The opinion
NEWMAN, Circuit Judge,
concurring:
I concur in all of the conclusions reached in Judge Mishler’s thorough opinion for the Court and in virtually all of his opinion, but add these additional views in an attempt to clarify analysis of employer motivation cases under §§ 8(a)(1) and (3) of the Act, especially in the aftermath of the Board’s significant opinion in Wright Line, A Division of Wright Line, Inc., 251 NLRB No. *42 150 (slip op. Aug. 27, 1980), [1980] Lab.L. Rep. (NLRB Dec.) (CCH) ¶ 17,356.
In Wright Line, the Board surveyed the assortment of pronouncements in prior decisions of the Board and the Courts of Appeals and concluded that the correct framework for determination of employer motivation is the model adopted by the Supreme Court in Mt. Healthy City School District Board of Education v. Doyle, 429 U.S. 274 , 97 S.Ct. 568 , 50 L.Ed.2d 471 (1977). Applying Mt. Healthy to the context of a § 8(a)(3) claim, the Board announced that the General Counsel must make “a prima facie showing sufficient to support the inference that protected conduct was a ‘motivating factor’ in the employer’s decision. Once this is established, the burden will shift to the employer to demonstrate that the same action would have taken place even in the absence of the protected conduct.” Slip op. at 20-21, [1980] Lab.L.Rep., supra, at 32,469. Thus, the Board has rejected all vestiges of the “in part” test of motivation and adopted the “but for” test, with the burden on the employer, after a prima facie case is shown, to establish that he would have taken the same action if only the valid ground had existed. If the employer fails to sustain his burden, then it can be concluded that he would not have acted “but for” the protected conduct, and only in that event is such conduct a legally significant cause of the employer’s action; if the employer satisfies his burden, then he avoids liability, even though he may in fact have relied “in part” on the protected conduct when he decided to take action.
The Board suggests in Wright Line that “an additional benefit which will result from our use of the Mt. Healthy test is that the perceived significance in distinguishing between pretext and dual motive cases will be obviated.” Slip op. at 20 n.13, [1980] Lab.L.Rep., supra, at 32,472 n.13. That suggestion bears examination in this case, because Judge Mishler’s opinion analyzes this case as one of pretext, rather than dual motive, thereby implicitly indicating that the distinction serves a useful purpose.
To see why this is so, at least to some extent, requires a clear understanding of the analytical processes involved (1) in examining whether a ground asserted by an employer is a “pretext” and (2) in applying the “but for” test to an employer’s dual motivation. Simply stated, “pretext” analysis asks, “What happened?” “But for” analysis asks, “What would have happened?” When we say an asserted ground is a pretext, we mean that we do not believe the employer when he says that he relied upon that ground in taking the challenged action. 1 When we say an employer has not sustained his burden under the “but for” test, we mean that he has not proven that he would have acted the same way if only the valid ground had existed. Of course, “but for” analysis asks the hypothetical question as to what would have happened in order to reach a conclusion as to the legal significance of what did happen, but the fact that both approaches answer the same ultimate legal question should not obscure the different factual inquiries each pursues.
*43 The Board is correct in suggesting in Wright Line that “but for” analysis can serve as the sole analytical model, but this does not eliminate the distinction between “but for” and “pretext” analysis. Relying on “but for” analysis subsumes “pretext” analysis within “but for” analysis in some cases and renders it unnecessary in others. In every case, the Board can ask, “Would the employer have taken the same action if only the valid ground had existed?” If the answer is no, the employer loses, and in that event the Board need not determine whether the employer really did rely on the valid ground in the actual case, or whether the employer really would have relied on the valid ground in the hypothetical case where only the valid ground existed. If the answer to the “but for” inquiry is yes, the employer wins, but in order to answer yes, the Board must have concluded that in the hypothetical case where only the valid ground existed, the employer really would have relied on it, and this conclusion will be reached only if the Board concludes that in the actual case the employer really did rely on the valid ground. Thus, to answer the “but for” question in the employer’s favor, the Board will have to have made a “pretext” analysis and found in the employer’s favor.
Conversely, the Board can start with pretext analysis, in which event it will have to move on to “but for” analysis only in some cases. In every case the Board can ask, “Did the employer really rely on the asserted, valid ground?” If the answer is no, the employer loses, and the Board need never make the hypothetical inquiry necessary for “but for” analysis. If the answer to the “pretext” inquiry is yes, the case is then one of dual motivation, and the Board must then move on to “but for” analysis and ask, “Would the employer have relied on the valid ground if it had been the only existing ground?”
The point is that the Mt. Healthy “but for” formula does not always eliminate the distinction between “pretext” and “dual motivation” cases. The further point is that analysis of any § 8(a)(3) case can begin with either the “pretext” inquiry as to what actually happened or the “but for” inquiry as to what would have happened. Whichever inquiry is first made, a no answer ends the case, and the employer loses; a yes answer obliges the Board to move on to the other inquiry, or implicitly to have considered it.
