The Ossification of American Labor Law and the Decline of Self-governance in the Workplace

Abstract

This article argues that the ineffectuality of American labor law and the shrinking scope of collective representation and collective bargaining are partly traceable to the law’s “ossification”—to its having been essentially sealed off for several decades from democratic revision and renewal and from local experimentation and innovation. The elements of this process of ossification, once assembled, make up an imposing set of barriers to innovation. The basic law has been cut off from legislative revision at the national level by Congress; from “market”-driven competition by employers; from the entrepreneurial and creative energies of private litigation; from variation at the state or local level by representative or judicial bodies; from changing constitutional doctrine; and from emerging transnational legal norms. Moreover, the National Labor Relations Board—the designated institutional vehicle for adjusting the labor laws to modern conditions—is increasingly hemmed in by the age of the text and the large body of judicial interpretations that has grown up over the years. The resulting statutory scheme is drastically out of date and out of sync with the needs of 21st century workers and labor markets.

Introduction

The law of the workplace in the United States is vast and dynamic. The proliferation of employment discrimination laws, individual employee rights, and regulations of terms and conditions of work has spawned a new field of legal study and an expanding area of legal practice and accounts for a growing share of the courts’ dockets. At the same time, private sector labor law—the law that governs workers’ efforts to advance their own shared interests through self-organization and collective protest, pressure, negotiation, and agreement with employers—has shrunk in its reach and its significance and is clearly ailing. Evidence of morbidity abounds, but it is enough here to cite the drastic diminution of organized labor, and hence the ambit of collective bargaining, to about eight percent of the private sector workforce. The labor laws have failed to deliver an effective mechanism of workplace representation and have become nearly irrelevant to the vast majority of private sector American workers.

Yet, if we strip the existing model down to its essentials—the right of employees to form associations and seek shared workplace objectives, free from employer interference and retaliation, and the right to bargain collectively with their employers about terms and conditions of employment—it is hard to see how we could do without it. The right of employees to associate, communicate about shared concerns, form organizations for mutual aid, and peacefully seek shared objectives has achieved the status of an international human right, and it has not been seriously questioned in the United States—in principle—for generations. Moreover, in this post-command-and-control era, collective bargaining would seem to represent a highly adaptable third way between the harsh regimen of individual freedom of contract and the rigidities of central regulation.

So why is the system of collective bargaining under American labor law, the essentials of which are so essential, so moribund? And why have labor unions, the labor law’s designated vehicles of employee representation in collective bargaining, shrunken so dramatically? Scholars have advanced several answers, not mutually exclusive but competing for emphasis: increasingly brazen employer resistance and the law’s inadequate response to that resistance; structural economic change, including deindustrialization and increasingly global and competitive product markets; a mismatch between the interests of both employees and employers and traditional adversarial unionism; and unions’ own complacency and lack of commitment to organizing for several crucial decades. Herein I advance a complementary argument about the labor laws themselves: The ineffectuality of American labor law, and the shrinking scope of collective representation and collective bargaining, is partly traceable to the law’s “ossification.” The core of American labor law has been essentially sealed off—to a remarkably complete extent and for a remarkably long time—both from democratic revision and renewal and from local experimentation and innovation. The basic statutory language, and many of the intermediate-level principles and procedures through which the essentials of self-organization and collective bargaining are put into practice, have been nearly frozen, or ossified, for over fifty years.¹

The elements of this ossification process are mostly familiar; yet once assembled they make up an imposing set of barriers to innovation. Most importantly, a longstanding political impasse at the national level has blocked any significant congressional revision of the basic text of the National Labor Relations Act, enacted in 1935, since 1959. Moreover, the basic text itself contains additional built-in obstacles to change. The original and unrevised federal ban on “company unions” has inhibited the development of competing, employer-sponsored forms of employee representation. The absence of a private right of action to enforce basic employee rights excludes the energies of private plaintiffs and their attorneys, and the creative role of courts, that have together so dramatically changed the landscape of “employment law.” And the National Labor Relations Board—the designated institutional vehicle for adjusting the labor laws to modern conditions—is itself increasingly constrained by the age of the text, the encrustation of binding precedent, and the growing dissonance between the collectivist premises of the New Deal labor law regime and the surrounding legal landscape.

Other possible avenues of revision and variation from outside the congressional scheme have also been closed or never opened. Most importantly, the broad implied federal preemption of state and local laws affecting collective labor relations blocks democratically inspired reforms or variations at the state and local level, as well as state common law innovation. The stubborn imperviousness of labor law to constitutional scrutiny has insulated labor law from an evolving body of legal norms that have extended robust protection to nonlabor protest and remade the law of public employment. Resistance to transnational legal authority, endemic to American law, has insulated labor law from the potential influence of international human rights or other labor rights such as those that have forced the reexamination of national labor laws in Europe.

One is hard pressed to identify any other major American legal regime—any other body of federal law that governs a whole domain of social life—that has been so insulated for so long from significant change as labor law. The result of this many-faceted phenomenon of ossification is a labor law regime that has fallen badly out of sync with dramatic changes in the labor force, the organization of work, and global product markets. Nor do the many facets of ossification seem likely to give way anytime soon. So, for the foreseeable future, proposals to improve the governance of the workplace are unlikely to find a toehold within the traditional domain of “labor law”; and even proposals for reform from without will need to steer clear of the large and durable roadblock that labor law creates.

