569 F.2d 315 | 5th Cir. | 1978
At issue in the present cases is the scope of the Equal Employment Opportunity Commission’s subpoena powers in investigations of individual instances of race and sex discrimination under Title YII of the Equal Employment Opportunity Act, 42 U.S.C. § 2000e et seq. In both cases individuals complained to the EEOC’s Jackson, Mississippi, office about relatively narrow factual situations. The EEOC, in its attempts to investigate these complaints, issued subpoenas for broad statistical information as to the respective employers’ entire work force. In each case the district court granted only partial enforcement of these sweeping requests, and the propriety of these partial denials of enforcement is now drawn before us.
Both cases turn on the EEOC’s wish to obtain data in the form known to it as “workforce breakouts.” Along with its requests for the personnel files of the individual complainants and general hire and fire data, the EEOC subpoenaed facility-wide “workforce breakouts” — the entire current personnel lists of the complained-of facilities, showing each employee’s race and sex, job classification, labor grade, date of hire, and pay rate. Although GMAC and Packard supplied some materials about the individual complainants and GMAC also supplied some limited plant-wide data about other job applicants, both employers refused to supply the plant-wide workforce breakouts.
When the EEOC sought to enforce its subpoenas, the district court refused to require facility-wide workforce breakouts; instead, the district court granted partial enforcement in each case, tailoring disclosure
At the outset, it is clear that we may review on appeal the partial denial of the EEOC’s petition for enforcement of its investigative subpoena. This court has long entertained such appeals, treating them as reviewable final orders. See, e. g., New Orleans Public Service, Inc. v. Brown, 507 F.2d 160 (5th Cir. 1975); Parliament House Motor Hotel v. EEOC, 444 F.2d 1335 (5th Cir. 1971); Georgia Power v. EEOC, 412 F.2d 462 (5th Cir. 1969). This is because an EEOC investigative subpoena is not necessarily part of a continuing judicial proceeding, and, like other administrative subpoenas, its grant or denial may as a practical matter end the controversy between the parties. See Ellis v. ICC, 237 U.S. 434, 35 S.Ct. 645, 59 L.Ed. 1036 (1915); FTC v. Texaco, Inc., 170 U.S.App.D.C. 323, 517 F.2d 137 (1975), cert. denied, 431 U.S. 974, 97 S.Ct. 2939, 2940, 53 L.Ed.2d 1072, reconsidered en banc, 555 F.2d 862 (D.C.Cir.1977); Genuine Parts Co. v. FTC, 445 F.2d 1382 (5th Cir. 1971); International Brotherhood of Electrical Workers Local 5 v. EEOC, 398 F.2d 248 (3d Cir. 1968), cert. denied, 393 U.S. 1021, 89 S.Ct. 628, 21 L.Ed.2d 565 (1969).
The standard by which we are to review such decisions, however, is somewhat less clear. This court has said that an order enforcing an EEOC investigative subpoena for documents is subject to the same standards as any other subpoena for documents under Fed.R.Civ.P. 45(b). New Orleans Public Service v. Brown, supra. We have said that this standard is “abuse of discretion,” Brown v. Thompson, 430 F.2d 1214 (5th Cir. 1970), but we have also implied that the standard may be “clearly erroneous,” United States v. Sun First National Bank of Orlando, 510 F.2d 1107 (5th Cir. 1975); see also Causey v. Ford Motor Co., 516 F.2d 416 (5th Cir. 1975); and in the Title VII context, we have even implied that a trial court’s decision might be overturned when the court of appeals merely “disagrees” with the lower court’s denial of subpoena enforcement. Burns v. Thiokol Chemical Corp., 483 F.2d 300, 306 (5th Cir. 1973). This disparity of language merely reflects the fact that a district court’s enforcement of an EEOC subpoena may take into account at least two distinct types of criteria, one type based on the statute and the other on the language of Fed.R.Civ.P. 45(b). Title VII provides that the EEOC, in connection with its investigatory activities, is to have access to any evidence “relevant” to the charge under investigation. 42 U.S.C. § 20006-8.
In denying enforcement of subpoenas for plant-wide workforce breakouts, the district court based its decision on the criterion of relevance rather than the equitable criterion of burdensomeness, finding that the EEOC had not shown the relevance of the broad breakout data “at this stage in the investigation.” Since the question of relevance in this instance is essentially a factual determination concerning the interrelation or lack thereof of different groups of facts, we must uphold the district court’s determination unless it is clearly erroneous.
The EEOC maintains that the district court took too narrow a view of relevance in judging the requested statistical information. It points out that statistical information may be used to establish that the treatment of a particular employee follows a general pattern of employer discrimination, McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). And as this court has said in a leading case on the subject, Georgia Power, supra, race and sex discrimination is inherently class discrimination, and the EEOC must have comparative data in order to investigate an employment discrimination charge. 412 F.2d at 468.
All this may be so, but it does not help us in determining precisely what statistical and comparative data are to be deemed “relevant.” Certainly, at a minimum, those statistical materials must be deemed relevant that are necessary for the EEOC effectively to carry out its statutory investigatory duties, and a district court may not construe relevance so restrictively as to disable the Commission in carrying out those duties. New Orleans Public Service, Inc., supra, at 165; Local 104, Sheet Metal Workers International Association v. EEOC, 439 F.2d 237, 242 (9th Cir. 1971); see also United States v. Bisceglia, 420 U.S. 141, 151, 95 S.Ct. 915, 43 L.Ed.2d 88 (1965). Moreover, we must look to the particular purpose of the investigation in question. United States v. Humble Oil & Refining Co., 518 F.2d 747 (5th Cir. 1975). In the context of an investigation of an individual complaint, it might well be most natural to focus .on that employing unit or work unit from which came the decision of which the individual complainant complains; within such a unit the EEOC might well need a wide spectrum of statistical data in order to illuminate the general policies bearing on the complainant’s situation. See Joslin Dry Goods v. EEOC, 483 F.2d 178, 184 (10th Cir. 1973); see also Parliament House, supra at 1340. But in the absence of some showing by the EEOC to the contrary, it is not immediately evident that this unit need be or is in this case the entire facility or that an entire facility-wide workforce breakout bears on the subject matter of these individual complaints. We therefore cannot deem clearly erroneous the district court’s finding that the plant-wide workforce breakouts sought were not shown to be relevant at the then current stage of the investigation. The district court’s order requires the defendants to disclose data along lines similar to those laid down in Georgia Power, supra. Moreover, the order leaves it open to the EEOC to show, if it can, the relevance of the plant-wide workforce breakouts at some later stage in its investigation.
One further issue concerns the GMAC investigation alone. The EEOC subpoena required that GMAC produce its current “EEO-1 report.” This is a report that employers must file annually with the EEOC, showing race, sex and other data about employees in several broad employment categories. The district court denied enforcement of the subpoena on the ground that the document was already in the
AFFIRMED.
. Rule 26 of the Fed.R.Civ.P., concerning the scope of discovery in ordinary civil cases, of course also limits discovery to material “relevant” to the subject matter of the proceeding.