Hostname: page-component-cd9895bd7-dk4vv Total loading time: 0 Render date: 2024-12-22T11:10:35.188Z Has data issue: false hasContentIssue false

Whose Work Counts? Congressional Republicans and the Battle over Employment Status, 1947–48

Published online by Cambridge University Press:  09 March 2023

Eva Bertram*
Affiliation:
Politics Department, University of California Santa Cruz, Santa Cruz, California, USA
*
Corresponding author: Eva Bertram Email: ebertram@ucsc.edu
Rights & Permissions [Opens in a new window]

Abstract

Conflicts over the employment status of Uber, Lyft, and other gig workers have made headlines in recent years. I argue that the conditions facing these workers and other independent contractors today are in many respects the result of policy decisions made seventy-five years ago, in hard-fought battles over which workers would—and which would not—be protected by New Deal social programs and labor laws for employees. In 1947–48, New Deal Democrats were poised to establish a more expansive definition of “employee,” extending eligibility to a range of workers excluded by more restrictive common law standards. The Republican-led 80th Congress thwarted the attempt to expand coverage, however, by blocking administrative initiatives, reversing court rulings, and redefining employment-based eligibility for federal labor and social protections. Their actions redirected policy on employment relations, restricting the reach of New Deal protections in the post–WWII economy and shaping the terms of subsequent conflicts over employment status in ways that have left broad power and discretion in the hands of employers.

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2023. Published by Cambridge University Press

The question of employment status in the workforce—and in particular, whether one is an employee or a “self-employed” independent contractor—has received growing political attention in recent years.Footnote 1 The question arises primarily in relation to workers’ protections and employer obligations under federal or state law. Are Uber and Lyft drivers employees, and therefore entitled to state or federal minimum wages? Are those companies responsible for contributing to state unemployment and disability programs for their drivers? Do home health aides have the right to bargain collectively, as employees? The questions have triggered bitter struggles at the state level over the status of such workers, in courts, legislatures, and ballot initiatives in states such as California and Massachusetts.Footnote 2 At the federal level, the decision to extend unemployment benefits to cover gig workers and other independent contractors under the 2020 CARES Act in the first months of the COVID-19 pandemic reflected a growing recognition that these workers need and deserve basic social protections, especially in uncertain and precarious economic times.

Yet even as the issue of employment status has gained visibility, the terms of the public debate remain narrow, in two respects. The issue is perceived as a relatively recent and limited development, a by-product of the increasing numbers of workers opting to work in the gig economy as independent contractors or of employers seeking to hire contingent workers (of various kinds) in recent decades. And it is seen largely as a consequence of market forces—individual decisions made by business owners and by workers in increasingly competitive markets for goods and services—rather than the result of politics or policy.

I argue that the contemporary contest over workers’ status as employees or independent contractors is historically rooted and has been highly political from the beginning. In the analysis that follows, I show that the conditions facing independent contractors in the workforce today are in many respects the result of policy decisions made seventy-five years ago, in hard-fought battles over which workers would—and which would not—be protected by New Deal social and labor legislation. These decisions, like many in social policy, inscribed new hierarchies and disparities precisely as they sought to address existing forms of inequality and deprivation through new government protections and rights. The focus here is on a critical and largely unexamined policy struggle in the late 1940s that sharpened and hardened enduring distinctions and inequities in the workforce, between those deemed employees (and therefore covered under most social and labor legislation) and those designated independent contractors (and therefore generally not covered).Footnote 3

The question of how to distinguish between employees and independent contractors is seen largely as a legal and technical matter today.Footnote 4 But as the boundary lines of employment status were being drawn in the 1930s and 1940s, it was at the center of two highly charged political struggles. One was over the purposes and scope of the new system of social and labor legislation created in the New Deal: Who was in, who was out, who pays, and who decides. The second was over power in the labor market: To what degree would capital be left free to determine whether and when the labor it hired would be shielded by the new social protections. Employers sought not only to avoid the costs and regulatory burdens imposed by the laws, but also to preserve their right to draw on a pool of unregulated (nonemployee) labor at will.

The issue of employment status became much more consequential after the passage of New Deal social and labor legislation. Because eligibility for New Deal protections was limited to employees, the new laws dramatically raised the stakes of which side of the employee/independent contractor line a worker fell. Workers found themselves protected or unprotected depending on their designation. Employers faced major new costs and obligations for their employees under the new laws—and significant incentives to evade these costs by defining workers as independent contractors rather than employees when and where they could. As I demonstrate below, a brief window of opportunity opened in the mid-1940s, one that might have led to significantly expanded protections for a wide array of American workers. Federal administrators backed by the courts were poised to implement a more inclusive definition of employee for key New Deal programs, extending coverage to many who had been deemed independent contractors. But by 1948, congressional Republicans, backed by allies in business, had waged—and won—a sustained legislative campaign to block the effort, and the window slammed shut. Though it has received little attention in the scholarly literature, the stakes were evident to all at the time. The conflict consumed attention at the highest level of every branch of government, generating three veto messages, four Supreme Court rulings, and three pieces of legislation specifically addressing the question of when and if a worker was an employee or an independent contractor. When the dust settled, the Republican-led 80th Congress had left an enduring institutional legacy of exclusion and contention, and a labor market logic that advanced the prerogatives of employers over the protections of workers in the gray zone between employment and independent dealing.

For scholars of American political development, this account offers new insight into the transition to a modern liberal employment regime, and the mixed legacies of the New Deal more broadly. Karen Orren's work changed our understanding of the origins and development of the liberal order in the United States, by demonstrating the persistence of old common law principles of labor governance, operating alongside liberal democratic institutions well into the twentieth century. Orren charts the halting and belated transition from feudal to liberal labor relations; she concludes that the shift from “the regulation of employment by the law of master and servant to the regime of collective bargaining, and from the common law's embedded position in American government to the fully legislative polity” was only fully realized in the New Deal.Footnote 5 This account suggests that well after the New Deal, that transition was contested and incomplete. The conflicts over employment status examined here reveal the tensions, contradictions, and nonlinear character of the shift. Successive legislative, bureaucratic, and judicial decisions in the 1940s capture the back-and-forth struggle between those seeking to repudiate common law principles in search of more liberal and equitable interpretations—and those determined to preserve older master-servant conceptions in implementing the new social and labor legislation. In the end, these events suggest that James Atleson may be correct in concluding that core master-servant concepts were not fully displaced by modern American labor law, but “are still embedded in the relationship” between workers and employers decades later.Footnote 6

The developments of 1947–48 raise questions as well about the broader role and impact of the New Deal. A sizable literature has mapped the limits of New Deal liberalism, documenting, for example, the racial and gendered occupational exclusions that rendered New Deal social protections deeply inequitable.Footnote 7 Conservatives in Congress—Southern Democrats as well as Republicans—were key architects of provisions excluding agricultural and domestic workers from coverage. The impact of these exclusions fell disproportionately on Blacks and women, and even after a wider range of occupations gained coverage in subsequent years, legacies of exclusion and discrimination meant that these groups were incorporated into the U.S. welfare state and economy on profoundly inequitable terms. The analysis here exposes an additional axis of exclusion—employment status—operating alongside race and gender. Although they have received less attention, exclusions based on workers’ status as nonemployees have also often disproportionately affected women and people of color.Footnote 8

Scholarship on the New Deal has also explored the constraints imposed by economic assumptions built into New Deal social protections, through eligibility and benefit structures tied to employment. Theda Skocpol and Margaret Weir were among the first to show the importance of conditions of full employment—and the corresponding need for job guarantees—to ensure the effectiveness of social protections accessed through work.Footnote 9 The failure to achieve employment assurance policies during or after the New Deal limited the reach and relevance of the new protections.Footnote 10 This account shows that the promise of universal and effective social protections rested not only on the availability of jobs, but also on an inclusive definition of employee in the relevant statutes. The question was not just whether people could find work, but also whose work counted as employment.

If this analysis deepens our understanding of the illiberal exclusions and inequities built into New Deal legislation, it also reveals a more complex developmental story, one in which struggles in the 1940s feature more prominently in chronicling how and when the more universal and social democratic aspirations of the New Deal were undercut. New Deal proponents in these years sought to inscribe expansive definitions of employee, not only to advance liberal principles of equity and inclusion, but also to meet an institutional imperative: the financial viability of broad entitlements like Social Security required a growing pool of covered workers, creating a logic of expansion.Footnote 11 The 1947–48 congressional session thus emerges as a turning point in the development of employment relations and social provision. Had New Deal supporters succeeded in establishing a broader interpretation of employee, the category would have incorporated, over time, many workers (including nonwhite workers and women) who were ultimately excluded from coverage.Footnote 12 The actions of Republicans instead redirected policy in ways that significantly restricted the reach and scope of New Deal protections for workers.

The arc of policy development and the durability of the Republican initiatives examined here shed light on broader processes of institutional development and resilience.Footnote 13 At the macro-level, these events reflect the tensions and transformations occurring between a liberal order of labor regulation under construction in the 1930s and 1940s and an older master-servant order under threat.Footnote 14 This account suggests that the conflict was not settled by New Deal labor and social protection laws, but instead was channeled into new rounds of struggles over how to interpret and implement the new laws. At the meso-level, 1947–48 emerges as a critical juncture in the institutional development of employment relations, setting policy on a path-dependent course, detailed below.Footnote 15 As new incentives, interests, and expectations were created, the trajectory would prove increasingly difficult to reverse. In the end, congressional Republicans won a two-part victory in 1947–48: (1) They thwarted the attempt to expand coverage by redefining employment-based eligibility for federal labor and social protections, and (2) they circumscribed the role of the state at a critical point in the political development of capital-labor relations by reversing court rulings, blocking administrative initiatives, and leaving broad power and discretion in the hands of employers.

I examine the events of the 1940s in three phases, tracing a conflict that started in the federal bureaucracy, moved to the courts, and ended in the legislature. The first three sections below examine each phase in turn. The fourth section turns to the aftermath and institutional legacy of these developments, and the final section considers current struggles over employment status in the context of this history.

1. Round One: Administrators Confront Ambiguity

Round One began in the federal bureaucracy, as the lack of clarity and agreement on how to define employee for the purposes of coverage under the new social protections drew a wide range of actors from across the federal government into the ring and into a heated debate. It originated in three major pieces of legislation that established for the first time a national system of labor and social protections for American workers. The 1935 Social Security Act created social insurance programs that partially replaced wages lost to temporary unemployment or retirement. Employers were responsible for paying unemployment taxes on behalf of their employees under the new unemployment insurance program and for half of the contribution to the new old-age insurance program, with employees shouldering the other half. The 1935 National Labor Relations Act (NLRA), or Wagner Act, established the right of employees to create or join unions and to bargain collectively with their employers, and it required employers to recognize those unions and bargain in good faith. The 1938 Fair Labor Standards Act (FLSA) enacted federal minimum standards and protections for pay and working conditions. It set the first federal minimum wage, required overtime pay for eligible employees who worked more than forty hours a week, and required employers to keep records of hours worked and wages earned by their employees to ensure compliance with these regulations.Footnote 16

As written, these new laws applied only to workers who were employees. Yet the meaning of "employee" was left vague and imprecise, and Congress did little to clarify the distinction between employees and those hired as independent contractors. There was no singular or unified definition of employee: each of the New Deal laws contained its own language describing the workers covered under the statute. None were clear, and some were maddeningly circular. The word “employee” for the purposes of the FLSA, for example, included “any individual employed by an employer.”Footnote 17

This was not simply an administrative oversight. It was intensely political and was soon the focus of a protracted battle over whose work would count as “employment”—meriting protection under the new laws and imposing obligations on those who hired them—and whose work would leave them unprotected in the post–New Deal economy. The conflict, at root, was over the transition to the modern system of employment relations embodied in New Deal legislation. Business leaders recognized what was at stake. They had mobilized against key New Deal initiatives in the 1930s.Footnote 18 When the legislation passed despite their objections, employers’ opposition was channeled in part into administrative, legal, and legislative struggles over whether and which of their workers would be considered employees covered under the new laws. Far from a technical or legal question of definition, this opened a front in the post–New Deal conflict between capital and labor that has remained active ever since.

The dispute over employment status in these years focused primarily on the Social Security Act and the Wagner Act.Footnote 19 In the case of the Social Security Act, the contest over the definition of employee (and thus the scope of coverage) began in the legislative process that produced the law in 1935. The Roosevelt administration cast a very wide net in its proposed legislation: “The term ‘employment’ shall mean any employment . . . under any contract of hire, oral or written, express or implied.” But the House Ways and Means Committee stripped this broad language from the bill, leaving the final version of the act stating vaguely and unhelpfully, in several titles, that “employment means any service . . . by an employee for his employer.”Footnote 20

The trouble started soon after. Two agencies were charged with administering the new Social Security Act. The Social Security Board (later located in the Federal Security Agency [FSA]) oversaw benefits under the new social insurance programs, and the Bureau of Internal Revenue (located in the Treasury Department) governed collection of the required taxes.Footnote 21 It quickly became clear to administrators of both agencies that Congress's definition was not adequate to guide decision-making, particularly regarding the workers who fell into the nebulous gray zone between a standard employer-employee relationship and independent dealing. Was an outside salesperson for a company an employee, deserving of benefits under the Social Security Act? What about a taxicab operator? How should the decision be made?

