Dissenting Statement of Commissioner Harold W. Furchtgott-Roth
In Lutheran Church-Missouri Synod v. FCC, 141 F.3d 344 (1998), the United States Court of Appeals for the D.C. Circuit ruled that this Commission's Equal Employment Opportunity regulations denied the equal protection of the laws to persons seeking employment at broadcast stations. Those regulations also made broadcasters, the Court said, "involuntary participant[s] in a discriminatory scheme." Id. at 350. To have established and enforced a program that required regulatees to engage in the most historically odious sort of discrimination against potential employees -- discrimination based on race -- was a most grievous offense.
After careful consideration, I am not persuaded that the Commission's efforts to conform those regulations to the requirements of Equal Protection are adequate. Unfortunately, the revised regulations bear some of the same characteristics that led the Court of Appeals to find the original rules unconstitutional. Because these rules are not clearly constitutional, I cannot support their adoption. Moreover, I have doubts about significant parts of the Commission's theory of statutory authority for the regulations. Accordingly, I cannot support adoption of this Report & Order, however well-intentioned it might be.
I. The Regulations Are Susceptible To Reasonable Constitutional Doubt
The Order's conclusion as to the constitutionality of the outreach rules appears to hinge on the assertion that they are wholly race-neutral and thus not subject to strict scrutiny under Adarand Constructors, Inc. v. Pena, 515 U.S. 200 (1995). See, e.g., Report & Order at para. 210 (arguing that EEO requirements "do not raise equal protection concerns"); id. at para. 217 (contending that outreach requirement is "race-neutral and . . . not subject to strict scrutiny" and "raises no equal protection concerns"); id. at 218 (asserting that commenters' arguments against race- and gender-targeted recruiting "are moot"). For the reasons that follow, I must question whether this is correct.(1)
A. The Regulatory Scheme Is Not Neutral With Respect To Race And Gender
As a factual matter, the instant rules go further than simply requiring outreach to all people, without regard to race. In several places, the regulations expressly employ race-based classifications and require broadcasters to so classify persons for reporting purposes. Moreover, the Commission's enforcement plan undermines the asserted race-neutrality of the outreach requirement. Finally, the impact of the overall regulatory scheme on the behavior of broadcasters reaches all the way to hiring, not just recruiting, decisions; the scheme subtly impels broadcasters to make all such decisions with an eye toward achieving some level of racial representation, even "balance," of employees and applicants.
Under the specific EEO program requirements of Track A, a station potentially must "co-sponsor at least one job fair with organizations in the business and professional community whose membership includes substantial participation of women and minorities." 47 CFR section 73.2080(c)(2)(iii). A station could also be required to "list . . . each upper-level category opening in a job bank or newsletter of media trade groups whose membership includes substantial participation of women and minorities." Id. section (c)(2)(xii). Although the Commission discontinues its prior practice of requiring the use of minority- and women-specific referral sources -- suggesting that this action insulates the plan from being described as "targeted"outreach, see Report & Order at para. 218 -- these new requirements do essentially the same thing. Broadcasters no longer have to use minority- and women-specific groups as referral sources, cf. Lutheran Church, 141 F.3d at 351 (noting minority-specific referral source rule), but the menu of options includes a requirement that they sponsor job fairs and list jobs with such groups.
Pursuant to the alternative recruitment requirements of Track B, a station must maintain "data reflecting the recruitment source, gender, and racial and/or ethnic status of applicants for each full-time job vacancy filled" by the station. 47 CFR section 73.2080(d)(1). In addition, a station is required to include in its public file report "data reflecting, for each recruitment source utilized for any full-time vacancy. . ., the total number of applicants generated by that source, the number of applicants who were female, and the number of applicants who were minority, identified by the applicable racial and/or ethnic group with which each applicant is associated." Id. section (d)(2).
Finally, per the rule reinstated today, all stations must file FCC Form 395-B, the Annual Employment Report. Section V of that document requires the charting of employees by job category and by male and female groupings subdivided into "White (not Hispanic)," "Black (not Hispanic)," "Hispanic," "Asian or Pacific Islander," and "American Indian, Alaskan native." See Report & Order, Appendix D. A rule requiring broadcasters to place people in boxes on a chart with race and gender categories on its face uses race-based classifications.
Although the actual mandate that stations widely disseminate vacancy information makes no reference to race or gender, see 47 CFR 73.2080(c)(1)(i), the overall scheme adopted today pressures broadcasters to target potential applicants and possibly even employees on the basis of race and gender - whether proceeding under Track A or Track B.
The "self-assessment" rule, which applies under both Tracks, requires a station to "[a]nalyze its recruitment program on an ongoing basis to ensure that it is effective in achieving broad outreach to potential applicants, and address any problems found as a result of its analysis." Id. section (c)(3). Also, in order to have its license renewed, a station must have complied with all substantive EEO requirements, such as the outreach rules, during the prior license period. The FCC conducts compliance review at the time of license renewal.
By what measure does one test the "effectiveness" of outreach? According to the Order, one gauges the adequacy of outreach efforts by the number of women and minorities in applicant pools, and even in employment profiles.
Specifically, the Order provides that in order to "demonstrate" to the Commission that an outreach program under Track B "is inclusive, i.e., that it widely disseminates job vacancies," a station must "collect data tracking the recruitment sources, gender, and race/ethnicity of its applicant pools." Report & Order at para. 104. This information will allow the broadcaster and the Commission to "evaluate whether the program is effective." Id. But "[i]f the data collected does not confirm that notifications are reaching the entire community, [the Commission] expects a broadcaster to modify its program as warranted so that it is more inclusive." Id. (2)
The Commission noisily disclaims that proportionality with the local labor force will be the exclusive test for adequacy of applicant pool composition. At the same time, it admits that it will have to rely, at least in part, on the numerical representation of minorities and women in applicant pools in order to assess compliance with the outreach rule. Id. at para. 120 (denying proportionality requirement for applicant pools but stating that "few or no" minorities or women would indicate inadequate "inclusiveness").