Moreover, as this case illustrates, it will sometimes be more sensible for the Board to start with “pretext” analysis. In this case the Board did not ask, “Would the employer have discharged Fleming if the only existing ground had been his threat to Emanuelle?” Instead the Board pursued “pretext” analysis and asked “Did the employer really rely on the threat episode?” The Board answered that question no, and therefore ruled against the employer. By using “pretext” analysis and inquiring as to what actually happened, the Board avoided the need to make the more speculative “but for” analysis and to inquire as to what would have happened if the threat episode had been the only existing ground. Had there been substantial evidence to support the Board’s conclusion, “pretext” analysis alone would have sufficed.
An additional complication arises when the Board’s decision in a § 8(a)(3) case is reviewed on appeal. In this case, Judge Mishler’s opinion for the Court suggests that since the Board used “pretext” analysis, we need not engage in “but for” analysis, supra at 39. I can agree that we need not, but I believe we should. Moreover, if contrary to my view, we do not engage in “but for” analysis, then the more appropriate disposition of the appeal would be a remand. To take the second point first, we all agree that the Board made a “pretext” analysis in concluding that the employer did not in fact rely 'on Fleming’s threat when it decided to discharge him. We also all agree that the record lacks substantial evidence to support the Board’s conclusion, and we therefore decline to enforce the portion of the Board’s order resting on that conclusion. However, if we were to end our consideration at that point, we would, in effect, have decided that the employer really did rely on Fleming’s *44 threat. As discussed previously, this yes answer to the “pretext” inquiry does not end the case; it simply requires moving on to the “but for” analysis, since at this point all we have decided is that the employer had two motives in mind, Fleming’s threat and his protected organizing activity. 2 To sustain his discharge, it is necessary to find that the employer would have discharged Fleming if the threat had been the only existing ground. Therefore, if we are not going to make the “but for” analysis, we should remand to the Board to permit it to make that analysis and then review its conclusion to see if it is supported by substantial evidence.
However, I believe it is appropriate and preferable, at least in a case such as this, for the reviewing court to avoid the delay of a remand and move on to “but for” analysis. I think we can safely assume that since the Board concluded from the evidence that the employer did not really rely on Fleming’s threat, the Board, if reversed on that conclusion and obliged to move on to “but for” analysis, would inevitably conclude that the employer would not have discharged Fleming if the threat had been the only existing ground. If the Board thought there was no reliance on this ground, it could hardly think there would have been independently sufficient reliance. 3 Having found no substantial evidence to support the Board’s “pretext” conclusion, we should therefore assume the Board would also rule against the employer on a “but for” analysis and proceed to consider whether there is substantial evidence for that conclusion. As Judge Mishler’s opinion points out, supra at 39 n.12, there is not substantial evidence for that conclusion either, and for that reason enforcement of the order concerning Fleming is properly denied, without the need for a remand. The same reasoning applies to the portion of the order concerning the failure to rehire Carmichael. For these reasons, I join the Court’s opinion.
. In the Wright Line excerpt quoted by Judge Mishler, supra at 39, the Board maintains that the employer’s asserted ground is pretextual if the ground either “did not exist, or was not, in fact, relied upon.” It is probably not accurate and certainly not helpful to consider as pretextual all situations where the employer asserts a ground that does not in fact exist. “Pretext” analysis focuses on the employer’s thought process and inquires whether he is to be believed when he says that he relied on an asserted ground. If the employer believes the employee has committed some act of misconduct, and discharges him based on this belief, then the asserted ground is not a pretext, even if the employee in fact is innocent. Of course, the circumstances would have to provide substantial indication that the employee is not innocent before the employer’s belief could be credited. However, where the asserted ground for the employer’s action is some rule that does not in fact exist, “pretext” analysis is proper. If an employee is five minutes late and the employer discharges him, claiming he relied on a rule that such tardiness always results in discharge, evidence that such a rule does not exist obviously shows that the one instance of tardiness has been seized upon as a pretext for the discharge. Even in this example, the fact of tardiness exists; what does not exist is any rule concerning such tardiness. To say the fact of the rule does not exist is simply another way of saying the employer is not to be believed when he says he relied on the fact of tardiness.
. In saying this, I assume we are not prepared to say that the record lacks substantial evidence to support a conclusion that Fleming’s protected organizing activity was part of the employer’s motivation in discharging him.
. Indeed, in some cases, when the Board applies “pretext” analysis and concludes that the employer did not rely on the asserted ground, it may have intuitively applied “but for” analysis and implicitly concluded that the employer would not have relied on the asserted ground, if it had been the only existing ground.