Rising authoritarianism and the need to upend the failed Southern economic development model

For generations, Southern politicians backed by powerful business interests have promoted a Southern economic development model—characterized by low wages, regressive taxation, lax environmental regulations, a weak social safety net, and vicious opposition to unions—while claiming such policies will attract business and thereby generate regional economic gains. But data actually show a grim reality. The South lags all other regions on most indicators of economic health including job growth and wages, and Southern workers and their families experience significantly higher rates of poverty than in other parts of the country (Childers 2024a).

The truth is that this Southern economic development model was never designed to benefit most Southerners; rather, it is historically rooted in efforts of white plantation owners to retain their wealth following emancipation and ensure continued access to the labor of Black people for as little compensation as possible (Childers 2025). Foundational to these efforts was an authoritarian approach to state governance that suppressed popular democracy and worker organizing—an approach that also sanctioned prison labor, sharecropping, a century of Jim Crow laws, lynching, and other forms of state-sponsored terror and exploitation. Until partially challenged by federal legal and policy interventions won by post-WWII civil rights movements, many Southern states for decades held elections that served merely to provide a cover of legitimacy to one-party rule of white, wealthy elites—functionally excluding Black voters from the electorate and blocking working-class constituencies from any meaningful participation in governance (Mickey 2015; Perez 2024; Mast 2025).

Today, the Trump administration’s increasingly authoritarian actions echo this troubling Southern history. At their foundation, the administration’s approaches to bypassing constitutional checks and balances—while rolling back civil rights, worker rights, and environmental protections; terrorizing immigrant communities; deploying military troops in U.S. cities; and attempting to engineer election outcomes via gerrymandering and other forms of voter suppression—are rooted in authoritarian models developed and tested in the U.S. South, and that Black, brown, and immigrant communities across the country are no stranger to.

Recent attempts to terminate federal employee collective bargaining agreements, for example, are familiar to public employees in Southern states for whom collective bargaining has long been banned or severely restricted. The Trump administration’s use of military-style policing in communities across the country echoes Southern histories of weaponizing law enforcement (or National Guard troops) to suppress organizing and instill fear, while prioritizing the expansion of the carceral state over investments in housing, education, and public services. Trump’s efforts to override the authority of state officials mirror Southern state uses of abusive preemption laws to strip policymaking authority from local governments. And administration attempts to halt clean energy investments and environmental protections threaten to repeat harms familiar in Black and brown communities in the South, where corporations have insisted on lax environmental regulations that allow them to degrade air, water, and climate quality, while profiting from the exploitation of local natural resources and labor.

Seizing opportunities to reverse decades of anti-worker, anti-democratic policymaking in the South at a moment of rising authoritarianism in the U.S. is a daunting and unavoidably urgent challenge. It will require robust new forms of multiracial organizing and labor-community coalition building across a broad set of industries in the South. Labor-community coalitions can leverage community benefits agreements (CBAs) as a powerful tool to transform economic power relations in Southern workplaces and communities. Because CBAs are private agreements between labor-community coalitions and project owners, they do not rely on government action and can therefore shape economic outcomes of major projects even in otherwise hostile political environments. CBAs have traditionally been fought for and won by labor and community groups coming together and building necessary public pressure to hold developers, corporations, and elected leaders accountable for ensuring that public investments in major new developments truly benefit workers and communities.

In this report, we analyze the potential for labor-community coalitions to pursue strong CBAs that secure significant economic benefits for Southern manufacturing workers and communities, drawing on examples of existing agreements to model potential impacts. We examine the scale of recent public investments in Southern manufacturing and examine how strong CBAs on major publicly-subsidized private projects could improve the quality of newly created construction and production jobs; open up pathways to unionization; ensure equitable hiring and training opportunities for local residents; and address community needs such as child care, affordable housing, and natural resource protection.

We contend that upending the failed Southern economic development model and the authoritarian structures that underpin it will require building new forms of labor and community power to increase union density in the South. Well-known research shows that unions promote economic equality and help workers win improvements in pay, benefits, and working conditions (Economic Policy Institute 2021). But unions also powerfully affect people’s lives outside of work. They help foster solidarity, increase democratic participation, enable working-class communities to shape economic policies affecting their lives, and serve as a counterweight to corporate power in our economy and democracy (McNicholas et al. 2025). Historically, unions have been engines of resistance to entrenched and undemocratic power—mobilizing working people to challenge inequality, defend civil rights, and push back against authoritarianism in all its forms. For all these reasons, strengthening labor-community coalitions and pathways to unionization in growing Southern industrial sectors is not just good economic policy—it is also a democratic imperative amid national authoritarian backsliding.

Worker and community power can ensure new manufacturing investments yield good jobs and community benefits

The latest wave of manufacturing growth in the South presents both opportunities and pitfalls for workers and communities. Southern states continue to lure businesses—including large manufacturing facilities—with promises of low corporate tax rates, low wages, lax regulations, and massive public subsidies. The automotive manufacturing industry has been a key recipient of public subsidies, receiving billions of dollars from Southern states in recent decades (Childers 2024a; Todd 2021). This system of low taxation and corporate giveaways starves other essential public goods, like education and social safety net programs (Mast 2025b). Likewise, weak or nonexistent environmental regulations have contributed to toxic sites and resource degradation that disproportionately affect Black and brown families, reflecting often intentional decisions to site hazardous facilities in low-income communities of color (Bergman 2019).

Some announced manufacturing projects have been cancelled or reduced in size after the Trump administration’s slashing of federal supports for strategic industries, but many projects launched during the Biden administration continue to move forward. These manufacturing investments, both in traditional industries and nascent ones such as electric vehicle (EV) and EV battery manufacturing, are spurring significant job growth in some Southern communities. Yet past experience shows that new investments and resulting jobs are unlikely to generate economic benefits for most Southerners unless local residents are able to ensure that developers and corporations respect workers’ rights, protect local natural resources, and contribute a fair share toward addressing priority community needs.

Community benefits agreements can be powerful vehicles for communities to secure lasting local economic benefits from major industrial development, at both new and existing facilities. A CBA is a legally enforceable contract between a private developer or company and a local coalition—typically made up of labor, community, faith, environmental, and other grassroots organizations—that details how a project will benefit workers and the community, and in turn how the community will support the project (including via potential public investment). Benefits spelled out in a CBA can include commitments to strong labor standards; respect for workers’ rights to organize; equitable workforce recruitment, training, and hiring practices; affordable housing; environmental protections; or a broad range of other community-identified priorities. CBAs are a well-developed model for responsible community development—so far mostly, but not entirely, in regions outside the South—and have been used for many different types of major projects including sports stadiums, events centers, manufacturing plants, airports, transit projects, and more (WRI n.d.).

CBAs can likewise mitigate risks for project developers by ensuring local project support and addressing important concerns early on, whereas failure to engage local communities in major development decisions can otherwise lead to strong community opposition, interruption of development, obstacles to obtaining necessary siting permits or rezoning approvals, or significant legal costs. In an example from June 2024, developers shelved plans for a $1.3 billion data center in Indiana after facing significant local opposition over environmental concerns (Fazili et al. 2025).

Because a CBA is a private, legally binding agreement, it does not require government action and can be used to shape outcomes of major projects even in contexts (as in most of the South) where state legislators have preempted local governments from establishing their own job quality or environmental standards (EPI 2025a). That being said, state and local governments can still have a role in facilitating, negotiating, or enforcing community benefits. Cities like Detroit and Cleveland have ordinances requiring developers of projects using public resources to engage in a community benefits plan process (City of Detroit n.d.; City of Cleveland n.d.). In 2005, Atlanta passed an ordinance specifying worker and community benefits for the Beltline redevelopment (WRI 2025). However, government involvement in community benefits plans does not guarantee strong agreements on its own. A strong labor-community coalition remains essential for securing meaningful community benefits.

Another key strength of a CBA is that it can set standards across all stages of a project’s development to ensure long-term benefits for the community at large. Private developers or public entities sometimes negotiate Project Labor Agreements (PLAs) or Community Workforce Agreements (CWAs) with building trades unions and community partners to set wages, working conditions, and timelines for the construction phase of a complex development project. A CBA can be negotiated alongside a PLA to also ensure pathways to quality jobs for local residents during the operational phases of a project, including any future expansions of the facility or additions to its workforce. A CBA can also secure commitments to build affordable housing, strengthen environmental standards, and provide other benefits to the community such as child care, public parks, or other community spaces.

To be successful, a CBA must also include defined enforcement mechanisms that hold all parties to the agreement accountable. It must clearly establish the obligations of each party, metrics for measuring progress, and ongoing monitoring of compliance with the agreement’s provisions (Last 2025; PWF and CBLC 2016). If the company or the coalition fails to make good-faith efforts on the agreement’s commitments, an arbitration process is initiated. While monitoring of the agreement is an ongoing responsibility of all members of the coalition, providing a pathway for workers to organize in the operational phase of a project is of particular importance. A newly established union at the project site is well-positioned to monitor the commitments of the CBA and hold the company accountable over the long term.

Organizers and advocates should be clear-eyed that while strong CBAs can yield powerful economic outcomes, such agreements are by no means easy to win. There are generally no legal requirements for a particular company or developer to recognize or engage with a labor-community coalition, much less to agree to negotiate and implement a CBA. Building the broad-based, durable coalitions and leverage necessary to compel private interests to engage in CBA negotiations (and then to implement and enforce the terms of a CBA) is unavoidably a challenging, long-term, resource-intensive organizing project. And like any worthwhile organizing, the formation of strong, durable labor-community coalitions is itself a key outcome of successful CBA campaigns. Vastly expanding the capacity of broad-based coalitions and labor, faith, environmental, and other grassroots organizations to gradually build community and worker power in Southern communities is the most essential ingredient for transforming existing power imbalances and, ultimately, upending the failed Southern economic development model.

Indeed, recent initiatives to win CBAs in Southern states have proven so threatening to some corporate interests that they have sought to undermine them. In 2025, Tennessee Republicans passed legislation prohibiting any company that enters into a CBA from receiving state economic development funds—aiming to create obstacles to replication of a highly successful CBA covering Nashville’s soccer stadium, and to discourage a coalition of West Tennessee residents and allied groups calling on Ford and SK Innovation to negotiate a CBA covering its massive BlueOval electric vehicle and battery manufacturing complex (Abrams 2025). In Tennessee and elsewhere, however, labor-community coalitions are nonetheless continuing to organize to ensure that massive, publicly subsidized new facilities yield good jobs and community benefits.

A new wave of Southern manufacturing is an opportunity to transform working conditions in growing industries—and across the South

Growth in Southern manufacturing industries presents a significant opportunity for labor-community coalitions to shape labor standards and community benefits in new plants and facilities—and to shape economic outcomes for generations of Southern workers to come. In recent years, the South has seen a wave of manufacturing investments. Between 2017 and 2023, manufacturing construction doubled in the East South Central Census division (Alabama, Kentucky, Tennessee, and Mississippi) (O’Brien 2023). The West South Central division (Arkansas, Louisiana, Oklahoma, and Texas) has the highest amount of manufacturing construction spending of any division in the U.S. These investments are part of a long-term trend of manufacturing industries locating in the South, which in recent years was accelerated by large federal investments through the Inflation Reduction Act, Infrastructure Investment and Jobs Act, and CHIPS and Science Act. These federal investments included both direct public subsidies and tax credits to businesses that invested in key clean energy manufacturing industries such as the production of batteries, electric vehicles, solar panels, and wind energy products.

In contrast to the typical economic development approach of many Southern states, some recent federal investments have included incentives meant to encourage strong labor standards on projects receiving public funds. While the future of many of these investments (and accompanying incentives) is now uncertain, the U.S. has in the past two years experienced its largest investment in clean energy manufacturing ever, and much of that has occurred in Southern states.1 Since the third quarter of 2023, more than $125 billion worth of clean energy manufacturing investments were announced across Georgia, North Carolina, South Carolina, Tennessee, Kentucky, and Texas (CET 2025). Advancing even a portion of these projects would result in thousands of jobs for Southern workers.

Independent of the future of federal support for clean energy manufacturing, the South will likely continue to be the largest manufacturing employer of all U.S. regions. Figure A shows manufacturing employment by region in the United States since 1990. While manufacturing employment overall has fallen during the last three decades, the South has retained the largest share of manufacturing employment of any region. In 2024, 35% of U.S. manufacturing employment was in the South. Furthermore, since 2010, manufacturing employment in the South has grown by 17%, the quickest growth of any region.

Figure A

Manufacturing jobs are often considered to be well-paid, benefit-providing “middle-class” jobs, but there is nothing inherent to the sector that determines their quality. Manufacturing jobs in some industries became “good jobs” thanks to relatively high levels of unionization during the mid-20th century, which improved wages, benefits, and working conditions (Bayard et al. 2024; Rhinehart and McNicholas 2020). As Figure B shows, unionization in manufacturing has fallen in all regions since 1983, but the South has almost without exception had the lowest unionization rate of any region.

Conservative Southern policymakers have long been hostile to union organizing. For example, every Southern state except Maryland and Delaware has passed anti-union so-called right-to-work (RTW) laws, which make it harder for workers to form, join, and sustain unions. Southern states like Florida and Arkansas were among the first to pass such laws in the 1940s, amid a wave of big business backlash against new federal labor laws and white supremacist campaigns to maintain racial hierarchies and suppress multiracial worker organizing. RTW laws suppress unionization rates and, as a result, have driven down wages for both union and nonunion workers alike across the South (Sherer and Gould 2025; Childers 2023).

Figure B

In 2025, Southern manufacturing had a 6.7% unionization rate—slightly below the national unionization rate for private-sector workers (6.8%). Unionization in Southern manufacturing grew by more than a percentage point between 2024 and 2025, a notable one-year reversal of the industry’s long-standing unionization decline, consistent with overall union gains in the South (McNicholas, Poydock, and Shierholz 2026). Nevertheless, Southern manufacturing’s unionization rate remains well below the Midwest’s (11.2%), the region where manufacturing is the most heavily unionized. Unions have a strong impact on job quality because they leverage worker power collectively to raise wages, win benefits like health care and retirement, and enact other meaningful workplace improvements, such as improved health and safety standards. These benefits can extend beyond unionized workers themselves, helping set standards across a workplace, and with enough density, across an industry.

As unionization declines in an industry or region, so does job quality. For instance, as unionization rates have fallen in auto manufacturing, the pay advantage for auto workers compared with the median worker has declined significantly (Barrett and Bivens 2021). Figure C demonstrates how this relationship holds across regions in 2025. Manufacturing jobs in the South have a pay advantage of 7%, the lowest of any region. Southern manufacturing workers also experience the lowest median hourly pay of any region ($24.41).2

Figure C

The Southern economic development model clearly hurts the region’s workers by denying them their right to organize and suppressing their wages, but there are harmful spillover effects for their communities as well. Corporate tax breaks with no strings attached provide billions of dollars to corporations that could otherwise be used to invest in schools and other essential government services. These types of tax breaks might be worthy of consideration if manufacturing employers were required to create high-quality jobs for local workers and make long-term investments in local community development needs (i.e., housing, infrastructure, education, etc.). Without such protections, they are simply taxpayer-funded giveaways that often drain the very resources needed to develop the local workforce recruited by large new facilities.

Southern states enact little to no regulation of workplace safety or environmental pollution. This results in unsafe workplaces with greater levels of injury and death (Childers 2024a). Environmental pollution from manufacturing sites can negatively affect public health by contaminating water, air, and soil. New manufacturing investments also can mean significant changes to the demand for housing in a community. A new plant or factory can drive up the cost of living for nearby residents without yielding any economic benefits to a local community. Labor, community, and environmental groups need to collaborate on shared solutions to effectively address these intertwined challenges.

Labor-community coalitions can obtain commitments that ensure “economic development” means shared prosperity for all

Labor-community coalitions organizing around manufacturing projects can secure commitments that offer direct economic benefits to workers and communities, while also establishing groundwork for the growth of worker and community power in the area. While a campaign to win a CBA can be the impetus for forming a local labor-community coalition, the alignment and relationships built through this shared work can lead to longer-term, sustainable coalitions capable of transforming local and state power relationships.

The following section analyzes a set of commitments that can be included in a CBA for a manufacturing project. The CBA framework is flexible and allows for the inclusion of many different types of commitments prioritized by particular groups of workers, community members, and environmental groups. This report focuses on key types of commitments including union neutrality agreements, living wage floors, equitable workforce development practices (such as local or targeted hire policies and programs to expand pathways to apprenticeship training), affordable housing provisions, child care benefits, and environmental protections. Each type of commitment is analyzed in terms of its economic impacts and effectiveness in reshaping local economic development to ensure that public investments generate broadly shared community benefits.

The construction phase and Project Labor Agreements (PLA)

This report mostly focuses on community benefits for workers during the operational phase of a manufacturing plant. Nevertheless, it is just as vital to set high labor standards during the construction phase. Strong community benefits agreements are ideally developed in tandem with strong project construction labor standards set via project labor agreements (PLAs). A PLA is a multiparty agreement between a project owner and a coalition of labor unions that sets out labor standards and dispute resolution procedures to promote stability and efficiency on complex infrastructure projects while also ensuring the project will generate good jobs. PLAs ensure that construction projects run smoothly, are safer, and pay workers fairly (Mangundayao, McNicholas, and Poydock 2022). By setting negotiated wage and benefit levels for each type of work on a project, PLAs level the playing field in highly competitive construction bidding processes; they ensure that contractors base bids on their ability to deliver on quality and efficiency, rather than low-ball cost estimates that reflect intent to pay substandard wages or cut corners on safety. By standardizing wage and benefit levels and taking them out of the competition in the bidding process, PLAs incentivize the use of skilled union labor, which is 14% more productive than nonunionized construction work (McFadden, Santosh, and Shetty 2022). PLAs typically set wages, fringe benefits, and working conditions but can also include requirements to utilize certain numbers of apprentices, hire locally or from certain target worker populations, and/or provide child care or other benefits that open up pathways to good union construction jobs for members of underrepresented groups.

Several of the types of standards for construction workers typically included in a PLA have analogous labor standards in the operational phase. For instance, a CBA can secure commitments for local or targeted hiring and the development of registered apprenticeship programs in a manufacturing facility, extending equitable recruitment and high-quality training requirements that a PLA typically sets for construction into the operational phase of a project.

Removing obstacles to unionization: Neutrality and labor peace agreements

Protecting workers’ freedom to unionize has historically been key to turning manufacturing jobs into good jobs. This remains just as true today. However, like workers across the country, Southern manufacturing workers continue to face formidable obstacles—including weak labor laws, powerful anti-union corporations, and hostile politicians—to exercising their legally protected rights to form or join a union. Employers are charged with violating federal labor law in more than 40% of union elections and spend more than $400 million a year on “union avoidance” consultants (McNicholas et al. 2019; McNicholas et al. 2023). Because existing weak labor laws do not effectively deter employers from union busting, these tactics are treated by many employers as a normal cost of doing business—stacking the deck unfairly against workers seeking to exercise their rights to organize and collectively bargain.

Union neutrality agreements can help safeguard workers’ right to form unions free of the types of interference employers often deploy. Under a neutrality agreement, an employer agrees to remain “neutral” and not interfere with workers’ decisions on whether to unionize. Such agreements typically include joint commitments to a “card check” process for verifying whether a majority of employees have indicated interest in forming a union. Unions and employers sometimes also enter into a labor peace agreement, where unions agree not to engage in certain types of picketing, work stoppages, or other economic disruptions during the organizing process in exchange for employer neutrality.

Employers can also choose to commit to union neutrality as a matter of principle or company policy. Union neutrality—providing workers a more free and fair choice to decide whether to unionize—has been a key component of successful unionization drives in Southern manufacturing. To take two recent examples:

• In 2024, workers at the Volkswagen (VW) Chattanooga plant voted to join the United Auto Workers. Like many European corporations, the German-based VW has an established policy of maintaining neutrality in union election processes, although workers still voiced concerns that in its U.S. facilities, VW management tried to intimidate and dissuade workers from forming a union (Bomey 2024).
• In tandem with community benefits agreement negotiations with New Flyer in Anniston, Alabama, the United Steel Workers and Communications Workers of America negotiated three neutrality agreements with New Flyer and its subsidiaries in 2022. Over the two years that followed, these union neutrality agreements enabled workers to pursue five successful union drives, including at the New Flyer facility in Alabama (Last 2025; Sasha 2024).

Successful recent instances of union organizing in Southern manufacturing facilities have been powerful enough to generate their own backlash. Because of the threat that union neutrality agreements represent to the reigning Southern economic development model, several conservative state legislatures in the South have used model legislation developed by the American Legislative Exchange Council to pass laws intended to interfere with these agreements (Sachs 2024). While the legality of such measures remains in question and has not yet been tested, Alabama, Tennessee, and Georgia now all have legislation in place stating that employers who agree to a union neutrality agreement will be barred from receiving state economic development funds, disincentivizing companies from participating in these agreements (Stephenson 2024).

Importance of unionization to improve manufacturing jobs and wages

Securing unionization in Southern manufacturing can have significant wage benefits for workers. Unionized manufacturing jobs are more likely to provide family-sustaining wages. Unionization in manufacturing is associated with a 17.9% wage premium for workers (Scott et al. 2022). This means that compared with similar workers in terms of education, occupation, experience, race, and ethnicity, unionized manufacturing workers are paid almost a fifth more per hour than their nonunionized counterparts.

Table 1 translates this union premium into how much more unionized workers in the South could make on an hourly, annual, and plant-wide basis. The average nonunionized manufacturing worker in the South earns $34.50 an hour, so with the typical union premium, that worker would be earning an additional $6.18 an hour. If that worker works full time, year-round, the hourly premium translates to $12,846 more a year. To illustrate the potential impact of unionization in an entire plant, we take the example of the BlueOval auto manufacturing investment in Tennessee, which is projected to create 6,000 jobs (TN Office of Governor 2023). For a plant of that size, unionization could mean more than $77 million in additional wages for workers.

Table 1

Wage gains from successful unionization are not hypothetical for manufacturing workers in the South. For example, in 2024, workers at New Flyer in Anniston, Alabama, ratified a union contract with significant pay raises, with some workers gaining raises of up to 38% through 2026 (CWA 2024). Establishing a union contract with transparent pay ladders will also help New Flyer workers combat persistent pay gaps between white and Black workers in Anniston’s manufacturing industry (Erickson 2021).

The benefits of unionization go far beyond hourly wage increases. The workers at New Flyer also achieved significant gains in terms of vacation time and retirement contributions. Unionized workers secure critical benefits like health care and sick days at greater rates than their nonunion peers. Adjusting for differences in industry, sector, and region, union workers are 18.3% more likely to have employer-covered health insurance than their nonunion counterparts (EPI 2021). Almost 9 in 10 private-sector union workers have paid sick days, compared with less than three-fourths of nonunion private-sector workers (EPI 2021).

Unions also contribute to safer and healthier working conditions across a wide range of industries (Dean, McCallum, and Venkataramani 2022). By strengthening workers’ voice on the job, unions empower workers to report safety issues and demand better protocols. One example of this is that unionized construction sites experience significantly lower rates of Occupational Safety and Health Administration (OSHA) violations than nonunionized sites (Manzo IV, Jekot, and Bruno 2021). This is despite the fact that unionized workplaces actually experience greater rates of OSHA inspections than other workplaces, likely because many unions maintain active health and safety committees and because unionized workers have greater access to education on how to recognize safety hazards and are less afraid of reprisals from their employer for reporting them (Leigh and Chakalov 2021).

As the New Flyer agreement demonstrates, a strong CBA includes (or is negotiated in tandem with) union neutrality commitments ensuring that workers have a free and fair choice to unionize, without employer interference or retaliation. Securing a pathway to unionization can provide direct benefits to workers at a particular facility, while also increasing local organizing capacity and coalition strength for future negotiations over new projects and local development decisions. Not only is a new union a legally recognized institution that can monitor and hold the company accountable for commitments in the CBA, but it can also play a critical role in amplifying demands of workers and communities outside of the workplace and building power for working people more broadly.

Living wage floor

CBAs can also include commitments to minimum wage floors for the workers who will operate a new facility. For example, the 2018 Nashville Soccer CBA in Tennessee included a commitment to an hourly wage of at least $15.50 for stadium workers (SUN 2018). This provision set the stadium’s wage floor well above the minimum wage in Nashville, where workers—like all Tennessee workers and many across the South—are otherwise subject to the federal minimum wage of $7.25 an hour.

If a wage floor set by a CBA is high enough, it can help workers achieve a living wage in the place that they live. What constitutes a living wage must be determined by labor and community partners (Gould, Mokhiber, and DeCourcy 2024). For example, a living wage could be defined narrowly as covering the necessities for a single adult, or more broadly as including the needs of a working parent and their children. A living wage target must also make assumptions about nonwage income such as health care benefits and government transfers. Manufacturing workers in the South can also rightfully seek wages that not only cover bare necessities but provide the family-sustaining resources needed to be healthy and thrive.

Figure D shows the share of manufacturing workers in the South earning less than $30 an hour, or $62,400 a year in wages for a full-time worker. More than 3 in 5 (60.8%) manufacturing workers in the region earn less than $30 an hour. Around 80% of Southern Black and Hispanic manufacturing workers earn below the $30 threshold. Women in manufacturing are also more likely to earn below $30 an hour (71.8%) than men (59.1%).

Figure D

A $30 wage floor exceeds the minimum costs for a single adult in most jurisdictions in the U.S., but still barely covers needs for many families with children in manufacturing-dense counties nationwide. EPI’s Family Budget Calculator estimates living wage standards by county that cover modest but necessary costs families face like food, rent, and transportation in the United States. Table 2 shows three Southern counties with significant clean energy manufacturing investments in recent years (CET 2025). Each county has significant manufacturing employment, exceeding the U.S. average for manufacturing employment density. For each county, living wage standards from the Family Budget Calculator are listed for different family types. In Morgan County, Georgia, and Maury County, Tennessee, a single adult with a child must earn at least $30 an hour to cover basic needs. For a single economic provider to cover the costs of a four-person family, they must earn over $35 an hour in all the counties listed. These living wage standards indicate that a $30 wage floor would provide significant economic security for workers with smaller families or multiple wage-earners.

Table 2

A CBA that secures a strong living wage standard in a manufacturing facility can create a virtuous cycle that brings about greater prosperity in the area. Higher wages for low- and middle-income workers boost spending in the local economy because these workers spend a greater share of their paycheck than high-income workers (Anderson 2014). Other employers in the area might have to raise their wages to compete for workers with the CBA-bound employer. The establishment of a living wage also demonstrates to other workers in the area that higher wages are a feasible goal through collective action.

Local and/or targeted hire policies

Local and targeted hiring refers to policies that prioritize recruitment of individuals from the local community, or workers from specific groups who are otherwise underrepresented in a given workforce relative to local population demographics, such as women, people of color, veterans, low-income workers, formerly incarcerated workers, or workers with disabilities (Lawliss, Finfer, and Sherer 2022). A local hire policy can require that a certain percentage of hours worked on a project be completed by local workers. These policies can also require giving local workers the first option to apply for jobs on a project. For the prosperity created through manufacturing investments in the South to be shared equitably, it is important that local community members have access to the jobs that are created during both the construction and operation phases of a development. Workforce policies also should be designed to remove barriers to employment for groups of workers—especially workers of color and women—who have historically been excluded from many construction and manufacturing career opportunities. Increasing access to these well-paying jobs can increase economic mobility for workers with more limited opportunities.

Despite these benefits, some state policymakers have been hostile to local hire as a public policy. In 2015, Nashville voters passed a ballot initiative that required city-funded construction projects to dedicate 40% of construction hours to Nashville residents, with 25% of those hours going to low-income Nashville residents (Blair et al. 2020). The Tennessee state legislature then quickly passed a bill that preempted the city from creating its own local hire policy.

As Figure E shows, the harm of Tennessee’s preemption of local hire falls disproportionately on workers of color. The construction workforce in the Nashville metro area has a higher share of workers of color and immigrant workers compared with the state construction workforce overall. Black workers are 8.2% of the construction workforce in Davidson County, but 5.5% of the overall state workforce. More than half (51.5%) of construction workers in Davidson County are Hispanic, compared with less than a quarter (20.1%) of the state overall. Davidson County construction workers are also more than twice as likely to be immigrants (40.2%) than in all of Tennessee (14.8%). State preemption of local hire prevented Nashville from ensuring that public spending would benefit local workers. However, private agreements like CBAs offer an opportunity to incorporate local hire and/or targeted hire requirements into publicly subsidized developments, even in heavily preempted jurisdictions.

Figure E

In 2018, three years after the preemption of Nashville’s local hire policy, the labor-community coalition Stand Up Nashville was able to leverage $275 million in public subsidies for a new professional soccer stadium into a successful CBA (SUN 2018). The Nashville Soccer CBA included commitments to local hire for stadium workers, particularly workers from “Promise Zones,” i.e., high-poverty areas with fewer economic opportunities (SUN 2020). Through the CBA, Nashville Soccer Holding, LLC agreed to consider qualified Promise Zone resident referrals for jobs at the stadium. So far, the program has succeeded in hiring Promise Zone residents. In 2023, Nashville Soccer Club had hired 180 employees, 80 of whom were residents of Promise Zones (SUN 2023).

CBAs in the South and throughout the country are securing similar commitments to local and targeted hiring in clean energy and manufacturing investments. In Alabama, the New Flyer CBA commits the company to ensuring that at least 45% of new hires and 20% of promotions are members of “Historically Disadvantaged Groups” (Sabin 2022).4 In Massachusetts, a new offshore wind terminal entered into a CBA with the City of Salem—setting targets for hiring of local workers, workers of color, and women workers (Sabin 2024). The CBA for Maine Aqua Ventis, an offshore wind facility, includes local hiring opportunities for residents of Monhegan, Maine (Sabin 2017). 

These types of agreements help ensure that local residents benefit from large investments in their communities, particularly when policymakers have invested public dollars in the form of tax breaks or corporate subsidies to support a new facility. Ensuring local workers are prioritized in training programs and hiring processes for newly created jobs also helps community members stay in the area when housing costs are driven up by a large new manufacturing investment. And in the longer term, providing pathways for local workers to benefit directly from these investments strengthens the labor and community alliances needed to hold developers and corporations accountable over time.

Equitable workforce development through apprenticeships and pre-apprenticeships

In addition to local hire policies, which help create equitable pathways for local workers to secure good jobs at a manufacturing site, construction and manufacturing projects require a skilled workforce to operate safely and productively. A robust ecosystem of registered apprenticeship and pre-apprenticeship programs can help ensure both that employers find the skilled workers they need in a large new manufacturing facility, and that local workers can access pathways to newly created jobs.

Registered apprenticeship programs are training programs vetted by federal or state agencies to ensure use of high-quality, best-practice training standards and approved curriculum aligned with skills needed to succeed in a particular occupation. Registered apprenticeships combine paid on-the-job and classroom training and result in a recognized, portable credential certifying that a worker has the skills and experience necessary for a specific occupation. Pre-apprenticeship programs (also known as apprenticeship readiness programs) recruit and prepare participants for registered apprenticeships—often partnering with community organizations—to open pathways to apprenticeship for women, Black and brown youth, immigrants, workers with disabilities, or others historically excluded from skilled trades occupations. The best practice is for these apprenticeships and pre-apprenticeships to be joint programs between unions and employers, providing high-quality instruction tailored to industry needs and training that leads to placement in a high-quality job with wages, conditions, and benefits negotiated into a union contract. Often, a vital building block for successful manufacturing apprenticeship programs is the establishment of a unionized workforce at a facility.

Unlike lower-quality workforce development programs, registered apprenticeships pay workers fairly for their labor during their training—and in joint apprenticeship programs, the wages and benefits of apprentices are negotiated into a union contract and typically include scheduled increases as apprentices progress through the training program. Registered apprentices (across joint and non-joint programs) typically see their earnings increase 49% between the year before they enter the program and the year after completing it (Walton, Gardiner, and Barnow 2022). These increases in earnings are greater than for similar workers who do not enter the apprenticeship during the same time period (Katz et al. 2022). Apprenticeships can also be particularly attractive to workers because they are debt-free. Most apprentices (60%) consider debt avoidance the most important reason for choosing to enroll in an apprenticeship (Walton, Gardiner, and Barnow 2022).

Apprenticeships can be a powerful tool for increasing the diversity of construction and other industry workforces. While participation of women and workers of color in apprenticeships has grown in recent years, this growth has been painfully slow for decades (CEA 2024). Research finds that union-based (joint) apprenticeship programs have been more successful than other types of apprenticeships at increasing diversity in the construction industry (Ormiston and Bilginsoy 2024). Joint apprenticeships enroll a higher share of women, Black workers, and Hispanic workers than non-joint programs, and have higher program completion rates for all workers, including for women and workers of color. Community benefits agreements can secure commitments and partnerships that equitably grow this pipeline of workers and set enforceable local and targeted hiring goals which in turn spur diversification of construction and manufacturing apprenticeship programs.

For instance, the New Flyer CBA creates a partnership between the company and coalition partners to develop pre-apprenticeship and technical training programs that expand access to manufacturing jobs for workers with low incomes and from disadvantaged groups (Sabin 2022). For these programs to succeed, community groups and educational institutions must have an active role in shaping the programs and connecting workers to these opportunities. The development of a growing skilled workforce and a robust, high-quality workforce development ecosystem can in turn be a strong incentive for bringing more facilities to an area over time. In 2015, Polaris stated that a significant factor in its decision to choose Huntsville, Alabama, for a new production facility was the area’s skilled workforce (Polaris 2015). As more workers participate in high-quality training programs that lead to union jobs, the organized workforce of the region will grow, strengthening labor-community coalitions the next time there is an opportunity to shape new development in the region.

Child care

Child care is an essential but extremely costly expense for many working families across the South. Average annual infant care costs in the South range from $6,868 in Mississippi to $14,277 in Virginia.5 The Department of Health and Human Services recommends that 7% or less of family income go toward infant child care costs, but typical Southern families spend significantly more. In Alabama, infant care costs are 9.8% of median family income, while in Oklahoma the share is 15.4% (EPI 2025b).

Increasing access to high-quality, affordable child care not only makes work more accessible to parents (and especially to women, who on average continue to assume disproportionate care responsibilities), but is a powerful investment in children’s development that can help narrow class and racial inequalities (Morrisey 2020). In addition, child care workers tend to work for very low wages and experience poverty at greater rates than the typical worker.

A large manufacturing investment in a locality might produce a significant number of jobs, and in turn increase the demand of workers and their families to live nearby. This is likely to increase the need for child care services in the region. However, data show that child care employment has not kept up with manufacturing growth in Southern counties. Table 3 compares counties with high manufacturing density, where manufacturing employment makes up more than the national average (9% in 2009), with those with lower manufacturing employment density (EPI 2025c).

Table 3

Between 2009 and 2024, manufacturing employment in high-manufacturing-density counties in the South grew 15.9%, achieving faster growth than similar counties in the U.S. overall (12.1%). However, over the same period, child care employment only grew 4.5% in Southern high-manufacturing-density counties, far below the national rate of 14.2%. Child care employment growth in the South for low-manufacturing-density counties (22.3%) is also below the national level (28.5%). The South systematically underinvests in child care, despite its importance to a healthy economy in the region.

CBAs and PLAs have been used to secure both the construction of physical child care spaces and financial support for actual services. The Nashville Soccer CBA reserved 4,000 square feet for the development of a child care center (SUN 2020). In 2001, the CBA for the North Hollywood Commons mixed-use development project in Southern California secured a commitment to an on-site child care center. Fifty child care spaces at the center were reserved for low- and moderate-income families (Sabin 2001). In the Boston area, unions have secured Project Labor Agreements that seek to address the unique child care needs of the construction industry. The PLA for the Winthrop Center in Boston established a child care access fund to research, develop, and implement alternative child care models within the construction industry, with a particular focus on assisting single mothers with child care while supporting their career (NEREJ 2019).

