Modeling the Impact of Sectoral Bargaining for U.S. Workers

An important and consequential contradiction exists in the U.S. economy, specifically in the job market. Many workers say they would like to be represented by unions, but the share of American workers in unions or covered by a collective bargaining agreement is near an all-time low. This issue brief explores why that gap between workers’ rational desire for union representation exists, why coverage is lower in America than in most other advanced economies, and how policy that supports sectoral bargaining would have a high likelihood of helping to close that gap.

The assertion that workers’ desire for union representation is “rational” is based on well-established empirical findings that union coverage improves workers’ job quality through better compensation and working conditions. Workers covered by union contracts earn significantly more than similar nonunion workers and experience corresponding increases in access to health insurance, better on-the-job protections, and—particularly for workers without college degrees—large increases in lifetime earnings and wealth. Broadly speaking, research shows that one of the key roles played by unions is to ensure that their members end up with a fairer slice of the pie that their labors are helping to bake. Not surprisingly, nearly half of workers say they would like to join a union, and more than two-thirds of the public support unions.

Why, then, is coverage not higher?

Many structural barriers to forming unions and bargaining collectively exist in the U.S. labor market. Existing labor law is so weak that it in effect allows anti-union corporations to undercut worker organizing and bargaining through both legal and illegal means—so much so that just 1 in 7 organizing drives successfully secures a collective bargaining agreement, according to a 2008 study, a share that is much smaller than workers’ expressed preferences.

This means that despite the popularity of unions, union membership and collective bargaining coverage—particularly in the private sector—have steadily declined for decades. In 2025, U.S. private sector union membership was just 6 percent and the overall union membership rate was only 10 percent, down from an overall high of 34 percent in the mid-20th century and lower than at any point since the 1930s. Collective bargaining coverage was just 7 percent in the private sector and 11 percent overall. This decline has been near constant for several decades and has occurred in virtually every private sector industry—sectors exposed to and sheltered from trade alike and in services, manufacturing, and construction.

Several policy solutions, including those in the Protecting the Right to Organize Act (PRO Act) that would hold employers accountable when they violate workers’ rights,* can address the fact that union coverage in the United States is suppressed by anti-union forces. Based on the evidence presented below, the authors believe that promoting sectoral bargaining must also be part of the suite of needed reforms.

Sectoral bargaining is a type of collective bargaining that creates minimum standards and provides representation for all workers in a particular industry or occupation. It typically operates in conjunction with workplace-level bargaining—the current default for bargaining in the United States—setting sector-wide standards that workers can seek to bargain above at their worksite. Sectoral bargaining can go by different names including multiemployer and industry-wide bargaining, and countries support it in different ways, but all sectoral bargaining involves unions negotiating standards that cover workers at multiple worksites. Achieving sectoral bargaining in the United States will take significant reforms to federal labor law, though there are some emerging state models to build upon.

The evidence presented below suggests that promoting sectoral bargaining would significantly increase collective bargaining coverage—the percentage of workers whose wages and benefits are determined by a collective bargaining agreement, regardless of whether they are union members—and would also likely support higher levels of union membership. Indeed, the authors’ modeling suggests that moving to sectoral bargaining could more than double collective bargaining coverage in the United States, from about 11 percent to just less than 30 percent, and also significantly increase union membership.**

Such an increase in collective bargaining coverage would yield many advantages to the U.S. economy and labor market. First, it would close the “representation gap” between the coverage many workers want and the much lower level that they get in this country. Second, because coverage is associated with a fairer distribution of compensation, increased collective bargaining coverage would help reduce economic inequality and ensure workers have a better chance to benefit from their contributions to economic growth. Third, while Americans’ ongoing affordability challenges have been largely cast in terms of prices, those challenges cannot be met without higher real wages. Thus, the sectoral bargaining path to higher union coverage could be part of the solution to the affordability shortfall.

