of Title VII violation. Jersey Central Power & Light Co. v. Local Unions 327 et al., 508 F.2d 687, 706, ftn. 51 (3d Cir. 1975). It raises an inference that the racial imbalance is the result of discrimination, shifting the burden of going forward and the burden of persuasion to the defendant. United States v. Ironworkers Local 86, 443 F.2d 544, 551 (9th Cir. 1971), cert. denied 404 U.S. 984, 92 S.Ct. 447, 30 L.Ed.2d 367 (1971); United States v. Hayes International Corp., 456 F.2d 112, 120 (5th Cir. 1972). The defendant has objected to the Government’s use of statistics on the ground that there has been no expert testimony regarding the statistical significance of the disparity in percentages. It is also argued that the base populations are not perfectly suited for purposes of comparison (e. g., some of the populations are defined in terms of males over 16, whereas the minimum age for membership in Local 5 is 18) and that the percentages must be weighted in some way to reflect the fact that Local 5’s membership is not uniformly distributed throughout its jurisdiction. It takes no statistician, however, to recognize the significance of the disparity between 1% and any one of the several base percentages ranging from 13% to 43.3%. Regardless of how the percentages are weighted, the resulting figure will lie somewhere in this range. And the inference to be drawn from the gross disparity shown here is not seriously weakened by the fact that none of the base demographic groups precisely mirror the pool of black persons “available” for union membership. This case does not present a situation where special qualifications or credentials are required of persons seeking to fill job vacancies. The only requirements for membership in Local 5 are that a worker has completed his probationary period and be 18 years old (the high school education requirement has never been a bar). And no requirements exist for employment as a probationary helper. It is not a case, therefore, where it is inappropriate to take some fairly broad, unrestricted demographic base, such as the population at large or the male population over 16 years old, as the “relevant universe for comparison purposes.” See, Mayor of the City of Philadelphia v. Educational Equality League, 415 U.S. 605, 94 S.Ct. 1323, 39 L.Ed.2d 630 (1974); McAdory v. Scientific Research Instruments, Inc., 355 F.Supp. 468 (D.Md.1973). Of course, some of the defendants’ objections to the Government’s statistical evidence may merit consideration in shaping the remedial provisions of a final decree (e. g., if a black membership goal is to be set, what should it be?). But for the purpose of establishing a prima facie ease of Title VII violation, the statistics are abundantly convincing and need no extended treatment. Cf. United States v. United Brotherhood of Carpenters and Joiners, Local 169, 457 F.2d 210 (7th Cir. 1972), cert. denied 409 U.S. 851, 93 S.Ct. 63, 34 L.Ed.2d 94 (1972). 2. Local 5’s Proof of Non-Discrimination Local 5 has attempted in several ways to rebut the inference of discrimination raised by the statistical evidence. With respect to the individual cases of alleged discrimination, the Union has sought to show that its actions towards the various black individuals were always based on neutral union policy, and never on race. To explain the gross underrepre-sentation of blacks in its membership, Local 5 contends, in essence, that it really exercises much less control over work opportunities in the trade than the Government asserts, and that it is the employers who bear the brunt of the responsibility for the underrepresentation. a. Instances of Alleged Overt Discrimination Even if Local 5 could demonstrate to the Court’s satisfaction that, in each of the reputed instances of discrimination, the basis for its action (or inaction) was some neutral union practice and not race, this would not adequately rebut the inference of discrimination raised by the statistical data. The Union’s burden of explaining the gross underrepresentation of black workers in union membership is not met by the Union’s attempts to parry specific allegations of discrimination. United States v. Hayes International Corp., 456 F.2d 112, 120 (5th Cir. 1972). It is not enough for Local 5 to successfully negate any evidence of subjective intent to discriminate. For Local 5 to dispel the inference of discrimination, it must show that the policies and practices controlling the issuance of work permits, admission to membership, and so on, which have undeniably had discriminatory consequences, bear a manifest relationship to legitimate union interests. Griggs v. Duke Power Co., 401 U.S. 424, 432, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). This does not mean, however, that evidence of particular acts of discrimination is superfluous. A pattern or practice of discrimination may be established by statistical data, or by evidence of specific incidents involving black persons, manifesting a course of racially motivated conduct, or by any combination of the two. Obviously, evidence of discrimination in action strengthens the inference, based on statistics, that discrimination has occurred. Such evidence may also be important in evaluating Local 5’s contention that it is the employers who are ultimately