Clearly, then, applicant pools must achieve some numerical level of minority and women applicants in order for a station's outreach program to be deemed EEO compliant. The Commission declines to say, however, just what that composition is. Thus, the Commission makes plain its intent to use numerical data on the race and gender of applicants to evaluate outreach efforts -- and even vows to require heightened efforts of broadcasters' whose data is inadequate - but is strikingly silent on just how many minority and women applicants are enough. Eventually, the Mass Media Bureau will be forced to come up with some kind of processing guidelines for review of outreach programs.
Once one focuses on race and gender statistics, however, it is difficult to come up with anything other than proportionality, or some derivative of proportionality, as a calibrator of adequacy. The only other number with significance I can identify would be zero; one could say that the absence of minorities and women in applicant pools would establish noncompliance. Beyond zero, however, it is hard to say that any one number is materially more meaningful than another. Conversely, whatever the Commission requires to demonstrate "effective" outreach, it surely could not require more than proportionality.
Given the lack of any other guidance as to compliance with the outreach rule, rational broadcasters wary of regulatory trouble will strive for some showing of rough proportionality in their applicant pools. At the very least, they will strive to have at least one woman or minority in every pool; while this is not a proportionality requirement, it is a fixed number or quota. Cf. Lutheran Church, 141 F.3d at 390 (reasoning that "the fact that the Commission looks at more than 'numbers' does not mean numbers are insignificant" since "a station would be flatly imprudent to ignore any of the factors it knows may trigger intense review" and "can assume that a hard-edged factor like statistics is bound to be one of the more noticed . . . criteria").
The fact that the standard by which Track B outreach programs, neutral on their face, will be judged is by counting minority and female applicants that wind up at the station makes it hard if not impossible to call this regulatory plan truly "race-neutral." The Commission has built into the back end of its policy what it shrewdly omitted from the face of the dissemination rule - that is, a requirement of some minimum (though vaguely defined) numerical representation of minorities and women in applicant pools. (3) Cf. Lutheran Church, 141 F.3d at 390 (reasoning that "the Commission has used enforcement to harden the suggestion" in its regulations).
The Commission also makes clear that the records broadcasters must keep under Track A regarding the referral sources of ultimate hires, see 47 C.F.R. section 73.2080(c)(5)(ii), (v)-(vi), are "designed to provide a starting point for a broadcaster to analyze the success of its recruitment efforts." Id. at para. 118. But "if it appears that, despite a broadcaster's outreach efforts, an excessive number of hires or interviewees are coming from inside, 'word-of-mouth' recruitment sources, we will expect the broadcaster to consider whether its recruitment efforts are achieving a sufficiently broad outreach." Id; see also id. at para. 115 (stating that "[d]ata as to the recruitment sources of the broadcasters' interviewees and hirees . . . will be one source of information concerning a broadcaster's EEO efforts that we may, as warranted, utilize in determining whether a broadcaster has demonstrated compliance with our EEO rule").
Clearly, then, the outreach regulations do not stop at the line between recruiting and hiring, as the Commission repeatedly asserts. As I read the plain language of the Report and Order cited above, a broadcaster could engage in every single act of outreach required under Track A but still be deemed noncompliant for failing to hire from referral sources with sufficient frequency, instead hiring too many people by word of mouth. Thus it is not just outreach that is required for compliance. Instead, broadcasters operating under Track A must avoid hiring "through an insular recruitment and hiring process," thereby "replicat[ing]" a "homogenous workforce" in which "minorities and women are poorly represented." Id.. at para. 3. (4) And the data on referral sources of employees will be used to police those hiring decisions.
Thus, under Track A, broadcasters who are not discriminating against anyone in the hiring process - indeed, who have never discriminated against anyone -- are not free to decide to hire whoever they want, as the Commission asserts. The Report & Order makes clear that they are expected to hire a certain amount of employees from referral sources. These rules are clearly aimed at a broadcaster's employment decisions and are meant to affect the racial composition of his staff by preventing the "replication" of "homogenous" staffs. I do not see how this language can be squared with the Commission's repeated claim that it has no intention of regulating hiring or injecting race and gender considerations into such decisions, and that its rules create no preferential effects whatsoever in hiring.(5)
Another measure of the efficacy of outreach under either Track A or B, according to the Order, is station employment profiles collected on Form 395. In explaining why it collects this expressly race- and gender-based hiring data, the Commission states that the data is necessary "in order to assess . . . the effectiveness of the new rules in achieving our objective of inclusive outreach." Report & Order at para. 164. The Commission further explains that "an increase in the number of women and minorities employed in the broadcast . . . industr[y] would indicate that our EEO requirements are effective in ensuring outreach." Id. If these employment numbers do not prove satisfactory to the Commission, it "will not hesitate to propose changes to [the] EEO rules if industry trends suggest that [they] are not effective." Id.
In other words, if broadcasters do not achieve some minimum level of minority and female employment, the Commission will impose added regulation - and thus greater costs -- upon them. I do not think it can be denied that an express threat of greater industry regulation creates a strong incentive to achieve the Commission's stated desire. Again, left without any clear idea as to what those employment profiles should look like, the rational broadcaster - or industry as a whole - will probably set its sights on something approaching proportionality and, if not that, at least some minimum number of minorities and women.
Finally, the Commission takes the highly irregular step of keeping open the docket in this proceeding, notwithstanding the adoption of final rules and regulations. See id. at 229. While it does not "at this time" pursue a direct remedial approach to the employment of minorities and women, the Commission will permit the submission of information "germane" to such regulation, such as a national employment disparity study. Id. The Commission "will consider any [such] submissions" and "determine [whether] action is appropriate at a later date." This action is "to facilitate any additional proceedings upon further Order," id. at para. 238, and to "facilitate the submission of information relevant to employment disparities," FCC Press Release on EEO Regulations (Jan. 20, 2000).
