Download PDF

Under a system of government that diffuses power and makes institutional “[a]mbition . . . counteract ambition,”1 sudden power grabs by a federal agency are rare. Nevertheless, they do occur, particularly when they can be conducted “under the radar.” A lawsuit can be a very successful means for launching a power struggle without arousing much public attention. As Justice Scalia famously observed, most lawsuits involving the allocation of governmental power arrive in court “clad, so to speak, in sheep’s clothing: the potential of the asserted principle to effect important change in the equilibrium of power is not immediately evident, and must be discerned by a careful and perceptive analysis.”2

National Labor Relations Board v. Arizona, pending in federal district court in Arizona, is one such case. The issue on the merits is whether the National Labor Relations Act (“NLRA” or “the Act”) preempts an amendment to the Arizona constitution dealing with the right to vote by secret ballot in the election of bargaining representatives. This issue, however important it may be, is unlikely to attract widespread public attention. In fact, the controversy is unlikely to stir much interest even among lawyers. Preemption jurisprudence is a relatively narrow category of federal constitutional law, and the merits issue in National Labor Relations Board v. Arizona arises under only one branch of that jurisprudence. Moreover, the element of suspense is lacking, since the few observers who have taken notice of the preemption issue agree by and large that the National Labor Relations Board (“the Board”) has the better case on the merits.

This paper does not deal with the merits of the preemption issue in National Labor Relations Board v. Arizona. Instead, it focuses on several procedural arguments the Board advanced (explicitly or implicitly) to convince the court to decide the merits of its preemption claim. The procedural issues in the case warrant serious professional and public attention, partly because of their novelty, partly because the district court ruled in the Board’s favor on some of them,3 but perhaps most significantly because they reflect a disturbing conception of federal-state relations and of the proper allocation of authority within the federal government. If the Board’s positions on these procedural questions ultimately are sustained, National Labor Relations Board v. Arizona could have very troubling consequences for how governmental authority will be exercised in the future, not just over the states but over individual citizens, and in areas of the law far removed from labor relations.

Three procedural issues in the case deserve particular attention. First, the Board maintains that its request for a declaratory judgment presents a justiciable “case or controversy” for resolution by the district court. The arguments it advances in support of that contention burst the envelope of traditional standing doctrines, which generally prohibit judicial resolution of abstract legal questions at the request of a party with no legally cognizable interest of its own in the answer. Second, the Board effectively alleges that the very existence of the Arizona constitution’s “secret ballot” provision is a wrong of constitutional dimension, which means that the relief the Board seeks necessarily is a direct interference with a sovereign function of a state that resulted from its internal political process. Interference of this kind can be justified in certain carefully-defined circumstances. However, the litigating position adopted by the Board would expand those circumstances in a way that would diminish public accountability of government officials to an unprecedented degree. Third, in pursuing its case against Arizona, the Board has advanced arguments that call for expanding the bounds within which so-called “independent regulatory agencies” can act. Unlike the Attorney General or other heads of executive branch departments, members of the Board are not subject to full supervision by the President because they cannot be removed from office at will. In bringing this lawsuit, the Board nonetheless implicitly asserts that it can decide unilaterally when and how the Supremacy Clause should be “enforced,” and against which state laws. For the Board’s assertion of its power to be upheld, a court would be required to decide that a so-called “independent regulatory agency” can make discretionary policy decisions pitting the national government against a state government without express congressional authorization and despite the President’s limited authority to control how that power will be employed.

The Board clearly grasps the potential significance of its litigating position for augmenting the power of federal regulatory agencies and shielding the exercise of that power from control by an elected official. In its papers filed in opposition to Arizona’s motion to dismiss the complaint, the Board describes itself not only as an “independent agency of the United States,” but also as “a sovereign federal agency.”4 The Board presumably understands what sovereignty means, and that there can be only so much of it to go around in any given country. It must therefore also realize that judicial recognition of the Board’s asserted status necessarily would alter the division of powers, the separation of powers, or both.

Despite the extraordinary nature of the Board’s procedural arguments, it is fair to ask why any of those arguments matter. The traditional answer is that the Framers thought individual liberty would be protected both by federalism and by the Constitution’s separation of federal powers among three distinct branches of government, one of which would be under the control of a single individual.5 The Supreme Court has elaborated on this thought by explaining the linkage among three concepts: protection of all the political processes that are necessary for constitutional federalism, the principle of accountability underlying the structural norms ordained by the Constitution, and the promotion of individual liberty.6 The Board’s litigating posture in National Labor Relations Board v. Arizona asks the federal courts to ignore each of these concepts.

Background and Context

To fully appreciate how the Board’s procedural arguments in the case would enhance the powers of independent agencies and pose a threat to the individual liberties of citizens, it is first necessary to appreciate the extraordinary novelty of the Board’s litigating position. That appreciation, in turn, requires some background.

The National Labor Relations Act of 1935 established certain rights and responsibilities in connection with employment in the private sector. One such right springs from Sections 7 and 9, which taken together generally require that a private-sector employer bargain over the terms and conditions of employment with a representative of his or her employees if the representative was “designated or selected for the purposes of collective bargaining by a majority of the employees in a unit appropriate for such purposes,” and that such a representative will be the exclusive agent for bargaining purposes.7 The Section 7 obligations of employers are generally enforceable under Section 8(a)(5), which makes it an “unfair labor practice” as defined in the Act for an employer “to refuse to bargain collectively with the representatives of his employees, subject to the provisions of section 159(a).”8 Thus, a determination of what is sometimes called “majority status” logically precedes a determination of a duty to bargain.