The Congressional Impasse over Labor Law Reform

One of the most striking features of American labor law is the age of its basic governing text. The current National Labor Relations Act consists almost entirely of the original Wagner Act of 1935, together with the major Taft–Hartley amendments of 1947. The Landrum–Griffin Act of 1959 only tinkered with the labor-management provisions of the NLRA, mainly by extending the prohibition of secondary and recognitional picketing. The text of the NLRA has remained virtually untouched since 1959. To be sure, much of that basic text is open-textured, and susceptible to a range of interpretations. Its declaration of employee rights is constitution-like in its phrasing, and its prohibitory provisions are studded with terms like “discriminate,” “interfere,” and “coerce” that had to be given meaning over time. But many crucial interpretive decisions were rendered by the Supreme Court in the early years of the Act’s existence and have been in place—amendable only by Congress—since at least the 1950s.

Needless to say, much has happened since then. Women have flooded into the workforce and racial and ethnic diversity has burgeoned. Manufacturing has shrunk relative to the service and information sectors, and the technology of transportation and communication has increasingly eroded geographic constraints on product markets and the location of production. Mass production and stable workplace hierarchies have given way to more flexible, customer-centered production methods and semiautonomous team-based organizations. And laws regulating substantive terms of employment and protecting individual employee rights have proliferated, as the collectivist premises of the NLRA have acquired the patina of a historic relic. So how is it that no significant reforms have made their way into the labor law statute since 1959?

It is not for lack of criticism. While the basic statutory rights of workers under Section 7 of the Act to associate, to discuss their grievances, to form a union, and to bargain collectively over terms and conditions of employment are sound and apparently politically untouchable (in principle), many intermediate level doctrines and procedures—some embedded in the text, some established by judicial precedent—have attracted much criticism and might have warranted another look in the last half-century. Unions and their allies have been especially critical of the rules governing union organizing activity and employers’ typically tenacious opposition to that activity (Becker 1993). The law protects not only employers’ right to express their opposition to unionization but also the right to compel employees to listen to them in “captive audience” meetings and to exclude union representatives from the workplace altogether. Moreover, by granting employers the right to an NLRB election, as opposed to more informal methods of verifying majority support for unionization, the law affords employers time to campaign against union representation and opens the door to the use of coercive and dilatory tactics.

So, too, the law fails to effectively prevent or punish coercive and illegal forms of employer opposition, the incidence of which appears to have risen significantly since the 1970s. In particular, the Act has been faulted for its paltry and easily delayed remedies for anti-union discharges, remedies that can be viewed as a minor cost of doing business for an employer committed to avoiding unionization (Weiler 1983). Likewise, the employer’s “right” to permanently replace economic strikers, established by Supreme Court dicta in 1938, has become the key to aggressive de-unionization campaigns. The doctrine has rendered the strike useless and virtually suicidal for many employees and has become employers’ Exhibit Number One in union organizing campaigns (Getman 1998). Yet legislative proposals to afford unions some workplace access, to expedite and recalibrate the representation process, to fortify the Act’s remedies, and to ban the use of permanent striker replacements have failed.

Since the Taft–Hartley amendments, employers have had far less to complain about. But they have sought legislative relief from the law’s nearly 70-year-old ban on employer-sponsored “employee representation” plans. Nonunion employers have been urged and many have sought in recent decades to tap the collective creativity and knowledge and to respond to the shared interests of their workforces by pursuing forms of labor-management cooperation. Yet federal law—specifically, Section 8(a)(2) of the Act—prohibits them from initiating or promoting any organization of employees that “deals with” the employer regarding terms and conditions of employment. Despite many trees’ worth of scholarly criticism and reform proposals, nothing has emerged from Congress.

Why has labor law stood still in the face of decades of social, economic, and legal change? Legislative inaction stems primarily from the fact that, for many decades, both organized labor and especially employers have had enough support in Congress to block any significant amendment that either group strongly opposes. “Enough support” does not mean a majority; it means a minority that is big enough, well-organized enough, and committed enough to tie up a bill through the arcane supermajority requirements of the Senate—for example, through filibuster—or to sustain a presidential veto. So, for example, the Labor Law Reform Act of 1977, which sought to address persistent labor criticism of the representation process and the inadequate deterrence of employer misconduct, garnered both majority support and fierce opposition. The latter prevailed: The bill died in the Senate in 1978, after a 5-week Republican filibuster, and six unsuccessful efforts to end debate. Similarly, several bills have been introduced since 1989 to prohibit the permanent replacement of strikers. In 1992, and again in 1994, bills gained majority support in both houses of Congress, but succumbed to determined and loquacious opposition in the Senate, where it takes sixty votes to end debate. Organized labor, too, was able to stymie a congressional majority on the single occasion in recent decades when employers pressed for a significant amendment of the NLRA. The Teamwork for Employees and Managers Act (“TEAM Act”) would have loosened the ban on employer-sponsored employee representation plans and permitted employers to establish nonunion “employee participation programs.” But organized labor held fast to the conviction that these schemes were still—as they were in 1935—vehicles of subtle employer coercion and additional weapons in employers’ already sizable arsenal of anti-union tactics. The TEAM Act narrowly passed both houses of Congress in 1997, but supporters were unable to override President Clinton’s veto.