The numbers involved were sizable. By one estimate, more than 1.25 million workers were in the gray zone, not clearly employees or independent contractors. Many were in low-wage or unstable work. “This group included certain taxicab operators, private-duty nurses, owner-operators of leased trucks, industrial home workers, entertainers, newspaper vendors, contract loggers, commission oil plant operators, mine lessees, journeymen tailors, filling station operators, and more than 600,000 salesmen,” noted Wilbur Cohen, who was then Technical Advisor to the Commissioner for Social Security, and James Calhoon of the Bureau of Old-Age and Survivors Insurance.Footnote 22

The agencies drew up regulations to implement the act in 1936. Lacking clear guidance from Congress, they turned to common law conceptions that were used to guide rulings not in the new world of federal social legislation, but in tort cases dating to the previous century. In doing so, they were reaching back to the terms and conditions of an earlier employment regime. The common law distinction between employees and independent contractors had its origins in a premodern economy. The modern employer-employee relationship was constructed on the master-servant relationship of the preindustrial era, and the label “independent contractor” was linked to early notions of an “independent calling.” The term was used to indicate that such a worker was at liberty to work for multiple clients—perhaps as a skilled craftsperson or artisan—and was not dependent on a single master (or employer) to earn a living.Footnote 23 Courts were occasionally asked to address the issue beginning in the mid-nineteenth century, often to clarify whether an employer could be held liable for harm resulting from a worker's negligence. In what came to be called the common law “control test” of employment status, if a court found that the employer had the right to control in detail how the work was carried out, then the worker was judged to be an employee. If not, the worker was considered an independent contractor and the employer was not held responsible for any negligence.Footnote 24

However suited the common law control test might have been to addressing such liability questions, federal administrators soon recognized that it provided too little guidance to serve the new social legislation. They took the issue back to Congress as part of a broad set of recommendations that informed the first major amendments to the Social Security Act in 1939. The Social Security Board wanted a clarification of Congress's intent. Whom was the law intended to cover, among those in this gray zone of employment? “Old-age insurance coverage is at present limited by the undefined terms ‘employer’ and ‘employee,’” the board said. They recommended that “this provision be expanded . . . to cover persons who are for all practical purposes employees, but whose present legal status may not be that of an employee,” emphasizing in particular the thorny category of personal services, including insurance and traveling salespeople.Footnote 25

There were, however, deep differences over the question, and these emerged in hearings in the House Ways and Means Committee over the amendments in early 1939. Business representatives protested the expansion and sought specific exemptions from paying unemployment taxes for their workers. “The only thing we are asking,” testified an insurance company representative, “is that the law continue to be restricted to the employer-employee relationship of the common law.”Footnote 26 Union leaders countered that the administrative agencies (particularly Treasury) were using criteria that excluded too many workers.Footnote 27 The House ultimately proposed a modest amendment to the definition of employee to address the status of salespeople, noting that for these purposes, the “common-law concept of master and servant should not be narrowly applied.”Footnote 28 But the Senate refused to go along, and in the end no action was taken. The agencies were back where they had started.Footnote 29

By this time, a growing number of disputes over who exactly was an employee for the purposes of New Deal legislation were landing not only on the desks of federal administrators but also in the courts. And courts were issuing conflicting decisions, at times siding with the rulings of administrative agencies and at other times reversing those rulings. Beginning with a 1941 Texas case, momentum grew for a narrower interpretation, with court rulings increasingly focused on the specific language in individual contracts between employers and workers, particularly regarding how much control would be exercised over the performance of the work.Footnote 30 The Treasury Department began to adopt a narrower standard in its determinations of employment status, in line with this trend. The Social Security Board, however, continued to use the broader definition that both agencies had embraced until then, weighing a range of factors beyond the right to control. This created yet another contradiction for federal bureaucrats. In numerous cases, the Social Security Board determined that a given worker was an employee (and therefore eligible for benefits), while Treasury ruled that the worker was an independent contractor (and therefore did not collect taxes on the worker's behalf).Footnote 31 Meanwhile, administration officials observed that the rulings were beginning to have adverse consequences in the larger labor market. They were encouraging “certain employers to revise their contracts” with their workers “for the specific purpose of avoiding liability for Federal employment taxes,” reported Cohen and Calhoon.Footnote 32

By the early 1940s, the question of employment status was generating confusion and conflict across the federal government—in the courts charged with interpreting the laws, in the federal bureaucracy tasked with implementing the laws, and even among members of Congress who had made the laws. The first round of political skirmishes over the issue had drawn in every branch of government, yielding a round of court decisions, bureaucratic regulations, and congressional hearings, but no agreement. What appeared at the outset to be a minor if nettlesome technical and administrative matter for federal administrators was quickly becoming a consequential legal and highly political issue. For those workers whose status was unclear, meanwhile, the outcome of these conflicts would determine whether they would be covered by the array of New Deal protections, providing income assistance during retirement and jobless periods, minimum wages and overtime pay, maximum work hours, and the right to unionize and bargain collectively.Footnote 33

2. Round Two: The Supreme Court Steps In

Recognizing the disarray and disagreement that had erupted over employment status, the Supreme Court stepped into the fray in April 1944. “Few problems in the law have given greater variety of application and conflict in results than the cases arising in the borderland between what is clearly an employer-employee relationship and what is clearly one of independent dealing,” the Court stated in its ruling on NLRB v. Hearst Publications, Inc. Footnote 34

The Hearst case arose in Los Angeles when four daily newspaper publishers refused to bargain collectively with the union representing the news vendors who distributed their papers. The publishers argued that they were not required to do so because the news vendors were not employees entitled to organize and bargain collectively with the publishers under the terms of the Wagner Act, but were instead independent contractors. The National Labor Relations Board (NLRB), the agency charged with implementing the Wagner Act, reviewed the case and determined that the vendors were indeed employees. The Supreme Court ruled in favor of the NLRB, based in part on the fact that the vendors were working full-time for Hearst and other publishers at established locations, typically for extended periods of time, and were “dependent upon the proceeds of their sales for their sustenance.”Footnote 35 The Court pointedly rejected the publishers’ argument that in the absence of a clearer definition by Congress, the meaning of “employee” must be determined using common law standards. Instead, the Court ruled that “in doubtful situations,” more inclusive criteria should apply, with the definition of employee “determined broadly by underlying economic facts rather than technically and exclusively by previously established legal classifications.”Footnote 36 Far more expansive than the common law “control test,” this standard came to be called the “economic realities” test.Footnote 37

Hearst had immediate reverberations, yet the legal situation remained unsettled, with courts applying a range of standards to reach decisions. In 1944 and 1945, some 250 cases were decided, with some lower courts following the logic of Hearst and others issuing conflicting opinions.Footnote 38 Seeking to clear up the confusion and decisively settle the issue, the Court took on a series of cases intended to provide guidance to the lower courts and administrative agencies.Footnote 39 Three rulings were delivered in short order: United States v. Silk and Harrison v. Greyvan Lines were handed down on June 16, 1947, and Bartels v. Birmingham was decided one week later.

In each, the question at hand was whether the workers (coal unloaders and delivery drivers, truck drivers, and band members at dance halls) were employees or independent contractors. The decisions affirmed and expanded on the logic and conclusions of Hearst, underscoring three key points. First, the purposes of the relevant legislation (in this case, social security legislation) must be considered in determining who should be covered. If the purpose of the Social Security Act was to help relieve “burdens that rest upon large numbers of people because of the insecurities of modern life,” then coverage should be extended to the workers who face those burdens, particularly given the act's broad definition of employment, the Court said in Silk and Greyvan. The term “employee,” in short, should be “construed to accomplish the purposes of the legislation.”Footnote 40 Second, the “control test” was not enough. The full range of factors in a given situation should be evaluated to determine whether a worker is an employee or an independent contractor—including not only the degree of control exercised over the work but also whether the relationship was a permanent one, whether the work in question was integrated into the business as a whole, and other considerations.Footnote 41 And third, the determination of employment status should be based on the worker's actual relationship with the hiring entity, not technical definitions of that relationship under common law. The seminal conclusion was clearly stated in the Bartels ruling: “In the application of social legislation employees are those who as a matter of economic reality are dependent upon the business to which they render service.”Footnote 42

At issue in these cases was how to define the modern worker, caught at the interface of two competing regimes of employment relations—the premodern common law system and the liberal order advanced by the New Deal. Stepping into the breach, the Court rejected master-servant law as an appropriate interpretation of the contemporary capital-labor relationship, and it recast an employee as someone who was not a servant, under the formal control of a master/employer, but who was nonetheless dependent on that employer as a matter of “economic reality.”Footnote 43

For administrative purposes, the Court's decisions had largely affirmed the judgments of the Social Security Board and suggested that Treasury needed to revise and broaden its standards. Armed with a clear set of interpretations by the Supreme Court, the two agencies went back to the drawing board to craft a new set of regulations that rested on the “economic realities” principle underlying the Court's rulings.Footnote 44 As the agencies formed a joint committee later that year to align regulations with the Supreme Court's interpretation, there was reason to expect that the struggle over how to define employment for purposes of New Deal social legislation had been resolved in favor of a more expansive standard, promising broader coverage. Instead, the conflict entered a third phase, this one between Congress and the executive branch.

3. Round Three: The Republican Counteroffensive

Round Three unfolded in a fast-moving volley of legislative initiatives and veto messages issued between 1947 and 1948. After a major strike wave in 1945 and 1946, President Truman laid out a policy agenda on labor issues in his January 1947 State of the Union address. The president urged Congress to take up “the extension and broadening of our social security system,” arguing that the solution to the nation's labor strife “is to be found not only in legislation dealing directly with labor relations, but also in a program designed to remove the causes of insecurity felt by many workers in our industrial society.”Footnote 45 Neither the Truman administration nor congressional Democrats took steps to advance the president's position legislatively, however, and they soon lost control of the agenda.Footnote 46

A Republican-led Congress had been elected two months earlier for the first time in fifteen years, and congressional Republicans were spoiling for a fight over labor and social protections they considered too expansive and too burdensome for business. They had willing partners in conservative Southern Democrats, who had sought repeatedly to scale back union and labor protections legislated in the New Deal. Congress's initial shot across the bow came in the form of the Taft-Hartley Act, the Republicans’ response to the 1935 Wagner Act.Footnote 47 Introduced in April 1947, Taft-Hartley proposed major rollbacks and revisions of New Deal labor protections. Provisions to weaken union security, restrict closed shop and union shop agreements, outlaw a range of strikes and boycotts, and extend federal authority to curtail strikes drew outrage from Democrats and their labor allies, who saw the measure as an attempt to eviscerate the new system of union protections and subvert the power of organized labor.Footnote 48

Buried in the proposed legislation was a little-noticed amendment that explicitly excluded independent contractors from the Wagner Act's protections. The provision was not discussed directly in the weeks of congressional hearings and floor debates on the legislation, but it sent a clear signal that Republicans intended to push back hard on federal administrators’ progress toward a broader and more inclusive definition of employee in social legislation.Footnote 49 The House Ways and Means Committee report accompanying Taft-Hartley reflected a concerted effort to construct a clear record of congressional intent on the matter. The committee rejected the considered judgment of both the NLRB and the Supreme Court, suggesting that the issue was straightforward and Congress's position clear and obvious to any intelligent observer: “An ‘employee,’ according to all standard dictionaries, according to the law as the courts have stated it, and according to the understanding of almost everyone, with the exception of the members of the National Labor Relations Board, means someone who works for another for hire.”Footnote 50 Underscoring how common-sensical such a definition should have been (and by extension, accusing the NLRB of deliberate and inappropriate policy overreach), the committee stated pointedly, “Congress intended then [in 1935], and it intends now, that the Board give to words not far-fetched meanings but ordinary meanings. To correct what the Board has done, and what the Supreme Court, putting misplaced reliance upon the Board's expertness, has approved, the bill excludes ‘independent contractors’ from the definition of ‘employee.’”Footnote 51

This was in no sense a mere “correction,” nor was the meaning of “employee” clear and undisputed, as evidenced by the many cases brought before administrators and the courts in the years since the law's passage. But the amendment drew little attention in the sprawling high-profile debate over the larger Taft-Hartley bill, which liberal Northern Democrats and labor leaders viewed as a direct assault on “the heart of American industrial democracy” on behalf of business interests.Footnote 52 With Republicans in control of both houses and Southern Democrats joining their ranks in support of antilabor measures, the bill passed easily in early June.Footnote 53 President Truman's June 20th veto was overridden by the House on the day it was issued and by the Senate three days later.

It soon became clear that the independent contractor provision quietly slipped into Taft-Hartley was part of a larger Republican plan to reinterpret New Deal social and labor legislation in ways that limited coverage for workers and protected business prerogatives. Even before the ink was dry on Taft-Hartley, Republicans had advanced on another front. The charge was led by Representative Bertrand Gearhart, a Republican from Fresno, California. He introduced legislation to reverse a district court ruling that news vendors should be considered employees under the Social Security Act (as the Supreme Court had ruled they should be under the Wagner Act).Footnote 54 The legislative mechanism this time—in what would become a favored tactic—was an explicit exemption, or carveout, to exclude a particular group of workers.

