Union-Led Direct Democracy

Abstract

The existing legal order has failed the American labor movement. The National Labor Relations Act (NLRA) cannot guarantee the rights it promises, and the National Labor Relations Board's enforcement mechanisms are too weak to ensure compliance. At the same time, federal preemption of labor law-among the broadest in all American law-has foreclosed local and state governments from innovating upon the national framework. The Roberts Court, meanwhile, continues to roll back labor rights and administrative protections. Trapped between the decaying foundation of the NLRA and the rigid ceiling of federal preemption, the labor movement finds itself increasingly constrained by the law itself. But nascent forms of labor lawmaking have begun to slip out of these constraints. Building on prior scholarship, this Note presents a new model of subnational labor lawmaking that avoids federal preemption: union-led direct democracy. This model harnesses direct democracy-electoral processes that allow citizens to vote directly on laws, such as the initiative and the referendum-to expand organizing, bargaining power, and workers' rights. It relies upon Labor's natural strengths in grassroots mobilization and, due to its reach and replicability, likely dwarfs all other forms of subnational labor lawmaking. This Note thus provides an original and foundational perspective on the subject. Union-led direct democracy is not just a theoretical model. It has already improved the conditions for millions of member and nonmember workers, while helping grow the labor movement's rank and file. This Note argues that direct democracy can be good for unions-and that unions, in turn, can be good for direct democracy.

Headnote

Abstract. The existing legal order has failed the American labor movement. The National Labor Relations Act (NLRA) cannot guarantee the rights it promises, and the National Labor Relations Board's enforcement mechanisms are too weak to ensure compliance. At the same time, federal preemption of labor law-among the broadest in all American law-has foreclosed local and state governments from innovating upon the national framework. The Roberts Court, meanwhile, continues to roll back labor rights and administrative protections. Trapped between the decaying foundation of the NLRA and the rigid ceiling of federal preemption, the labor movement finds itself increasingly constrained by the law itself.

But nascent forms of labor lawmaking have begun to slip out of these constraints. Building on prior scholarship, this Note presents a new model of subnational labor lawmaking that avoids federal preemption: union-led direct democracy. This model harnesses direct democracy-electoral processes that allow citizens to vote directly on laws, such as the initiative and the referendum-to expand organizing, bargaining power, and workers' rights. It relies upon Labor's natural strengths in grassroots mobilization and, due to its reach and replicability, likely dwarfs all other forms of subnational labor lawmaking. This Note thus provides an original and foundational perspective on the subject.

Union-led direct democracy is not just a theoretical model. It has already improved the conditions for millions of member and nonmember workers, while helping grow the labor movement's rank and file. This Note argues that direct democracy can be good for unions-and that unions, in turn, can be good for direct democracy.

Introduction

The legal order has long shaped the American labor movement.1 Prior to the New Deal, this shaping took place in the collisions between unions and courts, as Lochner-era judges cracked down on the labor movement through anti-union precedent, strike injunctions, and court-ordered police crackdowns.2 The legal forces shaping the American labor movement today, however, are dramatically different than those a century ago. Rather than a dyadic relationship between courts and unions, American Labor is encircled by statutory law, agencies, courts, doctrines, and other institutions at the federal and state level. Not all of these are inherently hostile to organized labor, but- much like the courts a century ago-they have collectively restricted Labor to the point of paralysis.

To briefly review these forces of the law, we can begin with the National Labor Relations Act (NLRA), which provides the legal foundation for collective organizing. The NLRA guarantees the right for most workers to organize unions and collectively bargain over working conditions, along with a set of procedures to vindicate and regularize these rights.3 But since at least the early 1980s, observers have noted that the Act is fatally underpowered. The principal critique is that the NLRA gives employers too much power to interfere with unionization efforts, while offering unions few rights to push back on (often illegal) employer interference.4 Statutory remedies against unionbusting employers are extraordinarily weak, limited to no- or low-cost penalties such as injunctions, reinstatement, and backpay-and usually far cheaper for employers than allowing unionization.5 The Supreme Court has interpreted the NLRA as providing the Board with "essentially remedial" rather than "punitive" powers, and the Court has interceded when the National Labor Relations Board (NLRB) goes beyond these bounds.6 And because of its slow processes, delays, and backlogs for adjudicating abuses and certifying elections, the NLRB can take years to resolve the easiest cases.7 The result is that the NLRA "more and more resembles an elegant tombstone for a dying institution."8

But if the NLRA is the decaying foundation, then labor law preemption is the ceiling foreclosing state and local efforts to go beyond the Act. In a series of decisions, the Supreme Court created "one of the most expansive preemption regimes in American law," shutting down labor law experimentation at the state and local level.9 Though a few exceptions to the doctrine exist, only one is encoded by statute-to the detriment of Labor. The Taft-Hartley Act authorizes states to pass right-to-work laws, which bar unions from collecting fees from nonmembers who benefit from collective-bargaining agreements and union benefits.10 Another common formulation is that right-to-work laws allow individual workers to secede from unions, "even where the latter had won elections to represent all the workers," thereby undermining worker solidarity and collective bargaining power.11 Under right-to-work laws, workers are economically incentivized to quit unions since they receive the same services without having to contribute dues; meanwhile, unions are bled dry as they expend resources on nonmembers rather than negotiating better contracts and organizing new workplaces.12 And while there is no statutory federal right-to-work law, in 2018 the Supreme Court overturned four decades of precedent to constitutionalize right to work for state and federal public sector employees.13

What is the collective effect of these developments? The NLRA is not strong enough to incentivize employer compliance. The NLRB is too cumbersome to enforce labor rights. Federal preemption prohibits subnational policy innovation. Right-to-work laws eat away at union finances and membership. These forces of law have caught Labor in a vice grip-stymying rights, enforcement, organizing, and bargaining. Following a decades-old trendline, these dynamics have driven labor density to its lowest rate on record: 10.0% in 2023.14

But nascent forms of labor lawmaking have begun to slip out of these constraints. While federal preemption closes traditional forms of lawmaking at the state and local level (such as statutorily redefining the rules of union organizing),15 legal scholarship has identified ways in which unions are beginning to indirectly leverage subnational law.16 Missing from these prior accounts, however, is how unions have utilized direct democracy-defined here as electoral processes that allow citizens to vote directly on laws, such as the initiative and the referendum-at the state and local level to shiftthe forces of the law.

This Note explores a new model of subnational labor lawmaking that I term union-led direct democracy. Analyzed in depth for the first time in legal scholarship, union-led direct democracy deploys Labor's natural strengths in collecting signatures, educating the electorate, and turning out voters to create law that indirectly facilitates union organizing and bargaining.17 While some scholars have broadly discussed pro-labor initiatives, these discussions fail to fully explain the strategic significance for how and why unions participate in direct democracy.18 They also fail to situate union-led direct democracy within the broader context of subnational labor lawmaking and federal preemption. This Note thus provides a foundational perspective on union-led direct democracy as a form of subnational labor lawmaking.

Direct democracy represents an increasingly decisive arena to contest workers' rights. The statutory or constitutional initiative exists in twentythree states, covering nearly half the country's population,19 and has been used by unions to alter the working, organizing, and bargaining conditions for millions of workers.20 Like other forms of subnational labor lawmaking, union-led direct democracy likely evades federal preemption under existing doctrine.21 But unlike those forms, union-led direct democracy is broadly applicable in a wide range of contexts. As shown in this Note, the sheer scale of union-led direct democracy almost certainly dwarfs all other subnational labor law in prevalence as well as impact on workers.22

Beyond its presumptive legality, Labor's use of direct democracy is a normatively positive development. Union-led direct democracy translates popular support for economic justice into better working conditions for member and nonmember workers, thus countering laws and structures actively hostile to organized labor.23 The labor movement has successfully used direct democracy in states with right-to-work laws and legislatures that weaponize state preemption against pro-worker localities.24 Furthermore, certain features of union-led campaigns may address common problems in direct democracy, such as information problems in the electorate and the influence of special interests. Direct democracy can be good for unions, and unions can be good for direct democracy.

This Note is organized as follows. Part I provides an overview of the doctrine of labor law preemption as well as two previously identified models of subnational labor lawmaking: tripartism and social bargaining. Part II presents three case studies of union-led direct democracy. Two case studies take place in Los Angeles, involve local initiatives, and feature a single union. The third examines a series of statewide initiatives in Arizona driven by a coalition of labor unions. Part III argues the legal and normative case for union-led direct democracy. Part IV examines broader implications and limitations of Labor's involvement in direct democracy.

I. Slipping the Constraints: Federal Preemption and New Forms of Labor Lawmaking

The doctrine of labor law preemption is one of the broadest in the entire body of American law. Three interlocking Supreme Court decisions preempt nearly any direct state or local interference with the NLRA regime.25 This doctrinal development reflects the history of intense industrial conflict preceding the Act's passage during the nineteenth and early twentieth centuries.26

Despite this broad preemption doctrine, the labor movement has begun to indirectly create state and local labor law.27 Two primary models of this "new" labor law have been described in the literature. First, tripartism involves political exchanges between unions, employers, and governments that enable private reorderings more favorable than those provided by the NLRA.28 For example, a hospital may agree to card-check neutrality in an organizing campaign in return for the union's political support to acquire zoning permits for a new building, which the local government may choose to grant only with the union's blessing.29 Second, social bargaining describes public processes involving negotiations between subnational governments, business, and labor over regional or sectoral working standards.30 California's recently codified "Fast Food Council" exemplifies this model. That council convenes representatives of fast-food management, employees, labor, and state government to collectively establish industry wages and other working standards.31 These paradigms, as discussed in this Part, capture important developments in labor law but do not account for union-led direct democracy.

A. Federal Preemption of Labor Law

One cannot discuss the doctrine of labor law preemption as it exists today without acknowledging the intense history of conflict that explains its development. Interactions between organized labor, management, and the state in the late nineteenth and early twentieth centuries were often violent.32 From the 1877 Railroad Strikes and continuing through the first Red Scare after World War I, workers attempting to unionize or strike were often met with brutal repression.33 State militias such as the National Guard, local police forces, and private security forces like the Pinkertons frequently turned labor actions into mass casualty events.34

And the courts played a decisive role in this era, largely against Labor. Judges prolifically deployed injunctions and other penalties to break strikes and force unions back to work.35 Antitrust laws were reinterpreted to penalize collective actions.36 Courts endorsed the federal government's weaponization of immigration laws as a tool to target and deport some labor activists, labeling them "alien radicals" regardless of whether they posed genuine threats to national security.37

This judicial antagonism emerged in part from a radical transformation in judicial review at the turn of the century.38 Now referred to as the Lochner era after an infamous 1905 case in which the Supreme Court invalidated a New York worker protection law, this period heralded unprecedented judicial intervention against regulation as the judiciary regularly invoked a substantive due process right to contract to overturn economic and social legislation.39 Some state supreme courts had already adopted the Lochner mentality by the late nineteenth century, although there was considerable variance in the degree of state judicial hostility against organized labor.40

The courts' escalating interventions-overturning pro-worker legislation, strike injunctions, and other judicial remedies-legitimated and facilitated violence against unions and their members.41 Industrial relations were in a state of constant chaos as unions, courts, and employers (often violently) attempted to assert control.42 "The worst thing about this power," as one legal historian described the use of judicial review and other interventions in this period, "was that it was randomly and irresponsibly exercised. It could be neither predicted nor controlled."43

For its part, the Lochner-era labor movement came to think of the courts as central antagonists to its struggle for collective organizing and bargaining rights.44 In 1918 the Executive Council of the American Federation of Labor commissioned a study to address how "'[c]ourts of the United States without constitutional authority or legislative sanction have assumed the power to invade the prerogatives of the legislative branch of the government by unmaking and rendering invalid laws enacted by the people or their legislative representatives.'"45 Labor framed its relationship with the courts as a battle for fundamental democratic rights against a tyrannical judiciary, whose "'action[s] . . . would be impossible in any other democratically governed nation.'"46 The final study proposed several remedial propositions, including jurisdiction-stripping and the widespread adoption of direct democracy procedures, including the initiative, referendum, proportional representation, and ranked choice voting.47

This antagonism softened as the New Deal triumphed over the courts and the Lochner era waned-and, most importantly, after Congress passed the NLRA in 1935.48 The Act sought to promote industrial peace by "encouraging the practice and procedure of collective bargaining and by 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."49 With the history of intense labor conflict (often made worse by state judiciaries and authorities) in mind, Congress centralized the Nation's procedures for industrial conflict under the NLRB with the intent of creating a single national labor policy.50 But the Act leftunaddressed the proper relationship between these new federal labor rights and the role of states and cities to develop their own labor frameworks, as they had done prior to the New Deal.51

Filling in this gap, the Supreme Court fashioned an extraordinarily broad doctrine of labor law preemption over a series of decisions.52 Three stylized points in the doctrine are highlighted here. First, federal labor law preempts any state regulation that touches upon conduct "arguably" prohibited or protected by the NLRA.53 Second, preemption applies even for conduct not regulated by the NLRA, if such conduct falls within the zone of activities Congress intended to leave unregulated and "controlled by the free play of economic forces."54 Third, states are preempted from using their powers in a manner "tantamount to [labor] regulation," including in areas not traditionally preempted.55 Importantly, the Supreme Court's unique labor law preemption doctrine reflects the history of state hostility toward the labor movement in the pre-New Deal era.56

The Court developed its first landmark rule on state-federal conflicts of labor law in San Diego Building Trades Council v. Garmon.57 In that case, several unions peacefully picketed a nonunionized lumber yard to pressure the owners into employing union labor.58 The unions also pressured customers and suppliers to avoid doing business with the employer.59 The employer sued in California court, winning an injunction and monetary damages.60 Concurrently, the lumberyard owner filed a proceeding with the NLRB, but the Board declined jurisdiction over the case.61

Upon review, the Supreme Court overturned the state court's injunction and monetary damages, reasoning that Congress had entrusted labor policy administration for the entire country to the NLRB.62 That the NLRB had declined to exercise its authority was irrelevant and did not create an opening for state court intervention either: "[T]he failure of the [NLRB] to assume jurisdiction did not leave the States free to regulate activities they would otherwise be precluded from regulating."63 The Court concluded that any activity "arguably within the compass of § 7 and § 8 of the Act" falls outside state jurisdiction.64 Indeed, as the facts of Garmon show, the Court's sweeping standard originally applied even when the NLRB declined jurisdiction over labor disputes.65 Congress slightly narrowed Garmon preemption in the 1959 Landrum-Griffin Act by allowing states to assume and assert jurisdiction over labor disputes that the Board declines.66 But as soon as the Board asserts jurisdiction, state and local governments are presumably ousted.67

In Lodge 76, International Association of Machinists and Aerospace Workers v. Wisconsin Employment Relations Commission, the Court expanded preemption to encompass activities the Court believed were intended by Congress to be leftunregulated and controlled by the "free play of economic forces"-even those not "arguably" covered by Sections 7 or 8 of the NLRA.68 In that case, a Wisconsin union had its collective-bargaining agreement terminated by the employer, setting offnegotiations for a new contract.69 Before a new contract could be signed, the employer unilaterally imposed changes to the conditions of employment, such as eliminating the union's office from the plant, refusing to participate in dues collection, and increasing the workweek from 37.5 hours to 40 hours per week.70 The union responded with a strike action and a resolution binding members to refuse to work overtime, defined as any time exceeding 37.5 hours per week.71 The employer filed a complaint with the Wisconsin Employment Relations Commission, "charging that the refusal to work overtime constituted an unfair labor practice under state law."72 The Commission and an Examiner found that the claim was not preempted by the NLRA under the Garmon rule because "the concerted refusal to work overtime[] is not an activity which is arguably protected under Section 7 or arguably prohibited under Section 8."73

Though agreeing that the NLRA neither arguably protected nor arguably prohibited the conduct in question, the Court nonetheless found the Commission's rulings to be preempted.74 The Court reasoned that "Congress struck a balance of protection, prohibition, and laissez-faire in respect to union organization, collective bargaining, and labor disputes."75 Because Congress had been "rather specific" when outlawing "particular economic weapons," it would simply be inconsistent with the federal statutory scheme if states were allowed to fill in the space leftunregulated by Congress.76 The Court concluded that a refusal to work overtime fell within this third category of conduct that Congress intended to leave "unregulated and to be controlled by the free play of economic forces."77

Finally, in Wisconsin Department of Industry, Labor and Human Relations v. Gould Inc., the Court forbade the states from using their spending powers in ways that "function[] unambiguously as a supplemental sanction for violations of the NLRA."78 Gould involved a Wisconsin law that banned repeat violators of the NLRA from doing business with the state.79 Unlike in Garmon or Machinists, Wisconsin relied on its spending powers rather than its police powers to pursue its policy goals.80 The Court reasoned that the state's reliance on spending rather than regulatory powers was "a distinction without a difference . . . because on its face the debarment statute serves plainly as a means of enforcing the NLRA."81 It did not matter that Wisconsin operated as a market participant in search of services: "[F]or all practical purposes, Wisconsin's debarment scheme is tantamount to regulation."82 Adding up the collective force of the rules in Garmon (the "arguable" standard), Machinists (tools meant to be leftunregulated by Congress), and Gould (state intervention in forms tantamount to regulation), the federal preemption doctrine for labor law appears uniquely expansive.

The Court has nonetheless reserved narrow exceptions to the doctrine. First, states may impose civil and criminal damages when labor conduct becomes violent and imminently threatens public order.83 "State jurisdiction . . . prevail[s] in these situations because the compelling state interest . . . in the maintenance of domestic peace is not overridden in the absence of clearly expressed congressional direction."84 As a corollary, one line of NLRB precedent affirmed by the Supreme Court has held that there is no protection afforded by the NLRA in cases where employees who are engaged in collective actions fail to take reasonable steps to protect employers' property from foreseeable, imminent danger.85

Second, federal labor law does not preclude "States from regulating aspects of labor relations that involve 'conduct touch[ing] interests so deeply rooted in local feeling and responsibility that . . . we could not infer that Congress had deprived the States of the power to act.'"86 Third, federal labor law does not preclude laws where "the activity regulated was a merely peripheral concern of the [NLRA]."87 As Part III below will discuss, the Court has clarified the boundaries of these exceptions (especially regarding the states' traditional police powers to regulate minimum working conditions and standards), and their applications are highly relevant to union-backed campaigns in direct democracy.

