NLRB v. CNN America, Inc., No. 15-1112 (D.C. Cir. 2017)

Annotate this Case
Justia Opinion Summary

The Board applied for enforcement of its finding that CNN's replacement of its unionized contractor with a nonunion, in-house workforce violated the National Labor Relations Act. The D.C. Circuit held that the Board's determination that CNN and TVS were joint employers cannot stand because the Board applied a standard for determining whether companies were joint employers that appeared to be inconsistent with its precedents, without addressing those precedents or explaining why they do not govern. Therefore, the court reversed the Board's finding that CNN was a joint employer with TVS. The court affirmed the Board's remaining three unfair-labor-practice findings that did not depend on CNN's joint-employer status. The court affirmed the Board's application and denial of CNN's cross-petition for review in all other respects.

Download PDF
United States Court of Appeals FOR THE DISTRICT OF COLUMBIA CIRCUIT Argued February 23, 2017 Decided August 4, 2017 No. 15-1112 NATIONAL LABOR RELATIONS BOARD, PETITIONER v. CNN AMERICA, INC., RESPONDENT NATIONAL ASSOCIATION OF BROADCAST EMPLOYEES AND TECHNICIANS - COMMUNICATIONS WORKERS OF AMERICA, LOCAL UNION NO. 11 AND NATIONAL ASSOCIATION OF BROADCAST EMPLOYEES AND TECHNICIANS COMMUNICATIONS WORKERS OF AMERICA, LOCAL UNION NO. 31, INTERVENORS Consolidated with 15-1209 On Application for Enforcement and Cross-Petition For Review of an Order of the National Labor Relations Board Kannon K. Shanmugam argued the cause for CNN America, Inc. With him on the briefs were Kevin T. Baine, Paul Mogin, and Zachary D. Fasman. 2 Maurice Baskin, Michael J. Lotito, and Elizabeth Parry were on the brief for amici curiae Chamber of Commerce of the United States of America, et al. in support of CNN America. Joan E. Hoyte-Hayes, Supervisory Attorney, National Labor Relations Board, argued the cause for the National Labor Relations Board. With her on the brief were Richard F. Griffin, Jr., General Counsel, John H. Ferguson, Associate General Counsel, and Linda Dreeben, Deputy Associate General Counsel. Usha Dheenan, Attorney, entered an appearance. Keith R. Bolek argued the cause for intervenors. With him on the brief was Patricia McConnell. Before: GARLAND, Chief Judge, and KAVANAUGH and PILLARD, Circuit Judges. Opinion for the Court filed by Chief Judge GARLAND. Opinion concurring in Parts II and V and dissenting in part filed by Circuit Judge KAVANAUGH. GARLAND, Chief Judge: For many years, Cable News Network (CNN) used outside contractors to provide technicians to operate the electronic equipment at its Washington, D.C. and New York City bureaus. In 2003, CNN changed that longstanding arrangement, terminating its latest contracts and hiring a new in-house workforce. The National Labor Relations Board found that CNN’s replacement of its unionized contractor with a nonunion, in-house workforce violated the National Labor Relations Act in several respects. 3 The Board now applies for enforcement of its decision and order. CNN cross-petitions for review. For the reasons set forth below, we grant each request in part and deny each in part. I CNN is a leading television and online news provider. Its Washington, D.C. and New York City bureaus, from their inception, relied on outside contractors to operate the equipment necessary to produce and broadcast the news. Pursuant to exclusive service contracts -- known as Electronic News Gathering Service Agreements -- camera operators, sound technicians, studio technicians, and broadcast engineers (hereinafter, “technicians”) employed by the contractors performed much of the technical work at both bureaus. Those technicians were consistently represented by a union. In 1982, the Board certified the National Association of Broadcast Employees and Technicians (hereinafter, “the union”) as the collective-bargaining representative of the technicians staffing the Washington bureau. In 1985, the Board certified the same union to represent the New York bureau’s technicians. Over the following years, the union and the contracting companies employing the technicians entered into successive collective-bargaining agreements. When CNN switched contracting companies -- which happened several times -- the new company would hire nearly all of its predecessor’s employees and continue to recognize the union. In 1997, Asgard Entertainment Group successfully bid for the Washington contract and created Team Video Services (TVS) for the sole purpose of staffing the Service Agreement with CNN. In 2001, Asgard won the New York contract, which it also serviced through TVS. The Service Agreements at the two bureaus were materially identical. Broadly, they required 4 TVS to furnish CNN with technicians, as well as supervisors for those technicians, in exchange for CNN covering TVS’ labor expenses and paying a monthly management fee. The Service Agreements provided that TVS would supply full-time technicians for at least 40 hours per week, in addition to parttime technicians available 24/7 as needed by CNN. They also provided that CNN: could require changes in TVS staffing levels; could audit TVS’ books without cause or notice; would fund a 4% increase in salary expenses per year; would provide all equipment used by TVS technicians; and had the sole option to renew the agreement and could terminate it for any reason upon four weeks’ notice. On September 29, 2003, CNN announced that it was terminating its contracts with TVS and would begin directly hiring employees to perform the camera, studio, and engineering work at the Washington and New York bureaus. CNN expressed appreciation for TVS’ performance and service. It said, however, that it wanted a new workforce to allow it to take advantage of technological developments in the industry, particularly computer-related technology. When the union requested recognition and bargaining, and asked to discuss future employment prospects for bargaining-unit employees, CNN denied the requests. CNN named the process by which it would directly hire its new workforce the “Bureau Staffing Project.” According to CNN executives, the company planned to use a multistep “behavioral interviewing process” to hire over 200 skilled technicians for the two bureaus. Former TVS employees could apply for the new jobs, but they would have to go through the Staffing Project’s interviewing process. In the end, more than 100 TVS bargaining-unit employees were not hired and lost their jobs. 5 Three months after CNN officially terminated the Service Agreements, the union filed unfair-labor-practice charges with the National Labor Relations Board (NLRB). Three years later, in 2007, the Board’s General Counsel filed his own complaint against CNN. In November 2008, after an 82-day trial, an administrative law judge (ALJ) ruled against CNN in an 83-page opinion. The ALJ first determined that CNN had been a joint employer of TVS’ employees before the termination of the Service Agreements and was thus bound by TVS’ collectivebargaining agreements with the union. He further determined that CNN became a successor employer after it terminated TVS and hired a new workforce. The ALJ found that “the reasons given by CNN for its termination of its contracts with [TVS] and its implementation of the Bureau Staffing Project [were] pretextual. A major motive in these decisions was CNN’s desire to operate its Washington and New York bureaus without a union.” CNN America, Inc., 361 NLRB No. 47, at 51 (2008) (ALJ Op.). The ALJ also found that “the Bureau Staffing Project was a sham process,” during which “CNN engaged in widespread and blatant discrimination against [TVS] bargaining unit members.” Id. at 50. “CNN did so,” the ALJ found, “with the objective of depriving employees of [union] representation.” Id. On the basis of these findings, and others discussed below, the ALJ determined that CNN committed multiple violations of the National Labor Relations Act (NLRA). CNN appealed to the Board, which finally issued its decision in 2014, affirming the ALJ in all relevant respects. CNN America, Inc., 361 NLRB No. 47, at 1 & n.1 (2014) (Board Op.). The Board found that, as a joint employer, CNN: violated NLRA § 8(a)(3) and (1), 29 U.S.C. § 158(a)(3), (1), by terminating the Service Agreements out of anti-union animus; and violated § 8(a)(5) and (1), id. § 158(a)(5), (1), by failing to 6 bargain with the union about its decision to terminate those agreements. The Board further found that, as a successor employer, CNN violated § 8(a)(5) and (1) by failing to recognize and bargain with the union and unilaterally changing the employees’ terms and conditions of employment. The Board also found that CNN: violated § 8(a)(3) and (1) by discriminating against union members in its hiring process; and violated § 8(a)(1) on four occasions by making coercive statements through its supervisors. As a remedy for those violations, the Board ordered that, among other things, CNN: (1) provide backpay and benefits to all TVS technicians who either lost their jobs or received reduced wages as a result of CNN’s violations; (2) reinstate and provide necessary training to all TVS technicians who were discharged and not hired by CNN; and (3) recognize and bargain with the union.1 CNN filed a motion for reconsideration, which the Board denied. CNN America, Inc., 362 NLRB No. 38 (2015). The Board now applies for enforcement of its decision and order, and CNN cross-petitions for review. The union has intervened in support of the Board. 