Wrong Line: Proposing a New Test for Discrimination Under the National Labor Relations Act*

Joshua D. Rosenberg Daneri** and Paul A. Thomas*** September, 2022

Abstract:

There has long been a consensus among scholars and union-side practitioners that the National Labor Relations Act (NLRA) is under-enforced. As a result, employers often treat violations of the NLRA as a cost of doing business rather than a serious violation of a federal statute. Calls for reform have historically tended to propose legislative amendments to the NLRA to constrain employer conduct and impose greater consequences for discrimination violations. However, little attention has been given to improving the flawed legal test by which such discrimination is analyzed, Wright Line, 251 N.L.R.B. 1083 (1980), enforced 662 F.2d 899 (1st Cir. 1981), cert. denied 455 U.S. 989 (1982). In this article, we propose a new causation test that better addresses how adjudicators should weigh the evidence that Congress and jurists have deemed relevant for evaluating discrimination claims. Our test lightens the initial burden to establish a showing of discrimination, formalizes the employer’s defense burden, and then provides a rebuttal burden for the discriminatee. This is no radical departure from historical precedent. Rather, we argue that the original Wright Line decision itself contained hints of our test, but that adjudicators and practitioners alike have whittled Wright Line to an oversimplified shell at best and an ambiguous, complex inquiry at worst. Our test better fulfills the NLRA’s objective of promoting collective bargaining in the private sector by encouraging a deeper inquiry in NLRA cases’ pre-litigation investigative stage.

I. Introduction

There has long been a consensus among scholars and union-side practitioners that the National Labor Relations Act¹ (NLRA) is under-enforced.² As a result, violations of the NLRA are often treated as a cost of doing business rather than a serious violation of a federal statute.³ Calls for reform have historically tended to propose legislative amendments⁴ to the NLRA to constrain employer conduct and impose greater consequences for discrimination violations.⁵ But reform has proved chimerical; the last significant amendments to the NLRA of any sort were in 1959 and no major pro-union NLRA reform has ever been enacted.⁶

Despite the best efforts of the NLRB and union-side counsel, it is far too easy to defeat a charge of discrimination under the NLRA. Wright Line requires the NLRB’s General Counsel (sometimes called the GC) to establish, by a preponderance of the evidence, that activity protected by Section 7 of the NLRA was a motivating factor in the adverse action taken against the employee.⁷ This “initial showing” of discrimination typically requires activity protected by Section 7 of the NLRA, employer knowledge of that activity, and that an adverse action was motivated by animus against the very same protected activity.⁸ Upon establishing this initial showing, the burden then shifts to the employer to show that it would have taken the adverse action even absent the protected activity.⁹ As such, Wright Line permits an employer to defeat even strong evidence of discrimination through persuasive inferences of what the employer might have done under a counterfactual scenario. Thus, no matter how strong of an initial showing by the General Counsel, the employer can craft a narrative of an alternate reality through which it may escape liability. As we will explore in Part 5 of this Article, employers have developed an all-too-successful bag of tricks resulting in a considerable percentage of discrimination cases never making it to a hearing,¹⁰ let alone review by the Board—the NLRB’s five-person quasi-judicial panel appointed by the President of the United States—or a reviewing court. While amending the NLRA to require harsher consequences for a violation might increase legal expenses for violating employers, such amendments would do little to rein in some of these effective anti-union tactics.¹¹

In an effort to address the ineffectiveness of the current statutory protections against discrimination, some commenters have proposed principally procedural reforms. One such reform, discussed more fully below, would have the NLRB delegate to the General Counsel the authority to independently and swiftly seek injunctions under Section 10(j) of the NLRA, eliminating some of the bureaucratic stumbling blocks that ordinarily impede petitioning a district court for a preliminary reinstatement order.¹²

While reforms to NLRB procedure are both necessary and important, they will fail if the causation test for issuing a discriminatory-discharge complaint remains as strongly pro-employer as it currently is. No complaint, no injunction—it is that simple.¹³ As we explore in Part 5 of this Article, over 90% of the NLRA’s discrimination claims are disposed of in the NLRB’s 26 Regional Offices, which investigate charges and attempt to settle the dispute before any hearing.¹⁴ If the employer wins the battle because a Regional Director dismisses the case, the charging party’s only option is to file an appeal in the hopes that, several months later, the General Counsel remands the case to the Region.¹⁵ By then, the damage is done, and the General Counsel is unlikely to satisfy the “irreparable harm” prong that most jurisdictions require for a preliminary injunction. ¹⁶ If the Regions use a flawed test to analyze discrimination, the charging party may never get its day in court. Put bluntly: the problem is even worse than it seems, because facile analyses of charges under Wright Line make otherwise-meritorious cases slip through the cracks.

Based on the foregoing, efforts to improve the enforcement of the NLRA must address how Regional Directors conduct investigations and how they weigh the evidence presented by practitioners. Our proposed test would motivate discriminatees to file charges, require employers to provide more robust defenses, and formalize a second burden-shift back to the General Counsel. This rebuttal burden would require practitioners and Regional employees to seriously contend with potentially pretextual defenses, better aligning with how courts analyze Title VII discrimination cases. This Article argues that the test for discrimination under the NLRA can be improved without waiting for Congress to act, and that such reforms would complement other scholars’ recommendations to improve internal NLRB procedure. We propose a new causation test that better addresses how adjudicators weigh the evidence that Congress and jurists have deemed relevant for evaluating discrimination claims. Our intent is not to propose a radical departure from settled law. Rather, we argue that the original Wright Line decision itself contained hints of our test, but that adjudicators and practitioners alike have whittled Wright Line to an oversimplified shell at best and an ambiguous, complex inquiry at worst.

A. Proposed Test for Discrimination

We propose the following test for cases involving discrimination under Section 8(a)(1),¹⁷8(a)(3),¹⁸or 8(a)(4)¹⁹ of the NLRA: An initial showing of discrimination is established by (i) evidence of retaliation taken, or disparate treatment condoned, against an employee for activities that implicate the concerns within Section 7, and (ii) a showing of either (a) general knowledge that the employee was inclined toward Section 7 activity or (b) evidence that the employer exhibited generalized animus against Section 7 activity. Those showings fully satisfy the statutory requirements to make out an unfair labor practice under Section 8(a)(3) of the NLRA: (i) discrimination, and (ii) tendency to increase or decrease union support.

To determine the propriety of reinstatement, backpay, and other make-whole remedies, the burden would then shift to the employer to show that it engaged in the personnel action for legitimate reasons and “for cause” within the meaning of Section 10(c)²⁰ of the NLRA. Such a showing could be made through one or more of the following: the existence of a lawful policy applied toward the employee, a track record of enforcement of that policy toward similarly situated individuals or a showing that no such violations have occurred, documentation suggesting that the decision to implement the personnel action pre-dated the Section 7 activity, or a showing that the Section 7 activity was unprotected and no reasonable employee could have believed it to be protected.

Finally, the General Counsel can rebut the employer’s showing to establish that make-whole remedies are warranted. The General Counsel’s second burden is to establish that the employer’s defense was pretextual because it was based on untrue facts or reasons that were not honestly believed. In assessing pretext, at least the following factors would be considered: shifting defenses, particularized animus toward the Section 7 activity, disparate treatment, white sheep/black sheep or collateral damage considerations, and evidence that the employer targeted or monitored the employee with the deliberate intent to find an excuse to engage in the personnel action.

Each burden should be analyzed on its own merits, and the relative strength of one burden should not be used as grounds to lighten the production regarding another burden. Thus, even a weak initial showing would not lighten the employer’s burden to show that the adverse action was legitimate and “for cause,” nor should a strong initial showing lighten the General Counsel’s second burden to establish pretext. This would preserve the burden-shifting framework’s integrity and prevent the test from becoming a haphazard balancing test à la Wright Line.²¹

This Article begins with a history of Wright Line’s origins in Part 2, followed by a critique of each element of that case’s test from a theoretic and doctrinal perspective in Part 3 and 4. We then discuss problems with Wright Line from a practitioner’s perspective in Part 5, after which we describe relevant jurisprudence from other relevant statutes in Part 6. In Part 7, we detail the rationale behind each element within our proposed test, and in Part 8 we contrast our test with Wright Line by using one hypothetical case and one actual case. In Part 9, we address potential objections to our test before closing.

II. How Wright Line Came to Be

Before we begin to discuss the flaws in current law, it will help to get a sense of how the Board got to where it currently is. It may surprise the reader to learn that the Board employed no consistent causation test for the first forty-six years following the NLRA’s enactment.²² Perhaps it is impossible from our modern-day viewpoint to imagine a world in which the Board dispensed something closer to the rough justice meted out by arbitrators.²³

Whatever the reason for this relative lack of rigid doctrinal formulation, the Wright Line Board surveyed its cases and found that the test for causation had been variously described as requiring discipline motivated, “in part,” by the protected activities of the employee,²⁴ discipline where union animus was “the motivating or moving cause”²⁵; “the motivating factor,”²⁶ “the substantial, contributing factor,”²⁷ “motivated [the discharge] principally”²⁸; was “a substantial cause”²⁹; was “a substantial or motivating ground”³⁰; or motivated the discipline “in substantial part.”³¹ It also reviewed circuit court decisions, finding cases that had required the General Counsel to affirmatively prove that the discipline would not have been issued but for the employee’s union activity.³² Yet other cases—some from the same circuit courts—required only a showing that union animus was “in part” responsible for the discipline.³³ Other formulations included “a reasonable basis for inferring that the permissible ground alone would not have” motivated discipline,³⁴ “that anti-union sentiment played a part,”³⁵ that “the force of anti-union purpose was ‘reasonably equal’ to the lawful motive prompting conduct,”³⁶ and that discipline was not “predicated solely on” lawful motives.³⁷

In our view, this chaos flowed primarily from courts adopting a crabbed view of what constitutes an unfair labor practice, adrift from the actual text of the NLRA. Section 8(a)(3), once you omit the various provisos contained therein, reads that it shall be an unfair labor practice for an employer “by discrimination in regard to hire or tenure of employment or any term or condition of employment to encourage or discourage membership in any labor organization[.]”³⁸ Discrimination, in short, does not need to cause a change in “hire or tenure of employment or any term or condition of employment.” It only needs to be “in regard to” such a change.³⁹ Under the text of the NLRA, the General Counsel’s burden in a discrimination case is to prove that the employer engaged in discrimination “in regard to” some change in the employee’s job status or conditions. It is upon this reading that we, in Part 7 of this Article, base our argument for a new approach to causation under the NLRA.

Returning for the moment to the history leading up to Wright Line, the courts and Board seemed to muddle through until a line of First Circuit cases urged the Board to apply a “dominant motive” formulation of causation.⁴⁰ These cases, primarily written by Judge Bailey Aldridge, threatened to deny enforcement to Board decisions if the Board did not apply the dominant motive formulation.⁴¹ Other circuit courts then began picking up the formula in the late 1970s, to the point where the confusion began to create real doubt as to the Board’s ability to effectively enforce the NLRA.⁴² The Board’s hand was forced, but its response, though well-intentioned, did not go far enough.

With due respect to the eminent Judge Aldridge, the “dominant motive” test just does not make sense, even if one assumes that causation of discipline (and not merely union animus “with regard to” discipline) is required to make out an unfair labor practice. At various points the First Circuit described that test as setting forth a rule of “but-for” causation.⁴³ But “dominant motive” does not do that; it requires a kind of super-but-for causation in which the illegal motive must be not merely sufficient to cause discipline, but superior to all other causes of the discipline. This is simply not what but-for causation means.

Consider a hypothetical example of age discrimination. The employer is a modeling agency that places a strong emphasis on youthful good looks. The agency is willing to hire over-40 models in some circumstances but holds the bigoted belief that they are generally unattractive and a drag upon its business. The employer rates all applicants upon a standardized appearance scale of 0-100, and deducts 10 points from the score of all over-40 applicants because it believes the use of their likeness could result in poor sales. The agency only hires applicants who score above 50. John, a 41-year-old, applies and is scored at a 55 and offered a job before the employer learns his age. Upon learning John’s age, the employer rescinds the offer. The “dominant motive” here is that John was a fringe candidate as to the employer’s appearance standards in the first place; he lost 45 points of a possible 100 due to appearance rating, and only 10 points for his age. Yet the facts of the hypothetical make clear that his over-40 status was the proverbial final straw that broke the camel’s back.