responsible for the dearth of black union members. Finally, proof of discrimination against particular individuals may affect the nature of the relief to be granted by illustrating the forms of discrimination to be eradicated, the appropriateness of back pay awards, and so on. Local 5 has offered an apparently satisfactory explanation for five of the seven cases cited by the Government as examples of discrimination in the issuance of work permits. In the cases of Robert Scott, Harold Johnson, James Foster, Ronald Baldwin and Arthur Keitt, it appears that they were seeking permits in periods when Local 5 was issuing none, and the Union’s refusal to issue permits to those men may be attributed to its general practice of issuing no new permits when Union members or permit-holders are out of work. However, that general practice does not serve to explain the cases of Kenneth Reid and Jessie Oliver, both denied work permits at times when permits were being issued. There is no merit to Local 5’s argument that its explanation can only be defeated by showing that on the days that those black applicants were refused permits, Local 5 issued permits to white men. Their applications for permits were on file during periods when permits were being issued. No suggestion has been made that Local 5 routinely gives preference to applications currently being submitted over those on file. Local 5 has offered no account of its failure to refer Harold Johnson for employment despite his repeated requests. The Union issuance of new work permits to several white men during the period that Mr. Johnson was seeking referral was a clear departure from the Union’s stated policy of granting no new permits when its members or permit-holders are out of work. On the other hand, the fact that Jessie Oliver was not referred by Local 5 following his layoff can be satisfactorily explained by the apparent lack of work opportunities during that period. The instances cited by the Government as suggestive of discrimination in Local 5's initiation practices are not very convincing. Messrs. Oliver and Reid were never passed over at a regular initiation. And it is difficult to draw an adverse inference from the fact that the first regular initiation (at which both men were admitted) was not held until a year or two after they attained eligibility, since it appears that several of the six white men also initiated then had been eligible for an even longer time. Nor does the experience of Albert Patterson present an unambiguous case of discrimination. The special initiation scheduled for Patterson’s admission upon his return from hospitalization tends to dispel the Government’s suggestion that the Union was seeking to deny him membership. Finally, the Union’s strong reaction to Security decision to lay off several helpers before Kenneth Reid, a black probationary helper, is quite adequately explained by the layoff provisions of the collective bargaining agreement. And the cancellation of the mechanics exam, which a predominately white group of men had planned to take, cannot fairly be viewed as a further instance of racial discrimination. b. Local 5’s Rebuttal to the Statistically-Based Inference of Discrimination It is Local 5’s position that the real reason for the disproportionately few black union members is the Elevator Construction Contractors’ “historical refusal” to hire blacks. According to this thesis, the Union plays an insignificant role in recruiting new, inexperienced workers to become probationary helpers, and, eventually, union members. Local 5 contends that most of the men entering the trade for the first time are selected by employers and sent to the Union for work permits. If there are no probationary helpers or union members out of work, Local 5 routinely issues permits to the employer’s new recruits. Only very rarely, it is argued, does Local 5 itself select the name of an inexperienced worker from the applications on file, issue him a permit, and refer him to an employer. Since in the long run the racial composition of union membership depends on the selection of new permit holders, this argument ascribes to the employers the primary responsibility for the racial imbalance in Local 5’s membership. Related to this defense is the Union’s argument that it cannot be fairly accused of discrimination when statistics show that, in percentage terms, black applicants were granted work permits as frequently as white applicants. Put otherwise, the reason that Local 5’s black membership is not representative of the population of available workers is that the pool of applicants was not representative. To the extent that applications for work permits were submitted by workers recruited by employer, imbalance in the racial composition of the applicant pool was attributable to the employers’ hiring practices. These reasons, in the Union’s view, accounted for the low proportion of blacks in Local 5 in the years preceding the establishment of the Philadelphia Plan. Thereafter, elevator construction contractors with federal contracts were under economic pressure to increase the number of their minority employees, and, because of the very few black men then working in the trade, these employers recruited and hired new, inexperienced black workers. Local 5 