Gearhart's proposed legislation (H.R. 3997) amended the Social Security Act and the Internal Revenue Code to explicitly exclude news vendors from unemployment and retirement insurance coverage under the act.Footnote 55 The vendors, Republicans insisted, were independent contractors, not employees, and were therefore not entitled to benefits under the Social Security Act. The bill was backed by business leaders in the news publishing industry with much at stake, and the dispute flared in hearings before the House Committee on Ways and Means in June 1947. The employers saw a clear and urgent threat to their longstanding employment practices by an encroaching social security system, reflected in local Social Security Board and court rulings. Charles Arnn, vice president of a Los Angeles newspaper and representing “a committee of all metropolitan newspapers on the Pacific coast from the Canadian to the Mexican border,” told the committee:

It is obvious what the extension of this philosophy means. Fact is the local social security offices in Chicago has already moved to cover Chicago corner boys [news vendors] with the provisions of the Social Security Act. The same thing is true of New Orleans. Every metropolis is thus faced with a threat. In this emergency, legislative correction such as drafted in this bill offers the only practical remedy at this time.Footnote 56

Arnn and others testified that such coverage was neither appropriate nor feasible. It was not possible to obtain the necessary information or collect the requisite tax from the news vendors who ran corner stands; the result would be “interminable records and interminable administrative costs.”Footnote 57

What was really at stake, these business leaders understood, was an entire business model. Its destruction, they argued, would not only be bad for business but bad for the workers themselves. George Bertsch, representing both the Baltimore Sun and the Baltimore News-Post, said, “Not only do we consider the equitable collecting of such taxes a practical impossibility, but we are convinced that any effort to enforce such collection would result in the necessary elimination of that type of sale by many publishers, to an ultimate disadvantage and loss to the vendors themselves.”Footnote 58 In fact, warned Harry Price, representing the San Francisco Publishers Association, “extreme” actions such as providing coverage to news vendors or other “small businessmen” would endanger the very system Congress had established. “The social security system must be protected from extremists,” Price said. “Inclusion of the small businessman within its scope under the guise that he is an employee will, in time, only invite legislation in the opposite direction.”Footnote 59

Truman administration officials saw a very different threat. They believed that the political and economic survival of the social security system—still in its early years—rested on broad, even universal coverage. William Mitchell, Acting Social Security Commissioner, worried about the potential domino effect of exclusions such as this one on the system overall: “I am concerned primarily because to exclude a group of people . . . whose conditions of employment become rather similar to other groups” poses a risk that “by progressive steps, one by one, we might be setting a precedent in the exclusion of this group that would be picked up by others, and the sum total of which would be exclusions of substantial numbers.”Footnote 60

Disagreements over how inclusive and universal eligibility should be ran to the core of competing visions of the new social security system. They surfaced in an exchange between Commissioner Mitchell and Representative Thomas Jenkins (R-OH):

Mr. Mitchell: . . . The need for social insurance protection on the part of the excluded groups is great. . . . Without doubt, the social security program would be strengthened and placed on sounder ground by the inclusion of all these now excluded groups. . . .

Mr. Jenkins: …You mean to say there is no worker who should not be included? You mean that everybody ought to be included?

Mr. Mitchell: Yes, sir; our position is that under old age and survivors insurance, the closer we can get to universal coverage, the better the system will be.

Mr. Jenkins: …But when you say everybody ought to be in, you surely do not want to take that position, do you? Everybody?

Mr. Mitchell: Yes, sir; we believe that all groups should be included.Footnote 61

An equally heated debate arose over how decisions should be made regarding which workers would be included, and more importantly, who—Congress, federal agencies, or the courts—had the authority to make these decisions. Federal administrators explained that in borderline cases (such as the news vendors), determining employment status required a case-by-case assessment of individuals’ work circumstances.Footnote 62 “One of the most difficult problems in the administration of the social security program is just the one that is facing the committee tonight; that of distinguishing between the employee and the self-employed individual,” testified Alanson Willcox, Assistant General Counsel of the FSA. “There are all sorts of shadings and degrees, and it is not confined to the newspaper business by any means. It has been one of our major administrative problems. As you will remember, the Congress has never defined what is meant by ‘employment.’”Footnote 63 Representative Richard Simpson (R-PA) shot back, “When you hit these close ones, why do you not come back to Congress to get an explanation of them?” To which Willcox replied, “We hit them every day of the year.”Footnote 64

When Willcox described agency efforts to comport with court rulings and pledged that “we have done our level best to try to interpret the law as we see it,” he drew a sharp retort from Committee Chair Harold Knutson (R-MN), who accused the agency of bureaucratic overreach. “Like other bureaus, you are reaching out for more and more income and power,” he said, adding, “I have been here since before you were born, and it is clear to me that the tendency is to get more and more employees, because the more employees you have . . . the bigger your appropriations. That is what we have to fight all the time. We are fighting it now.”Footnote 65

The Republicans’ news vendor bill ultimately passed both houses in July 1947—only to be blocked by a pocket veto by President Truman on August 6. In a signal of the importance of the issue to the administration, Truman released a strong veto message:

H.R. 3997 would remove social security protection from news vendors who make a full-time job of selling papers and who are dependent on that job for their livelihood. They and their families are exposed to the same risks of loss of income from old age, premature death, or unemployment as are factory hands or day laborers. They unquestionably fall in the group for whose protection our social security laws were devised.

The president spoke not only to the moral imperative of an inclusive system of social security but also to the institutional logic of extending coverage: The structure of the system “demand[s] an expanded social security system” serving a growing number of workers. Truman warned pointedly that the measure would encourage additional employers “to seek exemption whenever they can allege that the law is inconvenient.” It would set a “precedent for special exemption,” he said, and would “open our social security structure to piecemeal attack and to slow undermining.”Footnote 66

The president's veto statement arrived on the heels of the three Supreme Court rulings (Silk, Greyvan, and Bartels) handed down in June 1947. Both the White House and the Court were now on the record affirming an expansive interpretation of which workers should be covered as employees and which should be excluded as independent contractors. By late 1947, despite Taft-Hartley, it looked as if the administration's position would prevail and a broader and more inclusive definition of employment relations would soon guide coverage under key New Deal social legislation. The vendor bill had been blocked, the Court had ruled decisively on the question of employment status, and the Treasury Department and the FSA, working together, had released and were poised to enact new regulatory guidance.

The agencies’ proposed regulations held the key to implementing the new and more inclusive interpretation. The regulations fundamentally redefined the boundaries of the employment relationship and of inclusion in the social security programs, in line with Supreme Court rulings. They stated plainly, “An individual performing such services for a person is generally an employee of such person unless he is performing such services in the pursuit of his own business as an independent contractor.” Employment status, the regulations said, was to “be determined primarily from the terms and purposes” of the Social Security Act, which were “to replace a part of the wage income lost through old age, premature death, or unemployment.” Criteria for making determinations were not to be limited to a common law definition of master and servant, relying instead on a range of factors and the “economic realities” approach. Perhaps most importantly, the regulations created a presumption of employment. In other words, a worker who performed services for and was economically dependent on a business was presumed to be an employee, and eligible for protections under the law, unless proven otherwise.Footnote 67

But things took an unexpected turn. The proposed regulations were published in the Federal Register on November 27, with final publication slated for January. At the eleventh hour, congressional Republicans intervened. The chairs of the Senate Finance and House Ways and Means Committees, Sen. Milliken (R-CO) and Rep. Knutson (R-MN), “asked the Treasury Department to defer releasing the regulations until Congress had time to study the question further,” and Treasury complied.Footnote 68

In short order, Republican legislators were back with a bolder and more comprehensive plan of attack. They remained intent on shutting down the effort to establish a more universal definition of eligible workers for purposes of social and labor legislation, by blocking or reversing actions taken by the courts and the bureaucracy. Rep. Gearhart again led the way, introducing two new bills in January 1948. One (H.R. 5052) was a new vendor bill, identical to the one Truman had vetoed. The other (H.J. Res. 296) was a more sweeping measure designed to nullify the new regulations proposed by the Treasury Department.Footnote 69 The Republican strategy was to stand firm on procedural grounds and common law tradition, even as Democrats called attention to the large numbers of workers who would be excluded from the social security system under the Republican plan. The strategy rested on three major claims.

The first was a matter of prerogative and procedure: it was Congress's prerogative, not the courts’ or the bureaucracy's, to define and decide who would be protected by the nascent system of social and labor protections. Their proposed legislation was not about denying social security coverage to particular workers, Republicans asserted, but about restoring congressional authority to decide how and when social security would be expanded. Representative Edward Jenison (R-IL) explained that the vendor bill “in no way indicates a desire to restrict the proper expansion of social security into every proper field. Rather it limits only unwarranted and unjustified expansion by court interpretation rather than by properly considered legislative action.”Footnote 70 Gearhart put it more forcefully, warning that Congress must not “stand mute and permit great questions like this to be legislated for us by the courts in judicial decisions, that is, supinely submit to judicial usurpation of the legislative prerogative.”Footnote 71 Though Democratic lawmakers and the president himself dismissed the claim as disingenuous, it galvanized support among moderate and conservative members of Congress concerned about judicial overreach and convinced that the New Deal had unduly expanded the power and authority of the executive branch.

Second, Republicans argued that if allowed to stand, the proposed new regulations—like the court rulings that informed them—would upend a long and venerable tradition of applying common law principles (particularly the control test) to determine employment status. This was an explicit defense of master-servant principles and a repudiation of New Deal attempts to establish a new standard governing employment relations. Republican leaders were carefully constructing a narrative about the past that sought to establish that Congress had been clear and consistent in its intentions and to impose the control test as the decisive criterion for eligibility. Congress needed to act now because the proposed regulations published by the Treasury Department would “apply tests other than the usual common law tests for determining an employer-employee relationship.”Footnote 72

Third, they argued that the result would be catastrophic, leading to regulatory chaos, undermining the ability of employers to conduct the nation's business, and harming the economy. “If the proposed regulations become effective,” the House Ways and Means Committee said, “endless confusion will result, existing rulings will be unsettled, and many types of relationships fixed by contract will have to be reversed at a time when full emphasis should be given to an increase of production and distribution.”Footnote 73 The aim, therefore, must be to avert this outcome, by both reestablishing the common law control test as the definitive determination of employee status and restoring the judgment of Congress to its rightful place over that of the courts and bureaucracy.

House Republicans took up the new vendor bill first. After quickly passing both houses, it drew another veto—and an even sharper veto message. “This legislation has far greater significance than appears on the surface,” Truman's statement began. “It proposes to remove the protection of the social security law from persons now entitled to its benefits.” The president condemned the bill for affording employers the discretion to devise employment relationships to serve their own ends: “This bill would make the social security rights of these employees depend almost completely upon the form in which their employers might choose to cast their employment contracts. Employers desiring to avoid the payment of taxes . . . could do so by the establishment of artificial legal arrangements governing their relationships with their employees.” Truman closed by describing the threat to the social security system: “The security and welfare of our nation demand an expansion of social security to cover the groups which are now excluded from the program. Any step in the opposite direction can only serve to undermine the program and destroy the confidence of our people in the permanence of its protection against the hazards of old age, premature death, and unemployment.”Footnote 74 This time, however, the veto was overridden and the law stood.

If the vendor bill was intended as a rebuke and corrective to the courts, then the second Gearhart bill was aimed at the administrative agencies. It was far more sweeping, and it would negate the proposed regulations submitted by the Treasury Department. Asked to explain the difference between the two bills, Gearhart said that while the first measure would affect only newspaper vendors, “the other measure, House Joint Resolution 296, would legislate a definition of master and servant, employer and employee, by whatever name you may choose to describe it, which would apply generally insofar as the social security system is concerned.”Footnote 75 Its authors called it the “Status Quo” Resolution, because it purported to “maintain the status quo” regarding employment taxes and social security benefits by explicitly excluding those who were not employees under common law rules.