These types of investments are vital supports for working families, particularly mothers, seeking to balance professional and care work. Combined with union neutrality for the child care workers at these facilities, commitments to providing child care can further elevate worker power in the region and help large new facilities recruit and retain the skilled, experienced workforces they need to succeed.

Affordable housing

Without strategies to address the housing needs of a community impacted by a new manufacturing investment, local residents can experience increased economic precarity or forced displacement. The local housing impacts of a large industrial investment can be complex. A significant manufacturing investment can make a local community more attractive as workers move into the area to be close to their place of work. Manufacturing investments are also likely to be paired with prospective real estate investments in anticipation of future development around the original project. State and local governments might use eminent domain and other purchasing mechanisms to secure land for roads and other new infrastructure. These dynamics can increase housing costs for residents, particularly renters who are most vulnerable to the impacts of housing speculation and prospective rent increases. For instance, the BlueOval development in West Tennessee is already reported to have increased property prices and housing rents (TCG 2023). Homeowners, particularly those with fixed incomes, can also be more burdened with housing costs as higher demand in the area increases property tax valuations (Payne 2019).

On the other hand, extreme proximity to an industrial site can expose residents to environmental hazards and noise pollution, and may be considered unsightly, which decreases property values (Currie et al. 2016; Upton and Talpur 2024). The exact distribution of these changes in demand for housing across a community will depend on the type of industry and any other types of development included in the project.

Industrial investments like manufacturing facilities tend to take place in rural and semirural areas, in part because land is relatively inexpensive (Wiley 2015). While the counties with a higher share of manufacturing employment tend to have lower housing costs than urban areas, housing affordability remains a significant issue for workers. On average, across high-manufacturing-density counties in the South, a two-adult, two-child household must cover more than $14,000 a year in housing costs.6 A large share of renters in high-manufacturing-density counties in the South still are cost-burdened by housing, meaning they spend more than 30% of their income on rent, utilities, and other housing costs. As shown in Figure F, across the Southern states, the share of cost-burdened households in high-manufacturing-density counties ranges from 28% in Arkansas to 47% in Florida. More than 2 in 5 (42%) of Texas renters in these counties are also housing cost-burdened.

Figure F

A strong CBA will secure commitments to build a certain number of affordable housing units or dedicate a share of housing at the site as affordable. The Nashville Stadium CBA created agreements that at least 12% of residential units in the development would be affordable and that 20% of those units would be three-bedroom units to accommodate families (SUN 2020). The Staples Center CBA in Los Angeles, California, was another successful example of strong affordable housing benefits. The 2001 agreement for the development of an expanded convention center, theater, and surrounding housing, hotel, and retail space secured commitments that 20% of housing units would be affordable. The developer also agreed to provide $650,000 in interest-free loans to nonprofit affordable housing developers in the local community (WRI 2001).

Even in situations where a labor-community coalition is unable to reach a final CBA with a company, coalition organizing around community demands can still deliver meaningful affordable housing victories. Between 2002 and 2006, a labor-community coalition in Denver pressured Cherokee Investment Partners to provide community benefits as part of their redevelopment of the site of the Gates Rubber Company. The coalition leveraged zoning changes necessary for the project and a potential subsidy package from the city to extract benefits including an affordable housing plan for hundreds of rental and for-sale affordable housing units (Ingram and Hong 2011; PowerSwitch Action 2025).

In 2005, the labor-community coalition organized by Georgia STAND-UP was able to attach community benefits to an Atlanta city ordinance allocating $2 billion in public funding for the Atlanta Beltline transit-oriented development project. The city resolution shaped by the coalition established an affordable housing trust fund and a goal of developing 5,600 affordable housing units (PowerSwitch Action 2025). As of 2024, more than 4,100 affordable units have been created as part of the project (Atlanta Beltline, Inc. 2024).

Labor-community coalitions can also pursue other land-use commitments beyond the development of affordable housing. The BlueOval Good Neighbors coalition in West Tennessee has demanded commitments to protect land for farmers in the area. The development of the Ford factory has pushed Tennessee’s Department of Transportation to pursue land for new roadways through purchase and eminent domain. The area targeted for new roadways is a majority Black farming community, and several farmers are engaged in lawsuits with the state over the state’s meager compensation offers for their land (Wadhwani 2023). The coalition has demanded that farmers be offered replacement land in exchange for their sold land, as well as the creation of a 10,000-acre community land trust (BlueOval Good Neighbors n.d.).

Creating or protecting affordable housing is essential for protecting the communities that are necessary for any effective labor-community coalition. Large developments can cause instability within the community as new residents arrive, and existing residents are buffeted by rising housing costs. Because of historic and ongoing racial discrimination in housing policy, labor policy, and real estate practices, the costs of these changes are most likely to impact Black and Hispanic workers. Black families and other workers of color are the most likely to be cost-burdened by housing (JCHS 2024). Creating housing for workers and families to remain in the area is vital for continued collective action to secure benefits from developers and hold those developers accountable for their promises.

Environmental standards, funding, and monitoring

Large-scale manufacturing projects often have significant environmental impacts, both during construction and once they are in operation. Air, noise, and groundwater pollution; harm to wildlife habitats; and residents’ exposure to toxic byproducts are just a few examples of common concerns, and these consequences can be severe when projects are approved without sufficient environmental consideration. The consequences of large manufacturing projects often disproportionately harm communities of color and low-wealth areas throughout the South (Brouk 2024). For decades, poor and Black residents in the region have been exposed to toxic chemicals, pollution, and other environmental dangers at alarming rates (Bergman 2019).

In 2021, the Tennessee governor approved the construction of a General Motors lithium battery supplier in the city of Spring Hill, on the banks of the Duck River. Though the project was seen as an economic success, the plant’s operation has taken a toll on the fragile river ecosystem. The lithium battery factory is not the only strain—just eight companies along the river drain tens of millions of gallons of water daily (Wadhwani 2024). This enormous water usage has lowered river water levels, threatened biodiversity, and harmed local tourism and recreation. Advocates for the river’s health blame the state’s prioritization of manufacturing expansion without regard to the long-term environmental or economic consequences for local residents or other existing local industries.

CBAs are a tool that may help community-labor coalitions address the environmental impacts of data centers in the South. Data centers are booming across the United States, but particularly in Southern states like Georgia, Texas, and Virginia (Walker and Goldsmith 2026). New centers are heavy users of water and energy, create noise and air pollution, and are driving up electricity costs nationwide both by increasing demand for energy and requiring utilities to invest in new infrastructure paid for by all ratepayers (Merchant and Guerra 2025; Bizo et al. 2021; AI NOW 2025; Reed 2025). For example, in Virginia, electric bills were on track to increase as much as 25% in 2025 because of data centers (Penn and Weise 2025).

Growing community concerns surrounding data centers could create leverage for labor-community coalitions to pursue CBAs and other community benefits strategies. In 2025, community opposition blocked or delayed $64 billion in data center projects across the nation (Data Center Watch 2025). As community resistance to data centers continues to grow, more developers may recognize the need to come to the table with local coalitions to negotiate binding commitments on environmental and economic outcomes to secure project approvals. A handful of localities have begun to create agreements with data center developers regulating water use and securing commitments to green energy use (Turner Lee and West 2026).

Past development projects provide examples of how communities have used CBAs to secure long-term commitments to clean energy transition and protection of local natural resources in a multitude of ways, from mandating that any new construction must meet specific sustainability standards to requiring companies to contribute a set dollar amount to a city’s renewable energy transition fund. In Virginia, the City of Richmond Resort Casino CBA ensured the developing and operating company would design and construct all project buildings to Leadership in Energy and Environmental Design (LEED) Silver standards and would use previously existing pavement where possible (WRI 2021). The agreement also required the developer to attempt to reduce the urban heat island effect by planting shade trees along sidewalks and using other landscaping methods (WRI 2021). These agreements can mitigate additional environmental harm in areas that have already been polluted. A CBA between the Town of Waterloo, New York, and Seneca Meadows, Inc. regarding a landfill expansion commits the waste management company to pay for the development of new public water lines and other potable water infrastructure if existing public water wells become contaminated (WRI 2005).

CBAs can also be used to expand the positive impact of an already climate-friendly project. In New York, a CBA with an offshore windfarm developer stipulates that the company must contribute $2 million to the town of East Hampton’s Ocean Industries Sustainability Program (WRI 2018). Additionally, Deepwater Wind South Fork, LLC must spend $200,000 to establish an Energy Sustainability and Resilience Fund to support East Hampton’s transition to 100% renewable energy (WRI 2018). CBAs with environmentally focused companies provide valuable opportunities for communities looking to address climate change, especially where state governments have failed to invest in environmental programs.

A CBA can achieve a variety of climate and environmental commitments from a company but is also a strong starting point for building local capacity to monitor resource use, pollution, and other environmental priorities. A strong coalition of community, labor, and environmental groups can play essential roles in implementing and enforcing CBA commitments in contexts where understaffed government agencies have limited ability to monitor or investigate pollution and other environmental harms. Instead, workers and community members are often the first to report harmful practices and safety concerns. A strong CBA can provide opportunities for labor and environmental groups to work together to monitor and protect worker and community health, natural resources, and ecosystems.

Conclusion

For decades, Southern economic policies shaped by dominant business and corporate interests have resulted in poor working conditions and failed to ensure that profits generated by publicly subsidized development are shared with local workers and communities. Confronting the deep, long-standing imbalances of power that have entrenched this failed economic development model will require significant organizing and coalition-building to increase the collective power of workers and community members to shape different outcomes from the latest Southern manufacturing boom. Building new forms of worker and community power will be equally necessary to counter escalating authoritarian actions of the Trump administration, which closely parallel many features of the failed Southern economic development model that by design prioritizes corporations over workers and communities.

Our analysis shows that community benefits agreements could be powerful tools for Southern labor and community groups building the shared power necessary to reshape local and eventually regional economies. When strong coalitions of labor, environmental, faith-based, and other grassroots community organizations are able to build the necessary power to bring a company or developer to the table to negotiate an enforceable agreement, such coalitions can secure measurable economic benefits like higher wages, respect for workers’ rights to unionize, local or targeted hiring, protection of natural resources, or more affordable housing. Such economic gains are beneficial in themselves, but they also raise expectations, build local capacity to pursue additional gains, and demonstrate to the community at large that local residents can shape their own economic futures, and that these types of victories are achievable in the face of the Southern status quo.

While the urgent project of upending the Southern economic development model will require vigorous and persistent organizing across many sectors and geographies, community benefits agreements are one key strategy for turning manufacturing jobs into good jobs, ensuring long-term local economic gains from new industrial investments, and even renewing democracy in contexts where it has long been suppressed. Forming strong, long-lasting labor-community coalitions is essential to winning concrete gains for local workers as well as reshaping the political fabric of Southern communities and increasing working people’s influence over broader state or regional economic policy decisions. Winning and implementing any strong CBA requires the formation of an empowered labor-community coalition, which ideally endures and gains greater strength, experience, and influence over time. Just as the economic benefits of unionization extend far beyond an individual workplace, establishing a strong CBA coalition can create broader positive impacts across a community or region—delivering higher-quality jobs; more equitable tax systems; stronger public services; and healthier, more inclusive political systems.

Acknowledgements

The authors wish to thank the AFL-CIO Center for Transformational Organizing for their partnership and invaluable contributions in the production of this report. The authors are also grateful to Athena Last and Ian Elder at Jobs to Move America and Ben Beach at PowerSwitch Action for their expert feedback.

Appendix

Appendix Table 1

Notes

1. Clean energy manufacturing includes manufacturing of batteries, electric vehicles, mineral products, solar energy products, and wind energy products.

2. Workers in Southern states experience lower wages than in other regions even after adjusting for cost-of-living differences (Childers 2023).

3. The facilities covered by these agreements included plants in Alabama, California, Kentucky, Minnesota, New York, and Wisconsin.

4. This category includes workers who are Black, Indigenous, and/or people of color; women; LGBTQ+ persons; systems-impacted people (formerly incarcerated people); persons emancipated from the foster care system; residents of Anniston, Alabama, lacking GED or high school diploma; and veterans.

5. Southern states excluding D.C., Delaware, and Maryland.

6. EPI analysis of Family Budget Calculator and Quarterly Census of Employment and Wages data.

References

Abrams, Cynthia. 2025. “As West Tennessee Community Pressures Ford for Promises, Lawmakers Curb ‘Community Benefits Agreements’.” WPLN News, March 25, 2025.

AI NOW. 2025. “North Star Data Center Policy Toolkit: State and Local Policy Interventions to Stop Rampant AI Data Center Expansion.” December 2025.

Anderson, Sarah. 2014. “Wall Street Bonuses and the Minimum Wage.” Institute for Policy Studies, March 12, 2014.

Aquiles-Sanchez, Pablo, and Laura Dresser. 2022. Worker Power Levels the Playing Field: Community Benefits for Public Subsidies in the Iron District. High Road Research Center, November 2022.

Aquiles-Sanchez, Pablo, and Laura Dresser. 2024. From Community Benefits, to Collective Bargaining, and Back: Building Worker Power in Milwaukee. High Road Research Center, March 2024.

Atlanta Beltline Inc. 2024. Relentless Momentum, Undeniable Results: 2024 Annual Report.

Barrett, Jim, and Josh Bivens. 2021. The Stakes for Workers in How Policymakers Manage the Coming Shift to All-Electric Vehicles. Economic Policy Institute, September 2021.

Bayard, Kimberly, Tomaz Cajner, Vivi Gregorich, and Maria Tito. 2022. Are Manufacturing Jobs Still Good Jobs? An Exploration of the Manufacturing Wage Premium. Federal Reserve Board, March 2022.

Bergman, Megan Mayhew. 2019. “‘They Chose Us Because We Were Rural and Poor’: When Environmental Racism and Climate Change Collide“. Guardian, March 8, 2019.

Bizo, Daniel, Rhonda Ascierto, Andy Lawrence, and Jacqueline Davis. 2021. Uptime Institute Global Data Center Survey 2021. Uptime Institute, September 2021.

Blair, Hunter, David Cooper, Julia Wolfe, and Jaimie Worker. 2020. Preempting Progress: State Interference in Local Policymaking Prevents People of Color, Women, and Low-Income Workers from Making Ends Meet in the South. Economic Policy Institute, September 2020.

BlueOval Good Neighbors. n.d. “Community Benefits Agreement Demands Summary.” Accessed September 5, 2025.

Bomey, Nathan. 2024. “Volkswagen on Factory Unionization Efforts: ‘Neutral Doesn’t Mean Silent’.” Axios, February 15, 2024.

Brouk, Allison. 2024. “Wood Pellet Manufacturing in the South Harms Communities and the Environment.” Earthjustice, October 11, 2024.

Clean Economy Tracker (CET). 2025. Clean Economy Tracker. Accessed July 15, 2025.

Childers, Chandra. 2023. Rooted in Racism and Economic Exploitation: The Failed Southern Economic Development Model. Economic Policy Institute, October 2023.

Childers, Chandra. 2024a. Southern Economic Policies Undermine Job Quality for Auto Workers. Economic Policy Institute, September 2024.

Childers, Chandra. 2024b. The Evolution of the Southern Economic Development Strategy. Economic Policy Institute, May 2024.

Childers, Chandra. 2025. The Ongoing Influence of Slavery and Jim Crow Means High Poverty Rates and Low Economic Mobility in the South. Economic Policy Institute, April 2025.

City of Cleveland. n.d. “Community Benefits Ordinance” (web page). Accessed December 4, 2025.

City of Detroit. n.d. “Community Benefits Ordinance” (web page). Accessed December 4, 2025.

Communication Workers of America (CWA). 2024. “A New Dawn: Union Victory at New Flyer in Anniston, Alabama.” May 20, 2024.

Council of Economic Advisors (CEA). 2024. “All Aboard the ApprenticeSHIP: Assessing the Changing Face of Registered Apprenticeships.” Biden White House Archives, November 20, 2024.

Currie, Janet, Lucas Davis, Michael Greenstone, and Reed Walker. 2015. “Environmental Health Risks and Housing Values: Evidence from 1,600 Toxic Plant Openings and Closings.” American Economic Review 105, no. 2 (February 2015): 678–709. doi: 10.1257/aer.20121656.

Data Center Watch. 2025. “$64 Billion of Data Center Projects Have Been Blocked or Delayed Amid Local Opposition.” Accessed February 24, 2026.

Dean, Adam, Jamie McCallum, and Atheendar Venkataramani. 2022. Unions in the United States Improve Worker Safety and Lower Health Inequality. Washington Center for Equitable Growth, December 2022.

Economic Policy Institute (EPI). 2021. “Unions Help Reduce Disparities and Strengthen Our Democracy” (fact sheet). Economic Policy Institute, April 23, 2021.

Economic Policy Institute (EPI). 2025a. “Workers’ Rights Preemption in the U.S.” Accessed July 1, 2025.

Economic Policy Institute (EPI). 2025b. “Child Care Costs in the United States.” Accessed July 1, 2025.

Economic Policy Institute (EPI). 2025c. “State of Working America data.” Accessed June 25, 2025.

Economic Policy Institute (EPI). 2026. “Family Budget Calculator.” Accessed February 15, 2026.

Erickson, Emily. 2021. Are Good Jobs Possible in the Deep South? Jobs to Move America, March 2021.

Fazili, Sameera, Pronita Gupta, and Doug Bloch. 2025. “How Community Benefits Agreements Can Reduce Project Delivery Risk.” UtilityDive, October 15, 2025.

Gould, Elise, Zane Mokhiber, and Katherine DeCourcy. 2024. What Constitutes a Living Wage? Economic Policy Institute, January 2024.

Hickey, Sebastian Martinez. 2024. “Leveraging the Gulf Coast Hydrogen Hub Community Benefits Plan to Empower the Texas Workforce.” Comments submitted on behalf of Economic Policy Institute to Texas Climate Jobs Project Community Commission on Truth and Transparency in Texas Hydrogen, December 5, 2024.

Ingram, Gregory K, and Yu-Hong Hong. 2012. Value Capture and Land Policies. Lincoln Institute of Land Policy.

Joint Center for Housing Studies of Harvard University (JCHS). 2024. America’s Rental Housing. Harvard University, 2024.

Katz, Batia, Robert Lerman, Daniel Kuehn, Jessica Shakespeare. 2022. Did Apprentices Achieve Faster Earnings Growth Than Comparable Workers? Brief prepared for U.S. Department of Labor, Employment and Training Administration. Rockville, MD: Abt Associates; and Washington, DC: Urban Institute, August 2022.

Last, Athena Nicole. 2025. A Case Study of the New Flyer of America Inc. Community Benefits Agreement. Jobs to Move America, July 2025.

Lawliss, Michael, Lew Finfer, and Jennifer Sherer. 2022. Using Local and Economically-Targeted Hire to Promote Good Jobs through the Infrastructure Investment and Jobs Act. Jobs to Move America, September 2022.

Leigh, J. Paul, and Bozhidar Chakalov. 2021. “Labor Unions and Health: A Literature Review of Pathways and Outcomes in the Workplace.” Preventative Medicine Reports 24, no. 101502. https://doi.org/10.1016/j.pmedr.2021.101502

Mangundayao, Ihna, Celine McNicholas, and Margaret Poydock. 2022. “Project Labor Agreements on Federal Construction Projects Will Benefit Nearly 200,000 Workers.” Working Economics Blog (Economic Policy Institute), February 9, 2022.

Manzo IV, Frank, Michael Jekot, and Robert Bruno. 2021. The Impact of Unions on Construction Worksite Health and Safety. Illinois Economic Policy Institute, November 2021.

Mast, Nina. 2025a. Forced Prison Labor in the “Land of the Free.” Economic Policy Institute, January 2025.

Mast, Nina. 2025b. How Anti-Worker Policies, Crony Capitalism, and Privatization Keep the South Locked Out of Shared Prosperity. Economic Policy Institute, June 2025.

McFadden, Michael, Sai Santosh, and Ronit Shetty. 2022. Quantifying the Value of Union Labor in Construction Projects. Mechanical Industry Advancement Fund, December 2022.

McNicholas, Celine, Margaret Poydock, Julia Wolfe, Ben Zipperer, Gordon Lafer, and Lola Loustaunau. 2019. Unlawful: U.S. Employers are Charged with Violating Federal Law in 41.5% of All Union Election Campaigns. Economic Policy Institute, December 2019.

McNicholas, Celine, Margaret Poydock, Samantha Sanders, and Ben Zipperer. 2023. Employers Spend More Than $400 Million per Year on ‘Union-Avoidance’ Consultants To Bolster Their Union-Busting Efforts (fact sheet). Economic Policy Institute, March 2023.

McNicholas, Celine, Margaret Poydock, Heidi Shierholz, and Hilary Wething. 2025. Unions Aren’t Just Good for Workers—They Also Benefit Communities and Democracy. Economic Policy Institute, August 2025.

McNicholas, Celine, Margaret Poydock, and Heidi Shierholz. 2026. Workers’ Resolve Drives Increase in Unionization in 2025. Economic Policy Institute, February 2026.

Merchant, Zach, and Adrian P. Guerra. 2025. “Data Center Boom Impacting Georgia’s Water Resources.” 11Alive, April 14, 2025.

Mickey, Robert. Paths Out of Dixie: The Democratization of Authoritarian Enclaves in America’s Deep South, 1944-1972. Princeton, N.J.: Princeton Univ. Press, 2015.

Mishel, Lawrence. 2017. “The Increased Diversity of New York City Union Construction Employment.” Working Economics Blog (Economic Policy Institute), January 19, 2017.

Morrissey, Taryn. 2020. Addressing the Need for Affordable, High-Quality Early Childhood Care And Education for All in the United States. Washington Center for Economic Growth, February 2020.

New England Real Estate Journal (NEREJ). 2019. “MP Boston and Building Trades Unions sign Project Labor Agreement for Winthrop Center.” NEREJ, November 15, 2019.

O’Brien, Connor. 2023. “Factory Boom in the Mountain West and Upper Midwest Accelerated into the Spring.” Economic Innovation Group, June 13, 2023.

Ormiston, Russell, and Cihan Bilginsoy. 2024. “Measuring Diversity in Construction Apprenticeship Programs.” Working Economics Blog (Economic Policy Institute), November 21, 2024.

Partnership for Working Families and Community Benefits Law Center (PWF and CBLC). 2016. Common Challenges in Negotiating Community Benefits Agreements and How to Avoid Them. January 2016.

Payne, Mat. 2019. “When Nowhere Becomes Somewhere: Gentrification in Rural Communities and How Proactive Community Planning and a Progressive Property Valuation System Can Stem the Tide.” Kentucky Law Journal 207, no. 4: 727–745.

Perez, Daniel. 2015. Voter Suppression Makes the Racist and Anti-Worker Southern Model Possible. Economic Policy Institute, October 2024.

Penn, Ivan, and Karen Weise. 2025. “Big Tech’s A.I. Data Centers Are Driving Up Electricity Bills for Everyone.” New York Times, August 2025.

Polaris. 2015. “Polaris Industries Inc. Announces Expansion of Manufacturing Operations.” January 9, 2015.

PowerSwitch Action. 2025. “Community Benefits Agreements Examples.” Accessed September 1, 2025.

Reed, Rachel. 2025. “How data centers may lead to higher electricity bills.” Harvard Law Today, September 2025.

Rhinehart, Lynn, and Celine McNicholas. 2020. Collective Bargaining Beyond the Worksite: How Workers and Their Unions Build Power and Set Standards for Their Industries. Economic Policy Institute, May 2020.

Sabin Center for Climate Change Law (Sabin). 2001. North Hollywood Mixed-Use Redevelopment Project Community Benefits Agreement.

Sabin Center for Climate Change Law (Sabin). 2017. Monhegan Plantation et al. and Maine Aqua Ventis Community Benefits Agreement.

Sabin Center for Climate Change Law (Sabin). 2022. New Flyer of America, Greater Birmingham Ministries, and Jobs to Move America Community Benefits Agreement.

Sabin Center for Climate Change Law (Sabin). 2024. City of Salem and Salem Wind Terminal LLC Community Benefits Agreement.

Sachs, Benjamin. 2024. “Hey ALEC, Be Careful What You Wish For.” On Labor, March 8, 2024.

Saha, Devashree. 2024. Community Benefits Snapshot: New Flyer Community Benefits Agreement. World Resources Institute, December 2024.

Scott, Robert, Valerie Wilson, Jori Kandra, and Daniel Perez. 2022. Botched Policy Responses to Globalization Have Decimated Manufacturing Employment with Often Overlooked Costs for Black, Brown, and Other Workers of Color. Economic Policy Institute, January 2022.

Sherer, Jennnifer, and Elise Gould. 2025. It’s Time for Colorado to Remove Barriers to Unionization. Economic Policy Institute, February 2025.

Stand Up Nashville (SUN). 2018. “Historic Community Benefits Agreement Reached!” SUN, September 4, 2018.

Stand Up Nashville (SUN). 2020. Nashville MLS Soccer Community Benefits Agreement.

Stand Up Nashville (SUN). 2023. Community Advisory Committee Community Benefits Agreement Annual Report 2023.

Stephenson, Jemma. 2024. “Alabama Senate Bill Would Punish Companies That Voluntarily Recognize Unions.” Alabama Reflector, March 27, 2024.

Tennessee Office of Governor. 2023. “Gov. Lee, Ford Celebrate Historic BlueOval City in West TN” (press release). March 23, 2023.

The Chesapeake Group, Inc. (TCG). 2023. Haywood Market Assessment Section for Growth Strategies. September 2023.

Todd, Patricia. 2021. The Hidden Costs of Alabama’s Tax Incentives. Jobs to Move America, August 2021.

Turner Lee, Nicol, and Darrell West. 2026. Why Community Benefit Agreements Are Necessary for Data Centers. The Brookings Institution, January 2026.

Upton, Greg, and Sarang Talpur. 2024. Literature Review on the Impact of Utility-Scale Solar on Housing Prices. Louisiana State University, August 2024.

Wadwhani, Anita. 2023. “Black Farming Community Fights to Get Fair Deal as State Takes Land for Ford Plant Roadways.” Tennessee Lookout, April 3, 2023.

Wadhwani, Anita. 2024. “Water Wars: Groups Challenge ‘Unsustainable’ Withdrawals from Duck River.” Tennessee Lookout, May 6, 2024.

Walker, Carla, and Ian Goldsmith. 2026. “From Energy Use to Air Quality, the Many Ways Data Centers Affect US Communities.” World Resources Institute, February 2026.

Walton, Douglas, Karen Gardiner, and Burt Barnow. 2022. Expanding Apprenticeship to New Sectors and Populations. Prepared for the U.S. Department of Labor, Employment and Training Administration. Rockville, MD: Abt Associates, August 2022.

Wiley, Jonathan. 2015. The Impact of Commercial Development on Surrounding Residential Property Values. J. Mack Robinson College of Business, April 2015.

World Resource Institute (WRI). n.d. “Database of Community Benefits Frameworks Across the US.” Accessed September 5, 2025.

World Resources Institute (WRI). 2001. Staples Center Community Benefits Agreement.

World Resources Institute (WRI). 2005. Community Benefits Agreement between the Town of Waterloo and Seneca Meadows Inc.

World Resources Institute (WRI). 2018. Community Benefits Agreement between Deepwater Wind and the Town of East Hampton.

World Resources Institute (WRI). 2021. Resort Casino Host Community Agreement by and between the City of Richmond, Virginia and RVA Entertainment Holdings, LLC.

World Resources Institute (WRI). 2025. Atlanta Beltline. Accessed September 29, 2025.

Zessoules, Daniella, and Olugbenga Ajilore. 2018. Wage Gaps and Outcomes in Apprenticeship Programs. Center for American Progress, December 2018.

Chairman DeFazio, Ranking Member Graves, committee members, thank you for inviting me to talk with you today. I am Adam Hersh, Ph.D., Senior Economist at the Economic Policy Institute, a nonpartisan, 501(c)3 nonprofit think tank in Washington, DC.

Today, I will talk about 2 things:

1. Why infrastructure matters, and
2. What impacts we can see and expect from recent major legislation to expand infrastructure investment.

First, I want to recognize that this Congress, over the past 18 months, passed three monumental pieces of legislation that are critical for American transportation and infrastructure and by so doing have started America on a path to higher, more broadly shared, and more sustainable prosperity. These are the American Rescue Plan Act (ARPA), the Infrastructure Investment and Jobs Act (IIJA), and the Inflation Reduction Act (IRA).

Considering that the previous administration along with the 115th and 116th Congresses failed to advance any new infrastructure agenda—despite grandiose pledges and repeated “Infrastructure Weeks”—these three acts mark not just monumental political achievements, but also the promise to fundamentally transform the American economy and improve everyone’s quality of life.1 When these spending plans are fulfilled, you will have touched the lives of every single person in America with upgraded and expanded access to safer roads, less time spent in traffic congestion, cleaner drinking water and sanitation, modernized schools, dependable and sustainable energy grids, cleaner air and better health, lower costs of doing business, a revitalized manufacturing sector, more money in their pockets because of the investments in America this Congress has made.

Still, all you have done is far from enough. We need to be doing much more to meet the needs of our current economy and provide the foundation for America’s future prosperity and security. America faces a yawning infrastructure deficit and if we don’t rise to meet this moment, we risk being left behind economically.

1. Why infrastructure matters

Infrastructure constitutes the essential public goods at the heart of our economy that allow people, goods, and ideas to be more easily exchanged, as well as to address the costs of negative externalities arising in such complex social organization.2 This is the “truck, barter, and trade,” that Adam Smith wrote about in his book, The Wealth of Nations; infrastructure lowers the cost of moving people and goods as well as creating more opportunities to trade information and spark innovation. Policy debates often focus on infrastructure’s impact in terms of the jobs and investment that can be created today. Yes, every job created directly in infrastructure construction creates an additional 17.8 jobs in other sectors of the economy and fuels domestic manufacturing.3 But infrastructure is even more essential to creating the opportunity and productivity that propels economic activity into the future, yielding economic dividends for years to come by connecting people, goods, and information together more efficiently.

Consider your own consumption of a very basic infrastructure good like bridges. You may have crossed a dozen or more to get to this hearing today. You probably cross dozens more on every trip home for your district work periods. You probably don’t notice them passing by while you are busy on your mobile phone, but they are essential to everyday life in America and they are in trouble. A Department of Transportation survey of nearly 620,000 bridges nationally finds that 3 in 5 are in less than “good” condition, while 2 in 5 are more than 50 years old.4 Practically, this means most American bridges have reached or are approaching structural deficiency or functional obsolescence. Some states are significantly worse off than the average: In West Virginia 77% of bridges are in less than good condition, and the shares drop to 73% in Kentucky and 66% in Pennsylvania (see Table 1). Even in relatively well-situated states like Ohio and Florida, 39% and 38% of bridges, respectively, are still problematic—totaling more than 15,000 bridges.

Table 1

Bridges are just one type of critical infrastructure. There are many more that are also essential in everyday life for American families and businesses. Until now, Congress has allowed America’s infrastructure to go to rot, failed to supply it in adequate amounts, and—whether by design or malign neglect—too often forced select communities most in need of affordable transportation to bear the costs of infrastructure projects while excluding them from the benefits. These include:

• Roads and other surface transportation assets
• Drinking, waste, and irrigation water systems
• Energy generation and transmission
• Public transit, passenger rail, and airports and aviation systems
• Coastal and inland waterways and ports
• Access to high-speed internet
• Conservation and public recreation space

America has been disinvesting in infrastructure assets for years, and our growing deficiencies impose staggering economic costs.5 The American Society of Civil Engineers (ASCE) estimates that the loss of functionality from America’s depreciated infrastructure assets will cost the United States $10 trillion in GDP, 3 million jobs, and $2.4 trillion in lost exports by 2039 due to increased costs of doing business, lost time and wasted fuel, health impacts, and other individual costs that add up to a big deal in the aggregate.6

The ASCE projects that the U.S. economy will need $6 trillion in infrastructure investment, sustained over 10 years, may be too low. The aging infrastructure we have is ill-prepared to cope with increasingly frequent severe weather events, wildfires, and flooding we can expect moving forward as the climate warms—and certainly if we fail to limit that warming to 2 degrees Celsius above pre-industrial levels—which will cause it to deteriorate faster. And the ASCE analysis did not factor in additional investments that will be needed to deploy decarbonization technologies at the ambitious pace and scale needed to meet emissions targets. The U.S. Global Change Research Program estimated that, if unabated, climate change will permanently reduce U.S. GDP by 10 percent.7 Economic losses will result from harm to physical assets, reduced industrial and agricultural productivity, increased mortality and health impacts on labor force participation, and sociopolitical destabilization around the world. Other researchers estimate an additional $400–600 billion investment a year is needed to achieve carbon net neutrality.8

The good news is that you have the power to change this in ways that will yield outsized effects on the U.S. economy and the lives of families across this country. Research on the longer-term return on investment from public infrastructure finds that, on average, every $100 spent on infrastructure generates an additional $17 benefit, though some research finds a return on investment as high as 73%.9 The broad economic benefits may be even larger than these estimates suggest. Typical economic models, such as those used by the Congressional Budget Office (CBO) to score legislation, can paint a misleading picture by accounting for costs but not the full range of real-world benefits, making unrealistic assumptions about how people and markets behave, and assessing investment returns over too short a time horizon.

In the real world, infrastructure investments deliver an immediate economic surge, but also simultaneously achieve other objectives, for example:

• Expanding broadband internet access to rural and other neglected communities not only will create immediate jobs installing communications equipment, but also help bring to every corner of the country employment, education, health care, and social opportunities afforded by connectivity.
• Overhauling public water systems to eliminate lead and other toxics not only will create a lot of jobs and lower utility prices for families and businesses, but also yield lifelong impacts on educational attainment, earnings, and productivity for those living in affected communities.
• Reinvesting in and expanding sustainable public transportation systems not only will create direct jobs, but also open new opportunities for labor force participation and higher wages and productivity—connecting people to jobs that were literally out of reach—while reducing greenhouse gas emissions and improving air quality and therefore health and education outcomes.

This Congress has taken significant steps in the right direction to do this with ARPA, IIJA, and IRA, allocating new resources to these long-neglected foundations of national economic prosperity. But it is also important to note that the foundations of a dynamic and efficient economy go deeper than hard physical infrastructure assets. The pandemic “she-cession”—job losses disproportionately affecting women—has laid bare how essential caregiving “soft” infrastructure also is for the overall economy and that inadequate and unequal access to quality care has caused preventable harm to individuals, families, and on aggregate economic performance. America’s lack of paid caregiving infrastructure represents a glaring obstacle to achieving the country’s full economic potential that future Congresses must address.