The ordinary means for ascertaining whether a majority concurs in a proposition is to hold a referendum of some kind, usually involving each individual’s expression of his or her sentiments, i.e., voting. Traditionally, the Board has expressed a decided preference for secret ballot elections conducted under the supervision of its agents.9 At the same time, there also has been something of a running (if syncopated) dialogue at the federal level regarding the extent to which a labor organization’s majority status might be or even must be recognized in the absence of an election under the auspices of the Board.10 The Board’s position is that Section 7 of the Act affords a second path to the consequences of majority status, which it describes as “voluntary recognition [by an employer] based on other convincing evidence of majority status.”11

Cards purporting to authorize representation can be signed under a variety of circumstances, including home visitation by union organizers.12 Thus, one driver of the debate over the “second path” described in the Board’s Opposition to Arizona’s motion to dismiss is concern over whether card signing results as reliably as secret ballot elections in the authentic expression of the individual’s preference as he or she immediately perceives it.13

Over the decades, the mutual pushback between the Board and Congress changed the line of scrimmage but did not result in a victory for one school of thought over the other.14 For example, Congress amended Section 9 in 1947 to provide that “[i]f the Board finds upon the record of [an unfair labor practice] hearing that . . . a question of representation exists, it shall direct an election by secret ballot and shall certify the results thereof.”15 Based in part on the language of this amendment, the Supreme Court held in Linden Lumber Division, Summer & Co. v. NLRB that “unless an employer has engaged in an unfair labor practice that impairs the electoral process, a union with authorization cards purporting to represent a majority of the employees, which is refused recognition, has the burden of taking the next step in invoking the Board’s election procedure.”16

Yet, notwithstanding the 1947 amendment of Section 9(c) and the decision in Linden Lumber, the Board has determined that majority status can be found in some scenarios based on the presentation of authorization cards bearing signatures equal to 50% plus one of the number of individuals in a bargaining unit. Arguably, this represents a dilution of the requirement for “convincing evidence” of majority status that the Board invoked in its papers in National Labor Relations Board v. Arizona. Elsewhere in its Proposed Opposition to the Motion to Dismiss in National Labor Relations Board v. Arizona, for example, the Board seems to blur the distinction between “evidence” and “convincing evidence” by asserting that the Section 7 right to representatives of their own choosing is enforceable “[i]f private sector employees can persuade their employer to recognize their choice of a representative on the basis of evidence of majority status.”17 A standard based on “evidence” means less than one might think, because an employer need not actually adjudicate majority status based on “evidence,” whether or not accompanied by persuasive argument regarding how to interpret or weigh the evidence. To the contrary, an employer and a union can agree in advance that majority status will be recognized based on the union’s accumulation of signed cards. The Board has generally held that neutrality and card-check agreements between a union and an employer are enforceable.18 As a result, combined neutrality and card-check agreements have eclipsed elections as a means of organizing new bargaining units.19

* * *

On November 2, 2010, against this general backdrop, voters in four states (Arizona, South Carolina, South Dakota, and Utah) adopted a “secret ballot” amendment to their state constitutions, each worded somewhat differently from the others. The amendment to Arizona’s constitution provides that “[t]he right to vote by secret ballot for employee representation is fundamental and shall be guaranteed where local, state or federal law permits or requires elections, designations or authorizations for employee representation.”20

On January 13, 2011, the Board’s Acting General Counsel wrote to the Attorney General of each of the four states “to apprise [them] of the National Labor Relations Board’s conclusion that a recently-approved amendment to [the state’s constitution] conflicts with the rights afforded individuals covered by the National Labor Relations Act . . . [and] . . . to explain the Agency’s position and to advise you that I have been authorized to bring a civil action in federal court to seek to invalidate the [a]mendment.”21 In a Fact Sheet issued on January 14, 2011, the Board took a slightly more modest position, stating that on January 6, 2011, it had authorized the Acting General Counsel to bring lawsuits against the states to enjoin the application or enforcement of the states’ secret ballot amendments “insofar as they conflict with the federal rights of private sector employees to designate a union to represent them.”22

The Attorneys General responded to the Acting General Counsel in a single letter dated January 27, 2011. They flatly rejected the invitation to concede that portions of their state constitutions were invalid, and pointed out ways in which the substance of the secret ballot amendments was consistent with the Board’s interpretation of the NLRA and with the freedom of association guaranteed by the First Amendment.23 In response, the Board’s Acting General Counsel offered to hold discussions at the staff level on the issues.24 However, these discussions came to an immediate and abrupt halt when the Acting General Counsel’s staff refused to discuss the merits of the Board’s position in the absence of a confidentiality and non-disclosure agreement.25 The Attorneys General made clear that they would not hold discussions regarding the validity of state constitutional provisions behind the backs of their fellow-citizens (to say nothing of their governors and state legislatures).26

On April 22, 2011, the Board’s Acting General Counsel wrote to the Attorneys General again, this time to say that he had “directed [his] staff to initiate lawsuits in federal court seeking to invalidate Arizona Constitution Article 2 § 37 and South Dakota Constitution Article 6 § 28 as preempted by operation of the NLRA . . . and the Supremacy Clause. . . .”27 However, the Board followed through on the Acting General Counsel’s threatened legal action by singling out Arizona as the first target. On May 6, 2011, the Board filed its complaint “seek[ing] a declaratory judgment . . . that Arizona Constitution Article 2 § 37 . . . conflicts with the rights of private sector employees under the National Labor Relations Act . . . and is therefore preempted.”28

The Standing Issue

Whether a state law is preempted by federal law is an abstract question unless and until some regulated party is actually affected by the alleged conflict between federal and state law. For this reason, it is one thing for a federal agency to announce its views on whether a state law is preempted by federal law, but quite another to seek the endorsement of those views by a federal court.