Republicans introduced the legislation with an appeal to congressional authority. The House Ways and Means Committee report accompanying the legislation opened by stating: “The issue involved in the proposed regulations is whether the scope of social security coverage should be determined by the Congress or by other branches of Government.” The policy stakes were substantial; nothing less than “the whole matter of social security coverage is pending before the Committee on Ways and Means and the Senate Finance Committee,” according to the committee. And the implications for business in the United States were dire, as the reach of the proposed regulations extended across the economy, well beyond news vendors and “into such diverse fields of normally independent operations as cattle, hay, feed, and grain buying, pulpwood and other logging, marketing of petroleum products . . . delivery, distribution and sale of household and other items and appliances to the ultimate consumer, and also sales of fire, casualty, and some other types of insurance.”Footnote 76

For their part, the administration and its congressional Democratic allies were more prepared and coordinated in their response to this initiative than they had been to the vendor bills. Led by Representative Herman Eberharter (D-PA), they met the Republican arguments with a stronger and clearer stance of their own. They challenged the pretext that the legislation sought to maintain the status quo. The title of the bill, argued four Ways and Means Committee Democrats in a minority report, was “grossly misleading.” The bill in fact proposed a major change. Its revised definition of employees and independent contractors would exclude from certain provisions of the Social Security Act “some 500,00 to 750,000 employees and their dependents who are now entitled to the protection of social security coverage under existing law.”Footnote 77

Reports and testimony by the Treasury Department and the FSA delivered the central elements of the Democratic counterargument. It boiled down to three points. First, the Republican bill would undermine the core purposes of and reverse progress toward a stable, reliable, and sustainable social security system. The system's purpose, FSA Administrator Oscar Ewing reminded Congress, “is to provide to those who look for their livelihood to their earnings from services to others . . . a minimum of protection against the risk of loss of those earnings by reason of temporary unemployment, retirement on account of age, or death.”Footnote 78 By limiting coverage, the legislation would “reverse the direction in which . . . the program should move. It has long been recognized by the President, the Congress, this Agency and other competent authorities in this field that the coverage of the act should be broadened rather than narrowed.”Footnote 79

Second, administration officials argued that it was neither viable nor wise to return to the common law standard that had been dominant prior to Supreme Court decisions in the Hearst, Silk, and related cases. There was no straightforward or consistent way for courts and agencies to apply the common law control test in implementing social security legislation. Part of the problem was that the evolution of the common law standard varied by state. Beyond that, courts interpreted the standard in significantly different ways. Ewing said, “The so called control test, often stressed as the determinative factor under the common law as it has developed, is often all but impossible to apply. Even those courts which tend to treat the ‘control test’ as determinative differ widely in their application of it.”Footnote 80

FSA General Counsel Willcox pointed to another problem—the bureaucratic disagreements that had arisen. Like the courts, administrative agencies had read the laws and court guidance in varying ways, and they had not always agreed on how to apply the regulations developed in 1936 to administer social security. In cases ranging from life insurance agents and home workers to mining lessees and construction workers, “the result has been the payment of benefits in some situations in which no taxes have been collected,” Willcox reported. “The Federal Security Agency has generally tended to hold that the individuals were employees, while the Bureau on Internal Revenue has tended to adopt a more restrictive view.”Footnote 81

Third, and perhaps most importantly, administration officials argued that the proposed regulations were necessary because the common law control test was generating deleterious labor market effects. It created incentives for some employers to manipulate the new laws, and it produced inequities. Acting Treasury Secretary Archibald Wiggins reminded the committee that

in the absence of any other guide, this test [the common law control test] was adopted by the Treasury Department in 1936, in the Department's original regulations under the Social Security Act. As experience developed under these regulations, however, it became increasingly clear that such a test permitted employers to avoid employment tax liability and deprive their workers of social security coverage by dressing up their relationship through so-called independent contracts but without, in any material sense, altering their relative economic positions.Footnote 82

The administrators furnished numerous examples of employer manipulations and evasions. In one case, a drugstore company converted former branch manager employees into licensees, who were then classified by a court as independent contractors “despite the fact that their economic relationship with the drug company remained virtually the same as when they were branch managers.”Footnote 83

Attempts to preserve the common law standard, administrators warned, threatened to generate serious inequities among workers and employers alike. They argued that even if self-employed workers were brought into social security under specific terms (as Congress was separately considering), it would be difficult for them to receive reliable coverage under unemployment insurance if the common law standard held. And if they were required to pay a higher rate of contribution to social security than workers considered employees (for whom employers pay half the social security tax), the policy would be unfair: “Since all of the workers in this area occupy the same economic status as ‘common law’ employees, it would be inequitable to make them pay more than their ‘common law’ counterparts for social security protection, particularly when it is considered that such excess represents a tax burden which should properly be borne by their employers.” Likewise, it would discriminate against employers who treated their workers as regular employees, because they would face higher tax burdens than those employers who, legitimately or not, classified their workers as independent contractors.Footnote 84

The debate played out on the floor of Congress and in hearings in the Senate Finance Committee, with testimony from no fewer than fifteen business representatives and letters, statements, and telegrams from others included in the record. This time, business mobilization on behalf of the Republican bill extended well beyond the news industry. Lawmakers supporting the resolution underscored the extensive communications they had received from businesses in their districts urging passage.Footnote 85 Indeed, M.W. Zucker, of the New York Commerce and Industry Association, noted proudly that the proposed Treasury regulations had been blocked from going into effect as scheduled in part as “a result of protests from employers and business groups throughout the country.”Footnote 86

Business leaders argued that the proposed regulations to clarify and broaden the scope of workers’ protections reflected undue intervention in the prerogatives of employers. Zucker testified that the New York Commerce and Industry Association “vigorously opposes” the regulations, saying that they “will increase latitude for administrative discretion and will, therefore, increase the uncertainty of business personnel.”Footnote 87 Robert Canfield of the American Pulpwood Association expressed concern that more employer obligations would follow: “If a man is an employee for social security purposes it will not be long before the claim is made that he is an employee for purposes of workmen's compensation, for purposes of tax withholding, for purposes of wage-hours controls, for tort liability.”Footnote 88

Far fewer labor representatives were included in the roster of speakers in the hearings, but those present made equally impassioned arguments in defense of the regulations.Footnote 89 Unions had been active in pressing administrative agencies and the courts to ensure that workers were protected as employees rather than considered independent contractors.Footnote 90 Now Nelson Cruikshank, Director of Social Insurance Activities of the American Federation of Labor, argued before the committee that the resolution would “adversely affect the security of thousands of men and women in the United States who, though they may not be according to some legal definition technically in the category of employee, nevertheless depend upon the returns from their labor for their daily livelihood.” He added that “it would be disastrous for Congress to take a backward step by adopting a legalistic and narrow interpretation of the employer-employee relationship.”Footnote 91

John Stanley, testifying on behalf of the United Office and Professional Workers of America, CIO, was more pointed: “What proponents of the exclusion are actually worrying about is that they have to begin now making their share of the contributions required by the act covering these workers. This they would like to avoid.”Footnote 92 Speaking directly to the testimony of Zucker, Stanley said, “The statement . . . boils down to an objection that, under the present law, some administrator will have to make an interpretation of whether or not an employee is covered. What this Association of employers wants is that the employer should make that determination. We can be sure what that will be and where that will leave the workers. That is right out in the cold.”Footnote 93

Despite the heated debate, the measure passed both houses. The president returned the legislation with another stinging veto message, challenging the pretext of the Republican claims:

It has been represented that the issue involved in this resolution is whether or not the legislative branch of government shall determine what individuals are entitled to social security protection. This is not the issue at all. The real issue is whether the social security coverage of many hundreds of thousands of individuals should be left largely to the discretion of their employers.

The president criticized “the sponsors of the resolution who would have us believe, for example, that a traveling salesman who devotes full working time in the service of one company and depends completely upon that company for his livelihood is not an employee of that company but is an independent businessman and does not need social security protection.”Footnote 94 In the final months of the Republican-controlled Congress, the veto was overridden, and the Status Quo Resolution became law.

4. Aftermath: A Political and Institutional Legacy

Democrats regained control of Congress in the November 1948 elections. By then, the damage had been done. The proposed Treasury Department regulations never went into effect, and the effort to redefine employment status for purposes of New Deal social and labor legislation collapsed.

Truman had not given up without a fight. He launched his 1948 election bid in Detroit with a Labor Day speech calling for the repeal of Taft-Hartley, and he traveled to Rep. Gearhart's home district to campaign against him (Gearhart lost).Footnote 95 Organized labor played a central role in Truman's upset victory over Republican Thomas Dewey, and he later recounted that on “arriving at the White House, I had a Cabinet meeting and a series of conferences to plan immediate repeal of the Taft-Hartley Act, as promised in the campaign.”Footnote 96 But the new Congress had majorities in both the Senate and House that had voted for Taft-Hartley, and public sentiment for unions had soured in the wake of the postwar strike wave. By early 1949, administration lobbyists determined that a majority of thirty-eight Senate Republicans and fourteen Southern Democrats were likely to vote to retain its major provisions. Divisions among Democrats and union opposition thwarted compromise proposals, and the prospects for a quick repeal or meaningful revision evaporated.Footnote 97 Calls for Taft-Hartley's repeal would surface repeatedly in Democratic platforms in the 1950s and 1960s, but with diminishing force and focus.Footnote 98

When major new amendments to the Social Security Act were enacted the following year, the Democratic Congress extended coverage under the social security retirement program to nearly 10 million additional people (including regularly employed agricultural and domestic workers). Included in this group were 4.7 million nonfarm self-employed workers.Footnote 99 But in crafting the 1950 amendments, Congress chose not to repeal the Status Quo Resolution and retained its terminology in defining the term “employee,” despite the administration's support for a more inclusive definition. Already bruised by previous defeats and defections, liberal lawmakers faced escalating pressure from employers on the issue. After the highly visible political conflicts of 1947–48, a broader and more vocal constituency of employers had mobilized. Business leaders were concerned that a more expansive definition “might be used eventually in other legislation, such as in labor relations, wage-and-hour, workmen's compensation, and tort liability,” according to Wilbur Cohen, and these concerns “brought many industries and lawyers into the controversy.” In his estimation, “more time and money were spent on this issue than on any other issue in the bill,” with the possible exception of disability insurance.Footnote 100

In the end, self-employed workers were incorporated on distinct and inequitable terms. Because they were considered their own “employers,” these workers were required to pay both the employer and the employee share of the social security tax. The social security system was thus expanded, but in ways constrained by the previous Republican majority in Congress. Extensions of full coverage were made on an occupation-by-occupation basis, rather than on more universal terms (such as an expanded definition of employee). And as Truman officials had warned in the 1948 debate over the Status Quo Resolution, self-employed workers were disadvantaged by the new rules.Footnote 101

The congressional Republicans’ counteroffensive of 1947–48 ultimately had consequences that reached far beyond the Truman years. It left an enduring institutional legacy—administrative, judicial, and legislative. Administratively, the failure to finalize and implement the revised Treasury regulations was more than a missed opportunity to enact more inclusive standards. Republicans had also introduced new and more limited legislative criteria for determining coverage. The impact on administrative decisions was almost immediate. By February 1948, Congress's rebuke of the NLRB and the Supreme Court's Hearst ruling was cited in NLRB rulings that denied workers the right to organize and bargain collectively, as in the case of newspaper carriers in Kansas City.Footnote 102 Additional NLRB rulings that spring and summer made clear that the board's reading was that “Congress intended to give these terms [employee and independent contractor] their conventional meanings and that the Board, in determining coverage under the Act, should follow the ‘ordinary tests of the law of agency’ . . . the familiar ‘right of control test.’”Footnote 103

The judicial legacy ran deeper. Although courts continued to issue varying opinions on the issue, the Taft-Hartley provision and the Status Quo Resolution were cited repeatedly as evidence of congressional intent on the question of coverage under the New Deal laws and as grounds for rejecting the Supreme Court's Hearst, Silk, and other more expansive interpretations. This emerged clearly in some of the most prominent Supreme Court cases addressing the distinction between employees and independent contractors, even decades later. These included the 1968 NLRB v. United Insurance Co. case (390 U.S. 254), which cited the Taft-Hartley amendment as a statement of congressional intent regarding determinations of employee status under the Wagner Act, and the 1989 Community for Creative Non-Violence v. Reid case (490 U.S. 730). The Court set out a sweeping standard in Reid, asserting that “when Congress has used the term ‘employee’ without defining it,” as it has done repeatedly in protective legislation enacted since the 1930s, the Court has determined that “Congress intended to describe the conventional master-servant relationship as understood by common-law agency doctrine.”Footnote 104 This line of reasoning was extended in the 1992 Nationwide Mut. Ins. Co. v. Darden case (503 U.S. 318). Justice David Souter cited both the Taft-Hartley Act and the Status Quo Resolution in Darden in concluding—contrary to the Hearst and Silk rulings—that common law principles, not economic realities, were the basis for determining employee status, and that the corrective purposes of legislation were not a determining factor.Footnote 105

There was a legislative legacy as well, one that extended beyond the two laws (the Social Security Act and the Wagner Act) amended in 1947 and 1948.Footnote 106 The actions of the 80th Congress also reverberated through subsequent legislation, as new laws imported existing restrictions and imprecise definitions of employee. From the 1960s through the early 1990s, Congress passed an array of new rights and protections for workers. Some of these introduced new antidiscrimination protections, such as Title VII of the 1964 Civil Rights Act, which prohibited employers from discriminating in hiring or classifying workers on the basis of race, color, gender, national origin, or religion. Title VII was followed by the Age Discrimination in Employment Act of 1967 and the Americans with Disabilities Act of 1990. Other laws, such as the 1970 Occupational Safety and Health Act, created new workplace protections, requiring employers to maintain healthy and safe working conditions. Still others expanded benefits, through the Employment Retirement Income Security Act of 1974, which established standards for employer-provided pension and other benefit programs, and the Family Medical Leave Act of 1993, which provided certain protections for employees requiring time off work due to medical issues or the birth or adoption of a child. The new laws reflected changing standards about acceptable conditions and conduct in the workplace, yet none of these protections applied to workers who were not employees, including independent contractors. And all defined the employees eligible for coverage in terms similar to—and just as vague and limiting as—the language used in earlier New Deal laws.Footnote 107

One question that arises is why Democratic leaders did not seize opportunities to legislate a more expansive definition of employee in the years after 1947–48, especially when they held the presidency, majorities in Congress, or both. Why, in short, did the narrow concept of employee inscribed by Republicans but opposed by New Deal Democrats prove so resilient? This brief history captures several self-reinforcing processes that followed in the wake of the Republican counteroffensive. As mechanisms of path dependence, each made a return to the expansionary reform path charted by New Deal Democrats more challenging and less likely, and further entrenched the Republican initiatives.Footnote 108

Perhaps most importantly, the developments of 1947–48 created new incentives and vested interests among those most directly affected, particularly employers. Facing increased costs and regulatory burdens for their employees under New Deal social legislation, employers recognized that there were major benefits to maintaining a vague and limited definition of employee in all such legislation. Successive congressional hearings organized by Republicans afforded employers repeated opportunities to broaden and mobilize their alliances and to sharpen their arguments and strategies. The news vendor bill drew together a coalition of newspaper publishers stretching from “the Canadian to the Mexican border” in 1947, for example, and by the time the 1950 amendments were debated, a wide range of “industries and lawyers” had come to recognize their own interests in the matter and were actively lobbying against a more inclusive definition of employee, spending record levels of time and money to pressure lawmakers.Footnote 109

This is, in part, a familiar story of how (business) interests exert pressure on the state to shape policy. But it is also an example of how the state itself—through policy—constructs new (or newly vested) interests. By producing a raft of new entitlements and protections for employees, New Deal laws forced administrative agencies and courts to fashion a modern concept of “employee.” This in turn created an interest in—and struggles over—that concept, for groups on both sides of the employment relationship.Footnote 110 Employers, particularly those outside of the large-scale, capital-intensive industrial sectors, developed a new vested interest in the 1940s in mobilizing against broad definitions of the term. Workers had an interest in mobilizing for such broad definitions and for their own inclusion in the category—but this type of mobilization was more difficult for the often-dispersed, low-paid, and non-unionized workers who would most benefit from inclusion.Footnote 111 Labor unions sought to broaden the definition of employee (and thereby the number of workers who were covered by social protections and who could unionize), but their focus was on combatting the existential threats they saw in Taft-Hartley's core provisions, and they were not as formidable a force on the question of employment status.