Figure 1 illustrates the basic premise that sectoral bargaining can boost bargaining coverage. Data from the Institutional Characteristics of Trade Unions, Wage Setting, State Intervention and Social Pacts (ICTWSS) database published by the Organisation for Economic Co-operation and Development (OECD) were used to compare how different bargaining systems correlate with the number of workers covered by collective agreements. This dataset provides indicators on a range of wage-setting characteristics, union density, bargaining coverage, and more on the developed economies of OECD member states as well as several emerging economies. Bargaining coverage averaged from 1960 to 2019 is plotted on the y-axis; the x-axis plots “level,” defined as the predominant level where bargaining takes place in a country. Level 1 is workplace bargaining, 2 is intermediate, 3 and above are sectoral, with 4 and 5 representing the most encompassing versions of sectoral bargaining.

The relationship is fairly close; the sectoral bargaining level explains 77 percent of the variation in bargaining coverage. The United States is, as expected, at the very low end of both variables, with a low bargaining level—meaning bargaining occurs in the most balkanized manner—and low bargaining coverage. In contrast, countries with sectoral bargaining have high coverage. (Note that this includes coverage of both union members and nonmembers.)

Though these data are often noisy and the relationships between the variables not always clear, Greece offers a useful case study of the impact of moving from sectoral bargaining to a more mixed system with the predominant level of bargaining closer to workplace level, as can be seen in Figure 2. This graph shows how Greece’s level of coverage changed over a 14-year window (unlike Figure 1, which averages across the entire 60-year span of the dataset) surrounding a series of anti-labor policy reforms—including the weakening of its sectoral bargaining system—that Greece made starting in 2010-2011 in exchange for loans to help bail the country out of a financial crisis following the 2008 global recession. As policy changes undermined sectoral bargaining, the shock to bargaining coverage was instantaneous and sharp: It fell from more than 80 percent to about 13 percent.***

However, most shifts in coverage following a change in the predominant level of bargaining are not this clear-cut, as illustrated in Figure 3. This figure also shows how bargaining coverage changed over a several-year period centered on major bargaining policy changes in the United Kingdom and New Zealand.

In the U.K., starting in 1980, then-Prime Minister Margaret Thatcher pursued a host of anti-labor policies, including some that weakened support for sectoral bargaining. This led to a steady decline in both union membership and collective bargaining coverage. Yet sectoral bargaining was still the predominant level of bargaining until 1987, when bargaining was decentralized to an intermediate level—alternating between workplace and sector. For several years, sectoral bargaining still predominated, likely due to the length of some contracts as well as historical culture and some remaining union power. The gradual nature of this decline, and correlation with a whole suite of other anti-labor policy changes, illustrates that while coverage and sectoral bargaining policy changes are clearly correlated, causality is hard to isolate.

New Zealand provides another example of reduced coverage predating a shift in the predominant level of bargaining. In the late 1980s and early 1990s, New Zealand undertook a series of policy changes to weaken unions and reduce sectoral bargaining, including a 1987 law aimed at shifting collective bargaining from the sector to the workplace level and a 1991 law that imposed harsh limits on collective bargaining in any form. Bargaining coverage began declining soon after the 1987 law, but the predominant level of bargaining did not change until 1992.

Figures 1-3 illustrate that sectoral bargaining is associated with higher bargaining rights coverage and that shifts toward more decentralized bargaining arrangements tend to coincide with declines in bargaining rights coverage. The authors argue that sectoral bargaining is a powerful policy to advance workers’ rights and welfare in the United States. This section will connect this descriptive evidence to the authors’ proposed policy solution using a simple statistical model that projects how U.S. coverage might change under a shift toward more centralized collective bargaining. These results strongly suggest that a move to sectoral bargaining would substantially raise bargaining coverage. However, because of data limitations, these results should not be read as a precise prediction but as an order-of-magnitude projection.