2. Benefits of ARPA, IIJA, and IRA Infrastructure Measures

ARPA delivered critical resources to American state, local, tribal, and territorial governments in a time of acute crisis so that they could maintain continuity in essential public and private transportation and infrastructure services when revenue streams tanked;10 expanded and accelerated local infrastructure projects to offset demand losses in other sectors of the economy; and provided support for struggling small businesses and families to keep the lights on. Moody’s analytics found that ARPA increased employment by more than 4 million jobs and nearly doubled the rate of GDP growth in 2021, and delivered sufficient aggregate demand to ensure that the Great Lockdown of 2020 did not repeat in a double-dip recession.11

The Infrastructure Investment and Jobs Act reauthorized funding for existing infrastructure and provided nearly $550 billion in new investments in surface transportation, public transit and rail, water, and broadband internet infrastructure, along with new investments in renewable energy and electric vehicles.12 By my estimates, the additional infrastructure spending in IIJA supports 772,400 jobs annually.13 Table 2 details the number of jobs associated with each program under the IIJA, with most jobs coming from road and surface transportation projects, though all program areas provide significant job creation effects.14 Table 3 looks at the kinds of jobs that IIJA will create broadly across the economy. Of course, the construction industry accounts for the largest share of employment, with nearly 176,000 jobs annually. But the investments also stimulate 175,000 jobs annually in the manufacturing sector, nearly 100,000 jobs in transportation industries, and more broadly across other industries through induced demand. Critically, Buy America and prevailing-wage provisions in this and other legislation help set a high standard for contractors to ensure that America’s investments create good jobs with fair wages and contribute to the revitalization of our manufacturing base. Moody’s Analytics estimates that, at the peak of the expenditures, U.S. GDP will be roughly 0.8 percentage points higher as a result of IIJA.15

Table 2

Table 3

Ink is still drying on the Inflation Reduction Act, but it is clear this is the most ambitious action Congress has taken to confront the impending climate change crisis with more than 100 programs that will restore, expand, and modernize a broad range of infrastructure systems. These investments will bring the costs of transportation, electricity and other utilities, building heating and cooling for families and businesses in America. The spending will do so while supporting domestic manufacturing and development of new domestic industries where U.S. businesses will lead innovation and critical components of energy goods and systems will better insulated from disruption in global supply chains. University of Massachusetts economists estimate that IRA will generate 912,000 jobs per year, on average, for 10 years from the public and private investments that the policies incentivize.16 Rhodium Group economists estimate that IRA will reduce energy costs for the average household by $730 to $1135 annually.17

Conclusion

While this Congress has taken great strides to tackle pressing crises while reinvesting in the infrastructure at the foundation of America’s long-term economic prosperity. Achieving our economic goals will require not just significantly more resources, but embracing new approaches to the public role in how we fund and incentivize infrastructure and technological investments. Thank you.

Endnotes

1. Emily Cochrane and Eileen Sullivan, “The Many Times It’s Been ‘Infrastructure Week’ in Washington.” New York Times. April 1, 2020. Accessed September 15, 2022. https://www.nytimes.com/2020/04/01/us/politics/coronavirus-infrastructure-week-timeline.html.

2. Technically, economists define a “public good” as non-rival and non-excludable, although in common usage public goods may also refer to rival, non-excludable goods.

3. Josh L. Bivens,  Updated Employment Multipliers for the U.S. Economy. Economic Policy Institute, January 23, 2019.

4. EPI analysis of Department of Transportation, National Bridge Inventory: Bridge Condition by County 2022. June 15, 2022. Accessed September 15, 2022. https://www.fhwa.dot.gov/bridge/nbi/no10/county22.cfm; ASCE, Making the Grade: America’s Infrastructure Report Card 2021: Bridges. Accessed September 15, 2022. https://infrastructurereportcard.org/wp-content/uploads/2020/12/Bridges-2021.pdf.

5. Sarah Ayres Steinberg and Adam Hersh. “New Ryan Budget Cuts Investments in America’s Future.” Center for American Progress. March 13, 2013. https://www.americanprogress.org/article/new-ryan-budget-cuts-investments-in-americas-future/; Josh L. Bivens, “The Potential Macroeconomic Benefits from Increasing Infrastructure Investment.” Economic Policy Institute. July 18, 2017.

6. American Society of Civil Engineers. 2021 Infrastructure Report Card. 2022. https://infrastructurereportcard.org/.

7. U.S. Global Change Research Program (USGCRP). Impacts, Risks, and Adaptation in the United States: Fourth National Climate Assessment, Volume II. 2018. https://nca2018.globalchange.gov/.

8. Robert Pollin, Shouvik Chakraborty, and Jeanette Wicks-Lim, Employment Impacts of Proposed U.S. Economic Stimulus Programs: Job Creation, Job Quality, and Demographic Distribution Measures. Political Economy Research Institute, 2021. https://peri.umass.edu/publication/item/1397-employment-impacts-of-proposed-u-s-economic-stimulus-programs.

9. See Josh L. Bivens, The Potential Macroeconomic Benefits from Increasing Infrastructure Investment, Economic Policy Institute. July 2017. https://www.epi.org/publication/the-potential-macroeconomic-benefits-from-increasing-infrastructure-investment/; James Heintz, “The Impact of Public Capital on the U.S. Private Economy: New Evidence and Analysis.” International Review of Applied Economics 24, no. 5 (2010), 619–32; Joseph Berechman, Dilruba Ozmen, and Kaan Ozbay, “Empirical Analysis of Transportation Investment and Economic Development at State, County and Municipality Levels.” Transportation 33 (2006), 537–551.

10. In addition to supporting public health, public security, and education services, and providing aggregate demand support more generally—through the business and household sectors—with indirect economic benefits for transportation and infrastructure industries.

11. Moody’s Analytics. “Global Fiscal Policy in the Pandemic.” Moody’s Analytics. February 24, 2022. Accessed https://www.moodysanalytics.com/-/media/article/2022/global-fiscal-policy-in-the-pandemic.pdf.

12. Adam S. Hersh, Build Back Better’ Agenda Will Ensure Strong, Stable Recovery in Coming Years. Economic Policy Institute. September, 2021. https://www.epi.org/publication/iija-budget-reconciliation-jobs/.

13. To be clear, these average annual number of jobs supported cannot be summed together over 10 years. If, for example, all of the spending ramped up in Year 1 and then persisted, then 772,400 jobs would be supported in the first year and then this number would persist but not grow. Over the 10-year window, one could cumulate these job numbers and classify them as “job-years”—a measure of total hours of work supported by this spending over the next decade. For more on the estimation methodology, see Adam S. Hersh, Build Back Better’ Agenda Will Ensure Strong, Stable Recovery in Coming Years. Economic Policy Institute. September, 2021. https://www.epi.org/publication/iija-budget-reconciliation-jobs/.

14. These tables are reproduced from a report that also analyzed the potential effects of President Biden’s broader Build Back Better agenda.

15. Moody’s Analytics, “Macroeconomic Consequences of the Infrastructure Investment and Jobs Act & Build Back Better Framework.” Moody’s Analytics. November 4, 2021. https://www.moodysanalytics.com/-/media/article/2021/macroeconomic-consequences-of-the-infrastructure-investment-and-jobs-act-and-build-back-better-framework.pdf.

16. Robert Pollin, Chirag Lala, and Shouvik Chakraborty, Job Creation Estimates Through Proposed Inflation Reduction Act. Political Economy Research Institute. August 2022. Accessed https://peri.umass.edu/publication/item/1633-job-creation-estimates-through-proposed-inflation-reduction-act.

17. John Larsen, Ben King, Hannah Kolus, Naveen Dasari, Galen Hiltbrand, and Whitney Herndon, A Turning Point for US Climate Progress: Assessing the Climate and Clean Energy Provisions in the Inflation Reduction Act. Rhodium Group. August 2022. Accessed https://rhg.com/research/climate-clean-energy-inflation-reduction-act/.

The growing evidence that high-pressure labor markets can narrow these gaps calls for macroeconomic policies that test the absolute limits of how low unemployment can be pushed. Macroeconomic policymakers frequently weigh the potential benefits of higher-pressure labor markets against the potential risks, namely accelerating price inflation driven by excessively fast wage growth. The potential to close race-based gaps in the labor market should be counted as a substantial benefit in these deliberations and should convince policymakers to take on more inflation risk. Furthermore, running labor markets at the maximum sustainable pressure will provide much-needed information on what else we need to do to foster racial equity in labor markets.

This paper explores the promise and limits of high-pressure labor markets in reducing racial labor market gaps and how too-slack labor markets have helped thwart progress in closing the gaps. It then draws lessons from these investigations for policymakers. Its main findings are:

• Reductions in the unemployment rate boost hourly wages of typical (median) Black workers more than they boost hourly wages of typical white workers.
  • In 2019, the median Black worker was paid 32.2% less in hourly wages than the median white worker, up from 28.6% in 1973. Had the unemployment rate averaged 1 percentage point less annually from 1973 to 2019, the median Black–white wage gap could have declined to 18.0%. If the unemployment rate had averaged 2 percentage points less (a very ambitious target), the median Black–white wage gap could have fallen to just 5.4% (an 80% reduction in the size of this wage gap).
• Reductions in the unemployment rate provide an even bigger relative boost for median Black annual earnings, by increasing both hours worked and hourly wages.
  • In 2019, the annual earnings of the typical (median) Black worker amounted to just 80% of the annual earnings of the median white worker. Had the unemployment rate averaged 2 percentage points less annually over the 1970–2019 period, the Black–white median earnings gap (measured as a ratio) could have essentially closed. Had the unemployment rate averaged 1 percentage point less annually over that period, the typical Black worker in 2019 could have been paid 90.1% as much as the typical white worker—reflecting a 50% decrease in the Black–white median annual earnings gap.
• The Black–white unemployment gap (how much, in percentage points, the Black unemployment rate exceeds the white unemployment rate) closes significantly when the overall economy has fewer idle resources (i.e., when potential output climbs closer to actual output, leading to a rise in the measured “output gap”). For example, the Black unemployment rate falls more than twice as much as white unemployment when the economy’s output gap rises by 1 percentage point.
  • Even this disproportionate reduction in Black unemployment might understate how equalizing overall economic tightening can be. When the output gap measure rises 1 percentage point, the share of Black persons who are employed (the Black employment-to-population ratio, or EPOP) actually rises nearly seven times as much as the share of white persons employed (the white EPOP). But because the share of Black persons who are either working or actively looking for work (the Black labor force participation rate) also rises faster than white labor force participation when the output gap improves, the Black unemployment rate reduction is muted relative to gains in employment.
• Sustained high-pressure labor markets may have more power than we thought to close the Black–white unemployment ratio. For decades, the Black unemployment rate has been, on average, roughly twice the white unemployment rate. This persistent and distressingly high Black–white unemployment ratio (the Black unemployment rate divided by the white unemployment rate) has traditionally been seen as much more resistant to closing with high-pressure labor markets. However, the Black unemployment rate has only been included in most data sets since the early 1970s and, since then, genuinely high-pressure labor markets have been quite rare. Pre-1970s data that provide a potential proxy for the Black–white unemployment ratio show that sustained high-pressure labor markets may well actually reduce it significantly.

The policy lessons from this data are clear. While overall wage and price inflation remain the proper targets of policy (employment measures are not reliable enough to make good policy guides), policymakers need to change how they balance those targets:

• As they weigh the potential benefits of higher-pressure labor markets against the risks, policymakers should count, on the benefits side, potential reductions in chronic racial gaps in labor market outcomes.
• More forbearance should be exercised as wages and prices rise during economic recoveries and expansions, and at a bare minimum wage and price targets should be kept symmetric over business cycles: Every year that sees wage and price inflation come in 1% below target must be matched by a year with wage and price inflation coming in 1% above target.
• The potential to close race-based gaps in the labor market should convince policymakers to take on more inflation risk than they otherwise would have (that is, they should wait for actual and sustained, rather than forecast, inflation to appear before raising interest rates).

Background on the unemployment and inflation trade-off

All else equal, policymakers should aim for an unemployment rate so low that it reflects only the transitory and voluntary shifts of workers in and out of work or between employers. However, because of the way policymakers have traditionally sought to affect the rate of unemployment, they have instead aimed for a rate that was high enough to avoid any chance, even remote, of sparking inflation.

The primary way policymakers influence the unemployment rate is through measures that change the pace of aggregate demand growth. Aggregate demand is economywide spending of households, businesses, and governments. When this spending is strong, employers need workers to produce the output of goods and services needed to satisfy customer demand, which keeps unemployment low and employment growth strong. When this spending lags, less output and hence fewer workers are needed to satisfy demand, so employment growth lags and unemployment rises.

If policymakers boost economywide spending too much, however, demand might outstrip the productive capacities of firms. As demand runs ahead of supply, this puts upward pressure on wages and prices as firms scrambling to meet demand find they need to hire more workers and can charge customers a bit more for scarce goods. This “inflation barrier” to further efforts to boost demand—the point of tightness in labor markets that sparks an upward drift of inflation—may well be hit before the unemployment rate that reflects only voluntary job transitions is attained.

This balancing between demand growth that is strong enough to keep unemployment low, but not strong enough to generate accelerating inflation, is a central problem of macroeconomic policy (often called stabilization policy). Traditionally, the entity doing this balancing in the United States has almost always been the Federal Reserve, which tries to spur demand primarily by lowering interest rates and can brake escalating demand by raising interest rates. However, the Great Recession exposed the extreme limits of the Fed’s ability to generate strong enough demand growth and has elevated the role of fiscal policymakers (Congress and the president) in boosting (or failing to boost) demand by adjusting spending levels and taxation in the economy.1

Far too often in recent decades, policymakers have erred in targeting—or at least unnecessarily tolerating—demand growth that was too weak to generate enough pressure in labor markets to give workers leverage in wage negotiations with employers.2 This toleration of low-pressure labor markets was often done in the name of keeping inflationary pressures in check. But given that genuine inflationary pressures in the U.S. economy have been extraordinarily rare since the 1970s, the targeting of too-weak demand growth has often been about guarding against even the risk of inflation. A growing body of recent research notes that the benefits of low unemployment are large enough to justify taking on substantially more inflation risk than has previously been tolerated.

The most obvious benefits of low unemployment are more job opportunities for more people and more hours of work available to U.S. families. A less obvious benefit, but one that shows up strongly in the data, is faster hourly wage growth for the vast majority of U.S. workers, a particularly important benefit given the anemic pace of wage growth for these workers in recent decades.3 Yet another increasingly discussed benefit of low unemployment is its ability to put sustained pressure on compressing race-based gaps in the labor market. The rest of this paper largely tries to put some empirical bounds on just how large this last benefit might be.

Why aim for ‘high-pressure’ labor markets and not ‘full employment’?

The Fed’s legal mandate is to pursue maximum employment consistent with price stability. Over the years “maximum employment” has often been referred to as “full employment.” However, there is no universally agreed upon definition of full employment. For some, full employment simply means that anybody who wants a job can find a job. For others, particularly macroeconomists, it means attaining the rate of unemployment (often called “the natural rate”) below which further increases in economywide spending will mostly lead to accelerating inflation rather than greater output. “High-pressure” labor markets just mean labor markets characterized by low unemployment, fast rates of job creation, rapid job-finding among the unemployed, and sustained effort by employers to keep their enterprises properly staffed. Labor markets can be “high pressure” yet still tolerate further reductions in unemployment without leading to unsustainable wage or price inflation, and the term “high-pressure” may better connote a continuum of labor market states rather than a single fixed point.

Further, old theories of “disguised unemployment” and new developments in advanced capitalist economies (the rise of “gig work”) argue that “full employment” and “high-pressure labor markets” might not always coincide.

Joan Robinson (1936) defined “disguised unemployment” as follows:

In a society in which there is no regular system of unemployment benefit, and in which poor relief is either nonexistent or “less eligible” than almost any alternative short of suicide, a man who is thrown out of work must scratch up a living somehow or other by means of his own efforts. And under any system in which complete idleness is not a statutory condition for drawing the dole, a man who cannot find a regular job will naturally employ his time as usefully as he may. Thus, except under peculiar conditions, a decline in effective demand which reduces the amount of employment offered in the general run of industries will not lead to “unemployment” in the sense of complete idleness, but will rather drive workers into a number of occupations—selling match-boxes in the Strand, cutting brushwood in the jungles, digging potatoes on allotments—[that] are still open to them.

The modern U.S. economy obviously does not totally lack relief for the unemployed, and the reach and influence of the gig economy is often wildly overstated. But it seems clear that one margin of survival that many U.S. workers draw on when regular work is slack due to weak aggregate demand is to engage in gig or otherwise irregular work. But gig work generally does not provide high-quality jobs or economic security. In some deeply unsatisfactory sense, the rise of gig work could theoretically help fulfill the promise of one definition of full employment—that anybody “who wants a job can find a job.” But, in an economy with measured unemployment kept low only by a large incidence of gig work, if policymakers boosted aggregate demand, it is highly likely that many gig workers would leave the gigs behind and look for and find more regular work. In short, describing labor markets as “high pressure” might better describe the condition that employers are competing actively among themselves to attract workers.

High-pressure labor markets and median racial wage gaps

Since 1979, wage gaps between Black and white workers have widened significantly. Figure A shows the gap in two ways: how much less in percent terms the median Black worker earns in hourly wages than the median white worker, and the percent by which the average hourly wage of Black workers is less than the average hourly wage of white workers, holding other characteristics constant. The latter, a regression-adjusted average gap, controls for educational attainment, gender, ethnicity, and age. Both gaps widened significantly over time, but the median gap started larger and has expanded more rapidly since the late 1970s.

Figure A

Previous work has indicated that median Black wage growth responds more strongly to changes in unemployment than does median white wage growth.4 Figure B confirms this. The figure shows the relationship between the unemployment rate and wage growth. Specifically, it shows the change in wage growth that occurs if the unemployment rate rises by 1 percentage point. For white median hourly wages, a 1-percentage-point increase in overall unemployment is associated with wage growth that is 0.52% slower. For Black median wages, wage growth declines by 0.76%. As the figure shows, the coefficient for median Black wage growth is nearly 50% larger than for median white wages.5

Figure B

Figure C uses the estimated coefficients from Figure B and calculates counterfactual median Black–white wage gaps under three scenarios: unemployment rates that averaged 1, 1.5, and 2 percentage points lower over the 1973–2019 period. These scenarios are plausible alternatives of what might have been under a policy regime that determinedly aimed for high-pressure labor markets. Over this period, the unemployment rate was high, averaging 6.2%. One closely watched measure of the unemployment rate consistent with stable inflation—the nonaccelerating inflation rate of unemployment (or NAIRU) estimated by the Congressional Budget Office (CBO)—averaged 5.3% over this same period, almost a full percentage point lower. Additionally, between 1947 and 1973, the unemployment rate averaged 4.7%, exactly 1.5 percentage points lower than in the post-1973 period, and inflation before the oil price shock of 1973 was generally contained. Finally, when unemployment fell more than 2 percentage points beneath the 1973–2019 average in the late 1990s, and again in 2018–2019, there was no marked uptick in wage or price inflation requiring that macroeconomic policymakers slow demand growth.

Figure C

We should be clear that examining a counterfactual that assumes substantially lower unemployment on average does not reflect an assumption that recessions never happen. Instead, it simply assumes that macroeconomic policymakers do their job and ensure that every period of above-average unemployment is matched by an equivalent period of below-average unemployment. Estimates of the natural rate of unemployment are not hard floors below which the economy is never meant to go; instead they are averages that the unemployment rate should fluctuate both above and below. Running the economy far above even too-conservative natural rate estimates over decades is a policy failure that can clearly be addressed.

Achieving and sustaining high-pressure labor markets since the early 1970s would have dramatically narrowed the median Black–white wage gap. Had unemployment averaged 2 percentage points less over the entire period, 80% of the median Black–white wage gap that appeared in 1973 could have been erased (as the gap shrank from 28.6% to 5.4%). With unemployment averaging just 1 percentage point less (essentially just hitting conventional measures of the natural rate of unemployment), the median wage gap could have fallen slightly (to 18.0%) rather than rising by almost 8 percentage points over this period. In short, high-pressure labor markets hold great potential to reduce this particular measure of racial inequality in the labor market.

The gap-narrowing power of high-pressure labor markets is even more evident when looking at median annual earnings. Annual earnings can be affected by tighter labor markets not only through higher hourly wages but also through increased hours worked during the year. As shown in Figure B, the decline in Black worker annual earnings associated with an uptick in the unemployment rate is an even larger decline than the decline in Black worker hourly earnings. Applying the same counterfactual scenarios of unemployment rates that average 1, 1.5, and 2 percentage points lower over the 1973–2019 period yields dramatic results for the median Black–white annual earnings gaps, shown in Figure D. In this figure, the gaps are presented as ratios—how much Black workers earn as a share of what white workers earn.

Figure D

Between 1970 and 2019, the ratio of Black to white annual earnings rose from 62.4% to 80.0%. If the unemployment rate had averaged 1 percentage point lower after 1973, then this ratio could have surpassed 90% by 2019. If the unemployment rate had averaged 2 percentage points lower, the Black–white annual earnings ratio would have essentially been 1, indicating near-complete equality in this measure.

High-pressure labor markets and gaps in employment and unemployment

As we have frequently noted, the Black unemployment rate has been, on average, roughly twice the white unemployment rate since 1972 (the first year that Black unemployment is measured by the Bureau of Labor Statistics). Further, as shown in Figure E, this rough 2-to-1 ratio prevails if one looks at the measure of “nonwhite” unemployment compiled by the BLS before 1972 (Black workers accounted for a very large majority of nonwhite workers over that pre-1972 period).

Figure E

The Black–white unemployment ratio shrinks only slightly if one adjusts for age, educational credentials, and gender composition of the workforces. In the figure, this ratio is calculated by comparing the “Black adjusted” line (see figure note) with the line for the white unemployment rate. For example, between 1976 and 2019, the overall Black–white unemployment ratio averaged 2.3 while the adjusted ratio averaged 2.0. This is an improvement for sure, but a depressingly small one, at roughly 15%.

One conclusion that can be drawn from Figure E is that the ratio of Black to white unemployment is pretty stubborn: It does not seem to fall quickly during periods of labor market tightness (when all rates fall together). However, even if the Black-to-white unemployment ratio never moved, the raw gap in unemployment rates between Black and white workers (simply the Black unemployment rate minus the white unemployment rate) would shrink rapidly during periods of overall labor market tightness, and would expand rapidly during periods of overall labor market distress. At a minimum, this means that Black workers see disproportionate gains and losses from effective and ineffective macroeconomic stabilization policy, respectively. Thus, getting macroeconomic stabilization policy right is a key issue for racial equity.

Figure F confirms this intuition, using the output gap as a proxy for overall economic, and thus labor market, health.6 The output gap is a measure of how fully the economy’s resources are being utilized at any given point in time (resources including potential workers). Specifically, it is calculated as the quotient of actual gross domestic product (GDP) divided by a measure of potential GDP (what GDP could have been had the economy’s resources been fully utilized), minus 1. When actual GDP is lower than potential GDP, the output gap is negative. As actual GDP falls further and further behind potential GDP, the gap measure becomes more negative; as it comes closer to potential GDP, it rises.

Figure F

Given this, as the gap rises, resource utilization increases and the overall unemployment rate generally falls. The figure shows the relationship between the overall output gap and Black and white labor market indicators. As can be seen, the Black unemployment rate is twice as responsive as the white unemployment rate to a change in the output gap: specifically, a 1-percentage-point increase in the output gap (moving actual GDP 1% closer to potential GDP) is associated with a 1.57-percentage-point decline in the Black unemployment rate, compared with a 0.64-percentage-point reduction in the white unemployment rate. Because the BLS measures the unemployment rate of Black workers only after 1971, we also include a measure of the responsiveness of what the BLS labels “nonwhite” unemployment—a series that goes back to 1954. In the years before 1972, Black workers made up more than 90% of those labeled nonwhite.7 The advantage of using a series with a longer historical perspective is that one can examine a period of very tight labor markets that were achieved in the mid-to-late 1960s. The overall unemployment rate, for example, fell to under 4% for three straight years in the late 1960s. In this longer time series, the overall responsiveness of the nonwhite unemployment rate to an increase in the output gap measure (-1.54) is quite close to the responsiveness of Black unemployment in the more recent series.

These differential rates of responsiveness translate into substantial closing of the gap between Black and white unemployment rates when the overall economy tightens up, with this gap defined simply as the Black unemployment rate minus the white unemployment rate (i.e., the gap is by how many percentage points the Black unemployment rate exceeds the white unemployment rate). The figure shows that each percentage-point increase in the output gap (i.e., a tightening of the economy and labor market generally) is associated with a 0.70-percentage-point reduction in the Black–white unemployment gap. Each percentage-point increase in the output gap is also associated with a 0.90-percentage-point reduction in the nonwhite–white unemployment gap. It is possible that relatively greater responsiveness of the nonwhite–white unemployment gap is due to the inclusion of workers who are not Black in the nonwhite unemployment calculation. It is also possible that the difference is due to the longer time series available with the nonwhite–white unemployment rate gap. By restricting this series to just post-1971 data points (to make it consistent with the Black unemployment rate coverage), the responsiveness of the nonwhite–white unemployment gap to a 1-percentage-point reduction in the output gap shrinks to 0.83 percentage points.

Race-based differentials in the responsiveness of labor market indicators to a change in the output gap are even larger when examining the responsiveness of the Black and white employment-to-population ratios. The Black EPOP rises by 1.2 percentage points as the output gap increases, while the white EPOP rises by 0.18 percentage points, just over a seventh as much.

If the ratio of Black to white unemployment rates was constant, then the change in the gap between these rates would also just equal this ratio multiplied by the change in the white unemployment rate. Given the relative stubbornness of the Black–white unemployment ratio (for example, as seen in Figure E), it might seem that it is essentially constant regardless of the state of labor market pressure. But it may not be.

Can high-pressure labor markets reduce Black–white unemployment ratios, not just gaps?

Looking at the Black and white unemployment rates over time—like those displayed in Figure E—can easily convince observers that the ratio of Black to white unemployment is nearly constant. In good times and in bad, the Black unemployment rate looks to be roughly twice the white unemployment rate. But there are actually some reasons for optimism—tempered, to be sure—that this ratio is not as unyielding to change as it seems. For one, there seems to be a shallow but steady downward trend in this ratio over time. For another, more detailed evidence indicates that the Black–white unemployment ratio may indeed respond measurably to high-pressure labor markets. That evidence is highlighted in the discussion of the next two figures, which show the overall unemployment rate and the Black–white unemployment ratio (Figure G) and the nonwhite–white unemployment ratio (Figure H) prevailing at business cycle peaks. Both show a clear positive relationship between the overall unemployment rate and the respective ratios (i.e., an increase in one measure coincides with an increase in the other). All else equal, this would indicate that a higher-pressure labor market overall does indeed put downward pressure on the Black–white unemployment ratio.

Figure G

Figure H

However, the positive relationship in Figure G is likely driven in some part by the trend in the Black–white unemployment ratio coinciding with the fact that, since 1976, later business cycles have consistently achieved lower unemployment rates. Between the business cycle peaks of 1979 and 2019, the Black–white unemployment rate ratio actually declined by a bit over 20% (as did the adjusted ratio, a ratio that estimates what the Black unemployment rate would have been had the composition of the Black labor force shared the same age, educational credentials, and gender mix as the white labor force). This trend likely would have led to successively lower Black–white unemployment ratios in 1989, 2000, and 2019 anyhow. But on top of this trend, unemployment rates in these business cycle years were successively lower over time. Given this, it is not clear if it is a given year’s unemployment rate or a long-running trend that drives the pattern in Figure G.

To test the connection, Figure H includes data on nonwhite unemployment back to 1959 and thus includes business cycles not characterized by uniformly lower overall unemployment rates over time. The strong positive relationship between rising overall unemployment and an increasing Black–white unemployment rate ratio still holds.

Figure I looks at the responsiveness of various labor market ratios (not gaps, as was analyzed above in Figure F) to changes in the output gap while controlling for a time trend. The first three data points come from a regression that used a lagged measure of the output gap. They show a significant decline in the Black–white and the nonwhite–white unemployment rate ratios associated with each percentage-point increase in the output gap (remember, as the economy improves and actual GDP gets closer and closer to potential GDP, the output gap rises). As before, this analysis includes a look at the coefficient on the nonwhite–white unemployment ratio from this regression just in the years after 1971 to see if some of the difference between its responsiveness and the responsiveness of the Black–white unemployment ratio is simply due to different timespans. The responsiveness of the nonwhite–white unemployment ratio is roughly same (but actually increases slightly) when just looking at the post-1971 period.

Figure I

The results from estimating the responsiveness of ratios of EPOPs and labor force participation rates (LFPRs) demonstrates a similar pattern as that shown by Black–white employment gaps. When the output gap rises, the Black–white ratio of EPOPs rises, meaning that the share of Black persons employed approaches closer to the share of white persons employed. A similar increase holds for the ratio of nonwhite-to-white EPOPs and, again, the responsiveness of this ratio falls a bit when only the post-1971 period is examined. The ratio of Black to white LFPR rises when output gaps rise as well, meaning Black labor force participation approaches closer to white labor force participation. Again, the responsiveness of the ratio of nonwhite–white LFPRs is greater than for the Black–white ratios, but much of this seems due to the different time periods; when only post-1971 years are included, the responsiveness is very similar.

The upshot of this examination is that there is some suggestive evidence that even key labor market ratios (not just absolute gaps) that compare labor market performance for Black and white workers might indeed narrow during periods of high-pressure labor markets. This more hopeful interpretation may have been missed by those looking only at raw measures of the Black and white unemployment rates over time (like those shown in Figure E).

First, the post-1971 period might not contain enough episodes of truly tight labor markets to allow the relationships between high-pressure labor markets and the Black–white unemployment ratio to be well estimated. Figure I has some slight suggestive evidence of this: The responsiveness of the nonwhite–white EPOP and LFPR is greater when the pre-1971 period is included. This pre-1971 period includes a long stretch in the 1960s when unemployment was beneath 4% for four straight years (1966–1969). Further, Aizer et al. (2020) found that many racial gaps in labor markets (like the Black–white earnings gap and the measures of occupation segregation) fell significantly in the 1940s. These declines were concentrated in areas with more defense spending. This spending not only contributed to tighter labor markets but also often came attached with anti-discrimination conditions. The authors could not disentangle the precise effect of each of these influences, but the large spillovers of reduced race-based gaps in industries not directly affected by the defense spending suggests a large role for high-pressure labor markets generally. And the U.S. labor market of the 1940s was high pressure in a way not seen since: The unemployment rate was below 2% for three straight years between 1943 and 1945. In short, since we have begun measuring the Black unemployment rate specifically, we just may not have seen enough periods of genuinely high-pressure labor markets to get a solid statistical read on what happens to the Black–white unemployment ratio when labor markets get truly tight and are kept that way for a sustained period of time.

Second, while the employment rate of Black workers rises significantly faster than for white workers as the output gap rises, the labor force participation rate of Black workers also rises faster, muting any disproportionate decline in unemployment for Black workers. If the measure is simple joblessness and not unemployment, then it seems clear that the Black–white jobless ratio clearly declines when labor markets tighten up.

Part of the reason why the stronger responsiveness of the Black–white EPOP to output gap increases translates weakly into a reduction in the Black–white unemployment ratio is likely due to different dynamics of labor force participation over the business cycle. A recent paper by Cajner, Coglianese, and Montes (2020) makes a significant contribution in our understanding of the cyclical behavior of labor force participation. Their main finding is that the LFPR is indeed affected by the state of labor market tightness, but that it responds substantially more slowly to positive or negative shocks than employment or unemployment. They further find that the Black LFPR responds substantially more strongly to a negative shock to the overall labor market. So when economic growth slows and the labor market develops slack, the rise in the Black unemployment rate relative to the white unemployment rate can be somewhat muted because of a larger labor supply response from Black workers. This means that the Black–white unemployment ratio may actually fall during recessions. Just to buttress this point, it is striking that the two lowest annual Black–white unemployment ratios on record occurred in 2009 and 2020—two of the worst years for economywide labor market health in the past 70 years or more.

As recoveries begin, Black EPOPs respond more strongly to improving economic conditions. All else equal, this should lead to a reduction in the Black–white unemployment ratio. But because the Black LFPR recovers more quickly than the white LFPR , the progress in reducing the racial unemployment gap is blocked by faster labor force growth among Black workers. By the time late in recoveries when labor markets are getting tight again, the white LFPR likely begins recovering more strongly, which allows for a reduction in the Black–white unemployment ratio. This modestly complicated series of dynamics likely explains part of why the salutary effect of lower unemployment rates on the Black–white unemployment ratio might be harder to detect in a simple eyeballing of trends. In 2019, the last business cycle peak, the Black–white unemployment ratio hit its lowest point at any business cycle peak on record. This likely reflects both a shallow but nontrivial downward trend over time and pressure that tight labor markets put on compressing the ratio. Additionally, the prolonged (if too slow) recovery following the Great Recession allowed ample time for the white LFPR to recover from the negative shock of the Great Recession and to stop putting downward pressure on the white unemployment rate.

Policy implications

The upshot of this examination is that sustained periods of high pressure in U.S. labor markets might significantly narrow racial gaps in unemployment and other key labor market measures. Given the long history of structural racism in the United States and the intentional policy efforts that created these gaps, it seems incumbent upon policymakers to use every tool available to try to close them. High-pressure labor markets look as promising as (or more promising than) any other tool. The large benefits—moral, political, and economic—of closing these labor market gaps call upon macroeconomic policymakers to consider them when assessing the benefits and costs of a “go for growth” strategy targeting high-pressure labor markets. To be explicit: The potential of more aggressive expansionary macroeconomic policy to help close race-based gaps in the labor market is worth taking on more risk of sparking inflation.

This policy recommendation for macroeconomic policymakers (including the Federal Reserve) to take on extra inflation risk in the name of narrowing racial gaps in the labor market is likely frustratingly imprecise to some. Some policymakers would prefer the clarity of, say, a numerical target for the Black unemployment rate. However, excess confidence in the ability of macroeconomic policymakers to use hard-and-fast ex ante labor market targets that precisely define “high pressure” has backfired in the past. Specifically, that unfounded confidence is a prime reason why labor markets were kept too slack for so long in recent decades, as hard targets such as estimates of the NAIRU turned out to be wrong, leading to unemployment rates in excess of what was needed for reasonable inflation control. Further, if using overall unemployment rates as precise labor market targets has proven to lead to unsatisfactory outcomes (and it has), using the Black unemployment rate as a specific target might be even worse, as one would be implicitly targeting not only the overall rate, but also how robustly the ratio between the Black and the overall rate changed as overall unemployment rose and fell depending on labor market conditions.

One of the most direct and thoughtful calls for having the Federal Reserve aim for narrower racial gaps in the labor market was by Bernstein and Jones (2020b). Their paper is often described as calling on the Fed to “target the Black unemployment rate,” but it does so only in the sense described above: It calls upon the Fed to consider the benefits of narrower gaps, and explicitly make them part of their criteria for decision-making. As the authors explain, “It is not just asking the chair to tell us about the gaps; it requires him or her to make closing them a part of their mandate” (Bernstein and Jones 2020a).

These sensible calls to narrow labor market gaps do raise an important question: Why is there reticence to demand that macroeconomic policymakers achieve a full elimination of labor market gaps? The answer is because it is unlikely that macroeconomic policy by itself can neutralize the centuries-long legacy of structural racism. This legacy has led to disadvantages Black workers face along numerous margins in the labor market, and while high-pressure labor markets can help to ameliorate these disadvantages, high-pressure labor markets likely cannot completely undo them before inflationary pressures require some moderating of expansionary policy.

For example, some of the gap in unemployment rates between Black and white workers represents differing levels of educational credentials. As we showed earlier (Figure E), adjusting the Black unemployment rate under a scenario that gives the Black and white workforces the same age and educational profiles does reduce the Black–white unemployment ratio by a small amount, around 15%. This gap in educational credentials obtained by Black and white workers is itself largely a function of historic discrimination, but it is unlikely to be solved simply by boosting aggregate demand. Further, even at the same level of educational credentials, it is certainly possible for the quality of educational investments to differ systematically between Black and white workers. Research has shown that educational investments are not only larger in white neighborhoods, but they have also been systematically reduced in schools with larger shares of Black students.8 Thus it seems likely that equalizing labor market outcomes will require interventions over and above expansionary macroeconomic policy to address the differences in education investment.

Despite these caveats, the results in this paper and previous research clearly show that expansionary macroeconomic policy can have profoundly equalizing effects. It also seems clear that policymakers have not fully accounted for these benefits when weighing benefits against the potential cost of sparking inflationary pressure. Further, ignoring these potential benefits may even result in worse analytical forecasting. Concepts like the natural rate of unemployment and the level of potential output for the U.S. economy often are estimated by assuming a given Black–white unemployment gap that does not close as the economy heats up.