In addition to defining and expanding organized interests, the events of 1947–48 shaped the calculus of Democratic leaders in ways that led them to progressively narrow their sights—and the prospects for expansionary reform. Republican victories in the highly visible struggles of these years signaled the relative strength of the conservative coalition, and as Democrats drew back, the political space for reform shrank.Footnote 112 Democrats’ response to diminished expectations and their own declining influence included individual defections from the Truman administration's position (reflected in Democratic vote counts on overriding Taft-Hartley in 1947 and replacing it in 1949) and strategic retreats in the party's legislative aspirations (reflected in the declining prominence of the issue in party platforms in the 1950s). President Johnson extended the Democratic retreat into the 1960s, even after his party reclaimed the White House and seated one of the most liberal Congresses in decades. Johnson had promised early action on Taft-Hartley repeal but moved the issue off the agenda, anticipating the reaction of opponents and concluding that the political costs would be too high.Footnote 113 Moving swiftly on repeal, he told labor leaders, would galvanize the conservative coalition, threatening his ambitious policy agenda and splintering his own coalition.Footnote 114 Each strategic calculation and accommodation by Democrats may have been politically rational, but each made subsequent accommodations more likely and a resumption of the broader reform agenda more difficult.

If Democratic lawmakers found the Republicans' legislative initiatives increasingly difficult to displace, administrators and courts found them difficult to disregard, particularly as the record of “congressional intent” grew. The institutional weight of the legislative record was a further factor contributing to the durability of the Republican definition of employment status. Congressional Republicans seized opportunities in 1947 and 1948 to deliberately and systematically build a record of “legislative intent.” Their aim was not only to impose new legislative restrictions but also to construct a narrative establishing the intent of Congress (including previous ones) to maintain limited and vague definitions of employee and to clearly ground interpretations of New Deal legislation in common law conceptions. The Republican narrative included explicit “corrections” of NLRB and Supreme Court rulings that reflected more liberal interpretations.

As administrators and the courts seeking congressional intent on the issue drew on the reconstructed legislative record, the Republican conception became more deeply ingrained. Moreover, when Democratic lawmakers crafted new legislation beginning in the 1960s, the legislative record again exerted institutional influence, as Democrats reached to the record to import old definitions into new worker protections. Drawing on existing legislative language was common practice in writing new laws, particularly in the absence of a pointed effort to challenge an existing definition, and by then, the Democrats were not so inclined. The role of the legislative record in drafting what would become Title VII of the Civil Rights Act of 1964 was a case in point. Language was a major point of contention in the congressional debate, but the focus was on language defining discrimination and the range of groups to be protected, among other things. The House Committee took care to spell out the definitions of terms they deemed “particularly important to an understanding of the scope of the act,” but stated that other terms—and they explicitly included “employee” here—“are defined for the purposes of this act in the manner common for Federal statutes.”Footnote 115

Together, the impact of these processes—new incentives, interests, and constituencies among employers; diminishing expectations and a calculated retreat by reformers; and the institutional effects of a growing record of legislative intent on administrators, the judiciary, and legislators themselves—helped entrench the Republican conception of employee. As they unfolded, positive feedback effects made shifting from the path more costly over time. The Republican victories and Democratic retreat thus closed a window for reform in the 1940s and set a trajectory that shaped subsequent policy. They did not settle the issue of employment status, however.

5. New Century, Old Struggles

The actions of the 80th Congress not only left a political and institutional legacy. They also shaped developments in a labor market that was growing increasingly insecure and precarious for many workers, particularly as companies responded to the effects of rising global competition and technological change beginning in the late 1970s. Incentives to hire independent contractors and other contingent workers, rather than regular employees, escalated in this shifting economic environment, and the number of cases in which employers “misclassified” workers as independent contractors grew significantly.Footnote 116

Although the issue received wider media attention with the rise of the gig economy in the 2000s, the trend began much earlier and its effects have been felt far beyond the app-based gig sector. Federal officials had noted this practice as early as the 1930s and 1940s and warned that it might become more widespread. As the use and misuse of independent contractor arrangements shot up in the changing labor market, federal officials in the early 1990s again drew attention to employers’ practices and to the problems created by existing law. The use of “independent contractors and part-time, temporary, seasonal, and leased workers has expanded tremendously in recent years,” noted the Commission on the Future of Worker-Management Relations, appointed by President Clinton's Secretaries of Labor and Commerce in 1993.Footnote 117 Restating concerns expressed by the Truman administration fifty years earlier, the commission warned that “current tax, labor and employment law gives employers and employees incentives to create contingent relationships not for the sake of flexibility or efficiency but in order to evade their legal obligations,” depriving workers of protections under labor and employment laws, and of the employer contribution to social security.Footnote 118 In a reprise of efforts to institutionalize a modern liberal conception of employment relations, the commission's recommendations echoed regulatory proposals and court rulings of the 1940s:

The definition of employee in labor, employment, and tax law should be standardized. Instead of the control test borrowed from the old common law of master and servant, the definition should be based on the economic realities underlying the relationship between the worker and the party benefitting from the worker's services.Footnote 119

Weeks before the commission's report and recommendations were published, however, the 1994 elections swept into office a highly conservative, Republican-controlled Congress—the first in over forty years—and no actions were taken on the recommendations. But attention was growing.

By 2000, a nine-state audit conducted for the U.S. Department of Labor found that between 10 and 30 percent of employers misclassified a portion of their workforce, affecting millions of workers.Footnote 120 A subsequent survey by the U.S. Government Accountability Office in 2009 found a 42 percent increase in the number of workers misclassified between 2000 and 2007.Footnote 121 More recent studies indicate that misclassification is particularly prevalent in industries ranging from construction, trucking, and retail, to hospitality, home care, and janitorial services.Footnote 122 There is also clear and consistent evidence that misclassification disproportionately affects lower-wage workers, women, and workers of color.Footnote 123

These trends have generated a new wave of political struggles over employment status, involving employers reliant on independent contractors, labor unions and advocacy organizations, and state and federal officials.Footnote 124 The recent round of conflicts demonstrates both the enduring relevance of and continued contestation over New Deal social and labor legislation. Entitlements created then continue to generate disputes—among administrators, courts, and legislatures—over the boundaries of employment status. The conflicts this time have been fueled in part by the impact of misclassification on state revenues, particularly in the wake of the 2001 and 2007–09 recessions. The downturns of the early 2000s simultaneously brought steep declines in state revenues and increased demands on state coffers to meet the rising need for unemployment benefits and other safety net supports. As researchers produced some of the first studies quantifying the scale of tax revenues lost to the states from employers who misclassify their workers (an estimated $175 million a year in lost unemployment funds in New York alone in the early 2000s), state legislatures began strengthening laws to prevent the practice. Between 2004 and 2012, twenty-two states took action to increase the penalties for employers who misclassified workers and/or to tighten the definition of who could be classified as an independent contractor.Footnote 125

Some states went further. In 2018, a California Supreme Court ruling (in Dynamex Operations West, Inc. v. Superior Court) required the state to use a standard known as the “ABC test” to determine employment status for purposes of state wage and hour laws. Among the competing legal tests (such as the control test and the economic realities test), the ABC test is one of the most expansive. It presumes that all workers for an employer are employees and only allows a business to classify a worker as an independent contractor if it can show that the worker meets a set of three demanding standards (labeled A, B, and C) demonstrating independence.Footnote 126 Despite fierce opposition from employers, the ABC standard was codified by the California state legislature in 2019 in Assembly Bill 5. The law went into effect in January 2020 and was estimated to affect nearly one million workers.Footnote 127 California joins four other states (Massachusetts, Nebraska, New Jersey, and Vermont) that have adopted this strong standard for purposes of wage and hour and related employment laws.Footnote 128 State and local officials also turned to the courts to compel changes in companies' practices: California, New Jersey, and Massachusetts, for example, filed suit against Uber and/or Lyft to recover the unemployment and disability taxes the companies avoided paying in recent years by misclassifying their workers and to compel them to comply with the states’ ABC laws.Footnote 129

Leading gig companies fought back forcefully against these state-level actions. As in the 1940s, business leaders immediately recognized that they had much to lose if forced to comply with expansive interpretations of employment status under New Deal laws and related state and local protections. Not only would they confront increased labor regulations, but they would also face an estimated 30 percent more in labor costs if their workers were designated employees.Footnote 130 The companies quickly staffed up to mount aggressive lobbying efforts to protect their prerogative to classify their workers as independent contractors and to avoid other legal and regulatory obligations. By 2016, Uber had engaged some 270 lobbyists in forty-four states.Footnote 131

The contemporary stuggle over employment status escalated in 2019 after the passage of AB5 in California. Having failed to stop the measure in the legislature and courts, Uber, Lyft, DoorDash, Instacart, and other companies responded by mounting a ballot measure campaign in California. They invested $220 million in the effort—the most expensive ballot campaign in the state's history—and won passage of Proposition 22 in November 2020. The measure imposed a “carveout” for their drivers, not unlike the carveout from payroll taxes that Republicans engineered for news vendors in the 1940s. It exempted “rideshare and delivery network companies” from the provisions of AB5 and other state labor laws, and set up a framework—designed by the companies—of limited protections for their workers.Footnote 132 In August 2021, a California state Superior Court judge ruled that Proposition 22 was unconstitutional, a decision the companies set out to reverse in both state and federal courts.Footnote 133 Even as Proposition 22 confronted challenges in the courts, gig company leaders wasted no time in launching similar campaigns in other states, including a ballot initiative in Massachusetts. The initiative was certified for inclusion on the 2022 ballot by the attorney general in September 2021, and an intense conflict over the measure unfolded between labor- and industry-backed coalitions before it was blocked by the Massachusetts Supreme Court in June 2022.Footnote 134

The arguments of the gig companies in these state-level struggles have been strikingly similar to those of earlier business leaders. Once again, employers warn that a more expansive definition of employee and limitations on their use of independent contractors would, according to court filings by the U.S. and California Chamber of Commerce, “lead to reduced workforce flexibility, slower economic growth, and higher unemployment.”Footnote 135 Uber, Lyft, Postmates, TaskRabbit, and DoorDash argued that they would be “hamstrung” by adoption of the ABC test in California; indeed, it would “decimate business” and have the impact of “stifling innovation and threatening the livelihoods of millions of working Californians.” The head of the California Chamber of Commerce, Allan Zaremberg, drew similarly stark conclusions, in language reminiscent of the debates of the 1940s. The proposed standard, he said, could have dire consequences. “People depend very much now on an on-demand economy,” he pointed out. But in “the worst case scenario, it isn't a viable business model anymore.”Footnote 136

Though less visible, disputes over the definition of employee are underway at the federal level as well. Since 2017, Republican lawmakers have introduced legislation in each Congress that explicitly seeks to complete the work of the 80th Congress by applying the narrower definitions of employee enacted in Taft-Hartley and the Status Quo Resolution to the Fair Labor Standards Act. In a rejoinder to Democratic attempts in the 1940s (and 1990s) to standardize definitions along modern liberal lines, the proposed legislation seeks standardization with common law principles, a goal the bill's drafters believe was achieved for the Wagner Act (through Taft-Hartley) and Social Security Act (through the Status Quo Resolution), but not the FLSA. The stated purpose of the “Modern Worker Employment Act” (S. 536 and H.R. 1523, 117th Congress) is thus to “amend the Fair Labor Standards Act of 1938 to harmonize the definition of employee with the common law.”Footnote 137

In President Trump's final weeks in office, his administration sought the same aim through administrative means. In January 2021, the Trump Labor Department published a new FLSA interpretation rule, citing the Taft-Hartley definition and Status Quo Resolution.Footnote 138 The incoming Biden administration promptly withdrew the Trump rule in May 2021, determining that its narrower definition of employee status was inconsistent with the purpose of the FLSA and would result in workers losing protections.Footnote 139 Employers soon joined the Republican effort, fighting back in the courts. Biden's move was challenged in a case brought by Uber and other industry groups, and in March 2022, a federal district court in Texas ordered that the Trump rule be reinstated on procedural grounds.Footnote 140 All three branches of government were again engaged in a contest over how to define employee under New Deal laws.

