Report | Unions and Labor Standards

Collective bargaining beyond the worksite: How workers and their unions build power and set standards for their industries

Download PDF

Editor’s Note: The content of this report was prepared prior to the COVID-19 pandemic, and the report does not reflect COVID-related impacts on the bargaining examples cited in the report.

What this report finds and why it matters

  • The current legal framework for collective bargaining is outdated and does not match the realities of today’s workplaces, which are geographically fragmented and dispersed and heavily reliant on contractors, staffing agencies, and franchises.
  • U.S. labor law is currently structured in a way that places significant obstacles in front of workers and unions seeking to bargain more broadly with employers in their industry to set standards.
  • Despite the obstacles, experience has shown that when unions have built strength through strong membership and density, they have been able to bargain more broadly than for a single worksite, setting standards for an entire industry or region in the process. Policymakers should enact straightforward reforms to labor law that would facilitate this broader bargaining.


The National Labor Relations Act (NLRA or Act)—the primary law establishing organizing rights in the private sector—has as its premise a lofty and admirable goal: “encouraging the practice and procedure of collective bargaining” between workers and their employers.1 Since the Act’s passage in 1935, millions of working men and women have won higher pay, better health care and retirement benefits, stronger health and safety protections on the job, and other important improvements through forming unions and using their collective strength in bargaining with their employers.2 Historically, strong unions have helped ensure that income growth is distributed broadly and not just to the richest households (see Figure A).3

Figure A

As union membership declines, income inequality rises: Union membership and share of income going to the top 10%, 1917–2017

Year Union membership Share of income going to the top 10%
1917 11.0% 40.3%
1918 12.1% 39.9%
1919 14.3% 39.5%
1920 17.5% 38.1%
1921 17.6% 42.9%
1922 14.0% 42.9%
1923 11.7% 40.6%
1924 11.3% 43.3%
1925 11.0% 44.2%
1926 10.7% 44.1%
1927 10.6% 44.7%
1928 10.4% 46.1%
1929 10.1% 43.8%
1930 10.7% 43.1%
1931 11.2% 44.4%
1932 11.3% 46.3%
1933 9.5% 45.0%
1934 9.8% 45.2%
1935 10.8% 43.4%
1936 11.1% 44.8%
1937 18.6% 43.3%
1938 23.9% 43.0%
1939 24.8% 44.6%
1940 23.5% 44.4%
1941 25.4% 41.0%
1942 24.2% 35.5%
1943 30.1% 32.7%
1944 32.5% 31.5%
1945 33.4% 32.6%
1946 31.9% 34.6%
1947 31.1% 33.0%
1948 30.5% 33.7%
1949 29.6% 33.8%
1950 30.0% 33.9%
1951 32.4% 32.8%
1952 31.5% 32.1%
1953 33.2% 31.4%
1954 32.7% 32.1%
1955 32.9% 31.8%
1956 33.2% 31.8%
1957 32.0% 31.7%
1958 31.1% 32.1%
1959 31.6% 32.0%
1960 30.7% 31.7%
1961 28.7% 31.9%
1962 29.1% 32.0%
1963 28.5% 32.0%
1964 28.5% 31.6%
1965 28.6% 31.5%
1966 28.7% 32.0%
1967 28.6% 32.0%
1968 28.7% 32.0%
1969 28.3% 31.8%
1970 27.9% 31.5%
1971 27.4% 31.8%
1972 27.5% 31.6%
1973 27.1% 31.9%
1974 26.5% 32.4%
1975 25.7% 32.6%
1976 25.7% 32.4%
1977 25.2% 32.4%
1978 24.7% 32.4%
1979 25.4% 32.3%
1980 23.6% 32.9%
1981 22.3% 32.7%
1982 21.6% 33.2%
1983 21.4% 33.7%
1984 20.5% 33.9%
1985 19.0% 34.3%
1986 18.5% 34.6%
1987 17.9% 36.5%
1988 17.6% 38.6%
1989 17.2% 38.5%
1990 16.7% 38.8%
1991 16.2% 38.4%
1992 16.2% 39.8%
1993 16.2% 39.5%
1994 16.1% 39.6%
1995 15.3% 40.5%
1996 14.9% 41.2%
1997 14.7% 41.7%
1998 14.2% 42.1%
1999 13.9% 42.7%
2000 13.5% 43.1%
2001 13.5% 42.2%
2002 13.3% 42.4%
2003 12.9% 42.8%
2004 12.5% 43.6%
2005 12.5% 44.9%
2006 12.0% 45.5%
2007 12.1% 45.7%
2008 12.4% 46.0%
2009 12.3% 45.5%
2010 11.9% 46.4%
2011 11.8% 46.6%
2012 11.2% 47.8%
2013 11.2% 46.7%
2014 11.1% 46.9%
2015 11.1% 47.2%
2016 10.7% 47.4%
2017 10.7% 47.9%


ChartData Download data

The data below can be saved or copied directly into Excel.

