Breaking Down or Charging Up? Local Political Repercussions from Clean-Energy Manufacturing Investments

Introduction

In 2022, the United States passed the largest climate bill in the country’s history, the Inflation Reduction Act (IRA). The IRA’s designers hoped that the investments it spurred would “create a new politics” (Schattscheider 1935) by generating salient social, political, and economic benefits for constituencies who would then mobilize to defend and expand clean energy technologies and the policies that support them (Breetz, Mildenberger, and Stokes 2018; Béland, Campbell, and Weaver 2022; Jacobs and Mettler 2018; Campbell 2002; Mettler 2007; Soss 2000).

For instance, the law’s 45X tax credit incentivized companies to build facilities to manufacture clean-energy technologies like solar panels, wind turbines, batteries, and electric vehicles. These factories would provide construction and manufacturing jobs, attract related suppliers, and generate economic growth associated with providing housing and services for the plants’ workers. Any political momentum associated with these investments could increase the resilience of the energy transition to political uncertainty.

The IRA reflected a benefits-forward strategy to unite important constituencies (Bergquist, Mildenberger, and Stokes 2020), avoid imposing costs and provoking backlash, and bolster constituencies that would support future more aggressive policies (Breetz, Mildenberger, and Stokes 2018; Hacker 2019; Meckling 2019). Advocates hoped that these policy design features would result in policy “embedding” (Patashnik 2023) as opposed to “retrenchment” (Stokes 2020) or “backlash” (Patashnik 2023). The political theory behind the IRA was quickly challenged, when the Republicans regained control of the White House and both chambers of Congress in 2024. The 2025 budget bill gutted the IRA’s grant and loan programs and sharply accelerated the phase-out of many of its tax credits (Plumer 2025).

Our research examines what went wrong for the political theory behind the IRA. We find, first, that the IRA failed to generate an electoral boost for the Democrats, the party of the president whose policies undergirded the investments it facilitated. Next, we use deep qualitative analysis to explain why. We examine whether and how key constituencies perceived impacts, linked them with policy, and mobilized politically. We find that relevant constituents perceived local benefits but did not reliably link project impacts with federal policy, partially due to the nature of political communications concerning the projects. Particularistic interests did not promote strong supportive mobilization, and forces of backlash were more strongly activated than expansionary forces.

The (Non)-Effect of Clean-Energy Manufacturing Investments in the 2024 Elections

We first examine how voters responded to clean-energy manufacturing investments. Specifically, we ask whether voters in counties receiving such investments supported or turned against the Democrats, whose policies facilitated an expansion of the U.S. clean-energy supply chain. The logic is that if people were aware of the projects, supported them, and linked them with Biden-era policies, this support could manifest in support for the President’s party during the election. We examine presidential vote choices between individuals living in counties that received new clean-energy manufacturing investments during Joe Biden’s presidency, compared with people living in otherwise similar counties that did not receive such investments. We observe no evidence of increased support for Democratic Presidential candidate Kamala Harris in communities that received clean-energy manufacturing investments (Figure 1).¹

Examining the Process of Local Policy Embedding or Backlash

 We use deep qualitative case studies to examine why clean-energy investments failed to generate public enthusiasm in the form of an electoral response, and, more generally, to understand the process by which such investments are interpreted locally. We focus on two electric-vehicle battery manufacturing facilities: the Blue Oval Battery Park (Blue Oval) in Calhoun County, MI and the Blue Oval SK Battery Park (BOSK) in Hardin County, KY. Both facilities were sponsored by Ford Motor Company, in the Kentucky case in partnership with the Korean battery manufacturer SK On. The plants were projected to generate 1,700 (MI) and 5,000 (KY) jobs. This amounts to one job for every 78 people living in Calhoun County and one job for every 22 people living in Hardin County. In short, the economic impact of these projects would be enormous.