Wright Line correctly rejected the erroneous “dominant motive” formulation but because the Board felt constrained to adopt a test that required but-for causation, it also abandoned the many cases holding that a discharge “in part” based on union activity was unlawful.⁴⁴ Instead, it repurposed an inapposite test from an entirely different area of employment law, namely discharges in violation of an employee’s constitutional rights by public-sector employers.⁴⁵ In Mount Healthy School District v. Doyle, the Supreme Court confronted a situation where an employee’s teaching contract was not renewed due to the confluence of two actions, one constitutionally protected and one unprotected.⁴⁶ Although the employee showed that his protected activity was a factor in the adverse action, the Court insisted that the employer be given an opportunity to show that it would have made the same decision regardless of the protected activity.⁴⁷ The Wright Line Board took this test and repurposed it into the now-familiar inquiry in labor cases, requiring a showing of three preliminary elements: (i) protected activity, (ii) the employer’s knowledge of the protected activity, (iii) and a showing or inference that the protected activity played a role in the adverse action.⁴⁸ Consistent with Mt. Healthy, the Board additionally allowed an employer to entirely escape liability by proving that it would have taken the same action regardless of the employee’s protected activity.⁴⁹

The Board erred. Regardless of whether Mount Healthy was correctly decided in the first instance, it dealt with an implied anti-retaliation cause of action at significant remove from the plain text of the First Amendment.⁵⁰ The imposition of rigid elements and causation requirements could theoretically be justified by courts’ unwillingness to broadly restrict the scope of government action where the Constitution does not clearly compel such restriction.⁵¹ By contrast, the Board had a reasonably clear statute before it which already reflected Congress’s eagerness to restrict the scope of private-sector employer action.⁵² That statute was given virtually no textual analysis in Wright Line.⁵³ As the next two sections show, courts and the Board have grafted a problematically rigid and inflexible series of requirements onto NLRA discrimination law, hindering the Board’s ability to remedy unlawful employer conduct.

III. Wright Line’s Initial Showing: Unnecessary Complexity Distracts from the Real Inquiry

Wright Line’s problems begin, so to speak, at the beginning, with the requirements it imposes upon the party claiming discrimination to even create a rebuttable presumption of unlawful conduct.⁵⁴ As typically articulated, there are three elements to the prima facie case, more recently couched as the “initial showing” or “initial burden”:

The employee must have engaged in protected conduct (or be perceived to have done so);
The employer must have had knowledge that the employee engaged in protected conduct; and
The employer must have had animus against the protected conduct.⁵⁵

Only if all three elements are shown does the burden shift to the employer to show that the employee would have been discharged regardless of their protected activity.⁵⁶ Each of these elements, as the Board has recently applied them, has deep analytical flaws, which we explore below.

Before diving into those flaws, it is worth taking a moment to question what the point of this exercise is to begin with. If the legal question to be answered is simply “is it more likely that employee D was disciplined for union activity than for the reason the employer gave?”, then why not just ask that question in every case, without any arbitrary matrices and elements? Although we advocate a radical stripping-down of Wright Line’s elements, we do not propose to go quite this far. At their best, proof structures can incorporate fundamental evidentiary presumptions, or heuristics, about the way the world works.⁵⁷ Imagine that a fire department is deciding when to respond to 911 calls and when to dismiss them as pranks. The department might adopt a structured decision-making tree with two elements: that if (1) there is a 911 call and (2) one can see smoke from the cupola of the fire station in the vicinity of the call location, then firefighters should be dispatched. Occasionally, the smoke will turn out to have been just a coincidence, but the vast majority of the time, the rule will lead to dispatch in circumstances where there is a real fire.

Where this becomes problematic is when heuristics turn into rigid requirements of boxes that must be checked in order to prove one’s case. At that point, proof becomes less about the overall picture and more about individual flyspecking of the evidence as to specific points of the initial showing. Difficult-to-meet mandatory proof structures such as Wright Line become an exercise in missing the forest for the trees. It is for this reason that we advocate a radical reimagining of Wright Line—not as a series of atomized elements, but as an analysis designed to reflect the way that the real world operates.

A. “Protected Activity”: Traps for the Unwary

Wright Line’s difficulties begin with the very first element of the initial showing—the requirement that an employee engage in activity protected by the NLRA. This Article would be twice its length if we were to engage in a full exegesis of what types of activity enjoy substantive protection under the NLRA, and the ways that such protections have been narrowed by the Board and courts.⁵⁸ To avoid such an extended digression, we will simply take the landscape of protected activities as a black box.

Even doing so, there are significant problems with Wright Line. For one thing, as the Board itself has identified, employees can be victimized in ways that violate labor policy even when they have not engaged in any protected activity. Sometimes the employer wrongly perceives them to have done so, such as when an employer fires an employee it thinks supports a union even though they do not or have not yet formed an opinion.⁵⁹ Sometimes the employer wants to cut protected activity off at the source with a “preemptive strike” aimed at preventing that activity from ever happening in the first place.⁶⁰ When these types of behaviors are properly identified they are generally subject to liability, but Wright Line’s misleading language has the potential to deceive parties and investigators about this area of the law.

More problematically, it is routinely the case for employees to engage in activity that employees could reasonably believe is protected, only to have it turn out not to be protected after-the-fact. Sometimes this is because activity is characterized as having an objective that isn’t employees’ place to complain about.⁶¹ Sometimes it is because employees use tactics that the Board or courts deem off-limits.⁶² Sometimes it is because employees are deemed to have exceeded the boundaries of acceptable conduct during the course of their protest.⁶³ And sometimes it is simply that employees guess wrong about the law and employers are allowed to punish them for doing so.⁶⁴

This approach is fundamentally ill-conceived. Employees do not carry law books with them to work,⁶⁵ and the interests of labor law are best served when any conduct that even arguably implicates the interests that the law protects is shielded from employer reprisals. Employers have enough power in the workplace without possessing the additional power to exact revenge against employees who, acting in good faith, commit trivial tactical or legal errors. Nowhere in the NLRA is the expanse of concerted activity for mutual aid and protection delimited. Section 1 of the NLRA declares the policy of the United States to be protecting “the exercise by workers of full freedom of association, self-organization, and designation of representatives of their own choosing, for the purpose of negotiating the terms and conditions of their employment or other mutual aid or protection.”⁶⁶ The usage of the word “or” here implies that matters other than mere terms and conditions of employment are meant to be encompassed. This expansive view of Section 7 was also underscored in Eastex, Inc. v. NLRB, where employees acting “through channels outside the immediate employee-employer relationship” was found to constitute protected activity within Section 7’s “mutual aid and protection” clause.⁶⁷

For these reasons, Wright Line’s first prong must be fundamentally rethought. We recommend that the Board reformulate the prong to only require employees to have engaged in activity that touches or concerns the policies underlying the Act’s substantive protection of union and other concerted activities (for short, “Section 7 activity”). So long as the employee has not acted in bad faith (such as by intentionally lying, resorting to physical violence or true threats of such, or making complaints that the employee knows are utterly baseless), their Section 7 activity should be shielded if it arguably relates to the objectives of the NLRA. The workplace should be a place for the free exchange of ideas, not one where employees cower in fear of reprisal if they do not phrase their words in the exact right way for the benefit of after-the-fact observers. Our position in this regard is consistent with the General Counsel’s expansive interpretations of the “mutual aid and protection” and “inherently concerted” doctrines.⁶⁸

B. “Knowledge”: An Element with No Purpose

As problematic as Wright Line’s first element is, it pales in comparison with the second, which requires the charging party to show that the employer had knowledge of the employee’s protected activity. This seemingly innocuous requirement sets another series of needless obstacles in the way of meritorious cases, and for unpersuasive reasons. Consider an employer that actually does not know or suspect that an employee had engaged in protected activity. How would that employer behave? Exactly as it would have even in the absence of the protected conduct. If that phrasing sounds familiar, it should—it is a mere restatement of the Wright Line affirmative defense!⁶⁹ When one has no information about something, then by definition that something cannot change one’s behavior.

So what, precisely, is the element of “knowledge” adding to Wright Line’s test? From a strictly logical point of view, almost nothing. Yes, making knowledge an element of the initial showing theoretically could change the outcome of a case where the decisionmaker thinks it is exactly as likely that the employer knew about the employee’s protected activity as that it did not know. But that will hardly ever be the case to begin with, and it is made even less likely when one realizes that knowledge is often inferred through the same kinds of circumstantial evidence that rebut the employer’s Wright Line defense.⁷⁰ “Knowledge” as an element does no independent work here.

What the knowledge prong unmistakably does do is provide a roadmap for unsympathetic adjudicators to pick apart strong cases on hyper-technical grounds. Take Gestamp South Carolina v. NLRB, for example.⁷¹ Two employees were fired in the context of a heated union drive, and one of the employees had been specifically threatened with discharge by a supervisor if the company found out about his protected activities.⁷² The other employee was fired for a trivial one-time error in entering his timecard.⁷³ The Administrative Law Judge who heard the case unsurprisingly found that these discharges—one following through on a prior threat, the other a transparently obvious pretext—were motivated by the employees’ union activity, and the Board affirmed the decision.⁷⁴ Yet the court seized on the Administrative Law Judge’s inability—due to vagueness in the employer’s own testimony—to identify exactly who had made the decision to fire the employees.⁷⁵ The court then used that gap in the record to reverse the Board on the sole ground that the Administrative Law Judge had not drawn a specific enough inference that the decisionmaker—whoever it was—knew of the employees’ union activity.⁷⁶ All of the circumstantial evidence that these terminations were illegally motivated was simply ignored, because the Administrative Law Judge had failed, in the court’s view, to properly check the “knowledge” box.⁷⁷

Nor is this the sole example; one can cite any number of cases where inability to prove knowledge was cited as the dispositive factor while similarly overwhelming circumstantial evidence that a termination was unlawful was ignored or discounted.⁷⁸ When a required element of a legal test allows the adjudicator to reach preposterous factual conclusions based upon trivial inadequacies of the prosecution’s initial showing, that element fails to achieve its purpose.

C. “Animus”: A Misnomer with Harmful Consequences

Wright Line’s third element is commonly shorthanded as “animus,” but this is in fact a misnomer.⁷⁹ What this element actually requires is some evidence that the action the employer took against the employee was motivated by the employee’s protected activity.⁸⁰ In Wright Line itself, the Board emphasized that it need not have been the “dominant” motive.⁸¹ Moreover, unlawful motive can be shown in a variety of ways; evidence of animus against protected activity is only one of a plethora of proof tools. Others include the commission of contemporaneous unfair labor practices,⁸² suspicious timing of adverse action in relation to protected activity,⁸³ inexplicably disparate treatment of similar behavior,⁸⁴ evidence that the employer rushed to judgment and did not seriously investigate whether the employee actually committed any wrongdoing,⁸⁵ failure to follow an employer’s own internal disciplinary guidelines or procedures,⁸⁶ and most importantly of all, pretext—the showing that the employer’s asserted reasons for disciplining an employee were false or not honestly believed.⁸⁷

What this seeming grab-bag of tools have in common is that all of them are heuristics that suggest that the employer is not acting as it normally would when dealing with a potential disciplinary problem, but rather is targeting an employee for their protected activity. When properly applied, the third Wright Line element is an ecumenical inquiry into the totality of circumstances surrounding an employee’s discipline, which takes into account not just the direct admissions of the employer’s personnel (though those are often supremely useful), but the natural and probable inferences that one can draw from the facts of the case.⁸⁸

When wrongly applied, however, Wright Line and its progeny can sometimes turn into a snipe hunt for “animus,” an ill-defined concept that allows the factfinder to close their eyes to virtually any evidence short of a direct admission that an employee was fired for protected activity. Cases have been dismissed on the grounds that: the employer lied about its real reasons, but the employee can’t prove exactly what the employer was trying to cover up;⁸⁹ the employer hated union supporters, but there’s no direct evidence that it hated this union supporter;⁹⁰ or that well-honed chestnut—the comparator employees were too different to warrant an inference of discrimination (where the degree of difference is essentially in the eye of the beholder).⁹¹ In a handful of instances before 2003, the Board even applied a fourth “nexus” requirement to the initial showing.⁹² Though the Tschiggfrie Board did not formally add the nexus requirement, it nonetheless emphasized that Wright Line is inherently a test of causation.⁹³ Instead of adding a nexus requirement, the Tschiggfrie Board essentially transformed the animus requirement into a nexus requirement.⁹⁴ This is quite a departure from the Supreme Court finding discharges for union activity to be violations of Section 8(a)(1), and finding it unnecessary to even reach the 8(a)(3) question because the “Defeat of [Section 7] rights by employer action does not necessarily depend on the existence of an anti-union bias.”⁹⁵

At the risk of overgeneralizing these cases, the underlying problem with their reasoning is a persistent failure to grapple with industrial reality. Indeed, the same failure pervades Wright Line’s test, which essentially presumes that an employer’s actions are lawfully motivated unless there is particularized evidence to the contrary.⁹⁶ But it blinks reality in a landscape dominated by aggressive anti-union campaigning.⁹⁷ In that world, most union organizing drives are met with employer opposition, often involving unlawful tactics.⁹⁸ And most employers resist good-faith bargaining when unions do manage to achieve certification.⁹⁹ This should be unsurprising; the economic incentives to avoid unionization and maintain unilateral control over working conditions, and the feeble-to-nonexistent costs associated with committing unfair labor practices,¹⁰⁰ mean that it would be virtually a breach of fiduciary duty for an employer to follow the law in a circumstance where flouting it promises greater profits. Yet Wright Line and its progeny take no account of these background circumstances, evaluating every discharge as if the economic incentives involved were identical. Faced with a recurring pattern of pervasive illegality, the Board and courts ignore industrial realities.