In fact, the Constitution imposes general limitations on the circumstances in which federal courts can rule on a matter of law, and some of those limitations apply to any would-be plaintiff, including a federal agency. To issue a ruling deciding a contested issue of law (or of fact, for that matter), a federal court first must have jurisdiction over a “case or controversy,” i.e., an actual dispute between parties whose legally cognizable rights and/or obligations with respect to each other actually might be altered by how the court decides the legal issue presented to it.29

Although the district court ruled against Arizona on the standing concerns it raised in its motion to dismiss, a powerful argument can be made that National Labor Relations Board v. Arizona does not meet the “redressability” component of the traditional standing doctrine. The Board and Arizona will stand in precisely the same place with respect to their own labor law rights and obligations vis-à-vis each other regardless of any possible outcome in the Board’s lawsuit. No matter what the outcome of the case may be on the merits, the district court’s decision cannot determine Arizona’s rights and obligations in its capacity as an employer under the Act, because the term “employer” is defined in the NLRA to exclude states.30 Nonetheless, the district court held that “[a] declaratory judgment that Article 2 § 37 is preempted would redress plaintiff’s injuries by rendering the amendment unenforceable and affirming plaintiff’s exclusive power to enforce § 7 and § 8 of the NLRA.”31 However, as a general rule, “[i]t is a violation of due process for a judgment to be binding on a litigant who was not a party or a privy and therefore has never had an opportunity to be heard.”32 Thus, a declaratory judgment in National Labor Relations Board v Arizona to the effect that its secret ballot amendment is preempted will not by its own force bar a single employer or employee from invoking it in any subsequent case.33

Despite the seeming futility of the judgment the Board seeks in this case, it would be a mistake to dismiss the Board’s decision to sue as harmless. That decision is an assault on important principles that keep the power of the federal judiciary, legislature, and executive separated. To be sure, one important purpose of the “case or controversy” requirement is to limit the authority of the judiciary alone by preventing the federal courts from deciding abstract generalities outside the context of adjudicating specific rights between specific parties.34 However, it also serves three broader purposes, each of which promotes the ideals of self-government by avoiding instances in which distinct constitutional powers might be blended, thereby reducing the accountability of elected officials. First, it deters the political branches from running to the judiciary for political cover or moral reinforcement, for example by seeking an advisory opinion on the constitutionality of pending legislation, thereby closing a route by which elected officials might escape accountability to the voters. Second, it limits the circumstances under which the judiciary might accept an invitation to shut down popular or political debate before a majority has coalesced around a specific proposal or course of action, thereby forestalling at least one means by which the processes of representative democracy might be short-circuited. Third, it prevents the courts from deciding what Chief Justice Marshall called a “mere question of right” at the behest of a petitioner who requests the judiciary “to control the legislature” of a state, a dubious enterprise that Chief Justice Marshall observed “savours too much of the exercise of political power to be within the proper province of the judicial department.”35

National Labor Relations Board v. Arizona does not arise from a “case or controversy” under established principles of law, because the Board’s own rights and obligations are not at stake in the case.36 In fact, the Board essentially admits as much by arguing that it has standing to vindicate the federal statutory rights of workers in Arizona under a doctrine called “parens patriae,” under which a sovereign can bring suit to assert the rights of its citizens.37

Endorsing the Board’s parens patriae argument would extend that doctrine far beyond its current or historical boundaries, because the Board does not have “citizens,” i.e., persons whose interests it is authorized to represent as against other governments. At best, the Board has various “constituencies,” but only in the sense of that word when used to describe interest groups. In any event, the Board itself does not take seriously the notion that it is trying to vindicate the rights of individuals employed in Arizona: it has opposed a motion to intervene in the case by dozens of such individuals on the grounds that the State of Arizona adequately represents the interests of its citizens.38

There is a second way in which the Board’s parens patriae argument could set a dangerous precedent. The NLRA establishes a wide range of rights and obligations on the part of employers, employees, and employee organizations. These rights and obligations are cast in general terms, which means that when it comes to the exercise of these rights by their holders, interests may differ. For example, the Act establishes an individual right to refrain from forming, joining, or assisting a labor organization.39 It also establishes a right to engage in “concerted activities,” including those that are alternatives to collective bargaining, for “mutual aid or protection.”40 It seems to follow that employees have a right to collaborate in resisting the selection of a supposed “bargaining representative” if they consider the method of selection illegitimate.

It is impossible to act in any representative capacity for a group consisting of members with conflicting interests unless the representative is authorized to ignore or subordinate (that is, to sacrifice) the interests of certain members of the group in favor of others. Thus, the Board’s parens patriae argument necessarily entails the assertion that it may determine which of various abstract statutory rights to protect at the expense of others (a process that would ultimately call for determining that one portion of its organic statute is not to be advanced in light of another).41 A federal regulatory agency endowed with this authority genuinely would be “sovereign” as opposed to merely “independent”—and not in a good way.

Separation of Powers and the Unitary Executive

The implications of the Board’s parens patriae argument serve as an introduction to a separation of powers issue implicit in the Board’s litigating posture. This issue becomes apparent only when one realizes that an agency not directly accountable to the President has brought an action against a state seeking to invalidate a particular state law as repugnant to the Federal Constitution without any established statutory authorization to bring such a suit. Viewed in that light, the assertion of agency power underlying the complaint in National Labor Relations Board v Arizona is genuinely revolutionary because it can be justified, if at all, only by some non-statutory authority to perform the quintessential executive function—that is, to “see” that the Supremacy Clause “be faithfully executed.” The Board has no such authority, nor could it.