Perceptions of Project Impacts

Job creation is the most direct benefit associated with clean-energy manufacturing projects. In both cases, our interview respondents nearly unanimously recognized job creation as an important, even transformative, benefit. Leaders from the construction and electrical industries emphasized that the Blue Oval projects were a “monstrous” opportunity for the regions. They highlighted the relative attractiveness of the Blue Oval jobs for construction and electrical workers: high pay, a long duration, apprenticeship opportunities, job security and benefits guaranteed by a Project Labor Agreement, and overtime opportunities. The IRA was designed to incentivize high-quality jobs during construction, through requiring that employers pay prevailing wages and offering a large tax-credit bonus for employing workers participating in apprenticeship programs (IRS 2024).

Leaders from the construction and electrical industries emphasized the importance of apprentice certification for workers’ career development and credited the IRA’s incentives with promoting this feature at the job sites. The construction and electrical workers unions in both places also experienced substantial growth in their membership as a result of the projects, and this growth in union membership is one way in which the projects could reshape politics (Schattscheider 1935; Pierson 1993; Campbell 2002; Mettler and Soss 2004).

Our respondents also viewed operations jobs positively, despite uncertainty about the nature of these jobs in the Michigan case. We conducted our interviews well before the Michigan plant had opened for operation. As a result, interview respondents did not think that potential workers had much if any specific knowledge about job opportunities during the plant’s operations. Nonetheless, against this backdrop of uncertainty, workers and workforce development professionals were optimistic about the quality of jobs that would be available at the plant, and about the fit between the local workforce and jobs available.

Many grounded their optimism in an assumption that the plant would be unionized, based on a contract agreement with the United Auto Workers (UAW) (United Auto Workers and Ford Motor Company 2023). This made the Blue Oval jobs attractive by comparison with jobs available with auto suppliers in the area, many of which are not unionized. Our conversations in Kentucky, where the plant had already opened when we visited, reinforce the optimism expressed in Michigan.

Workers highlighted competitive wages, positive workplace culture, comprehensive health care benefits, and paid time off as comparatively attractive features of their jobs at BOSK. In December 2025 Ford announced plans to pivot operations at both facilities away from electric vehicles and lay off a large number of workers from the Kentucky plant (Garcia and Dederer 2025a). Time will tell how the communities will react to the uncertainty that this change introduced.

Linking Impacts with Federal Policy

Most respondents did not see federal incentives as a necessary condition for the projects’ success. Instead, local leaders and project champions generally highlighted the states’ role in funding land purchases; connecting the sites to infrastructure; and providing incentives to Ford and SK On. Moreover, many expressed confidence that the Blue Oval facilities, and the EV transition more broadly, would proceed despite a potential rollback in federal incentives for the purchase and manufacture of electric vehicles. Some respondents expressed a view that the transition to

EVs is “inevitable.” They believed that the repeal of tax credits may slow the shift to EVs, but it will not stop the overall transition, nor will it cause Ford or SK On to cancel their investments. They grounded their confidence in sunk costs: the companies have already invested enough that they would not pull back on their investments at this late stage.

Some argued that the companies may adapt to market or policy changes by producing batteries for other applications, but no one anticipated layoffs and delays (Garcia and Dederer 2025a) from such a pivot. Others believed that a slowing of the transition to EVs could be beneficial for the community, by allowing time to plan for growth, or for the industry, by ensuring that only the most viable companies survive.

We also do not find that the Blue Ovals have led to generalized support for clean-energy incentives, whether from the state or federal governments. The local leaders who actively worked to attract the projects—who emphasized the pivotal role played by state funding–articulated particularistic support for state funding to be used for “this project.” When asked about a potential (at the time of our visits) rolling back of state or federal incentives for EV manufacturing, they responded, “What we’ve got, the money we need, it’s in place. No taking away incentives is going to change that.” This particularistic focus undermines the idea that the creation of local benefits would increase the durability of policies to promote the energy transition more broadly.