contends that during this same period there was a drastic slowdown in the construction industry, resulting in severe unemployment in the elevator construction trade. And, indeed, the evidence relating to the issuance of work permits suggests that October 1970 through March 1972 did mark a period of depression in the trade. The Union asserts that it was the collision of these two forces — the employers’ sudden need to put new black workers on their jobs and a decline in business activity which put union men out of work — together with the Union’s longstanding policy against issuing new permits when its men were out of work, which produced the series of incidents cited by the Government as examples of overt discrimination by Local 5. Thus, according to this version of the facts, the employers’ abandonment of their “historical refusal” to hire blacks could not immediately be translated into an increase in the number of black probationary helpers because of a Union policy which was neutral on its face and neutral in design. A fact which further supports this argument is that between August 1972 and April 1973, a period when new permits were again being issued (and, not insignificantly, the period during which the parties were preparing for trial), permits were freely issued to a large percentage (80-85%) of black applicants. 3. The Union’s Violation of Title VII It may well be that the instances of overt discrimination by the Union disclosed by the record in this case, considered in isolation, would form an insufficient foundation for a conclusion that a pattern or practice of discrimination had been established. The instances were not numerous, and for the most part appear to have been motivated by arguably legitimate Union concerns. Be that as it may, it is unnecessary in the present case to rely solely upon evidence of specific acts of discrimination. The statistical evidence demonstrating gross underrrepresentation of blacks in the Union’s membership, and in the elevator construction industry, more than adequately establishes a pattern or practice of discrimination in violation of Title YII. That conclusion is not weakened, and to some extent is strengthened, by the evidence of specific instances. The Union’s attempt to ascribe to employers in the industry the primary responsibility for the disproportionately low black involvement is unacceptable. While it is true that many, perhaps the majority of persons entering the elevator construction trade as probationary helpers do so under the sponsorship of employers, a significant proportion (25 to 35%) of these new entrants are selected by Local 5. But Local 5 has never yet selected a black man to be a new probationary helper. Thus, whatever may be said of the employers’ recruitment practices, the uncontradicted evidence discloses that Local 5 has exercised its partial control of work opportunities in the trade to the complete exclusion of blacks. The record demonstrates that approximately 75% of the applications for work permits on file with Local 5 were submitted by friends or relatives of Union members, and less than 4% by black persons. Therefore, apart from the Union’s discrimination in the selection of applicants from the applicant pool, the pool itself suffers from gross racial imbalance. Although the record does not permit any firm conclusion as to the extent to which this imbalance is attributable to the recruitment practices of the Union, as distinguished from those of the employers, it is clear that at least a share of the blame is attributable to the Union. The availability of work opportunities through the Union hiring hall system is apparently a matter which is publicized primarily by word of mouth, with the result that a very high proportion of applicants are related to or acquainted with Union members. This practice of limiting notice of employment opportunities to word of mouth communication has been condemned, when it operates to perpetuate the effects of past discrimination. United States v. Georgia Power, 474 F.2d 906 (5th Cir. 1973); Parham v. Southwestern Bell Telephone Co., 433 F.2d 421 (8th Cir. 1970). It is also clear, of course, that the operation of the “seniority system” has the effect of perpetuating the racial imbalance that the Philidelphia Plan was designed to rectify. The principal characteristics of the seniority system in this case may be summarized as follows: The exclusive hiring hall system works to the advantage of both Union members and permit men, by assuring that they will be rehired before new men are hired; and it favors the more senior workers, insofar as Union officials base the order of referral upon seniority in the trade. The layoff requirements protect men who have completed their probationary period over men who have not; and, to the extent that the Union succeeds in influencing employers’ decisions regarding the order of layoffs, workers with greater seniority are advantaged. In addition, the Union’s policies regarding the scheduling of initiations and mechanics’ examinations are designed to protect the interests of present members and mechanics, and thus may also be considered a part of the seniority program. In the Jersey Central Power & Light Co. case, supra, the Third Circuit Court of Appeals held that, at least