While Garmon, Machinists, and Gould still form the basis of the doctrine, a recent preemption case, Glacier Northwest, Inc. v. International Brotherhood of Teamsters Local Union No. 174, suggests that some members of the current Court may be open to a doctrinal makeover.88 There, the Court relied on the first exception discussed above and ruled on narrow grounds that the NLRA did not preempt an employer's tort claims against workers who did not take reasonable precautions to protect the employer's delivery trucks from imminent danger or destruction during a strike.89 Ruling against the union on fairly conventional doctrinal grounds, the Court found that the NLRA does not even arguably protect the union's allegedly tortious conduct.90

Justice Thomas's concurrence, however, went further than the majority's opinion: "[I]n an appropriate case, we should carefully reexamine whether the law supports Garmon's 'unusual' pre-emption regime. In doing so, I would bear in mind that any proper pre-emption inquiry must focus on the NLRA's text and ask whether federal law and state law 'are in logical contradiction.'"91 Given his track record, Justice Thomas's goal is almost certainly to weaken federal labor rights and protections,92 but his concurrence does not spell out what exactly comes after Garmon preemption.93 Furthermore, any dilution of federal preemption is a double-edged sword, as pro- and anti-union states would likely take advantage of the latitude to pass labor laws consistent with their preferences. The uncertainty around reforming labor law preemption is reflected in the strange bedfellows who support weakening the doctrine: The conservative Heritage Foundation has proposed policies loosening NLRA preemption, but so have some labor advocates.94 These advocates believe that pro-union states have more potential upside since anti-union, right-to-work states already tend to have fairly minimal union presences, whereas the former could push their union density rates significantly higher.95 Put simply, it is unclear what a post-Garmon world looks like-and whether relaxing labor law preemption would hurt or benefit the labor movement.96 But, as only Justice Gorsuch joined Justice Thomas's concurrence (and the majority decided the case narrowly with few changes to existing doctrine), overturning Garmon preemption seems unlikely for now.97

B. Emerging Forms of Labor Law

In response to this Court-fashioned preemption doctrine, the labor movement has begun using novel pathways at the state and local levels to indirectly experiment with labor law. Two models previously identified in the literature are reviewed here. The first, known as "tripartism," describes negotiations between workers, employers, and governments that produce private agreements through which unions and employers redefine the rules of organizing and bargaining.98 The second, known as "social bargaining," describes how unions mobilize as political actors to change the standards of employment at the sectoral level through public processes.99

1. Tripartism

The tripartite model describes how political exchanges between unions, employers, and state and local governments enable a reconstruction of collective organizing rules more favorable than those provided by the NLRA.100 Though subnational governments are preempted from directly legislating labor laws, they retain full powers in other areas of law such as zoning, tax policy, and governmental contracting that can be used in employer-union exchanges.101 For instance, a city may award zoning permits to a hospital for a new facility in return for employer noninterference with a union organizing campaign. Elected officials often accede to these tripartite exchanges, driven by a desire for labor peace, economic development, and reelection. Thus, under tripartism, unions can leverage their political influence over local and state officials in return for more favorable organizing and bargaining rules from employers.102

The NLRB affirmatively permits private reorderings of rules for union organizing and bargaining between employers and unions.103 As part of its policy honoring these voluntary agreements, the NLRB "will enforce such agreements, including agreements that explicitly address matters involving union representation."104 This reasoning relates to Machinists, where the Supreme Court recognized that Congress intended to leave certain conduct unregulated and "controlled by the free play of economic forces" in the labormanagement relationship.105

But the tripartite method of avoiding federal preemption raises several concerns.106 First, tripartite exchanges almost always require closed-door negotiations with government that are very difficult to observe, raising issues of political accountability.107 Second, the private nature of tripartism can obstruct meaningful judicial review.108 Third, tripartism gives rise to a "[p]olitics of [i]ndirection," as "political action for labor rights must be rechanneled to support the nonlabor issues that are of interest to employers."109 And fourth, tripartism carries "[c]ollateral [e]ffects" as "local governments rely on other areas of law to effect private reorderings of organizing rules," resulting in "repercussions for those other areas of law."110

Tripartism raises other concerns of strategy and efficacy in rebuilding the labor movement. Though it successfully avoids federal preemption, it may be limited in its use to already-established unions and with significant influence in local and state government. It is unclear, for instance, whether sectors with newly or poorly organized unions, such as Amazon warehouses,111 could take advantage of tripartite exchanges to organize workers. Whether from the perspective of good governance or good strategy, tripartism seems limited to those well-organized unions already established in state or local politics.

2. Social bargaining

Tripartism locates the nexus of negotiations in political exchanges: Policy in unrelated areas of law such as zoning are traded for private reorderings of organizing and bargaining rules.112 The social bargaining model, by contrast, locates the nexus of tripartite negotiations at the "sectoral, industrial, and regional levels."113 The NLRA statutorily defines the appropriate unit for collective bargaining to be "the employer unit, craftunit, plant unit, or subdivision thereof."114 Social bargaining avoids unit-based restrictions entirely; instead, unions, employees, and other representatives negotiate with employer representatives across entire industries or sectors outside the collective bargaining context.115 These negotiations occur through public processes, such as industrial commissions convened by state or local governments. By setting general standards for union and nonunion workers alike, social bargaining somewhat erases the distinction between employment and labor law-and as a result does not share the same goals as tripartism, which focuses on collective organizing and bargaining for unions alone.116

Several states have experimented with forms of social bargaining across industries or classes of workers.117 California created a "Fast Food Council" to establish labor standards and minimum wages for the industry, with strong support from the Service Employees International Union (SEIU).118 The Council consists of nine voting members: two representatives each from the fast food industry, fast food franchisees, fast food employees, and advocates for fast food employees, with a ninth unaffiliated member from the public.119 The Council seats two additional non-voting representatives from the California government.120 Similarly, New Jersey's Minimum Wage Advisory Commission consisted of business representatives and labor representatives nominated by the state's AFL-CIO affiliate.121 The "Fight for $15" campaign- organized and promoted by SEIU but led by unorganized workers-is a variation of the social bargaining model, as unions and workers put pressure on government and employers through other means to raise minimum wages.122

Though it has not been challenged, social bargaining likely survives labor law preemption. The Supreme Court "has never curtailed the ability of states and local governments to pass universally applicable employment legislation."123 The Court would have to dramatically alter the doctrine to sweep state and local employment law under the aegis of the NLRA.124 Furthermore, social bargaining does not implicate the rules set forth in the Court's canonical cases. Because it does not even "arguably" raise Section 7 or Section 8 concerns, nor touch upon unregulated economic tools available for unions and employers, neither Garmon nor Machinists applies.125 The Court could strike down social bargaining processes that explicitly elevate union power in state commissions under Gould, but the Court would have to stretch Gould to encompass appointment powers in addition to spending powers-an unlikely result.126 As long as social bargaining doesn't mandate benefits that redound exclusively to organized workers, the model should survive preemption.

Surviving federal preemption, however, has not insulated the social bargaining model from substantive critiques in the labor movement.127 By design, social bargaining skips over workplace organizing as its keystone strategy. So far, it is unclear whether social bargaining has produced more dues-paying members, higher revenues for unions, or spurred union organizing in a meaningful sense.128 Expensive and time-consuming campaigns such as the "Fight for $15" never resulted in traditionally unionized fast-food retailers.129 While SEIU in February of 2024 launched the duescollecting California Fast Food Workers Union (CAFFWU) as an outgrowth of the "Fight for $15" campaign, CAFFWU is a nontraditional minority union that cannot bargain and negotiate contracts with employers.130 Instead, CAFFWU's ostensible advantage for workers is that it provides a path for fast food employees to have their voices heard at the state's fast-food council.131 Other members of the labor movement consider the social bargaining tendency-and, in particular, state-mandated industrial commissions-to even undermine the revival of a true labor movement because the state (rather than unions) becomes the central lever to improve working conditions.132 The coming years in California may shed further light on some of these open questions.

II. Union-Led Direct Democracy: Three Case Studies

Missing from the tripartite and social bargaining models, however, are the unique ways in which unions have leveraged broad popular support for economic justice at the ballot box. Termed "union-led direct democracy" in this Note, these efforts rely on voter mobilization at the local and state levels to empower unionization.

As this Part shows, this model strategically utilizes direct democracy mechanisms to accomplish three broad objectives: facilitate the organizing of new workplaces; strengthen union bargaining and negotiating; and raise the material floor for all workers, member and nonmember alike. Like tripartism (and perhaps unlike social bargaining), union-led direct democracy has helped unionize workers with better contracts. Like social bargaining (and unlike tripartism), union-led direct democracy relies on public processes and a broader movement of workers inside and outside the labor movement to succeed.

This Part presents three cases of union-led direct democracy. The first looks at how UNITE HERE Local 11 wielded a local initiative in Los Angeles to improve working conditions for all workers while reserving special bargaining clauses (known as "supersession clauses") to strengthen the union's hand during bargaining. The second case illustrates how the same union, Local 11, utilized the threat of direct democracy to pressure the City Council to pass ordinances favorable to Labor and future organizing campaigns. The third case examines how a coalition of unions in Arizona created the state's first minimum wage and sick leave laws while altering hostile constitutional structures by chipping away at state preemption and right to work. These case studies demonstrate the reach and power of union-led direct democracy to create subnational labor law-empowering union organizing and bargaining while improving working conditions for millions of member and nonmember workers.

A. Los Angeles, I: Collective Bargaining Supersession Clauses

UNITE HERE Local 11 represents over 32,000 workers across Southern California and Arizona.133 Their members primarily consist of lower-earning service sector workers in hotels, restaurants, airports, and more.134 Since July 2023, Local 11 has led a series of highly successful rolling strikes across the Los Angeles area, delivering major benefits for its membership in the hospitality industry.135 According to the union, over 14,000 members across a dozen hotels have been involved in the strikes.136 Crucially, Local 11 has also built political power in city politics by electing one of its own longtime labor leaders as a candidate for the Los Angeles City Council, Hugo Soto-Martinez.137 Local 11 also receives strong support from other city council members: For instance, Councilmember Nithya Raman, along with now-Councilmember Soto- Martinez, was arrested while supporting a Local 11 hotel workers' protest.138

But when Local 11 cannot count on either collective bargaining or its support on the City Council, Local 11 has resorted to direct democracy. During the spring of 2022, Local 11 began a signature drive for a successful initiative to establish safety standards, raise wages, and provide other rights for hotel workers.139 The ordinance, titled the "Hotel Worker Protection Ordinance," created a set of universally-applicable measures, including requiring hotels to "provide a personal security device to each hotel worker," pay "a wage premium," and obtain informed consent "when their workload assignments exceed defined limits."140 Local 11 sponsored the measure, providing funding and logistical support for the campaign.141

The provision on personal security devices-also known as "panic button[s]"-was one of the union's primary goals.142 As the ordinance's statement of purpose explains, hotel workers have been "vulnerable to crimes and other threatening behavior, including sexual assault."143 Equipping workers with devices that could quickly summon security would help ensure the safety not just of Local 11's members, but of all low-wage hotel workers.144

While the initiative raised the floor for all workers, Local 11 organizers also included a provision to strengthen their own organizing efforts.145 Buried toward the end of the measure, the ordinance states that any part of the fair compensation provisions may be "waived pursuant to a bona fide collective bargaining agreement, but only if the waiver is expressly set forth in clear and unambiguous written terms."146 These "supersession clauses" are powerful tools for unions, allowing them to exchange costly workers' rights regulations in return for a first contract, employer recognition, or any other private reordering of the rules of union organizing and bargaining.147

Local 11 quickly collected the signatures it needed, giving the Los Angeles City Council the choice of either adopting the measure outright, putting it before voters at the next scheduled election, or scheduling a special election for the initiative.148 The Council decided to simply adopt the measure; by August of 2022, Mayor Eric Garcetti signed it into law.149 In less than six months and in a single initiative, Local 11 improved working conditions for thousands of hotel workers inside and outside unions. At the same time, it created new points of leverage to bargain with employers through regulatory supersession.

B. Los Angeles, II: Organizing New Workers and Workplaces

The Hotel Worker Protection initiative both improved conditions for workers in the low-wage hospitality industry and also created a more favorable bargaining environment.150 Supercession clauses, as discussed, provide new points of leverage for unions negotiating with employers. But the labor movement has also used direct democracy to organize new workplaces and pressure employers into remaining neutral during unionization campaigns. Local 11's application of direct democracy in this second case study illustrates these tactics.

The story begins with one of Los Angeles's most pressing crises: the growth and entrenchment of homelessness in the city. In 2024, roughly 45,000 people were homeless in the City of Los Angeles, with another 30,000 in the rest of Los Angeles County.151 In 2023, Los Angeles had the second-largest homeless population of any city in the United States.152 As rents have shot up and the availability of affordable housing has fallen behind, more people have landed in the streets.153 And contrary to many media narratives, the vast majority of people experiencing homelessness are from the communities they are in: 90% of homeless people in California are from California, and 75% lived in the same county where they were last housed.154

According to Local 11, the growth of Los Angeles's hospitality industry has contributed to the housing crisis.155 Housing units are frequently demolished to make way for hotels.156 Even if no housing units are destroyed, hoteliers build on land that could be "equally suitable for housing development and thus crowd out sites that could be used" for affordable housing.157 Luxury hotels may have spillover effects, driving rents and land prices up in surrounding areas-indirectly making housing less affordable.158 These issues are of pressing concern for Local 11's members, who largely cannot afford to live near where they work.159

Faced with a citywide homelessness crisis and acute internal demand for affordable, stable housing from its members, Local 11 launched the controversial "Los Angeles Responsible Hotel Ordinance" initiative campaign.160 The initiative proposed forcing hotels to accept a "fair market rate" from the city in return for lodging homeless people in their vacant rooms.161 The initiative would have prohibited hotels from discriminating against homeless people or refusing to participate in the program.162 It also contained language that would have required stricter permitting and licensing for hotel permits, which could be denied if "unscrupulous" hoteliers violate employment laws such as wage and hour standards.163 By January of 2022, Local 11 had gathered over 126,000 signatures, earning the initiative a spot on the 2024 ballot.164

Unsurprisingly, the initiative evoked intense opposition from the hotel industry; more surprisingly, the initiative also raised intense fears that it could pass.165 Behind the scenes, the Los Angeles City Council quickly reached out to union and industry representatives to arrange a compromise.166 On November 1, 2023, City Council President Paul Krekorian announced that an agreement had been reached: The union would withdraw the ballot initiative in return for the codification of certain portions of the proposed ordinance.167 The new compromise ordinance required developers to "replace any housing lost to hotel construction" and allow more community input into hotel developments.168 The deal also replaced the mandatory rooms-for-homelesspeople program with a voluntary version of the same program.169

But the Union's greatest win was that all new hotel developments must obtain a special permit from either the City Planning Commission or the City Council with a mandatory public review process.170 With its deep support from the Los Angeles City Council,171 Local 11 (in theory) could pressure the City's elected officials to withhold future hotel developments and permits in Los Angeles unless developers concede to card-check-neutrality agreements guaranteeing employer noninterference and union recognition. Local 11 deployed direct democracy as a tactical threat to both Los Angeles's elected officials and its hospitality industry, forcing a political outcome that greatly expanded the union's capacity to organize new workplaces across the City.

The Responsible Hotel Ordinance case study shows how the different models of subnational labor lawmaking are not necessarily exclusive. Indeed, tripartism, social bargaining, and union-led direct democracy may often merge or overlap. At first, Local 11 sought to pass the initiative by popular vote; upon earning a spot on the ballot, industry and local government sought to make a series of backroom tripartite exchanges to pass a compromise ordinance. Tripartism and union-led direct democracy complemented one another to effectively empower union organizing at any future hotel development in Los Angeles. And similar to social bargaining, union-led direct democracy in this case study became a sword to proactively force the City Council and the hotel industry to accommodate Local 11's demands through public processes. While the three models are analytically distinct, they also share many characteristics and can be used in conjunction to deepen union density and organize new workers.

C. Arizona: Hostile Constitutions and Hostile Legislatures

The two Los Angeles case studies focus on local direct democracy and a single union. The final case study examines how union coalitions have utilized state-level direct democracy to remake Arizona's constitutional structure of home rule, state preemption, and right to work over a series of initiatives.

1. Round I: Creating a minimum wage in Arizona and beating back state preemption

Arizona has a history of hostility to the labor movement. Article XXV in Arizona's constitution proclaims a "[r]ight to work . . . without membership in [a] labor organization,"172 representing one of the earliest examples of a state right-to-work provision.173 Only about 4% of wage and salary workers are unionized in Arizona-less than half the national average.174 The Grand Canyon State has produced iconic anti-union politicians such as Barry Goldwater;175 his eponymous think tank, the Barry Goldwater Institute, based in Phoenix, continues to spearhead anti-union litigation and advocacy today.176

In part because of these challenging conditions, unions have relied on direct democracy to advance economic justice in the state. Until 2007, Arizona did not have a minimum wage law and instead defaulted to the federal minimum wage.177 This changed in 2006, when the union-backed Arizona Minimum Wage Coalition collected just under 200,000 signatures to place Proposition 202, the "Arizona Minimum Wage Act," on the statewide ballot.178 At the time, the $5.15 per hour federal minimum wage had not been raised in nearly a decade.179 The initiative proposed establishing Arizona's first minimum wage law to be set at $6.75 per hour.180 The modest change in the minimum wage would benefit an estimated 145,000 families.181 Several unionaffiliated individuals provided testimony supporting Proposition 202 in the state's 2006 ballot pamphlet.182 The Arizona Minimum Wage Act was passed by a nearly two-to-one margin in favor.183

Proposition 202 also contained a liberal home-rule provision allowing local governments to set their own, higher minimum wages-especially important in light of Arizona's status as a mixed home-rule state.184 Under Article IX, Section 20 of the Arizona Constitution, local governments can democratically elect to become home-rule municipalities, granting themselves broad powers over budget and policy.185 Courts tend to interpret non-home rule municipal powers according to what is known as "Dillon's rule"-an interpretive principle restricting a local government's power only to actions authorized explicitly by the state legislature.186 Local governments constrained by Dillon's rule can act only when there is a clear legislative delegation of power.187 Proposition 202's embedded home-rule provision circumvents Dillon's rule by granting every locality-whether it has acted through Article IX, Section 20 or not-the power to raise local wages or worker benefits.188

The Republican-controlled state government attempted to undercut the Arizona Minimum Wage Act through assertions of state preemption powers.189 First, in 2013, the Arizona Legislature passed House Bill (HB) 2280, which sought to preempt local wage regulation by forbidding "further regulation [of compensation] by a city, town or other political subdivision of this state."190 But in FlagstaffLiving Wage Coalition v. Arizona, a state court invalidated that measure for violating Arizona's initiative procedure, thus affirmatively granting localities the power to raise the minimum wage.191

The state government then passed HB 2579 in 2016, which again attempted to circumvent Proposition 202 by preempting localities from regulating "benefits" for workers.192 This time, UFCW Local 99 (one of the original union backers of Proposition 202) sued the state.193 And once again, the court ruled against the state's measure, holding that it violated "express limitations on legislative changes to voter-approved laws."194

2. Round II: Raising the wage again-and bringing supersession to Arizona

By 2016, another union-led coalition set its sights on raising the state minimum wage again by initiative. Proposition 206, also known as the "Fair Wages and Healthy Families Act," proposed raising the minimum wage to $12 per hour by 2020.195 The initiative also created Arizona's first "earned paid sick time" law.196 Unlike Proposition 202, the Fair Wages and Healthy Families Act contained a collective bargaining supersession clause: "[E]mployees covered by a valid collective bargaining agreement" may choose to waive the earned paid sick time requirements if done "expressly" and "in clear and unambiguous terms."197 The ballot pamphlet included testimony and support from AFSCME Local 2384, UFCW Local 99, the labor-affiliated Central Arizonans for a Sustainable Economy (now known as the Worker Power Institute), and a variety of civic society organizations.198 The initiative received additional backing and endorsement from the Arizona AFL-CIO, UNITE HERE, Pima Area Labor Federation, the Arizona Building and Construction Trades Council, and Living United for Change in Arizona.199 In a year where Donald Trump beat Hillary Clinton by nearly four percentage points, Proposition 206 passed with 58% of Arizonans voting in favor.200 According to one estimate, Proposition 206 raised the wages of 779,000 workers and provided statemandated paid sick leave to nearly a million residents.201