1 Member Miscimarra concurred in part and dissented in part. He concluded that CNN: (1) did not qualify as a joint employer and therefore did not violate the NLRA by terminating the Service Agreements, 361 NLRB No. 47, at 31-42; (2) did qualify as a successor employer, however, and therefore had an obligation to bargain with the union once it hired its new workforce, id. at 28 n.1; (3) violated § 8(a)(3) through some individual discriminatory hiring decisions, but not through a general hiring scheme motivated by antiunion animus, id.; and (4) violated § 8(a)(1) through three coercive statements by supervisors, id. He also dissented in part from the Board’s remedial order. Id. at 29. 7 II The Board first had to decide whether CNN was a joint employer of TVS’ employees. The Board’s affirmative answer to that question led it to find that CNN committed two unfair labor practices: terminating the Service Agreements due to antiunion animus, in violation of § 8(a)(3) and (1); and failing to bargain with the union before terminating the Service Agreements, in violation of § 8(a)(5) and (1). If, as CNN argues, the Board wrongly concluded that CNN was a joint employer, then those two unfair-labor-practice findings must fall away. This is so because, if CNN was not a joint employer of TVS’ employees, it would not have been bound by the collective-bargaining agreement between TVS and the union, and its termination of the Service Agreements would thus have been lawful. See Computer Assocs. Int’l, Inc. v. NLRB, 282 F.3d 849, 852-53 (D.C. Cir. 2002); see also Computer Assocs. Int’l, Inc., 324 NLRB 285, 286 (1997) (“[A]n employer does not violate Section 8(a)(3) by ceasing to do business with another employer because of the union or nonunion activity of the latter’s employees.”), enforcement denied on other grounds, 282 F.3d 849 (D.C. Cir. 2002). We conclude that the Board’s determination that CNN and TVS were joint employers cannot stand. This is not because we find that the two companies lacked a joint-employer relationship. Rather, it is because the Board applied a standard for determining whether companies are joint employers that appears to be inconsistent with its precedents, without addressing those precedents or explaining why they do not govern. Our conclusion does not bar the Board from finding CNN to be a joint employer by applying a different standard or sufficiently explaining the one it did apply. It means only that we cannot enforce the Board’s determination in this proceeding. 8 A Citing two 1984 decisions, the Board began its jointemployer analysis by setting forth the governing standard it intended to apply: “The Board will find that two separate entities are joint employers of a single workforce if the evidence shows that they ‘share or codetermine those matters governing the essential terms and conditions of employment.’” CNN America, Inc., 361 NLRB No. 47, at 3 (quoting TLI, Inc., 271 NLRB 798, 803 (1984)) (emphasis added); see id. (citing Laerco Transp., 269 NLRB 324, 325 (1984)); see also ALJ Op., 361 NLRB No. 47, at 52-53. In two sentences in a footnote, the Board acknowledged that its subsequent 2002 opinion in Airborne Express “stated that the test for joint-employer status requires ‘direct and immediate’ control by the putative joint employer over employment matters.” 361 NLRB No. 47, at 3 n.7 (quoting Airborne Express, 338 NLRB 597, 597 n.1) (emphasis added). But the Board noted that the case Airborne Express cited for that proposition, TLI, Inc., had “ma[de] no mention that control over employment matters must be direct and immediate.” Id. Apparently sidestepping the “direct and immediate” control requirement, which it never mentioned again, the Board concluded that CNN and TVS were joint employers.2 The Board’s decision was issued by a three-member panel. After the panel’s decision, but before the briefing of this appeal, the full Board sat in another case to consider “its current standard for assessing joint-employer status.” Browning-Ferris, 362 NLRB No. 186, at 1 (2015). In Browning-Ferris, the full Board canvassed a 30-year history of its joint-employer cases -- 2 The NLRB’s appellate brief acknowledges that the Board did not apply the “direct and immediate” control standard. NLRB Br. 28. 9 a period beginning with TLI and Laerco Transportation and running through Browning-Ferris itself -- and concluded that the prevailing standard during that period required an employer’s exercise of “direct [and] immediate” control. Id. at 13. “Most significantly,” the Board said, the Board’s decisions have implicitly repudiated . . . reliance on reserved control and indirect control as indicia of joint-employer status. The Board has foreclosed consideration of a putative employer’s right to control workers, and has instead focused exclusively on its actual exercise of that control -- and required its exercise to be direct, immediate, and not “limited and routine.” Browning-Ferris, 362 NLRB No. 186, at 13. Among other cases, the Board said, Airborne Express had “held that ‘[t]he essential element in [the joint-employer] analysis is whether a putative joint employer’s control over employment matters is direct and immediate.’” Id. at 14 (quoting Airborne Express, 338 NLRB at 597 n.1) (emphasis added). Having established that the existing standard was “direct and immediate” control, Browning-Ferris then went on to criticize that standard. Following an extensive discussion, the Board concluded that “the current joint-employer standard is not mandated by the Act and . . . does not best serve the Act’s policies.” Id. at 15. “[W]e will no longer require,” the Board continued, “that a joint employer not only possess the authority to control employees’ terms and conditions of employment, but must also exercise that authority, and do so directly, immediately, and not in a ‘limited and routine’ manner.” Id. at 19. “Accordingly, we overrule Laerco, TLI, A&M Property, and Airborne Express, . . . and other Board decisions, to the extent that they are inconsistent with our decision today. The 10 right to control, in the common-law sense, is probative of joint-employer status, as is the actual exercise of control, whether direct or indirect.” Id. Under its revised standard, the Board said, “two or more entities are joint employers of a single work force if they are both employers within the meaning of the common law, and if they share or codetermine those matters governing the essential terms and conditions of employment.” Id. B The difference between the case now before us and Browning-Ferris should be apparent. In Browning-Ferris, the Board carefully examined three decades of its precedents and concluded that the joint-employer standard they reflected required “direct and immediate” control. It then criticized that standard. Thereafter, it forthrightly overruled those cases and set forth, as “a new rule” for identifying joint employment, a standard quite similar to the one the Board in the case before us claimed had been the standard all along. Id. at 3; see id. at 19. This an agency may do, as long as it provides a reasoned explanation for its change of course. See FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009). In fact, whether the Board did so in Browning-Ferris is a question at issue in a petition for review of that decision that is currently pending before another panel of this court. See Browning-Ferris Indus. of Cal. v. NLRB, No. 16-1028 (D.C. Cir. filed Jan. 20, 2016). In the case on review here, however, the Board did none of those things. In characterizing the prevailing joint-employer standard, it did not grapple with its precedents in the manner of Browning-Ferris. It did not explain why it thought precedents that seemed to “focus[] exclusively on [the employer’s] actual exercise of . . . control -- and require[] its exercise to be direct [and] immediate,” Browning-Ferris, 362 NLRB No. 186, at 13, 11 instead supported a more flexible “share or codetermine” standard.3 Indeed, it did not even mention many of the important precedents at all, including those that expressly used the “direct and immediate” control formulation.4 Nor did the Board mention the fact that, just three months earlier, its own General Counsel’s amicus brief in Browning-Ferris had said: “[T]he Board [has] made clear that the essential element in its current analysis is ‘whether a putative joint employer’s control over employment matters is direct and immediate.’” Amicus Brief of the General Counsel at 8, Browning-Ferris Indus., Docket No. 32-RC-109684 (June 26, 2014) (quoting Airborne Express, 338 NLRB at 597 n.1) (emphasis added in brief). And because it did not acknowledge the precedent suggesting that 3 See, e.g., Flagstaff Med. Ctr., 357 NLRB 659, 666-67 (2011) (finding no joint-employer status where an entity made recommendations about which employees to hire and fire, supervised employees daily, and evaluated employee performance); Am. Prop. Holding Corp., 350 NLRB 998, 1000-01 (2007) (finding no jointemployer status where a company had the right to approve hiring decisions and supervised employees onsite); G. Wes Ltd. Co., 309 NLRB 225, 226 (1992) (finding no joint-employer status where a company supervised employees onsite on a daily basis); So. Cal. Gas Co., 302 NLRB 456, 461-62 (1991) (finding no joint-employer status where a company provided detailed specifications of work to be done, directed employees as to where and when to perform, and gave out assignments). 4 See In re Wiers Int’l Trucks, 353 NLRB 475, 487 (2008) (“The essential element in this analysis is whether the putative joint employer’s control over employment matters is direct and immediate.” (quoting Airborne Express, 338 NLRB at 597 n.1)); Summit Express, Inc., 350 NLRB 592, 592 n.3 (2007) (“[W]e find that the contract terms, by themselves, do not establish direct and immediate control over the terms and conditions of employment . . . required to prove a joint employer relationship . . . .”). 