An oft-cited example is the Congressional Budget Office (CBO) estimate of the natural rate of unemployment. The CBO assumes the overall unemployment rate reached in 2005 is consistent with the economy’s natural rate. It then takes group-specific unemployment rates that prevailed in 2005 and allows the overall natural rate to change only as the group-specific shares of the labor force change over time due to demography or immigration flows (Shackleton 2018). In some sense, this method implicitly assumes that group differences in unemployment that prevailed in 2005 are set in stone. Some have gone so far as to call this assumption racist. This seems wrong. The existence of the gaps is evidence of racism. But it would be odd indeed to ignore them entirely when doing forecasting given how persistent they have been. Instead, it seems that this assumption of ever-persisting gaps is evidence more of pessimism (much of it arguably well-earned) than of racism. 

But if these gaps do indeed close further as labor markets enter high-pressure periods, then any estimate of the natural rate of unemployment can be lower and estimates of potential output can be higher than one would otherwise forecast. In essence, we will never know the full extent of other policy interventions that need to be done to foster racial equity until we have fully maximized the reach of high-pressure labor markets. To put this another way, we won’t even know the size of the Black–white unemployment gap until we are sure we have reached the lowest rate of overall unemployment consistent with sustainable inflation. And yet for the vast majority of years over recent decades, we have not been close to this minimum unemployment rate.

In recent months, handwringing about the possibility of eventually “overheating” the U.S. economy due to excessively generous fiscal stimulus has begun. It is true that we are not completely certain about how low unemployment can go or how much the economy’s supply side will respond to growth in aggregate demand—so signs of overheating should indeed be monitored. But we should be very cautious about premature declarations of overheating. We have not seen sustained and broad-based wage and price inflation in the U.S. economy for decades. And we now know much more than in previous years about just how equalizing a high-pressure labor market can be, both for compressing wage growth among low-, middle-, and high-wage workers and for closing race-based gaps in labor market measures. These benefits are utterly enormous, and maximizing them is worth the risk of being very patient before aiming to deflate high-pressure labor markets through policy.

Endnotes

1. See Bivens 2016 for the central role of fiscal policy in conditioning economic growth after the Great Recession of 2008–2009.

2. See Bivens and Zipperer 2018 for some evidence of this.

3. See Mishel and Bivens 2021 for the central role of low-pressure labor markets in suppressing wage growth for most of the post-1979 period. See Gould 2020 for a broad overview of wage trends over the same period.

4. See Wilson 2015 for evidence on this.

5. While the differences between the coefficients are not statistically significant at most conventional levels, the difference in magnitude is economically large and is consistent and robust across the various regression specifications and time periods.

6. We switch to using an output gap measure for the state of the overall economy in this section because there is an arithmetic relationship between the overall unemployment rate and disaggregated measures of unemployment and employment by group. When, for example, the Black unemployment rate falls, this will by definition also reduce the measured overall unemployment rate. Our output gap measure does not have any arithmetic relationship to disaggregated labor force measures, so we use it for the rest of this paper.

7. See Hobbs and Stoop 2002 for evidence of this.

8. See Derenoncourt 2021 for evidence that as neighborhoods’ share of Black residents increased over time, public investments shifted more heavily toward policing and incarceration and white students saw higher enrollments in private schools. See Johnson 2011 for evidence that racial segregation led to lower resources for students in schools with higher shares of Black students.

References

Aizer, Anna, Ryan Boone, Adriane Lleras-Muney, and Jonathan Vogel. 2020. “Discrimination and Racial Disparities in Labor Market Outcomes: Evidence From WWII.” National Bureau of Economic Research (NBER) Working Paper #27689, August 2020, https://doi.org/10.3386/w27689.

Bernstein, Jared, and Janelle Jones. 2020a. “The Federal Reserve Could Help Make the Job Market Fairer for Black Workers.” Washington Post, June 15, 2020.

Bernstein, Jared, and Janelle Jones. 2020b. The Impact of the Covid-19 Recession on the Jobs and Incomes of Persons of Color. Groundwork Collaborative and Center on Budget and Policy Priorities. Policy Futures Report. June 2020.

Bivens, Josh. 2016. Why is Recovery Taking So Long—and Who’s to Blame? Economic Policy Institute. August 2016.

Bivens, Josh, and Ben Zipperer. 2018. The Importance of Locking in Full Employment for the Long Haul. Economic Policy Institute. August 2018.

Bureau of Labor Statistics, Consumer Price Index (BLS-CPI). 2021. Public data series for various years accessed through the CPI National Databases and through series reports. Accessed January 2021.

Bureau of Labor Statistics, Current Population Survey (BLS-CPS). 2021. Public data series for various years accessed through the CPS National Databases and through series reports. Accessed January 2021.

Bureau of Labor Statistics, Productivity and Costs by Major Sector. (BLS-LPC). 2021. Public data series for various years accessed through the LPC National Databases and through series reports. Accessed January 2021.

Cajner, Tomaz, John Coglianese, and Joshua Montes. 2020. “The Long-Lived Cyclicality of the Labor Force Participation Rate.” Working Paper. October 2020.

Congressional Budget Office (CBO). 2021. Historical Data and Economic Projections (online database). Accessed February 2021.

Derenoncourt, Ellora. 2021. “Can You Move to Opportunity? Evidence from the Great Migration.” Working Paper. February 2021.

Economic Policy Institute (EPI). 2021a. Current Population Survey Extracts, Version 1.0.18, https://microdata.epi.org.

Economic Policy Institute (EPI). 2021b. State of Working America Data Library.

Flood, Sarah, Miriam King, Renae Rodgers, Steven Ruggles, and J. Robert Warren. Integrated Public Use Microdata Series, Current Population Survey: Version 7.0 [data set]. Minneapolis, MN: IPUMS, 2021. https://doi.org/10.18128/D030.V7.0

Gould, Elise. 2020. State of Working America 2019: A Story of Slow, Uneven, and Unequal Wage Growth Over the Past 40 Years. Economic Policy Institute. February 2020.

Hobbs, Frank, and Nicole Stoop. 2002. Demographic Trends in the 20^th Century. U.S. Census Bureau. November 2002.

Johnson, Rucker. 2011. Long-Run Impacts of School Desegregation and School Quality on Adult Attainments. National Bureau of Economic Research Working Paper #166664.

Mishel, Lawrence, and Josh Bivens. 2021. Identifying the Policy Levers Generating Wage Suppression and Wage Inequality. Economic Policy Institute. May 2021.

Robinson, Joan. 1936. “Disguised Unemployment.” The Economic Journal 36, no. 182): 225–237.

Shackleton, Robert. 2018. “Estimating and Projecting Potential Output Using CBO’s Forecasting Growth Model.” Congressional Budget Office Working Paper Series. February 2018.

Wilson, Valerie. 2015. The Impact of Full Employment on African American Employment and Wages. Report for the Full Employment Project at the Center on Budget and Policy Priorities. March 2015.

A strong domestic steel industry is critical to U.S. national defense, to the health of America’s critical infrastructure, and to the competitiveness of many domestic manufacturing industries. Beyond supplying high-quality steel in sufficient quantities to meet national defense needs, the U.S. steel industry also plays a critical role in supporting the welfare of other industries essential to the broader health and operation of the economy and government. For decades, chronic global steel supply gluts have undermined the U.S. steel industry with surging imports to U.S. markets undercutting prices, domestic production, employment, and investments. This oversupply jeopardizes the fundamental health of the U.S. steel industry—one of the cleanest and most energy-efficient steel industries globally.

Global steel surpluses are the result of chronic global excess steelmaking capacity in major exporting countries, including China, India, Brazil, Korea, Turkey, the EU, and other nations, much of it from state-owned and state-supported enterprises that are heavy polluters. In 2018, the United States determined that steel imports posed significant risks to national security and imposed a 25% tariff and other trade remedies on certain steel products under Section 232 of the Trade Expansion Act of 1962. This report examines the impacts of these measures on domestic steel production and consuming industries, and it recommends that these measures be retained until a multilateral solution to the problem of global excess steel capacity can be achieved.

Key conclusions of this report include:

• The U.S. steel industry is a vital component of the American economy. In 2017, prior to Sec. 232 import measures, the U.S. steel industry supported nearly 2 million jobs that paid, on average, 27% more than the median earnings for men and 58% more than the median for women.

• Global steel markets are plagued by chronic excess capacity. Measured by the Organisation for Economic Co-operation and Development (OECD), global excess capacity is 5.8 times the productive capacity of the entire U.S. steel industry. Massive overcapacity driven by subsidies and other anti-competitive policies can only be disposed of by these producers flooding U.S. and other markets with exports, posing material harm to U.S. steel producers and risking the U.S. industry’s ability to maintain operations, grow, and invest in areas essential to national defense, critical infrastructure, and broader economic welfare.

• The economic picture for U.S. steel producers brightened considerably beginning in 2018 until the pandemic began. Following implementation of Sec. 232 measures in 2018—and prior to the global downturn in 2020—U.S. steel output, employment, capital investment, and financial performance all improved. In particular, U.S. steel producers announced plans to invest more than $15.7 billion in new or upgraded steel facilities, creating at least 3,200 direct new jobs, many of which are now poised to come online. In addition, more than $5.9 billion was invested by nine firms in plant acquisitions as part of industry restructuring to increase efficiency, preserving additional jobs at those facilities.

• Administrations dating back to the mid-20th century have worked to mitigate the effects on U.S. steel producers of unfair global practices. For decades, unfair trade practices have threatened the U.S. steel industry with repeated crises. In this context, the recent Sec. 232 import measures simply continue a long thread of executive policy actions to provide relief for the damages wrought on U.S. producers by unfair competition and global surplus capacity in steel. For example, President Obama pressed the excess capacity issue through diplomatic channels at the G20 and in the U.S.-China Strategic and Economic Dialogue and under U.S. law, overseeing 370 trade remedy actions on imported steel products.

• China has massively and rapidly expanded its steel production capacity. China, the world’s largest steel producer, used subsidies and other forms of distortionary government support to expand steel capacity by 418%, or 930 million metric tons (MMT), since 2000, such that by 2019 it controlled just shy of half of global steel capacity. Chinese steel firms are also investing in developing capacity overseas, including in Europe, Asia, and Latin America, in efforts to evade trade enforcement actions.

• Countries across several continents followed in China’s footsteps, developing more excess capacity. Rapid growth in overcapacity is not limited to China. Other major steel-producing countries achieving rapid capacity growth between 2000 and 2019 include India, Turkey, Iran, South Korea, Vietnam, Russia, Brazil, Mexico, and Taiwan, with increases ranging from 8 MMT in Taiwan to 95 MMT in India. These are all countries where the state dominates or plays a significant role directing steel and other heavy industries, where government policies provide trade-distorting support to steel producers, or where producers have histories of unfair trade the in U.S. market. Governments are also intervening in markets to maintain capacity, including in the EU.

• Rapid expansion elsewhere comes with falling domestic production. In the United States, by contrast, total steel production capacity fell by 5.5 MMT to 110 MMT in 2019, with world market share shrinking to less than 5% in 2019 from 10% in 2000.
• Section 232 measures delivered near-immediate benefits. Once implemented in 2018, such Sec. 232 steel import measures as 25% tariffs on imported steel and import quotas on select countries helped curb U.S. steel imports by 27% by 2019. Import penetration of the U.S. market fell to 26% of all steel consumed in the United States in 2019, from 35% in 2017.
• Section 232 measures have had no meaningful real-world impact on the prices of steel-consuming products (such as motor vehicles). Econometric analysis shows that price changes in basic steel products had statistically zero or economically negligible causal effects on prices of “downstream,” or steel-using goods, including new motor vehicles, construction equipment, electrical equipment and household appliances, motor vehicle parts, nonresidential construction goods, food at home, and durable goods more broadly—industries accounting for the majority of U.S. steel consumption. This lack of impact is unsurprising, given that steel is just one cost in a long list of inputs to production.
• Widespread exclusions to Section 232 measures mitigate positive economic impacts. Despite benefiting U.S. steel producers and having no discernible impact on steel consumers, Sec. 232 import measures have been progressively undermined by nearly 108,000 product-specific exclusions through July 2020 alone and broad, countrywide tariff exemptions for roughly one-third of all imports.
• Jobs, national security, and the steel industry itself are at risk if Section 232 measures are discontinued or weakened in the post-pandemic economy. The diminished global economic outlook as the world emerges from the COVID-19 pandemic means that the brief reprieve from a global supply glut and nascent recovery enjoyed by U.S. steelmakers is likely to evaporate. Premature relaxation or elimination of Sec. 232 measures, in the absence of any concrete measures to eliminate excess capacity and trade-distorting policies that contribute to the global steel glut, would put the U.S. steel industry at risk, imperiling new investments and hundreds of thousands of good jobs in steelmaking and in other indirect and induced jobs supported by steelmaking activity.
• Relaxing or reversing Section 232 measures also would provide an advantage for low-priced, high carbon-polluting producers overseas.
• A permanent global solution is the best answer. The Biden-Harris administration should press for a permanent multilateral solution to the chronic problem of excess global steel production capacity. But until such a solution is achieved, national security concerns and ensuring a sustainable economic recovery for the steel industry require the continuation of comprehensive Sec. 232 import measures and other policies to preserve the U.S. steel industry.

Introduction

In January 2018, the U.S. Department of Commerce (Commerce) concluded an investigation determining that imports of steel products pose significant risks to U.S. national security and the industry’s ability to maintain operations, grow, and invest in areas essential to national defense, critical infrastructure, and broader economic welfare under Section 232 of the Trade Expansion Act of 1962 (BIS 2018). Sec. 232 provides the president with authority to impose restrictions on products for which an investigation determines that the quantity or circumstances of imports to the United States “threaten to impair the national security” (CRS 2020).1 Beyond supplying high-quality steel in sufficient quantities to meet national defense needs, the U.S. steel industry also plays a critical role in supporting the welfare of other industries essential to the broader health and operation of the economy and government.

Following the Commerce determination, President Trump authorized tariffs of 25% on imported steel products in March 2018.2 The move also provided flexibility in implementation with respect to country of origin and product coverage and allowed domestic parties to petition for exclusion from tariffs where substitute domestic-sourced products were insufficiently available.3 This action follows a continuous thread of presidents—including President Obama—seeking to redress unfair trade practices that for decades have kept the U.S. steel industry on the brink of crisis.

President Biden and his administration undoubtedly will want to reevaluate the policies inherited from their predecessors. To provide perspective for this reevaluation, this report reviews recent developments in global steel markets and analyzes the economic impacts of Sec. 232 steel import measures to assess their efficacy in reversing the long-term trends undermining U.S. steel producers, as well as for evaluating the relative costs and benefits of this policy. Specifically, we examine the effects of Sec. 232 measures on:

• the decades-long problem of chronic global surplus capacity in steel plaguing U.S. producers
• the economic viability of U.S. steel producers
• downstream consumers of steel products
• expected effects of prematurely relaxing or removing Sec. 232 measures

The results presented here demonstrate that Sec. 232 measures on imported steel products remain an important and necessary policy tool. The U.S. steel industry is critical not just for national defense, but also for infrastructure sectors, including electricity systems and equipment, transportation infrastructure and equipment, food and agricultural systems, water systems, energy security and independence, and metal-making and other advanced manufacturing uses. It is also a vital component of the American economy. In 2017, prior to the Sec. 232 import measures and the pandemic, the U.S. steel industry supported nearly 2 million jobs that paid, on average, 27% more than the median earnings for men and 58% more than the median for women (Schieder and Mokhiber 2018; AISI 2018).

Currently, the United States has an excessive dependence on unreliable foreign sources to supply national needs. In 2020, the pandemic and resulting economic contraction showed the dire consequences of reliance on uncertain foreign supplies for personal protective equipment, critical medical goods, and supplies of many other essential products. Policymakers should heed this sober warning when considering how to secure the future for U.S. steel production.

Policy action under Sec. 232 follows decades of a mounting crisis for U.S. steel producers that risks their continued ability to meet the needs of national defense, critical infrastructure, and the broader domestic economy. Steel producers support good-paying, middle-class jobs both directly and indirectly in related industries and throughout local communities where they serve as anchors for regional economies. In 2001, a similar Commerce investigation found “no probative evidence” that imported semi-finished steel products threatened U.S. producers (Bureau of Export Administration 2001). This determination resulted in severe negative consequences for the domestic industry—soon thereafter, nearly 40 U.S. steel producers declared bankruptcy (CRS 2003).

The threat to U.S. steel producers has only worsened in the intervening period, as chronic overcapacity in foreign steel-producing industries has become a permanent feature of global steel markets, driven by countries supporting their national industries on noncommercial terms. A flood of underpriced imports to the United States and third-country markets has done significant harm to U.S. producers and put the future viability of U.S. steel production in jeopardy.

Section 232 measures on imported steel products serve as a last resort to preserve the U.S. steel industry and domestic industrial base. To be certain, the best policy outcome would be for President Biden to achieve a permanent, multilateral solution to the chronic problem of global excess steel capacity. But the failures of decades-long efforts to eliminate global overcapacity through multilateral diplomatic engagement, coupled with foreign governments’ failures to address persistent and growing excess capacity, leave U.S. policymakers to choose between Sec. 232 measures and losing an industry critical for national security and broader economic well-being. Our analysis finds the choice is clear: President Biden should maintain these measures while pursuing multilateral efforts to achieve a long-term solution to unfair competition in global steel. Backtracking on Sec. 232 measures now, without a global solution to surplus capacity, would leave the U.S. industry and steelworkers in an even more precarious situation as more steel production and good-paying American jobs are moved offshore, including to countries with the worst environmental records.

Chronic global overcapacity threatens U.S. steel industry

Over the past several decades, chronic conditions of oversupply have come to define global steel markets—there is significantly more capacity to produce steel than there is demand for steel around the world. This chronic excess capacity is a direct result of policies pursued in many countries to support domestic steel producers on anti-competitive terms, with negative consequences for producers elsewhere around the world. It is also due to the basic economics of production in highly capital-intensive industries like steel, which encourages firms to maintain high levels of production capabilities. For decades, the United States has sought multilateral solutions to this persistent problem to little avail. Scant progress on the excess capacity issue made through diplomatic channels, and continued deterioration of the situation faced by producers operating on a commercial basis, left few other viable options for U.S. policymakers.

Surplus capacity puts downward pressure on prices for steel products, squeezing producer profit margins to an extent that threatens the ability of firms to service debts; to invest in research and development in more advanced products and cleaner production technology; to maintain workers’ jobs, compensation, and retiree pensions; and even to remain financially solvent. Businesses incur both fixed costs and variable costs in the course of steel production. Variable costs change with the quantity a firm produces, whereas fixed costs must be incurred no matter how much a firm produces. For example, in the case of steel, variable costs include the cost of material inputs like iron ore, scrap, and coal, as well as electricity and compensation for workers. However, capital-intensive industries like steel face enormous fixed costs for investments in production facilities and equipment that dominate total costs of production.4

The capital intensity of steel production has several economic consequences that contradict textbook economic models of production and competition. First, in industries like steel, the capital-intensive nature of production means that producers face increasing returns to scale—the more raw steel that is produced, the more efficient it is to produce additional output—such that the minimum efficiency of scale for entering the market with competitive costs is so large as to create a nontrivial addition to industrywide capacity (Crotty 2002). That is, in order to be viable, steelmakers must maintain large production capacity and, when expanding capacity, must add capacity in large chunks. Second, because fixed costs of production dominate variable costs, it is almost always desirable for producers to operate near full capacity in order to minimize the average cost of production. For producers in many countries, production exceeds what can be consumed in domestic markets, and the excess must be disposed of through exports.

Finally, the capital invested in fixed assets is quite specific, meaning the equipment cannot be easily redeployed to other uses outside of steel production, as is typically assumed in textbook models of economic competition. This means that, typically, productive capacity of financially nonviable steel producers is not removed from the market, but rather acquired by other producers in better financial standing. Thus, the market mechanism of price competition and creative destruction does not work well to self-regulate excess capacity in the industry (Crotty 2002). In fact, the OECD finds that foreign governments maintain policies and implement barriers that prevent the contraction of steelmaking capacity during economic downturns (Rimini et al. 2020). Combined, these features of the steel industry create incentives for producers to build big and run hot, no matter what other producers in the market do. But when all producers follow this logic, the result, in aggregate, is chronic overinvestment in productive capacity.

In order to maintain the viability of national steel industries under such financial conditions, many countries have instituted policies designed to maintain and expand production on noncommercial terms or other policies impermissible under international trade rules like the World Trade Organization (WTO)’s Agreement on Subsidies and Countervailing Measures. Commerce and the U.S. International Trade Commission, as well as the WTO, regularly find such measures do significant material harm to U.S. producers operating on a commercial basis, discussed in further detail in the box below. At the time of the Sec. 232 report, Commerce had authorized 164 orders on steel imports for illegal dumping or trade-distorting subsidies by 40 countries, with another 20 ongoing investigations (BIS 2018). Some foreign producers also benefit from other policies favorable to domestic industries but not explicitly prohibited by international agreements, such as discretionary regulatory forbearance of environmental standards, discussed later in the report, in the section “Retreating from Section 232 measures would squeeze vulnerable producers, increase greenhouse gas emissions.”

As a result, international disputes over steel capacity and multilateral efforts to resolve them are not new. The European Coal and Steel Community was formed in the aftermath of World War II to resolve continental tensions over steel production, providing a foundation for the European Union. The United States has been involved in international steel diplomacy since at least the Lyndon Johnson administration. In 1989, President George H.W. Bush launched efforts to reach a global agreement to abolish steel production subsidies. In the late 1990s, President Clinton initiated a “Steel Action Plan” in response to a flood of underpriced steel imports being dumped in the U.S. market. On a bilateral basis, President Obama pressed steel capacity issues with China for years through the Strategic and Economic Dialogue. He also moved multilateral partners to launch the Global Steel Forum at the 2016 G20 leaders’ summit, and to find common ground and establish a level playing field through the decades-old Organisation for Economic Co-operation and Development (OECD)’s Steel Committee (White House Office of the Press Secretary 2017).

Despite these efforts, capacity for global steel production continues to substantially exceed global demand for steel products, as shown in Figure A. In 2000, the peak year before a recession and the year before China acceded to the World Trade Organization, global excess capacity of 282 million metric tons already exceeded production by one-third of total output (850 MMT). With surplus capacity already at substantial levels, capacity growth outstripped steel production growth for the next decade and a half. From 2000 to 2015, production volume increased by 91% to 1,625 MMT, while excess capacity grew 166% to 752 MMT.

Figure A

By the mid-2010s, total world production capacity stabilized near 2,400 MMT, and increased demand for steel products led production to increase and capacity utilization rates to rise. However, by 2017 excess capacity still remained high, at 616 MMT, and capacity utilization remained below the level in 2000. Only beginning in 2018 and 2019, coinciding with Sec. 232 measures, did world capacity utilization surpass the level in 2000. The global economic slowdown in 2020 resulting from the COVID-19 pandemic once again sent excess steel capacity up and dragged the capacity utilization rate down. By 2020, excess capacity reached 633 MMT, or the equivalent of 5.8 times total U.S. production capacity.

That world production capacity stabilized after 2014 belies significant changes in the composition of steel production capacity by country. Figure B illustrates these changes in the composition of global steel supply by plotting the production capacities of the world’s largest steel-producing countries and country groups in 2000 on the horizontal axis against the percentage change in steel capacities in these country and country groups from 2000 to 2019 on the vertical axis; the size of each bubble indicates each country’s relative share of global steel capacity in 2019. China, the world’s largest steel producer, expanded production capacity by 418% since 2000, such that by 2019 it controlled just shy of half of global steel capacity.

Just the additional capacity installed in China since 2000 exceeds the combined capacity in 2019 of all other individual countries depicted in Figure B. During this time, U.S. capacity contracted 5.5 MMT, and its global market share was cut in half to less than 5% in 2019 from 10% of world capacity in 2000. Although Chinese producers are the largest culprits driving chronic excess steel capacity, they are far from alone in aggressive expansions that have displaced other producers and reshuffled the structure of world production. Other major steel-producing countries achieving rapid capacity growth between 2000 and 2019 include India (95 MMT, 280%), Turkey (30 MMT, 151%), Iran (27 MMT, 300%), Korea (26 MMT, 47%), Vietnam (22 MMT, 2,036%), Russia (21 MMT, 31%), Brazil (17 MMT, 51%), Mexico (9 MMT, 46%), and Taiwan (8 MMT, 40%). Each of these countries features state-dominated or state-directed economies, trade-distorting government policies supporting steel producers, or a history of shipping unfairly traded steel products to the U.S. market.

Figure B

A multilateral solution to the chronic problem of global excess steel capacity remains essential. But until that time, the inefficacy of market mechanisms to address surplus overcapacity and national policy distortions introduced by foreign trade partners will continue plaguing U.S. steel producers, risking the industry’s survival at a scale necessary to meet national security demands.

Section 232 measures improve industry conditions, spur investments and jobs

Given that the problem of global excess capacity for U.S. steel producers is clear, policymakers should ask: “Are Section 232 measures on imported steel working to improve their conditions?” In considering this question, it is important to understand that the effectiveness of relief has been undermined by considerable “leakage” from Commerce-granted exclusions and broad countrywide exclusions that have curtailed tariff coverage on imported steel. Nevertheless, our analysis demonstrates that Sec. 232 measures remain critical to the long-term prospects of U.S. steel producers. A survey of publicly available sources reveals that following implementation of Sec. 232 measures, U.S. steel producers announced new investments, upgrades, plant expansions, and reopenings of idled facilities in at least 15 states, including plans to invest more than $15.7 billion in new or upgraded steel facilities, creating at least 3,200 direct new jobs, many of which are now poised to come online (see Appendix Table 1A). In addition, more than $5.9 billion was invested in plant acquisitions by nine firms, as part of industry restructuring to increase efficiency, preserving additional jobs at those facilities (see Appendix Table 1B).8

Individual anecdotes provide a suggestive initial glimpse at the effects of Sec. 232 steel import measures. But a more systematic assessment of available data demonstrates that the import measures coincided with improving conditions for U.S. producers—prior to the pandemic-related global recession beginning in 2020. Relief from the pressure of anti-competitive steel imports facilitated recovery of industrywide sales margins (a measure of profitability), production and capacity utilization rates, and a resurgence of new investment in steel industry fixed assets. Importantly, as discussed in the next section, these measures achieved improvements for U.S. steel producers without causing harm to downstream consumers of steel products in the United States.

From the trough of the Great Recession in 2009, U.S. steel imports rose sharply from 14.7 MMT to 40.2 MMT by 2014, as seen in Figure C. A series of nearly 69 new anti-dumping and countervailing duty determinations between 2014 and 2016 curbed the inflow of steel imports to 30 MMT in 2016—temporarily (USITC 2021).9 However, many foreign producers evaded these import surge measures by relocating steel production and processing to third countries, and imports climbed once again, reaching 34.5 MMT in 2017. But the Sec. 232 measures successfully slowed the pace of imports in 2018 and 2019, when imports fell to just 25.3 MMT. Overall, the volume of steel imports fell 27% between 2017 and 2019—before the pandemic’s “Great Lockdown” slowed U.S. and global economic activity. Separate data analysis shows that import penetration of the U.S. steel market fell to 26% of all steel consumed in the United States in 2019, from 35% in 2017. As a result, the rate of capacity utilization for U.S. steel producers rose to 80% in 2019 from 72% in 2017 (WSA 2020a; OECD 2020a). Commerce (BIS 2018) found that an 80% capacity utilization, sustained over the business cycle, is a critical threshold for U.S. steel producers to achieve long-term financial viability.

Figure C

Sec. 232 measures placing tariffs and quotas on foreign steel products were intended to create some breathing room for U.S. steel producers to recover market share and sustainable financial conditions enabling them to increase domestic production—which they did. The Sec. 232 measures have afforded the U.S. steel industry an opportunity to recover to a level of financial performance not experienced since before the Great Recession (Figure D), although this recovery has been undermined as exemptions from Sec. 232 measures allowed “leakage” of uncovered imports, and as recession from the pandemic’s 2020 Great Lockdown set in. Following the Great Recession of 2007–2009, U.S. steel producers strained to achieve profitability. From the third quarter of 2009 through 2016, net income for the U.S. steel industry averaged just $73 million. Over the same period, net income as a share of sales—a measure of profitability—averaged 0%. In 2018, the year Sec. 232 measures were first imposed, net income in the steel industry reached $7.9 billion, or 6.4% of sales—its highest level since the real estate construction boom that preceded the Great Recession. Since then, however, the domestic industry has faced serious challenges. In 2019, the industry’s net income receded to $2.9 billion, and in 2020 it sunk back into negative territory, posting losses with the pandemic-induced global recession.

Figure D

U.S. steel producers recovered with the Sec. 232 measures, bringing idled capacity back online with expectations for improving market conditions. However, more recently, the erosion of import coverage under Sec. 232 measures has coincided with declining prices and financial performance in the industry. Although the Sec. 232 measures initially covered all steel imports, Commerce has granted nearly 108,000 product exclusion requests from Sec. 232 measures as of July 2020 (CRS 2020; U.S. Department of Commerce 2021). A number of significant steel-producing countries, including Argentina, Brazil, Canada, Mexico, and South Korea, also obtained outright exemptions from Sec. 232 measures or quantitative quotas to replace import tariffs. These exclusions and exemptions significantly curtailed the coverage of Section 232 measures, although the measures remain significant in reversing the trend of declining viability of the U.S. steel industry. Today, a majority of steel products are imported to the United States either on a duty-free basis or under Sec. 232 product exclusions.

The U.S. steel industry’s initial recovery under Sec. 232 measures and the expectations of relief from conditions of chronic global excess capacity helped draw new investments into U.S. steel production (Figure E). New investment, adjusted for inflation, surpassed $5 billion in 2018 and reached nearly $5.9 billion in 2019. However, the dwindling coverage of Sec. 232 measures mentioned above and resulting decline in net income seen in Figure D will make it difficult for the industry to sustain this investment trend and could put many producers in further financial jeopardy. As discussed earlier, capital-intensive investments to upgrade and expand production are long-lived fixed costs that only can be reversed at prohibitively high cost. Firms that have made substantial new investments under the expectations of strong domestic demand and continuing Sec. 232 import relief may be deterred from future investments in technological upgrading and be squeezed by debt service commitments; those exploring expansion will likely shelve their plans.

Figure E

Despite a 25% tariff, the Sec. 232 measures had a limited effect on U.S. import prices of steel products, as seen in Figure F. The product categories in Figure F represent roughly three-fourths of total U.S. steel imports. Unit prices for imports of most steel products increased from 2017 to 2018—the year Sec. 232 import measures began. But then, import prices fell in 2019 and again in 2020, such that overall, averaged across all products, the import price of steel fell to $833 per metric ton in 2020 from $845 per metric ton in 2017.

Figure F

Sec. 232 import measures coincided with and contributed to an increase in prices for steel products in the U.S. market, as can be seen in Figure G, comparing prices paid to domestic steel producers relative to those paid by U.S. steel consumers purchasing comparable products on international markets for import. Unsurprisingly, both U.S. producer and import prices follow a common trend, although imports generally are lower priced than U.S.-made steel, as excess capacity and trade-distorting foreign government policies depress global prices. As the world emerged from the Great Recession in July 2009, particularly with China’s outsized stimulus investments in infrastructure and real estate construction (Hersh 2014), steel prices around the world began rising sharply. Steel demand was so strong that it pushed up prices for key steel inputs globally, including iron ore and coal (World Bank 2020). Then, as discussed in Section 2 above, expanded world steel production and surplus capacity through the middle of the last decade began driving prices down.

Figure G

Following implementation of Sec. 232 measures, Figure G shows domestic steel prices rose faster than U.S. import prices. This is due to a combination of the Sec. 232 measures, other trade remedies—including anti-dumping and countervailing duty orders—and the appreciation in value of the U.S. dollar relative to foreign currencies, making foreign products comparatively less expensive in dollar terms. These factors drove a wedge between domestic and foreign prices, which enabled U.S. steelmakers to achieve more sustainable operating margins. As input prices again eased in late 2018, steel prices fell in the U.S. market and globally—although they likely would have fallen further were it not for Sec. 232 import measures.

In recent months, U.S. and foreign steel prices are on the upswing—likely a temporary phenomenon caused by the lag between increasing demand as parts of the world economy recover from the Great Lockdown and the re-employment of steelmaking capacity—which, for blast furnace operations in particular, can take time and may occur only after market conditions create confidence that a facility can operate at a high level of capacity for a sustained period. In this environment, maintenance of Sec. 232 import measures will remain critical to ensuring the economic stability and financial viability of the U.S. industry. Country- and product-specific trade remedies, though significant, on their own have proven insufficient to abate the risk to the U.S. steel industry from anti-competitive imports and chronic excess production capacity.

Steel consumers face negligible effects from Section 232 measures

An important concern in assessing the impacts of Sec. 232 measures on imported steel products is how these measures affect downstream industries and consumers of products that use steel inputs. Indeed, as Sec. 232 measures were going into effect, a group of business lobbying associations representing downstream users sent a joint letter to the U.S. Trade Representative expressing this concern and claiming “significant harm” from this policy (Industry Week 2018). Our analysis in this section shows this claim proved incorrect.

Critics of import measures more broadly, including those levied in 2018 against China for unfair trade practices pertaining to technology transfer, intellectual property, and innovation (USTR 2018), often point to a recent Federal Reserve study purporting to find that tariffs are associated with negative outcomes for the U.S. manufacturing sector (Flaaen and Pierce 2019). However, this analysis should be treated with a healthy dose of skepticism due to myriad methodological issues that introduce statistical bias and call into question the validity of their findings.10 Weaknesses of Flaaen and Pierce’s (2019) results are illustrated in their own Figures 4 and B3, which demonstrate that import protection has no statistically significant impact on manufacturing employment, industrial output, or producer prices for virtually all of the period under consideration.

Given the inherent shortcomings of Flaaen and Pierce 2019, we implement an empirical strategy focused more narrowly on steel products and explicitly evaluating the causal effect of changes in the price of steel inputs on the prices of goods using steel. Our econometric analysis demonstrates that this relationship ranges from statistically insignificant (i.e., not statistically different from zero effect) to negligible. In other words, the statistical evidence does not support claims of harm from Sec. 232 measures that were predicted by certain steel-using businesses. This fact should not be surprising: Even in the downstream industries consuming the most steel, steel inputs amount to a minor share of overall production costs.

Harm to downstream industries would occur if Sec. 232 measures significantly increased steel prices, causing increased costs for producers or consumers of primary steel-containing goods, and then those costs squeezed profit margins or consumer welfare—by forcing consumers to either pay more for or consume less of a given product. To assess this linkage between steel input prices and end-user prices, we employ standard, related, and time-tested econometric techniques known as Granger causality analysis and vector autoregression (Granger 1969; Sims 1980). Vector autoregression (VAR) is a statistical method for modeling a system of variables and their interrelationship and co-evolution over time. In this case, we model (1) the price of primary steel inputs, (2) the price of steel-consuming products, and (3) the effective federal funds rate.11

Granger causality analysis uses the VAR model to test for evidence of a statistically causal relationship between the variables in the model. If past values of variable 1 are shown to significantly predict current values of variable 2, then it can be concluded that variable 1 “Granger-causes” variable 2. While the price variable used in this modeling includes the effects of Sec. 232 tariffs and quotas, the results of the statistical test are not limited to the effects of Sec. 232 measures, but rather evaluate whether a change in prices resulting from any factor causes a change in the price of the steel-using good. Technical discussion of this methodology and detailed results are presented in Appendix 2.

