This case is again before this court on appeal by the Miami Association of Firefighters, Local 587, International Association of Firefighters, AFL-CIO. The litigation arose from a consent decree entered in 1977 between the City of Miami and the United States that implemented an affirmative action plan for hiring and promotion of blacks, Hispanics and women within the City of Miami Fire Department. In 1989 Local 587 moved the district court to dissolve or modify the consent decree on grounds that it had served its purpose of remedying vestiges of discrimination. The district court refused to dissolve the decree and Local 587 appealed.

We vacated and remanded the case to the district court for consideration of whether the decree should be terminated or modified. U.S. v. City of Miami, 2 F.3d 1497 (11th Cir.1993). On remand the district court was to examine "whether the current underrepresentation of favored groups in the promotional ranks of the Fire Department is a vestige of past discrimination or a result of other neutral causes, such as a nondiscriminatory seniority system and other factors we have discussed." Id. at 1508. We concluded that it was necessary to ascertain the proportion of qualified applicants from each favored group before determining whether affirmative action should continue with respect to both initial hires and promotions. Id. at 1507.

Pursuant to this court's instruction the district court ordered that any party wishing to continue the consent decree file a brief and present evidence in support of its continuation. At a December 1993 hearing the district court denied Local 587's motion to dissolve and scheduled a second evidentiary hearing at which the City was to provide the court with expert testimony on the issue of whether the underrepresentation of blacks, Hispanics and women in the promotional ranks of the Department was a vestige of past discrimination. SR 1:132-133.

At that hearing the district court accepted the report and testimony of the City's labor economics expert, Dr. Mark Bendick, and concluded that the underrepresentation of favored groups in the promotional ranks of the Department was the result of past discriminatory practices. Dist.Ct.Ord. at 5-6. The court ordered that promotions be made according to an alternating promotion procedure. The first available promotion would be given to the highest ranking eligible individual regardless of race or sex, and the next promotion would be given to the highest ranking individual from a list of eligible black, Hispanic and female candidates. Id. at 11. This was to continue until the underrepresentation for each group at a particular rank was eliminated. Id. at 11-12.

The district court concluded that the City had not presented sufficient evidence to demonstrate that the Department's entry-level hiring exam was job-related and consistent with business necessity. Id. at 5. The City was directed to cease using the entry-level exam and to develop lawful selection procedures. Id. at 7.

Local 587 contends that the evidence was insufficient for the district court to conclude that the Department's entry-level hiring exam had an adverse impact on women and minorities. We need not reach the merits of this issue because we hold that Local 587 lacked standing to challenge hiring portions of the consent decree. Thus, the issue of hiring was not properly before the district court.

Second, Local 587 asserts that evidence relied upon by the district court does not provide a basis for continuing the promotional provisions of the consent decree. We agree. Dr. Bendick's testimony lacks probative value as a matter of law, therefore, we hold that the district court erred in retaining jurisdiction over the promotion claims. We REVERSE and REMAND to the district court with instructions to grant Local 587's motion to dissolve the 1977 consent decree provisions relating to promotions.

Local 587 is the exclusive collective bargaining agent for firefighters and employees presently employed by the Miami Fire Department. Because Local 587 does not represent applicants for entry-level positions in the Department, no member of Local 587 would suffer harm from the hiring relief ordered by the district court. Thus Local 587 lacks standing to contest hiring provisions of the 1977 consent decree.

An organization can sue based on injuries to itself or based on injuries to its members. However organizations lack standing to sue on behalf of non-members. For a union to have standing in its representational capacity it must show that:

(a) its members would otherwise have standing to sue in their own right;

(b) the interests it seeks to protect are germane to the organization's purpose; and

(c) neither the claim asserted nor the relief requested requires the participation of individual members in the lawsuit.

Hunt v. Wash. State Apple Adv. Com'n., 432 U.S. 333, 343, 97 S. Ct. 2434, 2441, 53 L. Ed. 2d 383 (1977). Individual union members have standing to sue in their own right if:

(1) there was a threatened or actual injury in fact;

(2) there is a causal relationship between the injury and the challenged conduct; and

(3) there is a likelihood that the injury will be addressed by a favorable decision.

Northeastern Fla. Chapter, Assoc. Gen. Contractors of Amer. v. Jacksonville, 508 U.S. 656, 663, 113 S. Ct. 2297, 2301, 124 L. Ed. 2d 586 (1993).