The supersession clause in Proposition 206 raised the ire of some of the historical nemeses of organized labor.202 Several opposing arguments to Proposition 206 in the 2016 Arizona Ballot explicitly called out the "special carve-out" that the unions reserved for themselves.203 After it passed, the Arizona Chamber of Commerce sued the state and raised two claims: first, that Proposition 206 violated Arizona's single-subject rule; and second, that the measure created new costs without providing new sources of revenue to fund those added costs, violating a state-specific budgeting provision.204 The Goldwater Institute filed an amicus brief, with an entire section arguing that "Prop. 206 is . . . designed to advance the political strength of unions, not to benefit workers."205 The Arizona Supreme Court, however, rejected both claims and upheld the constitutionality of the Fair Wages and Healthy Families Act.206

Over the course of a decade, a coalition of unions created Arizona's first minimum wage, raised that minimum wage, and then established earned paid sick leave-all while beating back state preemption multiple times and creating expansive home rule powers for localities to regulate wages. Propositions 202 and 206 have collectively benefited hundreds of thousands-if not millions-of Arizonans.207 Propositions 206 also embedded a statewide collective bargaining supersession clause, creating a new foothold for unions in one of the least unionized states in the nation.208 Building offthis string of victories, the Arizona labor coalition turned its sights to an even bigger challenge: repealing Arizona's right-to-work amendment.209

3. Round III: Repealing right to work

Right-to-work provisions entitle employees in unionized workspaces to all the benefits of a union contract and bargaining without paying any dues,210 including expensive services such as prosecuting worker grievances. As mentioned above, this undermines the basic principle of collective organizing, as workplaces with majority support for unionization may still lose individual workers who decide to opt-out of the union.211 This, in turn, forces unions to expend their limited resources on nonmembers rather than on organizing new workers and workplaces.212 Furthermore, workers in right-to-work states consistently receive lower wages and benefits than workers in other states, even after controlling for a variety of faí ctors.213 By one count, twenty-six states today have right-to-work laws on the books.214

Arizona was one of the first states in the country to pass a right-to-work amendment, Article XXV, adding it to the state constitution before the Taft- Hartley Act passed-but a coalition of labor unions recently sought to overturn that legacy.215 The effort began in January of 2023, when progressive legislators in the Republican-controlled state legislature attempted to refer an amendment to voters repealing Article XXV.216 While this first attempt died in committee,217 a coalition of unions picked up where the Legislature stalled to collect signatures and campaign for a statewide initiative, the Arizona Works Together Act, that sought to repeal right to work.218 The initiative would have replaced Article XXV with an affirmative right "to self-organize, to form, join, or assist labor organizations, to bargain collectively . . . and to engage in other concerted activities for the purpose of collective bargaining or other mutual aid or protection."219 The effort represented the first documented effort to overturn a state's right-to-work law via initiative.220

Unlike the previous propositions, the Arizona Works Together campaign ultimately fell short of the needed signatures to appear on the 2024 ballot.221 But the Act represented a first-of-its-kind application of union-led direct democracy. Building offLabor's coalitional experiences with Propositions 202 and 206, the campaign sought to alter hostile, anti-union constitutional structures by rolling back right to work.222 Had it been successful, the campaign would have knocked down barriers for the Arizona labor movement to grow its ranks and bargain more effectively.223

The campaign's failure to collect enough signatures may imply limitations for union-led direct democracy that Part IV will explore. While it is too soon to definitively explain the loss, its failure may have also been attributable to its poor timing, as most resources in that electoral cycle were concentrated on the presidential and Senate campaigns, as well as Proposition 139, a high-profile measure that codified a right to abortion.224 And the Arizona labor movement itself may not have been especially focused on the Arizona Works Together Act, as some of the state's largest unions did not throw their weight or membership behind the effort.225 Though the state's labor coalition fell short with the Arizona Works Together Act, one thing is certain: Union-led direct democracy has fundamentally shifted the material conditions for hundreds of thousands of Arizonans.

III. Good for Unions, Good for Direct Democracy

The case studies demonstrate that union-led direct democracy is a viable pathway for unions to change the legal order. This Part argues that the use of direct democracy to bolster organized labor is a normatively positive advance, creating opportunities for a stronger, broader labor movement. Furthermore, certain features of union-led campaigns may curb common defects in direct democracy, such as information problems in the electorate and the influence of profit-driven special interests. And by relying on the electorate, unions have successfully translated popular support for economic justice into a more inclusive labor movement covering workers inside and outside unions. As often applied, this model almost certainly evades federal preemption. As a result, direct democracy can be good for unions-and unions, in turn, can be good for direct democracy.

A. Strengthens Organized Labor and Workers' Rights

Union-led direct democracy strengthens the labor movement through at least three broad categories of mechanisms: leverage and pressure; altering statutory and constitutional structures; and raising the floor for all workers. These categories can variably apply toward union organizing, bargaining, or both, and thus provide opportunities for bigger and stronger unions. The case studies illustrate all three categories in action.226

1. Leverage and pressure

First, union-led direct democracy creates new forms of leverage and pressure on employers and subnational governments. During collective bargaining, leverage and pressure may strengthen Labor's hand by providing new chips for unions to exchange with employers.227 The Hotel Worker Protection initiative and Arizona's Proposition 206, for example, both included supersession clauses that favored collective-bargaining agreements and allowed unions to negotiate beneath or above those initiatives.228

Leverage and pressure can also directly produce more dues-paying union members.229 In the second case study, this mechanism allowed Local 11 to achieve an ordinance that effectively gives the City Council veto power over new hospitality projects.230 Because of Local 11's strong influence on the City Council, the ordinance provides the union with an additional tool against developers who oppose or interfere with organizing efforts.231 Union-led direct democracy could thus be deployed to strongarm employers into cardcheck neutrality and first contracts, compensating for the NLRA's feeble remedies for employer intransigence.232

Unions can also deploy direct democracy to raise costs for employers. This form of leverage weakens employers' financial position to union-bust or engage in other hostile, bad-faith tactics. Though underdiscussed in the case studies, the Responsible Hotel Ordinance illustrates this dynamic.233 There, the explosive rooms-for-homeless initiative would have imposed massive costs on developers and hoteliers in Los Angeles-costs that could not easily be passed onto customers and that would have weakened the industry's ability to resist union campaigns.234 These potential costs encouraged the hotel industry to negotiate a compromise ordinance with Local 11.

2. Altering statutory and constitutional structures

Direct democracy provides a pathway for unions to alter hostile statutes and constitutional structures at the subnational level. As seen in Arizona, this mechanism is particularly useful in red or purple states with anti-worker state legislatures but more favorable general electorates; in states with extreme gerrymanders; and in states with pro-worker localities but more hostile state governments.235 This mechanism relies most crucially on the assumption that voters have policy preferences that are not aligned with those of their elected officials, creating a space for unions to mobilize popular support. While the direct material benefits for member and nonmember workers are fairly clear from the Arizona case study, the potential to reverse state right-to-work laws shows one way that unions can expand membership through this mechanism.236

3. Raising the floor for all workers

Direct democracy also strengthens the labor movement by raising the floor for all workers and thus raising the ceiling for union demands in contract negotiations. At first glance, it may not be obvious why raising base working conditions redounds to unions. Most union contracts require wages, benefits, and standards significantly above the floor that could be set by initiative. Creating a minimum wage and then raising the minimum wage in Arizona, for example, directly benefited "almost nobody who belong[ed] to a union," according to the lead coordinator of Propositions 202 and 206, former Arizona AFL-CIO Director Rebekah Friend.237 Instead, as she explained in an interview for this Note, raising the floor for all workers "helps you with bargaining . . . . When you go in to negotiate a contract, and people on the streets are making almost as much as your union members, it gives you leverage in bargaining."238 In other words, initiatives that benefit all workers also raise the ceiling for demands that unions can make of their employers. This third category parallels some of the qualities of the social bargaining model, where member and nonmember workers may develop shared interests and understandings of their material conditions.

B. Strengthens Direct Democracy

Unions can benefit from direct democracy, but the relationship also goes in the other direction. Because of their advantages in communication, voter turnout, and forging organic constituencies, union-led direct democracy may curb-if not cure-two common defects of direct democracy: information problems in the electorate and the influence of profit-driven special interests.

1. Informational deficiencies

Information problems proliferate in American direct democracy. Some initiatives contain technical, complex subject matter requiring a relatively well-informed citizenry.239 Published information on a proposed measure may be hard to access, and ballot language itself may be vague or confusing.240 Some states have tried countering these problems by mandating ballot pamphlets with clear arguments for each side, simple titles, and neutral descriptions.241 However, efforts to convey information are oftentimes contested and allegedly biased, leading to further voter confusion; indeed, at least one judge has openly critiqued a particular ballot summary, writing, "[t]he ballot label as currently drafted is misleading."242 Voter overload from too many ballot measures may depress voter turnout or interest in learning about any ballot measures.243 According to one poll from a nonpartisan think tank, for instance, three in four California likely voters find referendums "too complicated or confusing for voters to understand."244

Information problems in the electorate have caused major headaches for courts trying to interpret the meaning of ballot measures. Consider, for example, the Michigan Supreme Court's struggle to interpret a 2004 constitutional amendment passed by initiative that defined marriage as "the union of one man and one woman . . . for any purpose."245 In National Pride at Work, Inc. v. Governor of Michigan, the state supreme court had to decide whether the amendment prohibited public employers from providing health insurance benefits to qualifying employees' same-sex partners.246 Complicating the case was the fact that the counsel for the initiative's original political committee, Citizens for the Protection of Marriage (CPM), had explicitly "asserted that the amendment would not prohibit public employers from providing health-insurance benefits to domestic partners."247 CPM had widely distributed brochures stating that the initiative "is [o]nly about [m]arriage" and "not about rights or benefits";248 made public statements in national papers such as USA Today that "[t]his has nothing to do with taking benefits away";249 and had even declared at a public certification before the Board of State Canvassers that the amendment would not bar public employers from providing benefits to employees' same-sex partners.250 But if all this were not enough, the voters themselves seemed to favor the amendment only if it narrowly concerned itself with marriage and not benefits: Polling showed that while 50% of likely voters favored the amendment, 65% specifically disapproved of barring public employers from providing domestic-partner benefits.251

Nonetheless, upon review, the Michigan Supreme Court interpreted the proposition as disallowing public employers from providing benefits to samesex civil unions.252 The court disregarded voters' common understanding of the initiative because "[t]he role of this Court is not . . . to speculate about how these statements may have influenced voters. Instead, our responsibility is, as it has always been in matters of constitutional interpretation, to determine the meaning of the amendment's actual language."253

While the National Pride court demonstrates the challenge of discerning popular intent based on the information problems surrounding voter-enacted laws, union involvement in direct democracy attempts to remedy informational deficiencies earlier in the electoral process, before courts attempt to divine voters' popular intent.254 Because of their reliance on strongly organized membership, unions effectively communicate the content, meaning, and importance of ballot measures. Empirical analyses similarly show that unions tend to significantly boost voter turnout, signifying a more informed and engaged electorate.255 Without the luxury of deep pockets, unions rarely have the means to blanket the airwaves with misleading ads. Instead, successful unions leverage their thousands of members to communicate in workplaces, churches, and communities to educate the electorate-thus channeling popular support by making their case directly in the streets, door-to-door, face-to-face.256

But as with many aspects of union-led direct democracy, Labor's ability to solve informational deficiencies may also be limited in many situations. In the context of California, for example, the vastness of the state combined with high costs of voter outreach and direct democracy procedures means that Labor will have less impact statewide than in local initiative campaigns. Even relatively well-organized unions in more manageable states like Arizona struggle to overcome information barriers in the electorate.257 As long as the labor movement remains diminished in size, it will have greater potential to curb informational problems in the electorate in local rather than statewide initiatives, and in cheaper, smaller states rather than more expensive, larger states.

2. Influence of special interests

Related to the informational issues from the previous Subpart, direct democracy can often devolve into a mere proxy for whichever side of a campaign has more money. As one scholar has noted, "money is virtually always sufficient for success" to qualify for the ballot;258 rich, powerful, and profit-seeking entities often use direct democracy to forward their own narrow self-interests.259 In one particularly egregious example, Uber and Lyftspent over $205 million-the most expensive initiative campaign in California history-to pass Proposition 22, creating a special carveout for those companies to classify their workers as independent contractors rather than employees.260 Unsurprisingly, a recent poll found that a large portion of likely California voters believed that "the referendum process in California is controlled by special interests-either 'a lot' (56%) or 'some' (40%)."261

Special interests have become so dominant in direct democracy that a professionalized "initiative industrial complex" has arisen to help prospective clients access the ballot.262 Procedurally, campaigns must receive a statutorily defined number of signatures in order to qualify for the ballot.263 Signature requirements, in theory, demonstrate broad popular support for an initiative, justifying a local or statewide vote.264 But rather than relying on organic and authentic popular interest, special interests plow huge sums of money to pay signature-gathering companies and circulators to contrive public support- sometimes, even lying to voters about the actual purpose of an initiative petition to wring out signatures.265 Direct democracy has become so thoroughly commodified that costs per collected signature can vary depending on the estimated wealth of the campaign-fluctuating, in one sampling, from a low of $3.42 per signature for California Proposition 15 changing commercial property tax structure to a high of $11.87 for San Francisco District Attorney Chesa Boudin's recall.266

Unions serve as a natural counterweight to the influence of special interests in direct democracy due to their natural constituencies. As shown in Part II above, union-led coalitions mobilize their membership to quickly gather signatures and then build support and awareness among the broader electorate.267 By relying on their organized membership as an "army" of volunteer signature gatherers, unions can channel authentic popular support without resorting to the lies and misinformation common to the paidsignature- gathering industry.268 And while unions may occasionally buttress their volunteers with some professional direct democracy services, their limited resources incentivize unions to make the most of their deep support among member and nonmember workers.269

There are some who claim that the labor movement itself is a special interest no different from corporations, but these arguments (if taken in good faith) are misleading for at least two reasons.270 First, because of the constraints of the NLRA and its strict preemption doctrine discussed in Part I above, unions cannot propose initiatives that directly and specifically strengthen their own position through the law.271 While Uber and Lyftare free to drop $200 million to create special carveouts from California's employment laws for their own drivers, the rules in Garmon, Machinists, and Gould prevent unions from doing anything so brazenly self-serving.272 Direct democracy, as a result, structurally provides corporations with far more opportunities for selfinterested action than it does for unions, which can only operate within the narrow band of preemption exceptions.

Second, unions are not special interests because they do not act solely-or even mostly-in their own interests. The case studies in Part II above show that union-led direct democracy is often about benefiting all workers, union members and nonmembers alike, while tweaking the law to facilitate union organizing and bargaining.273 As the lead coordinator of Propositions 202 and 206 and former Arizona AFL-CIO Director Rebekah Friend explained in an interview for this Note, the minimum wage propositions her coalition championed impacted almost no union members. Instead, the purpose was to "raise all boats" for working Arizonans: "There was [a belief] that this was just the right thing to do. This is what the union movement is about: helping families."274

3. A political theory of union-led direct democracy

Direct democracy, in the forms discussed in this Note, originated from early twentieth-century efforts to counter the domination of state and local politics by corporate power. In California, for example, the initiative, referendum, and recall were born from Governor Hiram Johnson's populist platform opposing the influence of the Southern Pacific Railroad.275 First elected in 1910, Governor Johnson explained the impetus behind California's direct democracy in his first inaugural address:

How best can we arm the people to protect themselves hereafter [from the railroad corporations]? If we can give to the people the means by which they may accomplish such other reforms as they desire, the means as well by which they may prevent the misuse of the power temporarily centralized in the Legislature, and an admonitory and precautionary measure which will ever be present before weak officials . . . . This means for accomplishing other reforms has been designated the "Initiative and the referendum," and the precautionary measure by which a recalcitrant official can be removed is designated the "Recall."276

At its best, direct democracy provides the electorate with the necessary tools-a sword in the form of the initiative and the referendum, and a shield in the form of the recall-to hold their elected officials accountable and responsive.277

But these tools, alone, are not enough to ensure voters' preferences are expressed at the polls: Political processes do not guarantee particular political outcomes. As the preceding Subparts discuss, direct democracy elections can be low-information events, often muddied by the involvement of special interests.278 Winning at the polls, according to the political theorist E.E. Schattschneider, requires a very different kind of calculus: "What happens in politics depends on the way in which people are divided into factions, parties, groups, classes . . . . Every shiftof the line of cleavage affects the nature of the conflict, produces a new set of winners and losers and a new kind of result."279

Union-led direct democracy is so powerful precisely because it can shiftthese cleavage lines in two ways.280 First, it creates and mobilizes new constituencies within the electorate. Indeed, "acts of representation can change the subject of politics in the fullest sense of the word: they change how people identify themselves by redefining what they care enough to fight for and whom they count as allies."281 The Responsible Hotel Ordinance study illustrates this dynamic. Though the housing crisis in Los Angeles remains the city's biggest challenge, the City Council arguably ignored the aggravating role of the booming hospitality industry-namely, the knocking down of affordable housing and consumption of scarce land for large, non-housing developments.282 When traditional methods of political pressure failed, Local 11 used direct democracy to make those political beliefs salient by representing workers and voters frustrated with the resultant impact of hotel development on housing affordability, homelessness, and gentrification. At first, the City Council disregarded Local 11's efforts, but when the union passed the petition hurdle to place its initiative on the ballot, the Council had little choice but to address and compromise on the issue.283 Confronted with a political grenade of an ordinance, the City Council quickly organized negotiations to find a political compromise that could vindicate the interests of workers and voters.284 By assuming the mantle of representation for all city residents concerned with affordable housing and major hotel developments, Local 11 shifted cleavage lines and mobilized new constituencies in its favor.