The exception to this general tendency is the leaders we spoke with from the electrical and construction trades. These leaders did see the link to federal policy, quite clearly. One union leader worried that the accelerated expiration of consumer-facing EV incentives would amount to “killing the electrical vehicle industry.” Another attributed to the IRA Ford’s support for apprenticeship programs for electrical workers involved in the construction of the Blue Oval facility in Michigan.

One union leader reported difficulty in persuading his members that the Trump administration would roll back Biden-administration policies that benefit workers. His perspective reflects the increasing disconnect between the political views of union leaders and members (Hertel-Fernandez 2024) and highlights a need to support local leaders in communicating the relevance of policy levers for beneficiary groups.

The Blue Ovals, Political Communication, and Elections

Political communication could facilitate the perception of a link between project impacts and public policy, but the federal legislators representing these communities did not draw strong and consistent links between federal policy and local benefits. On the date that the Blue Oval Battery Park was announced, Michigan Senators Debbie Stabenow (D) and Gary Peters (D) announced the project on Twitter. Michigan District 7 Representative Tim Walberg (R) did not acknowledge the Blue Oval announcement, suggesting that partisan interests overpowered the incentive to highlight an economically beneficial project in his district. Stabenow linked the project to the IRA in one tweet, but this was the only instance in which she or any other legislator linked these projects to federal support.

Instead, in the years following the announcement of the projects, the Republican legislators representing these districts drew a clear and consistent negative connection between federal policy and state or local impacts. Walberg, along with Kentucky Senators Rand Paul and Mitch McConnell and Representative Brett Guthrie, highlighted Chinese dominance of the EV supply chain; raised doubts about consumer demand for EVs, workforce opportunities associated with them, and government policies to promote them; and focused on the negative impacts of federal investment spending and auto regulations on the state or district (Polarization Research Lab 2025).

In the contest to replace Stabenow upon her retirement, Republican Senate Candidate Mike Rogers lambasted the project in Michigan. Rogers’ Democratic opponent, Elissa Slotkin, largely declined to comment on the project, rather than defending it against Rogers’ attacks (Pero 2024). Slotkin prevailed in this election, but we could not claim that support for Blue Oval or clean-energy manufacturing in general contributed meaningfully to her victory. Whereas Figure 1 shows that clean-energy manufacturing projects did not influence voting in the 2024 federal elections, we do find that the projects were relevant in local elections.

In Marshall, the city and township where the Hardin County project is located, the August 2024 local primary election presented an opportunity to mobilize. A slate of three candidates for the Marshall Township board focused their campaign on the incumbents’ support for Blue Oval. They appealed to local concerns about the project’s aesthetic, environmental, and other unwanted impacts. The challengers prevailed in this election, replacing the treasurer and both trustees by wide margins (Marshall Advisor & Chronicle 2024).

We find no evidence that BOSK had any relevance to elections in Hardin County. This is at least partly due to jurisdictional differences between the sites. In Marshall, local officials were involved in planning, permitting, and buying and transferring property, but local officials played a much smaller role in decisions that could spark backlash in the Kentucky case. The state purchased the properties for the project directly, over two decades ago. Glendale, where the site is located, is an unincorporated city, without elected officials or any jurisdictional authority over the site. This means there was no local decision-making for project opponents to mobilize against, either retrospectively or prospectively. Organized local pushback may have failed to materialize in Glendale because opponents lacked a relevant political window during which to act.

Constituency Organization

The preceding discussion suggests that opposition to these projects was more relevant than support for them, in election campaigns. We next discuss the organization of relevant constituencies as a contributing factor to this balance of interests in the political sphere. Interview respondents with expertise in workforce preparedness and training told us that worker recruitment would be one of the key challenges associated with implementing these projects. Importantly, this implies that, prior to the plants’ opening, the worker constituency targeted by the IRA’s job-quality incentives was not concentrated in beneficiary communities and thus, by definition, lacked the potential to mobilize in support of the policies at the heart of the law.

