Workers' Rights and Labor Law Reform: What Each Position Is Protecting

In 1954, approximately 35 percent of American workers belonged to a union. In 2024, that number was 9.9 percent — a record low, down from a record low the year before. Private sector union density is now 5.9 percent. If you round to the nearest whole number, it is 6.

This is not a story about economic inevitability. The National Labor Relations Act was passed in 1935. Within fifteen years, union density had more than doubled. By the mid-1950s, American unions had negotiated the eight-hour workday, employer-provided health insurance, defined benefit pensions, and paid vacation into the standard employment relationship — benefits that spread through wage-setting norms to non-union workers as well. The postwar middle class was not built by market competition alone. It was built in part by collective power to bargain.

The collapse of that power is the subject of genuine dispute. Labor scholars point to legal architecture: decades of employer legal challenges, regulatory rollback, and the particular vulnerabilities of an election-based organizing system that allows employers to contest, delay, and appeal at every step. Business economists point to structural economic change: manufacturing's decline, the shift to services and technology, globalization. Both are correct about something, and neither fully explains what the other describes. The debate is not primarily about whether unions are good or bad — it is about what the legal framework should be, who holds power in workplace governance, and whether the current system reflects a genuine equilibrium or a failure that has been systematically produced.

What labor law reformers are protecting

The proposition that the NLRA's procedural architecture has been converted from a framework that enables organizing into one that enables defeating it — and that this conversion has not been accidental. The Wagner Act of 1935 gave workers the right to organize and bargain collectively and created the NLRB to enforce that right. What it did not do was immunize the organizing process against the employer's vastly greater resources for contesting it. In a typical NLRB election campaign, the employer can require workers to attend mandatory meetings on company time, known as captive-audience meetings, at which managers deliver anti-union presentations. Workers can hold no equivalent meeting on company time. Employers can distribute anti-union literature through normal company channels. Workers cannot. The election itself typically occurs forty to sixty days after a petition is filed — a period during which employer opposition is at its most intense. Firing a union organizer triggers an NLRB complaint and, years later, a back-pay order minus interim earnings. For large employers, this is a calculable cost of doing business, not a deterrent. The Economic Policy Institute has documented that employers are charged with violating federal law in 41 percent of union election campaigns.

The specific evidence that winning an election is not the same as winning a union — and that the post-election terrain is where most organizing campaigns go to die. Workers at the Staten Island Amazon fulfillment center voted 2,654 to 2,131 to join the Amazon Labor Union in April 2022 — the first successful Amazon organizing campaign in American history. Four years later, they still have no first contract. Amazon has filed objections, appealed NLRB decisions, and litigation has continued through every available venue. Starbucks Workers United has organized over 520 stores and 11,000 workers across 40 states; the NLRB has docketed 771 unfair labor practice charges against Starbucks, which the agency's general counsel described as "almost certainly the largest in 90 years of NLRB history." Research suggests that only about half of workers who vote to unionize ever win a first contract. Labor law reformers are protecting the recognition that a legal framework that allows this outcome, repeatedly and predictably, is not neutral — it has a thumb on the scale, and the thumb belongs to the employer.

The case for the Protecting the Right to Organize Act: that restoring what the NLRA was supposed to guarantee requires updating both its substantive rights and its remedial framework. The PRO Act, reintroduced in 2025 as H.R. 20, would prohibit captive-audience meetings, allow workers to request card-check recognition when a majority signs authorization cards (rather than requiring an election the employer can contest), impose civil monetary penalties up to $50,000 per violation and $100,000 for repeat offenses, override right-to-work laws in the 27 states that currently allow workers covered by a union contract to opt out of paying dues, and expand the definition of "employee" to cover many workers currently misclassified as independent contractors. The AFL-CIO argues that these provisions do not expand labor law — they restore the balance the Wagner Act was meant to create and the Taft-Hartley Act of 1947 and subsequent regulatory and judicial rollback have dismantled. What this position protects is not the union movement as an institution but the basic principle that workers should be able to organize without their employer legally deploying superior resources against them in the process.