12 “direct and immediate” control was the existing standard, it certainly did not forthrightly overrule it.5 Such “[s]ilence in the face of inconvenient precedent is not acceptable.” Jicarilla Apache Nation v. Dep’t of Interior, 613 F.3d 1112, 1120 (D.C. Cir. 2010). “An agency’s failure to come to grips with conflicting precedent constitutes ‘an inexcusable departure from the essential requirement of reasoned decision making.’” Ramaprakash v. FAA, 346 F.3d 1121, 1125 (D.C. Cir. 2003) (quoting Columbia Broad. Sys. v. FCC, 454 F.2d 1018, 1027 (D.C. Cir. 1971)). Indeed, it is “elementary that an agency must conform to its prior decisions or explain the reason for its departure from such precedent.” Gilbert v. NLRB, 56 F.3d 1438, 1445 (D.C. Cir. 1995). “[A]n agency changing its course must supply a reasoned analysis indicating that prior policies and standards are being deliberately changed, not casually ignored, and if an agency glosses over or swerves from prior precedents without discussion it may cross the line from the tolerably terse to the intolerably mute.” Greater Boston Television Corp. v. FCC, 444 F.2d 841, 852 (D.C. Cir. 1970).6 Because the Board crossed that line here, we must set aside its finding that CNN was a joint employer. See, e.g., E.I. Du Pont De Nemours & Co. v. NLRB, 682 F.3d 65, 70 (D.C. Cir. 2012). 5 Nor has this court overruled it. In Dunkin’ Donuts Mid-Atlantic Distribution Center, Inc. v. NLRB, we did enforce the Board’s jointemployer finding without invoking the “direct and immediate” control requirement. 363 F.3d 437, 440-41 (D.C. Cir. 2004) (enforcing In re Aldworth Co., 338 NLRB 137 (2002)). But the Board decision on review in that case predated Airborne Express, and no party argued that “direct and immediate” control was the proper standard. 6 See Fox Television Stations, Inc., 556 U.S. at 515; LePage’s 2000, Inc. v. Postal Regulatory Comm’n, 642 F.3d 225, 233 (D.C. Cir. 2011); Dillmon v. Nat’l Transp. Safety Bd., 588 F.3d 1085, 1089-90 (D.C. Cir. 2009). 13 And as a consequence, we must vacate the two unfair-laborpractice findings that rested on CNN’s joint-employer status. See Computer Assocs. Int’l, 282 F.3d at 853. We emphasize that nothing in our holding in this case precludes the Board from adopting a “share or codetermine” standard that takes into account a putative employer’s indirect control of a group of workers. As we have noted, the validity of the Board’s rejection of the “direct and immediate” control requirement in Browning-Ferris is at issue in the pending petition for review of that decision. Nor does anything in our holding preclude the Board, on remand, from applying the “direct and immediate” control standard and concluding that CNN satisfied that standard. But it did not do so in the proceedings in this case, and this court lacks authority to resolve the case by applying that standard itself. As the Supreme Court held over 70 years ago in SEC v. Chenery Corp., “[t]he grounds upon which an administrative order must be judged are those upon which the record discloses that its action was based.” 318 U.S. 80, 87 (1943); see Michigan v. EPA, 135 S. Ct. 2699, 2710 (2015); see also Williams Gas Processing - Gulf Coast Co. v. FERC, 373 F.3d 1335, 1345 (D.C. Cir. 2004) (“It is axiomatic that we may uphold agency orders based only on reasoning that is fairly stated by the agency in the order under review . . . .”). III Although we cannot affirm the NLRB’s finding that CNN was a joint employer with TVS, the Board’s three remaining unfair-labor-practice findings do not depend on CNN’s jointemployer status. The first of the three survives if CNN was a successor employer to TVS, even if the two were not joint employers. The Board found that CNN was a successor, and on that basis found that CNN violated NLRA § 8(a)(5) and (1) by failing to bargain with the union after it completed its hiring and 14 became the technicians’ employer. “When the Board concludes that a violation of the NLRA has occurred, we must uphold that finding unless it ‘has no rational basis’ or is ‘unsupported by substantial evidence.’” Bally’s Park Place, Inc. v. NLRB, 646 F.3d 929, 935 (D.C. Cir. 2011) (quoting United Mine Workers of Am., Dist. 31 v. NLRB, 879 F.2d 939, 942 (D.C. Cir. 1989)); see 29 U.S.C. § 160(e). And because the Board largely adopted “the ALJ’s findings and conclusions as its own, we apply the same deferential standard to those findings and conclusions.” Weigand v. NLRB, 783 F.3d 889, 895 (D.C. Cir. 2015). NLRA § 8(a)(5) makes it an unfair labor practice for an employer “to refuse to bargain collectively with the representatives of [its] employees.” 29 U.S.C. § 158(a)(5). A new employer, however, “generally assumes an obligation to bargain with the representative of its predecessor’s employees only if the new employer is considered a ‘successor’ to the old.” Waterbury Hotel Mgmt., LLC v. NLRB, 314 F.3d 645, 653 (D.C. Cir. 2003). An entity qualifies as a successor employer when: (1) it “does not make a ‘significant change’ in the ‘essential nature’ of the business, and (2) ‘a majority of the new [employer’s] employees were employed by the predecessor.’” Capital Cleaning Contractors, Inc. v. NLRB, 147 F.3d 999, 1005 (D.C. Cir. 1998) (quoting Elastic Stop Nut Div. of Harvard Indus., Inc. v. NLRB, 921 F.2d 1275, 1281 (D.C. Cir. 1990)); see Waterbury Hotel Mgmt., LLC, 314 F.3d at 653; see also Fall River Dyeing & Finishing Corp. v. NLRB, 482 U.S. 27, 41-43 (1987). CNN does not contend that it made a significant change in the essential nature of TVS’ operations. And for good reason. As the Board found, after CNN took over from TVS, “CNN continued the same business operations with employees who performed the same work, at the same locations, and using the same equipment, as the TVS technicians.” Board Op., 361 15 NLRB No. 47, at 21; see also ALJ Op., 361 NLRB No. 47, at 108. The only question, then, is whether a majority of CNN’s new employees were previously employed by TVS. Normally, answering this question requires an exercise in counting. But when “a successor refuses to hire predecessor employees because of anti-union animus, the Board presumes that but for such discrimination, the successor would have hired a majority of incumbent employees.” Waterbury Hotel Mgmt., LLC, 314 F.3d at 655; see Capital Cleaning Contractors, Inc., 147 F.3d at 1008. “In effect, when a successor refuses to hire its predecessor’s employees based upon anti-union animus, the successor loses the right unilaterally to set the initial terms and conditions of employment; it must first bargain with the union.” Capital Cleaning Contractors, Inc., 147 F.3d at 1008. The Board took this tack in finding that CNN was a successor employer -- and thus violated § 8(a)(5) by refusing to bargain -because it determined that CNN discriminated against TVS employees (who were invariably union members) in its hiring process. See Board Op., 361 NLRB No. 47, at 18, 21.7 To determine whether an employer engaged in discriminatory hiring, the Board employs a burden-shifting analysis known as the Wright Line test. See Wright Line, 251 7 The Board and ALJ also found that CNN was a successor employer because the historical bargaining unit at each TVS bureau -consisting of field camera, field audio, engineering, and studio technical employees, ALJ Op., 361 NLRB No. 47, at 49-50 -remained appropriate, and a majority of each unit’s employees were previously employed by TVS. See Board Op., 361 NLRB No. 47, at 1 n.1 & 18 n.36; ALJ Op., 361 NLRB No. 47, at 105, 108. Member Miscimarra agreed that the record supports this finding, 361 NLRB No. 47, at 28 n.1, 42 & n.24, as do we. As a consequence, CNN incurred a duty to bargain as a successor employer even absent discrimination. 16 NLRB 1083 (1980), enforced, 662 F.2d 899 (1st Cir. 1981).8 First, the General Counsel must show that the employer’s hiring decisions were “motivated by anti-union considerations.” Waterbury Hotel Mgmt., LLC, 314 F.3d at 651 (quoting Teamsters Local Union No. 171 v. NLRB, 863 F.2d 946, 955 (D.C. Cir. 1988)). “The Board may rely on both direct and circumstantial evidence in resolving this question of fact.” Id. “Once the General Counsel has established that the employer was in fact motivated by anti-union animus, the Board must find a violation of the Act unless the employer can show that ‘it would have taken the [same] action regardless of the existence of such animus.’” Id. (quoting Elastic Stop Nut Div. of Harvard Indus., Inc., 921 F.2d at 1280). In this case, the Board found that the “evidence of animus . . . [was] overwhelming, as [was] the evidence that CNN’s explanations for its conduct were pretextual.” Board Op., 361 NLRB No. 47, at 18. It therefore concluded that CNN’s failure to hire over 100 TVS technicians was unlawfully discriminatory. We review NLRB findings of discriminatory motive with considerable deference. Fort Dearborn Co. v. NLRB, 827 F.3d 1067, 1072 (D.C. Cir. 2016) (“Our review of the Board’s conclusions as to discriminatory motive is even more deferential, ‘because most evidence of motive is circumstantial.’” (quoting Inova Health Sys. v. NLRB, 795 F.3d 8 See also NLRB v. Transp. Mgmt. Corp., 462 U.S. 393, 401-04 (1983) (approving Wright Line test); Shamrock Foods Co. v. NLRB, 346 F.3d 1130, 1135 (D.C. Cir. 2003); Tasty Baking Co. v. NLRB, 254 F.3d 114, 125-26 (D.C. Cir. 2001). 