Again, this none too subtly suggests that if subsequent studies show a "disparity" in the employment of minorities and women -- not of broadcasters' failure to make job information widely available to any and all persons -- the Commission will take further regulatory action. Indeed, by keeping the docket open in this proceeding, the Commission actively invites such submissions and keeps the possibility of further rulemaking very much alive; no future Commission need obtain a majority vote in order to initiate a rulemaking on this matter. Had the docket simply been closed, as normally occurs when final rules are adopted, nothing would have prevented parties from filing any studies they wish with the Commission. But by leaving the docket conspicuously open, the Commission keeps the motor on this regulatory vehicle running, allowing for immediate reentry onto the regulatory fast track. The drone of that motor is more than background noise for broadcasters; it is a constant reminder of an express threat of more regulation.
In short, the Report & Order attempts to walk an excruciatingly fine constitutional line. It deletes the requirement that broadcasters use minority- and women-specific referral sources, but replaces that with other race- and gender-specific recruiting requirements, such as the job fair and job listing rules. And although the Commission does not use race or gender classifications in the text of the outreach requirement, it makes clear that in enforcing the regulations it will expressly consider the race and gender composition of applicant pools in order to assess the "effectiveness" of a station's outreach under Track B. Under Track A, the Commission intends to track hiring from referral sources and indicates that a broadcaster's failure to hire from referral sources with sufficient frequency will present a regulatory problem. The Commission has also made clear that race- and gender-based employment data will be used to assess the effectiveness of the rules under both Track A and B, promising more regulation and less discretion for broadcasters if the current regime proves to achieve results that are, in its opinion, inadequate. Finally, the Commission declines to close this proceeding, inviting the filing of information on employment "disparities" for minorities and women.
Does this system influence or encourage hiring based on race? A reviewing court very well might find that it does. Given the realities of the overall scheme and the Commission's self-avowed purpose of influencing the racial composition of broadcast employment ranks, I for one see a real risk that these regulations and the accompanying Order will operate to "pressure--even if they do not explicitly direct or require--stations to make race-based hiring decisions." Lutheran Church v. FCC, 154 F.3d at 491; see also Community-Service Broadcasting of Mid-America, Inc. v. FCC, 593 F.2d 1102, 1116 (D.C. Cir. 1978) (en banc) (discussing the "variety of sub silentio pressures and 'raised eyebrow' regulation[s]" to which broadcast licensees are subject and holding that facially neutral regulations can be invalid if they increase the likelihood of self-regulation "to avoid official pressure and regulation"); Writers Guild of America, West v. FCC, 423 F. Supp. 1064, 1098, 1105, 1117 (C.D. Cal. 1976) (finding that informal "jawboning" by agency officials is judicially reviewable), vacated and remanded on jurisdictional grounds sub nom., Writers Guild of America v. ABC, 609 F.2d 355 (9th Cir. 1979) (agreeing that "the use of these techniques by the FCC presents serious issues involving the Constitution, the Communications Act, and the APA"), cert. Denied, 449 U.S. 824 (1980). Indeed, one of the stated goals of the regulations is to affect the racial design of the employment force at broadcast stations, as well as the racial composition of station owners, in order to promote "diversity of programming." See, e.g, Report & Order at para. 59 (asserting nexus between employment and programming diversity).
Even if the regulations do not influence or encourage hiring based on race, they certainly impel recruiting based on race. This is due to the use of applicant pool data to evaluate the adequacy of outreach programs, as explained above. The Commission has not eliminated race-based decisionmaking under the EEO regulations, rather it has moved such decsionmaking one step back in the employment process.
The Order opines that such decisionmaking is harmless, however, because no one is injured when the pool of applicants is merely expanded. This view goes more to standing than to the merits of the Equal Protection issue. In any event, the substantive problem with this view is that it assumes an infinitely expandable pool of recruits, applicants, and interviewees. That assumption, while not without rhetorical appeal, is open to doubt.
At some point, a broadcast station, just like any other business, must draw the line on how many people it can afford, in terms of time and money, to recruit and interview. And when a station draws that line, these regulations might cause it to leave candidates not of the Commission's preferred race or gender standing on the other side. Those persons who are not selected as recruits or interviewees stand less of a chance of getting the job, of course. Thus, a person may be denied an opportunity to compete for the job on the same basis as all others - that is, they may be passed over for an interview or not recruited for a position based on their race. In this way, they have been harmed by a governmental scheme that incents the broadcaster, in order to achieve an acceptable applicant pool composition, to prefer one person as a recruit because of their race. See Texas v. Lesage, Sup. Ct. Slip Op. 98-1111 (Nov. 29, 1999) ("[A] plaintiff who challenges an ongoing race-conscious program and seeks forward-looking relief need not affirmatively establish that he would receive the benefit in question if race were not considered. The relevant injury in such cases is 'the inability to compete on an equal footing.'") (quoting Northeastern Fla. Chapter, Associated Gen, Contractors of America v. Jacksonville, 508 U.S. 656, 667 (1993)); see also Comments of Institute for Justice at 4-5 (explaining harm caused by outreach rule).
For the foregoing reasons, I think the Commission's outreach rules are not merely cognizant of race and gender in the way that, for example, prohibitions on discrimination are. Instead, they classify people based on their race and gender, require broadcasters to do so, and encourage broadcasters to prefer people of a particular race or gender over others as recruits, and even as employees. This regulatory scheme is not clearly race- or gender-neutral with respect to the distribution of benefits in the employment process.