What current framework defenders are protecting

The secret ballot as a foundational protection for workers themselves — and the claim that removing it in favor of card check exposes workers to coercion from union organizers rather than protecting them from coercion by employers. The U.S. Chamber of Commerce and the National Federation of Independent Business, the two primary business lobbies opposing the PRO Act, center their opposition on the secret ballot argument: that card-check recognition allows union organizers to pressure workers face-to-face to sign authorization cards, without the confidentiality that the election process provides. When workers vote in an NLRB election, they vote anonymously. When they sign a card, they do so in front of the person asking them to sign. Opponents argue that the election requirement is not an obstacle to organizing — it is a safeguard for the worker who wants to decline without social consequence. The NFIB reports that 70 percent of its members oppose repealing right-to-work laws, which they characterize as protecting workers' freedom not to fund an organization whose activities they may oppose. What this position protects is the argument that worker freedom cuts in both directions: freedom to organize, and freedom to decline.

The structural economic argument that union density decline reflects genuine changes in how work is organized — not only legal obstruction — and that the comparison to 1954 obscures more than it reveals. The 1954 peak of union density was concentrated in manufacturing, mining, construction, and transportation — industries with large, stable workforces concentrated in single facilities, doing repetitive tasks with relatively homogeneous interests. The postwar economy that produced that density no longer exists. Manufacturing's share of employment fell from about 30 percent in the 1950s to under 9 percent today. The growth sectors — healthcare, technology, professional services, food service — have different organizational structures, more varied employee interests, and more geographic diffusion. Business advocates argue that lower union density in these sectors partly reflects this structural change and partly reflects that many workers, particularly in high-skill technology and professional services, prefer individual negotiation because their bargaining power as individuals is substantial. A software engineer at a large tech company has options that a 1950s autoworker did not. What this position protects is the recognition that the question "why aren't more workers unionized?" does not have a single answer, and that attributing the entire decline to employer legal strategy overstates how monolithic a 1.5-million-company economy is.

The concern that the PRO Act's provisions — particularly the override of right-to-work laws, the independent contractor reclassification, and the elimination of captive-audience meetings — would create compliance costs and legal uncertainty that fall disproportionately on small employers. The NFIB emphasizes that its membership consists largely of businesses with ten to fifty employees — enterprises that lack the HR infrastructure of a large corporation, cannot absorb the same legal costs, and are more dependent on workforce flexibility. The PRO Act's "ABC test" for contractor classification, modeled on California's Assembly Bill 5, would reclassify many workers currently engaged as independent contractors as employees entitled to NLRA protections — a significant operational change for industries like landscaping, construction, and home services that depend on flexible labor arrangements. Business groups are protecting the claim that labor law reform designed with large employers in mind tends to be most burdensome for small ones, and that the costs of compliance appear in the form of reduced hiring, not increased union density.

What sectoral bargaining advocates are protecting

The structural insight that enterprise-by-enterprise collective bargaining has a ceiling that most American workers can never reach — and that fixing the NLRA's procedural defects does not solve this deeper problem. Even if the PRO Act passed tomorrow and was fully enforced, the organizing model it would strengthen requires workers to organize one employer, one facility, one election at a time. Amazon has approximately 1.5 million U.S. workers spread across hundreds of warehouses, delivery stations, and fulfillment centers. Starbucks has 16,000 U.S. stores. Building a union at each location, separately, through a process that employers can contest and delay, is a decades-long project. And the benefits of that organizing are confined to the workers at organized locations — workers at non-union facilities, which will always be the majority, get nothing. Sectoral bargaining advocates — labor economists like Jake Rosenfeld, author of You're Paid What You're Worth, and policy researchers at the Economic Policy Institute — argue that the enterprise bargaining model can never replicate the wage-setting function that high union density performed in the postwar era: setting pay norms across entire industries that disciplined wage competition downward for non-union employers as well.