17 68, 80 (D.C. Cir. 2015))).9 As discussed below, we conclude that substantial evidence supports the Board’s findings. A To support its determination that anti-union animus motivated CNN’s hiring decisions, the Board pointed to several kinds of evidence. To begin, the Board found that CNN supervisors deliberately renamed every bargaining-unit job category, merged job functions, and drafted new position qualifications with the “purpose of getting out from under the Union’s jurisdiction” and of “minimizing the significance of the [TVS employees’] prior experience when they applied for the ‘new’ jobs.” Board Op., 361 NLRB No. 47, at 19. Camera and audio field technicians became photojournalists; engineers combined duties with information-technology specialists to form the new BIT/Engineering division; and studio, control room, and qualitycontrol technicians became studio operators, audio designers, and floor directors. Id. at 9. In practice, however, those title changes were devoid of substance: new CNN employees performed the same duties and used the same equipment and technology as the former TVS employees. See ALJ Op., 361 NLRB No. 47, at 51, 108. Regarding the newly required “qualifications,” CNN claimed, for example, that photojournalists would need to be more adept at “nonlinear editing” -- that is, editing on a computer rather than on tape, id. at 73 -- than were TVS’ field technicians. The ALJ found, however, that CNN’s emphasis on 9 See, e.g., Traction Wholesale Ctr. Co. v. NLRB, 216 F.3d 92, 99 (D.C. Cir. 2000); Vincent Indus. Plastics, Inc. v. NLRB, 209 F.3d 727, 734 (D.C. Cir. 2000). 18 nonlinear editing was pretextual because the skill was only “marginally important to the performance” of the relevant jobs and was one that TVS “applicants could have acquired with minimal training.” Id. at 62; see id. at 73. Indeed, the supervisor who orchestrated these qualification changes revealed his true motive in an email to other supervisors, writing: “the Photojournalist [position qualifications] . . . should emphasize the use of DV cameras (since this isn’t within NABET [union] jurisdiction now).” Id. at 73 (emphasis added). On the basis of a detailed examination of the hiring experience of a large number of specific applicants in both D.C. and New York, see id. at 64-101, the ALJ also found “[c]ompelling evidence that the Bureau Staffing Project was a sham,” id. at 64. Non-TVS applicants consistently received preferential treatment over TVS applicants, who were union members. See id. at 62. Hiring managers interviewed non-TVS applicants who either had been deemed unqualified by recruiters or had never been screened at all. Board Op., 361 NLRB No. 47, at 19. Non-TVS employees who had never even applied for positions were interviewed and hired. ALJ Op., 361 NLRB No. 47, at 62. An entire category of nonunion engineers who had previously worked as CNN satellite truck operators received jobs without undergoing the “behavioral interviewing process” that CNN touted as objective and impartial. Board Op., 361 NLRB No. 47, at 19 n.38; see ALJ Op., 361 NLRB No. 47, at 65-66. Hiring managers rarely, if ever, consulted with CNN supervisors who were familiar with the work of TVS camera operators, but when they did, “they ignored favorable assessments they received.” Board Op., 361 NLRB No. 47, at 19. At the same time, they routinely solicited favorable assessments of non-TVS applicants. Id.; see ALJ Op., 361 NLRB No. 47, at 72. In light of this and similar evidence, the Board reasonably concluded that “CNN’s hiring managers’ inconsistent application of their ostensibly objective guidelines 19 of ‘behavioral interviewing’ evinced discriminatory motivation.” Board Op., 361 NLRB No. 47, at 20.10 There was also considerable evidence of “disparate treatment in favor of non-TVS applicants with little experience.” ALJ Op., 361 NLRB No. 47, at 62. This was sometimes done under the guise of the so-called “growth candidate” program, where non-TVS applicants, “many lacking in the skills necessary for their positions, were often hired over much higherrated TVS employees.” Board Op., 361 NLRB No. 47, at 20. The Board found that “CNN’s emphasis on growth candidates [was] a poorly concealed effort to refuse to hire TVS employees.” Id.; see ALJ Op., 361 NLRB No. 47, at 75 (“[T]he designation of ‘growth candidates’ was a device by which to avoid hiring too many TVS bargaining unit members . . . .”).11 10 In further support of his characterization of the Bureau Staffing Project as a sham, the ALJ noted that “CNN conducted a secret hiring process apart from the [Bureau Staffing Project] that none of its witnesses mentioned when testifying,” in which CNN hired individuals “who did not participate in the [Bureau Staffing Project] process.” ALJ Op., 361 NLRB No. 47, at 66-67. The ALJ also credited the testimony of an independent human-resources expert CNN employed to coordinate the Washington hiring, who testified that she noticed red flags throughout CNN’s hiring process: for example, supervisors inexplicably altered spreadsheets calculating applicant scores and turned a blind eye to negative professional references for non-TVS applicants. Board Op., 361 NLRB No. 47, at 20; see ALJ Op., 361 NLRB No. 47, at 64 n.42 (noting that “[t]he record also shows that applicants’ interview scores were sometimes changed for unexplained reasons”). 11 See also ALJ Op., 361 NLRB No. 47, at 64 (“In several cases, high-level CNN officials directed the hiring of inexperienced applicants over much more experienced, qualified [TVS] unit employees.”); id. at 68 (“[T]here appears to be no correlation between 20 But the evidence upon which the Board relied was not merely circumstantial. In addition to the above, the Board pointed to statements by four CNN supervisors that provided direct evidence of the employer’s overt, anti-union bias. See Waterbury Hotel Mgmt., LLC, 314 F.3d at 652 (concluding that it was proper to infer from supervisor statements that “hiring decisions were motivated by anti-union animus”); see also W & M Props. of Conn., Inc. v. NLRB, 514 F.3d 1341, 1349 (D.C. Cir. 2008) (same). First, the photojournalist manager at the New York bureau, Jeff Kinney, told a former TVS cameraman “that because of his prior relationship with [TVS] and the Union, CNN was not going to be able to offer him freelance work.” ALJ Op., 361 NLRB No. 47, at 61. Second, New York operations manager Lou Strauss made clear that CNN had no intention of recognizing a union, confirming that employees could safely assume that the union “won’t be back at CNN.” Board Op., 361 NLRB No. 47, at 19 n.37; see ALJ Op., 361 NLRB No. 47, at 61. Third, Washington executive producer Danielle Whelton emphasized to a TVS cameraman that there would be “no union” at the Washington bureau following the termination of the Service Agreements. Board Op., 361 NLRB No. 47, at 19 n.37; see ALJ Op., 361 NLRB No. 47, at 59. Finally, on the day CNN publicly announced the termination of the Service Agreements, New York bureau chief an applicant’s interview scores and their ranking at the debriefing sessions.”); id. at 99 (“It is virtually impossible in certain cases to discern any relationship between CNN’s hiring decisions, the scores applicants received during their interviews and the entries [CNN hiring managers made] on the butcher blocks.”). 21 Karen Curry called a meeting to address employees. She told them that CNN had opted to terminate the Service Agreements because TVS “came with rules and regulations,” and that by “getting rid of [TVS], [CNN] can have more control of the technical people.” Board Op., 361 NLRB No. 47, at 19 n.37. Interpreting this statement in the context provided by an employee who attended the meeting, the ALJ found that Curry “was communicating at least implicitly an intention to get rid of [the union]” and “impart[ing] a coercive message to CNN employees . . . that CNN in general will not tolerate a union.” ALJ Op., 361 NLRB No. 47, at 59.12 Together, CNN’s manipulation of job titles and qualifications, its frequent deviations from its vaunted Bureau Staffing Project and “behavioral interviewing” protocols to favor non-TVS applicants, and anti-union statements by CNN supervisors, were more than sufficient evidence of anti-union animus to satisfy the General Counsel’s threshold burden. Added to this, as further evidence of animus, was the pretextual nature of the justifications CNN gave for its hiring decisions, which we discuss in Part III.C below. B To resist the conclusion that it was motivated by anti-union animus, CNN advances the following arguments. 1. CNN maintains that the Board improperly cited CNN’s termination of the Service Agreements with TVS as evidence of discriminatory hiring. Its argument amounts to the following syllogism: CNN was not a joint employer. Because it was not 12 See also ALJ Op., 361 NLRB No. 47, at 60 (finding further evidence of animus in Washington bureau chief Kross’ statement that “[the Union] would not be a part of CNN after [December 5]”). 