The NLRB’s statutory authority to litigate is confined to two areas. Under various subsections of the act, it can sue to enforce its orders and/or to prevent an unfair labor practice, and it can seek enforcement of its subpoenas in a federal court.42 Since a state is not a labor organization, a state cannot commit an unfair labor practice as defined by Section 8(b), the Section relating to unfair labor practices by labor organizations. Moreover, since the term employer is defined in Section 2(2) to exclude the states, a state cannot commit an unfair labor practice as defined in Section 8(a), the Section relating to unfair labor practices by an employer. Thus, it does not appear that the NLRB has statutory authority to bring litigation of the kind contemplated here.43

But imagine that the Board had such a power. Its exercise necessarily would involve discretion regarding the enforcement of federal law. “The authority to bring such suits [i.e., suits to enforce federal law as parens patriae] includes the discretionary authority not to bring them, if the responsible officers of the government are of the opinion that a suit is not warranted or would be of disservice to the national interest.”44 The exercise of discretion in the enforcement of the law is traditionally thought of as a quintessentially executive function. Indeed, National Labor Relations Board v. Arizona involves at least two individual examples of the Board’s exercise of executive discretion. First, it chose to pursue its case against Arizona but not against any of the other states it threatened to sue. Second, it chose to promote what it perceives to be the interests of Arizona employees who might wish to have a bargaining representative recognized through alternatives to secret ballot elections and to oppose the interests of Arizona employees who have a right under the Act not to engage in any concerted activity. Yet federal executive power involving such a degree of discretion may be wielded only by the President or an executive officer of the United States, and an executive officer of the United States must be removable at will by the President or by someone who the President can remove at will.45

To be sure, so-called “independent regulatory agencies” are permitted to play a role in the administration of federal law, despite the President’s inability to remove the heads of those agencies at will. However, this exception to the normal rule for the exercise of executive power applies only to the conclusion (or at any rate the polite fiction) in Humphrey’s Executor that what the independent agencies do is entirely “quasi-legislative” or “quasi-judicial.”46

Judging from the majority opinion written by Chief Justice Roberts in Free Enterprise Fund v. Public Co. Accounting Oversight Board, the Supreme Court may be losing its enthusiasm for the reasoning behind Humphrey’s Executor. To begin with, Chief Justice Roberts leaves no room for doubt that the President’s removal power stems from the Faithful Execution Clause, not solely from the Appointments Clause. Moreover, the majority opinion in Free Enterprise Fund studiously avoids even using the terms “quasi-legislative” or “quasi-judicial” except in quotations.47 In any event, to recognize that an independent agency’s assertion of discretionary authority to seek declaratory judgments regarding the constitutionality of state laws is at best constitutionally dubious, it suffices to observe that no court has ever tried to justify agency action on the grounds that it was merely “quasi-executive.” To do so would be to diminish the authority (and therewith the accountability) of the only federal officeholder whose authority is conferred by the people of the United States as a whole. A concept even remotely approaching a diminution of that authority by transferring some part of it to an agency has no place in the constitutional design.


And so we come to the third constitutional novelty in the Board’s litigating position. The Board implicitly asserts that the very existence of Arizona’s secret ballot amendment gives rise to a “case or controversy” and triggers the Board’s authority to approach a federal court for declaratory relief. This assertion is based in part on the possibility that the amendment might be invoked affirmatively or defensively to support or defeat a claim. However, the assertion also is based on the Board’s contention that “the enactment of the Amendment has a . . . chilling effect on the free exercise of organizational rights.”48

In a letter dated January 13, 2011, to the Arizona Attorney General, the Acting General Counsel’s office elaborated on its concerns about the potential chilling effect of the Arizona amendment. The letter makes clear that the Board’s concern applies not only when the amendment became law, but even before it became law:

I understand that the Amendment adopted by the voters in November is not technically in effect and must still be proclaimed by the Governor of Arizona. A.R.S. Const. Art 4 Pt. 1 § 1(5). Accordingly, I am hopeful that, after a review of the NLRB’s legal position, Arizona might be willing to take voluntary measures to ensure that the Amendment will not be proclaimed, and that the public will be so notified. . . .

In light of the significant impact of this Amendment, I request that any response to this letter on behalf of Arizona be made within two weeks. Absent any response, I intend to initiate the lawsuit.

In similar situations, where offending enactments have not yet ripened into actual enforcement actions, the courts have nonetheless permitted suits to bar their enforcement where a danger exists that public knowledge of the provision may result in “self-censorship; a harm that can be realized even without an actual prosecution.” . . . That principle is applicable here since it is foreseeable that widespread public knowledge of the Amendment will deter some employers from granting voluntary recognition or abiding by their commitments to recognize a union on the basis of a card check.49

On a chilling effect theory as expansive as the one advocated by the Board, there would seem to be no reason in principle why the Board could not seek injunctive relief to preclude a referendum on a state constitutional amendment or to preclude legislative action to propose such an amendment. (Ripeness considerations might convince a court not to proceed in such a case, but ripeness is generally regarded as a prudential doctrine, not a constitutional one, so the court would have the authority to decide whether or not to stay its hand.) If the courts endorsed this model of agency authority, the political processes of state and local governments would continue solely at the sufferance of a host of federal agencies—some accountable to an elected official, others not—each with what amount to censorial powers, including the power of discretionary interdiction. The Framers declined to give that power even to the federal courts. There can be no doubt that they would have denied it to an “independent agency.”