The international evidence that industry-wide bargaining achieves high coverage rates without requiring high union density — and that this is not a theoretical proposal but a demonstrated alternative in multiple functioning democracies. Germany's sectoral collective bargaining system covers approximately 54 percent of workers through industry-wide agreements, despite union density of only about 17 percent. The mechanism is extension: agreements negotiated between unions and employer associations can be extended by the government to cover all workers in an industry, whether organized or not. Denmark achieves even higher coverage. Australia operated an industry-based arbitration system through most of the twentieth century — covering large portions of its workforce through awards that set minimum terms — before moving to enterprise agreements in the 1990s. In 2023, California's FAST Recovery Act created a Fast Food Council — a sectoral wage board with worker and employer representatives — that set a $20 minimum wage for the fast food industry statewide. This is the American historical precedent: New York used wage boards in the 1930s to set minimum wages by industry before the federal minimum wage existed. What sectoral bargaining advocates are protecting is the argument that this path is available, that it has worked, and that the enterprise-bargaining debate is a debate about infrastructure that cannot scale to the actual problem.

The wage inequality connection: that the collapse of sectoral wage-setting norms — not just union density — explains why wage growth has decoupled from productivity since the 1970s. Rosenfeld's central argument in You're Paid What You're Worth is that unions historically did not raise wages only for their members. They set norms. In industries with high union density, non-union employers matched union wages to avoid organizing. Unions lobbied for minimum wage increases that raised floors across the economy. They publicized pay scales that workers in other industries used as reference points. When that norm-setting function collapsed, it collapsed for everyone — not just for organized workers. Real wages for non-supervisory workers stagnated for nearly forty years while productivity rose steadily. The gap between productivity and compensation grew precisely during the period in which union density fell most sharply. Sectoral bargaining advocates are protecting the case that restoring this connection requires an instrument capable of operating at scale — and enterprise bargaining, however reformed, cannot do that.

What worker ownership and codetermination advocates are protecting

The argument that the adversarial labor-capital bargaining framework is itself the source of the problem — and that transforming workers from counterparties into stakeholders changes the underlying dynamic rather than just the terms of the fight. Richard Wolff, economist at the New School and author of Democracy at Work, argues that collective bargaining, however strong, leaves the fundamental power structure of the capitalist enterprise intact: workers negotiate over the terms of their subordination but do not govern the enterprise. A union can win better wages and working conditions but cannot prevent the company from automating away those jobs, relocating to another country, or being sold to a private equity firm that extracts value through layoffs and facility closures. Worker ownership and codetermination advocates argue that the focus on bargaining rights, however necessary, is fighting over the distribution of decisions workers don't make, rather than changing who makes them.

The German codetermination model as evidence that worker governance rights at the enterprise level are compatible with competitive, globally successful economies. Germany's Mitbestimmungsgesetz of 1976 requires that companies with 2,000 or more employees have 50 percent worker representation on their supervisory boards — the body that oversees management, approves major strategic decisions, and appoints the executive. (In practice, shareholder-side chairs hold tie-breaking authority, giving shareholders a slight structural advantage in deadlock, but the requirement ensures workers are inside the room where decisions are made, not outside it.) German companies routinely outperform on long-term investment, apprenticeship training, and employee retention — outcomes that codetermination advocates attribute partly to the internal check that worker representation provides against short-term shareholder pressure. The Mondragon cooperative federation in Spain — approximately 80,000 workers who are employee-owners in a democratic governance structure — has maintained a CEO-to-worker compensation ratio capped at approximately 9:1, compared to the U.S. median of roughly 300:1. What codetermination advocates are protecting is not a utopian vision but a demonstrated institutional arrangement: workers with formal governance rights produce different outcomes than workers with only bargaining rights.

The American precedent: that worker ownership is not foreign to U.S. economic history and that a significant infrastructure already exists to expand it. Approximately 6,500 companies in the United States are majority employee-owned through Employee Stock Ownership Plans (ESOPs), covering roughly 14 million workers. The National Center for Employee Ownership documents consistent evidence that ESOP companies have higher productivity, lower turnover, and better employee wealth accumulation than comparable conventionally owned firms. The Worker Ownership, Readiness, and Knowledge (WORK) Act has been introduced in Congress to expand ESOP access and funding. Some economists, including Daron Acemoglu and others associated with the Roosevelt Institute, have proposed mandatory board-level codetermination for large U.S. corporations modeled on the German system. What this position protects is the claim that worker ownership addresses a structural question — who governs the enterprise — that collective bargaining cannot resolve no matter how strong the legal framework supporting it becomes.