Admittedly, a part of this is Congress’s fault. Section 8(c) of the NLRA prohibits the Board from using an employer’s statements as evidence that it has committed an unfair labor practice unless those statements amount to threats of reprisal or promises of benefits.¹⁰¹ The fact that an employer has publicly attacked its employees for exercising their legal rights or declared its implacable opposition to the exercise of such rights is obviously highly probative evidence in any case where the employer is alleged to have violated those rights by discharging an employee. But repealing Section 8(c) is a matter for legislation, not litigation.¹⁰² Still, even if Section 8(c) limits the degree to which the Board can make use of a particular employer’s virulent hostility to employees’ legal rights, it says nothing about the Board’s ability to take notice of general behavior by employers faced with certain types of hot-button factual situations such as an organizing drive.

We accordingly urge a dramatic relaxation of the initial showing burden with regard to the element of “animus” or motivating factor. To obtain a trial on the real question at hand—whether or not this discharge, more likely than not, was the result of union activity—the only proof that should be required is either (1) evidence that the employer was aware of the employee’s proclivity toward union organizing (or any other concerted activity that, if successful, would naturally tend to weaken the employer’s control of the workforce or its bottom-line profits) or (2) a minimal showing of circumstantial evidence of the employer’s hostility to protected activity writ large. Once that minimal hurdle is satisfied, the case should go to trial.¹⁰³ The presumptions and frameworks that such a trial should employ are discussed in more detail in Part 7 of this Article below.

IV. Wright Line’s Affirmative Defense: Letting Employers Off the Hook Even When They Take Actions with Improper Motives

Even assuming that an employee’s charge of discrimination survives the gauntlet of obstacles that have been set up for it at the initial-burden stage, the employer is still—under the Mt. Healthy balancing test repurposed in Wright Line—given an opportunity to escape scot-free by showing that it would have made the same decision despite the employee’s protected activity.¹⁰⁴ This affirmative defense is at least an improvement on the Wright Line initial showing insofar as it reflects a totality-of-circumstances analysis rather than a rigid list of required elements, but it nevertheless remains inadequate to cope with the landscape of discriminatory conduct in the modern workplace.

For one thing, it is not consistent with the statute. As explained above, the NLRA does not require but-for causation, it requires only discrimination that would tend to increase or decrease employees’ willingness to engage in protected activity.¹⁰⁵ Section 10(c) limits liability for unfair labor practices only in circumstances where an employee has been discharged for cause, which requires the employer to not merely show that it would have fired an employee regardless of the employee’s union activity, but that such a firing would actually have been justified.¹⁰⁶ For this reason, cases holding that an employer can escape liability based upon a mistaken belief that an employee committed misconduct are wrongly decided, except where the employer’s action was completely free of unlawful motivation (or, to put it another way, cases where the affirmative defense is unnecessary because the General Counsel has never satisfied his or her initial burden).¹⁰⁷ A mistake is not “cause.”

Furthermore, because the Wright Line affirmative defense rests on an essentially standardless counterfactual, it tends toward reasoning by inference; there will almost never be evidence that directly bears on the question of what an employer would have done in a hypothetical world. This causes the affirmative defense to needlessly rehash elements of the initial showing to the point where it is frequently difficult to tell the two analyses apart. Among the elements that have been found pertinent to the question of whether an employer would have acted in the absence of union activity are the giving of a pretextual explanation,¹⁰⁸ disparate treatment of an employee relative to others,¹⁰⁹ the timing of the adverse action,¹¹⁰ and contemporaneous unfair labor practices.¹¹¹ All of these should sound familiar because they are functionally the same elements used to establish “particularized” animus under Tschiggfrie and its predecessors.

But the most egregious error of Wright Line’s affirmative defense is how it treats cases where the defense is established: it dismisses those cases outright. The Board breezily justified this outcome by claiming that “should the employer be able to demonstrate that the discipline or other action would have occurred absent protected activities, the employee cannot justly complain if the employer’s action is upheld.”¹¹² This is untrue—regardless of technicalities of causation, we firmly believe that a reasonable person would feel aggrieved when an employer acted against them based in part on illegitimate considerations. This is also irrelevant given that the purpose of the NLRA is not to assuage the feelings of employees, but to enforce a public policy of separating employees’ union activity from their job tenure.¹¹³ Regardless of what happens with that particular employee, it is head-scratching that the employer is not enjoined from repeating its unlawful behavior. While not necessarily a strong deterrent, the presence of an unfair labor practice adjudication against an employer is relevant in many circumstances.¹¹⁴ The Supreme Court expressly recognized in NLRB v. Transportation Management that the Board “might have considered a showing by the employer that the adverse action would have occurred in any event as not obviating a violation adjudication but as going only to the permissible remedy[.]”¹¹⁵ In that vein, Title VII’s mixed-motive provisions are so worded; a showing that an employee was disciplined “for cause” defeats only damages, not injunctive relief or an award of attorney fees.¹¹⁶

In short, the question of whether an employee was fired “for cause” should have no independent relevance to whether an unfair labor practice occurred.¹¹⁷ The only relevant question is whether the employer took an adverse action motivated in some fashion by protected activity. Of course, the existence of a legitimate-if-credited explanation for the employer’s actions would tend to show that its actions were not badly motivated, which is why the test we propose continues to evaluate it. But the factfinder cannot and should not lose sight of the ultimate question to be answered.

V. Problems with Wright Line from a Practitioner’s Perspective

Thus far, we have focused mainly on flaws in how the Board and the courts have analyzed discrimination cases. However, the vast majority of NLRB cases are disposed of at the NLRB Regional office level. Indeed, a typical Region finds merit to only about a third of charges, of which over 90% settle before trial.¹¹⁸ But despite these seemingly rosy settlement statistics, differential approaches to investigations and employers’ legal tricks may disincentivize employees from filing charges in the first place.

A. Quasi-judicial vs. Prosecutorial Approaches to Investigation

Here, we detail the peculiarities of investigating discrimination charges at the Regional level, depending on whether the investigatory approach is quasi-judicial or prosecutorial. These characterizations are based on public data,¹¹⁹ are consistent with our personal experiences having worked under various Regional Directors, and are deliberately over-generalized to show the wide range of possible approaches to Regional investigations. A more streamlined discrimination test could provide Regions with better guidance as to how to exercise their considerable discretion.

The NLRB Casehandling Manual encourages a thorough investigation that “must reveal the totality of the circumstances” while also “avoid[ing] unnecessary expenditure of time and energy.”¹²⁰ Charging parties must give sworn affidavits and provide documentary evidence lest their charges be dismissed for lack of cooperation.¹²¹ In contrast, charged parties’ (whether unions or employers) mere assertions in position statements are often taken at face value¹²² because the NLRB lacks the resources to compel full cooperation by charged parties. In some instances, a Region may take advantage of its plenary authority under the statute to issue investigative subpoenas to uncooperative parties.¹²³ Ultimately, “[t]he Regional Director has the final authority and responsibility to make all casehandling decisions within the Regional Office.”¹²⁴

Quasi-judicial approaches leave no stone unturned in the investigation. Agents take lengthy and detailed testimony from charging-party witnesses and liberally issue investigative subpoenas to neutral witnesses whenever necessary. Regional staff weigh all evidence, much in the way a judge might, before determining whether the case has merit. In effect, the case is pre-litigated before a complaint might issue. Because a merit determination requires strong evidence, one might presume that this approach affords considerable leverage in settlement negotiations, as charged parties can reasonably conclude that their chances of victory at trial will be slim.¹²⁵ On the other hand, in an effort not to prejudge cases, this approach may discourage Board agents from settling prior to a merit determination, requiring additional expenditure of resources for weak cases. The quasi-judicial approach can be burdensome for charging parties, who could be discouraged by the duration and degree of personal investment in the process. Those same charging parties, however, may feel reassured that their story was meaningfully heard and digested.

Alternatively, agents may take a more prosecutorial approach toward a given case, characterized by aggressiveness and greater risk-taking. Under this approach, only a colorable showing of a violation is required to warrant a merit finding, even if the charged party has a plausible defense. Complaints are issued more liberally because comparatively weaker evidence will still result in a merit finding by a Regional Director. Taking literally the NLRB case handling manual’s instruction that only a judge can make credibility determinations, prosecutorial approaches give charging parties their day in court instead of exhaustively investigating a charge only to conclude that the charge ultimately lacks merit.¹²⁶ Thus, this approach envisions a strictly prosecutorial role by the Region on behalf of the charging party, after which judges, the Board, and the courts (rather than the Region itself) are left to wrangle with the minutiae and credibility issues.

We do not wish to express an opinion on the relative merits of the quasi-judicial or prosecutorial approaches. There is no clear evidence that either approach produces better results for employees.¹²⁷ And all Regions, even the most conservative or aggressive, will employ a combination of approaches depending on the type of case.¹²⁸ Rather, we aim to show that the legal standard for assessing NLRA discrimination should be formulated with the pre-litigation investigative stage in mind. Wright Line provides no guidance for investigators to assess what constitutes a proper defense to a showing of discrimination, nor does it cue the investigator to probe into possible pretext. In Part 8 of this Article, we provide an example of how, even irrespective of a quasi-judicial or aggressive prosecutorial approach, Wright Line’s superficiality can result in anti-Section 7 discrimination never being litigated.

B. Employers’ Tactics

Another issue with Wright Line from a practitioner’s perspective is the numerous tricks skilled employer counsel have at their disposal. Based on personal experience during our careers in the field, we suspect that some attorneys are advising employers to target secondary union supporters who are less vocal or conspicuous, to create plausible deniability by virtue of leaving the lead activists alone.¹²⁹ Other employers may discharge such activists alongside innocent bystanders to prove a lack of disparate treatment and broad application of an employer’s policy.¹³⁰ The latter are called white sheep/black sheep or collateral damage cases, which are notoriously difficult to prove and will be addressed in greater detail when we discuss our proposed General Counsel rebuttal burden. Finally, experienced counsel take advantage of the fact that Regional Directors do not always subject assertions and documentary evidence contained in employer position statements to the same level of scrutiny directed toward charging parties.¹³¹ Accordingly, during the pre-litigation investigation, it is common for employers to provide very few details about the contexts in which other employees may have been subjected to adverse actions pursuant to a policy, or the deliberative process through which the employer ultimately decided to undertake the adverse action.¹³² This lack of detail makes it difficult for Regions to assess whether a personnel action was undertaken “for cause” within the meaning of Section 10(c) of the NLRA, which lamentably is not a factor that the Board and the courts even consider when applying Wright Line.