22 a joint employer, its decision to terminate the Service Agreements was legal. And because the decision to terminate was legal, that decision cannot evidence animus. The conclusion of CNN’s syllogism, however, does not necessarily follow from its premises. A decision that by itself does not flout the law can still constitute evidence of an employer’s underlying anti-union animus. Here, the Board did not point to the fact of the termination alone, but rather to “evidence [that] CNN’s claim that it brought the work in-house in order to keep up with technological change was . . . pretextually false.” Board Op., 361 NLRB No. 47, at 19. And as the Board said, there is no dispute that “evidence of pretext may be used to show discriminatory motivation.” Id. (citing Lucky Cab Co., 360 NLRB No. 43, at 4-5 (2014)); see Fort Dearborn Co., 827 F.3d at 1075; Pioneer Hotel, Inc. v. NLRB, 182 F.3d 939, 947-48 (D.C. Cir. 1999). If the NLRB could reasonably find (and it could) that a “major motive in [the decision to terminate the contracts] was CNN’s desire to operate its Washington and New York bureaus without a union,” ALJ Op., 361 NLRB No. 47, at 51, it could also reasonably conclude that this same desire carried over into CNN’s subsequent hiring decisions for those bureaus, see id. at 58 (concluding that “the decision to embark upon the Bureau Staffing Project was part of an overall plan motivated by antiunion animus”). Although CNN’s anti-union animus may not have made the termination decision itself unlawful (because the Board erred in its joint-employer analysis), there was no reason to believe that its animus had dissipated by the time the new hiring began. But even if animus associated with terminating the Service Agreements could not be the basis for finding discrimination in hiring, that would not undermine the Board’s bottom-line 23 conclusion that anti-union animus pervaded CNN’s hiring process. The Board found that “[t]he evidence of animus in this case [was] overwhelming.” Board Op., 361 NLRB No. 47, at 18. Although it listed termination of the Service Agreements as one piece of that evidence, it also pointed to the numerous other pieces of evidence detailed in Part III.A above, including: the CNN supervisors’ anti-union statements, the deliberate modification of position qualifications to “minimiz[e] the significance of the [TVS employees’] prior experience when they applied for the ‘new’ jobs,” the “numerous instances of interviewing / debriefing / hiring disparities that adversely affected TVS applicants,” and the hiring of “growth candidates as a poorly concealed effort to refuse to hire TVS employees.” Id. at 19-20. The Board never suggested that the termination of the Service Agreements was necessary to its animus finding or that subtraction of that factor would reduce the “overwhelming” evidence to a level below the necessary threshold. To the contrary, the Board stated that the “principal evidence of [CNN’s] unlawful discrimination” was the manner in which it staffed the D.C. and New York bureaus, including the position reclassifications and actual hiring practices. Id. at 19 (emphasis added). “[W]hen an agency relies on multiple grounds for its decision, some of which are invalid, we may nonetheless sustain the decision as long as one is valid and the agency would clearly have acted on that ground even if the other were unavailable.” Bally’s Park Place, Inc., 646 F.3d at 939 (quoting Casino Airlines, Inc. v. Nat’l Transp. Safety Bd., 439 F.3d 715, 717 (D.C. Cir. 2006) (internal quotation marks omitted)). Whatever weight the Board accorded to CNN’s termination of the Service Agreements, that factor clearly “was not necessary to its 24 decision.” Id. We therefore sustain the Board’s conclusion that CNN’s hiring process proceeded with discriminatory animus.13 2. CNN also argues that it could not have discriminated against union members because it hired a majority of the TVS employees who had worked at each bureau. Although superficially appealing, this theory lacks support in either our case law or Board precedent. To the contrary, in rejecting an identical argument, we held that where the record contained evidence of animus but the employer hired a majority of union members, “the more reasonable inference is that the Employer’s discriminatory design ultimately failed, not that it wasn’t tried.” Great Lakes Chem. Corp. v. NLRB, 967 F.2d 624, 628 (D.C. Cir. 1992). Here, too, it was reasonable for the Board to infer that CNN planned to hire a sufficient number of former TVS employees to lend it an air of impartiality, while avoiding the number that would impose a bargaining obligation. Specifically, the Board noted that, although CNN heralded the fact that TVS employees constituted a majority of the employees it hired for the historical TVS bargaining units, CNN believed that the historical units were inappropriate for its new employees. Instead, CNN assumed that the NLRB would accept a much broader “wall-to -wall” unit composed of all production staff. Board Op., 361 NLRB No. 47, at 18 n.36. The Board found that CNN 13 Our conclusion that CNN’s termination of the Service Agreements was not necessary to the Board’s decision is further supported by our examination of the ALJ decision, with which the Board agreed. See Board Op., 361 NLRB No. 47, at 1 & n.1. The ALJ listed a host of reasons for concluding that the Bureau Staffing Project “was discriminatorily motivated,” but did not expressly include termination of the Service Agreements among them. ALJ Op., 361 NLRB No. 47, at 62. 25 endeavored to ensure that former TVS employees would not make up a majority of that broader unit, and that this deliberate effort to avoid hiring a union majority ran afoul of § 8(a)(3). Id.; see Great Lakes Chemical Corp., 967 F.2d at 628; see also U.S. Marine Corp., 293 NLRB 669, 669-73 (1989) (finding that an employer who hired a majority of union members nonetheless acted with animus).14 In a variation of the above argument, CNN points to its expert’s testimony that it was “actually biased in favor of TVS unit employees” because those employees were four times more likely to receive job offers than other applicants. ALJ Op., 361 NLRB No. 47, at 121 n.190. The ALJ, however, rejected this “startling conclusion” because the expert failed to account for: (1) the ways in which CNN deviated from its normal hiring process to accommodate non-TVS applicants; (2) “the possibility that TVS applicants were better qualified than nonTVS applicants because they had been doing the jobs for which they were applying for years”; and (3) “the fact that almost 100 percent of the [nonunion] CNN incumbents who were subjected to the [Bureau Staffing Project] kept their jobs.” Id. The expert’s analytical errors -- including her failure to confront some of the most compelling evidence of CNN’s bias against TVS employees -- could certainly cause a “reasonable factfinder” to doubt the validity of her conclusion. OzburnHessey Logistics, LLC v. NLRB, 833 F.3d 210, 217 (D.C. Cir. 2016) (quoting Bally’s Park Place, Inc., 646 F.3d at 935). 14 The historical bargaining unit at each TVS bureau consisted of field camera, field audio, engineering, and studio technical employees. ALJ Op., 361 NLRB No. 47, at 49-50. The Board ultimately concluded that the historical units, not CNN’s proposed “wall-to-wall” unit, were appropriate. See supra note 7. 26 In any event, even if CNN’s hiring of union members (or even union leaders) weighed against a finding of discrimination, the ALJ and Board reasonably found that the other, “overwhelming evidence” of discrimination “outweighed” the hiring numbers. ALJ Op., 361 NLRB No. 47, at 62. Because the Board fairly regarded the evidence of animus as particularly strong, it acted within its discretion in determining that such evidence outweighed the claim of CNN’s expert. See Alden Leeds, Inc. v. NLRB, 812 F.3d 159, 166 (D.C. Cir. 2016) (“Although [the employer] argues that the record contains evidence that is contrary to the Board’s findings and supports its position, ‘[t]he question before us is not whether substantial evidence supports the [employer’s] view, but whether it supports the Board’s.’” (quoting Wayneview Care Ctr. v. NLRB, 664 F.3d 341, 352 (D.C. Cir. 2011))). 3. CNN further claims that, in focusing on its hiring of applicants who lacked the technical expertise of TVS applicants, the Board disrespected “CNN’s business judgment to value [photo]journalistic skills and potential over traditional (and increasingly irrelevant) technical background.” Reply Br. 18. That claim is unfair. The ALJ did not dispute CNN’s business judgment. He simply found no evidence to support CNN’s contention that non-TVS applicants possessed greater journalistic skills than TVS applicants. There was, he said, an “[a]bsence of credible evidence regarding the hiring decisions made for photojournalists.” ALJ Op., 361 NLRB No. 47, at 68.15 15 See ALJ Op., 361 NLRB No. 47, at 69 (“[T]he deliberations [about photojournalists] . . . are poorly documented. CNN’s witnesses generally recall very little of what occurred, and their testimony is often inconsistent.”); id. (“It is . . . very unclear what actually transpired . . . [in the hiring of photojournalists] and on what basis the final decisions were made.”); id. at 88 (“[E]xactly when, how and on what basis decisions were made with regard to the hiring of New York 27 Nor was there evidence that CNN’s new “photojournalists” had materially different responsibilities than TVS’ “field technicians.” See id. at 112. To the contrary, “[m]ost employees continued to spend most of the day performing the same tasks and using the same skills they had used in their work for [TVS].” Id. at 108.16 4. Finally, CNN maintains that, even if anti-union animus infected specific hiring decisions, the NLRB failed to find that “CNN had systematically manipulated or disregarded its neutral hiring policy,” or even “that the [discriminatory] disparities were widespread.” CNN Br. 60. Although it is true that the NLRB did not use the phrase “systematically manipulated,” it used other formulations to the same effect. See, e.g., ALJ Op., 361 NLRB No. 47, at 55 (“I find that the entire Bureau Staffing Project was discriminatorily motivated . . . .”); id. at 75-76 (finding that “CNN manipulated its hiring decisions to obtain . . . a mix of TVS and non-TVS applicants that in conjunction with CNN’s plan to pack the bargaining unit, would allow it to decline to recognize [the union]”). And CNN’s claim notwithstanding, the NLRB did in fact adopt the ALJ’s finding that the discriminatory disparities were “widespread.” Id. at 50 (“During the [Bureau Staffing Project], CNN engaged in widespread and blatant discrimination against [TVS] bargaining photojournalists remains a mystery.”). 16 See ALJ Op., 361 NLRB No. 47, at 109 (“[W]ith regard to the photojournalists, the evidence shows that for at least 6 months after the end of the [TVS] contract, all photojournalists were performing the same work as TVS field technicians for the vast majority of the workday. The same is true for most photojournalists even after the first 6 months.” (internal citation omitted)). 