In contrast to the programs established in the Order, the proposal submitted by the Broadcast Executive Directors Association (BEDA) provides an example of a race- and gender-neutral outreach plan that would present no Equal Protection problem. See Report & Order at para. 82. In its final proposal, BEDA suggested that stations, among other things: post notices of full-time vacancies either directly or through its State Broadcast Association to any group that asks in writing to receive such notification; advertise full-time vacancies over the air, in local newspapers of general circulation, or on the internet; and, if using the internet, promote the website on the air. Stations would not be required to document the race and gender of applicants or interviewees or maintain records as to the source of referrals. (6)
These requirements effect the broad dissemination of information, and they do so without regard to the race or gender of the recipients of that information, job applicants, or ultimate hires. This, in my view, is what the phrase "race-neutral" means. This, in my view, is the kind of plan that doubtless "merely require[s] stations to implement racially neutral recruiting and hiring programs," Lutheran Church, 141 F.3d at 351, and thus does not trigger equal protection concerns. As one commenter succinctly put it, and as BEDA's plan shows, "broad outreach does not necessitate race-conscious action and can easily be accomplished through race-neutral means." Comments of Institute for Justice at 9. The Commission, however, never explains why an unquestionably neutral program would be inadequate to meet its stated goals, but instead continues to insist upon the collection and use of race and gender data and race- and gender-specific regulation. To my mind, a program such as that put forth by BEDA would have been the wiser constitutional course.
B. The Legal Precedent for the Constitutionality of the Rules Is Weak
Upon examination of the cases cited by the Commission as support for the constitutionality of these regulations, that precedent appears relatively weak. At best, the constitutionality of targeted outreach appears to be an open question in the vast majority of federal judicial circuits, including the D.C. Circuit. See Lutheran Church, 154 F.3d at 492 ("Whether the government can encourage - or even require - an outreach program specifically targeted on minorities is. . . a question we need not decide."). At worst, targeted outreach requires race-based decisionmaking, triggering strict scrutiny under the Equal Protection Clause.
Contrary to the assertion in the Order, it is simply not true that "courts have consistently held that recruitment measures designed to expand the applicant pool, and that do not favor anyone in the applicant pool on the basis of race, are race-neutral and are not subject strict scrutiny." Report & Order at para. 217 & n. 352. (7) I address seriatim each case cited for this proposition. (8)
Raso v. Lago neither addressed nor decided the standard of review for targeted outreach programs under the Equal Protection Clause. That case involved the interaction of a Massachusetts law granting a preference to former residents of Boston's "West End" for apartments in a new housing development, federal housing regulations requiring developers to engage in minority outreach for residents, and an extant consent decree governing the development that required a particular racial composition of residents. The state law preference for former "West Enders," which was entirely race neutral, had the effect of creating a preference for whites in the new development because most West Enders were white. HUD felt that implementation of the state law preference directly conflicted with its regulations as well as the consent decree, to which it was a party. HUD thus negotiated an agreement whereby the statutory preference for West Enders would be curtailed.
HUD's targeted outreach regulations were not the focal point of the challenge in that case. As the Court observed, "outreach efforts are not the real source of the plaintiffs' problem--rather, it is the partial loss of their preference." 135 F.3d at 17 n. 8. As to the plaintiffs' ancillary attack on the regulations themselves, the Court rejected it on the ground that it had been "essentially abandoned on appeal." Id. at 17. So, to the extent that the outreach regulations were themselves challenged, the Court did not address that claim on the merits but instead found it waived. (9)
Duffy v. Wolle, 123 F.3d 1026 (8th Cir. 1997), did not uphold targeted outreach requirements against Equal Protection attack, or decide the applicable standard for such review, either. Instead, that case involved a reverse discrimination hiring claim under Title VII. Although the Court reasoned that affirmative efforts to recruit minorities and women result in no harm to others, it did so in the course of deciding that the existence of such a program was not enough to support a finding that the defendant employer's asserted reason for hiring someone other than the plaintiff was pretextual. See 123 F.3d at 1038 (holding that "we [do not] believe that the [defendant's] alleged interest in obtaining a pool of diverse applicants can support a finding of pretext" in hiring). Because the plaintiff's other evidence of "pretext" was also unpersuasive, the employer was not liable for employment discrimination under Title VII. This case thus did not hold that outreach measures are "race-neutral," regardless of group targeting, and thus not subject to strict scrutiny.
Neither does Ensley Branch NAACP v. Seibels, 31 F.3d 1548 (11th Cir. 1994), stand for the asserted proposition. The question presented there was not whether the targeted outreach required by the consent decree in that case was subject to strict scrutiny, but whether the race-based hiring "goals" in the decree violated Equal Protection. In the course of ruling those goals unconstitutional, the court said that the fact that the city had engaged in "race-neutral" efforts to solve its employment problems did not save their hiring goals under strict scrutiny. In describing those prior efforts, the court observed that the city "actively encouraged blacks to apply for jobs" and that "the consent decrees themselves required strengthened recruitment of blacks and women." Id. at 1571. Admittedly, the Court described, in dicta, targeted recruiting programs as "race-neutral" in the course of evaluating another aspect of the decrees' constitutionality. But it did not decide the question - as no party raised it - whether the targeted recruiting was itself constitutional, or what the applicable standard of review for such a program would be.
The same is true of Peightal v. Metropolitan Dade County, 26 F. 3d 1545 (11th Cir. 1994). Again, in the course of determining "whether a race-conscious remedy" - specifically, a hiring preference -- was "narrowly tailored to serve a compelling state interest," the Court undertook the "initial inquiry" of asking whether the government had met its obligation first to consider race-neutral measures. Id. at 1557. As in Seibels, the Court characterized a recruiting program aimed at minorities and women as "race-neutral" and concluded that the city had therefore met its obligation to take such measures before introducing the hiring quotas. Id. at 1557-1558. But no challenge was made to the recruiting program itself, and this case therefore never held that such programs are race-neutral for purposes of analyzing their constitutionality under the Equal Protection Clause.