VI. Looking to Other Statutes for Hints of a Better Test

Having established that Wright Line is deficient for historic, doctrinal, and practical reasons, where might one turn to find a suitable replacement? While imperfect, we believe that jurisprudence from other discrimination statutes at least contains some hints. Hence, we analyze the Supreme Court’s decisions in Mt. Healthy,¹³³ McDonnell Douglas Corp. v. Green,¹³⁴ Price Waterhouse v. Hopkins,¹³⁵ and their progeny for inspiration.

A. First Amendment

As noted in Part 2 of this Article, the Wright Line test originated from Mt. Healthy City Bd. of Ed. v. Doyle, a First Amendment case. Fred Doyle, an untenured teacher in a public school district, was discharged shortly after he made critical telephonic remarks about a school dress code to a radio station.¹³⁶ Doyle also engaged in other significant misconduct, including a verbal altercation with another teacher, using profanity toward students, and making a lewd gesture toward female students.¹³⁷ Doyle sued the Mt. Healthy School District Board of Education for violating his rights under the First and Fourteenth Amendments to the United States Constitution. The District Court, affirmed, by the Court of Appeals, held that the telephone call was protected by the First Amendment and that because it played a substantial role in the decision to discharge him, Doyle was entitled to reinstatement and backpay.¹³⁸ The Supreme Court, however, was concerned that such a remedy “could place an employee in a better position as a result of the exercise of constitutionally protected conduct than he would have occupied had he done nothing.”¹³⁹ The Court reasoned:

This is especially true where, as the District Court observed was the case here, the current decision to rehire will accord “tenure.” The long-term consequences of an award of tenure are of great moment both to the employee and to the employer. They are too significant for us to hold that the Board in this case would be precluded, because it considered constitutionally protected conduct in deciding not to rehire Doyle, from attempting to prove to a trier of fact that quite apart from such conduct Doyle’s record was such that he would not have been rehired in any event.¹⁴⁰

Thus, the Court proposed a test whereby the plaintiff bears the initial burden to show that his conduct was constitutionally protected and that his conduct was a substantial or motivating factor in the employer’s decision not to rehire him, after which the burden shifts to the employer to show, by a preponderance of the evidence, that it would have reached the same decision as to respondent’s re-employment even in the absence of the protected conduct.¹⁴¹

Mt. Healthy’s context is distinguishable from those of a typical NLRB discrimination charge in three respects. First, a reinstatement remedy would have granted tenure to Doyle at his public school. As such, the remedy had a longer-term impact than would the ordinary discharge under the NLRA, after which an employer can still reasonably discharge a reinstated employee for lawful reasons.¹⁴² Second, the case involved the free-speech provisions of the First Amendment rather than a statute that expressly protects against a certain type of discrimination.¹⁴³ Thus, the Court was evaluating conduct that an employee had the right to engage in, but without a corollary prohibition on the employer’s adverse action. The lack of such a prohibition arguably would grant an employer greater freedom to engage in an adverse action. Finally, Section 1 of the NLRA states that the purpose of the Act is to promote Section 7 rights, and Section 8(a)(1) and 8(a)(3) make it unlawful to interfere with those rights.¹⁴⁴ Unlike the First Amendment, the NLRA implicitly contemplates an inherent power differential between employers and employees that necessitates government intervention (the NLRB’s General Counsel and Board) to protect certain kinds of employee behavior. No such entity exists for the singular purpose of enforcing the First Amendment.

B. Title VII of the Civil Rights Act of 1965

Discrimination in the context of discriminatory hiring under Title VII of the Civil Rights Act of 1965 is often analyzed under McDonnell Douglas Corp. v. Green.¹⁴⁵ A prima facie case of discrimination under that statute is established by producing evidence that (1) plaintiff is protected by the applicable employment discrimination statute; (2) she applied for a job for which she was qualified; (3) she was not hired; and (4) the position remained open and the employer continued to seek applicants with qualifications like those of plaintiff.¹⁴⁶ If this burden is satisfied, there is a weak rebuttable presumption of discrimination.¹⁴⁷ The employer may then rebut that presumption by producing evidence of a “legitimate, nondiscriminatory reason” for the adverse action.¹⁴⁸ If the employer satisfies this minimal burden, the burden of production shifts back to the plaintiff to prove that the employer’s proffered reason was pretextual.¹⁴⁹

McDonnell Douglas is significantly different from Mt. Healthy for four reasons: (1) it expressly addresses disparate treatment; (2) it does not require an explicit examination of the employer’s motives; (3) it requires that employers proffer legitimate reasons for adverse actions as opposed to merely other reasons; and (4) it shifts the burden back to the plaintiff to establish that the employer’s defense was pretextual. Nonetheless, McDonnell Douglas has been criticized by scholars for a slew of reasons,¹⁵⁰ including the ambiguous role that the prima facie case plays in later stages of the analysis,¹⁵¹ the tortification of the test with reliance on the weak tort principle of res ipsa loquitor,¹⁵² and the difficulty of evaluating causation in mixed-motive cases.¹⁵³ Still, by evaluating disparate treatment rather than motive, this test requires the parties to evaluate the personnel action in a broader and more practical context. Further, it forces the adjudicator not to accept employers’ defenses at face value, requiring an assessment of whether the reason was legitimate, and if so, whether there nonetheless might be reasons to believe that that reason was pretextual.

Curiously, McDonnell Douglas is not mentioned in the Supreme Court’s recent landmark Bostock v. Clayton County decision, in which the Court held that discrimination against sexual orientation or gender identity constituted discrimination based on sex.¹⁵⁴ Noting that Title VII’s prohibition on sex discrimination uses the terms “because of” without the qualifiers “solely” or “primarily,” the Court reasoned that it could find a violation if sex was one of several motivating factors in the personnel action, even amid other lawful factors.¹⁵⁵ Ipso facto, if “because of” encompasses situations where an employer’s action is motivated by more than one cause, then the NLRA’s use of the even broader “in regard to” must operate similarly.

After McDonnell-Douglas, in Price Waterhouse v. Hopkins, the Court clarified that Title VII cases generally fall into two categories.¹⁵⁶ For pretext cases, where the issue is the employer’s true motive for a personnel action, the McDonnell Douglas framework applies.¹⁵⁷ For mixed-motive cases, where there is direct evidence that forbidden factors were considered in reaching an employment action, the Mt. Healthy test applies.¹⁵⁸ Significantly, the Court in Price Waterhouse mentioned NLRB v. Transportation Management Corp., where the Court affirmed Wright Line, emphasizing the virtue of consistency across the various anti-discrimination statutes.¹⁵⁹

We view Price Waterhouse as taking a wrong turn here. The search for a “true motive” will forever remain elusive for both epistemic and practical reasons. Indeed, courts have struggled mightily to apply the distinction,¹⁶⁰ and scholars have had equal difficulty in parsing it.¹⁶¹ We accordingly reject the pretext/mixed-motive distinction of Price Waterhouse in our proposed test.

C. Age Discrimination in Employment Act of 1967

The Supreme Court’s decision in Babb v. Wilkie illustrates the importance of scrutinizing the precise language within discrimination statutes.¹⁶² The federal-sector provision of the ADEA states, in relevant part: “All personnel actions affecting employees or applicants for employment who are at least 40 years of age . . . shall be made free from any discrimination based on age.”¹⁶³ The Court held that “free from any discrimination based on age” warranted broader consideration than the ADEA’s private-sector counterpart,¹⁶⁴ which provides that “It shall be unlawful for an employer. . .to fail or refuse to hire or to discharge any individual or otherwise discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s age”.¹⁶⁵ Thus, the Court in Babb v. Wilkie held that for federal-sector ADEA claims, a mere showing that age was considered in a personnel decision is sufficient for relief, in contrast to the “but-for” test required in all private-sector employment discrimination statutes.¹⁶⁶ One might reasonably infer that the Court would hold that the NLRA, which also does not use the terms “because of,” would entail similarly broad protections.

D. Lessons from Other Statutes

The foregoing discussion should make clear that NLRA discrimination jurisprudence has fallen behind that of other employment-discrimination statutes. Analogous statutes have developed tests that have evolved beyond the oversimplified Mt. Healthy considerations and wrestle more deeply with statutory text. The Supreme Court takes very seriously the precise wording of the causal language within these statutes and how that wording bears on evidentiary burdens. Rather than being awkwardly retrofitted to align with these separate jurisprudences,¹⁶⁷ the NLRA needs its own test that properly recognizes the subtleties of its language and contemporary industrial realities. In the following section, we explain how our proposed test is better tailored to the language within Section 8(a)(3).

VII. Explaining Our Proposed Test

The Mt. Healthy test is inadequate for proper enforcement of the NLRA, and the Board’s jurisprudence on discrimination has failed to engage in the kind of close reading of the statutory language that courts have engaged in for similar discrimination statutes. The NLRA needs a test that relies more heavily on language found directly in the statute: whether there is evidence of interference, restraint or coercion;^[170]whether the conduct encourages or discourages membership in any labor organization;¹⁶⁸ whether discrimination occurred “in regard to” a term or condition of employment;¹⁶⁹ and whether the personnel action was “for cause.”¹⁷⁰

Whereas the NLRA protects union-like activity in addition to union activity (by virtue of the “concerted activities for mutual aid and protection” clause), the other statutes do not have similar breadth. Thus, the plain language of the NLRA protects not only union supporters but also the constellation of collective workplace activities that might germinate in response to employer overreach. Such a framework differs from the static, identity-based categories contemplated by Title VII and the ADEA, in which significant work is required to differentiate between discrimination based on belonging to a protected category (e.g., race), and conduct that may be implicated by or associated with that category (e.g., racially stereotypical behavior or appearance). Those statutes also require balancing different interests such as diversity and the need to address and remedy systemic racism and historic patterns of discrimination.¹⁷¹ In some respects, the NLRA bears a closer relationship to disability discrimination, which expressly encompasses not just identity but also history (it protects employees with a record of having been disabled) and employer perceptions (protecting employees who are regarded by an employer as disabled).¹⁷²

It follows that the NLRA needs a uniquely tailored test that properly assesses discrimination rooted in the broad protections contemplated in Section 7. Such a test must assess conduct and use proper heuristics to account for the interaction between employee and employer interests. In the sections that follow, we explain why our test would be an effective replacement for Wright Line, evaluating each of our proposed burdens in turn.

A. Initial Showing

As we explained in Part 2 of this Article, the broad language of the NLRA justifies an expansive view of what might ultimately be deemed “protected” activity under the statute. Indeed, Section 8(a)(1) of the Act makes it an unfair labor practice for an employer to “interfere with, restrain, or coerce employees in the exercise of the rights guaranteed by Section 7.”¹⁷³ Discharges for concerted, rather than union, activity are violations of Section 8(a)(1)—not Section 8(a)(3).¹⁷⁴

Moreover, the knowledge, animus, and adverse-action prongs can be abused by adjudicators to dismiss otherwise meritorious cases. The knowledge prong is redundant at best, especially when there is evidence of particularized animus toward the discriminatee. In a similar vein, the “animus” prong, while reasonable at first blush, has been interpreted by the Board and the courts to imply a causal nexus requirement.¹⁷⁵ Causality is difficult for researchers to prove through controlled laboratory experiments, let alone by jurists on the basis of testimony in an NLRB proceeding.¹⁷⁶ The nexus requirement, too, is prone to questionable outcomes.¹⁷⁷ Additionally, the “adverse action” requirement connotes concrete employment decisions rather than the sometimes-subtle forms of retaliation that would nonetheless implicate Sections 8(a)(1) and 8(a)(3) of the NLRA.¹⁷⁸

We therefore propose that an initial showing of discrimination be established by (1) evidence of retaliation taken or disparate treatment condoned against an employee for activities that implicate the concerns within Section 7; and (2) a showing of either general knowledge that the employee was inclined toward Section 7 activity or that the employer exhibited generalized animus against any Section 7 activity. Rather than functioning as a rebuttable presumption, the initial showing is sufficient in and of itself to establish the elements of an unfair labor practice under Section 8(a)(3), let alone Section 8(a)(1) and (4). That initial showing would thus be the opening of a deeper remedial inquiry under Section 10(c) of the Act as to whether the employer discharged the employee “for cause” so as to make the employee ineligible for a make-whole remedy including reinstatement and backpay. While we recognize that Transportation Management compels the initial showing to be a burden of persuasion, such a burden can be satisfied without resorting to futile analyses of causality.¹⁷⁹ This initial showing would eliminate any implication of a causality or nexus requirement, and it would reserve the more nuanced elements of the analysis to the second and third burdens.