We summarize the results of this causal analysis in Table 1. Each row of the table presents a separate VAR model relating the price of a steel-containing product with the price of its most relevant primary steel input(s) and reports the causal effect found on end-use product prices. The end-use products investigated represent the U.S. industries consuming the largest volume of steel products: nonresidential construction, motor vehicles, motor vehicle parts, construction machinery, electrical equipment and household appliances, and food processing (food consumed at home). We also evaluate the possible impacts of Sec. 232 steel measures at a broader level by modeling the effects of steel product prices on aggregated prices for durable goods.

As shown in Table 1, this analysis finds no discernible effect of steel prices causing price changes in new motor vehicles, motor vehicle parts, construction machinery, electrical equipment and household appliances, or, broadly, durable goods. These results, therefore, suggest that even if Sec. 232 measures caused an increase in the price of steel products, one would not expect a significant impact on the price of downstream goods. For prices of nonresidential construction goods and food consumed at home, the price of relevant steel inputs is found to be statistically significant in causing changes in the prices of steel-using products.12 While finding a statistical relationship between steel input prices and final goods prices, the same analysis shows that the economic significance of the impact is negligible: A 1% increase in steel input prices caused a 0.1% change in the price of construction goods and a less than 0.05% change in the price of food at home. However, as discussed in Appendix 2, causal analysis suggests the relationship between steel inputs and construction goods actually runs in the opposite direction, with demand for construction goods driving prices in the market for intermediate inputs.

Table 1

To recap, while conceptually a relationship exists between input prices and final goods prices, econometric analysis of the causal relationship between prices finds effects ranging from statistically zero to essentially nothing. Sec. 232 measures simply did not have a meaningful, real-world impact on prices for steel-consuming products. This fact should not be surprising. Even in the industries that consume the largest volumes of steel products, steel is just one cost in a long list of inputs to production. Despite these industries accounting for the lion’s share of steel consumption in the U.S. economy, the cost of their steel inputs is minor relative to their gross production. As shown in Table 1, the steel content as a share of total production ranges from 1% in food consumed at home to 9.8% in the motor vehicle parts industries. Illustrating the point in dollar terms, the average passenger car contains roughly 900 kg of steel (WSA n.d.). At a current cost of $1,048 per metric ton, the steel inputs amount to just 2% of the sales price for the average new U.S. car (Steel Benchmarker 2020; Kelley Blue Book 2020). In contrast, electronics components make up roughly 40% of a new car’s price (Deloitte 2019).

Retreating from Section 232 measures would squeeze vulnerable producers, increase greenhouse gas emissions

Thus far, we have seen that Sec. 232 import measures have helped improve market conditions for U.S. steel producers amid chronic global excess capacity that threatens their financial viability. We also have seen that the impact of these measures on steel-consuming U.S. industries has ranged from zero to economically insignificant. Furthermore, the benefits of this policy have eroded since it began as more steel imports have been exempted from the Sec. 232 regime. As the world looks to move forward from the economic shock of the Great Lockdown caused by COVID-19, it is clear that eliminating or even further relaxing the steel import measures likely would pose serious economic consequences for U.S. steel producers. Two important points are noteworthy here.

First, a slow and uneven recovery from the 2020 economic downturn is expected, with global demand for steel products uncertain. The International Monetary Fund (IMF) recently revised down its global economic growth forecast for 2021; it projects “limited progress toward catching up to the [expected] path of economic activity for 2020–2025” (IMF 2020).13 Families around the world have suffered deep economic scarring from lost jobs and income and depleted savings—not to mention, tragically, the many who have lost prime wage-earners. Millions of people worldwide who contracted the virus are likely to suffer long-term effects, reducing prospects for employment and earnings and allocating a larger share of disposable income toward health care services and away from goods consumption. At the same time, the downturn and its long-lasting effects have dampened public-sector revenues at a time when governments have undertaken unprecedented expenditures meeting the public health crisis and providing social protections. The enduring effects of this shock will dampen, in the near term, a recovery of household consumption and, in turn, business investment. In the longer term, the human toll will dampen prospects for economic potential, dragging down investments in human and fixed asset capital and the productivity growth these investments provide.

Economic recovery, of course, is contingent on how well world governments abate the global health crisis, but it is clear that even under optimistic scenarios, demand for steel production will remain muted for some time. U.S. steel demand declined 16% in 2020, and in 2021 it is expected to remain more than 10% below 2019 levels (WSA 2020c). Globally, steel demand declined 6.4% from 2019 to 2020 and is forecast to remain nearly 3% below 2019 levels (OECD 2020c). At the same time, countries have not retreated from policy efforts to prop up national steel industries (see text box, “Widespread government interventions drive unfair trade in steel products”) and are continuing to install additional productive capacity. The OECD (2020b) projects that by 2022, producers will add as much as another 3% of steelmaking capacity worldwide, concentrated in Asia and the Middle East.

Together, these trends point to increased excess steelmaking capacity and lower capacity utilization rates that would drive prices down and squeeze U.S. steel producers who face competition with imports produced on a noncommercial basis. These are exactly the pressures Sec. 232 measures are designed to address, in the absence of multilateral agreements to manage excess capacity. Retreating from these measures now, particularly after many U.S. companies committed to new investments in production (Figure E; Appendix 1), would leave U.S. steel producers in untenable financial positions, further jeopardizing their capacity to meet national security needs.

Second, a significant ancillary benefit of Sec. 232 import measures has been to divert steel production to more environmentally sustainable producers. Relaxing Sec. 232 measures would reverse this progress as the world looks to decarbonizing and achieving net-neutral emissions by midcentury. The U.S. steel industry is one of the cleanest and most energy-efficient steel industries globally. A 2019 report measuring the CO2 emissions intensity of steel industries in 15 major steel-producing countries ranked U.S. steelmakers among the least CO2 intensive industries—with industries in Brazil, Canada, China, France, Germany, India, Japan, South Korea, and other countries having higher CO2 emissions intensity (Hasanbeigi and Springer 2019, Figure 14).

Even this analysis understates that difference in environmental impact, as it does not account for the substantial pollution from ocean freight required to transport raw materials and finished products in supply-chain webs around the world before foreign steel products can reach the U.S. market (ENVI 2020). If Sec. 232 measures are relaxed, it is precisely the most polluting national steel industries, in countries that have rapidly expanded capacity at the expense of more efficient producers, that stand to capture marginal changes in market share. And as excess capacity further squeezes prices and profit margins, firms will face difficulty investing in new technologies to allow for greener steel production and will risk being shut out of markets as consumers develop preferences for low-carbon products.

Conclusion: The Section 232 trade measures helped slow the flood of unfair imports that was squeezing the U.S. steel industry without hurting downstream steel-using producers and consumers

Surging steel imports have undermined domestic steel production, prices, employment, profits, investments, and the fundamental health of the U.S. domestic steel industry. Global steel surpluses are the result of chronic global excess steelmaking capacity in major exporting countries. The steel Section 232 trade restraints imposed in 2018, including both tariffs and quotas on imports from selected countries, helped slow the flood of steel imports. Following imposition of these measures, U.S. steel output, employment, capital investment, and financial investment all improved.

Meanwhile, statistical analysis in this report has demonstrated that Section 232 measures have had no economically significant impacts on the prices of downstream products. Despite the benefits of the Section 232 tariffs for the domestic steel industry and its workers, and the minimal impacts of trade restraints on downstream industries, these measures have been progressively weakened by nearly 108,000 product-specific exclusions and broad tariff exemptions for a number of countries.

The domestic steel industry is just beginning to emerge from the depths of the COVID-19 recession with a steep hill to climb, given widening excess global steel capacity. With the right policies and major investments planned by the new administration in economic rebuilding, clean energy, and infrastructure construction, U.S. steel producers can be poised for a substantial upswing in employment, output, and investment that fuels growth in clean, efficient, state-of-the-art domestic steel production. The window to this opportunity could be slammed shut by the premature and unplanned elimination of the Section 232 import measures.

Acknowledgments

The authors thank Jori Kandra for technical and research assistance and Colleen O’Neill and Lora Engdahl for editing assistance. This research was made possible by support from the Partners for American-made Steel.

About the authors

Adam S. Hersh, Ph.D., is director of Washington Global Advisors, LLC, and a research associate at the Political Economy Research Institute, University of Massachusetts Amherst. Hersh was formerly chief economist for the Congressional Joint Economic Committee, Democratic staff; senior economist at the Franklin and Eleanor Roosevelt Institute and the Center for American Progress; a visiting scholar at Columbia University’s Initiative for Policy Dialogue and the Shanghai University of Finance and Economics; and a research fellow at the University College, London’s Institute for Innovation and Public Purpose. He is a contributing author with Joseph Stiglitz of Rewriting the Rules of the American Economy: An Agenda for Growth and Shared Prosperity (2015).

Robert E. Scott joined the Economic Policy Institute in 1996 and is currently director of trade and manufacturing policy research. His areas of research include international economics, trade, and manufacturing policies and their impacts on working people in the United States and other countries, the economic impacts of foreign investment, and the macroeconomic effects of trade and capital flows. He has published widely in academic journals and the popular press, including The Journal of Policy Analysis and Management, The International Review of Applied Economics, and The Stanford Law and Policy Review, as well as The Los Angeles Times, Newsday, USA Today, The Baltimore Sun, The Washington Times, and other newspapers. He has also provided economic commentary for a range of electronic media, including NPR, CNN, Bloomberg, and the BBC. Mr. Scott has a Ph.D. in economics from the University of California, Berkeley.

Endnotes

1. 19 U.S.C. §1862; https://www.law.cornell.edu/uscode/text/19/1862.

2. Adjusting Imports of Aluminum Into the United States, 83 Fed. Reg. 25849–25855 (March 15, 2018).

3. Adjusting Imports of Steel Into the United States, 83 Fed. Reg. 25857–25877 (March 15, 2018).

4. The capital-to-labor ratio for primary metals producers is 76% higher than for durable goods manufacturing industries overall. See BLS (2020).

5. This section is based, in part, on information summarized in Examples of Policies and Practices Contributing to the Global Excess Capacity Crisis, a report by the American Iron and Steel Institute and the Steel Manufacturers Association included at the end of this report.

6. There are two anti-dumping orders in place against Canadian steel products, and there are both anti-dumping and countervailing duty orders in place against wind towers, a major steel-using product. South Korean steel orders include six countervailing duty orders and 26 antidumping orders. EPI analysis of USITC (2021).

7. Countries receiving Chinese direct foreign investment in steel include Cambodia, Malaysia, Indonesia, Myanmar, Pakistan, the Philippines, Bolivia, Vietnam, the United Kingdom, and the Netherlands. See OECD (2020b).

8. These are direct steelmaking jobs; the investments would also generate indirect employment through the goods and services procured in expansion products, as well as induced employment generated by the incomes from direct and indirect employees.

9. The USITC (2021) lists 276 anti-dumping and countervailing duties in effect on steel products (categories ISM, ISO, and ISP) as of December 28, 2020, and of those, 69 orders went into effect between 2014 and 2016.

10. It is worth briefly considering several reasons why. First, the core explanatory variable—“import protection”—ignores the actual incidence and evolution of protection over time as more products received exclusions from tariffs. Second, their statistical model explicitly embraces violations of the core assumptions on which the statistical method is built, biasing the results. In particular, equation 7 specifies measures of import protection, input costs, and foreign retaliation as “independent” variables associated with the dependent variables of manufacturing employment, output, and producer prices. In fact, as Flaaen and Pierce appropriately theorize, input costs and foreign retaliation are, at least in part, caused by import protection. Finally, Flaaen and Pierce’s analysis conflates the effects of Sec. 232 import measures with Sec. 301 trade remedies. Conditions of chronic excess global steel capacity—explained in Section 2 above—mean that market conditions are significantly different for steel products than for other manufactured goods, suggesting that pooling data for steel products and other manufactured goods more broadly is inappropriate and may bias estimates of the statistical significance.

11. The federal funds rate—the interest rate at which depository institutions borrow and lend federal balances held at Federal Reserve Banks—is the primary target for Federal Reserve monetary policy actions and is linked both in theory and in practice to changes in price levels as well as to the level of demand for goods and services across the economy.

12. The causal effect of steel prices on food-at-home prices shows only weak statistical significance, at the 90% probability threshold; the model for other significant goods found 95% to 99% probability.

13. What’s more, as dour as the IMF’s assessment is, their forecasts are notorious for being overly optimistic. See Rosnick and Weisbrot (2007).

14. Producer Price Index by Commodity: Metals and Metal Products: Iron and Steel (WPU101); Producer Price Index by Commodity: Metals and Metal Products: Cold Rolled Steel Sheet and Strip (WPU101707); Producer Price Index by Commodity: Metals and Metal Products: Hot Rolled Steel Sheet and Strip, Including Tin Mill Products (WPU101703); Producer Price Index by Commodity: Metals and Metal Products: Hot Rolled Steel Bars, Plates, and Structural Shapes (WPU101704); Producer Price Index by Commodity: Metals and Metal Products: Steel Wire (WPU101705); Consumer Price Index for All Urban Consumers: New Vehicles in U.S. City Average (CUUR0000SETA01); Producer Price Index by Commodity: Transportation Equipment: Motor Vehicles Parts (WPU1412); Producer Price Index by Industry: Construction Machinery Manufacturing (PCU333120333120); Producer Price Index by Commodity: Inputs to Industries: Net Inputs to Nonresidential Construction, Goods (WPUIP2312001); Producer Price Index by Industry: Electrical Equipment and Appliance Manufacturing (PCU335335); Consumer Price Index for All Urban Consumers: Food at Home in U.S. City Average (CUSR0000SAF11); Personal consumption expenditures: Durable goods (chain-type price index) (DDURRG3M086SBEA); and Effective Federal Funds Rate (FEDFUNDS).

15. Motor vehicle parts manufacturing requires significant inputs from both hot-rolled sheet and strip as well as hot-rolled bars, plates, and structural shapes. Electrical equipment and household appliances require significant inputs from both cold-rolled sheet and strip and carbon steel wire. Therefore, these products are modeled as a four-equation VAR of the form

where and are 4 × 4 matrices of coefficients.

Appendix 1: New and expanded U.S. steel production under Section 232 measures capacity

Appendix Table 1A

Appendix Table 1B

Appendix 2: Methodology for analyzing causal relationship between steel prices and steel-consuming industries

This appendix outlines the methodological approach for assessing how Sec. 232 measures on imported steel products may affect downstream industries and consumers of products that use steel inputs. Harm to downstream industries and consumers could occur if Sec. 232 measures caused an increase in prices for steel products paid by U.S. users of steel and if those price increases were passed through to producer or consumer prices for steel-embodying goods. In order to assess this possibility, we evaluate a more basic question: Do changes in prices of basic steel products cause changes in steel-using products? This question asks whether any change in steel prices is a significant determinant of goods prices that use steel as an intermediate input, irrespective of what factors cause a change in steel prices.

Data and methodology

To evaluate this question, we estimate reduced-form vector autoregressions (VARs) that model the variables of interest as an interrelated system that co-evolves over time (Sims 1980). The VAR is an attractive analytical tool because it does not force an assumed structural form onto the data. Each variable in the system is modeled jointly as a function of its past values and the past values of the other related variables in the system. After estimating the system, we can evaluate causal relationships between the variables by testing whether past values of one variable are statistically significant determinants of the current value of another variable, following Granger (1969).

Our variables of interest are (1) prices for steel products, (2) prices for steel-using products, and (3) the effective federal funds rate—the interest rate at which depository institutions borrow and lend reserve balances held at Federal Reserve Banks.14 This interest rate is the primary target for Federal Reserve monetary policy actions and is linked both in theory and in practice to changes in general price levels, as well as to the level of demand for goods and services across the economy via the Taylor Rule (Taylor 1993). Data are observed monthly and drawn from the Federal Reserve Bank of St. Louis’s FRED Economic Data, spanning December 2001 to January 2020, or two business cycle expansions, other than for steel wire, for which available data begin in July 2004. Univariate analysis with a modified Dickey-Fuller test (Cheung and Lai 1995) fails to reject the null hypothesis of a unit root for each variable under consideration. While the individual variables are nonstationary (integrated of order one, or first-difference stationary), tests with Johansen’s procedure show that there is no cointegration—or a stable, long-run relationship—between the variables (Johansen 1995), and the system can be modeled with a VAR, as opposed to a vector error correction model.

The VAR model consists of

where   is the natural log of price at time of the relevant steel product input price,   is the natural log of the price of the steel-using product, and is the natural log of the effective federal funds interest rate. The pairings of steel product input prices and steel-using product prices are given in Section 4, Table 1.15 The model estimates parameters , to , and , which are, respectively, a vector of constant terms, 3×3 matrices of coefficients relating the current dependent variable to past values of the independent variables, and a vector of randomly distributed residual with mean zero and uncorrelated across time.

The specific number lags of the dependent and independent variables specified varies for each set of steel product and steel-consuming goods modeled, and they are chosen with some subjectivity, though guided by minimizing a battery of statistical tests, including the likelihood ratio test, the final prediction error, Akaike’s information criterion, Schwarz’s Bayesian information criterion, and the Hannan and Quinn information criterion (Neilsen 2001; Lütkepohl 2005). Results were robust to alternative lag-length specifications. The VAR parameters were estimated simultaneously by the “seemingly unrelated regression” method of Zellner and Theil (1962). Post-estimation, the statistical assumptions were tested to confirm that the VAR parameters are stable (with eigenvalues lying within the unit circle), and that the residual is normally distributed and not serially correlated, indicating that the models are well-specified.

The specific parameters estimated that define the structures of VARs are typically of less concern than how the system behaves when there is an exogenous change in one of the variables. In this case, we are concerned whether a change in the price  causes a change in , evaluated with a Granger (1969) causality test. This evaluates the hypothesis that the coefficients on are jointly statistically significant in determining  against the null hypothesis that the coefficients are all equal to zero. If the test statistic exceeds a critical value at a 95% probability or higher, we can reject the null hypothesis and conclude that Granger-causes . In the event we identify a significant causal relationship, then the system of equations making up each VAR can be used to simulate the effect on of a shock to  by simulating an impulse response function.

Results

Appendix Table 2 reports the Wald test statistic χ² and the associated probability for rejecting the null hypothesis of zero causal effect for each pair of prices. For the majority of end-use products considered, we find no statistical evidence that steel input prices affect the price of end-use products (<95% probability). This means that a change in steel prices is expected to have no effect on the price of end-use goods. We do find statistically significant causal effects (>95% probability) of steel input prices on the prices of nonresidential construction goods and food at home.

Appendix Table 2

For end-use goods experiencing a causal effect of steel prices, we estimate the impact of a 1% increase in steel input prices using an orthogonalized impulse response function, with results summarized in the final column of Appendix Table 2. For each end-use good, the shock from an initial change in steel prices reaches its maximum impact on end-use prices in the following one to two months, then gradually dissipates to zero over the ensuing months, meaning there is no permanent effect on prices.

These were not the only statistically significant causal relationships identified in the VAR modeling. In a majority of the models, Granger analysis finds that the effective federal funds rate has a causal effect on steel product price levels, as theory would predict. We also find that prices of nonresidential construction goods have a causal effect on prices of hot-rolled bars, plates, and structural shapes—more than five times the size of the effect of hot-rolled bar prices on nonresidential construction goods—suggesting that demand for construction projects leads demand, and therefore pricing, of intermediate inputs to construction.

Appendix 3: Countries of concern—examples of policies and practices contributing to the global excess capacity crisis

Interventionist policies by governments around the world have driven a buildup of excess steel production capacity. Because China is the largest source of global excess steel capacity, the crisis is frequently mischaracterized as “just a China problem.” However, as a report from the American Iron and Steel Institute and the Steel Manufacturers Association shows, numerous countries contribute to global overcapacity through state interventions that commonly include: the provision of low-cost inputs, subsidized loans and equity infusions, grants, tax breaks, support for acquisition of overseas raw materials, export restraints on domestically produced raw materials, state-led debt restructuring and other corporate reorganizations, local content requirements, transnational subsidies for establishing third-country operations, and other measures that forestall the exit of inefficient capacity. Read the report, Examples of Policies and Practices Contributing to the Global Excess Capacity Crisis, to learn how global steel overcapacity is fueled by government policies in South Korea, Japan, Vietnam, Indonesia, the Russian federation, Brazil, the Netherlands, Germany, the United Kingdom, Italy, Canada, and Mexico.

References

American Iron and Steel Institute (AISI). 2018. The Economic Impact of the American Iron and Steel Industry. May 2018.

American Iron and Steel Institute (AISI). 2020. Comments Regarding Foreign Trade Barriers to U.S. Exports for 2021 Reporting [Docket Number USTR—2020—0034]. October 2020.

Bureau of Economic Analysis (BEA). 2021. “Gross Private Domestic Investment: Fixed Investment: Nonresidential: Implicit Price Deflators” [Excel file]. Federal Reserve Bank of St. Louis. Accessed January 3, 2021.

Bureau of Export Administration. 2001. The Effect of Imports of Iron Ore and Semi-Finished Steel on the National Security. October 2001.

Bureau of Industry and Security (BIS). 2018. The Effect of Imports of Steel on the National Security: An Investigation Conducted Under Section 232 of the Trade Expansion Act of 1962, as Amended. January 2018.

Bureau of Labor Statistics (BLS). 2020. “Input-Output Data for the U.S. Economy for the Historical Years 1997–2019, and for the Projected Year 2029” [zipped Excel file]. Accessed January 25, 2021.

Bureau of Labor Statistics (BLS). 2021a. “Import Price Index (End Use): Iron and Steel Mill Products” [Excel file]. Federal Reserve Bank of St. Louis. Accessed January 1, 2021.

Bureau of Labor Statistics (BLS). 2021b. “Producer Price Index by Commodity: All Commodities” [Excel file]. Federal Reserve Bank of St. Louis. Accessed January 1, 2021.

Bureau of Labor Statistics (BLS). 2021c. “Producer Price Index by Commodity: Metals and Metal Products: Iron and Steel” [Excel file]. Federal Reserve Bank of St. Louis. Accessed January 1, 2021.

Bureau of Labor Statistics (BLS). 2021d. “Various Price Index Series” [Excel file]. Federal Reserve Bank of St. Louis. Accessed January 1, 2021.

Cheung, Yin-Wong, and Kon S. Lai. 1995. “Lag Order and Critical Values of a Modified Dickey-Fuller Test.” Oxford Bulletin of Economics and Statistics 57: 411–419.

Congressional Research Service (CRS). 2003. The American Steel Industry: A Changing Profile. Publication no. RL31748, November 2003.

Congressional Research Service (CRS). 2020. Section 232 Investigations: Overview and Issues for Congress. Publication no. R45249, August 2020.

Crotty, James. 2002. “Why There Is Chronic Excess Capacity.” Challenge 45, no. 6: 21–44.

Deloitte. 2019. Semiconductors—The Next Wave. April 2019.

European Commission (EC). 2021. “Actions Against Exports from the EU.” Accessed February 8, 2021.

European Parliament, Committee on the Environment, Public Health and Food Safety (ENVI). 2020. Greenhouse Gas Emissions from Shipping: Waiting for Concrete Progress at IMO Level. September 2020.

Federal Reserve Bank of St. Louis (FRED). 2021. “Effective Federal Funds Rate” [Excel file], Board of Governors of the Federal Reserve System. Accessed January 4, 2021.

Flaaen, Aaron, and Justin Pierce. 2019. “Disentangling the Effects of the 2018–2019 Tariffs on a Globally Connected U.S. Manufacturing Sector.” Federal Reserve Finance and Economics Discussion Series 2019–086. December 2019.

Granger, Clive. 1969. “Investigating Causal Relations by Econometric Models and Cross-Spectral Methods.” Econometrica 37, no. 3 (August): 424–438.

Hasanbeigi, Ali, and Cecilia Springer. 2019. How Clean Is the U.S. Steel Industry?: An International Benchmarking of Energy and CO2 Intensities. Global Efficiency Intelligence, November 2019.

Hersh, Adam. 2014. Assessing China’s Economic Reform Agenda. Center for American Progress, May 2014.

Industry Week (IW) Staff. 2018. “Trade Letter: Tariffs Have Caused ‘Significant Harm to Manufacturers.’” November 20, 2018.

International Monetary Fund (IMF). 2020. World Economic Outlook, October 2020: A Long and Difficult Ascent. October 2020.

Johansen, Soren. 1995. Likelihood-Based Inference in Cointegrated Vector Autoregressive Models. Oxford: Oxford University Press.

Kelley Blue Book. 2020. “Average New-Vehicle Prices up 1.3% Year-Over-Year in November 2020, Down 1.2% from Last Month, According to Kelley Blue Book.” PRNewswire. December 1, 2020.

Lütkepohl, Helmut. 2005. New Introduction to Multiple Time Series Analysis. New York: Springer.

Neilsen, Brent. 2001. “Order Determination in General Vector Autoregressions.” Economics Papers 2001–W10, Economics Group, Nuffield College, University of Oxford.

Organisation for Economic Co-operation and Development (OECD). 2020a. “Steelmaking Capacity” [Excel file], OECD Steelmaking Capacity Portal. Accessed December 31, 2020.

Organisation for Economic Co-operation and Development (OECD). 2020b. Latest Developments in Steelmaking Capacity. June 2020.

Organisation for Economic Co-operation and Development (OECD). 2020c. Short Range Outlook 2020–2021. September 2020.

Rimini, Michele, Anthony de Carvalho, Fabien Mercier, Valentina Burrai, Benjamin Liebman, and Timothy de Stefano. 2020. “Barriers to Exit in the Steel Sector.” OECD Science, Technology and Industry Policy Papers, no. 93: 1–62. October 2020.

Rosnick, David, and Mark Weisbrot. 2007. Political Forecasting?: The IMF’s Flawed Growth Projections for Argentina and Venezuela. Center for Economic and Policy Research. April 2007.

Schieder, Jessica, and Zane Mokhiber. 2018. “By the Numbers: Income and Poverty, 2017.” Working Economics Blog (Economic Policy Institute), September 12, 2018.

Sims, Christopher. 1980. “Macroeconomics and Reality.” Econometrica 48: 1–48. January 1980.

Steel Benchmarker. 2020. “Price History: Tables and Charts” [PDF]. Accessed December 28, 2020.

Taylor, John. 1993. “Discretion Versus Policy Rules in Practice.” Carnegie-Rochester Series on Public Policy 39: 195–214.

U.S. Census Bureau. 2020a and various years. “ACES Tables” [Excel file], Annual Capital Expenditures Survey Tables: 4a, 4b. Accessed January 3, 2021.

U.S. Census Bureau. 2020b and various years. “U.S. Imports for Consumption of Steel Products (FT900A)” [Zipped Excel file], U.S. International Trade Data. Accessed January 11, 2021.

U.S. Census Bureau. 2021. “Quarterly Financial Report” [Excel file]. Accessed January 11, 2021.

U.S. Department of Commerce. 2021. Section 232 Steel and Aluminum Published Exclusion Requests. Accessed February 5, 2021.

U.S. International Trade Commission (USITC). 2021. “Antidumping and Countervailing Duty Investigations Datasets: AD/CVD Orders” [Excel file]. Accessed January 3, 2021.

U.S. Trade Representative’s Office (USTR). 2018. Report on China’s Acts, Policies, and Practices Related to Technology Transfer, Intellectual Property, and Innovation. March 18, 2018.

White House Office of the Press Secretary. 2017. “Fact Sheet: The Obama Administration’s Record on the Trade Enforcement.” January 12, 2017.

World Bank. 2020. “‘Pink Sheet’ Monthly Prices: December 2020” [Excel file]. Accessed December 28, 2020.

World Steel Association (WSA). 2020a. “Total Crude Production of Crude Steel” [Excel file], Steel Data Viewer. Accessed December 31, 2020.

World Steel Association (WSA). 2020b. “Crude Steel Production Monthly” [Excel file], Steel Data Viewer. Accessed December 31, 2020.

World Steel Association (WSA). 2020c. World Steel Short Range Outlook, October 2020. October 15, 2020.

World Steel Association (WSA). 2020d. Steel Statistical Yearbook 2020: Concise Version. December 4, 2020.

World Steel Association (WSA). n.d. Steel in Automotive.

World Trade Organization (WTO). 2020. Anti-dumping Sectoral Distribution of Measures: by Exporter 01/01/1995–30/06/2020. June 30, 2020.

Zellner, Arnold, and Henri Theil. 1962. “Three-Stage Least Squares: Simultaneous Estimate of Simultaneous Equations.” Econometrica 29: 54–78. January 1962. https://doi.org/10.2307/1911287.

What this report finds: Past EPI research has shown that unionized workers in the United States have higher wages and better benefits than nonunionized workers, and that states with “right-to-work” laws (laws weakening unions financially) have lower wages—for both unionized and nonunionized workers—than states without RTW laws. This new analysis shows that Montana, a non-RTW state that is considering an RTW law, has had a higher unionization rate, faster wage growth, and faster job growth than its RTW neighbors.

Why it matters: Despite promises that an RTW law would create jobs, Montana workers would not see employment gains were an RTW law enacted. Instead, an RTW law would curb their ability to collectively push for better wages, benefits, and working conditions.

Summary and key findings

State legislators in Montana are considering enacting a “right-to-work” (RTW) law—a misleadingly named policy that is designed to make it more difficult for workers to come together in a union to negotiate for better wages, benefits, and working conditions.1 Since workplace improvements secured by unions typically spill over into nonunionized workplaces, RTW would have far-reaching harmful consequences for Montana workers—both those who are in unions and those who are not.

Despite the name, right-to-work laws do not confer any sort of right to a job. Rather, they dilute worker bargaining power by making it illegal for a group of unionized workers to negotiate a collective bargaining contract (a contract governing workplace wages, benefits, and working conditions) that includes “fair share fees.” A contract with fair share fees requires all employees who enjoy the contract’s benefits to pay their share of the costs of negotiating and enforcing it. Under an RTW law, employees who don’t join a union but who are still a part of the collective bargaining unit would get all of the benefits of union membership without paying their fair share of the costs. By making it harder for unions to collect these fair share fees, RTW laws aim to shrink union resources. Shrinking union resources impedes the ability of unions to negotiate better wages, benefits, and working conditions—and makes it harder for unions to help workers organize new unions or maintain existing ones.

Some supporters of RTW laws falsely claim that these laws ensure that no one is forced to be a member of a union or pay to advocate for political causes they do not support. But those things are already illegal under federal law. RTW laws are targeted specifically at fair share fees, which can only cover the costs of union representation, not political advocacy.

Proponents of RTW laws also claim that they boost employment by creating a “business-friendly” environment to attract employers from other states, but this is an empty promise. RTW laws have not succeeded in boosting employment in states that have adopted them. Further, arguing that adopting RTW laws will make states more appealing to businesses reveals the true intentions of RTW proponents: undermining unions to lower wages.

This report summarizes existing EPI research on the impact of RTW laws and presents new EPI analysis comparing wage and employment data for Montana with nearby RTW states. Together these findings show that the claims that RTW would boost the state’s economy and attract new businesses are completely without merit.

Key findings:

• In Montana, a typical worker who is represented by a union contract has a higher hourly wage than a typical nonunion worker. Specifically, the median hourly wage of union workers in Montana is $22.85, compared with $16.95 for nonunion workers.
  • The wage advantage for union workers persists even when adjusting for education and other characteristics that partially explain why union workers have higher earnings. While sample sizes are too small to provide such regression-adjusted wages of union versus nonunion Montana workers, previous EPI research shows that nationally, unionized workers are paid 11.2% higher wages, on average, than nonunionized workers in the same industry and occupation with similar education and experience. This wage advantage is referred to as the “union premium” because, by controlling for other factors known to affect earnings, it isolates that part of the wage difference that can be attributed to union status.
• RTW laws are associated with lower wages and benefits for both union and nonunion workers.
  • Both unionized and nonunionized workers in RTW states are paid 3.1% less, on average, than workers with similar characteristics in non-RTW states, according to previous EPI research. If this average pay penalty for being in an RTW state were applied to the pay of a median or typical full-time, full-year Montana worker, it would amount to a $1,143 loss in annual earnings.
  • Since the Great Recession, median wages have grown faster in Montana (12.9% from 2007 to 2019) than in its neighboring RTW states (8.2% from 2007 to 2019).
• By weakening unions, RTW laws fuel growing economic inequality. EPI’s regularly updated national tracker shows that as union membership declines, the share of income going to the top 10% increases.
• RTW laws are associated with reduced unionization. In Montana, 6.5% of private-sector workers are represented by a union, compared with only 4.0% of private-sector workers in neighboring RTW states.
  • Nationally, 8.5% of private-sector workers in non-RTW states are represented by a union, well above the average private-sector unionization rate in RTW states (5.2%).
• RTW laws have not boosted employment in states that have adopted them. In fact, rigorous studies have found that RTW laws have no causal impact on job growth or unemployment, contrary to the claims of RTW proponents.
  • In recent years, private-sector job growth in Montana has outpaced that of its RTW neighbors. From 2007 to 2019, employment grew 10.1% in Montana compared with 8.6% in neighboring RTW states.

Note that all findings in this report are for public- and private-sector workers, unless explicitly stated otherwise. While the Montana RTW law would apply to public- and private-sector unions, in effect it targets private-sector unions because of a Supreme Court case that prohibited fair share fees in union contracts covering government workers.2 “Unionized” workers are workers covered by a collective bargaining contract.

RTW laws were created to weaken unions, not guarantee a job

RTW laws were first enacted in the late 1940s in response to the large expansion of union representation that occurred in many U.S. industries during the late 1930s and early 1940s. The laws originated in the Jim Crow South, with the initial push widely credited to Texas businessman, lobbyist, outspoken racist and anti-Semite Vance Muse and the Christian American Association (CAA). Muse and his allies viewed the prospect of solidarity between Black workers and working-class white workers as a threat to the racist social hierarchy in the South, and the political dominance of wealthy white plantation owners and industrialists. In the 1940s, the CAA succeeded in passing a variety of anti-union laws in the South, including RTW laws, as a means of helping ensure that workers remained divided along racial lines (Kromm 2012; Pierce 2017).

Although the phrase “right to work” implies some expansion of workers’ rights or a guarantee of employment, RTW laws do neither of these things. They were designed to subvert the growth in unions by restricting unions’ resources and making it harder for workers to negotiate and enforce contracts with employers. Under the federal law that protects the right of private-sector workers to engage in collective action (the National Labor Relations Act or NLRA), unions must represent all workers covered by a union contract, even those who choose to not be a part of the union.3 Because this representation can often be costly, unions typically require nonmembers protected by the union contract to pay “fair share fees”—a percentage of regular union dues that covers only the most basic costs of union representation. In states with RTW laws, however, unions are forbidden from requiring workers who have opted out of the union—but still enjoy the union’s benefits—to pay fair share fees. In practice, this results in nonmembers at unionized businesses receiving all the benefits of union representation—higher wages and benefits, legal protection, and representation in the case of a dispute with an employer—without having to contribute anything toward the union’s costs. This “free riding” starves unions of resources, diminishing their ability to negotiate new contracts and organize new groups of workers.

In many other settings, membership fees to cover an organization’s operating costs are common, and not the affront to freedom that RTW proponents claim them to be. For example, condo and homeowner associations require fees to cover the collective costs of upkeep and improvements. Similarly, attorneys cannot appear in court unless they are dues-paying members of the state bar association. Yet in RTW states, unions are forced to provide all the benefits of a union contract for free to workers who choose not to pay their fair share.

Proponents also sometimes claim that RTW laws ensure that no one is forced to be a member of a union or to pay to advocate for political causes they do not support. However, federal law already prohibits workers from being required to join a union as a condition of employment and forbids unions from spending membership dues on political activity. RTW laws limit fair share fees, which specifically cover the costs of union representation, not political advocacy.