Second, union-led campaigns in direct democracy shiftcleavage lines through repeated acts of policymaking. These campaigns translate popular (but unsatisfied) demands for economic justice into the legal order.285 Over time, repeated policy successes create "feedback effects" that solidify the relationship between constituents and their representatives-thus fulfilling political demands that otherwise go unmet and delivering for constituents where elected officials fall short.286 Iterative policymaking goes hand-in-hand with the creation and mobilization of new constituencies to shiftpolitical cleavage lines.287The initiatives in Arizona demonstrate how repeated acts of policymaking remake political landscapes. There, the coalition of unions used a series of initiatives to form and reinforce new political blocs by establishing a minimum wage, raising the minimum wage, creating paid sick leave, and attempting to repeal Arizona's right-to-work status-in spite of a hostile, antiworker state legislature.288 Recent scholarship has traced the rise of a "new preemption" at the state level that not only removes entire subject areas from local regulation but also punishes local officials or governments with punitive measures, "effectively nuk[ing] local power . . . and existential[ly] threat[ening] local self-government."289 Local governments and officials can receive criminal and civil fines, state aid cutoffs, and liability if they cross into preempted policy areas.290 This aggressive and punitive form of state preemption arises directly from the "interacting polarizations of Republican and Democrat, conservative and liberal, and nonurban and urban . . . . The notso- irresistible force of cities pushing progressive agendas increasingly runs into the immovable object of conservative state resistance, manifested by aggressive preemption."291

And few states have seen as aggressive preemption as Arizona. In 2016, the Grand Canyon State passed Senate Bill 1487, frequently referred to as "the mother of all local preemption bills,"292 which empowered the state attorney general to investigate any official action by a local government that arguably contradicts state preemption law and withhold state funding from localities the attorney general considers to be out of compliance.293

The initiatives in the Arizona case study were carefully designed to overcome state preemption by remaking the typical cleavage lines between red legislatures and pro-worker localities.294 As the lead coordinator for Propositions 202 and 206 confirmed, those initiatives' home rule provisions were central to the coalition's strategy because of the state government's Republican tilt.295 Before these initiatives, statewide political dynamics submerged the interests of pro-worker localities.296 The multiyear direct democracy campaigns in Arizona, however, remade these cleavage lines by elevating unmet demands in the electorate that sought to expand workers' rights-demands that were ignored or opposed by the Legislature.297

Direct democracy can be a powerful tool to bridge the gap between what the electorate wants and what elected officials do. Mere direct democracy or union involvement is not sufficient: Truly successful union-led campaigns utilize mobilization and policymaking to "politicize lines of conflict that mute long-standing cleavages and form new ones to remake patterns of alliance and enmity."298 In Los Angeles, this meant activating new constituencies and representing submerged concerns about hotel development; in Arizona, this meant solidifying new political blocs around economic justice over a series of interconnected initiatives. The effects of the Arizona campaigns continue to reverberate, pushing the state to the leftin several key elections whose candidates ran on economic justice and pro-worker platforms.299 By forcefully shifting the terms of political debate for elected officials and the electorate, union-led direct democracy has remade the bounds of constituencies and upended preexisting political arrangements.300

C. Avoids Federal Preemption

The most basic appeal of union-led direct democracy is its presumptive legality. To be sure, union campaigns in direct democracy are not per se legal under the current preemption doctrine: A ballot proposal to ban unions in a state, for example, would not be acceptable under the current preemption doctrine.301 Rather, the initiatives discussed in the case studies evade federal preemption for two reasons. First, they comply with the rules in Garmon, Machinists, and Gould since they do not even arguably cross into the NLRA's statutory provisions and do not utilize state powers in a way tantamount to regulation of industrial relations.302

Second-and more importantly-they exploit the doctrine's narrow exceptions by creating law that applies broadly to all workers. In clarifying the traditional police powers exception and the peripheral concerns exception, the Supreme Court has held that laws regulating general labor standards for unionized and nonunionized workers alike do not trigger federal preemption.303 In Fort Halifax Packing Co. v. Coyne, the Supreme Court found that a Maine statute requiring employers to provide severance payments to employees in the event of a plant closing was not preempted by the NLRA because "its establishment of a minimum labor standard does not impermissibly intrude upon the collective-bargaining process."304 The Court reasoned that even though "the Maine statute gives employees something for which they otherwise might have to bargain," the NLRA itself was "concerned with ensuring an equitable bargaining process, not with the substantive terms that may emerge from such bargaining."305 The Court also emphasized that "pre-emption should not be lightly inferred in this area [relating to state regulation of labor standards], since the establishment of labor standards falls within the traditional police power of the State."306 Fort Halifax relied on the exceptions discussed in Part I above, which protect the traditional police powers of the state and statutes peripherally touching upon the NLRA.307 The initiatives in the case studies similarly slip the constraints of federal preemption because they merely regulate minimum labor standards like the Maine law at issue in Fort Halifax.308

To the extent that the Supreme Court has provided guidance on the contours of a minimum labor standard, the initiatives discussed in the prior Subpart pass with flying colors. In Metropolitan Life Insurance v. Massachusetts, the Court defined minimum state labor standards as "affect[ing] union and nonunion employees equally, and neither encourag[ing] nor discourag[ing] the collective-bargaining processes that are the subject of the NLRA."309 The Hotel Worker Protection Ordinance from the first case study applied to all domestic workers in the hotel industry, the Responsible Hotel Ordinance applied to all hotels of a certain size, and Propositions 202 and 206 in Arizona applied to all employers of a certain size.310 Because they regulate labor standards with general applicability-and neither encourage nor discourage collective bargaining processes-these initiatives almost certainly evade federal preemption under existing doctrine. Limited case law in the lower courts similarly suggests that challenges to state or local labor laws that regulate basic labor standards, passed by initiative or by legislature, almost universally survive preemption.311

Collective-bargaining supersession clauses are a closer call than minimum labor standards. However, these provisions also survive preemption under the Supreme Court's jurisprudence, at least for now.312 In a footnote to Livadas v. Bradshaw, the Court rejected the contention that the NLRA preempts these supersession provisions: "Nor does it seem plausible to suggest that Congress meant to pre-empt such opt-out laws, as 'burdening' the statutory right of employees not to join unions by denying nonrepresented employees the 'benefit' of being able to 'contract out' of such standards."313 The Ninth Circuit and courts within have relied on Livadas to reject two separate challenges to these collective-bargaining supersession clauses (both involving California ordinances backed by UNITE HERE Local 11).314 Though not per se guaranteed by the preemption doctrine, union-backed campaigns in direct democracy will remain legal and available for subnational labor lawmaking as long as they comply with the rules in Garmon, Machinists, and Gould, and fall within the doctrine's clear exceptions.315

IV. Implications and Limitations of Labor's Role in Direct Democracy

Parts II and III above make the case that direct democracy can be good for unions, and that unions can be good for direct democracy. The strategy has delivered major victories in economic justice for hundreds of thousands of member and nonmember workers across Southern California and Arizona.316 It provides tools and points of leverage for unions to organize new workers and bargain more effectively with employers.317 So far, it has succeeded in evading federal preemption-and in beating back hostile state legislatures and constitutional structures.318 The strategy also jumpstarts unresponsive political processes and systems by directly translating popular support for economic justice into the legal order.319 And as a means of more honestly conducting direct democracy, union involvement may curb information and special interest problems.320 However, the strategy has several limitations, ranging from procedural and substantive rules to more practical considerations. This Part argues that union-led direct democracy should be seen as one possible tool for unions to further their aspirations and strategies, rather than as a cure-all to revive the labor movement.

A. Procedural and Substantive Limitations in Direct Democracy

Union-led direct democracy is restricted by various procedural and substantive limitations. First, the labor movement cannot use direct democracy in states that do not have direct democracy. Twenty-three states have either the statutory or the constitutional initiative (or both).321 For the twenty-six states without any kind of initiative at all, union-led direct democracy either cannot function or must rely on (often) hostile legislatures for referenda. Union-led direct democracy, then, suffers from a scalability issue nationwide.

Second, doctrinal or constitutional bars without federal analogues also exist in almost every direct democracy state, which could be used to stymie unions. Clear-title, single-subject, original-purpose, and public-purpose rules are the most common.322 Largely self-explanatory, these rules, respectively, require that an initiative's title be clearly understood; that the initiative be concerned only with a single subject or topic; that a proposed law not be altered too drastically over the course of its passage such that it loses its original purpose; and that the law serve a public purpose rather than private gain.323 These rules may be required as a matter of constitutional, statutory, or state common law, depending on the state.324 Furthermore, even in states with these rules, they may apply exclusively to the legislature or certain direct democracy processes. For instance, Arizona's single-subject rule does not apply to voter-approved initiative measures and only applies to acts of the state legislature325-but it does apply to referenda originating in the state legislature sent to the people for approval.326

Generally, these rules are meant to improve transparency, fairness, and democratic accountability in the state legislative process, but the defining feature of these procedural limitations is their doctrinal flabbiness. Courts inconsistently enforce these rules and inconsistently apply rationales to invalidate laws.327 And as direct democracy has become more popular in recent years, litigants have become more incentivized to use these low-risk, lowreturn claims as a final opportunity to attack legislation.328 Indeed, Proposition 206 from the Arizona case study highlights how litigants deploy these tools.329 Opponents to Proposition 206 alleged that the proposition violated the state's constitutional single-subject rule and another related procedural bar, the separate-amendment rule.330 Those claims were tossed out by the court, but it is not a stretch to see how elected, partisan state judges with more ideological motivations could rule otherwise in cases of union-led direct democracy.331

Beyond procedural limitations, union-led direct democracy faces substantive limitations due to federal preemption, as already discussed in depth.332 Under existing preemption doctrine, direct democracy cannot rewrite the NLRA's rules of organizing for unions or force employers to abide by elections.333 For the most part, union-led direct democracy depends on raising the floor for all workers, while creating leverage points for unions to organize and bargain, either with employers (first case study), with government (second case study), or by altering constitutional structures such as home rule, state preemption, and right to work (third case study).334 But points of leverage are likely less powerful than the private reorderings in the tripartite model-and potentially less expedient than wage boards in social bargaining.

B. Rising Costs and the "Professionalization" of Direct Democracy

Rising costs associated with the professionalization of the initiative process may further limit the usefulness of direct democracy for labor in the coming years. Even two decades ago, academics and analysts voiced concerns about the emerging "initiative industrial complex" in California-a constellation of professional signature-gathering companies, political consultants, lawyers, advertisers, and deep-pocketed interests that put the initiative out of reach for many organic democratic interests.335 From all current indications, California is merely ahead of the curve rather than an outlier: Arizona's initiative in recent years has professionalized and become significantly more expensive as well.336

And as initiatives become more expensive due to professionalization, Labor may find itself outmatched by an increasingly organized and wealthy opposition. In 2018, SEIU put forward Proposition 8, a California ballot initiative that would regulate kidney dialysis clinics, cap profits, and limit patient expenditures. Proposition 8 raised the most money for one side of an initiative in history at that point.337 Labor groups contributed over $18 million in support of the campaign in the hopes of weakening the dialysis duopoly in California and facilitating union organizing, while the opposition spent $111 million-including $100 million just from DaVita and Fresenius, the duopoly that controls about 70% of the market in California.338 Two years later, the aforementioned Proposition 22 broke the record again for the most expensive initiative in history as Uber and Lyftspent over $200 million (ten times more than Labor's spending) to create a special carveout to classify their workers as independent contractors rather than employees.339 In both initiatives, Labor lost against a tidal wave of corporate spending that outspent them several times over.340

As these trends continue, cash-strapped unions may find themselves increasingly unable to mount serious initiative campaigns, at least at the state level. The future of union-led direct democracy, accordingly, may largely take place at the local level, where direct democracy remains more affordable and accessible. Indeed, the lead coordinator of the Arizona minimum wage initiatives acknowledged as much, noting that "it's become much more difficult, because it's become much more expensive" to mount successful statewide campaigns.341

C. Direct Democracy's Discontents

While unions are having a moment of nationwide popularity, local populations antagonistic to the labor movement may vote against worker rights at the ballot box.342 Indeed, the potential for reactionary outcomes in direct democracy is not a new concept: As Derrick Bell Jr. pointed out nearly half a century ago, at the height of the Boston busing crisis, voters in direct democracy do not face the same moderating incentives as elected officials to deliberate, compromise, and consider community input from minority factions.343 "[B]ecause it enables the voters' racial beliefs and fears to be recorded and tabulated in their pure form, the referendum has been a most effective facilitator of that bias, discrimination, and prejudice which has marred American democracy from its earliest day."344 Writing in the 1970s, Bell was rightly horrified by the prospect of direct democracy as he witnessed white Bostonians chuck rocks at the heads of Black children bused to newlyintegrated schools.345 And these fears continue to have salience: Only a few decades ago, 60% of Californians voted for a proposition steeped in anti-Latino rhetoric to ban nearly all bilingual education in the state.346 Even more perplexing, Californians in 2024 rejected Proposition 6, which would have banned slavery and involuntary servitude as punishment in the state's prisons.347

But Bell's assumption that voters will use the anonymity of the ballot box to inevitably vote for their worst impulses is wrong.348 Granted, direct democracy without guardrails is a high-risk proposition, and this Note does not remotely argue that unrestrained direct democracy on its own leads to progressive outcomes. Rather, one argument of this Note is that outcomes in direct democracy elections are reflective and contingent upon political cleavages, and that union-led campaigns can shiftthose cleavage lines.349 Bell appears to accept voters' preferences as a given: Constituencies preexist elections, and their preferences cannot be meaningfully changed.350 But the political theorist Lisa Jane Disch argues differently:

[A] constituency is not a demographic fact but a political achievement. Constituencies are not simply 'out there,' except in the most formal sense as an aggregate defined by district lines. Acts of representation call them into being. They take shape out of an amorphous plurality of social relations when a movement, a leader, a message, or an event solicits their participation in [a political] conflict.351

Union-led direct democracy is so powerful precisely because it can re-create constituencies out of the electorate where none existed previously.352 There is nothing per se in direct democracy that suggests voters will support proworker causes. But as discussed above, when unions mobilize members and voters, they redefine the lines of political conflict.353 Policy successes feed back into representation, entrenching new groups into existence.354 And this is what Bell misses: While some constituencies, in some places, under certain conditions will vote for reactionary policies, these constituencies are not fixed and do not preexist the cleavage lines that politicians, activists, unions, and others create in the public realm. Nothing in democracy, direct or otherwise, is inevitable.

Conclusion

Until the 2000s, the labor movement rarely utilized direct democracy to advance its agenda.355 When it did participate, its involvement was defensive, opposing measures proposed by hostile adversaries, rather than actively promoting its own preferred laws.356 This began to change in the mid-2000s as unions began to compete in direct democracy-a realm that had previously been associated with right-wing populist measures such as the taxpayer revolts of the 1970s.357 This uncertain foray marked the emergence of a "new pattern": union-led direct democracy.358

Since those first statewide initiatives nearly two decades ago, the labor movement has refined its approach to direct democracy. Today, union-led direct democracy, as described in this Note, creates subnational labor and employment law, strengthening Labor's hand in organizing, bargaining, and establishing workers' rights. Because of the constraints of labor law preemption, these laws must conform to narrow doctrinal rules and apply broadly to all or most workers, such that they cannot be read to conflict with the NLRA.359 To win their campaigns, unions mobilize their members into low- or no-cost canvassing armies that quickly collect petition signatures, educate the broader electorate, and drive up voter turnout.360 And through repeated acts of representation, unions solidify broader political blocs of member and nonmember workers in direct democracy campaigns.361

Union-led direct democracy has myriad strategic benefits. It has improved the working conditions of millions of member and nonmember workers through a diverse array of popularly-chosen laws-from panic buttons for vulnerable hotel workers to establishing state minimum wages and paid sick leave.362 This model of subnational labor lawmaking succeeds in the bluest localities as well as in right-to-work states with hostile laws and legislatures. Because of its deeply grassroots nature, union-led campaigns curb common issues in direct democracy, such as information deficiencies and special interest influence. And its scale and potential applicability across twenty-three states and thousands of localities means that union-led direct democracy almost certainly dwarfs all other forms of subnational labor lawmaking (such as tripartism and social bargaining) combined.363

The second Trump administration is in its early days. If some indicators are to be believed, the federal government could enter a period of renewed apathy (or antipathy) to workers.364 Nearly half a century ago, in the wake of a rightwing turn at the Supreme Court, Justice Brennan contemplated the nature of our federal system and subnational governments as a "double source of protection" for Americans:

State constitutions, too, are a font of individual liberties, their protections often extending beyond those required by the Supreme Court's interpretation of federal law. The legal revolution which has brought federal law to the fore must not be allowed to inhibit the independent protective force of state law-for without it, the full realization of our liberties cannot be guaranteed.365

The right to industrial democracy may come under attack at the federal level, but it can be strengthened and multiplied across subnational governments through union-led direct democracy. Given the deep uncertainty of the coming years, this may be one of Labor's best chances for survival-and eventual resurgence.366

Footnote

1. William E. Forbath, The Shaping of the American Labor Movement, 102 HARV. L. REV. 1109, 1112-13 (1989). By extending Forbath's discussion, this Note endorses his view that the legal order-broadly speaking, any interaction with statutory regimes, the judicial system, and enforcement-has played a major role in shaping and developing the American labor movement and its strategies. See id. at 1116.

2. WILLIAM E. FORBATH, LAW AND THE SHAPING OF THE AMERICAN LABOR MOVEMENT 38- 39, 59-97, 105-18 (1991) (providing the classic account of late nineteenth century and early twentieth century conflicts between labor and the judiciary-as well as the courts' use of police, troops, and guards to break labor actions-resulting in a distinctly less radical labor movement); see also Lochner v. New York, 198 U.S. 45, 52-53 (1905) (invalidating a law related to maximum hour regulations and ushering in an era of frequent judicial invalidation of state and federal economic legislation); FORBATH, supra, at 38 ("By the turn of the [nineteenth] century state and federal courts had invalidated roughly sixty labor laws. During the 1880s and 1890s courts were far more likely than not to strike down the very laws that labor sought most avidly.").

3. See 29 U.S.C. §§ 151-169 (codifying the right to strike). The NLRA categorically exempts certain kinds of workers from coverage, including public-sector employees, agriculture workers, domestic workers, independent contractors, and individuals employed by their parents or spouses, among others. Id. § 152. Since the early 2000s, the National Labor Relations Board (NLRB) has asserted jurisdiction over labor relations in tribal enterprises but has interpreted the NLRA as disallowing jurisdiction over traditional tribal or governmental functions. See Note, Tribal Power, Worker Power: Organizing Unions in the Context of Native Sovereignty, 134 HARV. L. REV. 1162, 1168-69 (2021) ("In 2004, the Board formally overruled Fort Apache to hold that Native nations are 'employers' under section 2(2) of the Act . . . . The San Manuel Board nonetheless held that a 'blanket assertion of jurisdiction' over Native nations was inappropriate as a policy matter. It therefore introduced a new rule: when Native nations operate in the 'particularized sphere of traditional tribal or governmental functions,' the Board should decline jurisdiction." (quoting San Manuel Indian Bingo & Casino, 341 N.L.R.B. 1055, 1063 (2004), aff'd, 475 F.3d 1306 (D.C. Cir. 2007))).

4. See, e.g., Benjamin I. Sachs, Despite Preemption: Making Labor Law in Cities and States, 124 HARV. L. REV. 1153, 1162-63 (2011); Paul Weiler, Promises to Keep: Securing Workers' Rights to Self-Organization Under the NLRA, 96 HARV. L. REV. 1769, 1769-70 (1983). Weiler argues, in one of the most cited articles in labor law, that the NLRA's weak remedies, slow processes, delays, and procedural form allow employers to disregard workers' labor rights, and thus have directly contributed to the decline of union density from a high of around 35% in 1945 to less than 25% in 1980, shortly before the article was published. Weiler, supra, at 1771-1803. Union density has continued to fall precipitously and Congress has failed to pass labor law reform in the decades since, driving union density down to 10% in 2023. See infra text accompanying note 14. For a more structural critique of the NLRA and labor rights in general, see Matthew Dimick, Counterfeit Liberty, CATALYST (2019), https://perma.cc/F49X-DTNJ ("Even labor law scholars, who once viewed labor law as a path of liberation for the labor movement, now see it as an ossified millstone around its neck. Recommendations for the reform and renewal of labor law therefore abound. In nearly all of these recommendations, there is no question that the law can and should play a fundamental role in revitalizing the labor movement. Indeed, labor law's current flaw according to these recommendations is not the rights they provide, but only the 'weakness' of these rights.").