in some circumstances, collectively-bargained seniority provisions do not violate Title VII, even though they perpetuate the effects of past discrimination. The ramifications of that decision will be considered in connection with determining the appropriate relief to be granted in this case. 4. Relief. The Government’s proposed decree calls for sweeping remedial action, including the establishment of black membership and referral goals, fundamental modification of key features of the seniority system, and the implementation of recruitment and training programs designed to attract minorities to the trade and to enable them to become mechanics. Local 5 has characterized the relief sought as “drastic.” a. The 23% Black Membership Goal. The establishment of specific membership or referral goals is a form of affirmative relief authorized by § 707(a) of the Civil Rights Act of 1964, 42 U.S.C. § 2000e-6(a), notwithstanding the anti-preferential treatment provision of § 703(j), 42 U.S.C. § 2000e-2(j). United States v. IBEW, Local Union No. 212, 472 F.2d 634 (6th Cir. 1973); United States v. Ironworkers Local 86, 443 F.2d 544 (9th Cir. 1971), cert. denied 404 U.S. 984, 92 S.Ct. 447, 30 L.Ed.2d 367 (1971); Asbestos Workers, Local 53 v. Vogler, 407 F.2d 1047 (5th Cir. 1970). The Government has proposed a black membership goal of 23%, based on the several demographic measures of available black workers in Local 5’s geographic jurisdiction. No deadlines for meeting either this or intermediate goals have been suggested, and it appears that the sole purpose of the 23% figure would be to determine the effective life of the decree. For this limited purpose, the 23% black membership goal is reasonable and may properly be adopted without further qualification of the statistical evidence. The parties are of course free at any time to seek a reduction or extension of the life of the decree by showing that no further affirmative action is needed. b. Modification of the Seniority System. Jersey Central Power & Light, supra, considered, in a somewhat unusual procedural posture, the relationship between collectively-bargained seniority rights and affirmative action programs authorized by Title VII. There, the Third Circuit discussed (1) whether the disproportionate impact of bona fide seniority provisions on minorities itself constituted a violation of. Title VII, requiring judicial modification of the seniority system, and (2) whether in a case where there had been past discrimination and where bona fide seniority provisions would operate to perpetuate the effects of that discrimination, a court was empowered to modify the seniority system as a part of the remedial action ordered in a Title VII suit. Under its interpretation of the legislative history of § 703(h), the Third Circuit considered itself obliged to answer both these questions, as presented in the factual context of that case, in the negative. The seniority system in this case is neutral on its face. And it has not been alleged or proven that the seniority system was intended or designed to disguise present discriminatory practices. A mere awareness of the disproportionate impact of the seniority system on black persons — an awareness which Local 5 officials now certainly must possess — does not affect the bona fide character of at least those provisions which were contained in collective bargaining agreements in effect when Title VII was passed. The layoff requirement contained in the new standard agreement is apparently such a seniority provision. However, the instant case presents problems not addressed by the Jersey Central Power & Light decision. First, some of the aspects of the seniority system (e. g., the Union’s control of additions to the ranks of its members and mechanics through scheduling initiations and mechanics’ exams) simply reflect union policy and practice, and are not contained in the collective bargaining agreement. Presumably, expectations based on informal union-management practices do not enjoy the same protection as contractually-based seniority rights. Second, one provision of the new standard agreement (viz, the requirement that employers hire experienced workers before new, inexperienced men), though apparently a longstanding practice, was not incorporated in the collective bargaining agreement until after the enactment of Title VII. It is one thing to say that Congress did not intend to create chaos by disturbing collectively-bargained seniority rights existing when Title VII was passed; it is quite another to say that Congress intended to give management and labor a free hand to perpetuate patterns of discrimination by post-enactment agreements. Third, to some extent the seniority system here operates on an industry-wide scale. The Jersey Central Power & Light case involved plant- or company-wide seniority provisions. By granting to members of neighboring locals of the International Union rehire rights with respect to employment opportunities in Local 5’s jurisdiction, the perpetuating effects of the exclusive hiring hall system may be prolonged to an intolerable extent. i. Layoffs. The Government asks that the decree include a provision enjoining Local 5 from enforcing the standard agreement or any custom or practice which would require an