B. Employer’s Burden

Under Wright Line, an employer meets its burden by showing that it would have engaged in the same action absent the protected conduct. The purported goal is to prevent scenarios in which a bad employee is placed in a better situation than they would have been had they engaged in similar misconduct, but not protected conduct. Despite this good intention, the practical effect of the employer’s Wright Line burden has been to encourage skilled counsel to proffer inferential evidence of what the employer could have done, rather than direct evidence of what the employer would have done or, in fact did.¹⁸⁰ But even assuming that an employer approaches its Wright Line burden in good faith, this test unjustifiably traps the parties and the adjudicator in a hypothetical, counterfactual world, with no clear guidance on how the employer operates in reality. Our proposed modification of the employer’s burden eliminates this guesswork, instead requiring an evaluation of the employer’s policies and direct evidence linking the adverse action to a violation of these policies.

Another problem with the employer’s Wright Line burden is that it gives excessive deference to the doctrine of at-will employment. If an employer provides a petty reason or fails to give the employee the reason for the adverse action, an employer may still meet its burden.¹⁸¹ At-will employment forms the unfortunate starting point in all jurisdictions except for Montana, and it gives employers considerable leverage both in employment decisions and as litigants.¹⁸² We believe that an employer should be held to a higher standard: if an employer must provide a “legitimate . . . reason” in Title VII cases analyzed under McDonnell Douglas, it should have to do so for NLRA cases as well.¹⁸³ This is consistent with Section 10(c) of the NLRA, which permits an employer to escape liability by showing that a discharge was “for cause.”

Requiring an employer to provide a legitimate reason for the adverse action is also consistent with NLRB v. Burnup & Sims. There, the Supreme Court held that when an employer discharges an employee for misconduct allegedly committed during Section 7 activity, a violation will be found if it can be shown that the employee did not actually engage in the stated misconduct.¹⁸⁴ In other words, the Court has recognized that even an employer’s good-faith belief that an employee engaged in misconduct is insufficient to evade liability when the alleged misconduct occurred in the course of protected activity.

But Burnup & Sims does not go far enough. It is natural in some workplaces for interactions to become heated, provoking unbecoming employee conduct that would reasonably prompt an employer to take an adverse action. Such scenarios were historically evaluated by the Board under Atlantic Steel Co.,¹⁸⁵ using a four-part test to determine whether an act of misconduct was so egregious as to render a loss of protection.¹⁸⁶ Cases abound in which the Board found a violation despite an employee’s use of profanity,¹⁸⁷ yelling,¹⁸⁸ or vulgarity.¹⁸⁹ True to form, the Board in General Motors LLC eliminated the Atlantic Steel test and held that Wright Line alone was sufficient to analyze employee outbursts during Section 7 activity. ¹⁹⁰ Hence, without Atlantic Steel, an employee who yells and insults their supervisor in the course of Section 7 activity is unlikely to fare well if the employer has evidence that the employer routinely disciplines employees who engage in similar misconduct. Notwithstanding, even under Atlantic Steel, let alone General Motors, the adjudicator can sua sponte determine that an employee’s otherwise unambiguous Section 7 activity can ultimately lose protection under the Act because of conduct that crosses the line.

The debate over loss of Section 7 protection articulated in General Motors evinces another problem with the current test: the concept of “protection” in discrimination cases is manifold, leading to ambiguity as to which concept is being invoked, when it should be invoked, and who bears the burden of proving it. Depending on the context, the concept of protection is invoked as a prong in analyzing whether concerted activity is undertaken for mutual aid and protection, as the first prong in the establishment of an initial showing under Wright Line, or to evaluate those Atlantic Steel scenarios determining whether an employee’s misconduct renders otherwise unlawful employer activity permissible under the circumstances.

Furthermore, neither Atlantic Steel nor Burnup & Sims have ever protected employees from the more subtle form of discrimination in which an employer merely takes note of an employee’s protected activity, then commences a search for later “cause” to discharge.¹⁹¹ The original Wright Line decision itself involved such a fact pattern.¹⁹² A tactically astute employer, rather than firing an employee for an outburst during protected conduct, might simply wait until the employee’s next performance review and then discharge them for generally having a bad attitude or not being a “team player.”¹⁹³

The General Motors Board did get one thing right—this is all far too complicated.¹⁹⁴ Our proposal places the burden squarely on the employer to argue that conduct was unprotected and unreasonable. We seek to simplify the question of protection by having it appear only in the employer’s burden. While our proposal would not necessarily resolve the issue of ambiguity, it would still clarify who bears the burden and how it should be raised. In practice, it is already customary for employers to argue why conduct is unprotected in addition to providing a Wright Line defense, both during the Regional investigation and at trial. Our test formalizes this practice and makes explicit to the adjudicator which party is ultimately responsible for proving lack of protection.

We thus propose that when the General Counsel establishes an initial showing of discrimination (and therefore an unfair labor practice), the burden should then shift to the employer (for remedial reasons) to show that it engaged in the challenged actions for legitimate reasons and “for cause” within the meaning of Section 10(c) of the NLRA. In evaluating this burden, the Board should consider the existence of a lawful policy applied toward the employee, a track record of enforcement of that policy toward similarly situated individuals or a showing that no such violations have occurred, documentation suggesting that the decision to implement the personnel action pre-dated the Section 7 activity, or a showing that the Section 7 activity was unprotected and unreasonable. Our test not only squares better with extant Title VII law, but it makes explicit the elements that 85 years of jurisprudence have evinced as relevant considerations when assessing the validity of an employer’s defense.¹⁹⁵

C. Second General Counsel Burden

Despite a long history of Board jurisprudence on the matter, the Wright Line test does not explicitly require the adjudicator to assess pretext. A significant minority of NLRB discrimination cases fail to mention the concept of pretext at all. On April 17, 2021, we ran a Westlaw search for all Board decisions citing to Wright Line (excluding decisions by Administrative Law Judges), which yielded 3,812 results. Of those results, 2,177, or 57%, included at least one variant of the root word “pretext.” Accordingly, it is incumbent upon the General Counsel to proffer pretext evidence, or upon an Administrative Law Judge or the Board to proactively cite pretext jurisprudence when applying Wright Line.

Curiously, the Board in the original Wright Line decision itself implicitly added a prong to analyze pretext. Though the employer’s defense was colorable at first blush, the Board ultimately ruled in favor of the General Counsel due to evidence of disparate treatment, departure from past practice, and a predetermined plan to discover a reason to fire the discriminatee.¹⁹⁶

Because pretext is expressly evaluated in McDonnell Douglas for Title VII cases, we see no reason not to embed a pretext analysis into the test for NLRA discrimination. To assist the adjudicator, our test makes explicit the type of information that might constitute pretext. We propose that if an employer meets its burden to show that the personnel action was taken for cause, then the burden should shift to the General Counsel a second time to show that the employer’s defense was pretextual because it was based on untrue factual claims or reasons that were not honestly believed. We now turn to the thorny issue of pretext and how we propose it should be analyzed.

Pretext has a long history as a marker of discrimination.¹⁹⁷ But it is an under-defined, inconsistently applied, and oft-overlooked factor in NLRA discrimination cases.¹⁹⁸ In Electrolux Home Products,¹⁹⁹ the Board amplified the ambiguity by stating that pretext alone is insufficient to constitute discriminatory motive or to overcome an otherwise satisfactory Wright Line burden. The Electrolux Board rejected traditional approaches to analyzing pretext, reasoning that:

When an employer has offered a pretextual reason for discharging or disciplining an alleged discriminatee, the real reason might be animus against union or protected concerted activities, but then again it might not. It is possible that the true reason might be a characteristic protected under another statue (such as the employer’s race, gender, religion, or disability), or it could be some other factor unprotected by the Act or any other law, which would be a permissible basis for action under the at-will employment doctrine.²⁰⁰

The Electrolux decision is at odds with a litany of Board decisions finding that while pretext cannot necessarily compel an 8(a)(3) finding,²⁰¹ pretext nonetheless calls the employer’s Wright Line defense into question and could even provide inferential evidence of discriminatory motive.²⁰²

The Board needs to do better, analytically speaking. Though the concept of pretext can be as variably interpreted as the concept of protection, efforts have been made to narrow the concept down to categories that help make sense of what pretext looks like and how it should factor in the analysis. Expanding upon Kathleen M. Kelly’s two categories,²⁰³ Michael Hayes identified three categories of pretext:

(1) The record makes clear that the facts underlying the employer’s proffered reason(s) for its challenged action did not exist,

(2) There is no support in the record that the facts underlying the employer’s proffered reason(s) did exist, or

(3) The record shows that the facts underlying the proffered reason(s) did exist, but the record shows that these facts were probably not the real reason for its challenged action.²⁰⁴

Hayes proposes that only the first and third categories should permit an inference of unlawful motive for the General Counsel’s initial burden because they imply deception or concealment at the employer’s peril.²⁰⁵ Hayes posits that the adjudicator should not infer unlawful motive in second-category cases where witness demeanor or deficiencies in record evidence are insufficient to prove unlawful motive, even though second-category pretext may factor into an overall determination that the employer failed to meet its Wright Line burden.²⁰⁶ Further, Hayes identifies several types of evidence commonly used to support pretext in third-category cases: departure from past practice,²⁰⁷ shifting or inconsistent reasons,²⁰⁸ and rushing to judgment by failing to investigate the circumstances.²⁰⁹

Our solution to the ambiguity of the pretext analysis is to explicitly mention the following factors in the test itself: shifting defenses, disparate treatment, particularized animus toward the Section 7 activity, white sheep/black sheep considerations, and evidence that the employer targeted or monitored the employee with the deliberate intent to find an excuse to engage in the personnel action. Some of our proposed criteria for pretext, like shifting defenses and disparate treatment, are so intuitive that one may argue that they are unnecessary to include. Once again, the lack of any mention of these criteria in the current test requires that the charging party be savvy enough to present evidence related to these criteria; we would prefer not to create unnecessary guesswork. More importantly, for adjudicators who may not accord much weight to shifting defenses or disparate treatment, embedding these concepts into the test ensures that these critical considerations not be given short shrift. While we agree with Hayes that departure from past practice and rushing to judgment raise suspicions, we believe that such factors should instead be used to assess the employer’s step-two burden because the employer itself is best equipped to address those questions. We thus do not include departure from past practice or rush to judgment in our list for pretext considerations. We will now explain the rationale for the factors that we chose to include in the list of pretext considerations.

1. Particularized Animus

The Board has long waffled on the extent to which statements exhibiting unlawful motive need be particularized. Until the Board’s 2019 Tschiggfrie decision,²¹⁰ anti-union comments privileged by Section 8(c) could nonetheless be used to establish discriminatory motive,²¹¹ and those comments did not require a particularized nexus to the discriminatee.²¹² As we have explained, we propose that generalized animus be sufficient to open the inquiry. However, it should go without saying that contemporaneous hostile statements by officers, agents, or representatives of an employer evince discriminatory motive even where an employer has proffered a colorable reason for the adverse action. It is the closest thing to a smoking gun available in NLRA discrimination cases, and the Board should treat it as such.

2. White Sheep/Black Sheep or Collateral Damage Considerations

White sheep/black sheep or collateral damage cases are among the hardest to prove. These involve innocent casualties who suffer discrimination so that the employer can claim that no disparate treatment occurred: fire the vocal union supporter pursuant to a policy, but also fire another poor soul pursuant to that same policy just to show that the union supporter was not singled out.²¹³ It is difficult to prevail on a white sheep/black sheep theory because the Board does not always have the resources, or the access, to dig deeper into the comparators proffered by an employer to meet a Wright Line burden. Indeed, we found only fifty Board cases that cited Majestic Molded Products v. NLRB, the first case to explicitly mention the black sheep/white sheep theory.²¹⁴ Because it is difficult enough to prove at trial that the black sheep suffered discrimination, let alone any of the white sheep, we suspect that most white sheep/black sheep cases get dismissed by Regional Offices of the NLRB and thus never make it to trial.