These reform efforts were stymied because existing legislative structures enable committed and well-organized congressional minorities to block or hijack statutory reforms they oppose. But why does labor law reform in particular manage to provoke such committed and cohesive opposition? Some of the answers suggest a certain symmetry. In particular, both sides are well organized and both perceive high stakes: Employers are fighting for flexibility and managerial prerogatives that they claim are crucial to their economic survival, whereas unions are fighting for their very existence in the face of aggressive managerial resistance and long-term attrition. Only a serious crisis in industrial relations seems capable of breaking through the resulting impasse (Hyde 1990). The seeming symmetry is misleading, however. The political power of employers to resist labor law reform in Congress is grounded in capital’s indispensability to economic well-being and in its mobility within and across national borders and is fortified by the unusual unanimity with which business opposes reforms that might facilitate unionization (Pope 1999). There is also a deeper asymmetry that is reflected not only in the legislative process but in the typical labor dispute: Unions cannot live if they kill their “hosts”; they cannot thrive if employers do not. But employers can thrive without unions; indeed, many would prefer to see unions die out altogether and are willing to do their part to bring that about. Finding common ground is difficult in that sort of contest.

Organized labor is faltering badly under the existing, antiquated regime. It has a far greater need for labor law reform, but less political capacity to secure it, than employers have. So while organized labor has mounted several major efforts at labor law reform in the past thirty years, employers—equipped as they are with all of the political advantages of organization, internal unity, access, and wealth—have mostly been willing to bide their time, batting down periodic reform proposals that might tip the scales in unions’ favor, and watching union strength ebb away. That is not to say that statutory reform is a hopeless pipedream.² But there are stubborn structural reasons why it has been so elusive.

Ossification from Within: Built-in Obstacles to Change and Variation

The New Deal Congress that enacted the Wagner Act did not mean for its hard-won victory to be easily undone; it built in some barriers to change. In particular, Congress sought to insulate the new labor relations regime as much as possible from some highly suspect sources of change and variation—from employers and from the courts—and to entrust the administration of the Act to a new federal agency, the NLRB. But over the past half-century, these built-in obstacles to change have grown anomalous and out of sync with surrounding legal and economic developments, while the Board has become increasingly circumscribed in its ability to adjust the Act to modern conditions.

The Ban on Employer-sponsored Forms of Employee Representation

One potential avenue of extralegal renewal and experimentation was deliberately shut down by Congress in 1935 and has never been reopened. Employers seeking to meet (or deflect) workers’ demands for a voice in their working conditions while advancing their own interests in productivity and profits might have devised—and did devise, before 1935—a variety of alternative mechanisms for employee representation. Those mechanisms—for example, a works-council with representatives of employees and management—might have competed with independent unions for employee loyalty and perhaps spurred innovation within the labor movement. We might think of this as a “market” mechanism of reform. But the single most controversial provision of the Wagner Act closed down that potential avenue of change.

Company unions–employee organizations established and controlled by management—loomed in 1935 as among the chief barriers to independent union representation (Barenberg 1993). What is now Section 8(a)(2) of the Act thus made it unlawful for an employer “to dominate or interfere with the formation or administration of any labor organization or contribute financial or other support to it.” Section 2(5), in turn, defined “labor organization” broadly to include “any organization of any kind, or any agency or employee representation committee or plan, in which employees participate and which exists for the purpose, in whole or in part, of dealing with employers concerning grievances, labor disputes, wages, rates of pay, hours of employment, or conditions of work.” The ban thus reached not only organizations that purported to bargain collectively, or that masqueraded as unions, or that operated to coerce employees or discriminate against union activists or to evade the duty to bargain with the labor organization chosen by the majority, but it also reached some employee representation plans that were favored by employees and that worked well, by all accounts, within the particular organization (Jacoby 1989). The result was a choice—constrained and far from free—between full-fledged independent representation and no collective representation.

The law does permit some employee involvement programs: those aimed at improving “quality” and “productivity” that do not deal regularly with terms and conditions of employment;³ one-way communication and suggestion schemes that do not involve give-and-take between employees and management;⁴ and schemes that delegate to groups of employees full managerial authority over hiring, scheduling, discipline, or other employment issues.⁵ In short, it makes room for schemes that allow employees to do either more or less than “deal with” employers. But it does not permit employer sponsorship or support of institutionalized forms of give-and-take, consultation, cooperation, or negotiation over the things that matter most to employees at work. It rules out a wide swath of potentially valuable forms of employee involvement in workplace decision making.