Sources: Data on union density follows the composite series found in Historical Statistics of the United States; updated to 2017 from Income inequality (share of income to top 10%) data are from Thomas Piketty and Emmanuel Saez, “Income Inequality in the United States, 1913–1998,” Quarterly Journal of Economics 118, no. 1 (2003), and updated data from the Top Income Database, updated March 2019.

Copy the code below to embed this chart on your website.

But the NLRA has fallen short of its goal. For decades, the percentage of private-sector workers covered by a union contract has steadily declined, and it now stands at just over 7% of the private-sector workforce—less than a third of what it was 40 years ago.4 The overall union membership rate is now lower than when the NLRA was first enacted.5

One of the consequences of this decline in union strength is a corresponding decline in the ability of unions in a particular sector or industry to set broad wage and benefit standards covering a large percentage of workers in that sector or industry. When unions were stronger, they were able to align the structure of collective bargaining with the corporate structure in their industry and negotiate agreements with employers that established wage standards for an entire industry.6 Union contracts established wages for unionized workers, and nonunion employers raised wages to stay competitive. In this manner, unions helped raise wages for all workers, both union and nonunion.

When union density is high, nonunion workers benefit from higher wages

When the share of workers who are union members is relatively high, as it was in 1979, wages of nonunion workers are higher. For example, had union density remained at its 1979 level, average weekly wages of nonunion men in the private sector overall would be 5% higher (that’s an additional $2,704 in earnings for year-round workers), while weekly wages of nonunion men in the private sector without a college education would be 8%, or $3,016 per year, higher.7

U.S. labor law is currently structured in a way that places significant obstacles in front of workers and unions seeking to bargain broadly with employers in their industry to set standards for their industry. Specifically, the NLRA establishes a single worksite, and at most a single employer, as the default unit for bargaining.8 Workers and unions can try to win a broader bargaining unit, such as a multi-facility bargaining unit of the same employer, but to do so they need to persuade the National Labor Relations Board (NLRB) of the appropriateness of the larger unit and organize support from a majority of employees in the bigger unit.

Current law allows employers to participate in the NLRB’s bargaining unit determinations, and employers use this process to manipulate the bargaining unit and delay and defeat union organizing drives. Employers often seek to add employees to the union’s proposed bargaining unit, not because they want to bargain with a larger unit, but to dilute the union’s support by adding employees the union has not yet organized.9

Under current law, workers and unions cannot insist that employers in their industry bargain together on a multi-employer basis with the union or a group of unions.10 Workers and unions are constrained in taking this multi-employer approach even though it would both coordinate bargaining within a sector or industry and prevent employers from pitting workers and unions at different locations against one another.

Despite the obstacles erected by the law and the problems created by anti-union employer tactics and declining union density, many unions have nevertheless been able to win and maintain bargaining that covers workers beyond an individual workplace.11 Unions have achieved this through national agreements, through pattern bargaining, through negotiating master contracts, through multi-employer bargaining, and through campaigns that use both policy changes and bargaining power to cover more workers.12 This report discusses various examples of these practices and the reasons why these practices are not as prevalent as they once were, and concludes by offering policy suggestions that would facilitate broader bargaining by giving workers more power to define the bargaining structure.

Current law of collective bargaining

When workers seek to form a union, they petition the NLRB to hold a representation election for a particular group of workers—the “bargaining unit.”13 The NLRB then reviews whether the proposed bargaining unit is “appropriate.” This analysis centers on whether workers in the proposed bargaining unit share a “community of interest,” in other words, whether they share common interests and experience at the workplace such that it is reasonable for them to bargain together with the employer over their wages, hours, and working conditions.14

The NLRB’s long-standing view is that a “single-facility” unit is presumptively appropriate.15 This rule stems from language in the NLRA describing potential bargaining units as “the employer unit, craft unit, plant unit, or subdivision thereof.”16 Typically the single facility is a single workplace, although workers and unions can seek a multi-location unit, or even a national unit, and seek to persuade the NLRB of the appropriateness of such a unit. The analysis turns on whether the work, workforce, supervision, and labor relations at the various facilities are sufficiently interrelated to justify a multi-facility unit.17 The NLRB’s bargaining unit determinations are rarely overturned.

Once the NLRB’s bargaining unit determination is made, the union must win a representation election among the employees in the unit.

The NLRA’s language has been interpreted over the years to mean that the largest possible unit that the NLRB can mandate is an employer-wide unit: a wall-to-wall bargaining unit encompassing a single employer’s employees. The courts have ruled that the NLRB does not have the authority to order a multi-employer bargaining unit, even if several similar facilities are situated near one another and the workers want to bargain together.