RTW laws undermine unions, which are critical for raising wages and combating inequality

By restricting union resources, RTW laws make it harder for workers to exercise their right to organize a union and collectively bargain. EPI analysis of pooled 2017–2019 Current Population Survey Outgoing Rotation Group (CPS-ORG) microdata finds that states with RTW laws have lower unionization rates than non-RTW states (EPI 2021). On average, only 7.8% of workers in RTW states were covered by a union contract, whereas in non-RTW states, 14.9% of workers were covered by union contracts.

This pattern is the same when looking only at the private sector. Just 5.2% of private-sector workers in RTW states are union members or are covered by a union contract, compared with 8.5% in non-RTW states.

Even after controlling for other factors that can be related to unionization—industry, occupation, education, age, gender, race, ethnicity, and foreign-born status—private-sector workers in RTW states are still 5.6 percentage points less likely to be union members or to be covered by a union contract than their peers in non-RTW states4 (Jones and Shierholz 2018).

Figure A shows that Montana and its neighbors mirror this nationwide pattern. While 6.5% of private-sector workers in Montana are represented by a union contract, the average private-sector unionization rate in the neighboring RTW states is just 4.0%.5

Figure A

By lowering the share of the workforce represented by unions, RTW laws aim to weaken workers’ bargaining power with their employers, lowering wages and benefits, and allowing business owners and shareholders to capture more of the income generated by the firm. The employer-employee relationship always has a fundamental power imbalance, which companies often exploit to deny workers their fair share. By coming together in a union to negotiate with their employer, workers can negotiate higher wages, better benefits, safer worker conditions, and a more democratic workplace.6

The benefits of this collective action are clearly shown by examining wage data; wages are consistently higher for union workers than nonunion workers. Figure B shows that is certainly the case in Montana, where workers who are represented by a union contract are paid $5.90 more per hour than their peers without a union contract. The typical unionized worker in Montana is paid $22.85 per hour, while a typical nonunion worker receives only $16.95 per hour.

Figure B

It is important to note that some of these wage differences are affected by other factors related to union membership. For example, union members tend to have higher levels of education and experience, resulting in higher wages that are not the direct result of collective bargaining. However, even after controlling for these factors, workers still see a significant union wage premium, as evident in national, regression-adjusted wage comparisons conducted by EPI last year. Nationally, unionized workers are paid 11.2% more in wages on average, compared to their nonunionized peers (workers in the same industry and occupation with similar education and experience) (McNicholas et al. 2020). This wage advantage is referred to as the “union premium” because, by controlling for other factors known to affect earnings, it isolates that part of the wage difference that can be attributed to union status. Unfortunately, sample sizes are too small to provide such regression-adjusted wages of union versus nonunion Montana workers.

Unions also help close wage gaps for Black and Hispanic workers, who typically receive a larger wage boost from unionization than white workers. Previous EPI research shows that white workers represented by unions are paid 8.7% more than their nonunionized peers who are white (workers in the same industry and occupation with similar education and experience), but Black workers represented by unions are paid 13.7% more than their nonunionized peers who are Black, and Hispanic workers represented by unions are paid 20.1% more than their nonunionized peers who are Hispanic (McNicholas et al. 2020).

Unionization boosts wages and working standards across industries for both union and nonunion workplaces, particularly in places where employers must compete for staff with unionized companies. When unions negotiate good wages and benefits for their members, other employers have to offer better compensation packages to attract and retain the workers they need. But in the same way that unions can raise standards across industries and economies, weakening unions can undermine them just as broadly. For this reason, RTW threatens not just unions, but also the nonunion workers who benefit from these positive “spillover” effects.

Across the country, workers (both unionized and not in unions) in RTW states tend to have lower wages, as shown in Figure C. The median wage in non-RTW states is $20.42, compared with $17.92 in RTW states. Even after controlling for a range of individual demographic and socioeconomic factors known to affect earnings as well as for state macroeconomic indicators, past EPI research shows that hourly wages for all workers in RTW states are 3.1% lower than hourly wages in non-RTW states (Gould and Kimball 2015). If this average pay penalty for being in an RTW state were applied to the median or typical full-time, full-year Montana worker, it would amount to a $1,143 loss in annual earnings. Similarly, workers in RTW states are less likely to have employer-sponsored health insurance and retirement benefits (Shierholz and Gould 2011).

Figure C

Figure C also shows that a typical worker’s wage—represented by the median wage, the wage of someone in the exact middle of the wage distribution—in Montana is similar to the wages of typical workers in neighboring RTW states. The median wage in Montana is $17.72 an hour, just below the median in the RTW states of Idaho, South Dakota, and Wyoming ($17.86). Breaking that three-state average down into individual states (not shown), the median in Montana is just below South Dakota ($17.88) and above the median in Idaho ($17.08). Wyoming’s median ($18.63) is a bit of an outlier, likely driven by the higher concentration of raw materials extraction jobs and public-sector jobs in the state—both sectors that tend to have higher wages.7

Although wage levels are similar across states in the region, Figure D shows that real (inflation-adjusted) wage growth in Montana outpaced its RTW neighbors in the wake of the Great Recession, particularly in from 2016 to 2019, prior to the COVID-19 pandemic. As of 2019, the typical worker in Montana is paid 12.9% more than they were in 2007. Wage growth over that same period averaged just 8.2% for the typical (median) worker in Montana’s RTW neighbors. Wage growth was particularly sluggish in Idaho (3.4%, not shown), with Wyoming also lagging behind (7.5%, not shown). This suggests that protecting workers’ right to organize will be critical in ensuring that Montana workers see substantial wage growth during the recovery from the COVID-19 recession.

Figure D

By strengthening workers’ bargaining power with employers, unions have always been a critical counterbalance against inequality. Yet, as shown in Figure E, with shrinking union membership since the 1950s—in part because of the proliferation of RTW laws and other anti-union policies—a growing share of income has gone to those at the top. By 2019, the top 10% was capturing nearly half of all annual income generated in the United States (McNicholas, Shierholz, and Poydock 2021).

Figure E

Rigorous research has shown that the decline in unionization throughout the country is directly responsible for between one-fifth and one-third of all the growth in inequality since the early 1970s (Western and Rosenfeld 2011).

RTW is not the job creator its proponents claim it to be

Proponents of RTW laws often claim that enacting such laws makes states more attractive to businesses seeking to relocate and consequently leads to stronger job growth. However, this is a false promise: Research shows that there is no causal link between a state’s RTW status and its job growth performance.8 For example, studies on the impact of Oklahoma’s enactment of RTW in 2001 found that the measure significantly reduced private-sector unionization, but had no measurable effect on the rate of employment growth (Eren and Ozbeklik 2015; Lafer and Allegretto 2011). Similarly, researchers at the University of Kentucky examined state economic performance across Southern U.S. states from 1964 to 2004 and found that right-to-work status had no relationship to state economic outcomes (Jepsen, Sanford, and Troske 2008).

Comparisons between non-RTW Montana and its RTW neighbors illustrate this point. As shown in Figure F, Montana added jobs at nearly an identical pace to its RTW neighbors as the nation emerged from the Great Recession. In fact, Montana has added jobs more quickly than its RTW neighbors since 2015. Relative to 2007, private-sector employment in Montana has increased by 10.1%. Over the same period, Montana’s RTW neighbors had 8.6% private-sector employment growth.

Figure F

Montana’s experience in reducing unemployment coming out of the Great Recession was also essentially the same as that of its RTW neighbors. Figure G shows that Montana’s unemployment rate peaked at 7.7% in 2010. The unemployment rate in the neighboring RTW states was lower at 6.9%. Yet by 2016, unemployment rates across the region were within a few tenths of a percentage point. By 2019, Montana’s unemployment rate was down to 3.5%—the exact same rate as in South Dakota and Wyoming (not shown). Idaho’s unemployment rate spiked to higher levels than its neighbors during the Great Recession, but fell to 2.9% by 2019 (not shown), bringing the average of the RTW states to 3.3%. Given the nearly identical unemployment trajectories of Montana and its RTW neighbors, there is little evidence that having an RTW law provided any advantage to Montana’s neighbors in reducing unemployment.

Figure G

Conclusion

The RTW law proposed in Montana’s legislature represents a direct attack not just on the state’s union workers, but on all of Montana workers who benefit from the better wages, benefits, and labor standards that unions advocate for in the workplace and in the state house. If Montana’s legislators care about strengthening their economy, then they should focus on empowering the workers who drive economic growth. RTW laws do just the opposite by undermining workers’ ability to come together collectively to negotiate and advocate for good jobs, safe workplaces, and shared prosperity.

Rather than embracing a policy designed to weaken unions, Montana lawmakers should instead pursue policies that address the state’s real needs. That includes raising pay for low-wage and middle class workers through such policies as raising the minimum wage and expanding access to overtime, addressing the workplace challenges exposed by the COVID-19 pandemic by expanding access to paid sick leave and affordable child care, and making investments in vital public services and state infrastructure through education and health care investments. Such policies would set Montana up for a strong recovery and ensure a more productive and equitable economy for Montana’s future.

Endnotes

1. The proposed right-to-work bill in Montana is HB 251.

2. Janus v. American Federation of State, County, and Municipal Employees, Council 31 et al. https://www.supremecourt.gov/opinions/17pdf/16-1466_2b3j.pdf.

3. Twenty-five states and the District of Columbia have laws that comprehensively protect public-sector workers’ right to form and join unions (McNicholas et al. 2020).

4. While this difference certainly reflects the impact of RTW laws, it may also in part reflect the effect of unobservable factors that are correlated with both RTW and unionization at the state level, such as other anti-union policies or practices.

5. Throughout this report, “RTW neighbors” refers to Idaho, South Dakota, and Wyoming. North Dakota is also a right-to-work state, however, the oil and gas boom that occurred there following the Great Recession led to extraordinary job growth making the state an outlier on nearly all economic measures. Since the North Dakota’s atypical experience in the wake of the Great Recession reflects unique access to natural resources, not broader macroeconomic conditions or policy, including it would distort any comparisons between the region’s RTW and non-RTW states.

6. For a full discussion of the benefits of unionization, and the obstacles that workers face when organizing, see McNicholas et al. 2020.

7. In Wyoming, 7.2% of workers are employed in the “natural resources and mining” industry and 23.8% work in government. Nationally, these sectors make up just 0.5% and 15.0% of the workforce respectively. Both of these sectors have higher median wages—$26.42 for natural resources and mining and $24.74 for public administration—than the overall workforce ($19.33), according to EPI analysis of Current Population Survey Outgoing Rotation Group microdata (EPI 2021).

8. Studies that have purported to show positive employment effects from RTW laws typically fail to control for a host of factors that would affect employment, such as the education level of the workforce to the proximity of transportation hubs to a state’s natural resources or level of manufacturing. See Lafer and Allegretto 2011.

References

Bureau of Labor Statistics, Current Employment Statistics (BLS-CES). Various years. Public data series accessed through the CES National Databases and through series reports. Accessed February 2021.

Bureau of Labor Statistics, Current Population Survey (BLS-CPS). Various years. Public data series accessed through the CPS Databases and through series reports. Accessed February 2021.

Economic Policy Institute (EPI). 2021. Current Population Survey Extracts, Version 1.0.13, https://microdata.epi.org.

Eren, Ozkan, and Serkan Ozbeklik. 2015. “What Do Right-to-Work Laws Do? Evidence from a Synthetic Control Method Analysis” Journal of Policy Analysis and Management 35, no. 1 (July 2015): 173–194.

Gould, Elise, and Will Kimball. 2015. “Right-to-Work” States Still Have Lower Wages, Economic Policy Institute, April 2015.

Jepsen, Christopher, Kenneth Sanford, and Kenneth R. Troske. 2008. Economic Growth in Kentucky: Why Does Kentucky Lag Behind the Rest of the South? Center for Business and Economic Research, University of Kentucky, 2008.

Jones, Janelle, and Heidi Shierholz. 2018. Right-to-work Is Wrong for Missouri: A Breadth of National Evidence Shows Why Missouri Voters Should Reject RTW Law. Economic Policy Institute, July 2018.

Kromm, Chris. 2012. “The Racist Roots of ‘Right to Work’ Laws.” Facing South, December 2012.

Lafer, Gordon, and Sylvia Allegretto. 2011. Does ‘Right-to-Work’ Create Jobs? Answers from Oklahoma, Economic Policy Institute, February 2011.

McNicholas, Celine, Lynn Rhinehart, Margaret Poydock, Heidi Shierholz, and Daniel Perez. 2020. Why Unions Are Good for Workers—Especially in a Crisis Like COVID-19: 12 Policies That Would Boost Worker Rights, Safety, and Wages. Economic Policy Institute, August 2020.

McNicholas, Celine, Heidi Shierholz, and Margaret Poydock. 2021. Union Workers Had More Job Security During the Pandemic, but Unionization Remains Historically Low. Economic Policy Institute, January 2021.

Pierce, Michael. 2017. “The Origins of Right-to-Work: Vance Muse, Anti-Semitism, and the Maintenance of Jim Crow.” Labor Online (the Labor and Working Class History Association), January 2017.

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

Shierholz, Heidi, and Elise Gould. 2011. The Compensation Penalty of “Right-to-Work” Laws. Economic Policy Institute, February 2011.

Western, Bruce, and Jake Rosenfeld. 2011. “Unions, Norms, and the Rise in U.S. Wage Inequality.” American Sociological Review 76, no. 4 (August 2011): 513–537.

• The revenue shortfall is smaller than forecast in the middle of 2020 because the economic fallout of the COVID-19 shock has fallen so heavily on low-wage workers. While this has blunted the revenue loss because low-wage workers pay fewer taxes, it has increased fiscal demands on spending by an abnormally large amount.

• The fiscal demands on the spending side of state and local budgets should not be defined simply by what these governments provided in public investments and safety net spending in the pre-COVID status quo. The need to “build back better” is real and should influence future plans for state and local spending.

• Building back better will require more public investments—particularly in education and safety net programs. Crucially, large educational investments are needed just to keep educational quality constant. These investments have lagged in recent decades.

• Currently, the federal aid to state and local government in the Biden administration’s American Rescue Plan (ARP) provides two utterly crucial functions: It is the major provision that offers potential financing for public investments, and it smooths out the disbursements of aid, allowing the aid to be distributed more gradually and hence buoy growth for a longer stretch of time over the coming years. Just stripping this aid out (or significantly reducing it) would hence be extremely harmful to the overall effectiveness of the ARP.

The low-wage focus of the crisis has been good for state and local revenue, but desperately calls for more spending

State and local revenue shortfalls in 2021 so far look significantly smaller than those forecast in the middle of 2020. Smaller shortfalls are good news, but they don’t justify inaction or mean these state and local governments are facing no forthcoming fiscal stress. The shortfalls forecast in the middle of 2020 would have been, by far, the largest in U.S. history. The shortfalls at the time were large enough to—by themselves—cause a jobs shortfall of over 5 million jobs if they weren’t filled. This jobs shortfall would have been larger than total employment loss of any recession since the Great Depression except for the 2008–2009 Great Recession. This was emphasized this week in a report by Moody’s Analytics. In highlighting the reduced estimates of these revenue shortfalls, Moody’s noted: “This resulted in smaller shortfalls than originally anticipated in most states, but the overall magnitude of the revenue stress is still historic.”

Moody’s—as well as many other sources—have noted two key reasons why the aggregate state and local shortfall has been smaller than forecast before: (1) Federal income support for households has been historically large over the past year, buoying tax collections at the state and local levels; and (2) much of the economic damage of the COVID-19 shock has fallen far harder on low-wage workers than in past recessions. As they note, “It has been unique in the way that budgets are being impacted, with a substantially larger share of overall stress coming via spending pressures than is usual, particularly among social service programs.”

While the focused damage on low-wage workers has blunted state and local revenue losses, it has greatly increased demands on the spending side of state and local budgets relative to previous recessions. To take just one example, Medicaid enrollment growth over the pandemic looks set to be larger than growth during previous recessions. As Medicaid is one of the largest drivers of state budgets, this highlights how needs have grown on the spending side of budgets.

‘Build back better’ will require more public investments

Crucially, the spending needs facing state and local governments should not be defined by the pre-COVID status quo. The Biden administration has often highlighted the need to “build back better.” This should mean a better functioning and more generous social safety net and increased public investments. The Moody’s report is very clear that its current estimates of the fiscal stress facing state and local governments do not include this aim to build back better. For example, the report notes that:

This undoubtedly represents a best-case scenario, as it accounts only for fiscal stresses resulting from the weaker pandemic economy and does not account for extra spending pressures outside of social services. For example, it does not include new needs for implementing IT enhancements or facility upgrades to account for social distancing measures or virtual workplaces. It also, outside of Medicaid, does not account for public health spending increases as a result of the pandemic.

By far the largest public investment financed by state and local governments is education, both K–12 and higher education. It is a well-known finding in economics that the cost of providing public education is likely to rise faster than overall economic growth over time. This finding, sometimes known as “Baumol’s Law,” highlights the fact that some sectors in the economy—like education or health care or other in-person services—are more resistant to automation and other sources of productivity growth. This means that the relative cost of these sectors will rise over time. In turn, even if the inflation-adjusted output of these sectors does not rise, their rising relative costs lead to them accounting for a larger share of total output. Given that many of these productivity-resistant services are those provided by the public sector, this means that the public sector share of economic output should rise over time.

One concrete manifestation of this can be seen in teachers’ compensation. Public school teachers are highly educated and skilled. Pay for workers like them throughout the economy tends to rise steadily over time. If the pay of teachers does not rise in tandem, it will be harder and harder to continue to attract and retain enough skilled teachers. In short, simply to keep educational quality constant, it will require state and local spending on education to rise as a share of the overall economy over time, just to keep teacher pay from falling relative to developments for similarly skilled workers in the private sector. In recent decades, this has clearly not happened. In fact, over the recovery and expansion following the Great Recession, state and local spending on education actually declined in absolute magnitude for long stretches, and certainly did not rise as a share of the economy’s potential output.

This same logic of Baumol’s Law applies to health care, another of the huge expenses on state and local budgets. Additionally, the decades-long reduction in state and local funding for higher education has translated directly into rising tuition costs facing U.S. families. These costs have driven the increase in student debt over this time, which has in turn fueled fierce debate over debt forgiveness. Whatever the merits of debt forgiveness (and we think it is necessary), there is no defensible reason to not at least rein in the growth of future debts driven by ever-rising tuition costs. The single most effective step in reining in tuition cost growth and stopping future student loan debts from being racked up is precisely to increase state and local investments in higher education.

In short, defining shortfalls in the fiscal capacity of state and local governments should not be based only on a pre-COVID status quo that was inadequate in the first place. Instead, they should be defined based on the portfolio of safety net spending and public investments these governments should be providing.

The role of state and local aid in the American Rescue Plan is not easily replaceable

The proposed ARP has generated intense debate in recent weeks. The least convincing critique of it is that it is “too large.” More valid concerns are that it does not contain explicit public investment measures, and that it could in theory be too front-loaded and not leave enough fiscal support 12 or 18 months from its passage when the economy could still use it.

These last two concerns are largely addressed by the large role of federal aid to state and local governments in the ARP. Most public investment is currently directed by state and local governments, so providing them more fiscal capacity greatly expands the possibility for the ARP to provide a big boost to public investment. And the disbursement rate of the state and local aid in the ARP will be longer-lasting than some other parts of the plan. This is useful—we want the $1.9 trillion to be spread a bit over the next couple of years to avoid a whipsaw of aid dropping at once and then turning off.

If the state and local aid is just removed or significantly scaled back, then the package would contain substantially less support for public investments and be more front-loaded than is optimal. In short, scaling back this aid would make it a significantly worse package.

The concentration of economic pain on low-wage workers combined with extraordinary fiscal support over the past year have kept state and local revenue shortfalls from becoming as large as it was once feared. This should not, however, be taken as a reason for complacency. Instead, it should be taken as an opportunity to genuinely use the ARP (or something like it) to “build back better.”

Key takeaways:

• While the economy remains 10 million jobs below pre-pandemic levels and job growth is slowing significantly as the pandemic surges, the remaining suite of pandemic unemployment insurance (UI) programs are set to expire on December 26, even as one of the most important—the extra $600 per week—has already expired and millions of workers have already exhausted benefits or had them significantly slashed.
• The economic shock from COVID-19 has been ongoing long enough that roughly one-third of unemployed workers have been unemployed for 27 weeks or longer. Unemployment insurance benefits should not just be made much more generous, they should also have their durations extended substantially. Once again, this highlights that UI benefit generosity and duration should never be tied to arbitrary dates but should rather be dictated by economic conditions (preferably tied to employment rates).
• If these programs—including the extra $600—are reinstated and extended through 2021, and if the virus is brought under control so that economic growth for 2021 returns to being simply a function of aggregate demand growth, the economy would be boosted by 3.5% and 5.1 million more jobs would be added in 2021.

By the end of this month, several key pandemic relief programs will expire, ending the valuable lifeline provided to workers and their families experiencing job loss, facing eviction/foreclosure, or needing emergency paid sick days and family and medical leave. While spending is needed across multiple avenues to provide relief and recovery, this post examines the importance of reinstating and extending the suite of pandemic unemployment insurance (UI) programs enacted in the Coronavirus Aid, Relief, and Economic Security (CARES) Act: Pandemic Unemployment Assistance (PUA), Pandemic Emergency Unemployment Compensation (PEUC), and Pandemic Unemployment Compensation (PUC) payments.

If the effective safety net functions provided by these programs were maintained through 2021, millions of workers would be better able to avoid economic catastrophe while out of work due to the pandemic. Maintaining these programs’ effectiveness in providing relief and aid for recovery in the face of rising long-term unemployment rates over the next year requires adding additional weeks of UI eligibility duration.

All told, we find that if these programs’ effectiveness is maintained through 2021, and if the virus is brought under control so that economic growth for 2021 returns to being simply a function of aggregate demand growth, the economy would be boosted by 3.5% and 5.1 million more jobs would be added in 2021.

While this past summer’s rebound from the 22.2 million jobs lost in March and April started strong, job growth has since slowed considerably. The first dose of austerity was the expiration of the enhanced UI benefit in July—specifically, the PUC program that provided an extra $600 per week in benefits. Although the economy grew strongly in the third quarter based on momentum from businesses reopening and strong income support from the CARES programs, these PUC cuts will continue to take a serious toll on job creation going forward. As of October, the U.S. economy is still down 10 million jobs from where it was in February. If job growth continues to slow or even reverse course in the winter months as COVID-19 caseloads rise, states reshutter large swaths of businesses, and federal policymakers provide no additional aid to unemployed workers or state and local governments, it will be years before we return to anything resembling the pre-pandemic economy. It would be a tragedy to force U.S. workers to yet again wait a full decade between brief periods of tight labor markets that drive acceptable wage growth.

When the PUA program—which expanded eligibility to millions of workers usually excluded from state UI programs—and PEUC—which extended regular state programs an additional 13 weeks—expire on December 26 of this year, millions will be left out in the cold. In a Century Foundation report, Andrew Stettner and Elizabeth Pancotti found that 12 million workers will be on either PUA or PEUC when the programs expire in less than four weeks. Further, they found that 4.4 million additional workers will have already faced expiration of the benefits on one of those programs before December 26.

Long-term unemployment (unemployment for 27 weeks and over) has been rising quickly over the last several months, hitting 32% of total unemployment in October. It is likely that the rates of long-term unemployment will continue to rise in coming months—as happened in the aftermath of the Great Recession when long-term unemployment exceeded 40% of total unemployment for three years. This means that maintaining the protectiveness of the pandemic UI programs over the next year requires providing additional weeks of eligibility for workers who fall into long-term unemployment. Maintaining the effectiveness of these pandemic-related UI programs over the next year would help workers and their families keep their heads above water while breathing necessary life into the economic recovery.

We estimate the income gains, gross domestic product (GDP) growth, and employment growth that would result from maintaining the PUA, PEUC, and PUA programs through 2021. Like prior estimates of the loss in the $600 boost to UI benefits last July, we use the 2020 relationships between personal income from each UI program and the level of unemployment (or long-term unemployment) to project the income boost provided by continued UI support based on projections of unemployment through 2021. Based on that projected income boost, we then project GDP growth and job growth. Our methodology is described in further detail below.

Table 1 illustrates that these UI extensions would create huge economic gains: an income boost of $441 billion, a gain of 3.5% GDP, and 5.1 million jobs created or saved over the next year. Figure A breaks down the job gains by state.

Having more generous and longer-lasting UI benefits turn off in the midst of a recovery that is still 10 million jobs short of pre-recession levels illustrates why these benefits should be dictated by economic conditions, not by the whims of Congress. Implementing effective automatic stabilizers—both in UI and for programs like federal fiscal aid to state and local governments—should be a pressing priority for the incoming administration.

Table 1

Figure A

Methodology

Our estimates of the income, GDP, and employment boost from maintaining the pandemic UI programs use the relationship between unemployment (or long-term unemployment) and personal income from the PUA and PEUC programs, and then apply this relationship to predicted unemployment in each quarter of 2021. In general terms, we use the relationship between income from the variety of pandemic UI programs and unemployment (overall or long-term) as a proxy for how much money will be put into the economy if the pandemic UI programs are expanded or reinstated in 2021. Then, we use this estimated income boost to estimate growth in GDP and jobs.

Our estimates employ an adjusted unemployment rate, which includes workers who were misclassified as employed and workers who left the labor force for pandemic-related reasons since February 2020. This actually makes our estimates more conservative, as the income boost per percentage point of unemployment in 2020 is lower with the use of the adjusted (higher) unemployment rate in the denominator. The long-term unemployment rate is defined here as the percent of the labor force who is employed 27+ weeks as a share of the labor force. We assume the relationship between the adjusted unemployment rate and long-term unemployment rate and the Bureau of Economic Analysis Personal Income Data for July, August, and September holds for 2021 and apply these ratios to the predicted unemployment and long-term unemployment rate for 2021 to determine the PUA and PEUC personal income boost. Similarly, the personal income boost from PUC is determined by the relationship between personal income from the PUC and the adjusted unemployment rate in April through August (a conservative estimate given that PUC was only just ramping up in April and had officially ended at the end of July, though back payments were still being made in August).

Predicted unemployment rates for 2021 are estimated by applying quarterly changes in the Congressional Budget Office’s 10-year economic projections of the unemployment rate to our adjusted unemployment rate starting in the third quarter of 2020. To construct predictions of long-term unemployment for 2021, we assume that 40% of the unemployed will be facing long-term unemployment in 2021. This calculation is based on the fact that long-term unemployment as a share of the unemployed has been rising steadily over the last several months and recently hit 32% in October, and that long-term unemployment as a share of total unemployment peaked at 45.1% in the aftermath of the Great Recession and exceeded 40% for three years straight (between December 2009 to November 2012). It is certainly possible this is an understatement, and that long-term unemployment is even a bigger problem in 2021 than it was during the worst years of the Great Recession.

We apply a multiplier of 1.5 to the personal income boost for each UI program separately and divide by GDP to get the resulting percentage boost to GDP. We apply this percent change in economic activity to CBO’s prediction for payroll employment in each quarter of 2021. Normally GDP growth runs faster than employment growth early in recoveries. However, because so much of the job loss associated with the COVID-19 shock has been in sectors with low pay and high labor intensity, we think employment growth will respond more rapidly and robustly to a given increment of GDP growth than during normal recessions. The numbers in the chart are the average boost to personal income, GDP, and employment across all quarters of 2021 for the extension or reinstatement of each pandemic unemployment insurance program separately.

The state-level estimates allocate national employment effects to each state with a weight that is the simple average of a state’s current share of initial and continuing UI claims in that state (as of the Department of Labor Unemployment Insurance Weekly Claims dated November 19, 2020) and total nonfarm employment from the Current Employment Statistics averaged from November 2019 to October 2020. In the table, each state’s change in employment is also expressed as a percentage of its October 2020 employment level.

From: Robert E. Scott (Economic Policy Institute)

Submitted via email to transition team

For the U.S. economy to “Build Back Better” after the COVID-19 pandemic, the Biden-Harris administration must emphasize job creation in America’s manufacturing and construction sectors. Rebuilding U.S. manufacturing industries, and upgrading domestic infrastructure, can generate millions of high-wage jobs and reduce income inequality while also addressing racial injustice.

There are three key efforts needed to rebuild manufacturing and the U.S. economy:

1. Realign the U.S. dollar and address overseas currency manipulation.
2. Invest in infrastructure and renewable energy.
3. Rebalance U.S. trade.

PRIORITY ONE: Realign the dollar through a competitive currency policy

The single most effective tool for rebalancing trade is the adoption of a competitive dollar policy. The real value of the U.S. dollar—which has gained nearly 21% since mid-2014 alone—needs to fall by 25% to 30% in order to rebalance trade, according recent research.1 Dollar realignment would stimulate rapid export growth, resulting in surging domestic investment and job creation even as it also reduced import growth.

However, there are two reasons the dollar is currently overvalued.

The first is currency manipulation, which is the result of years of foreign central bank purchases of U.S. dollar assets. This has driven up demand for the dollar and helped to keep it overvalued. This currency manipulation can be addressed through government sanctions and intervention.

More recently, however, the extent of currency manipulation has decreased. Instead, during the last five years, excess private demand for U.S. assets from overseas investors has caused the dollar’s value to soar. This second reason, currency misalignment, can be addressed through market interventions.

Implementing a competitive dollar policy will stimulate rapid export growth, resulting in surging domestic investment and job creation, while limiting import growth. Eliminating America’s $864 billion annual goods trade deficit could create between 3.5 million and 6.6 million jobs over the next four years, including at least 1.4 million good manufacturing jobs.

Immediate steps for the new administration

• The next Treasury report on Foreign Exchange Policies of major trading partners (due April 2021) should label all countries meeting the criteria identified by Christopher Collins and Joseph Gagnon of the Peterson Institute for International Economics (PIIE) as currency manipulators. In addition, countries maintaining very large stocks of foreign exchange reserves—which also have a depressive effect on the values of their respective currencies—should also be labeled as currency manipulators (as explained by Dean Baker, senior economist with the Center for Economic and Policy Research). This includes China and Japan.
• In January, the president should immediately announce the suspension of tax waivers on foreign government holdings of U.S. financial assets in the United States, as proposed regarding China by PIIE’s Joseph Gagnon and Gary Hufbauer in 2011. This will also discourage foreign government holdings of U.S. assets and put downward pressure on the dollar. The U.S. should also deliver notice of canceling tax treaties with selected foreign governments. Taxes should then be withheld on income earned on Treasurys and other government assets, at an initial tax rate of roughly 30%. Significantly, other countries that should be subject to this taxation continue to hold huge foreign exchange reserves, most in dollar assets. These countries include China ($3.1 trillion), Japan ($1.4 trillion), Singapore ($500 billion), and South Korea, ($400 billion).
• The president should use his executive authority under the International Emergency Economic Powers Act (IEEPA) to impose a tax on foreign government owned or controlled holdings of U.S. financial assets as soon as possible. Announcing his intent to do so when canceling tax treaties would put currency manipulators on notice that the United States is serious about stopping such practices. The net of these taxes could be widened as needed, since many currency manipulators have stashed large amounts of their reserves in additional Sovereign Wealth Funds, including China ($1.4 trillion), and Singapore ($900 billion).2
• Taxation of foreign government holdings of U.S. assets would discourage currency manipulation. However, government demand for these assets may not be influenced by taxes on the income the assets earned. Therefore, the Commerce Department and the U.S. trade representative should continue to pursue currency countervailing duty (CVD) cases against individual products and countries. Such cases send a “shot across the bow” to currency manipulators in the absence of more comprehensive measures.
• The Biden administration should initiate offsetting purchases of the foreign assets of those countries found to be engaging in currency manipulation, purchases known as countervailing currency intervention (CCI). Overall, such purchases are the most effective tool available to remedy currency manipulation. The use of Exchange Stabilization Fund (ESF) assets by the Treasury and the Federal Reserve to engage in CCI was proposed by Bergsten and Gagnon in 2017. Substantial increases in ESF assets are required to engage in significant CCI intervention, and will require congressional authorization.
• The Biden administration must also address market-driven currency misalignment. It should do so by empowering the Federal Reserve to establish an exchange rate management policy designed to achieve and maintain balanced trade. This can be done by working with Congress to implement the bipartisan legislation proposed by Senators Tammy Baldwin and Josh Hawley in their “Competitive Dollar for Jobs and Prosperity Act” (S. 2357). The measure would impose a “Market Access Charge” on new foreign investor purchases of U.S. assets. It would also authorize a substantial increase in resources for the Treasury Exchange Stabilization Fund, needed to fight government-backed currency manipulation.
• Along these lines, the administration should also impose an initial emergency access charge, exactly as defined in S. 2357, under the IEEPA.

PRIORITY TWO: Rebuild U.S. infrastructure and begin the clean energy transition

The Biden-Harris plan for investments in infrastructure and climate programs, with a full “Buy America” commitment, can supercharge recovery for the U.S. economy. A $2 trillion, four-year plan of investments in these sectors, along the lines of that announced by President-elect Biden in July, would support between 3.4 million and 6.3 million total jobs. Nearly half (45.7%) of the 3.4 million direct and indirect jobs supported would be in good-paying manufacturing and construction sectors.

Immediate steps for the new administration and Congress

• Immediately revive and expand plans for the Moving Forward Act (MFA), an infrastructure bill that was developed by the House Transportation and Infrastructure Committee, and passed by the House on July 1, 2020.
• Develop an expanded plan for renewal of transportation infrastructure legislation, which can attract bipartisan support. To the extent possible, extensions to the MFA should incorporate Biden-plan goals for expanded U.S. auto production, investments in zero-emission public transit, building and housing investments, and research and development.
• Maximize employment and economic impacts as part of the recovery plan by ensuring that the MFA is entirely debt-financed until the economy has fully recovered (with the exception of current revenues in the transportation trust fund).
• Do not raise additional revenues to fund infrastructure investments until and unless we meet two criteria: estimates of total nonfarm employment exceed 152.5 million (the level reached in February 2020) and interest rates on short-term treasury securities exceed 3.5% on a sustained basis. Where possible, user fees should be relied on to fund public investments in infrastructure, as growing reliance on efficient and clean transportation sources will erode transport fuel tax receipts over time.
• To the extent possible, use the infrastructure bill to fund climate-friendly public investment (e.g., electrical grid upgrades, battery development/capacity, R&D for smart grids, etc.).
• Develop separate legislation to support popular incentives:, such as incentives for hybrid cars, e-cars, wind turbines, and solar power to help utilities meet state renewable guidelines; residential and commercial incentives for weatherization; incentives for appliance efficiency upgrades; consumer solar; renewable conversions for schools and public buildings; and, investments in innovation, agriculture, and conservation.

PRIORITY 3: Pursue trade and industrial policies that will rebalance U.S. trade

More than three decades of globalization have devastated U.S. manufacturing and the American working class. Trade-related job losses are just the tip of the iceberg. Globalization has also reduced median wages by roughly $2,000 per year for roughly 100 million working-class Americans. Joe Biden recognized these problems when he promised the United Steelworkers in May that he would not consider any new trade agreements “until we’ve made major investments here at home, in our workers and our communities.”