Many employers have become convinced of the productivity benefits, even the necessity, of employee involvement in workplace decision making. Indeed, given the low risk and limited sanctions associated with a legal challenge, some employers appear to be experimenting with alternative forms of employee representation, many of which push or exceed the limits of legality (Kaufman 1999). Still, it seems likely that a narrower ban would produce more experimentation with new forms of employee voice (and certainly more information about existing experiments). The new plans would be employer initiated and would presumably be designed to give employees some voice while retaining managerial control. They would be “cooperative,” and arguably coopted, by design. But that is roughly what many employees say they want, according to the findings of Richard Freeman and Joel Rogers: Over 70% of all workers surveyed believe that an employee organization could be effective only with management cooperation, and over 80% of nonmanagerial employees prefer an employee organization “run jointly” by employees and management. Indeed, nearly 60% prefer “joint employee-management committees that discuss and resolve workplace problems,” a form of organization that is clearly illegal in the nonunion workplace under Section 8(a)(2) (Freeman and Rogers 1999).

Employees’ “preference” for organizations designed to cooperate with management appears to be partly an “adaptive preference,” a recognition that employers oppose independent unions and are able to stymie their efforts (Freeman and Rogers 1999). Moreover, employer-initiated plans might well promise more than they deliver; they might give the appearance of influence without any real power to effect change. They might simply deal management additional trump cards in the fight to remain nonunion and accelerate the decline of genuine independent employee representation. One thing is clear: The choice made by Congress in 1935 has impeded what we might call a market response to both employee and employer demands for viable forms of employee representation in workplace decision making. Of course that is what the original Wagner Act set out to do. And that choice, whether or not it serves us well almost seventy years later, remains in effect.

The Lack of a Private Right of Action to Enforce Employee Rights

For the modern observer of the law of the workplace, an obvious and prolific channel of law reform is private litigation. Costly and burdensome though it may be, litigation is a powerful cauldron of legal change, in which aggrieved individuals combine forces with zealous attorneys to test out new legal theories on sometimes receptive and sometimes creative courts. The aggregate impact of private litigation under broadly worded federal and state statutes and hospitable common law doctrines has utterly transformed the landscape of “employment law,” and has drawn the anxious attention of employers seeking to avoid its snares. But that has not happened within labor law. If employment law has electrified employers, labor law has proven to be a rather low-voltage instrument. Why?

The NLRA represents a very early and important “wrongful discharge” law. Enacted in an era of upsurging confidence in the administrative state and ingrained distrust of courts, it contains no private right of action. Rather, an aggrieved person may file a charge with the NLRB’s prosecutorial arm, which makes an unreviewable decision whether to file an unfair labor practice complaint with the Board. The New Deal choice of administrative rather than judicial adjudication constrained the range of remedies: reinstatement, backpay, and other equitable remedies, but not compensatory or punitive damages of the sort that only juries could award. The original choice of administrative enforcement is understandable. In 1935, the courts were the last places to which New Dealers and union activists would have turned for the enforcement of labor rights (Forbath 1998).

Since the original choice of administrative remedies and procedures was made, however, wrongful discharge law has come into its own. The Civil Rights Act of 1964 ushered in an era of expanding employment legislation and litigation. The availability of attorneys’ fees and of substantial damages under a growing array of federal and state statutes spawned the growth of a small but energetic plaintiffs’ bar in the private and nonprofit sectors. State courts began to develop common law doctrines of wrongful discharge in violation of public policy, backed by the threat of tort damages. Only a smattering of six- or seven-figure verdicts is needed to get employers’ attention and turn their minds toward reforms that reduce the threat of litigation.

Yet the explicit and longstanding federal rights of association and self-organization and the prohibition of employer retaliation for the exercise of that freedom is backed only by an administrative complaint procedure and a comparatively paltry financial threat. The right to be free from coercion and retaliation based on union organizing or other collective dissent is widely flouted by employers, who perceive too great an economic threat in the rumblings of union talk and too easy and cheap a response in the discharge of union adherents (Freeman and Kleiner 1990; Weiler 1983). As things stand, employers can treat the small and confined risk of an unfair labor practice charge as a minor cost of doing business.

What if labor law had kept up with the times and added a private right of action for anti-union discrimination that the law already condemns? We might have had a “common law” of anti-union discrimination, with cross-fertilization from other wrongful discharge doctrines (Estlund 2002). And if the resulting doctrines and remedies packed enough of a punch, they might have induced employers to choose carrots over sticks in fighting unionization, or even to let employees decide the matter for themselves. It might, in other words, have helped to do what the law already purports to do. That would have shaken up labor relations considerably.

Tying Up and Tying Down the Board: Judicial Review and Other Obstacles to Reform from Within

So an integral element of the original Wagner Act scheme was to confine the role of courts and to concentrate the power to interpret and enforce the Act in the NLRB. Yet the role that appellate courts do play in the administration of the Act through their judicial review and enforcement of the Board’s orders has proven to be one of the primary impediments to the Board’s effectiveness as an agent of change from within the existing legal scheme. The Board’s ability to keep the Act up to date is severely constrained by a combination of statutory language, stare decisis, and second—guessing by Congress and the courts.

Sometimes the statutory text itself quite “plainly” stands in the way of agency innovation, as in the case of Section 8(a)(2). The Board has sought to make as much room as possible for noncoercive employee participation plans, but in light of the broad language of the original Act, as well as old and broad Supreme Court interpretations of that language, the Board has proclaimed itself powerless to introduce greater flexibility into the Act.⁶ Obviously, statutes are meant to constrain innovation; very old statutes may simply bind more tightly.