Though we do not pretend that our proposal would make these sorts of cases any easier to prove, we hope that embedding white sheep/black sheep considerations into the test would motivate NLRB agents and adjudicators to at least reckon with this unfortunate phenomenon. As stated by the Fourth Circuit, “[l]ayoffs intended to ‘discourage membership in any labor organization’ violate the NLRA, even if the employer wields an undiscerning axe, and anti-union employees suffer along with their pro-union counterparts.”²¹⁵ Similarly, the D.C. Circuit recently stated that “the Board has long held” that “an employer’s discharge of uncommitted, neutral, or inactive employees” either to cover up for discrimination against a targeted union-supporting employee or to discourage employee support for the union “violates Section 8(a)(3).”²¹⁶ Thus, the D.C. Circuit found that the knowledge requirement in Wright Line need not be particularized to the discriminatee so long as anti-union animus motivated the adverse actions, because the operative concern in the statute is preventing an employer from discouraging its employees’ interest in unionization, not protecting a single employee from retaliation because of their union sympathies qua individual employee.²¹⁷ We urge the Board to incorporate the D.C. Circuit’s rationale into the test itself.

3. Targeting or Monitoring

The final consideration that ought to be explicitly addressed when evaluating pretext is evidence that the employer targeted or monitored the employee with the deliberate intent to find an excuse to engage in the personnel action. As previously mentioned, the original Wright Line decision partially relied on targeting and monitoring to reject the employer’s initially colorable defense. The employer had claimed that it discharged the union activist for violating its policy against alteration or falsification of time reports.²¹⁸ Though the employer had produced evidence that the union activist had indeed made errors in his timesheets, the Board nonetheless found that the employer only discovered the employee’s errors because the plant supervisor ordered another supervisor to check on the activist—despite lacking cause to be suspicious.²¹⁹ Relying also on evidence of a departure from past practice²²⁰ and disparate treatment, the Board concluded that the employer engaged in a “predetermined plan to discover a reason . . . to rid the facility of a union activist” and found a violation of Section 8(a)(3).²²¹ Consistent with the Board’s original application in the Wright Line decision, we believe that the test should make clear that evidence of suspicious targeting or monitoring would ordinarily defeat an otherwise-plausible employer defense.

VIII. Application of the New Test in Practice

In this Part, we apply our test to a fictional scenario, and then to an actual NLRB case that illustrates the pitfalls of how Section 7 discrimination is typically analyzed. In both contexts, we show how our test would provide a more just result.

Before we delve into some examples of how our proposed test might play out, we should elucidate the process by which Unfair Labor Practice charges are investigated by the NLRB’s Regional Offices. Section 10(b) of the NLRA imposes a six-month statute of limitations for charges to be filed with the NLRB.²²² Any individual or organization may file such a charge, which tends to be a single-page document alleging the Section of the NLRA that was violated and that briefly describing the unlawful conduct. Charges are filed based on the location of the affected employees, and that location determines which of the 26 Regional Offices will process the charge. ²²³ Upon receipt by a Regional Office, the charge is docketed and assigned to a single NLRB agent, either a Labor-Management Relations Examiner (Field Examiner) or a Field Attorney. For cases alleging 8(a)(1) or 8(a)(3) discharges, especially those in union-organizing contexts, Regions typically strive to determine whether to issue a complaint within two calendar months.²²⁴

To initiate the investigation, the assigned NLRB agent contacts the involved parties to obtain a general understanding of the facts. Within two weeks, or even sooner in cases where an injunction under Section 10(j) may be implicated,²²⁵ a discriminatee is expected to provide a sworn affidavit to the agent and to submit documentary evidence such as discharge letters, pay stubs, prior discipline, and the employee handbook. If the affidavit suggests that the case might have merit, the agent will send the employer a Request for Evidence, a formal letter spelling out the allegations more specifically, requesting documents and responses to questions, and requesting that the employer fully cooperate by presenting its own witnesses for sworn affidavits. Full cooperation is rare, and the charged party generally submits a short position statement ignoring the detailed inquiries in the agent’s letter.²²⁶ Thereafter, the agent attempts to track down corroboration either through individuals controlled by the charging party, or through third parties—sometimes issuing investigative subpoenas to compel testimony by uncooperative parties.²²⁷ The agent will often provide the discriminatee a chance to rebut the employer’s defense before the case is presented to a panel of Regional managers for a decision. Panel composition differs by Region, but a panel will usually include at least the Regional Director, the investigating agent, and that agent’s first-line supervisor.

The Regional Director can render one of four decisions. First, they can find merit to the charge, which would initiate an attempt to settle before issuing any complaint for a hearing before an NLRB Administrative Law Judge. If a complaint is issued, the General Counsel becomes the complainant and thereafter acts in the public interest (not strictly the interest of the charging party).²²⁸ Second, the Regional Director can dismiss the charge, which provides a two-week window for the charging party to appeal the decision to the Office of Appeals in Washington, D.C. for a de novo review.²²⁹ Third, the Regional Director may defer the charge to the parties’ grievance-arbitration procedure as long as (1) the parties are covered by a collective bargaining agreement that provides for final and binding arbitration, (2) the dispute is covered by that agreement, and (3) the employer agrees to waive all timeliness requirements impeding the processing of the grievance under that agreement.²³⁰ Finally, for cases involving unique legal issues or relating to initiatives by the General Counsel to change extant law, the Regional Director may, and is sometimes required to, submit the case to the NLRB’s Division of Advice.²³¹ The Division of Advice, which ultimately answers to the presidentially appointed General Counsel, will render a decision after several months as to how the Region should proceed.

At any stage, the parties may enter into an informal Board settlement to remedy the charge almost identically to how an ALJ or the Board would remedy it through full adjudication. Further, the charging party may request withdrawal of a charge at any time and for any reason, including a private resolution between the parties that does not necessarily entail a traditional remedy.²³²

This process is unusual for a federal agency that enforces an employment-related statute. It involves a deep investigation by highly educated subject-matter experts before any formal litigation, and it precludes individuals from filing private actions in the courts. The entire process is cost-free for the discriminatee, including all litigation expenses. Our anecdotal experience informed by our dealings with counsel who practice in labor and employment law suggests that NLRB investigations are unusually expeditious yet can be quite thorough. To our knowledge, the NLRB is the only federal agency that, as a standard procedure, will take detailed sworn testimony from a charging party even before any assessment of the merits. In fiscal year 2020, the average time between the filing of a charge to a disposition was 73.8 days.²³³

In contrast, the Equal Employment Opportunity Commission (EEOC)—which enforces Title VII—initiates its process with an attempt toward mediation. If mediation is declined or unsuccessful, the EEOC takes an average of 10 months to investigate.²³⁴ Except in exceedingly weak cases or those for which there is no jurisdiction, or at the other end of the spectrum in the handful of cases that it litigates itself, EEOC investigations will normally result in a Notice of Right to Sue that authorizes the discriminatee to pursue a lawsuit in court, without the EEOC as a formal party.²³⁵ The EEOC thus has a significant financial interest in taking very few cases, because it may pass the costs of litigation onto individual charging parties. Unfortunately, the EEOC model appears to be more common than the NLRB model when it comes to the operations of other state and federal agencies that enforce employment-discrimination statutes.

A. Hypothetical Case of Industrial Industries, Inc.

Garfield Examinari is a Field Examiner in the NLRB’s Area 51 Regional Office. Examinari was recently assigned a charge alleging that Industrial Industries, Inc. discharged Rosie DeRiveter because of her support for United Riveting Riveters of America, in violation of Section 8(a)(3) of the NLRA. DeRiveter’s affidavit explains that her discharge came as a surprise because she was the employer’s top performer for over a decade, and that she only began receiving discipline after an organizing campaign during which she leafletted in the employer’s parking lot to promote unionization as a means of improving safety protocols on the assembly line. Since she began wearing union insignia, the employer disciplined her three times within the span of two months for placing widgets in the wrong receptacle. Corroborated affidavit testimony from her coworkers confirmed that this was common practice because the receptacle used by DeRiveter was better suited to the transfer of widgets to the next stage of the assembly line. After the fourth widget misplacement, the employer discharged DeRiveter pursuant to the fourth stage of its progressive disciplinary policy.

Thanks to advice by the experienced labor counsel at Lawyer & Attorney, P.C., none of the employer’s supervisors even so much as mentioned the union or the leafleting at the workplace. Instead, the employer hired consultants from Union Busters, LLC, a firm that specializes in subtle and covert tactics to identify prominent union supporters and disseminate right-wing and anti-union propaganda to their coworkers. But DeRiveter herself never interacted with the consultants, and the witnesses who did engage with them declined to cooperate in the investigation—out of fear of retaliation.

In its defense against the NLRB charge, the employer supplied Region 51 with a four-page position statement, the widget-placement policy, all of DeRiveter’s disciplinary documents, and two examples of employees who were discharged for widget-placement infractions several years ago. The employer failed to fully cooperate in the investigation, declining requests to take affidavits from key witnesses. The position statement contended that the employer lacked knowledge of DeRiveter’s union activity because employees routinely distributed leaflets in the parking lot regarding political campaigns and barbecues, including on the very day DeRiveter had been leafleting. None of those other employees had been disciplined. Further, DeRiveter herself admits to having used the wrong widget-receptacle, and the employer claimed to have video camera evidence thereof.²³⁶ There was no evidence of anti-Section 7 animus aside from the employer’s hiring of Union Busters, let alone direct evidence of unfair labor practices committed against DeRiveter, to establish that her union insignia or protected activity was a motivating factor in the decision to discharge her. And even assuming arguendo that the leafleting was a motivating factor, the employer contended that it met its Wright Line burden because it had a clear widget-placement policy, it followed a progressive disciplinary policy, and it had a record of disciplining employees pursuant to that policy.

Examinari’s docket contains twenty-five active cases, and DeRiveter’s is his weakest by far because of the employer’s Wright Line defense and the lack of explicit animus. He must make judgment calls as to which cases deserve the most resources, lest he waste taxpayer money by over-investigating cases that the Regional Director is unlikely to find meritorious under extant law. Thus, upon obtaining the employer’s defense, he submitted the case to the Regional Director with a recommendation to dismiss. The Regional Director dismissed the case, noting the weak evidence of knowledge, absence of evidence of particularized animus, and strong Wright Line defense. The union appealed the dismissal, but the General Counsel upheld the decision three months later.

Though the foregoing scenario might sound far-fetched, it is entirely plausible given the NLRB’s continued lack of resources and its push for more efficient case processing.²³⁷ Under the Trump administration, Regional offices faced precipitous cuts in staff due to early-retirement incentives and failure to fill vacancies, increasing investigating agents’ and trial attorneys’ workloads.²³⁸ From 2010 until 2019, the NLRB has experienced a 26% decline in staffing.²³⁹ As noted above, the NLRB has received flat funding of $274 million since 2014, amounting to a de facto decrease due to inflation.²⁴⁰ Given these resource constraints, a reasonable Regional Director might reserve only airtight cases for complaint issuance, despite the “preponderance of the testimony” standard imposed by the statute.²⁴¹ And, as we have explained, the oversimplicity of Wright Line might result in well-meaning Regions making determinations under an incomplete evidentiary record.²⁴²

How might DeRiveter have fared under our test? For one, her initial burden would have been perfectly fine, and thus Examinari would not have perceived her case as a weak one that could easily be overcome by any colorable Wright Line defense. She engaged in clear Section 7 activity and the employer had generalized, though not particularized, union animus. The weak evidence of employer knowledge of DeRiveter’s union activities would be less of a concern because knowledge would no longer be a required element of the initial showing. Accordingly, the RD would have found a violation of Section 8(a)(3), shifting the burden for remedial purposes.