Local 587 lacks standing to assert rights of potential hires because they are not union members. Nor can it assert standing based on that of its members. Individual members of Local 587 would lack standing because as present members or employees of the Department they could not suffer threatened or actual injury due to the Department's entry-level hiring practices.

In U.S. v. Miami, 664 F.2d 435, 445 (5th Cir.1981) (en banc) the Fifth Circuit held that the union representing members of the Miami Police Department lacked standing to challenge entry-level hiring provisions of an affirmative action consent decree:

Except in class actions, under Fed.R.Civ.P. 23, an appellant may not appeal on behalf of others who have chosen not to intervene or to appeal. Consequently, the FOP cannot challenge provisions of the decree that affect only the City and municipal employees outside the police department or applicants for positions in the police department who, while affected by the hiring goals in the decree, are not members of the FOP.

Id. As in this case, the union had no voice in hiring and did not represent job applicants. We hold that Local 587 lacks standing to challenge hiring provisions of the decree.

We remanded this case for the district court to determine whether vestiges existed in the promotional ranks of the Department. The City's proffer of expert testimony on that issue lacked probative value as a matter of law, and the court erred in accepting it. Local 587's motion to dissolve the consent decree must, therefore, be granted with regard to promotions.

Relevant expert testimony is admissible only if an expert knows of facts which enable him to express a reasonably accurate conclusion. See Jones v. Otis Elevator Co., 861 F.2d 655, 662 (11th Cir.1988); U.S. v. 0.161 Acres of Land, 837 F.2d 1036, 1040 (11th Cir.1988). Opinions derived from erroneous data are appropriately excluded. See Slaughter v. Southern Talc Co., 919 F.2d 304 (5th Cir.1990). In an employment discrimination context we have held that expert testimony should be excluded when it involves an assessment method not used by the decisionmaker. Hill v. Seaboard Coast Line R.R., 767 F.2d 771 (11th Cir.1985). A labor economist's calculations are improper when they are not based upon record evidence and it is confusing as to how the economist reached his result. Randolph v. Laeisz, 896 F.2d 964 (5th Cir.1990). Dr. Bendick's expert opinion suffers from all of these maladies. His report did not examine the legally relevant factors and was derived from erroneous and incomplete data.

In our 1993 mandate we directed that the district court look at the actual number of black, female and Hispanic applicants eligible for promotions relative to the number of such individuals actually promoted. We said that

[a]scertaining the proportion of qualified applicants from each favored group would seem necessary before determining whether affirmative action should continue with respect to both initial hires as well as promotions, since the base may be skewed because of a lack of interest by a particular favored group.

Miami, 2 F.3d at 1507. Dr. Bendick failed to look at the number of employees of a given race or sex eligible for promotion, nor did he evaluate passing promotional exam scores on the basis of race or gender. He had no idea how many applicants within a particular group of employees took or passed a promotional exam. Rather than making the proper comparison between the minority composition of the Department versus that of the relevant labor market, Dr. Bendick testified that his entire report was based on general census data. This analysis conflicts with this court's mandate and our statement in Ensley Branch, N.A.A.C.P. v. Seibels, 31 F.3d 1548, 1577 (11th Cir.1994) that the relative proportion of minorities in a junior position will generally be the most significant determinant of the proportion of minorities in the qualified applicant pool.

Additionally Dr. Bendick's analysis did not take into account the number of qualified minorities and women interested in promotion at each promotional rank. He declined to measure the interest of women in various Department jobs and improperly assumed that all employees would seek promotion upon eligibility at the same rate as their percentage in the general population. Although he concluded that the lower number of women in the Department was attributable to the City's "reputation for unfairness," he acknowledged that he had not done any research to determine what reputation for unfairness, if any, existed on the part of the City.

Further, Dr. Bendick's opinion was premised on erroneous data which he used to determine the relevant labor market. He incorrectly assumed that the City consistently enforced a residency requirement for continued employment when in fact it imposed no such requirement until 1993. As a result census data for Miami city limits was used in error. Dr. Bendick testified that if there were no residency requirement he would have used different census figures which would affect the relevant labor market and the analysis. R25-531-23. See also Williams v. City of New Orleans, 729 F.2d 1554, 1562 (5th Cir.1984) (rejecting Dr. Bendick's analysis of the relevant labor market as erroneous).

Because Dr. Bendick's testimony was insufficient as a matter of law, we hold that the district court erred in adopting it. The case is REVERSED and REMANDED to the district court with instructions to grant Local 587's motion to dissolve the 1977 consent decree provisions relating to promotions.