5. See, e.g., 29 U.S.C. § 160(c), (j); Weiler, supra note 4, at 1776-81 (detailing patterns of employer intimidation that had become common and enabled by the NLRA); id. at 1788-91 (finding that "the traditional remedies for discriminatory discharge-backpay and reinstatement-simply are not effective deterrents to employers who are tempted to trample on their employees' rights" because the costs of unionization, assuming a 20% union effect on wages, are far more expensive than the potential liability from terminating troublesome employees); DENNIS NOLAN, RICHARD BALES, RAFAEL GELY & BRAD AREHEART, LABOR LAW: COLLECTIVE BARGAINING IN A FREE SOCIETY 482-83, 489- 90 (8th ed. 2024) (describing the extraordinary weaknesses of NLRB remedies such as cease-and-desist orders, reinstatement, back pay, and notice orders). The NLRB also has the power to order employers to bargain with and certify unions, even without an election or when the election is lost, if the Board finds that "an employer's unfair practices in terms of their past effect on election conditions and the likelihood of their recurrence in the future" would "make a fair election an unlikely possibility." NLRB v. Gissel Packing Co., 395 U.S. 575, 579, 610, 614-15 (1969) (upholding a bargaining order in the case of one election lost by a union and endorsing the NLRB's authority to "issue a bargaining order without first requiring the union to show that it has been able to maintain its majority status"). Known as a Gissel order, this remedy has been greatly narrowed across the federal courts of appeals. Peter J. Leff, Failing to Give the Board Its Due: The Lack of Deference Afforded by the Appellate Courts in Gissel Bargaining Order Cases, 18 LAB. LAW. 93, 97-112 (2002). Under the Biden administration, the NLRB announced a new standard that dramatically expands application of Gissel orders. Cemex Constr. Materials Pac., LLC, 372 N.L.R.B. No. 130, 25-26 (Aug. 25, 2023) ("Under the standard we adopt today . . . . if the employer commits an unfair labor practice that requires setting aside the election, the petition (whether filed by the employer or the union) will be dismissed, and the employer will be subject to a remedial bargaining order."). Given the instability in administrative law and the second Trump administration, it is unclear whether the Cemex framework will endure.

6. See Republic Steel Corp. v. NLRB, 311 U.S. 7, 10-11 (1940).

7. See, e.g., Weiler, supra note 4, at 1776-81 (discussing the impact of delays in the administration of the law); Gay Semel, Viewpoint: The NLRB Is Underfunded and Understaffed-And That's a Big Threat to the Current Organizing Wave, LABOR NOTES (July 6, 2022), https://perma.cc/V89N-HWJM (arguing, for example, that Amazon's procedural tactics in delayed NLRB proceedings have undermined organizing efforts); see also Michael Cetta, Inc., 373 N.L.R.B. No. 129, 1 n.6, 4-5 (2024) (presenting a relatively straightforward case in which the NLRB took nearly nine years from the initial NLRB complaint to determine that a restaurant had illegally fired thirty-six striking workers and the precise amount of make-whole backpay owed to the workers).

8. Weiler, supra note 4, at 1769.

9. Sachs, supra note 4, at 1154-55, 1172 ("There is no question that labor law's exceptionally expansive preemption regime precludes state and local governments from attempting directly to achieve at a local level the reforms that have been blocked in Congress.").

10. 29 U.S.C. § 164(b); Timothy Noah, How State Right-to-Work Laws Screw the Working Class, NEW REPUBLIC (Sept. 13, 2023), https://perma.cc/PWG2-XM5S. One leading textbook has called the naming conventions for the labor statutes a "terminological mess." NOLAN ET AL., supra note 5, at 171. While the original NLRA was incorporated into the Labor Management Relations Act of 1947 (also known as the Taft-Hartley Act), and while the Labor Management Reporting and Disclosure Act of 1959 (also known as the Landrum-Griffin Act) further amended the labor laws, the statutes collectively are still usually referred to as the NLRA, unless referring specifically to the 1947 or 1959 amendments. Id. This Note follows those naming conventions.

11. GARY GERSTLE, THE RISE AND FALL OF THE NEOLIBERAL ORDER: AMERICA AND THE WORLD IN THE FREE MARKET ERA 28 (2022).

12. See ELISE GOULD & WILL KIMBALL, "RIGHT-TO-WORK" STATES STILL HAVE LOWER WAGES 2 (2015).

13. Janus v. Am. Fed'n of State, Cnty., & Mun. Emps., Council 31, 138 S. Ct. 2448, 2459-60, 2486 (2018).

14. Union Members-2024, BUREAU OF LAB. STATS. (Jan. 28, 2025), https://perma.cc/DH72- R7G6; Dan Burns, US Union Membership Rate Hits Fresh Record Low in 2023-Labor Dept, REUTERS (Jan. 23, 2024, 1:03 PM PST), https://perma.cc/9W7X-D3YD.

15. See infra Part I.A.

16. See, e.g., Sachs, supra note 4, at 1155-56; Kate Andrias, The New Labor Law, 126 YALE L.J. 2, 7-8 (2016).

17. This Note adopts Jane Schacter's definition of direct democracy as procedural mechanisms. See Jane S. Schacter, The Pursuit of "Popular Intent": Interpretive Dilemmas in Direct Democracy, 105 YALE L.J. 107, 113 (1995).

18. See, e.g., Kate Andrias, The Perils and Promise of Direct Democracy: Labour Ballot Initiatives in the United States, 34 KING'S L.J. 260, 260, 263-69 (2023) (discussing Proposition 22- "which exempted app-based drivers from" a California labor law that extended employee status to workers-as a starting point to understand the role of initiatives in setting pro-labor policy, and assessing whether ballot initiatives can be an important tool for workers); John G. Matsusaka, Is Direct Democracy Good or Bad for Corporations and Unions?, 66 J.L. & ECON. 83, 84 (2023) (asking whether direct democracy benefits corporations or unions more); Daniel J.B. Mitchell, Unions and Direct Democracy in California: A New Pattern Emerging?, CAL. POL'Y OPTIONS, 2008, at 197, 197, 207 ("[A]lthough direct democracy has been available for political interest groups for almost a century in California, unions have exhibited only limited use of it . . . . The question raised by [a 2007 union-backed initiative] . . . signif[ies] a new interest by labor unions in proposing ballot measures."); see also Andrew Elmore, Labor's New Localism, 95 S. CAL. L. REV. 253, 258-59, 269, 302 (2021). Elmore's discussion focuses on the rise of local labor lawmaking in cities, featuring alliances between unions, worker centers, and community organizations. While Elmore does not explicitly define "direct democracy," he appears to be referring to the ability of these alliances to partake in direct action and participation and pressure their local governments, such as in protests and social movements or local decision-making processes in city government. Id. at 269, 302. This Note, by contrast, discusses "procedural" direct democracy-that is, formal tools for voters to directly express their policy preferences at the ballot box, such as the various forms of the initiative, referendum, recall, and more. Though the same name, these are very different concepts. Furthermore, Elmore's discussion of preemption mainly refers to state preemption against cities-a topic discussed in this Note-but he does not cover federal preemption issues. Id. at 276-91.

19. See ALLIE BOLDT, DIRECT DEMOCRACY IN THE STATES: A 50-STATE SURVEY OF THE JOURNEY TO THE BALLOT 5 (2023) (finding the constitutional or statutory initiative in Alaska, Arizona, Arkansas, California, Colorado, Florida, Idaho, Illinois, Maine, Massachusetts, Michigan, Missouri, Montana, Nebraska, Nevada, North Dakota, Ohio, Oklahoma, Oregon, South Dakota, Utah, Washington, and Wyoming, and excluding Mississippi because it has lost the power of the initiative); Jessica Bulman-Pozen & Miriam Seifter, The Right to Amend State Constitutions, 133 YALE L.J. F. 191, 192 (2023). For an alternative metric, Kate Andrias found in 2023 that about 70% of the population lived in a state or city where some kind of direct democracy procedure was available. Andrias, supra note 18, at 269.

20. See infra Part II.C.2. While estimates of how many workers benefit directly from initiatives are hard to come by, the single proposition discussed below was estimated to have raised wages for over 700,000 workers and provided improved health benefits for nearly one million. See infra note 201 and accompanying text. Thus, the total number of workers nationwide who have benefitted from union-led campaigns is almost certainly in the millions.

21. See infra Part I.B.

22. While this estimate is speculative, it is still a conservative statement. Tripartism appears to be a highly episodic form of subnational labor lawmaking, dependent on the alignment of political and economic circumstances, whereas the social bargaining tendency remains nascent and contained to certain a few places. See infra Part I.B.

23. See infra Part III.C.

24. See infra Part II.C (showing that the Arizona unions purposefully designed their initiatives to beat back state preemption); Richard Briffault, The Challenge of New Preemption, 70 STAN. L. REV. 1995, 1997-2000, 2005-06 (2018) (discussing how some legislatures-and, especially pertinent for this Note, Arizona-have weaponized preemption against localities, particularly in red states with blue cities). Indeed, relevant for this Note, other scholars have called attention to Arizona as well. See, e.g., Richard C. Schragger, The Attack on American Cities, 96 TEX. L. REV. 1163, 1176 (2017).

25. See San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236, 246 (1959); Lodge 76, Int'l Ass'n of Machinists & Aerospace Workers v. Wis. Emp. Rels. Comm'n, 427 U.S. 132, 154-55 (1976); Wis. Dep't of Indus., Lab. & Hum. Rels. v. Gould Inc., 475 U.S. 282, 287-89 (1986).

26. See infra Part I.A.

27. See Elmore, supra note 18, at 256-57 (discussing broadly the shiftof "unions, worker centers, and allied organizations [toward] city-based strategies to strengthen and liftworkplace standards, using local law instead of or in addition to federal labor law, as 'labor localism,' and examines its effects on labor and local law").

28. Sachs, supra note 4, at 1155.

29. See id. at 1156 (illustrating tripartism with a similar example); see also NOLAN ET AL., supra note 5, at 468 (In a neutrality agreement, the employer agrees not to oppose the union's organizing campaign. . . . In a card-check agreement, the employer agrees to recognize and bargain with the union once the union presents evidence (usually signed authorization cards) of its majority status. . . . Neutrality agreements and card-check agreements usually go together [and] the combination constitutes a very effective means of organizing employees.").

30. Andrias, supra note 16, at 8-10. Andrias states that social bargaining is interchangeable with tripartism and corporatism. Id. However, this would contradict Sachs's understanding of tripartism as private agreements between parties rather than public processes. See Sachs, supra note 4, at 1156. For clarity and consistency across the understandings of both Sachs and Andrias, this Note retains separate analytical definitions for these concepts.

31. Fast Food Restaurant Industry Act, CAL. LAB. CODE § 1475(a)(1)-(2), (b) (West 2024).

32. For a history of the labor movement (including its violent suppression) during the late nineteenth and early twentieth centuries, see generally PHILIP DRAY, THERE IS POWER IN A UNION: THE EPIC STORY OF LABOR IN AMERICA 61-121 (2010) (describing a series of strikes in this period that often involved violent confrontations, including the Great Railroad Strike of 1877). See also WILLIAM G. ROSS, A MUTED FURY: POPULISTS, PROGRESSIVES, AND LABOR UNIONS CONFRONT THE COURTS, 1890-1937, at 20-21, 31, 200 (1994) (covering the related conflicts of the era while largely focusing on the interactions of these social movements and the courts); LAWRENCE M. FRIEDMAN, A HISTORY OF AMERICAN LAW 253-78 (3d ed. 2005) (speaking more broadly about the social, economic, and political instability of the late nineteenth century, including those related to labor).

33. See DRAY, supra note 32.

34. See, e.g., FRIEDMAN, supra note 32, at 420.

35. FORBATH, supra note 2, at 61-62 (estimating that at least 4,300 labor injunctions were issued between 1880 and 1930, with 15% to 46% of all sympathy strikes enjoined).

36. See, e.g., Loewe v. Lawlor, 208 U.S. 274, 285, 292, 305 (1908) (overturning a lower-court demurrer and finding that a complaint sufficiently alleged that the AFL's boycott constituted a restraint of trade under the Sherman Act, opening a line of attack on unions rooted in antitrust).

37. See, e.g., David Cole, Enemy Aliens, 54 STAN. L. REV. 953, 995-97 (2002) ("Many of the early targets [of the federal government against alien radicals] were members of the Industrial Workers of the World (IWW). Created in 1905, the IWW organized unskilled workers and was deeply critical of capitalism. After several prominent IWW strikes in 1916 and 1917, law enforcement turned to immigration law as a tactic for attacking the IWW.").

38. See FRIEDMAN, supra note 32, at 266 ("Judicial review of state statutes was a rare, extraordinary event in 1850; it was a common occurrence in 1900. What happened in state courts paralleled what happened in the federal courts."); FORBATH, supra note 2, at 38 (finding that "[b]y the turn of the nineteenth century state and federal courts had invalidated roughly sixty labor laws. During the 1880s and 1890s courts were far more likely than not to strike down the very laws that labor sought most avidly").

39. Lochner v. New York, 198 U.S. 45, 52-53 (1905); see DRAY, supra note 32, at 244-51.

40. FRIEDMAN, supra note 32, at 271 ("A passion for 'Lochnerism' was not evenly distributed across the country; the Illinois supreme court, for example, was exceptionally hostile to labor laws. Other state courts were more friendly. But what was general was the threat, the possibility of a contest in court.").

41. See FORBATH, supra note 2, at 38-49, 105-18.

42. FRIEDMAN, supra note 32, at 255-56, 264, 269-72.

43. Id. at 272. Other accounts of the impact of this era on labor disagree. Forbath's classic analysis finds that the judiciary's threats and interventions against both labor and proworker legislation reshaped and redirected American Labor away from its early, strident radicalism toward a strategy of minimal politics and minimal confrontation with the legal order. FORBATH, supra note 2, at 1-8, 37-38, 59-97; see also DRAY, supra note 32, at 243-51 (describing judicial activism against labor unions in the late nineteenth and early twentieth centuries).

44. See FORBATH, supra note 2, at 134 ("In some important ways, then, the courts had recast the outlook of labor's national leaders. The burden of the law's antagonism made abolishing the injunction and the courts' strict limitations on collective action the AFL's top political priority.").

45. See JACKSON H. RALSTON, STUDY AND REPORT FOR AMERICAN FEDERATION OF LABOR UPON JUDICIAL CONTROL OVER LEGISLATURES AS TO CONSTITUTIONAL QUESTIONS 1 (1919) (quoting AM. FED'N OF LAB., REPORT OF PROCEEDINGS OF THE THIRTY-EIGHTH ANNUAL CONVENTION 324 (1918)).

46. Id. (quoting AM. FED'N OF LAB., supra note 45, at 324).

47. Id. at 61-64.

48. See DRAY, supra note 32, at 456-57 (noting that President Roosevelt's landslide victory in 1936 persuaded the Supreme Court to refrain from obstructing the New Deal agenda).

49. 29 U.S.C. § 151.

50. Id. §§ 159-160; see Garner v. Teamsters, Chauffeurs & Helpers Loc. Union No. 776 (A.F.L.), 346 U.S. 485, 490 (1953) ("Congress did not merely lay down a substantive rule of law to be enforced by any tribunal competent to apply law generally to the parties. It went on to confide primary interpretation and application of its rules to a specific and specially constituted tribunal and prescribed a particular procedure for investigation, complaint and notice, and hearing and decision, including judicial relief pending a final administrative order. Congress evidently considered that centralized administration of specially designed procedures was necessary to obtain uniform application of its substantive rules and to avoid these diversities and conflicts likely to result from a variety of local procedures and attitudes toward labor controversies."); San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236, 242 (1959) ("But the unifying consideration of our decisions has been regard to the fact that Congress has entrusted administration of the labor policy for the Nation to a centralized administrative agency, armed with its own procedures, and equipped with its specialized knowledge and cumulative experience . . . .").

51. Garner, 346 U.S. at 488 ("[The NLRA] leaves much to the states, though Congress has refrained from telling us how much. We must spell out from conflicting indications of congressional will the area in which state action is still permissible."); see, e.g., Forbath, supra note 1, at 1218-20 (finding that "organized labor won over fifty anti-injunction laws from state legislatures during the first two decades of the [twentieth] century," among other state and local labor laws).

52. For conciseness, this review covers only a portion of NLRA preemption doctrine, and only insofar as it relates to federalism. While preemption analysis usually takes this vertical form of federal-state relations, labor law preemption is so broad that it even eclipses some horizontal actions taken by the executive branch. See, e.g., Chamber of Com. of the U.S. v. Reich, 74 F.3d 1322, 1334-39 (D.C. Cir. 1996) (holding that the NLRA preempted a presidential executive order that barred federal government contracts with employers that permanently replaced legally striking workers, and that NLRA preemption applies to "federal as well as state action").

53. Garmon, 359 U.S. at 245.

54. Lodge 76, Int'l Ass'n of Machinists & Aerospace Workers v. Wis. Emp. Rels. Comm'n, 427 U.S. 132, 144 (1976).

55. See Wis. Dep't of Indus., Lab. & Hum. Rels. v. Gould Inc., 475 U.S. 282, 289 (1986).

56. NOLAN ET AL., supra note 5, at 235 ("The Supreme Court created Garmon preemption to protect a nascent labor movement from destruction by potentially hostile and relatively uninformed state tribunals.").

57. 359 U.S. at 236.

58. Id. at 237.

59. Id.

60. Id. at 237-38.

61. Id. at 238.

62. Id. at 241-48.

63. Id. at 246.

64. Id. (emphasis added).

65. Id.

66. 29 U.S.C. § 164(1)(c), (2).

67. See NOLAN ET AL., supra note 5, at 234-35.

68. 427 U.S. 132, 135, 144, 147-48 (1976).

69. Id. at 133-34.

70. Id. at 134.

71. Id.

72. Id. at 135.

73. Id. at 135-36.

74. See id. at 155 ("The [Wisconsin Supreme] [C]ourt held further that the refusal to work overtime was neither arguably protected under § 7 nor arguably prohibited under § 8 of the federal Act, an analysis which, as developed, is largely inapplicable to the circumstances of this case . . . . [W]e hold further that the Union's refusal to work overtime is peaceful conduct constituting activity which must be free of regulation by the States if the congressional intent in enacting the comprehensive federal law of labor relations is not to be frustrated . . . ." (citation omitted)).

75. Id. at 140 n.4 (quoting Archibald Cox, Labor Law Preemption Revisited, 85 HARV. L. REV. 1337, 1352 (1972)).

76. Id. at 143-44 (citing NLRB v. Ins. Agents' Int'l Union, 361 U.S. 477, 498 (1960)).

77. Id. at 144, 150-51.

78. 475 U.S. 282, 288 (1986).

79. Id. at 283.

80. Id. at 287.

81. Id.

82. Id. at 289-90 ("Nothing in the NLRA, of course, prevents private purchasers from boycotting labor law violators. But government occupies a unique position of power in our society, and its conduct . . . is rightly subject to special restraints . . . . [I]t is far from unusual for federal law to prohibit States from making spending decision in ways that are permissible for private parties."). Most recently, the Supreme Court has reaffirmed-and, to some extent, expanded-Gould in Chamber of Commerce of the United States v. Brown, 554 U.S. 60, 64, 66 (2008), which struck down a California statutory scheme prohibiting some employers that receive state funds from using the funds to "assist, promote, or deter union organizing."