The conflict over employment status, in short, contnues. In the latest round, it is not Hearst but Uber and Lyft that are leading the fight to defend a business model that rests on their prerogative to hire and classify workers as they choose. It is not news vendors and door-to-door salespeople but workers ranging from Uber drivers and DoorDash delivery workers to home health aides and construction workers whose rights and protections are at stake. And the contest is playing out not only (or primarily) in Washington, DC, but at the state level. Yet the arguments made by business leaders, and the conditions and vulnerabilities faced by workers they classify as independent contractors, are remarkably similar to those of the 1940s, and the institutional context was forged by the conflicts of 1947–48. Now as then, the struggle is over how and where to draw the lines determining the scope of workers’ social protections and the limits of employer power in the labor market. As the numbers of those working in the gray area between standard employment and independent contracting have soared, so too have the stakes for both sides.

Competing Interests

The author declares none.

Acknowledgments

I received valuable comments on an earlier version of this article in a 2021 American Political Science Association panel on “Legacies of Work: Social Policy, Employment, and Regimes of Inequality.” I would like to thank the participants and chair of that session, as well as anonymous reviewers of this manuscript, for their comments and suggestions.

References

1 See, for example, Noam Scheiber, “Uber and Lyft Ramp Up Efforts to Shield Their Business Model,” New York Times, June 10, 2021; Rebecca Rainey, “Uber, Others See Lobbying Chance in DOL Lull on Gig-Worker Rule,” Bloomberg Law, Daily Labor Report, April 26, 2022; Nandita Bose, “U.S. Labor Secretary Supports Classifying Gig Workers as Employees,” Reuters, April 29, 2021; “Trump-Era Rule That Made it Harder for Gig and Contract Workers to Get Minimum Wage Is Withdrawn,” Washington Post, May 5, 2021; “Worker Classification: Employment Status Under the National Labor Relations Act, the Fair Labor Standards Act, and the ABC Test” (Congressional Research Service, Washington, DC, April 20, 2021), 14–27.

2 State-level conflicts unfolded in California over a 2018 State Supreme Court ruling (Dynamex), 2019 legislation to codify that ruling (AB5), and a 2020 industry-backed ballot initiative (Proposition 22) to exempt app-based drivers for rideshare and delivery companies from being classified as “employees” under that law. A similar, and also highly contentious, initiative slated for the 2022 ballot in Massachusetts was blocked by the Massachusetts Supreme Judicial Court.

3 For an earlier discussion of some of the themes developed here, see Bertram, Eva, “The Political Development of Contingent Work in the United States: Independent Contractors from the Coal Mines to the Gig Economy,” Journal of International and Comparative Labour Studies 5, no. 3 (September–October 2016): 2–34Google Scholar.

4 Much of the scholarly discussion and debate on employment status takes place in law review articles. Some of this literature devotes attention to the broader political context examined here. See, for example, Dubal, Veena B., “Wage Slave or Entrepreneur?: Contesting the Dualism of Legal Worker Identities,” California Law Review 105 (2017): 65123Google Scholar; Dubal, Veena B., “The Drive to Precarity: A Political History of Work, Regulation, and Labor Advocacy in San Francisco's Taxi and Uber Economies,” Berkeley Journal of Employment and Labor Law 38, no. 1 (2017): 73–136Google Scholar; Zatz, NoahBeyond Misclassification: Tackling the Independent Contractor Problem Without Redefining Employment,” ABA Journal of Labor and Employment Law 26, no. 2 (Winter 2011): 279–94Google Scholar; Linder, Marc, The Employment Relationship in Anglo-American Law: A Historical Perspective (New York: Greenwood Press, 1989)Google Scholar; Linder, Marc, “What Is an Employee? Why It Does, But Should Not Matter,” Law and Inequality 7 (1989): 155–87Google Scholar; Linder, Marc, “Dependent and Independent Contractors in Recent U.S. Labor Law: An Ambiguous Dichotomy Rooted in Simulated Statutory Purposelessness,” Comparative Labor and Law Policy Journal 187 (1999):187–230Google Scholar; Carlson, Richard R., “Why the Law Can't Tell an Employee When It Sees One and How It Ought to Stop Trying,” Berkeley Journal of Employment and Labor Law 22, no. 2 (2001): 295–368Google Scholar; Befort, Stephen F., “The Regulatory Void of Contingent Work,” Employment Rights and Policy Journal 10, no. 1 (2006): 245–57Google Scholar; Jost, Micah Prieb Stoltzfus, “Independent Contractors, Employees, and Entrepreneurialism Under the National Labor Relations Act: A Worker-by-Worker Approach,” Washington and Lee Law Review 68 (2011): 311–52Google Scholar.

5 Orren, Karen, Belated Feudalism: Labor, the Law, and Liberal Development in the United States (New York: Cambridge University Press, 1991), 19Google Scholar.

6 Atleson, James B., Values and Assumptions in American Labor Law (Amherst: University of Massachusetts Press, 1983), 178Google Scholar. Atleson's focus is not on the definition of employee, but on the enduring presence of master-servant conceptions in labor law. He concludes that “just as the nineteenth-century notion of contract was infused with older master-servant doctrines, a similar conclusion can be reached about modern American labor law” (180).

7 On the politics leading to and the consequences of occupational exclusions in New Deal legislation, see, for example, Katznelson, Ira, When Affirmative Action Was White: An Untold History of Racial Inequality in Twentieth-Century America (New York: W.W. Norton, 2005)Google Scholar; Katznelson, Ira, Geiger, Kim, and Kryder, Daniel, “Limiting Liberalism: The Southern Veto in Congress, 1933–1950,” Political Science Quarterly 108 (1993)CrossRefGoogle Scholar; Lieberman, Robert, Shifting the Color Line: Race and the American Welfare State (Cambridge, MA: Harvard University Press, 1998)Google Scholar; Quadagno, Jill, The Color of Welfare: How Racism Undermined the War on Poverty (New York: Oxford University Press, 1994)Google Scholar; Mink, Gwendolyn, The Wages of Motherhood: Inequality in the Welfare State, 1917–1942 (Ithaca, NY: Cornell University Press, 1995)Google Scholar; Brown, Michael K., Race, Money, and the American Welfare State (Ithaca, NY: Cornell University Press, 1999)CrossRefGoogle Scholar; Bertram, Eva, The Workfare State: Public Assistance Politics from the New Deal to the New Democrats (Philadelphia: University of Pennsylvania Press, 2015)CrossRefGoogle Scholar.

8 See, for example, Alexander, Charlotte S., “Misclassification and Antidiscrimination: An Empirical Analysis,” Minnesota Law Review 101, no. 3 (February 2017): 907–67Google Scholar, which presents evidence that “women and/or people of color are overrepresented in seven of the eight occupations at highest risk for misclassification” as independent contractors. See also Chris Benner, with Erin Johansson, Kung Feng, and Hays Witt, On-Demand and On-the-Edge: Ride-hailing and Delivery Workers in San Francisco, Institute for Social Transformation, University of California, Santa Cruz, May 5, 2020, https://transform.ucsc.edu/on-demand-and-on-the-edge/

9 See, for example, Theda Skocpol, “The Limits of the New Deal System and the Roots of Contemporary Welfare Dilemmas,” and “Brother, Can You Spare a Job? Work and Welfare in the United States,” in Skocpol, Theda, Social Policy in the United States: Future Possibilities in Historical Perspective (Princeton, NJ: Princeton University Press, 1995)Google Scholar. For an analysis of the constraints produced by delinking economic policy (including jobs policy) and social policy in the New Deal, see Weir, Margaret, Politics and Jobs: The Boundaries of Employment Policy in the United States (Princeton, NJ: Princeton University Press, 1992)Google Scholar. On subsequent attempts to enact full employment policies, see Mucciaroni, Gary, The Political Failure of Employment Policy, 1945–1982 (Pittsburgh, PA: University of Pittsburgh Press, 1990)Google Scholar.

10 See Bertram, Eva, “Doors, Floors, Ladders, and Nets: Social Provision in the New American Labor Market,” Politics and Society 41, no. 1 (2013): 29–72CrossRefGoogle Scholar.

11 Social Security officials repeatedly made this case beginning in the program's early years. For their argument about why extending coverage was necessary to “protect the financial soundness of the system,” see, for example, “Proposed Changes in the Social Security Act: A Report of the Social Security Board to the President and to the Congress of the United States,” Social Security Bulletin 2, no. 1 (January 1939): 4–19.

12 I am grateful to an anonymous reviewer for highlighting this point.

13 The study provides what Daniel Galvin might characterize as a “causes-of-effects, within-case analysis” of a significant shift in public policy to “uncover the origins of contemporary political” debates over the issue. Galvin, Daniel J., “Qualitative Methods and American Political Development,” in The Oxford Handbook of American Political Development, ed. M., Richard Valelly, Suzanne Mettler, and Robert C. Lieberman (New York: Oxford University Press, 2016), 209Google Scholar.

14 Karen Orren's Belated Feudalism addressed the conflicts created by the simultaneous operation of these old and new orders. For a broader discussion of the “tensions routinely introduced by the simultaneous operation, or intercurrence, of different political orders,” see Orren, Karen and Skowronek, Stephen, The Search for American Political Development (New York: Cambridge University Press, 2004), particularly 13–32CrossRefGoogle Scholar and 108–19.

15 On the distinctions between “macro,” “meso,” and “micro” levels of institutional analysis, see Sheingate, Adam, “Institutional Dynamics and American Political Development,” Annual Review of Political Science 17 (2014): 471–75CrossRefGoogle Scholar.

16 For the original text of the Social Security Act, see United States Social Security Administration, Social Security History, “1935 Social Security Act,” accessed November 21, 2022, https://www.ssa.gov/history/35actinx.html. For the National Labor Relations Act, see National Labor Relations Board, “Key Reference Materials,” accessed November 21, 2022, https://www.nlrb.gov/guidance/key-reference-materials. For the Fair Labor Standards Act, see FRASER (digital database), “Full Text of Fair Labor Standards Act of 1938,” accessed November 21, 2022, https://fraser.stlouisfed.org/title/fair-labor-standards-act-1938-5567/fulltext.

17 In addition to this definition of “employee,” the Fair Labor Standards Act defines “employ” as “to suffer and permit to work.” This formulation had been developed in pre–New Deal social protection legislation and is considered the broadest definition under the New Deal statutes. The Social Security Act, under several titles, defined “employment” as any service “performed . . . by an employee for his employer,” except for services excluded under the act, which included agricultural labor and domestic service. The National Labor Relations Act defined employee “to include any employee.”

18 See, for example, Delton, Jennifer A., The Manufacturers: How the National Association of Manufacturers Shaped American Capitalism (Princeton, NJ: Princeton University Press, 2000)Google Scholar; Gordon, Colin, New Deals: Business, Labor, and Politics in America, 1920–1935 (New York: Cambridge University Press, 1994)Google Scholar; Levine, Rhonda F., Class Struggle and the New Deal: Industrial Capital, Industrial Labor, and the State (Lawrence: University of Kansas Press, 1988)Google Scholar.

19 Congressional Republicans did not restrict the definition of “employee” for purposes of the Fair Labor Standards Act in the years examined here, but did narrow the definition of what counts as “work time” for purposes of the FLSA in the Portal-to-Portal Act of 1947.

20 For the broader definition of “employment,” see H.R. 4120, introduced on behalf of the administration by Rep. Doughton on January 17, 1935, 74th Congress, 1st Sess., 44, accessed November 21, 2022, https://www.ssa.gov/history/pdf/fdrbill.pdf.

21 Beginning in 1939, the Social Security Board was housed in the Federal Security Agency.

22 Cohen, Wilbur J. and Calhoon, James L., “Social Security Legislation, January–June 1948: Legislative History and Background,” Social Security Bulletin 11, no. 7 (July 1948): 6Google Scholar. The total civilian workforce in 1940 was 55.6 million; the nonagricultural civilian workforce was 37.9 million. “Technical Note, Labor Force, Employment and Unemployment, 1929–39: Estimating Methods,” U.S. Bureau of Labor Statistics, Monthly Labor Review, July 1948: 51. Estimating the number of independent contractors in this time period is difficult for several reasons. For a discussion and early estimates of exclusions from the NLRA that were legislated in the Taft-Hartley Act, see Rosenthal, Robert J., “Exclusions of Employees under the Taft-Hartley Act,” Industrial and Labor Relations Review 4, no. 4 (July 1951): 556–70CrossRefGoogle Scholar.

23 See Carlson, “Why the Law Can't Tell,” 301–303; see also Linder, The Employment Relationship in Anglo-American Law, part II.

24 See Befort, “The Regulatory Void of Contingent Work,” 249.

25 “Social Security Act Amendments of 1939,” House of Representatives, 76th Congress, 1st Sess., Report No. 728, 8.

26 Robert Hogg, representing the Association of Life Insurance Presidents, Hearings Relative to the Social Security Act Amendments of 1939, Committee on Ways and Means, House of Representatives, 67th Congress, 1st Sess., February and March 1939, 2, 1563. Some employers, including a representative of the Industrial Insurance Conference, requested targeted exemptions from paying unemployment taxes for their workers.