Republican Sovereignty

The Board is right about one thing: in the last analysis, National Labor Relations Board v. Arizona is a case about sovereignty. For those who believe that the residual sovereignty of the states includes certain immunities from federal interference in their political and/or governmental processes,50 the case represents either a new threat or a new opportunity. Which it turns out to be depends at least in part on whether the courts will recognize that the case raises both a grave federalism issue and an equally important separation of powers question. Recognizing that National Labor Relations Board v. Arizona involves both of the basic structural underpinnings of republican democracy could lead a court to decide the case on the basis of the general accountability principles that are the constitutional postulates of both federalism and the separation of powers. Those principles of accountability are vital to the sovereignty of republics.51

One reason to believe courts are basing their decisions on these principles can be found in a recent ruling from a district court, which coincidently also deals with sovereignty as a limitation on the Board’s authority. Chickasaw Nation v. National Labor Relations Board52 arose from two unfair labor practice charges brought by the Board against a federally-recognized Indian tribe, with respect to a casino operated under the auspices of the Chickasaw Nation’s executive branch.

A federally-recognized Indian tribe is “a domestic, dependent nation” under federal law, and therefore a type of sovereign.53 On that ground, the Chickasaw Nation brought an action in federal court alleging that the NLRA has no application to it because of its sovereign status as a tribe, and moved for injunctive relief to restrain the Board from conducting further proceedings on the unfair labor practice charges.54 The plaintiff argued that allowing the Board to exercise jurisdiction over the Chickasaw Nation would unlawfully interfere with its inherent powers as a sovereign, including its right to make its own laws and be governed by them.55

The Board’s principal defense did not engage the sovereignty issue directly. Instead, it relied on the “primary jurisdiction” doctrine, under which the federal courts decline to exercise their power to determine in the first instance whether a given controversy falls within the scope of an independent agency’s adjudicatory purview.56 In effect, the Board argued that the district court should decline to render any judgment and defer to the Board’s determination of the scope of its own authority with regard to the unfair labor practice charges against the casino operation.57 In support of this argument, the Board pointed to the provision of the NLRA which the Board urged gives it “[power] to prevent any person from engaging in any unfair labor practice . . . [unaffected] by any other means of adjustment or prevention . . . established by . . . law.” 58

The district court acknowledged that the NLRA’s grant of the power to prevent unfair labor practices had been interpreted to limit the authority of federal district courts to enjoin proceedings before the Board.59 However, the court held that federal regulatory schemes such as the NLRA do not apply to tribal governments exercising their sovereign authority absent express congressional authorization.60 It also found that the NLRA makes no specific reference to tribes and the legislative history does not indicate an intention to abrogate the sovereignty of recognized tribes by subjecting their executives to regulation under the Act.61

Chickasaw Nation v. National Labor Relations Board will not eliminate all agency over-reaching, but, along with National Labor Relations Board v. Arizona, it is an opportunity to establish a limitation on such conduct. To be sure, the limitation would come into play only in the rare event that agency action threatened an immunity that is a component of the residual sovereignty of a state or of the provisional sovereignty of a federally-recognized tribe. Nonetheless, for close to a century, it has been regarded as a major achievement to establish any limitation on the power of an independent agency. Moreover, whether the states’ status as sovereigns is a limitation on the power of independent agencies is merely a specific form of a more general (and more pressing) question: whether liberty-reinforcing and accountability-reinforcing constitutional norms are limitations on the extent of otherwise ungoverned agency power.


* J.D. Harvard Law School, 1980; Shareholder, Ogletree, Deakins, Nash, Smoak & Stewart, P.C. The author gratefully acknowledges the research and editorial assistance of his colleagues, Ashley Prickett Cuttino (J.D., University of South Carolina Law School, 2002), and James R. Silvers (J.D., Charleston School of Law, 2009). The opinions expressed in this paper are the author’s alone and neither reflect any opinion nor are intended as the expression of any opinion on the part of Ogletree, Deakins, Nash, Smoak & Stewart, P.C., or any of its attorneys or staff.





1  The Federalist No. 51 (James Madison).

2  Morrison v. Olson, 487 U.S. 654, 699 (1988) (Scalia, J., dissenting).

See Order, Nat’l Labor Relations Bd. v. Arizona, No. 2:11-cv-913, Dkt. No. 18, at 2-5, and 8-10.

See National Labor Relations Board’s Opposition to Motion to Dismiss, Nat’l Labor Relations Bd. v. Arizona, No. 2:11-cv-913, Dkt. No. 13, 2-40, at 3, 16 [hereinafter Proposed Opposition] (emphasis added). The Board’s Proposed Opposition exceeded the page limitations applicable in the Arizona federal court. The Board lodged the Proposed Opposition with the court clerk (who maintains it as part of the docket in the case) and submitted a motion for leave to file an Opposition exceeding the court’s page limitations. Although the Board’s motion was denied, the Proposed Opposition spells out the Board’s arguments in more detail than the shorter brief it ultimately filed. For that reason, this paper relies to a great degree on the Board’s Proposed Opposition.

Printz v. United States, 521 U.S. 898, 920, 922-23 (1997); U.S. Term Limits v. Thornton, 514 U.S. 779, 838 (1995); New York v. United States, 505 U.S. 144, 168-69 (1992).

See, e.g., Bond v. United States, 131 S. Ct. 2355, 2364 (2011) (“The Framers concluded that allocation of powers between the National Government and the States enhances freedom, first by protecting the integrity of the governments themselves, and second by protecting the people, from whom all governmental powers are derived.”); Free Enter. Fund v. Pub. Co. Accounting Oversight Bd., 130 S. Ct. 3138, 3154 (2010) (holding that where an executive officer of the United States is removable only for cause and only by other executive officers protected from removal at will by the President, the latter “can neither ensure that the laws are faithfully executed, nor be held responsible for [the officer’s] breach of faith”).

7  29 U.S.C. § 159(a).

Id. § 158(a)(5).