Several features of the data make this analysis challenging. First, in some cases, the independent variable—changes in the level at which bargaining predominates—lags behind changes in policy and bargaining coverage, as discussed in the above case studies. Second, although it is a relatively long time-series (60 years) and large sample (34 countries), there is very little change in the independent variable of bargaining status (workplace level, mixed, sectoral), limiting the authors’ ability to identify the impact of such changes. Third, and perhaps surprisingly, there are few examples in the OECD data of bargaining policy becoming more centralized, such as by going from more to less fragmented. (Yet in examples outside the ICTWSS database, countries have recently moved toward sectoral bargaining: Though Australia gutted its sectoral system in the 1990s, it adopted policies starting in 2022 to promote sectoral bargaining and has since seen early signs of increased union membership and collective bargaining coverage.) While the lack of examples of bargaining policy becoming more centralized is a challenge for the analysis, it also serves as a potent commentary on who has had the upper hand in labor policymaking over recent decades.

Even with these caveats, it is possible to learn more about the relationship in collective bargaining and coverage by using these OECD data to run regressions. Regressions attempt to isolate the impact of moving to a sectoral bargaining system by controlling for the effects of other variables, such as country characteristics and global economic trends. The numerous versions run of the model all led to a similar conclusion: Sectoral bargaining is associated with much higher collective bargaining coverage.

The preferred model is one that set country- and year-specific parameters, or “fixed effects,” to control for persistent country differences and yearly shocks to coverage that occurred across all countries, such as the global financial crisis in 2008. With only these fixed effects, sectoral bargaining is associated with a statistically significant boost to coverage of 30 percentage points. Because union density is very strongly related to both collective bargaining and coverage, authors also estimate regressions that include union density as a control. While adding union density as a control results in a smaller boost to coverage of 18 percentage points, the increase remains positive and statistically significant—suggesting that sectoral bargaining and union density each have their own impacts on coverage. Note that in a separate fixed effects regression, sectoral bargaining relative to U.S.-style workplace-level bargaining is associated with a 13-percentage point increase in union density.

The Appendix shows the results of these fixed effects regressions, as well as pooled regressions that do not account for differences between countries or years. The pooled model finds larger coefficients on sectoral bargaining. Additional analysis not shown in this paper with different and additional variables and techniques led to similar results.

This model also was used to project how U.S. bargaining coverage would change with a shift to sectoral bargaining, holding constant the U.S. baseline level of coverage. This matters because the United States differs in many ways, including in the structure of its economy and the arrangements between capital and labor, from its OECD peers, particularly the Western and Northern European countries that form the bulk of the sample. The fixed effects model can only partially control for such deeply embedded differences.

Thus, while the direction of the findings leads to confidence that sectoral bargaining would raise coverage in the United States, the magnitudes of the estimates are not guaranteed.

The simulation holds union density constant, ignoring any increase in union membership due to sectoral bargaining, and controls for the effects of union density on coverage. This enables a focus on the impact of sectoral bargaining directly on coverage. Reiterating the many caveats about the data noted above, and noting that the authors are making an estimate for 2025 coverage levels based on an analysis of data that only extend to 2019, this exercise is suggestive of what might be achievable in terms of coverage, while controlling for the United States’ long history, relative to other countries in the database, of suppression of union membership. If the United States adopted sectoral bargaining, the authors’ research shows, coverage would increase from about 11 percent of all workers (using 2025 data) to 29 percent of all workers.

The authors can then estimate the number of workers that would be covered by these new collective bargaining agreements. In 2025, 16.5 million workers were covered by collective agreements; this analysis suggests that this would increase to an estimated 42.4 million under these new collective bargaining agreements. This economically significant growth in the number of covered workers would then be accompanied with the empirically well-established benefits of bargaining rights coverage, as discussed below.

The results of this analysis by the Center for American Progress are consistent with other research. There is overwhelming evidence finding that sectoral bargaining systems increase the number of workers covered by collective agreements. As the researchers at the OECD explain, “collective bargaining coverage is high and stable only in countries” with sectoral bargaining. Sectoral bargaining leads to higher coverage for a number of reasons, including that it sets standards across multiple worksites, ensures workers do not lose representation as firms open and close, and limits the effectiveness of firm restructuring to avoid paying union wages. Sectoral bargaining is particularly effective at covering those who are hardest to reach through worksite-level bargaining, such as workers that are contracted out, in small worksites, employed part time, or in other nonstandard employment.