Next, rather than simply submitting the case for dismissal upon receipt of the employer’s colorable defense, Examinari might have combed through the evidence to evaluate pretext. In so doing, DeRiveter may have satisfied her rebuttal burden by showing that the employer targeted or monitored her with the deliberate intent to find an excuse to undertake the managerial decision. Examinari would have had to probe more deeply into the comparator evidence, interrogating why the employer had not disciplined employees pursuant to the widget-placement policy for several years but suddenly began taking the policy seriously upon the initiation of the union campaign. The deeper investigation may have revealed that the supervisor who enforced the widget-placement policy several years ago had long left the company, after which the new supervisor actively encouraged subversion of the policy in order to maintain production targets. Thus, our test would have compelled a deeper investigation and healthy skepticism, and it might have precluded Examinari from swiftly recommending dismissal to the RD shortly upon receipt of a colorable Wright Line defense. DeRiveter would have gotten her day in court.

Needless to say, this hypothetical example addresses the cases that never see an Administrative Law Judge or the Board because of the unique means by which the United States processes unfair labor practices under the NLRA. We now turn to a recent real-life example of how the Wright Line test is applied by a judge and the Board, which demonstrates how our proposed test might have produced a more just result.

B. Case Study: Watco Transloading, 369 N.L.R.B. No. 93 (May 29, 2020)

Since October 2013, Watco Transloading has facilitated the transfer of railborne petroleum products at a Philadelphia refinery where it employed about twenty-one employees.²⁴³ In 2014, a Watco supervisor witnessed car-person Dennis Roscoe and engineer John D. Peters sign authorization cards in the parking lot outside the facility’s exit gates, alongside an organizer for United Steelworkers Local 10-1.²⁴⁴ The Administrative Law Judge found that Watco violated Section 8(a)(1) of the NLRA by promising benefits, such as better weather gear and higher wages to employees, and by buying employees lunch on a more frequent basis than it had previously.²⁴⁵ The conduct occurred mere days before an NLRB election.²⁴⁶ The Administrative Law Judge also found that the Employer had discriminated against employees in violation of Section 8(a)(3) and (1) by discharging Peters and Roscoe, suspending Roscoe for 14 days, and issuing two warnings to Roscoe.²⁴⁷ The union lost the election, it did not file objections, and thus it never became the employees’ certified representative.²⁴⁸ The Board adopted the Administrative Law Judge’s promise-of-benefits finding, reversed her discrimination findings, and found an additional 8(a)(1) violation for Watco’s solicitation of grievances. We now focus on each discriminatee in turn to illustrate how our proposed test might have operated in this case.

1. Judge’s Decision Regarding Peters

Peters was one of the employer’s initial hires and possessed a valuable skillset.²⁴⁹ On August 1, 2014, Peters was the subject of threatening, harassing, and disparaging texts by a coworker named Leroy Henderson.²⁵⁰ He complained about these texts to a supervisor, but Watco never investigated the situation.²⁵¹ On August 4, a coworker falsely informed the employer’s human resources department that Peters had repeatedly referred to him as a homosexual and had called him a homophobic slur.²⁵² Peters denied having made the remarks.²⁵³ On August 5 and 6, the employer completed an investigation into Peters’s alleged misconduct.²⁵⁴ On August 21, Watco witnessed Peters and Roscoe sign authorization cards.²⁵⁵ On August 26, Watco informed Peters that he had violated the employer’s sexual-harassment policy and would be discharged.²⁵⁶ Peters had no prior warnings for similar behavior, and he was discharged upon a first offense despite Watco’s progressive disciplinary policy.²⁵⁷ After he was discharged, Peters filed an internal appeal, prompting Watco to continue its investigation of the harassment allegations against him.²⁵⁸ The internal appeal was ultimately dismissed.²⁵⁹

The Administrative Law Judge found Peters’s discharge violated Section 8(a)(3) and (1) of the NLRA.²⁶⁰ First, she found that it was “perfectly clear” that Watco knew Peters signed union authorization cards to coworkers.²⁶¹ The Administrative Law Judge also emphasized the incredibility of Watco’s witnesses, who fashioned an improbable sequence of the events leading up to the discharge and failed to produce documentary substantiation of the process by which they decided to discharge Peters.²⁶² Watco’s witnesses gave self-serving testimony that the decision was made during a conference call a week before the discharge, but there were no emails, notes, or memoranda regarding what was discussed in this call.²⁶³ The Administrative Law Judge felt compelled to credit Peters over Watco’s witnesses, as none of the employees interviewed by the employer during its investigation testified at the hearing, including the alleged victim of Peters’s conduct.²⁶⁴ Further, the Administrative Law Judge found it suspicious that Watco would renew its investigation of Peters’s alleged harassment after-the-fact, given that Watco had claimed to have already conducted an investigation that was sufficient to warrant discharging Peters.²⁶⁵

The Administrative Law Judge did not mention any policy prohibiting Peters’s conduct, even though the record contained an employee handbook with a relevant provision on sexual harassment. Nor did the Administrative Law Judge address the fact that Watco did not produce similarly situated employees who had been discharged for comparable conduct pursuant to the same policy, or upon a first offense pursuant to some other policy.

2. The Board’s Decision Regarding Peters

More than five years after Peters was discharged, the Board reversed the Administrative Law Judge’s finding of discrimination. Like the Administrative Law Judge, the Board did not even address whether other similarly situated employees had been discharged pursuant to similar conduct or similar employer policies. The Board failed to consider the Judge’s points regarding Watco’s failure to abide by its own progressive disciplinary system, and it did not mention Peters’s clean disciplinary record. Rejecting the General Counsel’s and the union’s contentions that Peters was subject to disparate treatment given coworker Leroy Henderson’s text messages, the Board stated that even though the texts were “opprobrious,” neither the General Counsel nor the union excepted the Administrative Law Judge’s failure to find pretext or disparate treatment.²⁶⁶ Characterizing Henderson’s “angry, vulgar text messages” as “a largely incoherent ramble,” the Board found Henderson’s conduct incomparable to Peters’s sustained and repeated in-person remarks in the workplace regarding his coworker’s sexual orientation.²⁶⁷ While the Board conceded that the close timing between the union activity and the discharge warranted additional scrutiny, it accused the Administrative Law Judge of “impermissibly impos[ing] her own judgment as to how and when the Respondent should have conducted the investigation and discharge of Peters.”²⁶⁸ Thus, Watco’s “actions here were consistent with a bona fide effort to determine the validity of serious allegations against Peters,” and the Board dismissed the complaint allegation.²⁶⁹

3. Applying Our Test to Peters

How would Peters have fared under our proposed test? As we have discussed above, any case involving union activity that arrives on the Board’s docket will usually have a strong initial showing of discrimination under Wright Line given the rigidities of the investigation process prior to which a decision to issue a complaint is made. As such, we focus on how the employer burden and second General Counsel burdens in our test might have produced a different result.

Under our proposed test, Watco would have failed to meet its burden regarding Peters’s discharge. Watco had a lawful sexual-harassment policy that it invoked to discharge Peters. However, the record testimony, especially given the credibility resolutions of the judge, suggest that Peters did not, in fact, engage in the infraction for which he was accused. Watco produced no evidence of a history of enforcing its policy against similarly situated individuals, and it did not show that no such incidents had previously occurred. Nor was there evidence that the employer had taken affirmative steps toward implementing the managerial decision before the Section 7 activity, of which the Administrative Law Judge found that Watco was aware. Finally, because Watco did not produce evidence that Peters engaged in that activity during working time or in working areas, there was no evidence that the Section 7 activity was unprotected. Accordingly, the burden would not have shifted back to the General Counsel, and Peters would have been eligible for a full remedy under the Section 8(a)(3) violation that had been established from the initial burden.

4. Judge’s Decision Regarding Roscoe

Roscoe began working as a car-person for Watco in April 2014.²⁷⁰ On July 29, he handed Terminal Manager Brian Spiller a letter about the employer’s failure to promote Black car-persons, complaining that white coworker Mike Onaskanych was promoted noncompetitively to a lead position and that two of Onaskanych’s sons were hired to do the same work as Black employees despite their relatively lower experience.²⁷¹ Spiller and other car-persons met with Roscoe regarding his letter, and Roscoe was told that the employer had no obligation to post positions so as to make employees aware of vacancies.²⁷²

On separate occasions in early August, Roscoe saw Shift Supervisor Joe Ryder and Onuskanych smoking in an area that posed a safety hazard.²⁷³ Watco had a policy, posted on bulletin boards, prohibiting smoking in that area. Roscoe confronted Ryder, who told Roscoe that he was the boss and that Roscoe could not tell him what to do.²⁷⁴ Roscoe also reported the behavior to another supervisor and recommended that that supervisor issue a memo to employees reminding them to smoke only in the designated smoking hut.²⁷⁵ On August 13, Roscoe reported the incidents to the Safety Coordinator on the site, reported on the employer’s website that employees were smoking in unauthorized areas, and sent the employer an email regarding the “life-threatening and hazardous situation.”²⁷⁶ The employer subsequently posted a notice to employees reminding them of the policy, and it also spoke to the employees who had violated the policy.²⁷⁷

On August 15, Roscoe worked past his end time to repair a train car and to brief his relief on the next shift regarding other necessary repairs.²⁷⁸ Because Roscoe’s cell phone was in the trailer, he missed certain texts sent by his supervisor, Ryder.²⁷⁹ Ryder later ordered Roscoe to come to the supervisors’ trailer, where Ryder told Roscoe to go home because Roscoe’s relief had arrived and Ryder didn’t want Roscoe working overtime.²⁸⁰ Roscoe explained why he needed to stay in order to fix a pin and complete paperwork.²⁸¹ Ryder agreed, and Roscoe worked for about another hour.²⁸²

On August 21, Peters and Roscoe signed authorization cards alongside a union organizer.²⁸³ That day, Roscoe was given a warning for insubordination due to the overtime incident on August 15, and another warning due to the quality of his work.²⁸⁴ The Administrative Law Judge found that there was no evidence that the employer knew of his union activity before he received his warnings.²⁸⁵

On September 23, ten days before the scheduled representation election, Watco sent Roscoe home following disputes with the lead car-person and a supervisor.²⁸⁶ According to Watco, Roscoe had accused his coworker of performing oral sex on management, and said as much with obscene gestures, in exchange for reporting to work on an unscheduled day to work overtime.²⁸⁷ According to Roscoe, his coworker’s presence was unnecessary for the work that day, and Roscoe threatened to call human resources.²⁸⁸ After sending Roscoe home, the employer conducted an investigation including interviews with seven employees.²⁸⁹ Only three of those seven testified at the trial, including Roscoe and Peters, and only Roscoe was asked at trial about the events of September 23.²⁹⁰

On October 2, the day before the election, Roscoe was called into work and given a final written warning, a 14-day unpaid suspension, and a placement on a performance-improvement plan because of his inappropriate gestures of a sexual nature.²⁹¹ On October 3, the union lost the election by a vote of 13-3.²⁹² The union did not file objections, and Roscoe returned from his suspension October 7.²⁹³ About October 9, coworker Henderson complained that Roscoe pulled his car next to Henderson’s and started cursing and threatening.²⁹⁴ By email dated October 11, Watco discharged Roscoe.²⁹⁵

The Administrative Law Judge found that all of Watco’s adverse actions toward Roscoe violated Section 8(a)(3) and (1) of the NLRA. She rejected the employer’s Wright Line defense of the two August 15 warnings, finding that Roscoe’s complaints about race discrimination and smoking safety constituted protected concerted activity.²⁹⁶ She also found that Roscoe’s August 15 conduct was consistent with past practice in the facility and thus did not constitute insubordination or poor work performance.²⁹⁷ Animus was established because the employer promised benefits in exchange for voting against the union, as well as the suspicious timing of Roscoe’s suspension amid the election.²⁹⁸ The Administrative Law Judge found that Watco’s defense of the suspension was pretextual because Watco did not have a good-faith belief that Roscoe made the obscene sexual gesture, emphasizing Watco’s reliance on hearsay and failure to call the employee directly affected as a witness.²⁹⁹ Finally, the Administrative Law Judge found that there were three competing versions of the reasons for Roscoe’s discharge, no documents to substantiate any of the versions, and no testimony from two relevant witnesses regarding the October 9 events.³⁰⁰ Those facts, plus a finding that the discharge was tainted by the 14-day discriminatory suspension, led the Administrative Law Judge to find that the employer had not met its Wright Line burden regarding Roscoe’s discharge.³⁰¹