Cf. Nat’l Labor Relations Bd. v. Gissel Packing Co., 395 U.S. 575, 596 (1969) (“The most commonly traveled route for a union to obtain recognition as the exclusive bargaining representative of an unorganized group of employees is through the Board’s election and certification procedures under § 9(c) of the Act (29 U.S.C. § 159(c)); it is also, from the Board’s point of view, the preferred route.”). By contrast, the preference for elections has come under sharp academic criticism. See, e.g., Joel Dillard & Jennifer Dillard, Fetishizing the Electoral Process: The National Labor Relations Board’s Problematic Embrace of Electoral Formalism, 6 Seattle J. Soc. Just. 819, 819 (2008).

10  See Sheila Murphy, A Comparison of the Selection of Bargaining Representatives in the United States and Canada: Linden Lumber, Gissel, and the Right to Challenge Majority Status, 10 Comp. Lab. L.J. 65, 69-70 (1988) (chronicling decisions before 1947).

11  Proposed Opposition, supra note 3, at 2.

12  See, e.g., Congressional Documents, Senator Specter Speaks on Employee Free Choice Act/Card Check (Mar. 24, 2009) (“Testimony shows union officials visit workers’ homes with strong-arm tactics and refuse to leave until cards are signed.”).

13  See, e.g., Nat’l Labor Relations Bd. v. Village IX, Inc., 723 F.2d 1360, 1371 (7th Cir. 1983) (“Workers sometimes sign union authorization cards not because they intend to vote for the union in the election but to avoid offending the person who asks them to sign, often a fellow worker, or simply to get the person off their back. . . .”). For a survey of the arguments pro and con, see A. E. Eaton & J. Kriesky, NLRB Elections versus Card Check Campaigns: Results of a Worker Survey, 62 Indus. & Lab. Rel. Rev. 157 (2009).

14  The Board itself has changed its views regarding the reliability of non-electoral expressions of an employee’s preference. In Dana Corp., 351 N.L.R.B. 434, 439 (2007), the majority of the Board found that “There is good reason to question whether card signings . . . accurately reflect employees’ true choice concerning union representation,” and held that for a period of forty-five days following an employer’s voluntary recognition of a bargaining agent, employees could file a decertification petition supported by a showing of interest by 30% of the bargaining unit. Four years later, the Board overruled Dana Corp., based in part on the assertion that empirical evidence had demonstrated that the majority opinion in Dana Corp. regarding the reliability of card-check “was wrong.”  Lamons Gasket Co., 357 N.L.R.B. No. 72, 2011 WL 3916075 (N.L.R.B.) *6 (Aug. 26, 2011). 

15  29 U.S.C. § 159(c)(1).

16  419 U.S. 301, 310 (1974).

17  Proposed Opposition, supra note 3, at 2 (citing NLRB v. Creative Food Design, 852 F.2d 1295, 1297-98 (D.C. Cir. 1988); NLRB v. CAM Indus., Inc., 666 F.2d 411, 414 (9th Cir. 1982). The difference between “evidence” and “convincing evidence” is particularly important because “evidence” is a term of art referring to anything that tends to make an assertion more likely or less likely to be true, i.e., to make the probability of the proposition’s accuracy different to any degree whatsoever. Cf., e.g., Fed. R. Evid. 401 (defining relevant evidence). Despite appearing to say in its Opposition that mere evidence can support a determination of majority status, the Board obviously recognizes that evidence of “majority support” must be “reliable.” Complaint, ¶ XII.

18  Joseph Z. Fleming & Daniel B. Pasternak, Mutuality Agreements: Innovative Approaches to the Use of Neutrality Agreements—A Unique Proposal for Compromise, in Employment and Labor Relations Law for the Corporate Counsel and the General Practitioner 5 (Am. Law Inst. ed. 2007); see Kroger Co., 219 N.L.R.B. 388 (1975) (seminal case concerning card-check recognition agreements); SL DC Mgmt., LLC d/b/a Hotel Del Coronado, 345 N.L.R.B. 306 (2005); In re Central Parking System, Inc., 335 N.L.R.B. 390 (2001); Charles I. Cohen, Joseph E. Santucci, Jr., & Jonathan C. Fritts, Resisting Its Own Obsolescence—How the National Labor Relations Board Is Questioning the Law of Neutrality Agreements, 20 Notre Dame J.L. Ethics & Pub. Pol’y 521, 521 (2006) (“The NLRB . . . has been willing to defer to private agreements that resolve union representation matters, rather than deciding the representation question through a Board-supervised secret ballot election.”).

19  Cohen, Santucci & Fritts, supra note 18, at 522.

20  Ariz. Const. art. 2, § 37. The amendments adopted by voters in South Carolina, South Dakota, and Utah each were worded somewhat differently. See S.C. Const. art. 2, § 12 (“The fundamental right of an individual to vote by secret ballot is guaranteed for a designation, a selection, or an authorization for employee representation by a labor organization.”); S.D. Const. art. 6, § 38 (“The rights of individuals to vote by secret ballot is fundamental. If any state or federal law requires or permits an election for public office, for any initiative or referendum, or for any designation or authorization of employee representation, the right of any individual to vote by secret ballot shall be guaranteed.”); Utah Const. art. 4, § 8 (“All elections, including elections under state or federal law for public office, on an initiative or referendum, or to designate or authorize employee representation or individual representation, shall be by secret ballot.”).