Many of the crucial provisions of the NLRA are written in open-textured language, however, and are manifestly open to a range of interpretations. In principle, the courts are instructed to defer to “reasonable” interpretations of the Act by the NLRB, and not to substitute their own judgment.⁷ Part of the justification for this deference lies in the flexibility it affords the implementing agency to adjust to changing conditions, new evidence, and changing policy preferences. In practice, the Supreme Court seems to accord rather little deference to the Board and to find any number of reasons to substitute its own reading of the statute. Consider, for example, the Board’s recurring efforts to reconcile the bargaining rights of “professional employees” with the statutory exclusion of “supervisors.” Twice the Board advanced a narrow reading of the supervisory exclusion that was aimed at preserving the bargaining rights of registered nurses and other professionals who direct the work of others based on their professional skills but do not exercise other supervisory functions; and twice the Supreme Court has rejected those interpretations as inconsistent with the text of the supervisory exclusion.⁸ The saga suggests how little latitude the Board has in seeking to make a very old statute function in the modern workplace—in this case to accommodate the interests of a growing white-collar sector of the workforce.

Even when the language at issue is plainly open to interpretation, the Board’s interpretive latitude is limited by the fact that the statutory language has often been interpreted before. And when the language has been interpreted by the Supreme Court at some point in the past 60-plus years, that is that; the Board gets no leeway to interpret the law differently in response to changes in the world.⁹ This was brought home in the Lechmere decision.¹⁰ The Board had declared that union organizers had a right to distribute union literature in the employer’s outdoor parking lot, which was open to the public, because the employer’s interest in exclusion from this property was outweighed by the employees’ interest in getting information about the union. But the Supreme Court, by a 5–4 vote, rejected the Board’s resolution of the issue in light of its own 1956 decision that allowed union organizers onto employer property only in exceptional circumstances where employees were “beyond the reach of reasonable union efforts to communicate with them.” The Board’s supposed expertise added no weight to its conclusion that what was reasonable in 1956—when employees typically lived close to each other and close to the workplace—was not reasonable in the 1980s—when cars were ubiquitous, long commutes common, and communal gathering places rare (Estlund 1994).

Finally, even when the Board is operating within the constraints of both language and precedent, the courts give the NLRB a rather short leash, one that often strangles innovation. Professor James Brudney has shown that the judicial leash is shortest—at least since the 1960s—when it is the most “collective” of labor’s rights that are at stake (Brudney 1996a). This is evident in First National Maintenance v. NLRB,¹¹ in which the Court overruled the Board’s holding that the employer had a duty to bargain with the union over a decision to close part of a business. The Court concluded that management’s interest in quick unilateral action trumped the interest of the union in protecting members’ job security. A more quotidian example of courts’ skepticism toward the most collective of labor’s rights is their frequent rejection of “bargaining orders” that require the employer to bargain with a union whose majority support has been eroded by serious employer misconduct during a representation campaign (Brudney 1996a).

The Board faces political as well as judicial constraints. Congressional oversight of the NLRB, maintained through appropriations and confirmation proceedings, has at times been perfunctory, bordering on neglect. But employers’ allies in Congress are easily roused to action by departures from the laconic pace of “business as usual.” For example, when an energetic General Counsel began to make frequent use of preliminary injunctions as the only effective remedy against aggressive employer interference in union organizing campaigns, employer allies refused to reconfirm him and held up Board appropriations over the issue (Brudney 1996a).

In spite of the constraints it faces, the Board in the past has successfully put forward some innovative doctrines aimed at addressing contemporary workplace realities. One example, unanimously approved by the Supreme Court in 1995, is the Board’s “salting” doctrine, holding that individuals who are employed by a union to organize a workplace, and who seek or gain employment at that workplace, are “employees” protected against employer discrimination in hiring or firing.¹² Another innovation was Levitz Furniture Co.,¹³ which raised the bar for employers’ unilateral withdrawal of recognition from an existing union: A “reasonable doubt” about the union’s majority status is no longer enough; the employer must now show that the union has actually lost majority support. The decision puts a much-needed speed bump in the path of employers’ de-unionization plans. But the fact that this cautious and incremental decision is touted as evidence of the Board’s vitality and willingness to innovate speaks volumes: The standard deviation must be small indeed if this decision was cause for excitement.

More recent “innovations” by the current Board have consistently narrowed the Act’s coverage, cut back on Section 7 rights, and loosened constraints on employer conduct (Hiatt and Becker 2006). These are not insignificant changes, but they are hardly evidence of vitality, for they further constrict the Act rather than adapting it to the needs of 21st century workers and workplaces.

In contemplating the maneuvering room left to the Board, I cannot shake from my mind the image of a car cautiously negotiating the quaintly narrow, cobblestoned streets of an old European town. Slight deviations in either direction may cause a noisy collision. But the resulting commotion cannot obscure the fact that travel is slow, bumpy, and narrowly confined by the rigid and aged walls on either side.