Employers may, if they wish, agree with workers’ request to bargain on a multi-employer basis, and there is a long history of this practice. It is voluntary on the employers’ part, but if employers have agreed to multi-employer bargaining, the NLRB will enforce this practice. With multi-employer bargaining, each participating employer agrees to designate an agent—typically an association—for purposes of collective bargaining, and each employer is then bound by the terms of the negotiated agreement. In the 1970s, an estimated 10% of private-sector workers were covered by multi-employer collective bargaining agreements.18

A leading expert has observed the “narrowness” of the single-worksite-and-single-employer approach, stating, “The NLRA, with its emphasis on firm-based organizing and bargaining, is mismatched with the globalized economy and its multiple layers of contracting.”19

In contrast to the NLRA’s bias for single-worksite bargaining units, the Railway Labor Act, which governs union representation in the railroad and airline industries, specifies that bargaining units are employer-wide, national units. Workers seeking union representation under the Railway Labor Act petition for a nationwide unit of all the employer’s workers in their particular class or craft, such as airplane mechanics, flight attendants, baggage handlers, or customer service agents. This enables each union in the railroad and airline industries—once it has won collective bargaining rights for the group through an election—to bargain nationally with a single national employer.

Not only is the current law biased in favor of single-facility bargaining units, but it also places obstacles in the way of workers and unions seeking to coordinate bargaining at multiple facilities. For example, current law restricts the ability of workers and unions to coordinate expiration dates for contracts covering different bargaining units at multiple facilities, even though common expiration dates would bring rationality and order to the bargaining process. Nor can workers picket or try to put economic pressure on a “neutral” employer other than their own in order to promote their objectives at the bargaining table—such activity is most likely to be ruled illegal as an unlawful “secondary boycott.” Finally, workers and unions are limited in their ability to bargain over the labor practices of the suppliers and contractors their employer hires to perform work. Unless these practices directly relate to the work and workers covered by the collective bargaining agreement, they are likely to be viewed as “permissive” subjects of bargaining, meaning that the employer is not legally required to bargain over them if it chooses not to.

How current law impedes broad bargaining

  • A single worksite is the default bargaining unit.
  • Workers cannot insist on multi-employer bargaining.
  • Workers cannot put economic pressure on a “neutral” employer.
  • Employers are often not required to bargain over terms of contractor and supplier employees (such bargaining may be “permissive,” i.e., voluntary on the part of the employer).
  • Workers are limited in coordinating bargaining across facilities on issues like contract expiration dates.

Examples of successful bargaining beyond a single worksite

Despite the legal impediments outlined above, workers and unions in many industries throughout the country have found ways to broaden their bargaining relationships with employers to cover multiple worksites and, in some cases, multiple employers. A variety of approaches are described below.

National bargaining

Unions can win certification of a nationwide bargaining unit of a single employer and bargain a single collective bargaining agreement covering all locations of that employer, or they can bargain nationally on a multi-employer basis. Examples of both follow.

Trucking industry

One of the most famous examples of national multi-employer bargaining that established standards for an entire industry is the Teamsters Master Freight Agreement, negotiated by Jimmy Hoffa in the 1960s.20 When it was first adopted, the Master Freight Agreement covered more than 450,000 drivers, had literally hundreds of signatory employers, and set standards for the entire trucking industry. However, the agreement has been undermined by deregulation and by consolidation in the industry. Nevertheless, the Teamsters are still able to bargain national agreements that raise standards for tens of thousands of freight industry workers. The Master Freight Agreement covers YRC Freight, Holland, and New Penn, which together employ approximately 24,000 truck drivers, dockworkers, and clerical workers at over 200 locations across the United States. In addition, the Teamsters have national agreements with ABF Freight, covering more than 8,000 workers at over 150 locations, and UPS Freight, covering 12,000 freight drivers and dockworkers. The Teamsters also represent nearly 300,000 UPS package car, air, and feeder drivers as well as loaders, sorters, and clerks across 400 UPS sites—making the contract covering these workers the largest private-sector collective bargaining agreement in the United States. In the goods transportation logistics industry, the Teamsters have national agreements with DHL covering nearly 5,000 workers at over 50 locations.21


After decades of organizing and struggle, the Communications Workers of America (CWA) achieved a national collective bargaining agreement with AT&T that allowed the union to bargain for 500,000 workers in the telecommunications industry. But subsequent deregulation and the breakup of “Ma Bell” into eight regional companies (“Baby Bells”) in the 1980s destroyed the national single-employer agreement, and nonunion companies quickly took advantage of deregulation to set up operations and undermine standards. Now CWA must bargain with AT&T for 11 individual and geographically dispersed bargaining units: six for workers in traditional wireline services, four for workers in mobile wireless services, and one focused on DSL customer service. CWA also bargains separately with the other regional bell companies, which became Verizon and CenturyLink. Today the union represents about 100,000 workers at AT&T across the country and another 50,000 at the remnants of the regional bells. All of the telecommunications operations compete with nonunion cable for the same broadband market, further eroding bargaining power.22