Academic research has also shown that sectoral bargaining systems are associated with more equal distribution of wages and reduced poverty, especially compared with an enterprise-level system of bargaining such as the one used in the United States. Some countries have weakened their existing centralized wage-setting systems, which increased earnings inequalities. More centralized bargaining systems have also been shown to reduce the gender wage gap since women’s pay tends to be closer to wage floors, and the decentralized wage-setting system of the United States is one reason why it has a larger gender wage gap than other comparable countries.

In addition, sectoral bargaining also increases productivity because it creates a level playing field that forces firms to compete based on innovation and quality rather than low labor standards.

Research also finds that sectoral bargaining tends to increase union membership, with one study finding that “sector-level bargaining, as opposed to plant-level or national-level bargaining, turns out to be the most conducive to union growth.” However, there are nuances to the research on union density. For example, features of some sectoral systems, such as automatically extending collective bargaining agreements to workplaces without unions, can potentially disincentivize union membership (although, in many countries with these mechanisms, union membership is far higher than in the United States). There are also outliers, such as France, that have very low union membership because of unusual sectoral systems that hinder membership in a variety of ways. Ultimately, the research generally finds that sectoral bargaining increases union membership, though policy details matter, and complementary policies that facilitate recruitment and retention can be necessary to help support very high union membership in sectoral systems.

In short, previous research shows that sectoral bargaining increases collective bargaining coverage, boosts union membership, and leads to improvements in workers’ lives and the economy. Furthermore, though this report is the first that quantifies anticipated changes expected from enacting a sectoral bargaining system in the United States, previous research is consistent with the direction and scale of CAP’s findings. Indeed, some research suggests even larger impacts. For example, a 2023 paper that incorporated sectoral bargaining into its model of the relationship between union density and the number of workers covered by a union contract in developed countries found that increasing the degree of wage-setting centralization increases coverage by 15 percentage points on its own, and incorporating an automatic sectoral bargaining extension mechanism increases coverage by 17 additional percentage points.

Both CAP analysis and previous extensive research create an empirically muscular case that moving to sectoral bargaining would increase union coverage, which in turn would raise the living standards of millions of workers, helping to close the gap between what they currently earn and their contribution to the economy’s growth. What, then, needs to change so that workers can have a chance to achieve the union density and collective bargaining that so many say they want?

First, it must be easier for workers to join unions and bargain collectively, closing the representation gap between workers wanting and achieving collective bargaining. This starts with the PRO Act, which would create real financial penalties when employers illegally union bust and provide for first contract arbitration to prevent delay tactics. Labor law reforms should also include policies that more actively facilitate organizing, such as those that make it easier for workers to learn about unions and create additional incentives for union membership.

Second, the United States needs to promote sectoral bargaining. This means changing federal law to get rid of barriers that make it hard for workers to bring multiple employers to the bargaining table, expanding the use of prevailing wage-style policies that set minimum sectoral standards based on union contracts, and creating new sectoral bargaining pathways. While these would be a big change from current federal law, there are a number of state models to build upon.

The authors would like to thank Larry Mishel and Jesse Rothstein for their feedback on this report.

* The PRO Act would make several additional changes including increasing the number of workers with collective action rights, limiting tools employers can use to union bust, and adopting new procedures to make sure unions can reach a first contract.

** Membership (workers who join unions) and coverage (workers covered by a collective bargaining agreement) differ under the current system largely due to “right-to-work” laws and could differ further under sectoral bargaining.

*** It is important in this analysis to distinguish between union density—the share of the workforce that are direct members of unions—and collective bargaining coverage, which includes nonmembers covered by contracts. In several countries it is not uncommon to have higher coverage than density. In Greece, union density also declined over this period—from 20 percent in 2010 to 13.4 percent in 2020—though not to the same degree as coverage.

These tables show the results of the analysis. Table A1 includes the results of time- and country-level fixed effects regressions of collective bargaining coverage on whether or not the country has a sectoral bargaining system. Table A2 shows the results of regressing union density on bargaining system under both fixed-effects and pooled models. Table A3 shows the results of pooled regressions of bargaining coverage on bargaining system.