28 unit members. CNN did so with the objective of depriving employees of [union] representation.”).17 CNN insists that the NLRB should have “sought to quantify” the hiring disparities it found. CNN Br. 60. But we have affirmed Board findings of generalized animus in hiring even where the Board did not provide a statistical summary of the employer’s anti-union practices. See Great Lakes Chemical Corp., 967 F.2d at 628. Simply put, where there is “broadly damning . . . evidence that the Employer acted out of general animus toward the Union” -- whether qualitative or quantitative -- the Board can presume that “any [union] employee not hired was turned away because of his union adherence.” Id. For the reasons discussed above, this case fits the mold.18 *** 17 See also Board Op., 361 NLRB No. 47, at 19 (“As the record . . . shows, . . . there were numerous instances of interviewing / debriefing / hiring disparities that adversely affected TVS applicants. In every job category, . . . hiring managers ignored ostensibly governing protocols intended to ensure the objectivity of the behavioral interviewing process.”). 18 In any event, the NLRB did in fact quantify the hiring disparities it found. The ALJ listed the exact number of additional TVS employees CNN would have hired absent its anti-union discrimination: 46 additional TVS employees in the Washington bureau and 63 additional TVS employees in the New York bureau. ALJ Op., 361 NLRB No. 47, at 102. The ALJ arrived at these numbers by listing the name of every non-TVS employee who received a job based on preferential treatment at the expense of a more qualified TVS employee, and explaining the nature of that preferential treatment. Id. Thus, CNN’s claim that the NLRB “steadfast[ly] refus[ed]” to quantify CNN’s hiring irregularities, Reply Br. 24, is simply incorrect. 29 In sum, despite CNN’s arguments, we remain convinced that substantial evidence supports the Board’s finding that antiunion animus was a motivating factor in CNN’s hiring decisions, thus satisfying the first step of the Wright Line test. C “Where, as here, the General Counsel makes a strong showing of discriminatory motivation, the employer’s rebuttal burden is substantial.” Bally’s Park Place, Inc., 646 F.3d at 936. To satisfy that Wright Line burden, CNN offered two neutral justifications for its hiring decisions that assertedly showed it would have made the same decisions regardless of any animus. The Board, however, reasonably found that those justifications were pretextual. First, CNN contended that it needed to overhaul its workforce in order to adapt to technological changes in the television-news industry. But this explanation is belied by the fact that CNN’s operations did not materially change when the Service Agreements ended. In representative testimony, one former TVS employee hired by CNN testified that “everything in her new job was the same as her TVS job.” Board Op., 361 NLRB No. 47, at 15; see also, e.g., id. at 16; ALJ Op., 361 NLRB No. 47, at 112. And the ALJ found that, “[w]hile CNN employees performing what was bargaining unit work may use some newer equipment and may have been given some additional duties, the work they performed was essentially the same as the work they performed for [TVS].” ALJ Op., 361 NLRB No. 47, at 108. Moreover, to the extent that some changes occurred, “[t]here [wa]s no evidence that any [TVS] employee . . . could not have adapted to the technological changes that CNN was undertaking.” Id. at 51. To the contrary, the TVS employees 30 whom CNN hired generally showed more technological proficiency than the non-TVS hires. See id. at 63. As the Board noted, the TVS employees “had lived through substantial technological changes, . . . [and] CNN never terminated or directed the termination of any TVS unit employee for failing to keep up with those changes or inability to perform the work.” Board Op., 361 NLRB No. 47, at 19. Indeed, “when it terminated the TVS contracts, . . . CNN personnel went out of their way to praise the abilities of the two bargaining unit work forces.” Id. In the face of this evidence, it was hardly unreasonable for the Board to conclude that replacing veteran technicians with a cohort of inexperienced and underqualified newcomers was an unlikely strategy for keeping up with a world of increasing technological complexity. Second, CNN contended that it based its hiring decisions on the neutral “behavioral interviewing” process that it adopted at the outset of the Bureau Staffing Project. But even assuming this process were facially neutral, we have detailed in Part III.A how frequently CNN deviated from it to favor non-TVS applicants. Where, as here, an employer “applied its neutral criteria inconsistently,” it is reasonable for the Board to conclude “that the hiring criteria served as little more than pretext for weeding out . . . union employees.” Waterbury Hotel Mgmt., LLC, 314 F.3d at 653; see also Board Op., 361 NLRB No. 47, at 20.19 The Board found that “CNN’s reasons for failing to hire the TVS technicians were all pretextual, and that it . . . therefore failed to establish that it would not have hired the technicians 19 See also W & M Props. of Conn., Inc., 514 F.3d at 1349 (affirming Board finding that an employer failed to satisfy its Wright Line rebuttal burden when it deviated from its alleged “neutral and objective hiring criteria,” thereby rendering those criteria “illusory”). 31 absent its union animus.” Board Op., 361 NLRB No. 47, at 21. Because substantial evidence underlies that finding, and because “continuity of the business enterprise and the work force was established,” id., we sustain the Board’s finding that CNN was a successor employer and violated § 8(a)(5) by refusing to bargain with the union. See Waterbury Hotel Mgmt., LLC, 314 F.3d at 653-55; Capital Cleaning Contractors, Inc., 147 F.3d at 1007. IV The Board found that CNN’s discriminatory hiring cost the company in another respect as well. Such discrimination not only caused CNN to become a successor employer and therefore liable for failing to bargain. It also amounted to an independent violation of NLRA § 8(a)(3). Section 8(a)(3) makes it an unfair labor practice for an employer to discriminate in hiring in order to “discourage membership in any labor organization.” 29 U.S.C. § 158(a)(3). The Board applies the same Wright Line framework in determining whether an employer violated § 8(a)(3) through discriminatory hiring as it applies in determining whether an employer became a successor employer through discriminatory hiring. See Waterbury Hotel Mgmt., LLC, 314 F.3d at 651-53, 655 (sustaining finding that an employer violated § 8(a)(3) because its “hiring decisions were motivated by anti-union animus,” and that it became a successor employer for the same reason); Capital Cleaning Contractors, Inc., 147 F.3d at 100507 (same). Accordingly, because we have already sustained the Board’s successorship finding on the basis of discriminatory hiring, see supra Part III, we likewise sustain its finding that CNN violated § 8(a)(3) by discriminating in hiring. 32 V Finally, the Board found that CNN violated NLRA § 8(a)(1) through the no-union statements of supervisors Curry, Strauss, Whelton, and Kinney, the content of which we have already set forth above. See supra Part III.A. Section 8(a)(1) makes it an unfair labor practice to “interfere with, restrain, or coerce employees in the exercise of” their § 7 rights. 29 U.S.C. § 158(a)(1).20 An employer’s statement violates § 8(a)(1) if, “considering the totality of the circumstances, the statement has a reasonable tendency to coerce or to interfere with those rights.” Tasty Baking Co., 254 F.3d at 124. “Ordinarily, an employer’s statement that it will not have a union at its plant does not violate section 8(a)(1).” Williams Enters., Inc. v. NLRB, 956 F.2d 1226, 1234 (D.C. Cir. 1992).21 “A successor employer’s statement that it will not have a union at its plant, however, does violate section 8(a)(1).” Id. Before conducting its hiring process, a successor employer “does not know whether it will have a duty to recognize and bargain with the predecessor’s union because it does not know whether it will hire a majority of the predecessor’s employees.” Id. “Therefore, any statement that it will be nonunion ‘indicates to the applicants that [it] intends to discriminate against the [predecessor’s] employees to ensure its nonunion status.’” Id. 20 NLRA § 7 provides that “[e]mployees shall have the right to self-organization, to form, join, or assist labor organizations, to bargain collectively through representatives of their own choosing, and to engage in other concerted activities for the purpose of collective bargaining or other mutual aid or protection.” 29 U.S.C. § 157. 