Construction and electrical jobs materialized quickly after the projects’ announcement, and these jobs were high-quality compared with other local opportunities. However, the structure of the building trades may limit the extent to which construction and electrical workers are mobilizable in support of place-based industrial policy. A union leader expressed a distinction between the organization of labor in trades (such as electrical and construction workers) versus sectors (such as auto workers). Skilled trades workers feel more loyalty to their union, relative to their employer.

This is largely because jobs in the trades are fundamentally temporary, and the union plays the role of training and placing workers in new jobs. This lends the union more allegiance with workers. The higher loyalty that trades workers feel to their union, relative to their employer, is important because it could undercut the importance of any particular employer, job, or, crucially for this context, policy associated with a specific employment opportunity.

This could, in turn, weaken the incentive to take political action in defense of any given job. Of course, workers’ loyalty to the union also lends the union strong influence, relative to the employer, in persuading and mobilizing workers. One union leader reported that building trades union members reliably showed up to county commission meetings and other events to support Blue Oval. However, he also acknowledged their show of force was hindered by low density of union membership in the rural, agricultural community hosting the project.

Overall, the sparseness of local union membership prior to the projects’ initiation constrains workers’ power. While job creation is the clearest first-order benefit of these projects, the worker constituency was not (yet) sufficiently locally concentrated in the early phases of the investments to exert strong influence. By contrast, residents who experienced or anticipated costs were physically present, locally attached, and primed for activation (as, for instance, in the Marshall Township election).

Conclusion

Our analysis points to several actionable insights for policy design and communication. The leaders and workers we spoke with were enthusiastic about the opportunities presented by the Blue Oval projects, but they did not view federal policy as pivotal to their success. This perspective reveals that the link between local benefits and place-based industrial policy are not self-evident without a robust communication effort.

Federal legislators did not connect the dots between government incentives and local benefits. To be sure, Michigan Governor Gretchen Whitmer experienced some pushback in the media for claiming credit prospectively for local job creation that later fell short of projections (Hendrickson 2022). Her experience shows that credit claiming or communicating the links between local benefits and government policy is a complicated and risky proposition in an uncertain political and economic context. Even so, the experience highlights a need to develop effective strategies for communicating the benefits of submerged policies.

Of course, hyperpartisanship disincentivized most of the legislators representing these districts from communicating the benefits of the investments. In the early days of IRA implementation, it became clear that the bulk of the investments it facilitated were going to districts represented by Republicans (Gaffney 2024; E2 2024; Tamborrino and Siegel 2023). Consistent with recent political science scholarship (Lee 2016), in a political context in which elections are decided on a knife’s edge, the risk of crediting the other party outweighed the benefit of highlighting an economic opportunity for the district. In the future, policy champions need a communication strategy that does not rely on legislators bucking their partisan interests.

The location of these projects in primarily agricultural areas may have limited their potential for generating enthusiasm among the worker constituency most immediately impacted. We observed strong enthusiasm for the projects among representatives of the local “growth machines” (Logan and Molotch 2007), whereas the workforce for the facilities simply was not yet present locally, in a concentrated way. This limited worker mobilization potential.

Through its tax credit bonuses for locating projects in communities historically reliant on fossil fuel industries, the IRA was designed to ensure that benefits accrued to communities most vulnerable to the transition away from fossil fuels. In theory, this could neutralize opposition from and potentially build support among displaced workers. But nothing about the energy community designation ensured that potential workers were present, organizable, or mobilizable. One could imagine a design that could ensure this, for instance by providing a bonus to companies who locate facilities in communities whose unemployment rate exceeds a specified threshold. Additionally, consistent with critiques of “deliverism” as a political strategy (Hertel-Fernandez 2025), policy could include resources for organizing and empowering workers in the short term.