Paper industry

The United Steelworkers have used their union’s density and strength to build a national bargaining relationship with International Paper (IP). In the past, the union’s relationship with IP was contentious, and bargaining was fragmented among many different regions and locals, but the union now bargains with IP over two national agreements setting wages and benefits. One agreement covers 5,800 workers at 17 paper mills, and the other agreement covers 4,700 workers at 55 box plants around the country. Site-specific issues are then bargained at the local level. The union represents workers at about 70% of IP’s mills and 60% of IP’s box plants. In contrast, the union represents workers at only four of 18 Kimberly Clark facilities, and the union has not yet been able to win national bargaining at Kimberly Clark.23


Under the Railway Labor Act, unions that have won an election and collective bargaining rights for a craft or class of employees bargain a national agreement for that national bargaining unit with a railroad or airline. The unions are sometimes able to set standards that other employers follow. For example, the collective bargaining agreements reached by unions on the freight railroads (class 1), where there is high union density, set a pattern for unionized commuter rail and smaller railroads.24

Coordinated national bargaining

Unions can join together in a coalition to bargain with a nationwide employer on behalf of different groups of employees.

Manufacturing (General Electric)

For decades, a coalition of unions has bargained with General Electric through a coordinated bargaining council. In the 1980s, this bargaining covered 40,000‒50,000 GE workers, and the collective bargaining agreement set a pattern that would apply to other manufacturers of electrical appliances, equipment, and component suppliers. Because of corporate changes, downsizing, and loss of jobs to outsourcing and trade, currently only approximately 6,600 workers are covered by collective bargaining agreements at GE.

Pattern bargaining

Under pattern bargaining, a union will bargain with an initial employer to reach an agreement that then becomes the pattern for subsequent agreements with other employers in the industry.

Automakers GM, Ford, and Fiat-Chrysler

Perhaps the most well-known example of pattern bargaining involves the United Auto Workers union and General Motors, Ford, and Fiat-Chrysler. The collective bargaining agreements negotiated through bargaining with the three automakers not only set wages and benefits for tens of thousands of autoworkers, but also have an impact on nonunion auto manufacturers, which try to keep their wages competitive with their unionized counterparts.25 Under the UAW’s approach to pattern bargaining, the union bargains with all three auto companies and then picks a bargaining target to set the standard. In 2019, the target was General Motors. After failing to reach an agreement by the expiration of the contract, 49,000 GM workers went on strike for six weeks in the fall of 2019 in an effort to win a better agreement from GM. As a result of the strike, workers protected their health care benefits and won a substantial pay raise in the form of base wage increases and an accelerated progression to top pay for newer workers.26 The agreement set the pattern for the industry, and the UAW was able to reach an agreement with Ford very quickly after the GM settlement, and later reached an agreement with Fiat-Chrysler.

A challenge faced by the UAW (and other unions in their respective industries) is when employers try to evade the terms of the collective bargaining agreement by establishing new operations outside the scope of the agreement. For example, GM formed a new company, Lordstown Motors, with partner LG Chem, to build a new facility next to the existing Lordstown, Ohio, plant, rather than doing the work at the Lordstown plant under the UAW‒GM agreement.27


UNITE HERE, the hotel and hospitality industry union, has developed pattern bargaining with major hotel chains that approaches national bargaining. The benefits of this approach are evident from what UNITE HERE members were able to achieve at the Marriott Corporation in the fall of 2018. Seven different UNITE HERE locals in seven locations—Detroit, Boston, San Francisco, Oakland, San Jose, San Diego, and Hawaii—bargained separate contracts, but because the prior collective bargaining agreements expired at the same time, the workers had more leverage than they would have had if they had been bargaining in a single location. The workers went on strike at Marriott hotels in each of the locations, demanding better pay and benefits, under the banner of “One Job Should Be Enough.” Workers won substantial raises, improvements in their pensions, and strong protections against sexual harassment, among other achievements. The agreements covered 7,700 of UNITE HERE’s 20,000 members working for Marriott.28 Because the agreements have such broad coverage for workers at Marriott facilities across the country, the agreements set a standard for the industry, meaning that thousands more hotel workers employed by different companies in the seven locations won the same or similar improvements.


In the aerospace manufacturing and services industry, the Machinists Union bargains major national contracts with Boeing and Lockheed Martin that establish standards for the industry that the Machinists seek to achieve at other manufacturers.29

Master contracts

Unions will sometimes negotiate a master contract with an employer or an employer association and then insist that newly organized employers sign on to the master agreement rather than negotiating an individual agreement. This arrangement is commonplace in the construction and entertainment industries, but is also used by other unions in other industries. (See examples of multi-employer bargaining below.)

Multi-employer bargaining

Unions in several industries have bargaining relationships with groups or associations of employers—an arrangement that allows them to negotiate wage and benefit standards across an industry or geographic area. In addition to the Teamsters’ Master Freight Agreement described above, examples from other industries are outlined below.