employer to lay off a black probationary helper before a union member or another probationary helper with more seniority in the trade. Under the standard agreement, contractors are obligated to lay off employees in the following order: First, probationary helpers; second, “transient employees”; third, “mechanics and helpers who permanently live in the area.” The requirement that Local 5 probationary helpers be laid off before other Local 5 workers is a bona fide seniority provision within the protection of § 703(h), as interpreted by the Jersey Central Power & Light case. It is my view, however, that the requirement that probationary helpers of Local 5 be laid off before non-probationary workers of other locals, a seniority provision with the potential for perpetuating effects more far-reaching than those considered by the Third Circuit in that case, should not be sanctioned. The decree will enjoin Local 5 from enforcing this aspect of the standard agreement’s layoff provision. Although not required by the collective bargaining agreement, employers are sometimes exhorted by Local 5 officials to favor more senior union members over less senior members, and card helpers over 70% helpers, in making layoff decisions. To the extent that this practice is followed by employers, it will tend to perpetuate the existing racial imbalance. Local 5 will be enjoined from attempting in any way to cause employers to lay off men in order of seniority, except by enforcing the terms of the standard agreement to the extent allowed above. ii. Referrals. With regard to the referral of mechanics, both the Government and Local 5 agree that the order of referral should be based on the date of unemployment of a worker, with the mechanic out of work the longest referred first, and so on, irrespective of the mechanic’s race. There is disagreement, however, over what rights employers should have to select a particular man from the list of unemployed mechanics, and to reject a mechanic referred by the Union. The Government seeks to limit employers’ selection rights to blacks only, while Local 5 would continue the practice currently provided for in the standard agreement, which permits employers to choose any man, black or white. With respect to employers’ rejection rights, the Government’s proposed degree provides that if a black mechanic is rejected upon referral, no white mechanic lower on the unemployment list can be referred until that black person has been employed (blacks lower on the list could be referred). Again, Local 5 would allow rejections within the terms of the present collective bargaining agreement. Those terms purport to circumscribe the right of an employer to continually reject a particular worker. The Government’s proposed modification of the standard agreement would affect the employment rights of contractors rather than the seniority rights of labor. The apparent purpose of these changes is to prevent employers from exercising their selection and rejection rights in a discriminatory fashion. However, the employers have not been joined as defendants. In my opinion, it would be improper to modify provisions of the collective bargaining agreement governing the rights of employers in their absence, particularly in light of the principles of the Jersey Central Power & Light decision. Moreover, I do not believe that this modification is essential to an effective remedial decree. There is less agreement concerning the provisions to govern the referral of helpers and probationary helpers. The Government proposes that Local 5 be ordered to make referrals with the goal of referring one black worker for every white worker. Under the suggested procedure, first black helpers would be referred, then black probationary helpers, and finally black applicants for work permits, with the order of referral within each category based on the helper’s date of unemployment or the applicant’s date of application. No order of referral of white men is specified. The aims to be'served by the 50% referral goal appear to be twofold. In the first place, an increase in the Union’s black membership is not per se the ultimate purpose of this action. Membership in Local 5, or status as one of its probationary helpers, is valuable only insofar as it leads to employment in the elevator construction trade. Second, since a permit holder must work a certain number of hours to complete his probationary period, work referral is essential if black probationary helpers are to attain the limited form of seniority provided for in the standard agreement and to become Union members. Under the 50% referral procedure, Local 5 would be obligated in some situations to issue permits to new, inexperienced (black) men and to refer them for employment before referring (white) helpers or probationary helpers out of work. This result would also follow from two other provisions of the Government’s proposed relief. One would require Local 5 to issue a permit to refer out of order any black applicant specifically requested by a contractor. The other would require the Union to issue a permit to any black person who has obtained an employment offer from a contractor. The Government’s proposals would contravene the provisions of the collective bargaining agreement now in effect, whi