That leaves one district court decision, Shuford v. Alabama State Board of Education, 897 F. Supp 1535 (M.D. Ala. 1995). Unlike the foregoing cases, this one does analyze the constitutionality of affirmative outreach programs. And it does reason that procedures that only expand the applicant pool - that are "inclusive" as opposed to "exclusive" -- and that do not affect hiring decisions are not subject to equal protection analysis.
By its own admission, however, the district court's reasoning "presents a new method of looking at affirmative action." Id. at 1551. The court candidly recognized not just the novelty of its analysis, but that the analysis actually "is a deviation from general affirmative-action case law." Id. at 1556. It is this portion of the opinion that the Order cites. See Order at para. 24 n.4. If this is the best case that can be cited for the Commission's legal theory of the race-neutrality of targeted outreach, then the precedent for that theory is weak, to put it mildly.
Of course, the above-discussed cases are from the lower federal courts. The last word
from the Supreme Court on "race-based decisionmaking" is Adarand Constructors, Inc. v.
Pena, 515 U.S. 200 (1995). The Court has never since suggested that the phrase "race-based decisionmaking," as used for purposes of Equal Protection analysis, means anything
other than what its plain terms indicate: the making of choices
-- whether to fire, pass over for a promotion, hire, interview, or recruit a person -- on account of an individual's race. Of course, the D.C. Circuit has already expressed its skepticism that Adarand can be limited to hiring decisions: "Under Title VII, courts have distinguished between 'preliminary' and 'ultimate' employment decisions. . . . [but] the Equal Protection Clause would not seem to admit such a de minimis exception." Lutheran Church, 141 F.3d at 351.
II. The Commission's Theory of Statutory Authority Is Problematic
The Commission's argument for statutory authority to regulate broadcast employment appears to rest in large part on a ratification theory. (10) See Report & Order at para. 21. Under the cases cited by the Commission, however, see id. at para. 26 & n.39, it is not the agency practice itself that Congress blesses post hoc, but rather the agency's construction of a particular statutory provision, pursuant to which it has purportedly acted, that Congress implicitly accepts.(11) Thus, although section 334 is indeed an explicit legislative recognition of the Commission's EEO practices at the time of its enactment in 1992, there must have been some other grant of authority under which the Commission promulgated the EEO regulations, the construction of which Congress then ratified when it enacted section 334.
In fact, the express and exclusive provisions relied upon by the Commission in adopting EEO rules were the "public interest" provisions of Title III. See Nondiscrimination Employment Practices of Broadcast Licensee, 13 FCC 2d 766 (1968); 18 FCC 2d 240 (1969); 23 FCC 2d 430 (1970); 44 FCC 2d 735 (1974). The Report & Order even acknowledges that the "ratification" argument necessarily relates back to the "public interest" parts of the Communications Act. See Report & Order at para. 26 (arguing that Congress has long known of the Commission's position that "it has authority under the public interest mandate to adopt and enforce EEO rules") (emphasis added).
Even if accepted as a legitimate method of statutory construction, all that the majority's ratification theory proves is that Congress in section 334 acquiesced in the Commission's historic reading of the public interest language in Title III. Although it is difficult to untangle the Commission's ratification argument from its section 334 argument, it seems that, at bottom, the Report & Order relies upon the "public interest" standard as an ultimate source of asserted authority for these regulations. (12) To the degree that it does so, the Report & Order is not on firm legal ground. And, for reasons described below, section 309(j) does nothing to improve that situation.
A. Both Prongs of the Commission's "Public Interest" Standard Suffer From Legal Flaws
Originally, the EEO rules were adopted in the "public interest" of furthering the general national policy against employment discrimination, as evidenced in Title VII of the Civil Rights Act of 1964. Later, the Commission stated that the regulations were meant to create diversity of programming. See 13 FCC 2d 766. See Nondiscrimination in Employment Practices (Broadcast), 60 FCC 2d 226, 229 (1976), reversed on other grounds, Office of Communications of the United Church of Christ v. FCC, 560 F.2d 529 (1977). Since then, the Commission has alternated between these goals as independent rationales or cited them as complementary aims.
Most recently, in the Lutheran Church litigation, the Commission clung solely to the diversity of programming goal, disavowing any reliance on non-discrimination. See 141 F.3d at 354. Today, it takes the opposite tack, deliberately downplaying the diversity of programming rationale. See Report & Order at para. 4 (non-discrimination goals "would be sufficient in themselves to warrant non-discrimination and outreach requirements" but the rules "also serve an important, constructive function in fostering greater diversity of viewpoints and programming").
Over time, and as described in detail below, each rationale has been drawn into question by the courts as a basis for employment discrimination rules. This, of course, explains the Commission's veering back and forth between the Scylla and Charybdis of its "twin aims." In the end, I am not sure that these regulations - in so far as they derive from the "public interest" sections of the Communication Act--will make it safely through the statutory strait.
1. The Scylla of Non-Discrimination.
As noted above, the original and exclusive policy goal of EEO regulations was founded on a pure nondiscrimination principle. Reliance upon the non-discrimination theory was perhaps fine, as a legal matter, in the 1960s. But reliance upon the goal of non-discrimination pursuant to statutory "public interest" provisions is now questionable under Supreme Court and D.C. Circuit precedent.
In NAACP v. Federal Power Commission, 425 U.S. 662 (1976), the Supreme Court held that the "public interest" provision of the Federal Power Act did not confer statutory authority upon the Federal Power Commission to regulate the employment practices of its regulatees. The Court stated that its "cases have consistently held that the use of the words 'public interest' in a regulatory statute is not a broad license to promote the general public welfare." Id. at 669.