83. San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236, 247 (1959).

84. Id.

85. Glacier Nw., Inc. v. Int'l Bhd. of Teamsters Loc. Union No. 174, 143 S. Ct. 1404, 1413 (2023) (citing Bethany Med. Ctr., 328 N.L.R.B 1094, 1094 (1999) ("The Board has held concerted activity indefensible where employees fail to take reasonable precautions to protect the employer's plant, equipment, or products from foreseeable imminent danger due to sudden cessation of work.")).

86. Lodge 76, Int'l Ass'n of Machinists & Aerospace Workers v. Wis. Emp. Rels. Comm'n, 427 U.S. 132, 136 (1976) (quoting Garmon, 359 U.S. at 244).

87. Garmon, 359 U.S. at 243.

88. See 143 S. Ct. at 1417 (Thomas, J., concurring in the judgment).

89. Id. at 1413-14 (majority opinion) (noting that while some workers merely walked offthe worksite or coordinated with their employer before striking, others "abandoned their fully loaded trucks without telling anyone-which leftthe trucks on a path to destruction unless [the employer] saw them in time to unload the concrete").

90. Id. at 1414.

91. Id. at 1417 (Thomas, J., concurring in the judgment) (citation omitted) (quoting Merck Sharp & Dohme Corp. v. Albrecht, 139 S. Ct. 1668, 1681 (2019) (Thomas, J., concurring)).

92. Justice Thomas's hostility to the federal protection of workers' rights is almost too obvious to argue, but for a brief sampling, see United States v. Lopez, 514 U.S. 549, 584 (1995) (Thomas, J., concurring) (arguing for a return to a narrower, pre-New Deal doctrine for the Commerce Clause, which delivered opinions like Hammer v. Dagenhart, 247 U.S. 251, 275-76 (1918), in which the Court struck down federal child-labor laws); Allstates Refractory Contractors, LLC v. Su, 144 S. Ct. 2490, 2490 (2024) (Thomas, J., dissenting from the denial of certiorari) (calling into question the constitutionality of the Occupational Safety and Health Act, which regulates workplace safety and protections); and Sackett v. EPA, 143 S. Ct. 1322, 1358 (2023) (Thomas, J., concurring) (arguing for a pre-New Deal conception of the Commerce Clause that would revive the production/commerce distinction, in which the Commerce Clause permits regulating regulating only trade). See also Benjamin Sachs, Glacier and Justice Thomas' Preemption Breadcrumbs, ONLABOR (June 2, 2023), https://perma.cc/9GVC-X7HQ (speculating that Justice Thomas may be interested in replacing Garmon with a "strict form of conflict preemption," but that his concurrence in Glacier Northwest is too short and vague to know for certain).

93. See Glacier Nw., 143 S. Ct. at 1416-18 (Thomas, J., concurring).

94. Jonathan Berry, Department of Labor and Related Agencies, in HERITAGE FOUND., MANDATE FOR LEADERSHIP: THE CONSERVATIVE PROMISE 605 (Paul Dans & Steven Groves eds., 2023), https://perma.cc/JG2V-W3JM (supporting the issuance of preemption waivers from the NLRA and FLSA for state and local government). For an articulation of the pro-labor view for less preemption, see Moshe Z. Marvit, The Way Forward for Labor Is Through the States, AM. PROSPECT (Sept. 1, 2017), https://perma.cc/4Y7Y-2EQ8. Pro-labor critiques of the case mainly focus on how Glacier Northwest opens the door for employers to sue unions for economic losses during strikes. See, e.g., Jane McAlevey, How Should Workers Respond to the Supreme Court's Ruling in Glacier Northwest?, NATION (June 1, 2023), https://perma.cc/F72LF3YM.

95. Marvit, supra note 94.

96. Sachs, supra note 92 ("This is all on the order of speculation. We can't know what Thomas has in mind from [his Glacier Northwest concurrence]. We can't know who else might join him. We can't know how the doctrine would play out in real life. And given the malleability of all legal tests there's the real possibility that this new standard would be applied in asymmetric ways, permitting anti-union state laws while prohibiting pro-union ones.").

97. For a discussion of how Garmon preemption has evolved, see International Longshoremen's Association v. Davis, 476 U.S. 380, 393 (1986) ("[Post-Garmon cases] serve only as more precise demarcations of the scope of Garmon pre-emption. They have not redefined the nature of that pre-emption in any way. A claim of Garmon pre-emption is a claim that the state court has no power to adjudicate the subject matter of the case, and when a claim of Garmon pre-emption is raised, it must be considered and resolved by the state court.").

98. Sachs, supra note 4, at 1155.

99. See Andrias, supra note 16, at 8-10.

100. Sachs, supra note 4, at 1155.

101. The starting point here is that "[t]he powers not delegated to the United States by the Constitution, nor prohibited by it to the States, are reserved to the States respectively, or to the people." U.S. CONST. amend. X. Cities' powers to regulate in these areas vary according to the state constitution in question. See infra Parts II.C.3, III.A.

102. Sachs, supra note 4, at 1171-72 ("Thus, the type of political exchange involved in tripartite labor lawmaking can be understood as an example of this broader phenomenon, an example in which unions leverage their state and local political power-rather than their economic power-to secure employer agreement on an altered set of organizing and bargaining rules.").

103. See, e.g., Verizon Info. Sys., 335 N.L.R.B. 558, 559 (2001) ("The [National Labor Relations] Board will enforce [voluntary] agreements, including agreements that explicitly address matters involving union representation."); Penske Truck Leasing Co., 371 N.L.R.B. No. 113, 2-3 (2022) ("It is axiomatic that federal labor policy favors the arbitral resolution of disputes between parties to a collective-bargaining agreement and the parties agreed to follow that policy here. The Board too recognizes that national labor policy favors the honoring of voluntary agreements reached between employers and labor organizations, including agreements that explicitly address matters involving union representation.").

104. Verizon, 335 N.L.R.B. at 559.

105. Lodge 76, Int'l Ass'n of Machinists & Aerospace Workers v. Wis. Emp. Rels. Comm'n, 427 U.S. 132, 140 (1976) (quoting NLRB v. Nash-Finch Co., 404 U.S. 138, 144 (1971)).

106. See Sachs, supra note 4, at 1202-09 (summarizing concerns).

107. Id. at 1202.

108. Id. at 1204.

109. Id. at 1205.

110. Id. at 1208.

111. See, e.g., Chris Isidore, Amazon Faces New Union Test in North Carolina, CNN (Feb. 8, 2025, 2:00 AM EST), https://perma.cc/J9Y3-KMR4.

112. See supra Part I.B.1.

113. See Andrias, supra note 16, at 8-9.

114. 29 U.S.C. § 159(b).

115. Andrias, supra note 16, at 9 ("[Social bargaining] would locate decisions about basic standards of employment at the sectoral, industrial, and regional levels, rather than at the level of the individual worksite or employer.").

116. See id. at 68.

117. Kate Andrias, Social Bargaining in States and Cities: Toward a More Egalitarian and Democratic Workplace Law, HARV. L. & POL'Y REV. ONLINE, 2017, at 6-8.

118. Fast Food Restaurant Industry Act, CAL. LAB. CODE §§ 1474-1477 (West 2024); Lynn La, Fast Food Workers Get Raises from New California Law, CALMATTERS (Sept. 29, 2023), https://perma.cc/2FY9-LBEU.

119. Fast Food Restaurant Industry Act § 1475(a).

120. Id.

121. Andrias, supra note 117, at 7-8.

122. See Andrias, supra note 16, at 8, 52.

123. Id. at 91-92.

124. Id.

125. See supra Part I.A.

126. See Wis. Dep't of Indus., Lab. & Hum. Rels. v. Gould Inc., 475 U.S. 282, 289 (1986).

127. See Andrias, supra note 16, at 70-76.

128. Id. at 57-58 ("[T]he Fight for $15's leaders admit that they are aware of no clear path to unionization in its traditional sense . . . . [But] although the path to unionization is unclear, from close examination of the movements' efforts, a coherent vision of unionism-and of a legal framework to support it-emerges. That emerging framework rejects the old regime's commitment to the employer-employee dyad and to a system of private ordering. Instead, it locates decisions about basic standards of employment at the sectoral level and positions unions as social actors empowered to advance the interests of workers generally.").

129. Steven Greenhouse, 'The Success Is Inspirational': The Fight for $15 Movement 10 Years On, GUARDIAN (Nov. 23, 2022), https://perma.cc/YDH3-9UBR.

130. Alejandra Reyes-Velarde, California's Fast Food Workers Gain New, First-of-Its-Kind Union to Represent Them, CALMATTERS (Feb. 9, 2024), https://perma.cc/QY77-X5GT.

131. Id.

132. Andrias, supra note 16, at 70-76.

133. Who We Are, UNITE HERE! LOC. 11, https://perma.cc/ZV54-CNBY (archived Apr. 7, 2025).

134. Id.

135. See Sasha Abramsky, The New Labor Militancy of LA's Hotel Workers, NATION (Dec. 8, 2023), https://perma.cc/933Y-Q53K; 2023-2024 Contract Fight: Hotels with Contracts or Tentative Contract Agreements, UNITE HERE! LOC. 11, https://perma.cc/H6VA-9FFB (archived Apr. 7, 2025).

136. Abramsky, supra note 135.

137. See UNITE HERE Local 11 Endorses Hugo Soto-Martinez for Los Angeles City Council District 13, UNITE HERE! LOC. 11 (July 29, 2021), https://perma.cc/W9L9-5RQ4 ("As a server at the L.A. Grand Hotel in Downtown Los Angeles, Hugo led a successful union organizing drive. Since then, Hugo has dedicated the past 15 years to standing up for his colleagues and members . . . . As a leader of UNITE HERE Local 11, Hugo has played an instrumental role in efforts to mitigate the effects of illegal short-term rentals on the housing market.").

138. See Julia Wick, Jeremy Childs & Christian Martinez, L.A. City Council Members, California Assemblywoman Arrested as Hotel Workers Protest, L.A. TIMES (updated June 22, 2023, 11:12 PM PT), https://perma.cc/CU95-RX6Y.

139. L.A., CAL., MUN. CODE ch. XVIII, art. 2, §§ 182.00-182.10 (2024); see LA Hotel Workers' Petition for Better Conditions, Higher Pay, Moves Forward, KCAL NEWS (May 20, 2022, 6:34 AM PDT), https://perma.cc/E4QY-NB2M; UNITE HERE Local 11 Housekeepers Submit 100k Signatures on Groundbreaking Initiative to City of Los Angeles, UNITE HERE! LOC. 11 (May 2, 2022), https://perma.cc/3N77-5JUQ (discussing how "[o]ver one hundred housekeepers and other hospitality workers today turned in the petitions they have collected since late January to qualify their initiative for the November 2022 ballot").

140. L.A., CAL., MUN. CODE ch. XVIII, art. 2, §§ 182.00, 182.02(A)(1) (2022).

141. See Official Top Funders. Valid Only for April 2022: Workplace Security, Workload, Wage, and Retention Measures for Hotel Workers, UNITE HERE! LOC. 11, https://perma.cc/L6SW-C8QW (archived Apr. 7, 2025) (showing that Local 11 was a major funder of the initiative, and that petition circulation was paid for by Citizens for a Better Los Angeles 2022); see also UNITE HERE Local 11 Housekeepers Submit 100k Signatures on Groundbreaking Initiative to City of Los Angeles, supra note 139 (confirming that Local 11 and its membership were involved in signature-gathering for the petition).

142. See L.A., CAL., MUN. CODE ch. XVIII, art. 2, §§ 182.00, 182.01(M), 182.02 (2022).

143. See id. § 182.00.

144. See id. § 182.02.

145. See sources cited supra note 141 (describing that Local 11 membership helped organize the initiative).

146. L.A., CAL., MUN. CODE ch. XVIII, art. 2, § 182.09 (2022).

147. See CAROL ZABIN & ISAAC MARTIN, LIVING WAGE CAMPAIGNS IN THE ECONOMIC POLICY ARENA: FOUR CASE STUDIES FROM CALIFORNIA 15 (1999), https://perma.cc/H9SW-XFF8 (noting the usage of "opt out" clauses "that allow[] specific terms of the ordinance to be superseded by a collective bargaining contract").

148. See LA Hotel Workers' Petition for Better Conditions, Higher Pay, Moves Forward, supra note 139.

149. Hotels Required to Provide Workers with Personal Security Devices Under New Law Signed by Garcetti, ABC7 NEWS (July 8, 2022), https://perma.cc/LP8S-8VKW. Though outside the scope of this Note, the measure was adopted through a California-specific process called the "indirect initiative," in which proposed initiatives may be adopted outright by a local or county government. See CAL. ELEC. CODE § 9214 (West 2024). Indirect initiatives can be particularly effective tools when union locals have strong preexisting local political support, as in this case study.

150. See supra Part II.A.

151. 2024 GREATER LOS ANGELES HOMELESS COUNT: CITY OF LOS ANGELES, L.A. HOMELESS SERVS. AUTH. (2024), https://perma.cc/F9UQ-AN62; 2024 GREATER LOS ANGELES HOMELESS COUNT: LOS ANGELES COUNTY, L.A. HOMELESS SERVS. AUTH. (2024), https://perma.cc/B2YC-AXJ5.

152. U.S. DEP'T OF HOUS. & URB. DEV., THE 2023 ANNUAL HOMELESSNESS ASSESSMENT REPORT (AHAR) TO CONGRESS: PART 1: POINT-IN-TIME ESTIMATES OF HOMELESSNESS 2, 20-21 (2023), https://perma.cc/E3AT-BBM4 (focusing only on the Nation's fifty largest cities and showing that Los Angeles City and County have the second-largest homeless population after New York, and that Los Angeles City and County-with 71,000 homeless people-dwarfs Seattle, which ranks third with 14,000 homeless people).

153. See MARGOT KUSHEL & TIANA MOORE, BENIOFF HOMELESSNESS & HOUS. INITIATIVE, TOWARD A NEW UNDERSTANDING: THE CALIFORNIA STATEWIDE STUDY OF PEOPLE EXPERIENCING HOMELESSNESS 37 (2023), https://perma.cc/2949-J7YC. For a discussion of the critical and worsening shortage of affordable housing for lower-income tenants, see NAT'L LOW INCOME HOUS. COAL., THE GAP: A SHORTAGE OF AFFORDABLE HOMES 1 (2023), https://perma.cc/NA49-FS22 ("Between 2019 and 2021, the shortage of affordable and available rental homes for extremely low-income renters worsened by more than 500,000 units, or 8%.").

154. See KUSHEL & MOORE, supra note 153, at 5.

155. The Responsible Hotels Ordinance, UNITE HERE! LOC. 11, https://perma.cc/Q66W-SJ2Y (archived Apr. 3, 2025). For another supporting perspective of this view, see Zoie Matthew, Amid L.A.'s Housing Crisis, Developers Are Demolishing Affordable Homes to Build Upscale Hotels, L.A. MAG. (Sept. 3, 2019), https://perma.cc/ZFU4-PY54. For reporting on the Union's involvement in affordable housing issues, see Sasha Abramsky, Unions Are Stepping Up to Find a Solution to California's Housing Crisis, NATION (May 26, 2023), https://perma.cc/N9UV-7CA5.

156. See Matthew, supra note 155.

157. The Responsible Hotels Ordinance, supra note 155.

158. Matthew, supra note 155.

159. See Abramsky, supra note 155 (noting that Local 11 has included affordable housing goals in some of its contract negotiations with employers, including a dedicated fund to subsidize construction).

160. See, e.g., Michael Goldstein, Will Los Angeles Vote to House Homeless People Alongside Paying Hotel Guests?, FORBES (Sept. 8, 2023, 5:30 AM EDT), https://perma.cc/674CJ8WQ ("Is the luxury hotel an endangered species? In Los Angeles, hotel operators say it may be if the 'Responsible Hotel Ordinance' passes on the March 2024 ballot."); Susan Shelley, The Bizarre Politics of Los Angeles' Hotel Ballot Measure, L.A. DAILY NEWS (updated July 19, 2023, 9:23 AM PDT), https://perma.cc/G8LS-N8D5 ("Imagine the dismay of the people who work in the hotels if they have to manage that situation. Business travelers, tourists and visitors will be side-by-side in the corridors, elevators, lobby and breakfast room with people who have been relocated from a nearby tent encampment to enjoy the same accommodations, paid for by city taxpayers.").

161. See Initiative Measure to Be Submitted Directly to the Voters: Land Use and Replacement Housing Requirements for Hotel Developers; Program to Utilize Vacant Hotel Rooms for Unhoused Individuals; Police Permit Requirements for Hotels. Initiative Ordinance., UNITE HERE! LOC. 11, https://perma.cc/G8VY-74SC (archived Apr. 3, 2025) [hereinafter Responsible Hotel Ordinance Initiative Measure] (proposing the initiative measure's text).

162. See id.

163. Id.

164. UNITE HERE Local 11 Hospitality Workers Call on ALIS Conference Hotel Executives to Help Solve Housing Crisis, UNITE HERE! LOC. 11 (Jan. 24, 2023), https://perma.cc/Q2K2- KHUQ ("UNITE HERE Local 11 members collected a record 126,000 signatures from LA residents to place the Responsible Hotel Ordinance on the March 2024 ballot."); Letter from Holly L. Wolcott, L.A. City Clerk, to Hon. Members of the L.A. City Council (July 20, 2022), https://perma.cc/R9A9-WP5U (certifying that the proposed ordinance was received by the City Clerk's office on June 10, 2022, with 126,006 signatures).

165. See, e.g., Goldstein, supra note 160 (describing the fears expressed by industry spokespeople about the ballot proposal); Noelle Mateer, Union Pulls Ballot Measure That Would Require LA Hotels to House Homeless People, HOTEL DIVE (Nov. 3, 2023), https://perma.cc/KY96-VYUF ("Los Angeles hoteliers and the American Hotel and Lodging Association [AHLA] decried the original ordinance . . . . [According to an AHLA spokesperson,] 'the dangerous "homeless-in-hotels" ballot measure would be a disaster for Los Angeles and its hotel workers' . . . . [and] was 'the worst idea we've ever heard.' ").

166. LA City Council President Krekorian Directs City Staffto DraftHistoric Ordinance to Prioritize Housing over Luxury Hotel Development, UNITE HERE! LOC. 11 (Nov. 1, 2023), https://perma.cc/QC6U-TLL3.

167. Matthew Hall, Los Angeles City Advances Restrictions on New Hotel Development, SANTA MONICA DAILY PRESS (Nov. 20, 2023), https://perma.cc/3UEH-GRNQ.

168. Id.

169. Id.

170. See id.; L.A., CAL., MUN. CODE PLAN. & ZONING ch. I, art. 2, § 12.24(U)(4) (2024).

171. See supra note 138 and accompanying text.

172. ARIZ. CONST. art. XXV.

173. Right to Work States Timeline, NAT'L RIGHT TO WORK COMM., https://perma.cc/WTB3- ZXFA (archived Apr. 7, 2025). Congress passed the Taft-Hartley Act in 1947, but Arizona was one of four states that had a right-to-work provision in its Constitution prior to the Taft-Hartley Act, having passed the provision by voter approval in 1946. Id.