It is tempting to imagine that IRA-facilitated projects could have generated greater enthusiasm with more time. The problem with this logic is that the passage of the IRA was probably not the binding constraint for the timing of either of these projects. Instead, the IRA’s tax credits were a strengthening tailwind for projects that were already underway. Both states had been courting auto manufacturers to occupy these greenfields for decades, and both states provided hefty incentive packages for the Blue Oval facilities. Indeed, Kentucky’s BOSK was announced in 2021, prior to the passage of the IRA (BOSK later received an award from the DOE Loan Programs Office and would have been eligible for the advanced manufacturing tax credit).

Of course, local interpretations likely do change over time as projects mature, and these changing interpretations could have political repercussions. We plan to continue studying the longitudinal effects of investments such as the ones we study here, in the wake of policy changes that have led to tremendous uncertainty and major changes around these and other projects (Gardner 2025; Garcia and Dederer 2025b). But the underlying dynamics here do not support the idea that passing the IRA earlier in the Biden presidency would have resulted in greater enthusiasm among mobilizable constituencies, by the time of the next election and subsequent policy-making processes.

References

Béland, Daniel, Andrea Louise Campbell, and R Kent Weaver. 2022. Policy Feedback: How Policies Shape Politics. Cambridge University Press.

Bergquist, Parrish, Matto Mildenberger, and Leah C Stokes. 2020. “Combining Climate, Economic, and Social Policy Builds Public Support for Climate Action in the US.” Environmental Research Letters 15 (5): 054019.

Breetz, Hanna, Matto Mildenberger, and Leah Stokes. 2018. “The Political Logics of Clean Energy Transitions.” Business and Politics 20 (4): 492–522.

Campbell, Andrea Louise. 2002. “Self-Interest, Social Security, and the Distinctive Participation Patterns of Senior Citizens.” American Political Science Review 96 (3): 565–574.

E2. 2024. Clean Economy Works: Inflation Reduction Act Two-Year Analysis. https://e2.org/wp-content/uploads/ 2024/08/E2-Clean-Economy-Works-IRA-Two-Year-Review_August-2024.pdf. Accessed: December 1, 2025, August.

Gaffney, Austyn. 2024. “The Clean Energy Boom in Republican Districts.” The New York Times (November). https://www.nytimes.com/2024/11/21/climate/ira-republicans-clean-energy.html.

Garcia, Joseph, and Alex Dederer. 2025a. “1,600 Workers Laid off at Glendale, Kentucky, Plant as Ford Continues Major EV Pivot.” Accessed: December 16, 2025, December. https : //www . whas11 . com / article /money / business/ford-blueoval-sk-glendale-facility-ev-battery-plant-storage/417-224e49fd-a723-429f -88e4-a61df7b7a2ab.

——. 2025b. “1,600 Workers Laid off at Glendale, Kentucky, Plant as Ford Continues Major EV Pivot.” Accessed December 15, 2025. https://www.whas11.com/article/money/business/ford-blueoval-sk-glendale-facility-ev-battery-plant-storage/417-224e49fd-a723-429f-88e4-a61df7b7a2ab.

Gardner, Paula. 2025. “Ford Retreats from EVs after Michigan Spends Big on Subsidies for Battery Plants.” Accessed December 17, 2025. https://bridgemi.com/business-watch/ford-retreats-from-evs-after-michigan-spends-big-on-subsidies-for-battery-plants/.

Hacker, Jacob S. 2019. “Medicare Expansion as a Path as Well as a Destination: Achieving Universal Insurance through a New Politics of Medicare.” The ANNALS of the American Academy of Political and Social Science 685 (1): 135–153.

Hendrickson, Clara. 2022. Whitmer Touts Growth in Auto Industry, but Jobs Data Doesn’t Show That. https://www. freep.com/story/news/politics/2022/07/01/whitmer-auto-job-growth-fact-check/7776867001/.

Hertel-Fernandez, Alexander. 2024. “Civic Organizations and the Political Participation of Cross-Pressured Ameri-cans: The Case of the Labor Movement.” American Political Science Review, 1–17.