Immediate steps for the next administration and Congress

• Establish a freeze on negotiating new trade agreements until the dollar is realigned and the U.S. goods trade deficit has been erased.
• Ensure that trade policy does not privilege corporate interests over workers. The proliferation of Investor State Dispute Settlement (ISDS) clauses inserted into international trade and investment agreements has created a global system of special courts exempt from any judicial appeal or review. These courts allow multinational companies to sue governments for any potential infringement on future profits, as documented by Todd Tucker in his book Judge Knot. These agreements have cast a pall over the ability of governments to regulate in their own legal and national interest. The U.S. must negotiate the elimination of ISDS clauses from most or all trade deals.
• Promote welfare-enhancing multilateral agreements negotiated by the USTR in areas such as labor rights and environmental standards while also rejoining the Paris Climate Accord.
• Pursue multilateral rules to address major international challenges, such as greenhouse gas (GHG) emissions. The U.S. should pursue binding agreements to reduce GHG emissions, especially with China—the world’s largest and most rapidly growing GHG emitter—earlier than called for in the current Paris Climate Accord.
• Ensure that the USTR works with the Commerce Department to aggressively combat overcapacity in steel, glass, paper, solar panels, and a host of other industries distorted by massive state subsidies and other illegal trade and industrial policies.
• Maintain Section 232 steel and aluminum tariffs until tariffs can be replaced by more comprehensive, global limits on unfair trade in these products. One model is to negotiate global tariff agreements to “wall off” products from countries with excess capacity. At present, overcapacity is widespread in many exporting countries, including Japan, Korea, Brazil, Turkey, and China.
• Eliminate tax evasion, corporate inversions, and tax havens that allow multinational enterprises to avoid corporate taxation. The Tax Cuts and Jobs Act of 2017 (TCJA) established new, lower tax rates for foreign investment by multinational corporations; this encouraged further offshoring. Consider adopting sales factor apportionment (SFA) to fully tax profits on all corporate sales in the United States, regardless of where production takes place or where corporations are domiciled. SFA techniques have been used for many years by states to fairly allocate taxation of corporate profits.
• Don’t tax U.S. consumers to protect the intellectual property rights of U.S. multinationals in China. This simply encourages more offshoring of U.S. jobs and factories. As Dean Baker has explained, stronger patent and copyright protections have transferred roughly $1 trillion annually from workers and consumers to corporations and the richest 10 percent.
• Strengthen “Buy America” requirements for all federal, state, and local purchases, as supported by the Alliance for American Manufacturing’s industrial policy proposals. Buy America requirements can greatly enhance the job creation and domestic output of public investments. Historically, Buy America preferences have been loosely enforced.
• Develop new standards and methods to maximize domestic job creation associated with any recovery act, clean energy, or infrastructure expenditures. One simple step would be to require bidders for large government contracts to submit job impact assessments, and to document these outcomes in post-project assessments.
• Establish a strong, whole-of-government program to reshore critical materials production. This includes bringing back to the United States the production of everything from pharmaceuticals to medical equipment to rare earth metals.
• Expand job training and workforce development programs, and consider adopting “flexicurity”-style programs—modeled after those developed in Denmark and other European countries—to support growth and renewal of America’s aging industrial workforce.
• Revamp America’s unemployment insurance (UI) system. Effective UI systems are industrial policies. Jobs not destroyed are much cheaper to restart than those that have been entirely eliminated through short-sighted labor policies. The COVID-19 crisis has demonstrated that America’s UI system is broken and in desperate need of repair. In contrast, many European governments have paid firms to keep workers on the payroll, so that when employers emerge from a downturn, they would be intact.
• Greatly expand R&D and Cooperative Extension Services. Fund the proposal by Simon Johnson and Jonathan Gruber in their book Jump Starting America to identify metropolitan areas that could be hubs of science and technology. Designate them to receive large-scale science and tech spending using the tripartite federal/state and local/private sector investment model to create dozens of national centers of manufacturing research and excellence. In addition, the Hollings Manufacturing Extension Partnerships, which have been targeted for extinction in the Trump administration, should be substantially expanded.
• Substantially increase domestic content requirements and require jobs impact statements by the US Export-Import Bank in its Buy America policies, which have been watered down beyond all recognition. These standards must be tightened and reformed.
• Reevaluate the costs and benefits of foreign direct investment in the United States. So-called “insourcing” (foreign investment in the United States) is dominated by foreign acquisition of U.S. companies, such as Lenovo’s purchase of IBM’s PC division in 2005. Such purchases have eliminated millions of U.S. jobs over the past three decades through layoffs, plant closures, and sell-offs. Furthermore, foreign multinational companies (MNCs) often buy domestic firms simply to distribute their own exported products. As a result, foreign MNCs are responsible for a large and growing share of U.S. trade deficits. Adding insult to injury, state and local governments are often involved in a race to the bottom for such investments, competing to offer tax abatements and infrastructure subsidies to attract foreign investors. The United States should consider banning tax abatements and infrastructure and other subsidies to foreign investors unless entities seeking subsidies can prove that they are effectively creating jobs.

Endnotes

1. Fred Bergsten, “Time for a Plaza-II?” in International Monetary Cooperation: Lessons from the Plaza Accord after Thirty Years, eds. (Washington, D.C.: Peterson Institute for International Economics, 2016), Table 14-5. Bergsten estimated that in order to rebalance U.S. trade, the real value of the U.S. dollar must fall by 26.5% on a trade-weighted basis against the currencies of major surplus countries, including the Euro Area countries, China, and Japan. In their 2020 working paper for the Coalition for a Prosperous America (“Modeling the Effect of the Market Access Charge on Exchange Rates, Interest Rates and the U.S. Economy”), Steven Byers and Jeff Ferry, using a macroeconomic model from the Federal Reserve, estimated that the dollar needs to fall by 27% to rebalance U.S. trade.

2. These funds include China Investment Corporation ($1.046 trillion) and the National Council Social Security Fund ($324 billion), and from Singapore, GIC Private Limited ($453 billion) and Temasek Holdings ($417 billion) according to the Sovereign Wealth Fund Institute (data downloaded from SWFI November 22, 2020).

By “stop the bleeding” we mean using fiscal policy to end the crisis of joblessness and restore the labor market to a reasonable degree of health. Failing to invest enough to sustain a healthy labor market would fatally compromise both the political and the economic ability to structurally reform the economy’s key institutions to create fairer outcomes. We have seen this failure before, with fiscal recovery efforts following the Great Recession of 2008–2009 that were insufficient and too short-lived. As a result of this austerity, it took a full decade for the labor market to return to even its pre–Great Recession health (which was too modest a benchmark to begin).1 It seems clear that a key reason why the Obama administration was unable to get ambitious reform efforts finished after its first year in office was the continued intense labor market distress.

This memo explains why policymakers need to pass roughly $3 trillion in debt-financed fiscal support now, with the first $2 trillion hitting the economy between now and mid-2022. This amount of upfront stimulus, combined with investments that ensure a very slow phaseout of this fiscal support, are needed to ensure a return to a high-pressure, low-unemployment labor market by mid-2022. Specifically, the memo calls on policymakers to take the following actions:

• Finance fiscal support with debt. The point of federal spending now is to boost aggregate demand growth (spending by households, businesses, and governments), so financing this support with taxes would make it less effective.2
• Aim for a high-pressure labor market by picking an ambitious unemployment rate target that constitutes labor market health. Pre-COVID, the unemployment rate hit 3.5% with no evidence that it was too low and likely to lead to a surge of inflation. We suggest targeting an overall unemployment rate of 3%.3 Running the economy at this low level of unemployment for a sustained period creates a high-pressure labor market, one with enough competition for workers to drive faster and more equal wage growth and reduce race-based disparities in wages and unemployment.
• Refuse to accept the self-defeating notion that the COVID-19 shock will leave (or has already left) permanent and unfixable economic scars. Policymakers should run the economy hot and see how much we can heal the damage that the COVID-19 shock inflicted on measures like potential output.4
• Avoid the premature and precipitous withdrawal of fiscal support by ramping up public investments in public goods that are appropriate to debt-finance even during times of full economic health. For the sake of future crises, we should also start building automatic triggers in things like unemployment insurance and aid to state and local governments. But, for the coming years, discretionary fiscal support should be continued. Specifically, to avoid tumbling down a fiscal cliff into an economic contraction in 2023, policymakers should ramp up public investments in such public goods as clean energy, energy efficiency, and early child care and education, and sustain these investments—and their debt-financing—at least through 2024.
• Finally, note that this $3 trillion in needed fiscal support is for hitting economic targets. Money is still clearly needed for virus containment and will be needed for rapid vaccine deployment in coming months. Public health measures are the most important part of the response to the pandemic, so whatever money can usefully help on this front should be added on top of this economic package of relief and recovery.

Below we elaborate on this proposal for $2 trillion in debt-financed fiscal support between now and the middle of 2022, followed by support on the order of $400 billion annually until the end of 2024.

Do not fear debt: Relief and recovery measures should be unambiguously debt-financed

The point of fiscal support is to increase spending by households, businesses, and governments to spur employers to hire. Tax increases put a drag on this spending. While the U.S. needs major progressive reform to taxes, as policymakers plan how to stop the economic bleeding caused by COVID-19, they should be thinking of how to maximize the plan’s stimulative effects, and this means financing with debt. Progressive tax reform, when it occurs, should target economic “bads” like emissions of greenhouse gases and the staggering number of socially unproductive financial transactions that funnel money to Wall Street. This reform should also aim to put a brake on rising inequality and reduce incentives for CEOs and other high earners to rig the rules of the economic game to transfer more money their way.5

But there is no need to think about revenue as policymakers plan on how to stop the economic bleeding caused by COVID-19. All relief measures should be debt-financed. Even the “back-end” relief measures that help to avoid a sharp fiscal cliff as the front-end stimulus winds down should be debt-financed. Because any such back-end measures could in theory come on line with the labor market largely healed, optimally they should be measures that make sense to finance with debt even during times of full employment. Two such measures that meet this requirement are investments to mitigate the emission of greenhouse gases and investments to shore up our woefully underfunded early child care and education system (Gould and Blair 2020).

The logic of using debt even in normal times to finance these investments is clear: The vast majority of their benefits will accrue to future generations. But for the effects of climate change or present-day malinvestment in education, these future generations will almost surely be richer than present generations. Given this, borrowing from future generations to finance these types of investments makes perfect sense. Some hold the misconception that debt incurred always constitutes a burden on future generations. It does not. The heirs of a society and economy also receive any assets financed by debt, which in this case would be a cleaner, more energy-efficient economy and a better-educated workforce.

Go big and stay big: Do not repeat mistakes of past crises

Recovery from the Great Recession of 2008–2009 was agonizingly slow, largely because the admirable initial round of fiscal recovery efforts was too small and ended too quickly. Policymakers did not acknowledge what has become increasingly clear: Long-run trends (such as rising inequality) are making it harder and harder for the U.S. economy to generate sufficient spending by households, businesses, and governments (aggregate demand) to absorb the economy’s resources (including potential workers).6 This “secular stagnation” means that fiscal support cannot be a one-shot “jump start” that can be pulled back quickly as private-sector demand generation roars to life. Instead, fiscal support must be large enough and last long enough that the boost provided ebbs only after the Federal Reserve begins raising interest rates. In short, there must be an attempt to overshoot the target to ensure that private-sector demand generation really is running fast enough when the last bit of fiscal support fades out.7

When setting the target we should avoid fulfilling the bleakest prognosis by setting it too low. Economists and policymakers often express concern that severe recessions can permanently “scar” the economy’s underlying productive capacity (sometimes called potential output), leading to slower growth for a long time—even after the short-run demand shortfall is rectified.8 We should not assume this, and in particular we should not assume it when addressing a shock that is still less than year old. Given the very promising signs of a vaccine on the horizon, there is every reason to think that economic normalization will be possible in 2021, as long as recovery is not held back by a shortfall of demand. This means that policymakers’ goal today absolutely should be to return economic growth to the trajectory it was on before the shock. This goal will inform the assessment of how much fiscal support is needed.

For insight on the level of fiscal support needed when the economy is depressed and interest rates are stuck at near-zero like today, policymakers often look to measures of the “output gap”—the difference between what GDP is projected to be in coming months and years under current conditions (actual GDP) and projections of what it would be were the economy running at maximum sustainable output (potential GDP).

Figure A shows forecasts made by the Congressional Budget Office (CBO) for actual gross domestic product (GDP), a forecast for potential GDP made in January 2020 (pre-COVID-19) and a forecast for potential GDP made in July 2020.

Figure A

CBO has projected a large decline in potential output growth for coming years, as shown by the distance between the top and middle lines in the figure. To the degree that this is driven by estimates of the effect of the COVID-19 economic shock, this degradation of the economy’s potential should not be taken as a given.9 Specifically, focus on the difference between the forecast of actual GDP by the middle of 2022 and the two forecasts of potential GDP. If one believed the July 2020 forecast of potential GDP (with the effect of COVID-19 baked in), then one would think the “output gap”—the difference between potential GDP and actual GDP averaged across every quarter between now (November) and the middle of 2022 (July)—is 4.1%. If one instead compares the forecast of actual GDP with the January 2020 forecast of potential GDP, the gap averaged across quarters in the same period is 7.7%.

For our goal of estimating the size of the stimulus needed to close the real output gap, we land between the two CBO estimates. Specifically, we estimate an output gap based on comparing the forecast of actual GDP with a forecast of potential GDP that is derived from projected unemployment rates. For these unemployment projections, we adjust CBO forecasts based on two factors: Unemployment has already fallen in the second half of 2020 more rapidly than CBO projected, yet reductions in labor force participation and the misclassification of unemployed workers as employed during the pandemic has led to official estimates of unemployment understating the true degree of labor market slack. Based on this adjusted unemployment projection, and applying historical relationships between unemployment and output gaps, we find that between now and the middle of 2022, there is an average output gap of 5.0%.

‘Going big’ means aiming for 3% unemployment

Additionally, CBO estimates of output gaps are based on assumptions that the unemployment rate should be roughly 4.5%. This actually represents some real progress. Before the Great Recession hit, CBO estimated this “natural rate of unemployment” to be 5.0%. But the unemployment rate reached 3.5% right before the COVID-19 shock, and yet there was no sign of economic “overheating”: Both interest rates and inflation were far below historical averages, and inflation continued to register below the Federal Reserve’s 2% target. Given the lack of recent overheating, and given especially that there are real benefits to running the economy “hot” after periods of steep recessions, policymakers should not aim to merely close the output gap, they should aim for a very low unemployment rate in the ensuing recovery. Specifically, it seems like 3.0% unemployment is a reasonable target.10 It is possible that 3.0% is too ambitious, and that economic overheating will set in before we hit that rate, and the Federal Reserve governors might feel compelled to raise interest rates. But it is far more likely that the economy can sit at 3.0% unemployment—or even below—for quite some time before inflation rises fast enough and for long enough to compel the Fed to step in.

Most crucially, however, a scenario involving fiscal support that is strong enough to prompt a monetary policy offset is a sign of success, not failure. The entire point of fiscal support is to push the economy to a point where private-sector sources of demand growth are strong enough to keep the economy pinned at a healthily low unemployment rate even as extraordinary fiscal support is withdrawn. Providing enough fiscal support to overshoot the goal of closing CBO’s estimated output gap and instead hitting an unemployment rate of 3.0% would require an extra 1% of GDP in fiscal support. Applied to our adjusted output gap of 5% identified above, this means fiscal support in the next couple of years should be closer to 6% of GDP, or even a bit above given that social distancing measures might reduce the multiplier effects of stimulus in the very near-term.

Don’t withdraw too quickly: Ending fiscal support should be done very gradually

In the era of chronic demand shortfalls, fiscal policy measures to fight recessions should not be seen as jump-starting a car—providing a one-time burst that can be immediately removed. Private sources of demand generation in recent decades have been quite weak. This means that if large fiscal support is removed quickly, the fiscal contraction can overwhelm private sources of growth and tip the economy back into recession.

Figure B compares CBO’s forecast of actual GDP with two possible scenarios of fiscal support. Under the “short-lived stimulus” scenario, ambitious support is provided between now and the middle of 2022 but is removed quickly thereafter. Under the “long-lived stimulus” scenario, the same level of ambitious initial support is provided, but it is supplemented by a more gradual drawdown between mid-2022 and the end of 2024. Under the short-lived fiscal support scenario, the economy enters a period of substantial contraction in early 2023 as the rapid fiscal contraction overwhelms trend sources of private demand growth. Under the long-lived fiscal support scenario, the back-end support allows the economy to return to trend growth without ever entering an outright contraction.

Figure B

Putting it all together: How much and what should be in relief and recovery measures?

Closing an output gap of roughly 5% (a more realistic estimate than CBO’s for the reasons outlined above) and then overshooting to push the unemployment rate to 3% requires fiscal support equivalent to roughly 6%–7% of GDP. Given the unique nature of the COVID-19 economic shock, the most pressing part of this aid should be to provide relief to those made jobless by the pandemic. A clear model is the enhanced unemployment insurance (UI) provisions that were part of the Coronavirus Aid, Relief, and Economic Security (CARES) Act passed in the spring. These provisions significantly increased the generosity of UI payments and also expanded the universe of people eligible to receive UI. These are incredibly well-targeted measures, and would also serve to provide a powerful economic stimulus when we are able to relax virus-induced social distancing measures (Bivens 2020c). For the purposes of this memo, we assume that UI benefits are increased by $600 per week between now and the middle of 2022, eligibility expansions are maintained over this period, and available weeks of benefits are extended. Given the projected path of unemployment, these extended benefits will cost on average $250 billion at an annualized rate between now and the middle 2021.

Another obvious pressing need for relief and recovery measures is federal aid to state and local governments. State and local governments have revenue sources that have been damaged enormously by the COVID-19 shock. They also provide the front-line public response in many areas related to health and education—and both health and education services have been put under enormous strain by the virus and the need to respond safely. For this memo, we assume that $500 billion at an annualized rate is provided to state and local governments between now and the middle of 2022.

Besides enhanced UI benefits, a range of other targeted safety net measures would provide crucial relief to families and a needed spur to recovery once social distancing measures relax. We assume $100 billion at an annualized rate is provided in enhancements to the Supplemental Nutrition Assistance Program (food stamps), low-income rental assistance programs, and other targeted safety net measures in coming years.

Finally, we provide for public investments in early child care and education and in mitigating greenhouse gas emissions that ramp up immediately and reach peak levels of $400 billion annually by the fourth quarter of 2021. Crucially, these investments do not begin ramping down in 2022. Instead, they remain—and remain debt-financed—at least through the end of 2024. While these public investments will take time to ramp up to full potential, a key virtue is that they provide a very gradual ramping down of fiscal support over time, hence avoiding any sharp contraction. These investments are what kept the “long-lived-stimulus” line in Figure B from ever becoming outright negative.

Relief and recovery measures are just stage one of what is needed from here

A very substantial near-term, debt-financed package of relief and recovery is needed just to allow U.S. families to survive the coming years and to push the labor market back to genuine health by mid-2022, while avoiding a fiscal cliff that would cause contraction soon thereafter. It is important to note that these estimates only include what it would take for a return to health of the economy. Obviously the most important issue in the coming months will concern virus containment and the rapid deployment of vaccines. These public health measures will likely be expensive to do well, and this money should absolutely be spent (and debt-financed). Any cost for these public health measures should go on top of what is estimated for economic relief and recovery in this report.

Labor market normalization should not constitute the limits of economic policymakers’ ambition. The COVID-19 shock did not just cause enormous economic harm, it also exposed huge weaknesses in the economy’s ability to guarantee economic security—both now and when faced with future shocks. These weaknesses were bred by decades of intentional underinvestment and this underinvestment will need addressing in myriad transformative ways.11

But unless we address the immediate crisis of the labor market, both the political and economic support for these longer-run measures to rectify past underinvestment will be crushed. Failure to engineer a rapid recovery from the Great Recession was politically poisonous to the prospects of progressive policy change more generally, as voters lost any confidence in the federal government’s capacity to effectively tackle big issues. On the economic side, there are numerous reasons why long-run investments to change the structure of the U.S. economy will have their greatest effects when the labor market is closer to healthy. Reforming social insurance systems, for example, is likely best done when the existing systems are not inundated by desperate claimants. Building up a stable, professionalized workforce for care work will become easier when the labor market is not so damaged that millions of potential workers will take nearly any job available. Granting workers greatly enhanced rights to join a union will likely lead to much better outcomes if it is not done in a context in which a large pool of unemployed workers gives employers a huge power advantage when bargaining. In short, stopping the economic bleeding first and fully is not a substitute or a distraction from the longer-run effort to create a fairer U.S. economy generally. It is instead a crucial first step.

As here and in earlier reports, the cautionary tale for future efforts is the failure to generate sufficient fiscal support for recovery from the Great Recession. Too many people blame the slow recovery from the Great Recession on largely irrelevant things like its association with a financial crisis. This is excuse-making that lets policymakers off the hook. The fact is that the too-slow recovery was driven by a fiscal policy response that was too weak. We should not make the same mistake following the COVID-19 recession.

Appendix A: Methods

This appendix explains how the effects of fiscal stimulus in Figure B were constructed. For the spend-out of unemployment insurance, we took the effect of enhanced UI benefits from May through July of 2020 on personal income, as measured by the Bureau of Economic Analysis (BEA) and divided by a measure of the unemployment rate in those months that accounted for misclassification within the Current Population Survey (CPS) and the falloff in labor force participation between January and those months. This gives a measure of additional UI payments per percentage point of unemployment. We then took a modified measure of CBO’s forecast for unemployment (CBO 2020b) and applied this measure to that forecast to obtain the extra UI spending that would result.12

For the measure of federal aid to state and local governments, we first use estimates of the likely state and local budget shortfalls caused by the COVID-19 shock between now and the middle of 2022.

For the measure of safety net programs, we use the roughly $100 billion included for these programs in the HEROES Act (Health and Economic Recovery Omnibus Emergency Solutions Act) passed by the House of Representatives in June.

The modified CBO unemployment forecast accounts for the fact that CBO’s last projection was made in July 2020, and unemployment as reported by the Bureau of Labor Statistics (BLS) has recovered more rapidly than CBO forecast it would. Concretely, CBO forecasts unemployment averaging 13.3% in the third quarter of 2020, but it instead averaged 8.8%. However, accounting for the pandemic-induced rise in labor force nonparticipation and the rising misclassification in surveys would boost this 8.8% unemployment rate to roughly 12.2%.13 From there, we allowed the unemployment rate to decline in percent terms at the same pace forecast currently by CBO.

We use this path of estimated unemployment rates that take account of pandemic-induced nonparticipation to back out an implicit output gap. Historically, each percentage point of unemployment over the CBO nonaccelerating inflation rate of unemployment (NAIRU) is associated with an increase in the output gap of 1.4 percentage points. This derived output gap is the one we identify in the paper as averaging over 5% between now (i.e., the third quarter of 2020—the latest date for which we have decent data) and the middle of 2022.

We assume that the full effect of UI benefits, aid to state and local governments, other safety net spending, and public investments hit the economy over six quarters. Although this is slightly more spread out over time than some other estimates of fiscal stimulus, a slower phase-in seems particularly appropriate in the current moment, as social distancing measures are almost sure to reduce any economic effect of relief and recovery spending for a couple of quarters at least (though these efforts will still be improving human welfare, if not GDP in full measure).

For multipliers we assume 1.5 for each provision. This is a small underestimate relative to some past research, but, again, these effects might be attenuated in the short run due to social distancing measures, so this seems like a safe estimate.

For our public investment measures, we assume they take a year of phasing in to reach the full $400 billion annualized spending. The difference between the “short-lived stimulus” and “long-lived stimulus” is simply that this fiscal support stays on line through the end of 2024.

Appendix B: Can automatic stabilizers substitute for the gradual phaseout of stimulus?

Some might wonder if putting triggers on key policies like UI and federal aid to state and local governments might substitute for passing fiscal support today that has a longer phaseout (as called for in this memo). It would absolutely be a good thing if new and improved “automatic stabilizers” that wind down only as key benchmarks of labor market health are reached were passed into law (see for example, Gould 2020). But even on top of much better triggers for making some of these policies automatic in the future, we also need to ensure long-lasting discretionary fiscal support in the coming years.

Automatic stabilizers, by definition, begin providing less fiscal support as the economy improves. In this way, unless the triggers for winding down are extremely powerful and long-lasting, automatic stabilizers are unlikely to provide enough fiscal support on their own to overshoot current estimates of the natural rate of unemployment and help push the economy all the way to 3% unemployment. Given the unique nature of the COVID-19 shock, a uniquely large and long-lasting fiscal support is needed to ensure that a fully healthy labor market is quickly regained.

If we pass powerful and well-calibrated triggers for programs like UI and for federal aid to state and local governments in coming years (and, again, we certainly should do that), the discretionary fiscal stimulus that will be needed in coming recessions will be far smaller. Most of the work in combatting the earliest phases of future recessions will be done by automatic programs.

Endnotes

1. Seen Bivens 2016 for documentation of the historically austere fiscal policy support given to recovery from the Great Recession.

2. See Bivens 2020a on the relationship between aggregate demand and debt-financed stimulus.

3. See Bivens 2020b on the lack of inflationary pressures stemming from a sub-4% unemployment rate in the pre-COVID-19 economy.

4. See Bivens 2019 and Girardi, Meloni, and Stirati 2018 on the potential long-run macroeconomic benefits of rapid and sustained aggregate demand growth.

5. See Piketty, Saez, and Stantcheva 2014 for evidence of the correlation between low top marginal tax rates and aggressive bargaining by high earners to increase their compensation.

6. See Summers 2014 for the first elucidation of this “secular stagnation” thesis. See Bivens 2017 for how it is likely driven in large part by rising inequality.

7. This logic is slightly different from, but largely related to, the arguments laid out in Krugman 2015.

8. See Ball 2014 for the logic and evidence that prolonged periods of demand shortfalls cause reductions in estimated potential output, as well as for some evidence that this “hysteresis” can be potentially reversed with a period of rapid demand growth.

9. Following the Great Recession, CBO began systematically downgrading its estimates of potential GDP growth as well. Even though this downgrading took place over a much longer time frame than the January-to-July period in which the latest potential output revisions took place, it likely was still quite misleading. Much of the decline in in potential output was likely driven by an aggregate demand shortage, and could have been substantially healed with a period of rapid aggregate demand growth (again, see Bivens 2019 and Girardi, Meloni, and Stirati 2018).

10. It is worth noting that an overall unemployment rate of 3% implies a Black unemployment rate of roughly 5%. While this would be a historically low Black unemployment rate, there likely would still remain a gap between white and Black unemployment. Macroeconomic policy alone is unlikely to completely erase the Black/white unemployment ratio, but macroeconomic policy can certainly make the gap much less pronounced.

11. For the overarching policy agenda indicating what is necessary to provide for a more efficient and far fairer U.S. economy, see EPI’s policy agenda, https://www.epi.org/policy/.

12. These unemployment adjustments follow the methodology used by Shierholz (2020).

13. For more on misclassification and other issues with the unemployment rate, see Shierholz 2020.

References

Ball, Lawrence. 2014. “Long-Term Damage from the Great Recession in OECD Countries.” National Bureau of Economic Research (NBER) Working Paper, May 2014.

Bivens, Josh. 2016. Why Is Recovery Taking So Long—and Who’s to Blame? Economic Policy Institute, August 2016.

Bivens, Josh. 2017. Inequality Is Slowing U.S. Economic Growth: Faster Wage Growth for Low- and Middle-Wage Workers is the Solution. Economic Policy Institute, December 2017.

Bivens, Josh. 2019. “Looking for Evidence of Wage-Led Productivity Growth” EPI Macroeconomics Newsletter, December 10, 2019.

Bivens, Josh. 2020a. “Debt and Deficits in the Coronavirus Recovery: Answers to Frequently Asked Questions” (fact sheet). Economic Policy Institute, July 2020.

Bivens, Josh. 2020b. “The Signal the Unemployment Rate Provides Can Change a Lot over Time.” EPI Macroeconomics Newsletter, January 31, 2020.

Bivens, Josh. 2020c. “Cutting Off the $600 Boost to Unemployment Benefits Would Be Both Cruel and Bad Economics.” Working Economics Blog (Economic Policy Institute), June 26, 2020.

Bureau of Economic Analysis. 2020. “Effects of Selected Federal Pandemic Response Programs on Personal Income” [online data table]. Published October 30, 2020.

Bureau of Labor Statistics (BLS). 2020a. “Table A-1. Employment Status of the Civilian Noninstitutional Population.” Employment Situation data tables. Accessed November 2020.

Congressional Budget Office. 2020a. Budget and Economic Outlook, 2020 to 2030. January 2020.

Congressional Budget Office. 2020b. Budget and Economic Outlook, 2020 to 2030. September 2020.

Girardi, Daniele, Walter Paternesi Meloni, and Antonella Stirati. 2018. “Persistent Effects of Autonomous Demand Shocks.” Institute for New Economic Thinking (INET) Working Paper, February 2018.

Gould, Elise. 2020. “The Unemployment Rate Is Not the Right Measure to Make Economic Policy Decisions Around the Coronavirus-Driven Recession.” Working Economics Blog, Economic Policy Institute, March 20, 2020.

Gould, Elise, and Hunter Blair. 2020. Who’s Paying Now? The Explicit and Implicit Costs of the Current Early Care and Education System. Economic Policy Institute, January 2020.

Krugman, Paul. 2015. “Rethinking Japan.” New York Times, October 20, 2015.

Piketty, Thomas, Emmanuel Saez, and Stephanie Stantcheva. 2014. “Optimal Taxation of Top Labor Incomes: A Tale of Three Elasticities.” American Economic Journal: Economic Policy 6, no. 1, 230–271.

Shierholz, Heidi. 2020. “Nearly 11% of the Workforce Is Out of Work with No Reasonable Chance of Getting Called Back to a Prior Job.” Working Economics Blog, Economic Policy Institute, June 29, 2020.

Summers, Lawrence. 2014. “U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound.” Presentation to the National Association of Business Economists (NABE).

The essential elements of this two-pronged strategy for rebuilding the domestic economy are detailed in this report and summarized here:

• Trade and industrial policies that dramatically boost U.S. exports and eliminate the U.S. trade deficit—now roughly $850 billion—within four years. At the heart of these policies are measures to end the overvaluation of the U.S. dollar and rebuild the competitiveness of U.S. manufacturing industries.
• A four-year, $2 trillion program of investments in infrastructure, clean energy, and energy efficiency improvements. This would include investments of $70.2 billion per year in schools and broadband, which would have substantial social benefits. Note also that virtually all (91.6%) of clean energy and energy efficiency investments are for manufactured products.

Following are the key findings of this report:

• Surging exports and major investment in infrastructure, clean energy, and energy efficiency would support between 6.9 and 12.9 million U.S. jobs annually by 2024. The lower-bound estimate includes direct and indirect jobs but not “respending” jobs created as consumers spend more in the economy.
• Of the 6.9 million direct and indirect jobs, at least 471,200 would be construction jobs and 2.5 million would be U.S. manufacturing jobs. Because the jobs supported would be concentrated in high-wage manufacturing (36.4% of jobs supported) and construction industries (6.8% of jobs supported), this strategy would help rebuild U.S. manufacturing and restructure the domestic economy away from low-wage service-sector work.
• Projections of rapid export expansion are not wishful thinking: they are based on the actual export performance in prior periods when the real value of the U.S. dollar was substantially reduced. And there is much room for the dollar to fall: its value has gained 21.4% since July 2014, stagnating U.S. exports and depressing domestic commodity prices, including farm products and incomes.
• Rapidly growing exports in this forecast—especially for U.S. durable goods—along with substantial demand for manufactured products arising from infrastructure and clean energy and energy efficiency investments would support rapid growth in output and employment in a wide range of industries. Rapidly rising demands for fabricated metal products, industrial machinery, computer and electrical products, and transportation equipment (including both motor vehicles and parts, and aerospace products), would generate substantial increases in demand for primary metals (ferrous and nonferrous) and other industrial materials. Production of U.S. energy-based products (crude oil, refined petroleum, and chemicals) also would increase rapidly.
• Within manufacturing, jobs supported would be in both durable and nondurable goods categories. Under the 6.9 million jobs scenario, rapidly growing sectors would include nondurable goods (367,600 jobs), and durable goods (2.1 million jobs). Within durable goods industries, the most jobs will be supported in nonelectrical machinery (436,700 jobs), fabricated metal products (383,700 jobs), transportation equipment (343,800 jobs), electrical equipment (302,700 jobs), and primary metals (248,000 jobs). Within primary metals, 69,900 jobs would be supported in the steel industry. Within transportation equipment will be substantial growth in motor vehicles and parts (188,800 jobs) and aerospace products (127,600 jobs).
• Many sectors outside of manufacturing would experience substantial job growth: transportation (603,400 jobs); agriculture, forestry, and fisheries (588,600 jobs); administrative and support services (454,900 jobs); professional, scientific, and support services (375,300 jobs); wholesale trade (337,100 jobs); and mining (201,400 jobs).
• Rapidly growing exports supported by trade and industrial polices combined with major public investments in infrastructure, clean energy, and energy efficiency would support rapid job creation in all 50 states and the District of Columbia. Jobs supported would be concentrated in regions that have been hardest hit by globalization and outsourcing. Six of the top 10 states in terms of jobs supported as a share of state employment are among the top 10 manufacturing states (as a share of total state employment),  including Wisconsin (6.16%, 181,000 jobs), Indiana (5.95%, 185,900 jobs), Iowa (5.91%, 94,500 jobs), Michigan (5.55%, 251,200 jobs), Ohio (5.51%, 302,400 jobs), and Kentucky (5.37%, 104,100 jobs). Other top-10 job gainers are in energy and resource-intensive states, including North Dakota (6.07%, 24,300 jobs), Wyoming (5.69%, 16,700 jobs), Oklahoma (5.62%, 98,200 jobs), and South Dakota (5.61%, 24,600 jobs).
• Our lower-bound estimate of 6.9 million jobs supported is conservative. The Congressional Budget Office projects that it will take more than five years for employment to return to its pre-recession levels (CBO 2020). In this kind of environment, increases in exports and deficit-financed public investments would generate additional rounds of respending and job creation in the domestic economy (Bivens 2014). Thus our upper-bound estimate of 12.9 million jobs, which includes about 6.0 million respending jobs, is plausible. It is important also to note that these jobs supported are jobs, not job years.2

Introduction: Policy proposals and modeling assumptions

This report evaluates a set of trade and manufacturing policy proposals developed by the Alliance for American Manufacturing (Paul et al. 2020). It also estimates the impacts of a package of infrastructure and clean energy proposals that is based on investments made under a detailed plan developed by the Sierra Club and other civil society groups but at a slightly smaller scale, and for fewer years. That plan, which was analyzed by the University of Massachusetts Amherst’s Political Economy Research Institute (PERI) (Pollin and Chakraborty 2020), is a 10-year plan that would invest $683 billion per year in the elements considered here.3 The plan analyzed here is a four-year, $2 trillion plan.

Trade flows and investment allocations for these activities were prepared in order to project output and employment changes over the 2019–2024 period and thus estimate the increased annual output and jobs supported by 2024, as described below.4

Trade (export promotion and currency rebalancing) projections

Trade projections in this study assume that currency realignment and an aggressive program of industrial policies for recovery result in elimination of U.S. trade deficits in 2024. Currency overvaluation makes U.S. exports more expensive (and suppresses prices of domestic commodities, including gains), while also acting like a subsidy to the cost of all imports (Scott 2020). The policies proposed here are based, in part, on proposals to prioritize industrial policy in the post-COVID-19 world (Paul 2020), which emphasize substantial investment in American-made infrastructure, the reshoring of critical supply chains, enhanced enforcement of Buy America laws, and aggressive enforcement of fair trade policies and the pursuit of high-standard trade agreements. The trade projections are based on actual market behavior in earlier periods of dollar realignment.

For exports supported by currency rebalancing and industrial policies, we examined prior periods of substantial dollar devaluation, including 1985 to 1991 (following the Plaza Accord of 1985) and 2002 to 2008 (the previous period of substantial dollar overvaluation).5 Total U.S. exports increased between 80% and 90% following each of those dollar realignments (Scott 2009 and 2017a). It is important to note that the real value of the U.S. dollar has gained 21.4% since July 2014, stagnating U.S. exports and depressing domestic commodity prices, including farm products and incomes (Federal Reserve Board 2020).