Building trades

The industry with the most widespread practice of multi-employer bargaining is the construction industry. By long-standing practice and tradition, unions in the building trades bargain multi-employer master agreements with employer associations in their craft, and employers, including newly organized employers, adopt the master agreement often by signing letters of assent agreeing to be bound by the master agreement. Most of these agreements are bargained locally or regionally by the individual construction unions with their counterpart employer associations. For example, locals of the International Brotherhood of Electrical Workers (IBEW) bargain master contracts with local chapters of the National Electrical Contractors Association (NECA). Bargaining is streamlined by the national IBEW and NECA through the development of standard contract language that is approved by both national organizations.30

In addition, the national building trades unions, through their trade department at the AFL-CIO, bargain national agreements with contractors for both construction and maintenance projects. For example, currently there are national building trades agreements covering 301 maintenance projects in 34 states involving 118 signatory employers.31

Commercial cleaners

Local 32BJ of the Service Employees International Union (SEIU) offers a compelling example of what workers and their unions are able to accomplish when they have density and bargaining power. The union—which represents workers in 12 states and Washington, D.C.—achieves major gains for workers in the property services industry through a combination of multi-employer bargaining, group bargaining, bargaining master contracts, and identifying policy levers to facilitate bargaining.

Recently, the union completed negotiations for approximately 75,000 commercial building cleaners up and down the East Coast. The New York City agreement alone covers 22,000 commercial cleaners.32 The union bargains with a multi-employer association or with groups of employers, and its agreements bind the signatory employers in all cities where the union has local agreements. In other words, by way of example, in the New York City agreement, employers in New York City agree to abide by the collective bargaining agreement in Philadelphia if they have operations in Philadelphia. In the recent bargaining, SEIU Local 32BJ won substantial wage increases, improvements in pensions, new protections against sexual harassment, and more. Employers also agreed to a union recognition process for cleaners in Miami, opening the door to extending collective bargaining protections to another 1,500 building cleaners in that city.

Grocery workers

The United Food and Commercial Workers International Union (UFCW) engages in multi-employer bargaining with the major grocery chains in Southern California. Bargaining used to cover more grocers, but because of mergers in the industry, only two major chains—Ralphs and Albertsons—now participate in the bargaining. In the fall of 2019, the union was able to reach an agreement covering 46,000 workers at more than 500 stores. The agreement provided for wage increases, preserved health care benefits, guaranteed more hours, and helped close the wage differential between job classifications.33 While only two major grocers were at the bargaining table, the collective bargaining agreement set a standard, and other local grocery chains—including Gelson’s, Stater Bros., and Super A Foods—have signed on to bargaining contracts with their workers that have comparable or better terms. A challenge for the union is when unionized grocers enter into partnerships and other business arrangements with new entities and use them to erode bargaining unit work—by, for example, contracting out work that would be done by bargaining unit members to companies like Instacart—or when unionized grocers create lower labor standards for chains in food deserts, as Kroger did with its Food 4 Less subsidiary.

Auto dealerships and auto repair shops

The Machinists Union has several regional multi-employer agreements covering hundreds of auto dealerships and auto repair shops in Chicago, San Francisco, New York, and other cities. These agreements set wages and benefits for thousands of employees.34

Food canneries

The Teamsters have a long-established multi-employer bargaining relationship with the Cannery Council, an association of food processors with operations in central California, including Del Monte and Heinz. The most recent collective bargaining agreement raised wages by more than 10% for the 12,500 workers under the agreement.35 With corporate consolidations and more automation in the industry, the Cannery Council agreement covers far fewer employers and workers than it once did. At one point, the agreement covered as many as 50,000 workers, but it now covers only about 25% of that number.

‘Supply chain’ bargaining

Under current law, workers and unions are limited in their ability to insist that their employer bargain with them over terms and conditions of employment for the employees of their employer’s suppliers and subcontractors. The current legal definition of “joint employer” is too narrow to bring employers together at the bargaining table, and employers are typically unwilling to bargain with their unions about the employment terms of their contractors.36 Nevertheless, there are exceptions. For example, the Machinists Union has negotiated with both a government contractor and subcontractor at the table and won agreements that cover employees of both employers. This approach is more efficient than bargaining separate agreements with two companies that are operating at the same facility, and it establishes common standards for the contractor’s and subcontractor’s employees.37

Using policy levers to facilitate bargaining

Unions have also been able to win better working terms and conditions for workers through campaigns for local city ordinances.

Airport services

In Philadelphia, SEIU Local 32BJ lobbied for, and won, passage of ordinances establishing a minimum wage and paid sick days for employees of contractors at the Philadelphia International Airport. The union then won recognition as the representative of 1,400 Prospect Airport Services and PrimeFlight Aviation Services employees who work as baggage handlers, wheelchair attendants, cabin cleaners, and more. The union was able to build on the minimum standards established by the ordinance and, in their first collective bargaining agreement, win provisions that exceed the requirements of the paid sick day ordinance.38

Domestic employees

Unless they work for an agency, domestic employees (such as nannies, house cleaners, and gardeners) are not covered by the NLRA, and their employment is dispersed throughout millions of individual households. Recently, worker advocates in Seattle won passage of a city ordinance that sets a minimum wage, meal breaks, and days off for domestic workers and establishes a Domestic Workers Standards Board, through which employers, domestic workers, and worker organizations meet to discuss other recommendations and standards for domestic workers.39 An estimated 33,000 domestic workers are covered by the law.40

Policy changes that would facilitate broader bargaining

As the examples listed above have demonstrated, when unionized workers have significant density within an industry, occupation, sector, or employer, they can overcome the obstacles to broader-than-single-worksite bargaining and win significant gains at the bargaining table with their employers—gains that not only benefit workers directly covered by the collective bargaining agreement, but also raise wages and set standards for nonunion workers in the area. The following national labor law reforms would strengthen workers’ bargaining power and enable them to bargain and set standards more broadly in their occupation, sector, or industry.