Under NAACP v. Federal Power Commission, then, the Commission can not just promote the policy of nondiscrimination - as laudable a goal as that is - but must promote goals with a "'direct relation,'"to the purposes of the Communication Act, id. (quoting New York Central Securities Corp. v United States, 287 U.S. 12, 24-25), such as the efficient distribution of radio spectrum. (13) Indeed, the Supreme Court went out of its way in that case to characterize the FCC's employment regulations as tied to the communications policy goal of "ensur[ing] that . . . licensees' programming fairly reflects the tastes and viewpoints of minority groups," id. at 670 n. 7, as opposed to a non-discrimination goal. Likewise, in express recognition of the import of NAACP v. FPC, the D.C. Circuit has described the Commission's EEO program as regulating "the employment practices of its licensees only to the extent those practices affect the obligation of the licensee to provide programming that 'that fairly reflects the tastes and the viewpoints of minority groups,' and to the extent those practices raise questions about the character qualifications of licensees." Bilingual Bicultural Coalition on Mass Media, Inc. v. FCC, 595 F2d 621, 628 (1977) (emphasis added). (14)
Justifying the EEO rules as furthering an aim of nondiscrimination under the "public interest"standard is thus clearly problematic.(15) Indeed, this proposition is so obvious that the Lutheran Church court - having undermined the legitimacy of the alternative rationale of diversity of programming -- noted the statutory authority issue sua sponte and remanded the question for our consideration. See 141 F.3d at 354, 356.
2. The Charybdis of Diversity of Programming.
In view of the foregoing caselaw on statutory authority to prohibit employment discrimination under the "public interest" standard, the Commission might like to turn to the alternative rationale of programming diversity, as it did in the Lutheran Church litigation. But as susceptible as the anti-discrimination rationale is to doubt, the diversity rationale is even more so.(16)
First, the Lutheran Church court criticized the Commission's vague use of the term "diversity" of programming. The Court of Appeals observed that "[t]he Commission never defines exactly what it means by 'diverse programming.'" 141 F.3d at 354. On remand, this Report & Order provides no clearer a definition of "diversity" than the Commission articulated in that litigation. In fact, the Report & Order does not even attempt to grapple with this issue by, for example, distinguishing between the sorts of "diversity" it might have in mind or explaining how it measures "diversity." Instead, it speaks of promoting programming that is "responsive to the interests of a diverse community," an entirely circular concept. Report & Order at para. 4. This definitional problem remains as real as ever.
Second, as the D.C. Circuit noted the first time around, "[its] opinion
. . . undermined the proposition that there is any link between broad employment regulation and the Commission's avowed interest in broadcast diversity." Lutheran Church, 141 F.3d at 356 (emphasis added). Although the Commission cobbles together anecdotal assertions by individual commenters as support for the claimed nexus between the race and gender of station employees and the station's programming, see Report & Order at para. 58, it is highly selective in its choice of quotations; many commenters denied the nexus as a matter of fact, and the Commission never explains why they are wrong or less credible and the others right or more credible. See id. at para. 56 & n.112. In any event, this smattering of personal beliefs provides scant support for the proposition that race and gender correlate with programming choices in a statistically significant enough fashion to justify the instant employment regulations. Cf. Lamprecht v. FCC, 958 F.2d 382, 393 (D.C. Cir. 1992) (holding that even under intermediate scrutiny "[a]ny 'predictive judgments' concerning group behavior and the differences in behavior among different groups must at the very least be sustained by meaningful evidence"). (17)
Consider also the overall relationship between the adopted regulations and the ultimate goal of programming diversity. It is even more attenuated than the essential link between the race/gender of employees and programming discussed above. To get from the regulated behavior to the goal of diverse programming requires numerous leaps of evidentiary logic: first, broad outreach will lead to applicant pools with a certain number of minorities women; second, such applicant pools will in turn create interviewee pools with more minorities and women then otherwise would exist; third, the composition of such interviewee pools will then affect the composition of the employees at broadcast stations; fourth, those employees might someday become owners of broadcast properties; and fifth, those owners will then program stations based on their personal race and gender. This is a daisy chain of hypotheticals.
Equally ill-supported is the specific link between the race and gender of low-level employees and programming output. To get from the coverage of non-editorial employees to its diversity goal, the Commission states its "belief" that "program content is not determined solely by the individuals at the station with authority to select programming, but may also be influenced by interaction between these individuals and other station employees, which exposes the former to the views and perspectives of the latter." Report and Order at para. 55. Upon what record evidence is this assumption based? None. In fact, the Commission later concedes that "it is impossible to establish from empirical evidence the connection between programming decisions and the backgrounds of the decisionmakers." Id. at para. 58 (emphasis added). If it is impossible to establish that connection in an empirical context, it is even harder to establish from an individual's purely subjective impression of events at a station. See, e.g., id. (relying upon commenter who states, without offering supporting facts, "I believe that having a diversified staff . . . has helped WNBC be more conscientious towards a wider range of programming and news views") (emphasis added). This is agency speculation of the idlest sort. Cf. Bechtel v. FCC, 10 F.3d 1875 (D.C. Cir. 1993) (finding preference for owners who manage stations to be without evidentiary foundation and thus arbitrary and capricious). Worse, it is not even speculation about the actual operation and management of broadcast stations; it is conjecture about social science -- namely, the potential psychological effects that exposure to one human being might have on another. (18)
Instead of real evidence, the Commission's assumptions about the relationship of race and gender to an individual's point of view seem based on impermissible stereotyping. See generally Lamprecht, 958 F.2d at 392-394. The Supreme Court has said this about making assumptions about individuals based on their gender: "Discrimination based on archaic and overbroad assumptions about the relative needs and capacities of the sexes forces individuals to labor under stereotypical notions that often bear no relationship to their actual capabilities." Roberts v. United States Jaycees, 468 U.S. 609, 625 (1984). For example, "[i]n the specific context of sex stereotyping, an employer who acts on the basis of a belief that a woman cannot be aggressive, or that she must not be, has acted [unlawfully] on the basis of gender." Price Waterhouse v. Hopkins, 490 US 228, 250 (1989).