For the projections in this report, we first assumed that exports in each of the individual industries that make up the traded goods portion of the U.S. economy—technically, the detailed, four-digit North American Industry Classification System (NAICS) traded goods industries—would grow at the rate experienced in the 2002–2008 period, with two exceptions, noted here.6 We assume that imports would grow at their actual rate in the 2014–2019 period.7 The initial projections would have resulted in a substantial trade surplus.8 To bring projected trade flows into balance, initial projected exports were then reduced in each sector by 15.5%, resulting in overall trade balance in 2024, as shown in the tables in this report.9

Investment and clean energy projections

The allocation of the four-year, $2 trillion package of investments in infrastructure, clean energy, and efficiency improvement programs was based on allocations developed by Pollin and Chakraborty 2020.10 That report assumes levels of public investment that are about 36% higher than is assumed here (here we look at overall spending of $500 billion a year versus overall spending of $683 billion per year in Pollin and Chakraborty 2020). But the allocations assumed here are in roughly the same proportions as in Pollin and Chakraborty 2020.11 Details of these allocations are summarized in Table 1. (Table 3 shows the allocation of infrastructure and clean energy and efficiency spending by industry.) It is important to note that schools and broadband investments represent $70.2 billion (28.1%), or more than one-quarter of proposed infrastructure investments, which would generate substantial social benefits.

Table 1

Overall economic and employment impacts of trade and investment proposals

Table 2 summarizes the overall impacts of all three components of the program proposed in this report. The top panel of the table shows the economic impact in billions of dollars, and the bottom panel shows the employment impact. The first set of rows in the top panel shows changes in trade flows from 2019 to 2024 resulting from the policies to end overvaluation of the U.S. dollar and rebalance trade. It is assumed that the real value of the U.S. dollar is reduced by approximately 25%, as discussed in the methodology appendix toward the end of this report. Total exports expand by 64.6% between 2019 (actual) and 2024 (projected), while imports increase by only 8.3%. As a result, goods trade balance is achieved in 2024, completely eliminating the U.S. goods trade deficit, which reached $854.3 billion in 2019.

Table 2

The second set of rows in the top panel shows the economic impacts of the fourth year of the new $2 trillion in infrastructure, clean energy, and energy efficiency spending in 2024, reflecting the assumption that public spending on infrastructure and clean energy and energy efficiency investments increases by $500 billion per year in 2021, 2022, 2023, and 2024 ($250 billion per year for each of these purposes). In 2024, the $854.3 billion in increased economic output from rebalancing trade combined with the additional $500 billion yearly spending on infrastructure and clean energy/energy efficiency yields an additional $1.354 trillion in total spending on domestic goods and services (some of which will include imported components). This represents an increase of approximately 6.8% of GDP. It is worth repeating that this increase in spending will only require $500 billion per year in new federal spending; the rest results from increased foreign purchases of U.S. products. The final element of increased demand shown in the top panel of Table 2 is $812.6 billion in induced respending: roughly, how much additional spending happens as the $1.354 trillion in spending makes its way to workers and consumers’ pockets and is respent on consumer goods and services. This figure assumes that there will be a macroeconomic multiplier of 1.6, i.e., a 60% boost to spending in the form of respending. Bivens (2014) reviews the economic literature on multipliers, and notes that infrastructure spending is found to have very high levels of economic multipliers.12 A multiplier of 1.6 is used for that study. Spending on clean energy products, and output from additional U.S. exports, also are likely to have very high multipliers, for similar reasons. Note that multipliers depend in part on the level of excess capacity (economic distress) in the economy. Thus we do not include the multiplier (induced or respending) effects in our main results (jobs supported by industry and by state), but we do include them for informative purposes in our upper-bound estimate of jobs supported and in Table 2 and Table 4.

The employment impacts of these policies are summarized in the bottom panel of Table 2. Note that the employment effects include direct jobs supported or created by a given level of output (an aggregate of all industries) and the aggregate indirect jobs in industries that supply goods to directly affected industries (think auto assembly jobs and the jobs held by those who make auto parts, steel, and rubber, or who provide accounting, finance, staffing, or other services to auto manufacturers).

The U.S. goods trade deficit in 2019 displaced 5.1 million jobs. If trade is balanced, the number of jobs displaced by trade flows is reduced to 1.6 million jobs, for a net gain of 3.5 million direct and indirect jobs supported, as shown in column 3 (see the text box, “Defining jobs: supported vs. created vs. job years”). The reason that there still are jobs displaced under balanced trade is that U.S. imports are more labor intensive, on average, than U.S. exports, as predicted by trade theory, so the U.S. experiences a net loss of jobs.

Infrastructure investments of $250 billion in 2024 would support 2.1 million direct and indirect jobs, and clean energy and energy efficiency investments would support an additional 1.3 million direct and indirect jobs. Overall, the combination of export promotion (balanced goods trade), and expanded public investments will support a total of 6.9 million direct and indirect jobs. In addition, to the extent that multiplier effects are generated by these activities as workers spend their incomes in the economy, up to 6.0 million additional jobs could be supported by these activities. (As noted earlier, multiplier effects are stronger when the economy is struggling than when it is at full employment.)

The fourth and last data column in panel two of the table shows the results of jobs supported or created per category if we break down the additional 6.0 million induced respending jobs by each of the three program areas: export promotion, infrastructure investment, and clean energy/energy efficiency investment. If induced (multiplier) effects are included, trade rebalancing could support an additional 3.1 million jobs, meaning trade rebalancing has the potential to support between 3.5 million jobs (column three) and 6.6 million total jobs (column four). If the overall adjustment in the trade balance is less, then total jobs supported would be smaller. For example, if the trade deficit falls by half, then net export growth will support between 1.8 and 3.3 million additional net jobs.

Similarly, a $250 billion annual increase in infrastructure spending could support an additional 1.8 million respending jobs, meaning the infrastructure spending has the potential to support between 2.1 million jobs (column three) and 3.9 million jobs (column four) when direct, indirect, and induced (respending) jobs are included. Finally, spending on clean energy and energy efficiency could support between 1.3 million and 2.5 million net new jobs. The overall results —roughly 6.3 million direct, indirect, and respending jobs supported—are comparable with Pollin and Chakraborty 2020, when multiplier effects are included.13

Overall, the programs summarized in Table 2 will support a grand total of between 6.9 million and 12.9 million new jobs (depending on the overall level of macroeconomic multipliers in 2024 and thus respending jobs) if the U.S. trade deficit is eliminated in that year.

Economic impacts by industry

Overall economic impacts of the three trade and investment proposals by industry are summarized in Tables 3 and 4. Table 3 reports changes in imports, exports, and the trade balance from implementing export promotion policies that eliminate the trade deficit by 2024, and Table 4 reports how the $500 billion in new spending on infrastructure, clean energy, and energy efficiency in 2024 breaks down by industry.

Table 3

The trade model is based on actual trade behavior during the 2002–2008 period, the last time the dollar experienced a sustained declined of about 25%. During this period, total U.S. goods exports increased 87.5%. The forecast assumes that exports at the industry level increased at the rate that prevailed in the 2002–2008 period (with few exceptions, explained in the notes and methodology appendix), and that imports in each sector increase at the rate that prevailed in the most recent 2014–2019 period (8.3%, in total, as shown in Table 2).14 Finally, assumed export growth in each sector is further reduced by 15.5% so as to achieve overall balance in goods trade in 2024. In other words, the model assumes that overall U.S. goods exports increase 64.6% between 2019 and 2024, as shown in the last column of Table 2.

Table 3 also reports each industry’s share of the overall import growth, export growth, and trade balance change between 2019 and 2024. In terms of net changes in the trade balance, 70.9% of the improvement in the goods trade balance (i.e, the decrease in the goods trade deficit) takes place in the manufacturing sector, 7.2% is in agricultural products, and 19.8% is in mining (oil and gas is a big contributor, alone responsible for 12.6% of the increase in goods trade). Within manufacturing, petroleum and coal products, and chemicals—both essentially “refined energy products”—are together responsible for 21.5% of the total improvement in the trade balance. Finally, 44.0% of the improvement in the trade balance occurs in durable goods industries, which support many good, high-wage jobs, as discussed in the next section.

Table 4 reports the industry breakdown of the $500 billion in spending for infrastructure, clean energy, and energy efficiency in 2024, as noted above, as well as the economic output generated by the $812.6 billion in respending induced by the $1.354 trillion in spending from the trade rebalancing and infrastructure and clean energy/energy efficiency investments. Spending allocations for infrastructure, clean energy, and energy efficiency are scaled to proposals outlined in Pollin and Chakraborty 2020. Overall, 32.5% of planned spending for infrastructure is for construction services, as shown in the addendum at the bottom of Table 4. Less than one quarter (22.8%) of infrastructure spending is for manufactured products. On the other hand, virtually all (91.6%) of clean energy and energy efficiency investments are for manufactured products.

Table 4

The respending allocations assigned to each industry in the last data column are based on personal consumption expenditure data from the Bureau of Labor Statistics input–output tables (BLS-EP 2020b).15 Respending is heavily weighted toward service industry purchases, and manufactured products account for just 12.4% of respending. These differences between the industry composition of investment spending and the industry composition of respending have important implications for the patterns of job creation in the model results, as discussed in the next section.

Job impacts by industry

Table 5 provides the industry breakdown of direct and indirect jobs supported by export promotion (rebalancing trade), infrastructure investments, and clean energy/energy efficiency investments.16 The last two data columns in the table report the total direct and indirect jobs from all three categories combined and the total jobs supported in each industry as a share of the overall total jobs supported (it excludes jobs from respending).17

Table 5

Overall, 6,895,200 jobs would be supported between 2019 and 2024 as a result of these three activities. More than one-third (36.4%) of the jobs supported would be in manufacturing, or 2,508,000 total jobs. In addition, 471,200 jobs (6.8% of the total) would be in construction. An overwhelming share (87.4%) of the 471,200 construction jobs supported are jobs supported by infrastructure investments (411,900).

Manufacturing and construction offer high wages with excellent benefits (Scott 2017b). Nearly half (43.2%) of the direct and indirect jobs supported by the programs outlined in this study would be in these high-wage industries (supporting a combined 2,979,200 jobs). Manufacturing and construction employed a total of 20,491,000 workers, or 13.4% of total nonfarm employment, in February 2020 (BLS 2020a). Thus, these programs, if enacted, would create a threefold increase in the rate at which the U.S. economy is generating good jobs for non-college-educated workers. This would help restructure the labor market toward more high-wage jobs for these workers.18 Competition for these workers also would help pull up wages for all workers with similar characteristics in other industries, by tightening the labor market for non-college-educated workers.

The addendum at the bottom of Table 5 illustrates some of the differences and relative strengths of these three proposals for rebuilding the economy. Nearly two-fifths (39.5%, or 1,386,400 jobs) of the jobs supported by rebalancing trade would be in manufacturing. Roughly one-fifth of jobs supported by infrastructure investment will be in construction. And among the three programs considered here, infrastructure investment supports the smallest share of manufacturing jobs (14.4%, or 298,800 jobs). Clean energy and energy efficiency investments would support 1,315,400 jobs, nearly two-thirds of which (62.6%, or 822,800 jobs) would be in manufacturing. This is an important result for those concerned that clean energy proposals will hurt employment. Clean energy proposals substitute capital, and especially manufactured goods, as inputs instead of energy; these proposals also substitute wages for profits—traditional energy industries such as oil are among the most profitable in the United States.19 To understand the potential benefits of clean energy job creation, consider that the coal mining industry in the United States employed only 50,400 workers, in total, in February 2020 (BLS 2020a). While targeted policies that help workers transition to new industries are clearly a necessary complement to these investment proposals, many of these workers displaced by shifting energy production easily could be absorbed by growing manufacturing industries in the United States if the clean energy proposal were implemented. Overall, 2.5 million manufacturing jobs would be created by these three proposals over the next four years, more than enough to absorb all workers displaced by reduced energy consumption.

A state-by-state breakdown of job creation

Rebalancing trade, rebuilding U.S. infrastructure, and investing in clean energy and energy efficiency would generate significant job growth in all 50 states and in the District of Columbia, as shown in Table 6 and the interactive map in Figure A. Job gains would range from 6.16 % of total employment (or 181,000 jobs supported) in Wisconsin down to 2.85% of employment (or 10,200 jobs supported) in Washington, D.C., as shown in Table 6, which ranks states by jobs supported, as a share of total state employment. In general, job growth would be concentrated in the manufacturing-intensive areas of the country in the upper Midwest and the South which have been hardest hit by globalization and outsourcing, and especially by growing imports from China (Scott and Mokhiber 2020). Certain energy-producing states (i.e., North Dakota, South Dakota, Wyoming, and Oklahoma) are also in the top 10 job-gaining states.

Table 6

Figure A

The model used in this study assumes that construction spending, which is prominent in the infrastructure proposal, will be proportional to current distributions of construction and manufacturing employment by state. Actual results could vary if infrastructure and clean energy spending are allocated based on need, and if spending programs are used to redress existing patterns of racial and gender discrimination. The past is not prologue, in these cases, despite the structure of the model revealed in Table 2. Policy can change the distribution of jobs shown.

Supplemental Table A at the end of this report provides total jobs supported per state ranked by the total number of jobs supported. Jobs supported are in general proportional to total employment, so the states with the largest populations (California, Texas, New York, Florida, and Illinois) make up the top five on this list. Supplemental Table B ranks states alphabetically, and reports the same results shown in Table 6.

Conclusion

Rebalancing trade by expanding exports, and expanding public investments in infrastructure, clean energy, and energy efficiency, are the keys to generating at least 6.9 million good jobs, rebuilding American manufacturing and the U.S. economy.

Acknowledgments

The author thanks Josh Bivens, Scott Boos, Riley Olson, Scott Paul, and Michael Wessel for comments, and Lora Engdahl for editing assistance. We also thank Robert Polin and Shouvik Chakraborter of the Political Economy Research Institute at the University of Massachusetts, Amherst, for additional details about their modeling assumptions. This research was made possible by support from the Alliance for American Manufacturing.

About the authors

Robert E. Scott joined the Economic Policy Institute in 1996 and is currently director of trade and manufacturing policy research. His areas of research include international economics, trade, and manufacturing policies and their impacts on working people in the United States and other countries, the economic impacts of foreign investment, and the macroeconomic effects of trade and capital flows. He has published widely in academic journals and the popular press, including the Journal of Policy Analysis and Management, the International Review of Applied Economics, and the Stanford Law and Policy Review, as well as the The Hill, Los Angeles Times, Morning Consult, Newsday, The New York Times, USA Today, The Baltimore Sun, and other newspapers. He also has provided economic commentary for a range of electronic media, including NPR, CNN, Bloomberg, and the BBC. He has a Ph.D. in economics from the University of California, Berkeley.

Zane Mokhiber joined EPI in 2016. As a data analyst, he supports the research of EPI’s economists on such topics as wages, labor markets, inequality, trade and manufacturing, and economic growth. Prior to joining EPI, Mokhiber worked for the Worker Institute at Cornell University as an undergraduate research fellow.

Daniel Perez is a research assistant at the Economic Policy Institute. He joined EPI in December 2019 and supports the work of EPI economists on trade, inequality, worker power, and more. As a research assistant, he compiles and analyzes economic data for media briefings, research reports, and policy proposals. Prior to joining EPI, Perez served as a research assistant for The University of California, Santa Cruz’s Income Dynamics Lab, studying development and political economy, and worked as programmatic assistant for ROC United, where he worked to improve labor market outcomes for low-wage and tipped workers. Perez also has worked in other industries, including food, wholesale, and education.

Appendix: Methodology

The trade, investment, and employment analyses in this report are based on a detailed, industry-based study of the relationships between changes in trade and investment flows and employment for each of approximately 205 individual industries of the U.S. economy, specially grouped into 44 custom sectors, and using the North American Industry Classification System (NAICS) with data obtained from the U.S. Census Bureau (2019) and the U.S. International Trade Commission (USITC 2020).

This model was developed to analyze the employment impacts of trade flows on the domestic economy by Scott and Mokhiber (2020). It is adapted and extended here to examine the impacts of other types of spending, including infrastructure, clean energy, and induced respending (personal consumption expenditures or PCE) and multiplier effects. The underlying input-output and employment requirements models used to study trade effects are perfectly well suited to the study of domestic investment changes as well.

The number of jobs supported or displaced by $1 million of exports, imports, or other spending for each of 205 different U.S. industries is estimated using a labor requirements model derived from an input–output table developed by the BLS-EP (2020a).20 This model includes both the direct effects of changes in output (for example, the number of jobs supported by $1 million worth of auto assembly output) and the indirect effects on industries that supply goods (for example, goods used in the manufacture of cars). So, in the auto industry for example, the indirect impacts include jobs in auto parts, steel, and rubber, as well as service industries such as accounting, finance, computer programming, and staffing and temporary help agencies that provide inputs to the motor vehicle manufacturing companies. This model estimates the labor content of trade or other spending using empirical estimates of labor content and goods flows between U.S. industries in a given base year (an input–output table for the year 2019 was used in this study) that were developed by the U.S. Department of Commerce and the BLS-EP. It is not a statistical survey of actual jobs gained or lost in individual companies, or the opening or closing of particular production facilities (Bronfenbrenner and Luce 2004 is one of the few studies based on news reports of individual plant closings).

Only nominal trade and expenditure data and nominal employment requirements tables are used in this analysis. Inflation and productivity growth were ignored, in the absence of complete price and productivity projections.

The steps followed to estimate the economic and employment impacts of investments in infrastructure, and in clean energy and energy efficiency, are similar to the steps followed to estimate the economic and employment impacts of trade.

Data requirements for trade and for investments

The text below follows the step-by-step process for developing the data for analyzing all three proposals, with Steps 1 through 3 applying only to trade flows.

Step 1. U.S. trade data are obtained from the U.S. International Trade Commission DataWeb (USITC 2020) in four-digit NAICS formats. General imports and total exports are downloaded for each year.

Step 2. Trade projections are developed based on actual market behavior in earlier periods, as described in the text, above.

Step 3. To conform to the BLS Employment Requirements tables (BLS-EP 2020a), trade data must be converted into the BLS industry classifications system. For NAICS-based data, there are 205 BLS industries. The data then are mapped from NAICS industries onto their respective BLS sectors.

Step 4. Data on expenditures for investments in infrastructure, clean energy, and energy efficiency improvements and for respending were collected as described in the text and in tables 1 and 4, above. Expenditure data were translated into four-digit NAICS industries and then mapped onto their respective BLS sectors.

Step 5. Nominal domestic employment requirements tables are downloaded from the BLS-EP (2020a). These matrices are input–output industry-by-industry tables that show the employment requirements for $1 million in outputs in nominal 2019 dollars. So, for industry i the aij entry is the employment indirectly supported in industry i by final sales in industry j and, where i=j, the employment directly supported.

Analysis of trade and investment impacts

Step 1. Job equivalents. For the trade analysis, BLS trade data are compiled into matrices. Let [T₂₀₁₉] be the 205×2 matrix made up of a column of imports and a column of exports for 2019. [T₂₀₂₄] is defined as the 205×2 matrix of 2024 trade estimates. Define [E₂₀₁₉] as the 205×205 matrix consisting of the nominal 2019 domestic employment requirements tables. To estimate the jobs supported or displaced by trade, perform the following matrix operations:

[J₂₀₁₉] = [T₂₀₁₉] × [E₂₀₁₉]

[J₂₀₂₄] = [T₂₀₂₄] × [E₂₀₁₉ ]

[J₂₀₁₉] is a 205×2 matrix of job displacement by imports and jobs supported by exports for each of 205 industries in 2019. Similarly, [J₂₀₂₄] is a 205×2 matrix of jobs displaced or supported by imports and exports (respectively) for each of 205 industries in 2024.

A similar analysis is performed for infrastructure, clean energy, and energy efficiency investments, and for respending (PCE) as described above. The investments are all assumed to result in net increases in jobs supported by domestic spending.

To estimate jobs supported/displaced over certain time periods, we perform the following operations:

[J_nx19-24] = [J₂₀₁₉] − [J₂₀₂₄]

Step 2. State-by-state analysis. For states, pooled (five-year) estimates of employment-by-industry data are obtained from the Census Bureau’s American Community Survey (ACS) data for the 2013–2017 period (U.S. Census Bureau 2019) and are mapped into 44 unique census industries and seven aggregated total and subtotals, for a total of 52 sectors (including scrap, not part of the census analysis) (Data Planet 2019).21

Previous reports examining employment impacts of trade flows (Kimball and Scott 2014; Scott and Mokhiber 2018) relied on single-year estimates, based on ACS 2011 data, of employment by industry, state, and congressional district. This model has been completely reestimated in this version of the report with the newer ACS five-year data referenced above. These data provide substantially better detail, and greatly improved accuracy, in the form of much lower levels of variance for employment estimates at every level of detail in the model. The new estimates also reflect congressional district boundaries for the 115th Congress for most districts in the country. Boundaries changed in only a few districts in Pennsylvania and Colorado between the 115th Congress and the current 116th Congress.22

We look at net jobs supported from 2019 to 2024, so from this point, we use [J_nx19-24]. In order to work with 44 sectors, we group the 205 BLS industries into a new matrix, defined as [Jnew_19-24], a 44×2 matrix of job support numbers.

Jobs supported by infrastructure and clean energy/energy efficiency investments are added to net jobs supported by trade for the state analysis and combined into the separate vectors shown in Table 6 and Supplemental Tables A and B.

We define [St_2013-2017] as the 44×51 matrix of state employment shares (with the addition of the District of Columbia) of employment in each industry calculated from the ACS five-year employment estimates. We calculate:

[Stj_nx19-24] = [St_2013-2017]T [Jnew_19-24]

where [Stj_nx19-24] is the 44×51 matrix of job displacement/support by state and by industry. To get state total jobs supported, we add up the subsectors in each state.

Jobs supported by infrastructure and clean energy investments are added to net jobs supported by trade for the state analysis, shown separately in Table 6 and Supplemental Tables A and B, and then combined into one final vector for the calculation of total jobs gained as a share of total state employment.

Supplemental Table A

Supplemental Table B

Endnotes

1. The plans examined in this report have long been needed, but would be especially effective at the present time, due to the depressed state of the U.S. labor market (BLS 2020b).

2. See text box, “Defining jobs: Supported vs. created vs. job years,” and discussion there of jobs supported versus job years.

3. The PERI group has published a number of detailed studies of the impacts of clean energy programs at the state, national, and global levels, including Green Growth (Pollin, Garrett-Peltier, Heintz, and Hendriks 2014), and Climate Crisis and the Global Green New Deal (Chomsky and Pollin 2020).

4. The year 2019 is chosen as the base period for this study because that is the last year for which we have complete trade data.

5. See Scott 2009, especially Figure A, for further review of the history of the Plaza Accord and currency realignment in the 2002–2008 period. See Bergsten and Gagnon 2012 for an analysis of the impacts of currency manipulation on the U.S. economy and global trade flows. There are several tools available to combat currency manipulation and offset dollar misalignment (Scott 2017a). One of the most effective and direct methods is to tax foreign investment. Recently, Sens. Tammy Baldwin (D-Wis.) and Josh Hawley (R-Mo.) introduced bipartisan legislation to address the twin problems of an overvalued dollar and growing trade imbalances. Their bill would empower the Federal Reserve to tax new foreign purchases of U.S. stocks, bonds, and other assets—which could return the dollar to a competitive, trade-balancing level (Hansen 2017; Scott 2019).

6. See Methodology Appendix for discussion of NAICs industries and trade data sources. Actual exports of energy products increased extremely rapidly between 2002 and 2008, from a very tiny base, including crude oil (which increased 395%) and refined petroleum products (which increased 632%). By 2019, exports of these products had increased very substantially, to $95.7 billion and $93.8 billion, respectively. Use of historical growth rates for these sectors would have overwhelmed the forecast. Therefore, the initial forecast is that exports of each of these products would double between 2019 and 2024, and then adjust downward by 15.5% in 2024, as described in the text.

7. Total U.S. goods imports increased only 6.0% between 2014 and 2019. Currency realignment will increase the prices of imports, limiting additional consumption of imported products to at most recent trend growth in imports. Note that imports increased rapidly in the 2002–2008 period due to currency manipulation by China and other Asian countries, and extensive unfair trade policies, which limited the decline of the U.S. trade deficit in that period. We assume in this forecast that the dollar falls against all major surplus currencies here, including the Chinese yuan, Japanese yen, Korean won, and the euro, and that fair-trade enforcement otherwise prevents and unwinds unfair import trade. (Authors’ analysis of USITC 2020).

8. The initial projection resulted in a 94.8% increase in total exports between 2019 and 2024, using the weighted average of actual 2002–2008 growth rates, and an 8.3% increase in imports, resulting in an initial projected surplus of $496.7 billion.

9. It should be noted that the 2019–2024 period is one year shorter than the 2002–2008 period mentioned above, so it is reasonable to assume that future export growth will be less than in the reference period.

10. The Sierra Club (2020) plan is detailed but is at a higher spending level and for a longer period of time than the plan considered here, which is based on a four-year, $2 trillion climate and infrastructure investment proposal.

11. The authors thanks Robert Pollin and Shouvik Chakraborty for additional details about model assumptions (Chakraborty 2020). The final version of that report also evaluates a proposed investment of $186 billion per year in agricultural and land restoration investments that are not included here. The program considered here includes a much smaller component of agricultural programs, for energy conservation, as noted below. Individual modeling elements were converted from the IMPLAN 546 modeling format to the Bureau of Labor Statistics formal modeling of 205 individual industries of the U.S. economy; this conversion was implemented using IMPLAN to NAICS crosswalks.

12. The actual level of respending achieved could be higher or lower than shown in Table 2 and elsewhere in this report. The actual size of the multiplier will depend on the level economic activity when the spending takes place. See the text box, “Defining jobs: Supported vs. created vs. job years,” for discussion of the role of labor market tightness or slack on overall job creation. See also Bivens (2014) for a review of the literature on economic multipliers.

13. Pollin and Chakraborty (2020, Tables 1b and 2b) estimate that $683.1 billion in infrastructure and clean energy spending would support a total of 9.3 million new jobs, including direct, indirect, and induced spending. Our report estimates that $500 billion in infrastructure and clean energy and energy efficiency could support a total of 6.34 million jobs (including respending, Table 2, above). Adjusting for the 36.6% higher spending levels in Pollin and Chakraborty relative to this report’s $500 billion spending package, overall projections shown in Table 2 are about 7.6% lower, in terms of jobs per billion dollars of spending, which is likely explained by small differences in multipliers (induced spending) in the two models. In addition, the BLS model used here is based on 2019 input–output tables, and the IMPLAN model used by Pollin and Chakraborty is based on 2018 input–output data. See the methodology appendix for further details.

14. The model is based on trade flows at the NAICS 4-digit level, which are aggregated into the 205-industry BLS model used for this study, as described in the appendix. These data are further aggregated into 52 sectors for presentation in Tables 2–4 (some of which with no data are omitted from Tables 2 and 3).

15. The personal consumer expenditures vector is one of the components of the Aggregate Final Demand data set that is included with the BLS input–output matrix files, as a component of “Nominal dollar input–output data for 1997–2019” (BLS-EP 2020b).

16. The table provides detailed information on jobs supported by industry and within industries. An additional fact not provided in the table but rather from unpublished analysis of the data is that within primary metals, 69,900 new jobs would be supported in the steel industry (NAICS 3311 and 3312).

17. Four industries show net jobs displaced by trade, and in three of those industries that translates into jobs displaced in the trade plus investments total.

18. Manufacturing and construction employ a substantially higher share of non-college-educated workers than other sectors of the economy. For example, in 2009–2011, 47.7% of manufacturing workers had a high school diploma or less education, compared with 37.6% of workers in all industries (Scott 2013, Table 1).

19. Profits are much lower in manufacturing industries, which produce 91.6% of products in this study. Hence, substitution of clean energy equipment for energy products will increase the labor share of energy expenditures.

20. The model includes 205 NAICS industries. The trade data include only goods trade. Goods trade data are available for 85 commodity-based industries, plus information (publishing and software, NAICS industry 51), waste and scrap, used or secondhand merchandise, and goods traded under special classification provisions (e.g., goods imported from and returned to Canada; small, unclassified shipments). Trade in scrap, used, and secondhand goods has no impact on employment in the BLS model. Some special classification provision goods are assigned to miscellaneous manufacturing. Most trade in the special classifications provisions is small package trade that enters duty free, and involves products that are not classified.

21. The U.S. Census Bureau uses its own table of definitions of industries. These are similar to NAICS-based industry definitions, but at a somewhat higher level of aggregation. For this study, we develop a crosswalk from NAICS to Census industries, and we use population estimates from the ACS for each cell in this matrix. The ACS data we obtain from the Census Bureau for this project includes 44 unique sectors, plus subtotals for manufacturing, and for total employment. Trade and job loss coefficients are estimated using data only for the 44 unique sectors, across states and congressional districts.

22. According to the U.S. Census Bureau, only Colorado and Pennsylvania had congressional district boundary changes for the 116th Congress.

References

American Society of Civil Engineers (ASCE). 2017. 2017 Infrastructure Report Card.

Bivens, Josh. 2012. Green ‘Sequester’ is Already Costing U.S. jobs: Job Losses from Ongoing Clean-tech Cuts Will Rival Those from Defense Cuts. Economic Policy Institute, December 2012.

Bivens, Josh. 2014. The Short- and Long-Term Impact of Infrastructure Investments on Employment and Economic Activity in the U.S. Economy. Economic Policy Institute, July 2014.

Bergsten, C. Fred, and Joseph E. Gagnon. 2012. Currency Manipulation, the U.S. Economy, and the Global Economic Order. (Policy Brief 12-25), Peterson Institute for International Economics, December 2012.

Bronfenbrenner, Kate, and Stephanie Luce. 2004. The Changing Nature of Corporate Global Restructuring: The Impact of Production Shifts on Jobs in the U.S., China, and Around the Globe. Commissioned research paper for the U.S. Trade Deficit Review Commission.

Bureau of Labor Statistics (BLS). 2020a. “Employment, Hours, and Earnings from the Current Employment Statistics (National)” (Excel spreadsheets). Accessed September 21, 2020.

Bureau of Labor Statistics (BLS). 2020b. “Table A-11. Unemployed Persons by Reason of Unemployment.” Accessed September 29, 2020.

Bureau of Labor Statistics, Employment Projections program (BLS-EP). 2020a. “Nominal Domestic Employment Requirements Table for 2019” [Excel sheet, converted to Stata data file]. In Historical Employment Requirements Tables, 1997–2019 [data series]. Last modified June 10, 2020.

Bureau of Labor Statistics, Employment Projections program (BLS-EP). 2020b. “Inter-industry relationships (Input Output matrix): Nominal Final Demand Aggregate Data.” [Excel sheet]. Last modified June 10, 2020.

Chakraborty, Shouvik. 2020. Personal communication with Robert Scott, August 31, 2020.

Chomsky, Noam, and Robert Pollin, with C.J. Polychroniou. 2020. Climate Crisis and the Global Green New Deal: The Political Economy of Saving the Planet. London and New York: Verso.

Congressional Budget Office (CBO). 2020. “An Update to the Economic Outlook: 2020 to 2030 (10 year Economic Projections).” (Report), Congressional Budget Office, July 2, 2020.

Data Planet. 2019. “American Community Survey, 5-year Estimates: About the ACS 5-year Estimates” (web portal for exploring ACS data). Last updated December 18, 2019.

Federal Reserve Board. 2020. “Foreign Exchange Rates – H.10: Real Broad Dollar Index – Monthly Index” (data table). Accessed September 14, 2020.

Hansen, John R. 2017. “Why the Market Access Charge is Necessary to Fix Trade Imbalances.” Coalition for a Prosperous America. September 2017.

Kimball, Will, and Robert E. Scott. 2014. China Trade, Outsourcing and Jobs: Growing U.S. Trade Deficit with China Cost 3.2 Million Jobs between 2001 and 2013, with Job Losses in Every State. Economic Policy Institute Briefing Paper no. 385, December 2014.

Lee, Thea. 2020. “HEROES Act Provides Critical Relief and Recovery Measures to U.S. Workers.” (Statement). Economic Policy Institute, May 12, 2020.

Osterholm, Michael T., and Neel Kashkari. 2020. “Here’s How to Crush the Virus Until Vaccines Arrive: To Save Lives, and Save the Economy, We Need Another Lockdown.” New York Times, August 7, 2020.

Paul, Scott. 2020. Our American Manufacturing Plan Will Create 6.9 to 12.9 Million New Jobs by 2024, Alliance for American Manufacturing. October, 2020.

Pollin, Robert, James Heintz, and Heidi Garrett-Peltier. 2009. How Infrastructure Investments Support the U.S. Economy. Political Economy Research Institute, University of Massachusetts Amherst, January 2009.

Pollin, Robert, Heidi Garrett-Peltier, James Heintz, and Bracken Hendricks. 2014. Green Growth: A U.S. Program for Controlling Climate Change and Expanding Job Opportunities. Center for American Progress and Political Economy Research Institute, University of Massachusetts Amherst, September 2014.

Pollin, Robert, and Shouvik Chakraborty. 2020. Job Creation Estimates Through Proposed Economic Stimulus Measures.  Political Economy Research Institute, University of Massachusetts Amherst, September 2020.

Scott, Robert E. 2009. “Re-Balancing U.S. Trade and Capital Accounts.” Economic Policy Institute, Working Paper no. 286, December 2009.

Scott, Robert E. 2013. Trading Away the Manufacturing Advantage: China Trade Drives Down U.S. Wages and Benefits and Eliminates Good Jobs for U.S. Workers. Economic Policy Institute, September 2013.

Scott, Robert E. 2017a. Growth in U.S.–China Trade Deficit between 2001 and 2015 Cost 3.4 million Jobs: Here’s How to Rebalance Trade and Rebuild American Manufacturing. Economic Policy Institute, January 2017.

Scott, Robert E. 2017b. “We Still Haven’t Recovered Well-Paying Construction and Manufacturing Jobs.” Economic Snapshot, Economic Policy Institute, August 16, 2017.

Scott, Robert E. 2019. “Trade Wars and the Over-Valued Dollar.” The Hill, August 9, 2019.

Scott, Robert E. 2020. We Can Reshore Manufacturing Jobs, but Trump Hasn’t Done It: Trade Rebalancing, Infrastructure, and Climate Investments Could Create 17 Million Good Jobs and Rebuild the American Economy. Economic Policy Institute, August 2020.

Scott, Robert E., and Zane Mokhiber. 2018. The China Toll Deepens: Growth in the Bilateral Trade Deficit between 2001 and 2017 Cost 3.4 Million U.S. Jobs, with Losses in Every State and Congressional District. Economic Policy Institute, December 2018.

Scott, Robert E., and Zane Mokhiber. 2020. Growing China Trade Deficit Cost 3.7 Million American Jobs Between 2001 and 2018: Jobs Lost in Every U.S. State and Congressional District. Economic Policy Institute, January 2020.

Sierra Club. 2020. Millions of Good Jobs: A Plan for Economic Renewal. May 2020.

U.S. Census Bureau. 2019. “American Community Survey: Special Tabulation over 44 industries, Covering 435 Congressional Districts and the District of Columbia (115th Congress Census Boundaries), Plus State and US Totals Based on ACS 2013 5-year file” [custom tabulation, spreadsheets received November 26, 2019; Rhode Island data received January 14, 2020].

U.S. International Trade Commission (USITC). 2020. USITC Interactive Tariff and Trade DataWeb [database]. Accessed September 2020.