Passing the PRO Act

The Protecting the Right to Organize (PRO) Act removes obstacles to workers organizing, curtails employer interference in worker organizing, and establishes meaningful penalties when employers break the law.41 The PRO Act reins in employer efforts to gerrymander bargaining units to undermine union organizing drives by keeping employers out of the representation process entirely.42 The PRO Act contains a strong joint-employer standard43 that would enable workers and unions to bring the relevant employers to the bargaining table. It establishes a process for newly formed unions and employers to successfully negotiate a first agreement. It removes prohibitions against secondary strikes and boycotts—allowing workers to put economic pressure on a “neutral” employer, an employer other than their own. These measures and others in the PRO Act would meaningfully strengthen workers’ ability to form unions, bargain with their employers, and pursue the broader bargaining models outlined above.

Changing the law to give workers the power to designate multi-employer bargaining units and multi-union bargaining

The NLRA should be amended to allow workers to designate a multi-employer bargaining unit, or to tie several bargaining units together in multi-employer bargaining, with one or more unions. This bargaining could be either horizontal (within an industry) or vertical (to capture the supply chain). Currently multi-employer bargaining is at the employer’s option: Workers, unions, and the NLRB have no ability to insist on this format, even when it makes the most sense. The voluntary nature of multi-employer bargaining allows employers to pit workers and unions in one location against one another. The law should be changed to give workers and unions the ability to request multi-employer bargaining, with direction given to the NLRB to approve the request unless there are compelling reasons why the approach should not be followed.

Changing the law to facilitate coordinated (multi-union) bargaining

The NLRA should be amended to make clear that workers and unions in a common sector or industry may coordinate and insist on key contract terms—such as the term of the collective bargaining agreement, terms for the employer’s use of subcontractors, etc.—that help them build power within their sector or industry.

Changing the law to facilitate extension of contracts to new groups of workers

The NLRA could be amended to add provisions for extending the terms of a collective bargaining agreement to cover a group of workers newly organized by a union that has density in the industry.44 An example of this kind of extension is that provided under the Baigent-Ready Proposal, named for two special advisers to the British Columbia Minister of Labour. Under the proposal, a union in a sector (defined as a geographic area with similar enterprises doing similar work) with low union density would have the opportunity to seek certification for a multi-employer unit in the sector if the union could demonstrate support from at least 45% of workers at each location within the proposed unit. Certified unions would then file for individual elections at each worksite, and the collective bargaining agreement negotiated in the sector would automatically be extended to new facilities organized in the sector.45 This approach would facilitate extending wage and benefit standards to newly organized groups, and would save the workers, unions, and employers involved the time and expense of negotiating a new collective bargaining agreement.

Developing models for sectoral bargaining

Proposals have been advanced for a sectoral bargaining system in the United States, to assure the broadest possible collective bargaining coverage.46 Sectoral bargaining is used in many industrialized democracies, and it extends the benefits of negotiated agreements to all enterprises in a given sector.47 While the idea of sectoral bargaining in the United States is being further explored and developed, the examples outlined above show that when workers are able to form strong unions, they have the power to set standards for their industries. Policy reforms should be undertaken to facilitate this outcome.

About the authors

Celine McNicholas is Director of Government Affairs at EPI. She previously served as Director of Congressional and Public Affairs at the NLRB and Labor Counsel to the House Education and Labor Committee. Lynn Rhinehart is a Senior Fellow at EPI. She previously served as General Counsel of the AFL-CIO, a federation of 55 national and international labor organizations.


The authors gratefully acknowledge the assistance of the following individuals who provided the examples and information outlined in this report:

Jennifer Abruzzo, Special Counsel, and George Kohl, Former Special Assistant to the President (retired), Communications Workers of America; Larry Engelstein, Secretary-Treasurer, and David Prouty, General Counsel, SEIU 32BJ; Peter Ford, General Counsel, United Food and Commercial Workers (UFCW), and Margo Feinberg, for UFCW as Attorney with Schwartz, Steinsapir, Dohrmann & Sommers of Los Angeles; Leeann Foster, International Vice President, and David Jury, General Counsel, United Steelworkers; Neil Gladstein, Director of Strategic Resources, International Association of Machinists and Aerospace Workers; Iain Gold, Director of Strategic Research, and Cassandra Ogren, Research Director, International Brotherhood of Teamsters; Jennifer Kelly, Research Director, United Auto Workers; Joe McCartin, Professor of History, Georgetown University; Richard McCracken, McCracken, Stemerman & Holsberry, for UNITE HERE; Jon Newman, Esq., Sherman Dunn, for the International Brotherhood of Electrical Workers and North America’s Building Trades Unions; Richard Resnick, Esq.; Kelly Ross, Policy Director, AFL-CIO.