27 Samuel Ansell, representing the American Federation of Musicians, argued that the “failure of Congress” to provide specific definitions of employer and employee had “afforded room for the Bureau of Internal Revenue . . . to construe these terms as having the same technical meaning as ‘master and servant’ in the common law field of torts,” and that a more expansive definition was in order, as “the purposes of these two acts [National Labor Relations Act and Social Security Act] lie in the field of Social Legislation” (Hearings Relative to the Social Security Act Amendments of 1939, Committee on Ways and Means, House of Representatives, 67th Congress, 1st Sess., March and April 1939, 3, 1820–22).

28 The committee said, “It is the intention of this amendment to set up specific standards so that individuals performing services as salesmen may be uniformly covered without the necessity of applying any of the usual tests as to the relationship of employer and employee” (“Social Security Act Amendments of 1939,” 1, 8).

29 The Senate Finance Committee report stated, “It is believed inexpedient to change the existing law which limits coverage to employees.” Cited in House Report 1319, House Ways and Means Committee, 80th Congress, 2nd Sess. [hereinafter H. Rept. 1319], February 3, 1948, 10.

30 The relevant Texas case was Texas Co. v. Higgins 188 F.2d 636.

31 Cohen and Calhoon, “Social Security Legislation, January–June 1948,” 6.

32 Ibid., 6.

33 The debates over employee status have thus shaped access to a wide range of worker protections not just for independent contractors but also for other nonemployees; they are relevant, for example, to contemporary struggles over the unionization of workers ranging from home healthcare workers and port truck drivers to graduate students and student athletes. On the latter, see, for example, National Labor Relations Board, Federal Register 86, no. 48 (March 15, 2021), 14297; National Labor Relations Board, Office of the General Counsel, Memorandum GC 21-08, September 29, 2021.

34 NLRB v. Hearst Publications, 322 U.S. 111 (1944), 120.

35 Ibid., 116.

36 Ibid., 129.

37 For a comparison of the two tests, see Muhl, Charles J., “What Is an Employee—The Answer Depends on the Federal Law,” Bureau of Labor Statistics, Monthly Labor Review 125 (2002): 311Google Scholar.

38 Some lower courts continued to rule based on more restrictive precedents, such as those set by the 1941 Texas Co. v. Higgins case. Cohen and Calhoon, “Social Security Legislation, January–June 1948,” 6.

39 Thomas F. Broden Jr., “General Rules Determining the Employment Relationship under Social Security Laws: After Twenty Years an Unresolved Problem,” Temple Law Quarterly 33, no. 3 (Spring 1960): 309–10.

40 United States v. Silk, 67 S. Ct. 1463, 1466–67.

41 Ibid., 1469.

42 Bartels v. Birmingham, 67 S. Ct. 1547, 1550.

43 I am indebted to an anonymous reviewer for this observation.

44 Cohen and Calhoon, “Social Security Legislation, January–June 1948,” 7.

45 President Harry S. Truman, “State of the Union Address, 1947,” January 6, 1947.

46 See Dubovsky, Melvyn, The State and Labor in Modern America (Chapel Hill: University of North Carolina Press, 1994), 201Google Scholar.

47 Literature on labor history and politics has devoted substantial attention to the conservative assault on organized labor in the late 1940s. These works generally focus on the major provisions of the Taft-Hartley Act, which overshadowed the debate then unfolding over who was considered an employee. See, for example, Dubovsky, The State and Labor in Modern America; Gross, James A., Broken Promise: The Subversion of U.S. Labor Relations Policy, 1947–1994 (Philadelphia, PA: Temple University Press, 1995)Google Scholar; Lee, R. Alton, Truman and Taft-Hartley: A Question of Mandate (Lexington: University of Kentucky Press, 1966)Google Scholar; Lichtenstein, Nelson, “Taft-Hartley: A Slave Labor Law?Catholic University Law Review 47, no. 3 (1998): 763–90Google Scholar; Plotke, David, Building a Democratic Order: Reshaping American Liberalism in the 1930 and 1940s (New York: Cambridge University Press, 1996)CrossRefGoogle Scholar. See also Twenty Years of National Labor Legislation (Washington, DC: Congressional Quarterly, 1965). For a recent survey of and intervention in the literature on labor and employment policy and an analysis of the persistent impact of New Deal–era labor laws on contemporary developments in labor politics, including employment law, see Galvin, Daniel, “From Labor Law to Employment Law: The Changing Politics of Workers’ Rights,” Studies in American Political Development 33, no. 1 (April 2019): 50–86CrossRefGoogle Scholar.

48 For a summary of the major provisions of Taft-Hartley and the debate over the measure, see Twenty Years of National Labor Legislation, 17–20.

49 Title I, Section 2, of the House bill, entitled “Definitions,” proposed changes to the meaning of “employee” by adding to the list of those specifically excluded (a list that had originally included domestic and agricultural workers). The change that provoked the most attention was the exclusion of “foremen and other supervisory personnel.” The exclusion of independent contractors was included in the original House version of Taft-Hartley, introduced in April, though not in the Senate version. House Report 245, 80th Congress, 1st Sess. [hereinafter H. Rept. 245], 13–17. Marc Linder notes that the exclusion of independent contractors from the definition of “employee” in the final legislation was not directly discussed in the thousands of pages of congressional testimony in hearings on the Taft-Hartley legislation, nor was it mentioned in Rep. Fred A. Hartley's 1948 book, Our New National Labor Policy: The Taft-Hartley Act and the Next Steps (New York: Funk & Wagnalls, 1948). Linder, The Employment Relationship in Anglo-American Law, 223.

50 The report elaborated: “In the law, there has always been a difference, and a big difference, between ‘employees’ and ‘independent contractors.’ ‘Employees’ work for wages or salaries under direction supervision. ‘Independent contractors’ undertake to do a job for a price, decide how the work will be done, usually hire others to do the work, and depend for their income not upon wages, but upon the difference between what they pay for goods, materials, and labor and what they receive for the end result, that is, upon profits” (H. Rept. 245, 18).

51 The committee stated, “It is inconceivable that Congress, when it passed the act, authorized the board to give to every word in the act whatever meaning it wished” (H. Rept. 245, 18).

52 “Minority Report,” H. Rept. 245, 112.

53 Amendments from liberal Democrats were defeated, including a substitute bill offered by Senator Murray (D-MT) that did not include the independent contractor exclusion. For the relevant section of the Murray substitute, see Legislative History of the Labor Management Relations Act, 1947, (Washington, DC: National Labor Relations Board, 1948), II: 1443.

54 The relevant case was another involving Hearst, addressing employee status for the purposes of social security rather than for unionization (which was the focus of the 1944 Supreme Court Hearst case). Hearst Publications, Inc. v. The United States was decided by district court on December 31, 1946, and affirmed by appeals court in January 1947.

55 Newspaper Vendors, Hearings before the Committee on Ways and Means, House of Representatives, 80th Congress, 1st Sess., June 12, 1947, 1–2.

56 Ibid., 2, 4.

57 Ibid., 4, 7.

58 The “type of sale” to which he referred was one in which publishers sold papers to the vendors, and the latter resold them for profit, and the impact of eliminating it, he contended, would be to throw the vendors out of work. Newspaper Vendors, 28.

59 Newspaper Vendors, 10.

60 Ibid., 14.

61 Ibid., 11.

62 Ibid., 12-13.

63 Ibid., 17.

64 Ibid., 17.

65 Ibid., 18.

66 Harry S. Truman, “Memorandum of Disapproval of Bill To Exclude Newspaper and Magazine Vendors from the Social Security System,” August 6, 1947.

67 In addition to setting out specific factors to be used in determining employment status, the regulations specified a number of characteristics of a “typical independent contractor,” including maintaining a separate establishment, performing services to complete a specific job for a price agreed in advance, offering services publicly or to customers chosen by the contractor, performing work under the contractor's name or a trade name (rather than under the name of a customer), and operating an ongoing business that could be sold. Federal Register 12, no. 232 (November 27, 1947), 7966–69.

68 Cohen and Calhoon, “Social Security Legislation, January–June 1948,” 7.

69 See Broden, “General Rules Determining the Employment Relationship under Social Security Laws.”

70 U.S. Congress, Congressional Record—House, April 14, 1948, 4432.

71 Ibid., 4429.

72 H. Rept. 1319, 2. The Republican narrative about the history of the employer-employee relationship was captured in a “legislative history” appended to the report.

73 H. Rept. 1319, 2.

74 Truman Veto Message, Document No. 594, House of Representatives, 80th Congress, 2nd Sess., April 6, 1948.

75 U.S. Congress, Congressional Record—House, April 14, 1948, 4429.

76 H. Rept. 1319, 4.

77 Ibid., 12. The full title of the report, filed by the House Ways and Means Committee on February 3, 1948, was “Maintaining the Status Quo in Respect of Certain Employment Taxes and Benefits Pending Action by Congress on Extending Social Security Coverage.” The legislation that the report accompanied was H.J. Res. 296.

78 H. Rept. 1319, 17.

79 Ibid., 16.

80 Ibid., 17.

81 Ibid., 19.

82 Ibid., 13.

83 As an indication of how widely these manipulations might be used, Secretary Wiggins cited the advice published by a leading tax service, intended to assist employers seeking to “re-classify” their workers: “Many employers have taken steps to eliminate payroll tax liability on certain individuals by changing their status from that of employees to that of independent contractors. The types of employees where such change is feasible include, among others, salesmen, selling agents, factors, brokers, bulk oil operators, store managers, motion picture theater managers, and taxicab drivers.” The tax service addressed specific steps that employers must take to avoid legal liability when undertaking this change in the status of their workers. H. Rept. 1319, 13–14.

84 H. Rept. 1319, 14–15.

85 See, for example, the statement by Representative Harness (R-IN), U.S. Congress, Congressional Record—House, 80th Congress, 2nd Sess., February 27, 1948, 1888–89.

86 Social Security Status Quo Resolution, Hearings before the Committee on Finance, United State Senate, 80th Congress, 2nd Sess., April 1–2, 1948, 175.

88 Ibid., 173.

89 Representatives of both the AFL and CIO testified against the Status Quo Resolution, as did representatives of several member unions. See Social Security Status Quo Resolution. For a list of speakers, see pp. iii–iv.

90 In the Hearst case (1944), and in the court and the National Labor Relations Board actions leading up to it, for example, an extensive Amicus Curiae brief on behalf of the vendors (“newsboys”) was filed by the International Printing Pressmen and Assistants’ Union of North America (IPPAU-NA), AFL, 1944 WL 42809 (U.S.) (Appellate Brief), Supreme Court of the United States. And in the Kansas City Star Company case, decided by the NLRB in February 1948, citing the Taft-Hartley amendment to the NLRA regarding independent contractors, the Newspaper Carriers’ Cooperative Association of Greater Kansas City, affiliated with the IPPAU-NA, AFL, was the petitioner seeking recognition of a bargaining unit comprised of “all nonsupervisory home delivery carriers of the Company” in Kansas City. NLRB, Kansas City Star Co. 1948, 385. One notable exception to the general position of organized labor was that of the Newspaper and Periodical Vendors and Distributors Union, Local No. 468, AFL, based in San Francisco, which negotiated contracts with Hearst newspapers; the union sent telegrams in support of Representative Gearhart's news vendor bill, advocating the “sale of papers by adults under an independent contractor relationship.” U.S. Congress, Congressional Record—House, 80th Congress, 2nd Sess., April 14, 1948, 4429.

91 Social Security Status Quo Resolution, 53, 57.

92 Ibid., 181.

93 Ibid., 188.

94 Harry S. Truman, “Veto of Resolution Excluding Certain Groups from Social Security Coverage,” June 14, 1948.

95 Truman told Gearhart's constituents, “You have got a terrible Congressman here. He has done everything he possibly could do to cut the throats of the farmer and the laboring man” (“The Congress: Face of the Victor,” Time, November 15, 1948).

96 Gross, Broken Promise, 43. Repealing Taft-Hartley was perceived by many Democrats and union leaders as the most promising course to regain broader momentum on labor issues. A full repeal would have restored the Wagner Act definition of employee and signaled a repudiation of the Republicans’ interpretation of that definition. A partial repeal (of the right-to-work provisions, for example) would not necessarily have addressed the question of employment status.

97 Gross, Broken Promise, 46, 57. For contemporaneous accounts of the initial efforts to repeal Taft-Hartley, see B. W. Patch, “Revision of the Taft-Hartley Act,” in Editorial Research Reports 1948 (Washington, DC: Congressional Quarterly Press, December 1, 1948), vol. II, 843–62; and “Record of the 81st Congress (Second Session) (1950),” in Editorial Research Reports 1950 (Washington, DC: Congressional Quarterly Press, September 25, 1950), vol. II, 617–18.

98 The 1952 platform called for a full repeal, making the case in seven paragraphs; twelve years later, the language on Taft-Hartley was reduced to two sentences, focused on repealing its most egregious limits on unionization. For the Democratic platforms in 1952 and 1964, see “Text of Democratic Party Platform for 1952 Race as Adopted by the Convention,” New York Times, July 24, 1952; “Text of Democratic Party Platform's Domestic Section as Approved by Committee,” New York Times, August 25, 1964. The 1964 platform focused on repealing Section 14(b) of Taft-Hartley, which enabled the passage of state-level right-to-work laws, rather than repealing the act as a whole (which would include removing the independent contractor exclusion).