21 Such a statement, however, can provide evidence of a § 8(a)(3) violation for discriminatory hiring. See supra Part IV. 33 (quoting Kessel Food Markets, Inc., 287 NLRB 426, 429 (1987), enforced, 868 F.2d 881 (6th Cir. 1989)). Three of the four supervisor statements -- Curry’s, Strauss’s, and Whelton’s -- were made before CNN completed its Bureau Staffing Project hiring. The Board found that each of those statements gave employees the message, either explicitly or implicitly, that CNN’s new workforce would not have union representation. Board Op., 361 NLRB No. 47, at 22-23. Yet, as a successor employer that had not finished its hiring process, CNN could not have legitimately known that fact. It was therefore reasonable for the Board to conclude that the statements signaled CNN’s intent to discriminate against union members in its hiring. And that, in turn, was sufficient to support the conclusion that the statements had a reasonable tendency to coerce employees, thereby violating § 8(a)(1). See Williams Enters., Inc., 956 F.2d at 1234. But what about Kinney’s statement, which he made after CNN completed its Bureau Staffing Project hiring process? For a successor employer’s no-union statement to violate § 8(a)(1), the employer’s hiring process must either be upcoming or ongoing. See id. If the process has been completed, then the employer could know with reasonable certainty that its workforce would lack union representation, so a no-union statement would not generally suggest coercion. See id. Although it is true that CNN had finished hiring full-time employees through the Bureau Staffing Project when Kinney made his no-union statement to TVS employee Jonathan Smith, the record shows that CNN was still hiring freelance cameramen at that time. See ALJ Op., 361 NLRB No. 47, at 61. In February 2004, after CNN had completed the Bureau Staffing Project, Smith called Kinney about obtaining freelance camera work. Id. Kinney told Smith that “CNN was hiring cameramen 34 who owned their own gear.” Id. Smith responded that he had his own gear, and then asked Kinney “if his membership in the Union was a problem.” Id. Kinney replied, “[t]hat’s good to know,” and told Smith that “he would have to check with ‘higher-ups’” before extending Smith an offer, but that Smith would “be good to have around because of his maturity.” Id. Three weeks later, “Kinney informed Smith that because of his prior relationship with [TVS] and the Union, CNN was not going to be able to offer him freelance work.” Id. Thus, because the hiring process for freelance cameramen was ongoing, Kinney’s statement to a freelance applicant that CNN would not hire him due to his union membership constituted a fourth violation of § 8(a)(1). VI This brings us to CNN’s challenge to the Board’s remedy. NLRA § 10(c) authorizes the Board, upon finding an unfair labor practice, “to take such affirmative action including reinstatement of employees with or without back pay, as will effectuate the policies of [the Act].” 29 U.S.C. § 160(c). While the Board has “broad discretionary power under this section to fashion remedies,” Petrochem Insulation, Inc. v. NLRB, 240 F.3d 26, 34 (D.C. Cir. 2001), its orders must be “truly remedial and not punitive” because the Board lacks statutory authority to punish employers for their bad acts, Capital Cleaning Contractors, Inc., 147 F.3d at 1009. Nor may a remedial order impose “an undue burden” on an employer by requiring “a substantial outlay of new capital or otherwise caus[ing] undue financial hardship.” Regal Cinemas, Inc. v. NLRB, 317 F.3d 300, 315 (D.C. Cir. 2003) (internal quotation marks omitted); see Fibreboard Paper Prods. Corp. v. NLRB, 379 U.S. 203, 216 (1964). 35 CNN takes issue with three of the affirmative obligations the Board imposed in its remedial order: (1) that CNN provide backpay and benefits to all TVS employees, both those who lost their jobs and those who were hired at reduced pay, at the wage rate provided in the TVS collective-bargaining agreements; (2) that CNN reinstate and provide any necessary training to all TVS workers who were discharged and not hired by CNN; and (3) that CNN recognize and bargain with the union as the exclusive representative of the unit employees. We analyze these claims one at a time, as each presents its own unique set of issues. A In Capital Cleaning Contractors, Inc. v. NLRB, this court confronted a Board-ordered backpay remedy materially identical to that imposed here. Like CNN, the employer in that case took over a predecessor’s business and discriminatorily refused to hire its predecessor’s union employees. Capital Cleaning Contractors, Inc., 147 F.3d at 1005-07. As here, we affirmed the Board’s findings that the successor employer violated § 8(a)(1), (3), and (5). Id. at 1007. To remedy those violations, the Board ordered the employer to provide backpay to all injured employees “at the rate set by the [collective-bargaining agreement] between [the predecessor employer and the union] for the entire period from the violation . . . until such future time as [the successor employer] reaches a new agreement or an impasse with [the union].” Id. at 1010. The court concluded that the Board’s backpay remedy exceeded its authority because, rather than “restore the situation ‘as nearly as possible, to that which would have obtained but for the illegal discrimination,’” id. at 1009 (quoting Sure-Tan, Inc. v. NLRB, 467 U.S. 883, 900 (1984)), the remedy “[was] punitive,” id. at 1010. “By engaging in anti-union 36 discrimination,” the court held, “the successor loses only the right to set initial terms without first bargaining with the union; it does not lose the right to take an initial bargaining position with the union and to bargain hard from that point.” Id. at 1011. Based on this reasoning, the court announced the following limitation on backpay in successor-employer cases: The Board may require backpay based on the terms of the predecessor’s collective-bargaining agreement “only for ‘a period allowing for a reasonable time of bargaining.’” Id. (quoting Kallmann v. NLRB, 640 F.2d 1094, 1103 (9th Cir. 1981)). Thereafter, backpay must be based on the wages the successor employer “actually paid the new employees who did the work previously done by the [predecessor’s] employees,” unless the Board can show that the successor “would have agreed in negotiations with the Union to pay a higher rate than it had to pay for alternative labor.” Id. In the case now before us, the Board ordered CNN to provide backpay and benefits, under the terms of the TVS collective-bargaining agreement, to all TVS employees for the entire period since the termination of the TVS contracts. That remedy does not comply with the holding of Capital Cleaning. Nor does the Board’s brief dispute the point. Instead, it argues that, while the Capital Cleaning limitation applies to successor employers, it does not apply to joint employers. But whether or not that is correct, it cannot support the remedy in this case because we have vacated the Board’s finding that CNN was a joint employer. Capital Cleaning therefore binds us, and we remand for the Board to limit its backpay remedy in accordance with that precedent. 37 B CNN next challenges the Board’s order that CNN reinstate the 114 former TVS employees whom it did not hire and provide those employees with training. CNN contends that this remedy imposes an “undue and unfair burden” in light of changes at CNN since 2003, CNN Br. 68, and “has grave First Amendment implications,” id. at 71. CNN’s attacks on the Board’s reinstatement remedy are premature. In denying CNN’s motion for reconsideration, the Board explained that it was deferring resolution of the challenge to the reinstatement order to “the compliance phase of this proceeding.” CNN America, Inc., 362 NLRB No. 38, at 1. This court has consistently declined to consider challenges to remedial orders when the Board has “reserve[d] the issue for later consideration.” Scepter, Inc. v. NLRB, 448 F.3d 388, 391 (D.C. Cir. 2006); accord E.I. Du Pont de Nemours & Co. v. NLRB, 489 F.3d 1310, 1317 (D.C. Cir. 2007); Ark Las Vegas Restaurant Corp. v. NLRB, 334 F.3d 99, 107 (D.C. Cir. 2003). There is no reason to depart from that practice here. Citing Great Lakes Chemical Corp. v. NLRB, CNN contends that, despite the forthcoming compliance proceedings, “it is appropriate at this stage for the Court to review the burdens that any wide-scale rehiring would impose.” CNN Br. 68 n.9. But Great Lakes actually stands for the opposite conclusion. There, after the Board had said it would consider the employer’s challenge to a reinstatement order in compliance proceedings, we “reject[ed] as premature [the employer’s] present objection to the apparent breadth of the Board’s order.” Great Lakes Chemical Corp., 967 F.2d at 630. In holding that CNN’s challenge is premature, we express no view on its merits. If the Board retains a reinstatement order 38 after compliance proceedings, CNN will have the opportunity to present its arguments in a petition for review of that order. See id. C Finally, CNN challenges the Board’s order to recognize and bargain with the union. The bargaining order was impermissible, CNN maintains, because the Board failed to follow this Circuit’s requirement that such an order be accompanied by an explanation. We agree. “This court repeatedly has reminded the Board that an affirmative bargaining order is an extreme remedy that must be justified by a reasoned analysis that includes an explicit balancing of three considerations: (1) the employees’ § 7 rights [of self-organization and collective bargaining]; (2) whether other purposes of the Act override the rights of employees to choose their bargaining representatives; and (3) whether alternative remedies are adequate to remedy the violations of the Act.” Vincent Indus. Plastics, Inc., 209 F.3d at 738. “We have repeatedly held that if the Board wishes to impose an affirmative bargaining order, it must explain why that remedy is appropriate given the facts of that particular case.” Lee Lumber & Bldg. Material Corp. v. NLRB, 117 F.3d 1454, 1461 (D.C. Cir. 1997).22 This requirement applies with full force in the context 22 See, e.g., Scomas of Sausalito, LLC v. NLRB, 849 F.3d 1147, 1156 (D.C. Cir. 2017); Cogburn Health Ctr., Inc. v. NLRB, 437 F.3d 1266, 1273-75 (D.C. Cir. 2006); Douglas Foods Corp. v. NLRB, 251 F.3d 1056, 1065-67 (D.C. Cir. 2001); Flamingo Hilton-Laughlin v. NLRB, 148 F.3d 1166, 1170-73 (D.C. Cir. 1998); Exxell/Atmos, Inc. v. NLRB, 28 F.3d 1243, 1248-49 (D.C. Cir. 1994); Caterair Int’l v. NLRB, 22 F.3d 1114, 1122-23 (D.C. Cir. 1994); Peoples Gas Sys., Inc. v. NLRB, 629 F.2d 35, 46 (D.C. Cir. 1980). 