1. 29 USC 151.

2. Josh Bivens et al., How Today’s Unions Help Working People: Giving Workers the Power to Improve Their Jobs and Unrig the Economy, Economic Policy Institute, August 2017.

3. Heidi Shierholz, Working People Have Been Thwarted in Their Efforts to Bargain for Better Wages by Attacks on Unions, Economic Policy Institute, August 2019.

4. Data are for 2019. See Heidi Shierholz, The Number of Workers Represented by a Union Held Steady in 2019, While Union Membership Fell, Economic Policy Institute, January 2020.

5. Heidi Shierholz, Working People Have Been Thwarted in Their Efforts to Bargain for Better Wages by Attacks on Unions, Economic Policy Institute, August 2019.

6. See Josh Bivens et al., How Today’s Unions Help Working People: Giving Workers the Power to Improve Their Jobs and Unrig the Economy, Economic Policy Institute, August 2017. The report describes examples of unions setting standards for an industry or geographic area.

7. The wage estimates are in 2013 dollars and look at what wages would have been in 2013 had union density (the share of workers in similar industries and regions who are union members) remained at its 1979 levels. See Jake Rosenfeld, Patrick Denice, and Jennifer Laird, Union Decline Lowers Wages of Nonunion Workers: The Overlooked Reason Why Wages Are Stuck and Inequality Is Growing, Economic Policy Institute, August 30, 2016.

8. See 29 USC 159(b): “The Board shall decide in each case whether…the unit appropriate for the purposes of collective bargaining shall be the employer unit, craft unit, plant unit, or subdivision thereof.”

9. Celine McNicholas et al., Unprecedented: The Trump NLRB’s Attack on Workers’ Rights, Economic Policy Institute, October 2019.

10. Pacific Metals Co., 91 NLRB 696 (1950).

11. For a discussion of the ways employers legally and illegally work to defeat union organizing and union contract negotiation efforts, see Celine McNicholas et al., Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns, Economic Policy Institute, December 2019.

12. In general terms, a national agreement is a single collective bargaining agreement between a national employer with multiple facilities (such as General Motors) and the union representing the employer’s employees. The agreement is binding on all facilities where the union represents workers. Workers can also seek to bargain nationally on a multi-employer basis, but participation by employers is voluntary. A pattern agreement is a lead agreement with an employer that establishes wages, benefits, and other terms and conditions of employment that the union then takes to other employer(s) of workers who are represented by the union in order to attempt to persuade the other employer(s) to follow the pattern. There is no legal obligation on an employer to agree to a pattern agreement, only to bargain in good faith. A master contract is an agreement negotiated between a union and an employer or a group of employers setting wages, benefits, and other terms and conditions of employment for all workers covered by the agreement—workers who may work at many facilities for many different employers (for example, construction workers). A multi-employer agreement, as its name suggests, is an agreement between a union representing employees of several different employers and those employers, which agree to bargain together and be bound by the same agreement. These various categories of bargaining may overlap; for example, a master contract may also be a multi-employer agreement or a national agreement.

13. Workers may also seek recognition directly from their employer through a process known as card-check recognition or majority sign-up. When the union is recognized in this manner by the employer, workers do not need to go through the NLRB process.

14. National Labor Relations Board, “An Outline of Law and Procedure in Representation Cases,” (August 2012), 135–136.

15. National Labor Relations Board, “An Outline of Law and Procedure in Representation Cases,” (August 2012), 134.

16. 29 USC 159(b).

17. National Labor Relations Board, “An Outline of Law and Procedure in Representation Cases,” (August 2012), 135–136.

18. David Madland, How to Promote Sectoral Bargaining in the United States, Center for American Progress Action Fund, July 2019.

19. Kate Andrias, “The New Labor Law,” Yale Law Journal 126, no. 1 (October 2016): 2–100.

20. Teamsters, “Master Freight Agreement” (web page), accessed March 5, 2020.

21. Interview with Iain Gold, director of strategic research, and Cassandra Ogren, research director, International Brotherhood of Teamsters, conducted by Lynn Rhinehart on January 14, 2020.

22. CWA President Chris Shelton, remarks made during “Sectoral Bargaining: America’s Past, Present, and Future,” a session of the Future of American Labor: Initiatives for a New Era conference, Georgetown University Law Center, February 8, 2019.

23. Interview with Leeann Foster, United Steelworkers International vice president, conducted by Lynn Rhinehart on February 11, 2020.