Ossification from Without: Obstacles to Change from Outside the NLRA

So we find that the basic federal scheme of labor relations contains some built-in obstacles to innovation, and that this scheme, including the obstacles to change, has become entrenched at least since the 1950s by a political logjam in Congress. But our system of government distributes and diffuses power and the potential to bring about legal change both to other subordinate sovereigns through principles of federalism and to other branches of government through constitutionalism and the separation of powers. Moreover, domestic law is theoretically subordinate to the commands of international law. Yet there has been little or no innovation from these quarters either. The NLRA’s scheme of labor relations has been rendered largely impervious to change at the margins by state and local lawmaking, and to challenges from “above”—from constitutional and international law.

The Preemption of State and Local Labor Law

Labor law reform may be a nonstarter in Congress. But what about adaptations at the state and local level? In a federal system, state-by-state variations might ordinarily be expected and might permit experimentation around the edges of the national scheme. Successful experiments might have spread to other localities and might even have provided credible models for national reform. State and local lawmaking might also have given voice to popular discontent with the existing labor law regime that has no effective outlet at the federal level.

But even incremental state and local reform efforts run headlong into the wall of federal labor law preemption. Although the NLRA contains no express preemption provision, the Supreme Court has long held that Congress implicitly preempted a broad range of state laws. Modern labor law preemption essentially ousts states and municipalities from tinkering with the machinery of union organizing, collective bargaining, and labor-management conflict. Under “Garmon preemption,”¹⁴ by and large, states may not regulate activity that is arguably protected or arguably prohibited by the Act. And under “Machinists preemption,”¹⁵ states and municipalities may not weigh in on one side or the other of labor disputes by regulating activity that is clearly unregulated by the Act—that is clearly neither protected nor prohibited. Garmon and Machinists together virtually banish states and localities from the field of collective labor relations.

So, for example, states may not award additional remedies for conduct the Act prohibits; nor may they impose additional punishment on labor law violators, or even disfavor them in the award of state contracts.¹⁶ States may not extend a private right of action, with make-whole remedies, to an employee fired for seeking union representation, even though that tort remedy would fit comfortably within the contours of the tort of wrongful discharge in violation of public policy (Gottesman 1990). Affording remedies beyond what the Act prescribes is deemed to “conflict” with the federal scheme of limited remedies. The exceptional breadth of preemption doctrine rests on Congress’ presumed intent, as discerned by the Supreme Court in a series of rulings in the 1950s, 1960s, and 1970s. Professor Gottesman (1990) has shown that preemption doctrine came untethered from its statutory moorings and outgrew its justification during the 1950s, and that the far reaches of preemption doctrine are neither clearly implied nor clearly consistent with either national labor policy or principles of federalism. Yet preemption doctrine has become nearly as entrenched as the text itself.

It is hard to say for sure whether labor or management would have gained more from a narrower preemption doctrine, because it is hard to say how state and local governments would have exercised their broader authority over labor relations. But it is important to recall that labor’s basic rights are explicitly protected in the NLRA itself, and would thus be protected from infringement by even a narrow preemption doctrine—one that hewed more closely to the minimum demands of the Supremacy Clause. Management’s most important “rights,” by contrast, are not found in the Act but in state law itself. Some of those rights have been elevated to federal stature by the outer reaches of preemption doctrine, by implication from the law’s silence. But a narrower preemption doctrine would put them back in their proper place in state law, and would thus predictably afford more room for the regulation of employer conduct than for the regulation of employee and union conduct (Gottesman 1990).

A narrower preemption doctrine certainly would have afforded more room for experimentation and variation. As things stand, however, popular impulses and innovations that are stymied at the federal level have no outlet at the state and local level either. Whatever lessons might have been learned from state-by-state experimentation with the intermediate level principles of labor relations have gone unlearned.

The Deconstitutionalization of Labor Law

That brings us to the last two blocked avenues of potential labor law reform, which come not from within the statutory scheme but from “above”: The federal Constitution and international law both embody evolving fundamental norms of civil and human rights that might have cast a critical light on, or even supported a legal challenge to, aspects of American labor law. But that has not happened.

Before 1937, labor law had everything to do with the Constitution. The laissez faire construct of “liberty of contract” posited a right to sell one’s labor, and to buy the labor of others, on such terms as the market permitted, free from interference by the state. Minimum wage and maximum hours laws, laws legitimizing organized labor activity, and even laws prohibiting the discharge of union members were struck down by federal courts as unlawful interference with the right of employers and of workers to buy and sell labor on such terms as they saw fit.¹⁷ This was the unique American version of constitutional labor rights: The right to be free from the shackles of protective labor legislation.

But since the New Deal, the courts have had almost no resort to the Constitution as a source of critical scrutiny of the federal labor law regime. In particular, the Constitution has played virtually no role in expanding the statute’s protection or in challenging its suppression of collective action and expression by employees. It is not for lack of textual material. The labor movement long harkened to the 13th Amendment’s ban on involuntary servitude as a foundation for the rights to strike and to act in concert (Pope 2002). So, too, in the magisterial language of the 14th Amendment—the equal protection clause or the privileges and immunities clause—or in the First Amendment, with its freedoms of expression, association, and assembly, one might have found room for the development of workplace rights (Forbath 1985; Pope 1984).