174. Unions Members in Arizona-2023, U.S. BUREAU LAB. STAT., https://perma.cc/EH3UWZNN (last updated Feb. 29, 2024).

175. See Damon Stetson, Goldwater Is the Unions' No. 1 Election Target, N.Y. TIMES (Sept. 12, 1964), https://perma.cc/R43F-VVYV; John D. Pomfret, Labor and Goldwater; Reasons for Unions' Opposition to His Election Are Analyzed, N.Y. TIMES (Aug. 9, 1964), https://perma.cc/NN75-7FBV.

176. See, e.g., Parker Jackson, Goldwater Tells Federal Agency to Protect Workers' Rights from Union Power Grab, GOLDWATER INST. (Jan. 25, 2023), https://perma.cc/YCT8-87K4 ("The National Treasury Employees Union . . . is one of many big labor unions that wants to make it difficult for people to leave and stop paying dues, even though the U.S. Constitution and other federal laws protect workers and prohibit the unions' money grab. That's why the Goldwater Institute submitted comment to the Federal Labor Relations Authority last week opposing the National Treasury Employees Union's request for rules and policy changes that would allow it to prevent its members from leaving the union or stop paying union dues unless they formally opt out within a narrow, annual window of time.").

177. See Changes in Basic Minimum Wages in Non-Farm Employment Under State Law: Selected Years 1968 to 2024, U.S. DEP'T. LAB., https://perma.cc/6TEG-JXDN (last updated Jan. 2025).

178. JANICE K. BREWER, PUBLICITY PAMPHLET: BALLOT PROPOSITIONS & JUDICIAL PERFORMANCE REVIEW 105-06 (2006) [hereinafter 2006 OFFICIAL BALLOT PAMPHLET], https://perma.cc/F9KP-KLPT; Memorandum from Janice K. Brewer, Ariz. Sec'y of State (July 18, 2006), https://perma.cc/968C-XLD2 ("Arizona Minimum Wage Coalition, has filed a total of 15,311 petition signature sheets containing 196,778 signatures to initiative petition serial number I-13-2006, which are eligible for verification." (emphasis omitted)).

179. History of Federal Minimum Wage Rates Under the Fair Labor Standards Act, 1938-2009, U.S. DEP'T LAB., https://perma.cc/3MYD-CTHK (archived Apr. 7, 2025).

180. 2006 OFFICIAL BALLOT PAMPHLET, supra note 178, at 101.

181. See id.

182. Id. at 106.

183. See STATE OF ARIZONA OFFICIAL CANVASS, 2006 GENERAL ELECTION-NOVEMBER 7, 2006, ARIZ. OFF. SEC'Y STATE 15 (Dec. 4, 2006, 1:16 PM), https://perma.cc/TYE9-WZ2X.

184. See 2006 OFFICIAL BALLOT PAMPHLET, supra note 178, at 102-03 ("A county, city, or town may by ordinance regulate minimum wages and benefits within its geographic boundaries . . . . This article shall be liberally construed in favor of its purposes and shall not limit the authority of the legislature or any other body to adopt any law or policy that requires payment of higher or supplemental wages or benefits . . . .").

185. ARIZ. CONST. art. IX, § 20(1)-(2), (6), (9) (allowing local voters to approve "alternative expenditure limitation[s]" to the state-imposed expenditure limitations on municipalities, and thus permitting home-rule options over fiscal policy).

186. For a doctrinal overview of Dillon's rule and how it applies to cities, see GERALD E. FRUG, CITY MAKING: BUILDING COMMUNITIES WITHOUT BUILDING WALLS 27-50 (1999); and David J. Barron, The Promise of Cooley's City: Traces of Local Constitutionalism, 147 U. PA. L. REV. 487, 506-09 (1999).

187. Barron, supra note 186, at 508 ("Dillon's new rule precluded local governments from exercising any powers that had not been expressly granted by the legislature . . . .").

188. See 2006 OFFICIAL BALLOT PAMPHLET, supra note 178, at 102-03.

189. See, e.g., H.B. 2280, 51st Leg., 1st Reg. Sess. (Ariz. 2013) (codified as amended at ARIZ. REV. STAT. ANN. § 23-204 (2025)).

190. Id.

191. FlagstaffLiving Wage Coal. v. Arizona, No. CV2015-004240, at 2 (Ariz. Super. Ct. July 1, 2015).

192. H.B. 2579, 52d Leg., 2d Reg. Sess. (Ariz. 2016) (codified as amended at ARIZ. REV. STAT. ANN. §§ 23-204, 23-350, 23-357 (2024)).

193. United Food & Com. Workers Loc. 99 v. Arizona, No. CV 2016-092409, slip op. at 2-3 (Ariz. Super. Ct. Aug. 29, 2017); see also 2006 OFFICIAL BALLOT PAMPHLET, supra note 178, at 106 (showing that the President of UFCW Local 99 backed Proposition 202 before it passed).

194. See United Food & Com. Workers Loc. 99, slip op. at 12.

195. MICHELE REAGAN, ARIZONA 2016 GENERAL ELECTION PUBLICITY PAMPHLET 58, 73 (2016) [hereinafter 2016 OFFICIAL BALLOT PAMPHLET], https://perma.cc/DB48-LHCQ.

196. Id. at 60; Alia Beard Rau, Prop. 206: Arizona Voters Approve $12 Minimum Wage, AZCENTRAL, (updated Nov. 8, 2016, 9:37 PM MT), https://perma.cc/VJD7-K5DE (explaining that Proposition 206 raised the minimum wage and required employers to provide paid sick time); Arizona's Proposition 206: Much More than a Minimum Wage Increase, OGLETREE DEAKINS (Oct. 24, 2016), https://perma.cc/K6C4-Q87T ("The lesspublicized, but certainly more groundbreaking portion of Prop. 206, is its paid sick time (PST) component. Arizona law does not currently require any private or municipal employers to provide PST for their employees. Under Prop. 206, private employers and municipalities would be subject to [paid sick leave] requirements . . . .").

197. 2016 OFFICIAL BALLOT PAMPHLET, supra note 195, at 65.

198. Id. at 71, 73, 80; see also About Us, WORKER POWER INST., https://perma.cc/43F4-C27L (archived Apr. 3, 2025).

199. About Us, YES ON 206! HEALTHY WORKING FAMS., https://perma.cc/ZM6Q-9SX7 (archived Apr. 3, 2025).

200. 2016 General Election, OFF. ARIZ. SEC'Y STATE, https://perma.cc/4LZC-LNVS (last updated Nov. 21, 2016, 4:28:45 PM).

201. Our Ballot Measure Campaigns, FAIRNESS PROJECT, https://perma.cc/Z83S-BDG5 (archived Apr. 8, 2025).

202. 2016 OFFICIAL BALLOT PAMPHLET, supra note 195, at 84-85.

203. Id. ("One of the hypocrisies of this measure is that the minimum wage and paid sick time requirements would not apply to workers in unions. So if you're not a union member, tough luck! And if you're a union boss, what a deal!"); see also id. at 83 (featuring a statement from the Arizona Chamber of Commerce, which referred to "out-of-state labor union supporters" behind the proposition).

204. Ariz. Chamber of Com. & Indus. v. Kiley, 399 P.3d 80, 83 (Ariz. 2017). For a discussion of the single-subject rule, see notes 322-23 and accompanying text below.

205. Brief of Amicus Curiae Goldwater Institute in Support of Plaintiffs/Petitioners with Consent of Parties at 9-15, Kiley, 399 P.3d 80 (No. CV2016-0180902).

206. Kiley, 399 P.3d at 89.

207. Quantifying the collective impact of Propositions 202 and 206 lies outside of the scope of this Note. However, supporters argue that nearly 800,000 Arizonans would experience direct wage increases because of Proposition 206 alone as minimum wage earners. See DAVID R. BERMAN, UNDERSTANDING ARIZONA'S PROPOSITIONS: PROP 206, at 1 (2016), https://perma.cc/AY4Q-MV86.

208. See 2016 OFFICIAL BALLOT PAMPHLET, supra note 195, at 65-66.

209. See ARIZ. CONST. art. XXV; infra Part II.C.3.

210. GOULD & KIMBALL, supra note 12.

211. GERSTLE, supra note 11, at 28.

212. GOULD & KIMBALL, supra note 12.

213. Id.

214. Right to Work States, NAT'L RIGHT TO WORK LEGAL DEF. FOUND., https://perma.cc/7T8M-3QPJ (archived Apr. 3, 2025).

215. Right to Work States Timeline, supra note 173; ARIZ. CONST. art. XXV.

216. See Arizona House Concurrent Resolution 2008, LEGISCAN, https://perma.cc/5BCW-PECM (archived Apr. 3, 2025).

217. Id.

218. Endorsements & Coalition Partners, ARIZ. WORKS TOGETHER, https://perma.cc/4DRVLFL9 (archived Apr. 3, 2025) (including supporters like United Auto Workers, United Campus Workers of Arizona, BCTGM Local 232, IBEW Local 640, Starbucks Workers United, and Cannabis Workers Union).

219. State of Arizona: Application for Serial Number Initiative Petition; A.R.S. § 19-111, ARIZ. SEC'Y STATE, https://perma.cc/94B3-SE7M (archived Apr. 3, 2025).

220. Only one state, Michigan, has repealed right to work in the last six decades, doing so via legislative action. Joey Cappelletti, Michigan Becomes 1st State in Decades to Repeal 'Right to Work' Law, PBS NEWS (Mar. 24, 2023, 4:51 PM EDT), https://perma.cc/CFT3- FUD5. Direct democracy has stymied right-to-work efforts twice before at the state level. In Illinois, voters in 2022 approved a referendum establishing a strong constitutional right to collectively organize and bargain. ILL. CONST. art. I, § 25. In Missouri, voters used the veto referendum to reverse a right-to-work law passed by the state legislature in 2018. See 2018 Ballot Measures, MO. SEC'Y STATE, https://perma.cc/667G-AG8W (archived Apr. 3, 2025); Election Results, MO. SEC'Y STATE (Nov. 30, 2018), https://perma.cc/E2LS-WBAP.

221. See Arizona Repeal Right to Work Amendment (2024), BALLOTPEDIA, https://perma.cc/U6TE-WY9A (archived Apr. 3, 2025).

222. See id.; TJ L'Heureux, Coalition Launches Push to Repeal Arizona's "Right-to-Work" Law, PHX. NEW TIMES (Oct. 31, 2023), https://perma.cc/S39Q-D49F.

223. See James Feigenbaum, Alexander Hertel-Fernandez & Vanessa Williamson, From the Bargaining Table to the Ballot Box: Political Effects of Right to Work Laws 3 (Nat'l Bureau of Econ. Rsch., Working Paper No. 24259, 2018) (summarizing the literature on right-towork laws and finding that "[o]ne through-line seems to be that RTW laws weaken unions, either by reducing union organizing or density or labor's leverage more generally . . . . [and] [u]nion revenues are also hit" (citations omitted)).

224. Albert Serna, Jr., State Ballot Measures Attract More than $417 Million Ahead of Election Day, OPENSECRETS (Oct 30, 2024, 2:32 PM), https://perma.cc/S7PZ-TUXC (showing that over $17 million was raised in support of the proposition).

225. For instance, two of the largest unions in the state, the Arizona AFL-CIO and the United Food and Commercial Workers, did not endorse and back the campaign. See Endorsements & Coalition Partners, supra note 218.

226. See supra Part II.

227. See supra Part II.

228. See supra Part II.

229. See supra Part II.

230. See supra Part II.

231. See supra Part II.

232. In neutrality agreements, employers agree not to interfere in union organizing campaigns. Relatedly, in card-check agreements, employers agree to recognize a new union once the campaign has achieved a majority of worker support. Combined cardcheck- neutrality agreements are powerful tools to quickly organize workplaces. Indeed, some unions organize the vast majority of their workers through card-checkneutrality agreements, without any Board involvement. See NOLAN ET AL., supra note 5, at 468.

233. See supra Part II.B.

234. See supra Part II.B.

235. For an exploration of these dynamics, see generally Briffault, note 24 above (addressing the growing trend of aggressive state preemption policies, oftentimes targeted against more progressive blue localities by red legislatures); and Richard C. Schragger, The Attack on American Cities, 96 TEX. L. REV. 1163 (2018) (recounting aggressive state preemption against cities pursuing novel policies).

236. See supra Part II.C.

237. Telephone Interview with Rebekah Friend, Former Executive Director, Arizona AFLCIO (Feb. 14, 2024).

238. Id.

239. ELLIOT BULMER, DIRECT DEMOCRACY: INTERNATIONAL IDEA CONSTITUTION-BUILDING PRIMER 3, at 10 (2d ed. 2017).

240. See Bruce E. Cain & Kenneth P. Miller, The Populist Legacy: Initiatives and the Undermining of Representative Government, in DANGEROUS DEMOCRACY? THE BATTLE OVER BALLOT INITIATIVES IN AMERICA 33, 45-48 (Larry J. Sabato, Howard R. Ernst & Bruce A. Larson eds., 2001); see also Why Are Ballot Measures So Confusing?, N.Y. TIMES (Nov. 3, 2016), https://perma.cc/3FSR-8N5L (collecting several op-eds by different authors explaining several reasons why ballot measures may confuse voters, such as amateur writing from citizen-proposers and dark-money influences, and proposing several solutions, including plain-language requirements).

241. BOLDT, supra note 19, at 7-10.

242. See, e.g., Ben Christopher, Critics Demand Fairer Prop Ballot Labels and Summaries, but Lawsuits Tend to Flame Out, CALMATTERS (updated Aug. 7, 2020), https://perma.cc/6K5B-YFUA (illustrating how the efforts to convey accurate information via ballot pamphlet summaries can be confusing and result in litigation).

243. See BULMER, supra note 239, at 10.

244. Mark Baldassare, Californians Want Referendum Reforms, PUB. POL'Y INST. CAL. (June 7, 2023), https://perma.cc/Q2JH-WGHV.

245. Nat'l Pride at Work, Inc. v. Governor of Mich., 748 N.W.2d 524, 529 (Mich. 2008); MICH. CONST. art. I, § 25.

246. 748 N.W.2d at 529.

247. Id. at 541 n.22.

248. Id. at 547 (Kelly, J., dissenting) (emphasis omitted) (quoting CITIZENS FOR PROT. MARRIAGE, PROTECT MARRIAGE 30a (2004)).

249. Id. (quoting Charisse Jones, Gay Marriage on Ballot in 11 States, USA TODAY, Oct. 15, 2004, at A.3).

250. Id. at 546.

251. Id. at 547-48.

252. Id. at 529 (majority opinion).

253. Id. at 542.

254. For a discussion of the issues of judicial review of popular intent in direct democracy when information is hard to access or the electorate receives information from mass media sources, see Schacter, supra note 17, at 130-38, 147.

255. Richard B. Freeman, What Do Unions Do . . . to Voting? 34 (Nat'l Bureau of Econ. Rsch., Working Paper No. 9992, 2003) ("[A]s long as unions maintain millions of members and allocate resources to politics, they can maintain or increase their impact on elections even with declining density at work places. They can do this by contacting and influencing the millions of non-union persons favorably inclined to unions but whom unions cannot organize under current US labor laws and procedures."); see also TOVA WANG, UNION IMPACT ON VOTER PARTICIPATION-AND HOW TO EXPAND IT 1 (2020) ("Union members develop a sense of agency and efficacy, and gain skills and knowledge that research shows has a spillover effect beyond the individual members, especially to the members' households. It has been shown that areas with higher labor density have higher voter turnout rates. In other words, the effects may even spread to a wider geographic region."); Maximillian Alvarez, In Key Swing States, Union Members Are Democrats' Last Line of Defense, IN THESE TIMES (Nov. 8, 2022), https://perma.cc/ECZ5- JMZU (asserting that unions played a critical role in turning out the vote in Pennsylvania, Nevada, Arizona, and Georgia); C.M. Lewis, Unions Are the Heart of Arizona's Political Change, NATION (Aug. 6, 2021), https://perma.cc/Z2LB-F8GG (noting the transformational impact of unions in elections over several elections in Arizona).

256. See JANE F. MCALEVEY, NO SHORTCUTS: ORGANIZING FOR POWER IN THE NEW GILDED AGE 59-70 (2016) (arguing that in cases of successful organizing, unions make use of "whole worker organizing," reaching deep into workers' lives not just through their workplaces but also through their churches, neighborhoods, civic organizations, social media, children's schools, and more).

257. Telephone Interview with Rebekah Friend, supra note 237.

258. Elizabeth Garrett, Money, Agenda Setting, and Direct Democracy, 77 TEX. L. REV. 1845, 1849 (1999).

259. See BULMER, supra note 239, at 11.

260. Faiz Siddiqui & Nitasha Tiku, Uber and LyftUsed Sneaky Tactics to Avoid Making Drivers Employees in California, Voters Say. Now, They're Going National, WASH. POST (Nov. 17, 2020, 7:00 AM EST), https://perma.cc/XS6J-Y626; Grace Manthey, Prop. 22: Rideshare- Driver Measure Is Most Expensive in California History, ABC7 NEWS (Nov. 3, 2022), https://perma.cc/5K99-BY4U.

261. Baldassare, supra note 244.

262. J. FRED SILVA, THE CALIFORNIA INITIATIVE PROCESS: BACKGROUND AND PERSPECTIVE, at i (2000) ("The initiative process itself has undergone major changes over the last 25 years. Today, one could describe it as the 'initiative industrial complex,' given the number of companies providing services such as signature gathering, legal services, and campaign consulting that are now integral and apparently essential to the process. The days of romanticizing it as the 'citizens' initiative process are long over.").

263. See, e.g., CAL. ELEC. CODE § 9035 (West 2024) (requiring signatures from 5% of the total registered voters in the State for proposed statutes, and 8% for constitutional amendments).

264. See Elizabeth Garrett, Who Directs Direct Democracy?, 4 U. CHI. L. SCH. ROUNDTABLE 17, 20 (1997) (finding that "[a]ccording to some historians, in the early days of direct democracy, signature drives often were occasions for a great deal of public deliberation," but that the professionalization of signature drives in the present day undermines the organic demonstration of popular support associated with signature gathering).

265. Bob Egelko, Lawmaker, Advocates Are Trying Again to Rein in Paid Political Signature- Gatherers, S.F. CHRON. (Mar. 21, 2023, 4:41 PM), https://perma.cc/5ZQT-9BH2 (documenting several instances in which paid signature-gatherers lied about the purposes of petitions, such as one petition where gatherers told voters that the petition reduced drilling when, in reality, it removed limits on gas and oil wells); Will Fritz, Signature-Gathering Lawsuits Bring Some of the Industry's Biggest Tensions to Light, VOICE OF SAN DIEGO (Apr. 8, 2019), https://perma.cc/GB8N-SV2Y (discussing several lawsuits related to fraud and wage theftin the signature-gathering industry).