5. Board’s decision Regarding Roscoe

The Board reversed on all of the allegations regarding Roscoe.³⁰² Though it found his complaint about race discrimination to constitute protected concerted activity, it found a lack of animus against that activity because the Terminal Manager promptly discussed the race issues with coworkers and did not display animosity during that discussion.³⁰³ The Board found that Roscoe’s safety complaints were not concerted because there was no evidence that Roscoe discussed the matter with coworkers or contemplated group action.³⁰⁴ The suspension was consistent with Watco’s handbook provisions for the misconduct, which was substantiated by the employer’s investigation.³⁰⁵ Because of corroborated testimony that Roscoe cursed at Henderson and made threats, including a threat involving the welfare of his children, the Board found that the employer legitimately relied on a good-faith belief that Roscoe engaged in a dischargeable offense.³⁰⁶ The Board did not mention the employer’s progressive disciplinary policy or inquire as to whether employees were discharged for similar offenses. Nor did it substantively address the adverse inferences drawn by the judge as to witnesses who did not testify, despite clear Board law drawing adverse inferences in such circumstances.³⁰⁷

6. Applying Our Test to Roscoe

While Roscoe’s is a closer case due to his brief work history and arguably more serious misconduct, we nonetheless believe that our test would have resulted in an analysis that more thoroughly probed the material issues. For starters, the Board would not have been able to find that Roscoe did not meet his initial burden regarding his warnings on the basis of his smoking-related safety complaints not having been concerted. Those safety complaints at least implicate the concerns in Section 7 and therefore would satisfy the initial burden in our test, establishing a violation of Section 8(a)(1) and (3). Concededly, the employer could perhaps have met its second-stage burden regarding the discipline before the distribution of authorization cards by establishing a lack of concertedness for mutual aid and protection within the meaning of Section 7. Next, both the ALJ and the Board failed to mention the lack of comparator evidence and failed to consider the longstanding recognition that tempers run high amid the employee-employer power imbalance and thus the language of the shop is not the language of polite society.³⁰⁸ Finally, the ALJ summarily found Watco’s defenses pretextual and discredited Watco’s witnesses, yet the Board rejected such findings for questionable reasons. Our test would have required a thorough analysis of pretext that would have given the Board more information to consider. In sum, though Roscoe may not necessarily have received reinstatement and backpay under our test with all compositions of Board members, our test would likely have caused the case to have been litigated differently and would have at the very least, established violations of Section 8(a)(1) and (3). The record would have then contained richer material with which the Administrative Law Judge and Board could have engaged in a deeper analysis of the relevant remedial considerations.

IX. Responses to Potential Criticism

In this section, we respond to three critiques of our proposed test. The first critique is that our test would make it too easy for charging parties to initiate the NLRB process, saddling employers with frivolous expenses. Our response centers on Section 1 of the NLRA, arguing that increased costs are a small price to pay for the proper effectuation of the statute. Another critique is that our test is lengthy and cumbersome, undermining the doctrine of at-will employment by requiring employers to needlessly justify their entrepreneurial decisions. Our response is that it is not always possible to devise a test that is both true to the statute and succinct; Wright Line’s pithiness does not compensate for its inadequacy. Third, one could argue that our test is at odds with the Administrative Procedure Act (APA) and Supreme Court jurisprudence by requiring employers to proffer more substantial defenses. We argue that our test is consistent with the APA and goes no further than the limits of Section 10 of the NLRA. We will address each of these objections in turn, and then we will address a proposal from Professor Charles Morris and explain how it lacks teeth without an improvement to the Wright Line test.

A. Section 1 of the NLRA Justifies the Potential for Increased Litigation

Employers might object that our test lowers the bar for an initial showing of discrimination, which could result in more frivolous charges filed, more frivolous complaints issued, and therefore increased legal costs. However, charging parties who are aware of the NLRB process already file charges that lack merit, and it is not clear that they would possess such a nuanced understanding of labor law that they would file additional charges if our test were implemented.³⁰⁹ It is more likely that parties already err on the side of filing charges even when there is mere suspicion of Section 7 discrimination, irrespective of any considerations of how a Region will handle the charge and what the current Board law is.

At any rate, an increase in charge-filing rates would be consistent with the statutory purpose of the NLRA. Section 1 of the NLRA provides that it shall be “the policy of the United States to encourage collective bargaining and to guarantee employees the freedom of association.”³¹⁰ Expanding Section 7 rights, including through increased enforcement, is well within the scope of the statutory mission of the NLRB.³¹¹ The goal for our test is to bring to justice those employers who are already violating the plain language of the statute, not to penalize innocent bystanders. Even assuming arguendo that our test might increase the charge-filing rate, that would not necessarily entail a surge in frivolous charges. Therefore, our test would only result in increased legal costs for employers because they would no longer be able to easily escape the consequences of their already-discriminatory conduct.

B. We Strive for the Correct Test, Not One That is Expedient

Another objection is that Wright Line is an elegantly simple two-sentence test compared to our multi-paragraph analysis. Wright Line decisions can be rendered more quickly by adjudicators without getting tangled in the weeds that we would require those adjudicators to consider. Further, requiring employers to proffer more robust defenses might have them walk on eggshells not to violate the NLRA. Such would also incentivize excessive record-keeping in order to prevent false positives in situations where an employer is unable to meet its burden despite the true absence of anti-Section 7 animus. Accordingly, our test might disrupt the flow of commerce by causing employers to be more vigilant and cautious about hiring and firing, rendering obsolete the doctrine of at-will employment.

The reason this critique ultimately does not persuade us is, as noted above, that the NLRA is notoriously underenforced.³¹² Extant caselaw does not sufficiently deter employers from violating it. The prospect of additional 10(j) litigation, while promising from an enforcement standpoint, poses no additional deterrent because injunctive relief, being purely prospective in nature by definition, has no deterrent effect currently.³¹³ The only way to better encourage employers not to violate the NLRA is to adjust the economic incentives by making it more likely that unfair labor practices will be uncovered and cured.

Consider three hypothetical treatment options for a disease, none of which can be combined. Treatment A is inexpensive, painless, and 51% effective within weeks, but those for whom the treatment is ineffective invariably die. Treatment B is 90% effective, and the 10% for whom it is ineffective will not die despite continuing to exhibit symptoms, but the treatment takes several months and is expensive and painful. Treatment C is 75% effective, and the 25% for whom it is ineffective may either continue exhibiting symptoms or could—but will not necessarily—die. Relative to treatments A and B, treatment C is average in cost, duration, and painfulness. Now assume that the effectiveness-percentages of these treatments correspond to the rate at which true NLRA discrimination is captured by application of a given test.

We contend that treatment A is Wright Line, which may work more than half the time and with great efficiency, but leaves almost half of discriminatees with no recourse when it fails. Treatment B, on the other hand, would represent an overcorrection to Wright Line that is excessively burdensome to employers and creates “false positives” because it is unjustifiably forgiving to charging parties. Our test strives to be treatment C, which works the vast majority of the time without generating false positives and without adding significant financial cost, emotional burden, or delay.

The relative ease of applying Wright Line comes at the expense of accuracy. We do not think that this is a justifiable tradeoff for such important worker protections.

C. Our Test is Consistent with Supreme Court Jurisprudence

One final objection is that our test fails to accept extant law on at-will employment and the proper allocations of the burden of proof under the Administrative Procedure Act (APA). Section 7(c) of the APA states that “[e]xcept as otherwise provided by statute, the proponent of a rule or order has the burden of proof.” In Director, Office of Workers Compensation Programs, Department of Labor v. Greenwich Collieries,³¹⁴ the Supreme Court addressed whether Transportation Management³¹⁵ ran afoul of the APA by requiring the employer to satisfy a burden of persuasion even though the employer was not the moving party in the dispute. Greenwich Collieries clarified that § 7(c) of the APA is a burden of persuasion, not merely one of production, that is consistent with the Wright Line burden-shifting framework because the burden only shifts after the General Counsel itself has satisfied its own burden of persuasion.³¹⁶

So too with our test. The initial burden is a sufficient hurdle that would square with Greenwich Collieries for APA § 7(c) purposes. The key point is that the General Counsel’s burden is only to prove facts from which a tendency to discourage (or encourage, in rare cases) union activity can be drawn, consistent with the Board’s established understandings of employers’ economic incentives. Once that tendency is shown, the burden shifts to the employer to establish cause consistent with Section 10(c), only for remedial purposes.

The employer’s burden provides numerous avenues to escape liability that do not require excessive record keeping, including a showing that no similar policy violations have existed, or a showing that the decision to implement the personnel action pre-dated the Section 7 activity. Indeed, our test simply makes clearer the evidentiary means by which an employer’s motive for a personnel action—whether good, bad, or without reason—can be established. Finally, as we have previously argued, Section 10(c) of the NLRA already contemplates a just-cause requirement for remedial considerations. Our test would simply emphasize that long-ignored requirement.

D. Charles Morris’s Proposed Reforms

Pointing to the Railway Labor Act’s generally successful track record of blocking discriminatory discharges, Professor Charles J. Morris contends that increasing the injunctive activity of the NLRB would deter employers from discriminating against employees because of the costs and consequences of becoming a defendant in a federal district court. ³¹⁷ Morris proposes that the NLRB delegate the General Counsel the authority to independently and swiftly seek injunctions under Section 10(j) of the NLRA, eliminating some of the bureaucratic stumbling blocks that ordinarily impede petitioning a district court for a preliminary reinstatement order.³¹⁸

While we endorse Morris’s proposal, it will fail if the causation test for issuing a discriminatory-discharge complaint remains as strongly pro-employer as it currently is. No complaint, no injunction—it is that simple.³¹⁹ As we explored in Part 5 of this Article, over 90% of the NLRA’s discrimination claims are disposed of in the NLRB’s Regional Offices, who investigate charges and attempt to settle the dispute before any hearing.³²⁰ If the employer wins the battle because a Regional Director dismisses the case, the charging party’s only hope is to file an appeal in the hopes that, several months later, the General Counsel remands the case to the Region.³²¹ By then, the damage is done, and the General Counsel is unlikely to satisfy the “irreparable harm” prong that most jurisdictions require for a preliminary injunction. ³²² If the Regions use a flawed test to analyze discrimination, the charging party may never get its day in court. Put bluntly: the problem is even worse than Morris portrays it, because the limitations of Wright Line make cases where an employee was, in fact, discharged for union activity slip through the cracks.

X. Conclusion

We hope to have established that modifications to the Wright Line test are long overdue, better comport with the language of the NLRA, and honor relevant considerations in Supreme Court jurisprudence on employment discrimination. Our revisions would force adjudicators to engage in more thorough analyses that would better arrive at the truth, obviating lengthy and costly litigation. Contrary to others’ solutions for NLRA reform, our proposal would make its greatest impact during the investigative stage of the NLRB process. A lightened initial burden for NLRA discrimination will enable presently overlooked discriminatees to obtain relief pre-trial, which is when most NLRB cases are disposed of, and allow others to proceed to adjudication. While our proposal will not solve all the problems in the NLRA-discrimination arena, our experience as practitioners suggests that these modest revisions will prevent swaths of 8(a)(1) and 8(a)(3) violations from slipping through the cracks. Combined with necessary legislative reforms, our test would thrust the NLRA into the 21st century to fulfill its original objective of promoting collective bargaining.

* The analysis and views expressed in this article are solely our own and do not represent the views of the National Labor Relations Board, its General Counsel, or the United States Government. We give special thanks to former Board Member and former Regional Director Dennis P. Walsh for his feedback.

** Senior Labor-Management Relations Examiner, Region 4, National Labor Relations Board. Juris Doctor candidate (expected 2025), Temple University Beasley School of Law & Staff Editor at Temple Law Review; Master of Industrial & Labor Relations, Cornell University; Bachelor of Arts, New College of Florida.

** Supervisory Attorney, Contempt, Compliance and Special Litigation Branch, National Labor Relations Board. Juris Doctor, University of California-Hastings; Bachelor of Arts, Pomona College.