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 successor-employer issues. I therefore respectfully concur in part and dissent in part.
[[Image here]]
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 in-house. 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 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 collective-bargaining 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.
*766Here, 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 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., 818 U.S. 80, 87-88, 63 S.Ct. 454, 87 L.Ed. 626 (1943).
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 Miseimarra’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 Misci-marra’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 bar*767gaining units remained appropriate at CNN. The answer to both questions is yes.
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 successor-employer 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 redetermine 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 successor-employer issues.