39 of successor employers and incumbent unions. See Sullivan Indus. v. NLRB, 957 F.2d 890, 903 (D.C. Cir. 1992) (refusing to enforce the Board’s order requiring a successor employer to bargain with a union because “the Board ha[d] not explained why an affirmative bargaining order -- with its corresponding decertification bar -- [was] the appropriate remedy in th[at] case”). For its part, the Board has long disagreed with our Circuit’s requirement, see supra note 22, although on occasion its decisions have nonetheless combined an expression of disagreement with sufficient explanation to satisfy the requirement, see, e.g., In re Marion Hosp. Corp., 335 NLRB 1016, 1019-20 (2001), enforced, 321 F.3d 1178 (D.C. Cir. 2003). It did not do so here. Instead, it provided no explanation whatsoever for the bargaining order it imposed. We are thus left with no choice but to grant CNN’s petition on this point and remand for the Board either to vacate the order or provide sufficient justification. See Lee Lumber & Bldg. Material Corp., 117 F.3d at 1462 (“Case law in our circuit is as clear as it could be on this question. The Board, however, continues to ignore us. We continue to reverse.”). VII We deny the Board’s application for enforcement and grant CNN’s cross-petition with respect to the violations that flow from the Board’s finding that CNN was a joint employer because the Board failed to explain its departure from relevant Board precedents. We do likewise with respect to those elements of the Board’s remedial order that depart from this Circuit’s precedents. In all other respects, we grant the Board’s application and deny CNN’s cross-petition for review. So ordered. KAVANAUGH, Circuit Judge, concurring in part and dissenting in part: The majority opinion concludes that the Board erred in its analysis of the joint-employer issue. By contrast, the majority opinion upholds the Board’s analysis on the successor-employer issue. In my view, the Board erred in its analysis of both the joint-employer and the successoremployer issues. I therefore respectfully concur in part and dissent in part. * * * CNN contracted with TVS for certain technical services related to CNN’s television operations. CNN then decided to terminate its contracts with TVS and bring those services inhouse. In staffing up, CNN hired about 100 former TVS employees, most of whom were union members. But CNN did not consider itself a “successor employer” to TVS and did not bargain with the union that represented the newly hired CNN employees. (In the past, TVS had negotiated with the union representing those employees in the predecessor units at TVS.) There are two key issues in this case: (i) Were CNN and TVS joint employers of the TVS employees such that CNN could not lawfully terminate the TVS contracts without bargaining with the TVS employees’ union? (ii) Was CNN a successor employer to TVS because of CNN’s supposed discrimination against former TVS employees when CNN hired for its new in-house positions? The Board said yes to both questions. The majority opinion concludes that the Board’s first conclusion is not sustainable on this record, but the majority opinion upholds the Board’s second conclusion. I do not think that either of the Board’s conclusions is sustainable on this record. First, as the majority opinion persuasively explains, the Board erroneously concluded that CNN and TVS were joint 2 employers of the TVS employees. The Board did not analyze the CNN-TVS relationship under the Board’s “direct and immediate control” test. The Board has long used that test for determining joint-employer status. And that test was in place at the time the Board issued its decision in this case. The Board’s failure to apply that test (or to reasonably explain why it was not doing so) flouts basic principles of administrative law, as the majority opinion rightly concludes. Moreover, in my view, there is no plausible way to apply the “direct and immediate control” test and conclude that CNN and TVS were joint employers. Second, for several independent reasons, I also do not believe we can sustain the Board’s successor-employer conclusion on this record. (I therefore respectfully disagree with the majority opinion on that issue.) Under Board law, there are two relevant ways in which an employer may become a successor to the predecessor employer. First, a new employer is considered a successor employer when, among other things, a majority of the new employer’s work force in an appropriate unit consists of former employees of the predecessor employer. In that circumstance, a successor employer must bargain with the union but ordinarily is not bound by the prior collectivebargaining agreement. Second, a new employer may be considered a successor employer when it discriminates in hiring against the prior employer’s union-member employees. In that circumstance, the new employer may be bound by the prior collective-bargaining agreement, which may in turn require back-pay remedies. Here, in reaching its conclusion that CNN was a successor employer to TVS, the Board adopted the second approach. The Board reasoned that CNN had discriminated 3 against union members (the former TVS employees) in hiring for CNN’s new in-house positions. In reaching that discrimination conclusion, the Board relied in part on CNN’s purportedly unlawful termination of its contracts with TVS. The Board’s conclusion that CNN’s termination of its contracts with TVS was unlawful in turn rested on the Board’s conclusion that CNN and TVS were joint employers. As noted above, however, CNN and TVS were not joint employers under the “direct and immediate control” test, and we are remanding that issue back to the Board. It follows that we should also remand the successor-employer issue. The Board’s jaundiced view of CNN’s termination of the TVS contracts clearly infected the Board’s view of CNN’s subsequent hiring decisions with respect to the former TVS employees. If the Board on remand were to find that the termination was lawful, the Board would then have to assess whether CNN’s lawful termination of the TVS contracts could somehow still be considered evidence of CNN’s discrimination against the former TVS employees in its hiring decisions. The Board has not yet conducted that analysis. Therefore, I believe we must remand the successor-employer issue. The majority opinion disagrees, in part because it does not believe that the Board’s finding that CNN unlawfully terminated its contracts with TVS was necessary to the Board’s successor-employer conclusion. What we know on this record, however, is that the Board in fact relied on CNN’s purportedly unlawful termination of its TVS contracts as one basis for its conclusion that CNN was a successor employer. I do not think we can airbrush that part of the Board’s analysis out of the picture. See SEC v. Chenery Corp., 318 U.S. 80, 87-88 (1943). 4 Moreover, even if I am wrong about that point and we must reach the merits of the successor-employer issue, I respectfully disagree with the majority opinion’s merits analysis. I instead agree with Member Miscimarra’s dissent from the Board’s decision. See CNN America, Inc., 361 NLRB No. 47, at 42 (2014) (Miscimarra, dissenting). On this record, I do not see substantial evidence that CNN, when making hiring decisions, discriminated against former TVS employees. The statistics suggest that CNN favored former TVS employees. Indeed, CNN hired about 100 former TVS employees. If CNN really had some nefarious plan to discriminate against TVS employees in its hiring, CNN’s purported plan was, in Member Miscimarra’s words, “an abject failure.” Id. at 31. As Member Miscimarra further explained: The Board’s “finding of unlawful motivation is undermined by CNN’s actual hiring of a work force majority consisting of former TVS employees in both Washington, D.C., and New York, requiring CNN to recognize and bargain with the Unions in both locations.” Id. at 41. In short, on this record, the Board’s conclusion that CNN was a successor employer because of CNN’s purported discrimination against union members is not supported by the record and should not be sustained. That said, I agree with Member Miscimarra’s further conclusion that CNN would qualify as a successor employer under the Board’s traditional successor-employer analysis. See Capital Cleaning Contractors, Inc. v. NLRB, 147 F.3d 999, 1005 (D.C. Cir. 1998). That analysis depends on whether a majority of employees in the new CNN units were former TVS (union) employees, and whether the historical bargaining units remained appropriate at CNN. The answer to both questions is yes. 5 That raises a natural question: If CNN is a successor employer, why does it matter which way CNN qualifies as a successor employer? Money. Lots of money. As noted above, finding CNN a successor employer under the traditional test would have dramatically different consequences in terms of the remedies available in this case. In particular, under the traditional test, CNN would be subject to an obligation to bargain with the union going forward. Under the discrimination finding, however, CNN could also be liable for tens of millions of dollars of back pay to former TVS employees. So if CNN qualifies as a successor employer only under the traditional test and not under the discrimination test, that would make a huge difference in the real world. Under my view on the merits of the successoremployer issue, which was also Member Miscimarra’s view, CNN qualifies as a successor employer only under the traditional test. I would therefore remand to the Board for it to re-determine the appropriate remedies associated with the proper successor-employer conclusion. Bottom line: In my view, the Board jumped the rails in its analysis of both the joint-employer and the successoremployer issues.