Of course, these constitutional rights mostly operate against state action, not against employer interference. The one exception is the 13th Amendment; but that has long been construed to assure individuals’ ability to quit employment, and not, as organized labor would have it, to guarantee employees’ freedom from “wage slavery” and their rights of association and collective self-help. With the 13th Amendment relegated to the periphery of the modern economy, the state action requirement largely neutralizes the federal Constitution as a source of employee rights in the private sector. But that is not the whole story, for even where the NLRA itself regulates employee expression—as with the prohibition of peaceful picketing of neutral employers, for example—the Supreme Court’s Constitution has had almost nothing to say about it.

The conventional understanding of this constitutional silence points to the “switch in time” in 1937 from the highly intrusive judicial scrutiny of economic, and especially labor, legislation to a new era of deference to the legislature, and especially to Congress, in economic matters. The lesson might be encapsulated as “live by the sword, die by the sword”: The basic rights of American workers were won in a battle for legislative supremacy against the courts and their power of judicial review. When the political winds shifted against unions, deference to the “delicate balance struck by Congress” prevailed. It prevailed even when Congress passed laws abridging the freedom of speech of unions and employees, as it did with the Taft–Hartley restrictions on peaceful secondary picketing.

But the conventional story is too simple. For the Supreme Court in 1937 did not simply give up its extraordinary power of judicial review; it redirected it. The Supreme Court’s new constitutional project, haltingly undertaken at times, was to safeguard liberties of free expression and association, especially as they contributed to democratic politics, and to protect “discrete and insular minorities” whose access to the political process was impaired (Ely 1980). Initially it was unclear where labor activism fit in this dichotomized universe (Pope 2002). Was organized labor a social movement pursuing the cause of economic justice and equality, and thus deserving of energetic constitutional protection? Or was labor simply a market actor jostling for a bigger share of the economic surplus, and thus regulable by the legislature as it saw fit? In particular, how were the courts to view labor’s everyday forms of protest and appeals for public support—strikes, pickets, and boycotts in support of representation rights and collective bargaining gains?

For a short time after the enactment of the NLRA, it seemed that the former vision might prevail. In striking down a state ban on peaceful labor picketing under the First Amendment, for example, the Court proclaimed that the facts of a labor dispute were undeniably “matters of public concern.”¹⁸ But labor’s moment in the constitutional sun proved brief. The prolabor decisions all came against state and local actions. Once Congress extended its regulatory reach to unions in 1947, and federal preemption soon took hold, labor protest was relegated to the domain of economic activity, where deference to the legislature ruled the day.¹⁹ In hindsight, the Court’s brief experiment with constitutional prolabor rights seems to have served merely as an expedient transitional device in the transfer of authority over labor relations from states to the federal government.

Deference prevailed even after picketing and boycotts became constitutionally protected forms of protest in the hands of civil rights activists.²⁰ Organized labor’s role as the voice of the downtrodden had helped to spur the expansion of constitutional protest rights during the 1940s, and those decisions became crucial stepping stones for the further expansion of protest rights in the civil rights era (Klarman 1996). But labor did not, in turn, become the beneficiary of those expanded rights, at least not for picketing, its signature form of expression. Congress had struck a “delicate balance,” and the Court was loath to disturb it. The contrast between the Court’s vigilant protection of peaceful civil rights picketing and boycott activity and the deference to Congress’s ban on virtually identical activity by unions suggests a little of what has been lost in the deconstitutionalization of labor law (Pope 1984).

On the other hand, existing constitutional doctrine may also have worked—indeed, it has worked—more to the advantage of individual employers who oppose unions than to the advantage of unions as institutions. Among the few constitutional challenges to the Act that have been given a serious hearing, several have come at the behest of individual employees against unions. In particular, the Supreme Court has elaborated a rather robust First Amendment “right to refrain” from compelled association and political activity that casts a shadow over the Act’s provisions for “union security” and dues collection (Brudney 1996b). The quasi-constitutional “right to refrain” and its expansion over the years has worked a change in labor law—a change in the unsurprising direction of greater individual rights and weaker collective institutions. It shows one way in which the process of “ossification” is qualified, complicated, and uneven. The “right to refrain” decisions remind us that the Constitution is still a double-edged sword in the domain of labor law. It is thus difficult to predict where a more constitutionalized conception of labor rights would have led. But one can predict that it would have opened pathways of legal challenge and of change that have instead remained closed.

The Neglect of Transnational Labor Law

The last chapter of this story is the shortest. International human rights law has made no discernible mark on American domestic labor law. That is not because the United States rejects the existence of international human rights in the labor context; the rights of workers to associate freely and to form unions are long established and widely recognized human rights, which are recognized by the U.S. as such.²¹ But the U.S. does not embrace the full scope of the obligations that international law imposes on states with respect to those rights. If it did, or if there were more effective mechanisms for the enforcement of international norms, international law would provide critical traction for efforts to expand and invigorate the protection of workers’ associational activity.