99 Cohen, Wilbur J. and Myer, Robert J., “Social Security Act Amendments of 1950: A Summary and Legislative History,” Social Security Bulletin 13, no. 10 (October 1950): 3–14Google Scholar. The amendments also significantly increased the value of social security benefits and gradually increased both the employer and employee share of the social security tax.

100 Cohen, Wilbur J., “Aspects of Legislative History of the Social Security Act Amendments of 1950,” Industrial and Labor Relations Review 4, no. 2 (1951): 193CrossRefGoogle Scholar.

101 H. Rept. 1319, 14–15. The category “self-employed” includes independent contractors, as well as others who are not employees, such as some small business owners.

102 NLRB Case No. 17-R01701, decided February 26, 1948. See “Carriers Who Deliver Newspapers to Home Under Contract Barred to Unions by NLRB,” New York Times, February 28, 1948, 42.

103 Morris Steinberg, et al., 78 NLRB, No. 35 (July 14, 1948), 220–21. See also Southwestern Associated Telephone Co., 76 NLRB, No. 157 (April 2, 1948).

104 Community for Creative Non-Violence v. Reid 490 U.S. 730 (1989), 739–40.

105 Justice Souter called these congressional initiatives “legislative revisitations” of the Court's Hearst and Silk rulings. Nationwide Mut. Ins. Co. v. Darden case 503 U.S. 318 (1992), 325.

106 The NLRA today still excludes “any individual having the status of an independent contractor,” the provision added under Taft-Hartley in 1947. National Labor Relations Act, Section 2(152)(3). And the Social Security Act defines “employee,” in the wake of the Status Quo Resolution of 1948, as “any individual who, under the usual common law rules applicable in determining the employer-employee relationship, has the status of an employee.” Social Security Act, Section 210(j)(2).

107 For a summary of these laws and the standards used to determine eligibility for coverage, see U.S. General Accounting Office, “Contingent Workers: Incomes and Benefits Lag Behind Those of Rest of the Workforce” (GAO/HEHS-00-76, June 2000), 49–55. The impact of statutory language in determining worker coverage has emerged in studies comparing New Deal and post–New Deal laws. Title VII of the Civil Rights Act, the Age Discrimination in Employment Act (ADEA), and the Family Medical Leave Act (FMLA), for example, all incorporated the Fair Labor Standards Act's (FLSA's) vague definition of employee, but only the FMLA also included the FLSA's broader definition of “employ.” A side-by-side comparison of how courts interpret these three laws shows a mixed record on Title VII and the ADEA, with courts using various legal tests, including the common law control test and the economic realities test. Coverage under the FMLA, however, has been interpreted more broadly, primarily on the basis of the economic realities test. See Muhl, “What Is an Employee,” 3–11; Barron, Myra, “Who's an Independent Contractor—Who's an Employee,” Labor Lawyer 14, no. 3 (1999): 457–74Google Scholar.

108 This discussion draws in part on concepts in Paul Pierson's analysis of path dependence in “Power and Path Dependence,” in Advances in Comparative Historical Analysis, ed. James Mahoney and Kathleen Thelen (Cambridge, UK: Cambridge University Press, 2015), 123–46; and on the analysis of institutional resilience in Sheingate, “Institutional Dynamics and American Political Development,” 461–77. See also Pierson, Paul, “Increasing Returns, Path Dependence, and the Study of Politics,” American Political Science Review 94, no. 2 (June 2000): 251–67CrossRefGoogle Scholar.

109 Newspaper Vendors, 2; and Cohen, “Aspects of Legislative History,” 193.

110 For one study of how state policies shape interests, see Campbell, Andrea Louise, How Policies Make Citizens: Senior Political Activism and the American Welfare State (Princeton, NJ: Princeton University Press, 2003)CrossRefGoogle Scholar. Campbell writes, “Public policies can confer resources, motivate interest in government affairs by tying well-being to government action, define groups for mobilization, and even shape the content and meaning of democratic citizenship” (p. 1). My thanks to an anonymous reviewer for drawing attention to this connection.

111 The ability of employers and their organizations to prevail in many of the battles over employment status is a reflection, in part, of the structural advantage enjoyed by “well-organized groups and those with substantial resources and lobbying and legal capacities” in longer-term contestations over policies that are ambiguous and subject to interpretation. Hacker, Jacob S., Pierson, Paul, and Thelen, Kathleen, “Drift and Conversion: Hidden Faces of Institutional Change,” in Advances in Comparative Historical Analysis, ed. Mahoney, James and Thelen, Kathleen (Cambridge, UK: Cambridge University Press, 2015), 189, 197Google Scholar.

112 See Paul Pierson's discussion of the ways “signaling effects of outcomes at critical junctures” tend to produce increased power for the actors who prevail and diminished resources and expectations for those on the losing side. Pierson, “Power and Path Dependence,” 134–41.

113 For an analysis of multiple mechanisms of path dependence, including anticipated reactions and agenda control, see Pierson, “Power and Path Dependence,” 126–31.

114 For a detailed account of the legislative battles over repealing Taft-Hartley in 1965 and 1966 and an argument about the importance of Johnson's delay in creating conditions for conservatives to stage a successful filibuster to block repeal of Section 14(b) despite its passage in the House and majority support in the Senate, see Johnston, Travis, “A Crowded Agenda: Labor Reform and Coalition Politics in the Great Society,” Studies in American Political Development 29, no. 1 (2015): 89–105CrossRefGoogle Scholar.

115 House Committee on Education and Labor, “Equal Employment Opportunity Act of 1963,” House Report 570, 88th Congress, 1st Sess., July 22, 1963, 7. For a detailed account, see, for example, Vaas, Francis J., “Title VII: Legislative History,” Boston College Industrial and Commercial Law Review 7, no. 3 (April 1966): 531–58Google Scholar. See also Burstein, Paul and MacLeod, Margo W., “Prohibiting Employment Discrimination: Ideas and Politics in the Congressional Debate over Equal Employment Opportunity Legislation,” American Journal of Sociology 86, no. 3 (November 1980): 512–33CrossRefGoogle Scholar.

116 By 1989, a U.S. Government Accountability Office report found that 38 percent of examined employers had misclassified employees. U.S. Government Accountability Office, “Misclassification of Workers” (GAO-GGD-89-107, 1989), 2.

117 Commission on the Future of Worker-Management Relations, Report and Recommendations (Washington, DC: United States Department of Labor and United States Department of Commerce, December 1994), 35. The commission was also called the “Dunlop Commission,” for its chair, John T. Dunlop.

118 Commission on the Future of Worker-Management Relations, Report and Recommendations, 35–36.

119 Ibid., 36.

120 Planmatics Inc., Independent Contractors: Prevalence and Implications for Unemployment Insurance Programs (Washington, DC: U.S. Department of Labor, Employment and Training Administration, February 2000), iii.

121 U. S. Government Accountability Office, “Employee Misclassification: Improved Coordination, Outreach, and Targeting Could Better Ensure Detection and Prevention” (GAO-09-717, September 9, 2009).

122 See, for example, Ratna Sinroja, Sarah Thompson, and Ken Jacobs, “Misclassification in California: A Snapshot of the Janitorial Services, Construction, and Trucking Industries” (UC Berkeley Labor Center, March 11, 2019).

123 See, for example, Alexander, “Misclassification and Antidiscrimination,” 907–67.

124 Unions such as the Teamsters, Service Employees International Union, and UNITE HERE have pursued campaigns against worker misclassification, for example, and the AFL-CIO's top legislative priority in 2022, the proposed Protecting the Right to Organize Act, sought to address the issue by establishing an ABC test to determine employee status for purposes of the National Labor Relations Act. The National Employment Law Project, a leading worker rights advocacy organization, has also devoted extensive legal, research, and advocacy resources to addressing the issue of employee misclassification. See National Employment Law Project, “Turning ‘Gig’ Jobs into Good Jobs,” accessed November 22, 2022, https://www.nelp.org/campaign/turning-gig-jobs-into-good-jobs/.

125 Deknatel, Anna and Hoff-Downing, Lauren, “ABC on the Books and in the Courts: An Analysis of Recent Independent Contractor and Misclassification Statutes,” University of Pennsylvania Journal of Law and Social Change 18, no. 1 (2015): 55–57Google Scholar.

126 California's ABC standard states, at California Labor Code 2775(b)(1), “[A] person providing labor or services for remuneration shall be considered an employee rather than an independent contractor unless the hiring entity demonstrates that all of the following conditions are satisfied: (A) The person is free from the control and direction of the hiring entity in connection with the performance of the work, both under the contract for the performance of the work, and in fact. (B) The person performs work that is outside the usual course of the hiring entity's business. (C) The person is customarily engaged in an independently established trade, occupation, or business of the same nature as that involved in the work performed.”

127 California Legislative Analyst's Office, “The 2020-21 Budget: Staffing to Address New Independent Contractor Test” (February 11, 2020).

128 Nineteen states use a version of the ABC standard to determine eligibility for state unemployment programs. “Worker Classification: Employment Status Under the National Labor Relations Act, the Fair Labor Standards Act, and the ABC Test,” 14–27.

129 Matthew Haag and Patrick McGeehan, “Uber Fined $649 Million for Saying Drivers Aren't Employees,” New York Times, November 15, 2019, A21; Kate Conger, “California Targets Lyft and Uber in Labor Suit,” New York Times, May 5, 2020, B1; Kate Conger and Daisuke Wakabayashi, “2nd State Sues to Force Uber and Lyft to Treat Drivers as Employees,” New York Times, July 15, 2020, B3.

130 National Conference of State Legislatures, “Worker Misclassification,” accessed November 22, 2022, https://www.ncsl.org/research/labor-and-employment/employee-misclassification-resources.aspx.

131 By 2018, the companies had persuaded lawmakers in forty-one states to limit the authority of localities to set their own regulatory standards for transportation network companies, including regarding the employment status of their workers. Rebecca Smith, Joy Borkholder, Mariah Montgomery, and Miya Saika Chen, “Uber State Interference: How TNCs Buy, Bully, and Bamboozle Their Way to Deregulation” (National Employment Law Project, New York, January 18, 2018), 4.

132 For a critical examination, see Rebecca Smith, Brian Chen, and Rey Fuentes, “Rigging the Gig” (National Employment Law Project, New York, July 16, 2020).

133 Margot Roosevelt and Suhauna Hussain, “Prop. 22 Is Ruled Unconstitutional, a Blow to California Gig Economy Law,” Los Angeles Times, August 20, 2021. See also Joyce E. Cutler, “Uber Appeal of California Gig Classification Law at 9th Circuit,” Bloomberg Law Daily Labor Report, July 11, 2022. Pending their appeal of the ruling against Proposition 22, the companies did not have to abide by the terms of AB5. Political and legal conflicts continued to unfold over the question of whether other groups, particularly port truck drivers, should be granted a carveout from AB5. See, for example, Kurtis Lee, “Truckers Shut Down Oakland's Port in Protest over California Labor Law,” New York Times, July 22, 2022.

134 See, for example, “Uber, Lyft and Other Tech Companies Test Language for Potential Mass. Ballot Measure on Gig Workers,” Boston Globe, July 16, 2021; “Coalition Formed to Protect App-Based Workers’ Independence,” New Jersey Business and Industry Association, February 2, 2021; “Lyft Makes Largest One-Time Political Donation in Massachusetts History, Fueling Gig Worker Ballot Fight,” Boston Globe, January 18, 2022; “Court Halts State's Vote on Gig Work,” New York Times, June 15, 2022.

135 “Application for Leave to File Amici Curiae Brief of Chamber of Commerce of the United States and California Chamber of Commerce in Support of Petitioner Dynamex Operations West, Inc.,” December 4, 2015, 7, https://www.courts.ca.gov/documents/7-s222732-ac-chamber-commerce-usa-et-al-121015.pdf.

136 Josh Eidelson, “Gig-Economy Giants Ask California to Save Them from a Ruling That May Turn Their Contractors into Employees,” Los Angeles Times, August 6, 2018.

137 Though the FLSA definition of employee is as vague and limited as others, the law includes a more expansive definition of “employ,” which has led to a more expansive interpretation of coverage under the FLSA than under other New Deal laws. For the legislative language, see “Senators Scott, Blackburn Introduce Modern Worker Empowerment Act,” March 2, 2021, https://www.scott.senate.gov/media-center/press-releases/senators-scott-blackburn-introduce-modern-worker-empowerment-act-. The legislation was drafted and promoted by the industry-backed Coalition to Promote Independent Entrepreneurs, headed by Russell A. Hollrah (“Mission Statement,” accessed November 22, 2022, https://iccoalition.org/about-us/). For an extended version of his argument about the need to complete the work of the 80th Congress in aligning the definition of “employee” under New Deal legislation with the common law, see Hollrah, Russell A. and Hollrah, Patrick A., “The Time Has Come for Congress to Finish Its Work on Harmonizing the Definition of Employee,” Journal of Law and Policy 26, no. 2 (2018): 439–86Google Scholar.

138 Federal Register 86 (January 7, 2021), 1168–1248.

139 Federal Register 86 (May 7, 2021), 14027–14038.

140 See United States Department of Labor, Wage and Hour Division, “Misclassification of Employees as Independent Contractors,” https://www.dol.gov/agencies/whd/flsa/misclassification. See also “How the ‘Coalition for Workforce Innovation’ Is Putting Workers’ Rights at Risk” (National Employment Law Project, New York, July 2022).