24. Interview with Neil Gladstein, director of strategic resources, International Association of Machinists and Aerospace Workers, conducted by Lynn Rhinehart on January 15, 2020.

25. Noam Scheiber, “Nissan Workers in Mississippi Reject Union Bid by U.A.W.,” New York Times, August 5, 2017 (reporting on a UAW organizing campaign at Nissan, and noting that “veteran workers at the plant make about $26 per hour, typically only a few dollars less than veteran workers represented by the union at the major American automakers, and well above the median wage in Mississippi”).

26. Neal E. Boudette, “G.M. Workers Approve Contract and End U.A.W. Strike,” New York Times, October 25, 2019.

27. Neal E. Boudette, “Buyer of G.M. Lordstown Plant Promises Union Work and Wages,” New York Times, November 7, 2019; Interview with Jennifer Kelly, research director, United Auto Workers, conducted by Lynn Rhinehart on January 13, 2020.

28. Karen Schwartz, “Deal in San Francisco Ends Nationwide Marriott Strikes,” New York Times, December 5, 2018.

29. Interview with Neil Gladstein, director of strategic resources, International Association of Machinists and Aerospace Workers, conducted by Lynn Rhinehart on January 15, 2020.

30. National Electrical Contractors Association (NECA), “Agreement Language and Information” (web page), accessed March 5, 2020.

31. North America’s Building Trades Unions (NABTU), “Agreements” (web page), accessed March 5, 2020.

32. SEIU 32BJ, “Strike Averted: NYC Office Cleaners Reach Tentative Agreement” (press release), December 20, 2019.

33. Russell Redman, “Grocery Workers Ratify New Contract in Southern California,” Supermarket News, September 12, 2019.

34. Interview with Neil Gladstein, director of strategic resources, International Association of Machinists and Aerospace Workers, conducted by Lynn Rhinehart on January 15, 2020.

35. Teamsters Joint Council 7, “Best Contract Ever for Cannery Council” (press release), July 25, 2018.

36. The Trump administration issued a rulemaking in February that returns to a narrow joint-employer standard that limits when a firm can be found to be a joint employer and thus share liability for violations of the NLRA. See Celine McNicholas and Heidi Shierholz, “New Joint-Employer Rule Strips Workers of Bargaining Rights” (statement), Economic Policy Institute, February 25, 2020. For how the narrower definition of joint employer constricts bargaining, see Celine McNicholas and Marni von Wilpert, The Joint Employer Standard and the National Labor Relations Board: What Is at Stake for Workers? Economic Policy Institute, May 2017.

37. It is worth noting that a supply-chain bargaining arrangement in the clothing and textile industry has a long history, dating back to the early 20th century, when suppliers (“jobbers”) entered into collective bargaining agreements with contractors and workers. See Mark Anner, Jennifer Bair, and Jeremy Blasi, “Towards Joint Liability in Global Supply Chains: Addressing the Root Causes of Labor Violations in International Subcontracting Networks,” Comparative Labor Law and Policy Journal 35, no. 1 (January 2013): 1–43.

38. SEIU 32BJ, “Mayor Kenney Announces Landmark Agreement Between 32BJ SEIU and American Airlines’ Contractors” (press release), August 30, 2017.

39. City of Seattle, “Domestic Workers” (web page),, accessed March 5, 2020.

40. Kathy White, “Seattle Domestic Workers Bill of Rights Takes Effect July 1” (blog post), Compliance Poster Company, May 9, 2019.

41. Celine McNicholas and Lynn Rhinehart, “The PRO Act: Giving Workers More Bargaining Power on the Job,” Working Economics Blog (Economic Policy Institute), May 2, 2019; Protecting the Right to Organize Act of 2019, H.R. 2474, 116th Congress (2019‒2020).

42. Celine McNicholas et al., Unlawful: U.S. Employers Are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns, Economic Policy Institute, December 2019.

43. Celine McNicholas and Heidi Shierholz, “EPI Comments Regarding the Standard for Determining Joint-Employer Status,” comments submitted to National Labor Relations Board, December 9, 2018.

44. Extension contracts are common around the world, according to the International Labour Organization. See International Labour Organization, Collective Agreements: Extending Labour Protection, ed. Susan Hayter and Jelle Visser, 2018. The Baigent-Ready Proposal described in the text is a comparatively narrow approach to contract extension.

45. Sara Slinn, “Collective Bargaining,” Changing Workplaces Review, Ontario Ministry of Labour, 2015, 84–91; British Columbia Ministry of Labour and Consumer Services, Recommendations for Labour Law Reform (Victoria, B.C., 1992), Appendix 3, 35. See also David Madland, How to Promote Sectoral Bargaining in the United States, Center for American Progress Action Fund, July 2019.

46. Sharon Block and Ben Sachs, Clean Slate for Worker Power: Building a Just Economy and Democracy, Harvard Law School Labor and Worklife Program, January 2020.

47. International Labour Organization, Collective Agreements: Extending Labour Protection, ed. Susan Hayter and Jelle Visser, 2018.