International human rights standards require more than abstention from state repression of associational activity and more than the formal protection of such activity. International human rights advocates have thus criticized the notoriously patchy enforcement of basic labor rights in the United States. According to human rights observers, “[i]n a system replete with all the appearance of legality and due process, workers’ exercise of rights to organize, to bargain, and to strike…has been frustrated by many employers who realize they have little to fear from…a ponderous, delay-ridden legal system with meager remedial powers” (Compa 2000, p. 16). Given paltry remedies and long delays, “[m]any employers have come to view [legal sanctions] as a routine cost of doing business, well worth it to get rid of organizing leaders and derail workers’ organizing efforts. As a result, a culture of near-impunity has taken shape in much of U.S. labor law and practice” (Compa 2000, p. 10). Yet American legal institutions and decision makers have thus far been deaf to the claim that international labor law provides a potential model for American labor law or even a critical vantage point from which to view American labor law.

Unlike the peculiarly broad reach of labor law preemption or the peculiarly limited role of the federal constitution in labor law, the insulation of American labor law from transnational and international legal scrutiny is not peculiar to labor law. The same official resistance to the application of international human rights standards within American borders meets international criticism of the death penalty or of police brutality. As things stand, the labor arena is simply one of many in which the critical and transformative potential of transnational law remains untapped by domestic American legal institutions. And among the barriers to labor law reform, the resistance to transnational law is, to quote Pink Floyd, “just another brick in the wall.”

The Burdens of Ossification

Virtually all of the normal institutional channels of legal change are clogged or blocked in the case of labor law. The consequences of ossification are hard to discern with any degree of precision. For who knows what statutory reforms might have been enacted by a less paralyzed, less polarized Congress? Or what doctrinal innovations might have arisen through the adjudication of private claims? But from a few steps back, and given a few basic premises of the American labor laws, the overall consequences of ossification appear quite clear and quite lopsided.

The first premise is that employees—specifically, a majority of the relevant group of employees—must affirmatively choose collective representation. In other words, any new firm, and virtually any new operation or location, begins from a nonunion baseline. That baseline is not logically inevitable. A stronger attachment to industrial democracy might dictate employee representation within firms of a certain size, and require regular elections in which representatives may be chosen. Workplace democracy would then look a little bit more like political democracy, which guarantees voters a choice of representatives, not a choice about whether to be represented. But the nonunion baseline is sufficiently unquestioned in the American context to be taken for present purposes as a given. And it is a very big given. It means that, in an economy characterized by “creative destruction,” in which old firms shrink and die and new ones are born and grow, the forces of entropy alone tilt strongly against collective representation. The nonunion baseline means that, even without the threat of “decertification,” or the ouster of union representation, constant new organizing is necessary for unions to keep from disappearing from the scene.

We should also take it as given that union organizing takes place against the nearly uniform backdrop of employer opposition. Opposition to unions has a rational core, given the higher labor costs and lower profit margins that unionization tends to bring. That rational core is surrounded by layers of ideological attachment to unfettered managerial power and the prerogatives of ownership. But whatever historical and cultural contingencies may help to explain the depth of employer opposition to unions in the United States, that opposition is sufficiently enduring to be taken as a given here. Not only do most employers strongly prefer to operate nonunion, but they also have the economic power to impose that preference on their employees if the law does not effectively intervene. They own the workplace, and they effectively own the employees’ jobs under the prevailing American rule of “employment at will.”

Given that employees must make an affirmative choice to organize themselves, that employers almost invariably oppose that choice, and that they have the economic power to back up their opposition, it is evident—and it is a basic assumption of the Act—that employees need the law’s help in order to have a real choice with respect to self-organization. That being the case, weakness and ineffectuality in the law operate to fortify the nonunion baseline. And weakness and ineffectuality have been among the clearest consequences of the law’s obsolescence. Rigidity in the face of change has crippled the law’s capacity to protect employees’ associational rights. It has buttressed the nonunion baseline from which workplaces and workers begin, and allowed the “creative destruction” and the deliberate deunionization of organized workplaces to outrun the pace of new organizing. And on top of those failings, the law blocks many alternative avenues for collective employee voice that employers or states might have chosen to explore.

Conclusion

The forces underlying the decline of unionization and employee representation are many and complex. Increased, and increasingly boundaryless, competition in product markets as a result of falling technological, regulatory, and trade barriers to transportation of goods and services is undoubtedly a major factor in that decline; among other things, it helps explain the apparent escalation in employer opposition to unions. But to the extent that law is a factor at all in that decline—or that a different set of laws could have helped to slow or reverse the decline—the law’s ossification is a chief culprit.

American labor law has been largely insulated from both internal and external sources of renovation. It has been cut off from revision at the national level by Congress; from “market”-driven competition by employers; from the entrepreneurial energies of individual plaintiffs and the plaintiff’s bar and the creativity they can sometimes coax from the courts; from variation at the state or local level by representative or judicial bodies; from the winds of changing constitutional doctrine; and from emerging transnational legal norms. Even without knowing where any of these potential paths of change might have led, one can surmise that change or experimentation through one or more of these channels might have produced, over the past half-century, a body of labor law that was more responsive to the very different economic and social conditions that workers and employers face today.