I think it no less an instance of sex stereotyping to say that women are naturally interested in programming or seeing coverage of topics such as "breast cancer" and "premature child birth." See Report & Order at para. 58 (citing comments of Cathy Hughes as evidence of nexus between employment of women and program diversity). A woman's pursuit of these topics as a programming executive will depend largely on sex-neutral business factors such as her target audience. A woman's interest in these topics as a viewer will depend on personal factors such as her age or marital status; or maybe she would simply rather watch a financial report, a political talk show, or a documentary on international relations. I had hoped we were "beyond the day when an employer could evaluate employees by assuming or insisting that they matched the stereotype associated with their group." Price Waterhouse, 490 U.S. at 250. The Commission cannot simply assume - or require broadcasters to assume -- that a female station manager would be more likely to take a, say, kinder, gentler, more "feminine" approach to local news or have a certain point of view on political issues. See Lamprecht v. FCC, 958 F.2d at 395-396.
Even if it is true that a majority of women are interested in the kind of issues described above, what the law has always respected, whether under Title VII or the Equal Protection Clause, are the abilities and tastes of the individual. As Justice Ginsburg put it, "generalizations about 'the way women are,' estimates of what is appropriate for most women, no longer justify denying opportunity to women whose talent and capacity place them outside the average." United States v. Virginia, 518 U.S. 515, 520 (1996). And the above-described problems associated with generalizing about people based on immutable characteristics such as gender go to race-based rules with equal force. Cf. Lutheran Church, 141 F.3d at 355 (noting that the danger of perpetuation of invidious group stereotypes is "poignantly illustrated by this case," as "one of the NAACP's primary concerns was its belief that the Church had stereotyped blacks as uninterested in classical music").
In short, the diversity of programming rationale is riddled with definitional, empirical and, thus, legal flaws. Yet the anti-discrimination rationale has real problems under the Federal Power Commission decision. Choosing between them is like deciding whether to jump into the frying pan or the fire. To the extent that the Commission's section 334 argument and ratification argument relate back to the public interest language of Title III, those arguments are legally problematic.
B. Section 309(j) Provides No Authority For Employment Regulations
To buttress its statutory argument, the Commission also cites section 309(j) of the Communications Act as authority for the EEO regulations. See Report & Order at paras. 42-47. This statute provides no support at all for these rules.
It is true enough that Congress expressed certain policy preferences with regard to minorities and women in this section. See id. at para. 42. But those policy preferences are to be implemented, by the plain terms of the statute, "[i]n identifying classes of licenses and permits to be issued by competitive bidding, in specifying eligibility and other characteristics of such licenses and permits, and in designing the methodologies for use under this subsection," the "competitive bidding" section. 47 U.S.C. section 309(j)(3). This item is not a rulemaking to design systems for license auctions. Indeed, it is not about licensing at all. It is about employment practices of existing broadcasters. Section 309(j) simply has no applicability.
As set forth above, there is legitimate reason to doubt the constitutionality of these revised EEO regulations. In essence, the Commission continues to insist on the collection and use of race and gender statistics, whether for assessing applicant pools in order to evaluate the "breadth" of outreach, for assessing the "insular" nature of hiring, or for determining the overall adequacy of the regulations. Although there is no case law squarely against what the Commission has done, neither is there any in direct support of it, as the Commission claims. Even if this cloud of constitutional doubt were removed - for instance, by the adoption of a truly race- and gender-neutral plan such as BEDA's - the statutory authority for the rules, in so far as it is built on the public interest standard and section 309(j), is quite vulnerable. In the end, I cannot support these regulations.
2. See also id. at para. 8 (records concerning the race, ethnicity, and gender of applicants must be maintained in order to "monitor whether. . . outreach efforts have been successful in achieving broad outreach" and "[i]f the data collected indicates that outreach has not been inclusive, a broadcaster . . . will be expected to adjust its outreach program accordingly"); id. at para. 105 (describing the possibility of "verifying broad outreach using applicant pool data"); id. at para. 113 (describing collection of applicant pool data by station as necessary to "demonstrat[e] that its outreach efforts are inclusive"); id. at para. 115 (stating that "applicant flow data. . . will be one source of information concerning a broadcaster's EEO efforts that we may, as warranted, utilize in determining whether a broadcaster has demonstrated compliance with our EEO rule").
3. There are, of course, practical problems with using the number of persons who apply for a job to measure the number of people who received notice of the job. Cf. City of Richmond v. J.A. Croson, Inc., 488 U.S. 469, 507 (1989) (criticizing the "completely unrealistic assumption that minorities will choose a particular trade in lockstep proportion to their representation in the local population"). There are virtually infinite reasons why a person who hears about a job might not ultimately apply for it - perhaps they are already employed, or maybe they are not as interested in a broadcasting career (not the most dynamic sector of the communications industry today) as the Commission thinks they should be.
4. If the broadcaster refuses to hire applicants because of their race or gender, that is of course another thing, and wholly actionable under employment discrimination laws.