21  Letter from Hon. Lafe Solomon, Acting General Counsel, by Eric G Moskowitz, Assistant General Counsel, to Hon. Tom Horne, Attorney General of Arizona (Jan. 13, 2011), available at; Letter from Hon. Lafe Solomon, Acting General Counsel, by Eric G Moskowitz, Assistant General Counsel, to Hon. Alan Wilson, Attorney General of South Carolina (Jan. 13, 2011), available at; Letter from Hon. Lafe Solomon, Acting General Counsel, by Eric G Moskowitz, Assistant General Counsel, to Hon. Marty Jackley, Attorney General of South Dakota (Jan. 13, 2011), available at; Letter from Hon. Lafe Solomon, Acting General Counsel, by Eric G Moskowitz, Assistant General Counsel, to Hon. Mark Shurtleff, Attorney General of Utah (Jan. 13, 2011), available at

22  Nat’l Labor Relations Bd., State Constitutional Amendments Conflict With the NLRA (2011).

23  Letter from Hon. Alan Wilson, Attorney General of South Carolina; Hon. Mark L. Shurtleff, Attorney General of Utah; Hon. Tom Horne, Attorney General of Arizona; and Hon. Marty J. Jackley, Attorney General of South Dakota to Hon. Lafe Solomon, Acting General Counsel, National Labor Relations Board (Jan. 27, 2011) (citing, inter alia, Abood v. Detroit Bd. Of Educ., 431 U.S. 209, 233-35 (1977)), available at

24  Letter from Hon. Lafe Solomon to Hon. Alan Wilson, Attorney General of South Carolina; Hon. Mark L. Shurtleff, Attorney General of Utah; Hon. Tom Horne, Attorney General of Arizona; and Hon. Marty J. Jackley, Attorney General of South Dakota (Feb. 2, 2011), available at (“As you have unanimously expressed the opinion that the State Amendments can be construed in a manner consistent with federal law, I believe your letter may provide a basis upon which this matter can be resolved without the necessity of costly litigation. My staff will shortly be in contact . . . to explore this issue further.”).

25  David Montgomery, Feds May Sue over Secret Ballot Vote, Rapid City J., Mar. 18, 2011, available at

26  Letter from Hon. Alan Wilson, Attorney General of South Carolina; Hon. Mark L. Shurtleff, Attorney General of Utah; Hon. Tom Horne, Attorney General of Arizona; and Marty J. Jackley, Attorney General of South Dakota to Hon. Lafe Solomon, Acting General Counsel, National Labor Relations Board (Mar. 4, 2011), available at

27  Letter from Hon. Lafe Solomon to Hon. Tom Horne, Attorney General of Arizona; Hon. Alan Wilson, Attorney General of South Carolina; Hon. Marty J. Jackley, Attorney General of South Dakota; and Hon. Mark L. Shurtleff, Attorney General of Utah (Apr. 22, 2011), available at

28  Complaint at 1, Nat’l Labor Relations Bd. v. Arizona (D. Ariz. May 6, 2011).

29  Cf. Massachusetts v. Mellon, 262 U.S. 447, 480 (1923) (holding that a proceeding does not fall within the Supreme Court’s original jurisdiction merely because a state is a party to the proceeding, but only where the state is a party to “a proceeding of judicial cognizance”); Raines v. Byrd, 521 U.S. 811 (1997) (holding that despite alleging an institutional injury (namely a loss of the legislature’s power), Members of Congress as such lack standing to challenge the constitutionality of a federal statute to which they were opposed).

30  29 U.S.C. § 152(2).

31  Order, supra, Dkt. No. 18, at 6.

32  Parklane Hosiery Co., Inc. v. Shore, 439 U.S. 322, 327 n.7 (1979) (citing Blonder-Tongue Laboratories, Inc. v. Univ. of Ill. Found., 402 U.S. 313, 329 (1971) and Hansberry v. Lee, 311 U. S. 32, 40 (1940)). 

33  As discussed in more detail below, the Board asserts that the very existence of the amendment has a chilling effect on the exercise of workers’ rights under the Act. In this case, it may be immaterial whether the Board can marshal admissible evidence to prove this contention; since the Board did not name the Arizona Secretary of State as a defendant, the district court cannot effectively order that records of the amendment’s due adoption be burned or that its text be obliterated. See Ariz. Const. art. IV, pt. I, § 1(13) (setting out governor’s duty to proclaim the results of voting on proposed amendment to constitution, declaring such amendments as approved if voted for by a majority of electors voting); Ariz. Rev. St. 41-121(A)(2) (requiring the Secretary of State to “keep a register of and attest the official acts of the governor.”). Thus, in a very important sense, Arizona’s secret ballot amendment will remain “on the books” after the lawsuit to precisely the same extent as before the lawsuit.

34  Georgia v. Stanton, 73 U.S. 50, 75 (1867) (“This court can have no right to pronounce an abstract opinion upon the constitutionality of a state law. Such law must be brought into actual, or threatened operation upon rights properly falling under judicial cognizance, or a remedy is not to be had here.”).

35  Cherokee Nation v. Georgia, 30 U.S. 1, 20 (1831).

36  In its analysis of the standing question, the district court invoked the three-part test in Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992), i.e., according to which a plaintiff must show (1) that it suffered an injury in fact that is concrete and particularized and actual or imminent; (2) that there is a causal connection between the injury and the conduct complained of that is fairly traceable to defendant’s action; and (3) the likelihood that the plaintiff’s injury can be redressed by a decision in its favor. Order, supra, Dkt. 18, at 2.  The district court found that the amendment to Arizona’s constitution caused an “injury” to the Board, and “affects plaintiff’s legal rights by undermining its exclusive authority to administer the NLRA.” Id., at 6. This finding appears to extend the Lujan test for standing beyond its original scope, because Lujan defines an injury for purposes of that test as “an invasion of a legally protected interest.” Lujan, 504 U.S. at 561. An adjudicatory body such as the Board has jurisdiction to resolve disputes, but that jurisdiction is more properly described as a power than an interest. Moreover, the delegation by Congress to an administrative agency of exclusive authority to enforce a federal law does not create an “interest” in that authority in the traditional sense of that word, for example, the way that a federal land grant creates a property interest on the part of a grantee. If Congress subsequently narrowed a prior delegation of enforcement authority, a court would be unlikely to decide that the agency as such had standing to seek a declaratory judgment questioning the constitutionality of the cut-back. 