——. 2025. “Beyond Deliverism.” Accessed January 24, 2026. https://democracyjournal.org/magazine/76/ beyond-deliverism/.

Ho, Daniel, Kosuke Imai, Gary King, and Elizabeth Stuart. 2007. “Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference.” Political Analysis 15 (3): 199–236.

Ho, Daniel, Kosuke Imai, Gary King, and Elizabeth A Stuart. 2011. “MatchIt: Nonparametric Preprocessing for Parametric Causal Inference.” Journal of Statistical Software 42:1–28.

IRS. 2024. Frequently Asked Questions About the Prevailing Wage and Apprenticeship under the Inflation Reduction Act. https://www.irs.gov/credits-deductions/frequently-asked-questions-about-the-prevailing-wage-and-apprenticeship-under-the-inflation-reduction-act. Accessed: April 3, 2025.

Jacobs, Lawrence R, and Suzanne Mettler. 2018. “When and How New Policy Creates New Politics: Examining the Feedback Effects of the Affordable Care Act on Public Opinion.” Perspectives on Politics 16 (2): 345–363.

Kuriwaki, Shiro. 2025. “Cumulative CES Common Content.” Harvard Dataverse.

Lee, Frances E. 2016. Insecure Majorities: Congress and the Perpetual Campaign. University of Chicago Press.

Logan, John R, and Harvey Molotch. 2007. Urban Fortunes: The Political Economy of Place, with a New Preface. Univ of California Press.

Marshall Advisor & Chronicle. 2024. “Aug. 6 Primary Election Results.” Marshall Advisor & Chronicle (August).

Meckling, Jonas. 2019. “A New Path for U.S, Climate Politics: Choosing Policies that Mobilize Business for Decar-Bonization.” The ANNALS of the American Academy of Political and Social Science 685 (1): 82–95.

Mettler, Suzanne. 2007. Soldiers to citizens: The GI Bill and the making of the greatest generation. Oxford University Press.

Mettler, Suzanne, and Joe Soss. 2004. “The consequences of public policy for democratic citizenship: Bridging policy studies and mass politics.” Perspectives on politics 2 (1): 55–73.

Patashnik, Eric M. 2023. Countermobilization: Policy Feedback and Backlash in a Polarized Age. University of Chicago Press.

Pero, Olivia. 2024. China’s 2 EV Battery Plants in Michigan Spark National Security Concerns. https://www. dailysignal.com/2024/11/04/chinas-2-ev-battery-plants-in-michigan-spark-national-security-concerns/. Accessed: April 4, 2025, November.

Pierson, Paul. 1993. “When effect becomes cause: Policy feedback and political change.” World politics 45 (4): 595–628.

Plumer, Brad. 2025. “How the G.O.P. Bill Will Reshape America’s Energy Landscape.” Online, New York Times (July). https://www.nytimes.com/2025/07/03/climate/congress-bill-energy.html.

Polarization Research Lab. 2025. America’s Political Pulse. Accessed: November 24, 2025. search.americaspolitic alpulse.com.

Schattscheider, E.E. 1935. Politics, Pressures, and the Tariff. Prentice-Hall.

Soss, Joe. 2000. Unwanted claims: The politics of participation in the US welfare system. University of Michigan Press.

Stokes, Leah Cardamore. 2020. Short circuiting policy: Interest groups and the battle over clean energy and climate policy in the American states. Oxford University Press, USA.

Tamborrino, Kelsey, and Josh Siegel. 2023. Big winners from Biden’s climate law: Republicans who voted against it. hhttps://www.politico.com/news/2023/01/23/red-states-are-winning-big-from-dems-climate-law-00078420.

United Auto Workers and Ford Motor Company. 2023. UAW-Ford Agreement. Accessed: March 27, 2025. https://uawford.org/about-us/uaw-ford-contracts/.

1. We use a series of model specifications to ensure the robustness of our results to the influence of unobserved characteristics that may be associated with voters’ choices and firms’ investments’ decisions. [↩]