5. The breadth of this policy - which limits the ability of broadcasters to hire based on "word of mouth," without any evidence of past or present discrimination --is remarkable when compared to Title VII. Under Title VII, employment "practices, procedures, or tests neutral on their face, and even neutral in terms of intent, cannot be maintained if they operate to 'freeze' the status quo of prior discriminatory employment practices." Griggs v. Duke Power, 401 U.S. 424, 430 (1971). The Commission relies on a similar perpetuation-of-discrimination theory in order to tie the outreach rules to the non-discrimination (as opposed to the diversity) rationale. See Report & Order at para. 3 ("We believe that repeated hiring without broad outreach may unfairly exclude minority and women job candidates when minorities and women are poorly represented in an employer's staff. . . . Outreach in recruitment must be coupled with a ban on discrimination to effectively deter discrimination and ensure that a homogenous workforce does not simply replicate itself through an insular recruitment and hiring process."). Generally, however, there must first be a showing of "prior discriminatory employment practices," Griggs, 401 U.S. at 430, to warrant the inference of present, continuing discrimination. Yet the Commission presumes present discrimination on the part of those who make hiring decisions whenever minorities and women are "poorly represented" at stations, without any evidence whatsoever of past or present bad acts. Absent such evidence, the Commission's policy begins to look like one of "outright racial balancing." Lutheran Church, 141 F.3d at 355. Even on the Commission's own dubious logic, however, the outreach rules could apply only to stations with "poor representation" of the suspect classes, for only in those circumstances could there be any chance of the "replication" effect that the Commission seeks to prevent.
6. See Letter to Chairman William E. Kennard from Richard R. Zaragoza on behalf of the Broadcast Executive Directors Association, in the Matter of Equal Employment Opportunity, MM Docket Nos. 98-204 and 96-16, Dec. 29, 1999.
7. As discussed above, this characterization of the Commission's program as not creating pressures to favor any class of persons in the hiring process is subject to dispute.
8. Curiously, the Commission characterizes its outreach program as non-targeted, but relies upon cases involving mostly targeted programs. I address the cases in any event, since it appears to me that the regulations are, as a practical matter and as previously explained, specifically aimed at minorities and women.
9. In response to the primary challenge, the Court of Appeals held that HUD did not violate the Equal Protection Clause when it cut back on the statutory preference for West Enders in order to make housing available to all, regardless of race. There was no record evidence that the HUD plan distributed housing based on any racial classifications. 135 F.3d at 16. Rather, it sought to mitigate the effects of a law that might have subjected HUD to sanctions under the consent decree. Id. at 17 ("HUD's concern that the preference, in this instance, if unmodified, would restrict the preference to whites and subject HUD to sanctions under the consent decree" was not an illegitimate, race-based motive). In closing, the Court expressly noted the limited scope of its holding. Id. ("[I]t is one thing for HUD to insist that all apartments it subsidizes must effectively be open to all races" but "it would be quite another thing if HUD planned to impose this requirement only where the beneficiaries of the statutory preference were white.").
10. I do not doubt that section 634 of the Communications Act grants the Commission statutory authority to make EEO rules for cable systems, according to the dictates of that provision. Whether the statute itself is constitutional, and whether the Commission has complied with those dictates, are other matters.
11. See Haig v. Agee, 453 U.S. 280, 300 (1981) (holding that Congress had approved of administrative "interpretation [of the Passport Act of 1926]" as statutory authority for passport regulations); Lorillard v. Pons, 434 U.S. 575 (1978) (holding that Congress was presumed aware of the judicial construction of the Fair Labor Standards Act as requiring a jury trial when it incorporated sections of that law in the Age Discrimination in Employment Act); Zemel v. Rusk, 381 U.S. 1, 11 (1965) (holding that Congress had approved of administrative construction of the 1926 Passport Act's "broad rule-making authority" as basis for passport regulations); Norwegian Nitrogen Products Co. v. United States, 288 U.S. 294 (1933) (holding that Congress had acquiesced in administrative practice pursuant to 1922 Tariff Act and explaining that "administrative practice, consistent and generally unchallenged, will not be overturned except for very cogent reasons if the scope of the [statutory] command is indefinite and doubtful," thus presuming a contemporaneous statutory command).
12. For radio licensees, the public interest language, without any "ratification" of its construction, is all the Commission has; section 334 applies only to television licensees.
13. Some of the Commission's stated goals clearly fail this test. For example, the Commission asserts it belief that the regulations will "increase our understanding of those from different backgrounds, decrease the sense of isolation of minority groups, and help us build bridges across racial, ethnic and socioeconomic divides." Report & Order at para. 4. While perhaps part of a broad social or even religious agenda, these are not directly related to communications policy.
14. Not coincidentally, it was not until 1976 - just months after Federal Power Commission was handed down -- that the Commission first articulated an end other than nondiscrimination for its EEO policies. See Nondiscrimination in Employment Practices (Broadcast), 60 FCC 2d at 229 (EEO rules are meant to promote diverse programming).
15. To the extent that the Commission wishes to rely upon character qualifications, as opposed to pure non-discrimination or diversity of programming rationales, it is hard to see why all holders of Title III licenses would not be subject to this same understanding of character adequate to hold a federal license. Such a definition of "character" seems arbitrarily limited to broadcasters, as opposed to all Title III license holders, of which there are many other than traditional broadcasters.
16. No doubt this is why the Commission tries to link the outreach rule to the non-discrimination rationale, see Report & Order at paras. 3, 40 (discussing "perpetuation-of- discrimination" theory as basis for outreach rules), as opposed to the diversity rationale. As noted supra n. 5, the theory employed to justify the outreach rule as necessary to prevent discrimination goes far beyond even the broadest propositions of employment discrimination law.
17. Even if there were sufficient evidence of such a correlation, the D.C. Circuit has expressed its "doubt . . . that the Constitution permits the government to take account of racially based differences, much less encourage them." Lutheran Church, 141 F.3d at 392.
18. Where our judgments are afforded deference, it is with respect to areas within our agency's expertise such as, say, the technological nature of digital television. Employment matters do not fall within that zone of expertise and thus deference. Cf. Bechtel, 10 F.3d at 881 (stating that where "predictive judgments" underlying a policy concern an area beyond the Commission's expertise, deference to those judgments is not as warranted).