It is interesting to note that the Board’s complaint is not predicated on an injury to anything it identifies as its own legally-protected interest. In its Opposition, the Board raised for the first time an argument that Arizona’s secret ballot amendment “interferes with the Board’s own activities or operations,” but this argument amounts to suggesting that uncertainty will generate litigation or prolong disputes, which in turn will diminish the Board’s ability “to meet its performance goals.” Proposed Opposition, supra note 3, at 12. There is a large element of bootstrapping inherent in this argument, since the “performance goals” the Board refers to are ones the Board sets for itself. See Nat’l Labor Relations Bd., Fiscal Year 2010 Performance and Accountability Report, page prior to page 1, available at (“The Performance section compares the NLRB’s performance to its annual performance goals. . . .”).

37  See Alfred L. Snapp & Son, Inc. v. Puerto Rico ex rel. Barez, 458 U.S. 592, 600-07 (1982) (explaining the parens patriae doctrine).

38  NLRB’s Opposition to Motion to Intervene of Save Our Secret Ballot and 34 Individuals at 3, Nat’l Labor Relations Bd. v. Arizona (D. Ariz. 2011) (citing and quoting Arakaki v. Cayetano, 324 F.3d 1078, 1086 (9th Cir. 2003).

39  29 U.S.C. § 157.

40  Id.

41  Six weeks after filing its Proposed Opposition Brief, the Board issued a decision denying the legitimacy of its favoring one set of Section 7 rights over another. See Lamons Gasket Co., 2011 WL 3916075 (N.L.R.B. ) at *8-9 (reasoning that because the Board’s decision in Dana Corp. required posting notices of decertification petition rights, it “placed the Board’s thumb decidedly on one side of what should be a neutral scale,” and holding that “[s]uch administrative action is not appropriate under the Act”).

42  See 29 U.S.C. §§ 160-61.

43  The Board contends, and the district court agreed, that the Supreme Court’s decision in Nat’l Labor Relations Bd. v. Nash-Finch Co., 404 U.S. 138, 144-47 (1971), provides that authority. It does not. The case simply holds that in an action against an employer involving a specific unfair labor practice, the Board is authorized to seek a remedy—even if the remedy is prohibited by ostensibly applicable state law—if the state law is in turn preempted by the NLRA. The invalidation of the state law (actually, state court injunction) in Nash-Finch was merely a necessary step in the enforcement of Section 8 of the Act in a case the NLRB was statutorily authorized to bring and over which the federal courts had a basis for jurisdiction. Nash-Finch does not stand for the proposition that the NLRB has the authority to sue a state seeking a declaration of a law’s invalidity as an end in itself, and certainly does not stand for the proposition that the Board can pick and choose which of various actually- or supposedly-preempted statutes to sue over and which to leave be.

44  Georgia v. Penn. R.R. Co., 324 U.S. 439, 474 (1945) (Stone, C.J., dissenting).

45  Free Enter. Fund v. Pub. Co. Accounting Oversight Bd., 130 S. Ct. 3138, 3153 (2010); see also Myers v. United States, 272 U.S. 52, 164 (1926).

46  Humphrey’s Executor v. United States, 295 U.S. 602, 620 (1935).

47  See generally Free Enter. Fund, 130 S. Ct. 3138.

48  Proposed Opposition, supra note 3, at 24.

49  Letter from Hon. Lafe Solomon, Acting General Counsel, by Eric G Moskowitz, Assistant General Counsel, to Hon. Tom Horne, Attorney General of Arizona, at 2-3, 3 n.1 (Jan. 13, 2011) (citations omitted), available at

50  The attributes of the states’ residual sovereignty trump the exercise of congressional power. Alden v. Maine, 527 U.S. 706, 713 (1999) (Congress may not require the courts of a state to hear claims against the state arising under a federal statute otherwise within its enumerated powers because “the States’ immunity from suit is a fundamental aspect of [their] sovereignty.”). While this paper is not the appropriate place for an elaborate discussion of the residual sovereign immunities of the states from federal inference in their internal politics, it is worth noting that it was already well-established before the Revolutionary War that, as an attribute of its sovereignty, a nation-state is by right immune from another sovereign’s attempt to interfere in its internal governmental affairs. “It clearly follows from the liberty and independence of Nations that each has the right to govern itself as it thinks proper and that no one of them has the least right to interfere in the government of another.” Emmerich de Vattel, The Law of Nations or the Principles of Natural Law Applied to the Conduct and to the Affairs of Nations and of Sovereigns 131 (Charles G. Fenwick trans., Carnegie Inst. 1916) (1758) (Book II, Ch. iv, §54).

51  Deborah Jones Merritt, The Guarantee Clause and State Autonomy: Federalism for a Third Century, 88 Colum. L. Rev. 1, 41 (1988) (“A republican government . . . is responsible to its voters rather than to any outside agency.”); cf. Taylor v. Beckham, 178 U.S. 548, 571, 580 (1900) (holding that a state’s electoral process should be “free from external interference” and treating “interference of the general government,” i.e., from the national government, as a type of “external interference”).

52  No. 11-CV-506 (W.D. Okla., July 11, 2011), appeal pending, No. 11– 6209 (10th Cir., Aug. 11, 2011).

53  Id. at 2.

54  Id. at 3.

55  Id. at 4.

56  Id. at 6.

57  Id.

58  Id. at 5 (quoting 29 U.S.C. § 160(a)).

59  Id. at 6-7.

60  Id. at 9.

61  Id. at 7-8.