266. See Anna Tong, Chesa Boudin Recall Campaign Featured Exceptionally High Costs for Signature-Gathering, S.F. STANDARD (Apr. 19, 2022, 9:04 AM), https://perma.cc/9WKVG654 ("It's possible the signature-gathering company knew [some recall campaigns] had a ton of money, and they jacked up the price."); see also Garrett, supra note 258, at 1852- 53 (finding that "one direct mail firm in California promises qualification 'to 110%' " within forty-five days, with "a money-back guarantee" for its services).

267. See supra Part II.

268. Telephone Interview with Rebekah Friend, supra note 237.

269. Id.

270. See, e.g., Richard A. Epstein, Labor Unions: Saviors or Scourges?, 41 CAP. U. L. REV. 1, 8, 33 (2013) ("[N]othing in the history of labor law justifies the extraordinary set of legal privileges that they have received over the past 100 years . . . . The truly amazing feature of American labor history is the fact that politicians and judges were prepared to turn cartwheels to give unions special powers that only increased the total level of social dislocations . . . . The union movement, defended with such passion for its lofty ideals, is better understood as a scourge, not a savior, not by some narrow theory that exalts management over labor, but by any systematic evaluation of the overall negative impact that unions have had on the social welfare of a great nation."). For one counter to this view, see Eric Levitz, Unions Are Not a Special Interest Group, N.Y. MAG. (May 25, 2018), https://perma.cc/7373-5D69. For a constitutional view that would have allowed differential treatment of corporations relative to other actors like unions under a Massachusetts campaign finance law, consider First National Bank of Boston v. Bellotti, 435 U.S. 765, 809 (1978) (White, J., dissenting).

271. See supra Part I.

272. For comparison, the law at issue in Gould was merely supplemental enforcement to strengthen the national labor law enforcement whereby Wisconsin would not contract with repeat offenders of the NLRA, but the Supreme Court read that law as interfering too much with the careful statutory balance contemplated by Congress. Wis. Dep't of Indus., Lab. & Hum. Rels. v. Gould Inc., 475 U.S. 282, 287-89 (1986); see Siddiqui & Tiku, supra note 260.

273. See supra Part II.

274. Telephone Interview with Rebekah Friend, supra note 237.

275. See Indep. Energy Producers Ass'n v. McPherson, 136 P.3d 178, 190-91 (Cal. 2006) (recounting the origins of direct democracy in California as a popular means to counter the influence of poweful railroad corporations).

276. Hiram Johnson, Governor of Cal., First Inaugural Address to the Senate and Assembly of the State of California (Jan. 3, 1911) (transcript available in the California Governor's Gallery).

277. See id.

278. See supra Parts IV.B.1-.2.

279. E.E. SCHATTSCHNEIDER, THE SEMISOVEREIGN PEOPLE: A REALIST'S VIEW OF DEMOCRACY IN AMERICA 62-63 (1960) (emphasis omitted).

280. See generally LISA JANE DISCH, MAKING CONSTITUENCIES: REPRESENTATION AS MOBILIZATION IN MASS DEMOCRACY 18-33 (2021) (explaining how voter mobilization and policymaking shiftcleavage lines within Schattschneider's theory).

281. Id. at 19.

282. See Robin Urevich & Gabriel Sandoval, Checked Out: How LA Failed to Stop Landlords from Turning Low-Cost Housing into Tourist Hotels, PROPUBLICA (July 10, 2023, 9:00 AM EDT), https://perma.cc/U2T9-WURK.

283. See supra Part II.B.

284. See supra Part II.B.

285. See DISCH, supra note 280, at 24 (describing that policymaking "constitutes social relations, produces social 'group identification and consciousness,' and brings social groups into being by calling them 'into or out of the political process' " (quoting Suzanne Mettler & Joe Soss, The Consequences of Public Policy for Democratic Citizenship: Bridging Policy Studies and Mass Politics, 2 PERSPS. ON POL. 55, 57, 59 (2004))).

286. See id. To clarify, "representatives" here does not refer just to elected officials but also to unofficial representatives of political movements, such as unions and community organizations, that credibly lead to the creation of policy.

287. See id.

288. See supra Part II.C.

289. Briffault, supra note 24, at 2007.

290. See, e.g., id. at 1999-2003 (discussing the highly punitive and aggressive nature of new forms of state preemption); see also Elmore, supra note 18, at 276-86 (explaining the threat of state preemption against localities experimenting with pro-worker legislation and policy).

291. Briffault, supra note 24, at 1997-98.

292. Lydia Bean & Maresa Strano, Punching Down: How States Are Suppressing Local Democracy, NEW AM., https://perma.cc/6QNH-W577 (last updated July 11, 2019, 4:57 PM EDT).

293. S.B. 1487, 52d Leg., 2d Reg. Sess. (Ariz. 2016) (codified as amended in scattered sections of ARIZ. REV. STAT. §§ 41-43).

294. See supra Part II.C.

295. Telephone Interview with Rebekah Friend, supra note 237.

296. Briffault, supra note 24, at 1997-99; 2004-06.

297. See Lydia Saad, More in U.S. See Unions Strengthening and Want It That Way, GALLUP (Aug. 30, 2023), https://perma.cc/Z35K-5M5T (finding that unions currently receive a historical level of support in the United States, with 71% of Americans in 2022 approving of labor unions-the highest registered approval since 1965); Working People Want a Voice at Work, ECON. POL'Y INST. (Apr. 21, 2021), https://perma.cc/G2Z8-2JEH (finding that nearly half of workers say that if given the option, they would join a union).

298. DISCH, supra note 280, at 23.

299. A most recent example is the electoral victory of Democrat Ruben Gallego for Senate in Arizona, replacing the centrist Kyrsten Sinema, who voted against raising the Nation's minimum wage. Gallego ran as an unapologetically pro-union and proworkers' rights candidate and made economic justice a central plank of his platform. See Catie Edmondson, Kyrsten Sinema and the Thumbs-Down that Enraged the Left, N.Y. TIMES (updated Oct. 18, 2021), https://perma.cc/PCY9-SKLN; Charlotte Alter, 'Nobody Trusts Her.' Ruben Gallego on Challenging Kyrsten Sinema for Her Senate Seat, TIME (Jan. 25, 2023, 5:50 PM EST), https://perma.cc/MW3C-E44K (quoting Gallego that his breaking point with Sinema was after she voted to the right on a number of economic issues, including against a minimum-wage increase, negotiating drug prices with pharmaceutical companies, and programs opposed by private equity and hedge-fund interests); Sahil Kapur & Frank Thorp V, How Ruben Gallego Outperformed Harris, Dominated with Latinos and Won Arizona, NBC NEWS (Nov. 15, 2024, 3:06 PM PST), https://perma.cc/C85D-GGBY (describing how Gallego outperformed Harris by eight points and ran on economic issues relevant to working-class voters in Arizona).

300. See SCHATTSCHNEIDER, supra note 279, at 63 ("Since this is the process by which majorities and minorities are made, it may be said that every change in the direction and location of the line of clevage [sic] produces a new majority and a new allocation of power.").

301. See supra Part I.A.

302. See supra Part I.A.

303. Fort Halifax Packing Co. v. Coyne, 482 U.S. 1, 20-21 (1987).

304. Id. at 23.

305. Id. at 20-21.

306. Id. at 21.

307. See Lodge 76, Int'l Ass'n of Machinists & Aerospace Workers v. Wis. Emp. Rels. Comm'n, 427 U.S. 132, 136 (1976) (describing that federal labor law does not "preclude the States from regulating aspects of labor relations that involve 'conduct touch[ing] interests so deeply rooted in local feeling and responsibility that . . . we could not infer that Congress had deprived the States of the power to act' " (quoting San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236, 244 (1959))); see also Garmon, 359 U.S. at 243 (describing that federal labor law does not preclude laws "where the activity regulated was a merely peripheral concern of the" NLRA).

308. Coyne, 482 U.S. at 20-21; see supra Part II.

309. 471 U.S. 724, 755 (1985).

310. See supra notes 139-48 and accompanying text; L.A., CAL., MUN. CODE ch. XVIII, art. 2, §§ 182.00-182.01(I), (L) (2022); Responsible Hotel Ordinance Initiative Measure, supra note 161; 2006 OFFICIAL BALLOT PAMPHLET, supra note 178, at 101-03; 2016 OFFICIAL BALLOT PAMPHLET, supra note 195, at 58, 61.

311. See, e.g., BF Foods, LLC v. SeaTac, No. 13-2-25352-6, 2013 WL 6851515, at ·14-15 (Wash. Super. Ct. Dec. 27, 2013) (finding that a local initiative that established minimum employment standards for hospitality and transportation workers in the city, including a $15 per hour minimum wage, was not preempted by the NLRA, except for an antiretaliation provision), aff'd in part, rev'd in part sub nom. Filo Foods, LLC v. SeaTac, 357 P.3d 1040 (Wash. 2015); R.I. Hosp. Ass'n v. Providence, 667 F.3d 17, 23, 32 (1st Cir. 2011) (finding that a Providence ordinance regulating the hospitality business and changes of ownership was not preempted by the NLRA under Coyne, 482 U.S. at 20-21); Sachen v. Ill. State Bd. of Elections, 215 N.E.3d 977, 978 (Ill. App. Ct. 2022) (avoiding the preemption question by finding the petitioner's action invalid on other grounds, but affirming the trial court's order, which did address preemption); Am. Hotel & Lodging Ass'n v. Los Angeles, 834 F.3d 958, 964-65 (9th Cir. 2016) (finding that a wage ordinance was not preempted because "[b]y providing a basic minimum wage and time-offcompensation, the Wage Ordinance alters the backdrop of negotiations, not the mechanics of collective bargaining"); Cal. Grocers Ass'n v. Los Angeles, 254 P.3d 1019, 1030, 1037 (Cal. 2011) (finding that a grocery worker retention measure was not preempted because "we discern no evidence that Congress [in passing the NLRA] affirmatively intended to leave the subject of employee retention unregulated by states and municipalities"). But for examples of the opposite outcome, see Chamber of Commerce of the United States v. Bragdon, 64 F.3d 497, 498-99, 501 (9th Cir. 1995) (finding that a "prevailing wage" ordinance in Costa Contra County was preempted, because the prevailing wage formula depended substantially on "established collective-bargaining agreements," which amounted to an imposition of "substantive requirements [that do] affect the bargaining process"); and 520 S. Mich. Ave. Assocs. v. Shannon, 549 F.3d 1119, 1133-34 (7th Cir. 2008) (finding that a statute passed by the Illinois Legislature, the "One Day Rest in Seven Act," establishing labor standards in the hospitality industry, was "not a true minimum labor standard" and thus was preempted by the NLRA).

312. See Livadas v. Bradshaw, 512 U.S. 107, 132 n.26 (1994).

313. Id.

314. See Am. Hotel & Lodging Ass'n, 834 F.3d at 965 (relying on Livadas to uphold a supersession clause); Columbia Sussex Mgmt., LLC v. Santa Monica, 482 F. Supp. 3d 1002, 1013 (C.D. Cal. 2020) (upholding a supersession clause).

315. See supra Part I.A.

316. See supra Part II.

317. See supra Part II.

318. See supra Part II.C (showing how the union coalition in Arizona beat back state preemption and hostile legislative action).

319. See supra Part II.C.

320. See supra Part III.B.1.

321. BOLDT, supra note 19, at 4-5.

322. See, e.g., Maready v. Winston-Salem, 467 S.E.2d 615, 618, 625 (N.C. 1996) (finding no public-purpose violation in an economic tax incentive granted exclusively to private corporations); Barclay v. Melton, 5 S.W.3d 457, 459-60 (Ark. 1999) (finding an originalpurpose violation in a bill that changed from a tax credit to a tax surcharge assessment); Hammerschmidt v. Boone County, 877 S.W.2d 98, 103 (Mo. 1994) (en banc) (finding a single-subject violation because there was no "single subject core" shared between the statute's provisions on election procedures and county governance); Nat'l Solid Waste Mgmt. Assoc. v. Dir. of the Dep't of Nat. Res., 964 S.W.2d 818, 821-22 (Mo. 1998) (en banc) (finding a clear-title violation because the title, "relating to solid waste management," was underinclusive and failed to give any indication of how the bill addressed hazardous waste); Burns v. Cline, 382 P.3d 1048, 1051 (Okla. 2016) (finding a single-subject violation because a bill included one provision relating to abortion and three unrelated provisions relating to statutory rape, licensing and inspection for abortion facilities, and felony penalties for those who violate abortion statutes).

323. JEFFREY S. SUTTON, STEPHEN R. MCALLISTER, RANDY J. HOLLAND & JEFFREY M. SHAMAN, STATE CONSTITUTIONAL LAW: THE MODERN EXPERIENCE 767-68 (4th ed. 2023); Martha J. Dragich, State Constitutional Restrictions on Legislative Procedure: Rethinking the Analysis of Original Purpose, Single Subject, and Clear Title Challenges, 38 HARV. J. ON LEGIS. 103, 103- 04 (2001) (describing the clear-title, original-purpose, and single-subject rules).

324. Compare the single-subject and clear-title rule in Oklahoma's Constitution, see OKLA. CONST. art. V, § 57 ("Every act of the Legislature shall embrace but one subject, which shall be clearly expressed in its title . . . ."), with Wisconsin's common law publicpurpose doctrine, see State ex rel. Warren v. Nusbaum, 208 N.W.2d 780, 795 (Wis. 1973) (describing that the public-purpose doctrine requires that "[p]ublic funds may be expended for only public purposes," and that "[a]n expenditure of public funds for other than a public purpose would be abhorrent to the Constitution of Wisconsin").

325. Ariz. Chamber of Com. & Indus. v. Kiley, 339 P.3d 80, 88-89 (Ariz. 2017).

326. Hoffman v. Reagan, 429 P.3d 70, 73 (Ariz. 2018).

327. Dragich, supra note 323, at 122-24 (using Missouri Supreme Court jurisprudence to provide an example of how "rhetorical imprecision" and "analytical weakness" have muddied these concepts' doctrinal bases).

328. See id. at 106.

329. See Kiley, 339 P.3d at 83, 88-89 (rejecting litigants' efforts to overturn the initiative through the revenue-resource rule, the separate-amendment rule, and the singlesubject rule).

330. Id. at 83; ARIZ. CONST. art. 21, § 1; id. art. 4, § 13.

331. Kiley, 339 P.3d at 88-89.

332. See supra Part I.A.

333. See supra Part I.A (explaining that Garmon forbids any subnational law that even "arguably" touches upon the NLRA).

334. See supra Part II.

335. See, e.g., SILVA, supra note 262 ("The initiative process itself has undergone major changes over the last 25 years. Today, one could describe it as the 'initiative industrial complex,' given the number of companies providing services such as signature gathering, legal services, and campaign consulting that are now integral and apparently essential to the process. The days of romanticizing it as the 'citizens' initiative process are long over.").

336. DAVID R. BERMAN, INITIATIVE REFORM IN ARIZONA: EXPLORING SOME IDEAS 3 (2013), https://perma.cc/35RR-A4H3.

337. The Health 202: The Dialysis Industry Spent More than $100 Million to Beat a California Ballot Measure, WASH. POST (Nov. 12, 2018, 7:57 AM EST), https://perma.cc/43WPR77Z.

338. Id.

339. Siddiqui & Tiku, supra note 260.

340. Id.; David Dayen, The Dialysis Industry Is Spending $111 Million to Argue that Regulating It Would Put It Out of Business, INTERCEPT (Oct. 31, 2018, 8:00 AM), https://perma.cc/VG6P-79BR (finding that SEIU spent around $20 million on Proposition 8 against the dialysis industry's $111 million).

341. Telephone Interview with Rebekah Friend, supra note 237.

342. Cf. Derrick A. Bell, Jr., The Referendum: Democracy's Barrier to Racial Equality, 54 WASH. L. REV. 1, 13-22 (1978) (discussing broadly how direct democracy has often delivered reactionary, even racist policy outcomes).

343. Id.

344. Id. at 14-15.

345. See Morgan v. Hennigan, 379 F. Supp. 410, 482-83 (D. Mass. 1974) (ordering the desegregation of public schools in Boston through busing), aff'd sub nom. Morgan v. Kerrigan, 509 F.2d 580 (1st Cir. 1974); John Kifner, Violence Mars Busing in Boston, N.Y. TIMES (Sept. 13, 1974), https://perma.cc/9XVJ-8DGX. Bell was involved in Boston's efforts to desegregate its schools under Judge Garrity's orders. Derrick A. Bell, Jr., Serving Two Masters: Integration Ideals and Client Interests in School Desegregation Litigation, 85 YALE L.J. 470, 482-83 (1976). The violent resistance that resulted along with the mixed educational outcomes for Black students inspired his seminal article on the tension for lawyers who "serve two masters": their ideals of racial equality in schooling and the needs of their clients. See id. at 470-72 (introducing the paper's ideas); id. at 482-83 (discussing the Boston desegregation litigation); id. at 516 (concluding that "single-minded commitment to racial balance" has often failed to deliver quality educational outcomes for students).

346. Amanda Matas & James L. Rodríguez, The Education of English Learners in California Following the Passage of Proposition 227: A Case Study of an Urban School District, PERSPS. ON URB. EDUC., Summer 2014, at 44-47 (describing that CAL. EDUC. CODE § 300 (WEST 1998) codified a proposition banning bilingual education).

347. See Dyanna Castañeda Policarpio, Note, Behind and Beyond Bars: Children of Incarcerated Parents, 58 U.C. DAVIS L. REV. ONLINE 27, 50-51, 51 n.140 (2024).

348. See Bell, supra note 342, at 1 ("[T]he experience of blacks with the referendum has proved ironically that the more direct democracy becomes, the more threatening it is.").

349. See supra Part III.A.1.

350. See Bell, supra note 342, at 10-13.

351. DISCH, supra note 280, at 3-4.

352. See supra Part III.A.1.

353. See DISCH, supra note 280, at 19, 22-23 (discussing how mobilization creates new political cleavages and redefines the grounds for conflict).

354. Id. at 23-32.

355. See Mitchell, supra note 18, at 197 (writing, in 2008, that "although direct democracy has been available for political interest groups for almost a century in California, unions have exhibited only limited use of it-and that often on the defensive side," opposing ballot initiatives rather than proposing their own); see also id. ("The question raised by the 2007 action of UNITE HERE is whether the four referenda, whatever their specific fate, signify a new interest by labor unions in proposing ballot measures.").

356. Id.

357. Id. at 201 (noting that since Proposition 13, "direct democracy has tended to have an image of a device for the facilitation of right-wing populism").

358. Id. at 197, 207 (predicting presciently that while "California unions have tended . . . to focus on traditional candidate endorsements and legislative lobbying," and that "[t]he decision by UNITE HERE to file referendums in 2007 in the case of Indian gaming compacts . . . could signal a shifttoward more active use of direct democracy by organized labor in California").

359. See supra Part I.A.

360. See supra Part II.

361. See supra Part II.

362. See supra Part II.

363. See BOLDT, supra note 19, at 5 (showing that twenty-three states and nearly half of the country live in states with some form of the initiative).

364. See generally Berry, supra note 94, at 581 (outlining a right-wing agenda for a second Trump administration largely unfavorable to workers and unions).

365. William J. Brennan, Jr., State Constitutions and the Protection of Individual Rights, 90 HARV. L. REV. 489, 491, 503 (1977).

366. See Orin S. Kerr, A Theory of Law, 16 GREEN BAG 2D 111, 111 (2012).