Decarbonize America’s Growth Model

From a distance, American policymakers can seem disproportionately focused on “making stuff in America.” After all, this is a country that has been steadily offshoring production and jobs for decades, and where 9 in 10 Americans work in the service sector. The Trump and Biden administrations had particularly strong producerist rhetoric, but nearly every president before them had some version of this as well. If the US returns to attempting to decarbonize in 2029, policymakers need to have some tough conversations about whether they are getting the right balance between producerist and consumerist logics. We argue that one tool for doing so is growth model theory (GMT), which focuses on how demand drivers vary across countries.¹

Growth Models à l’Américaine

In a sense, a country’s growth model can be considered its business model: In short, how does a country typically “make money”?² A country’s growth is always driven by some combination of exports, consumption, investment, and government spending—namely, the four components of Keynesian macroeconomic models.

Crucially for policy purposes, a growth model perspective allows researchers to decompose the components of a country’s demand to understand what drives its economic growth. Often, scholars find that characteristics and outcomes cluster around countries’ specific sources of demand. For example, policymakers governing export-led growth models may focus on keeping wages low to encourage export competitiveness. By contrast, policymakers governing economies where growth is driven by domestic demand may expand credit markets to prop up household consumption.

What type is the US growth model? Growth decomposition and accounting of the US economy is a helpful first step. Notably, the primary growth driver for the United States is household consumption, at 54 percent of GDP. The primary sectors for growth are services, at 50 percent. The US growth model is thus typically categorized as “strongly consumption led,”³ driven by household spending that is propped up by debt and credit markets.⁴ The dominance of services, typical of countries once they’ve developed, has risen over the past few decades.

The US economy is built around consuming and servicing, not producing and exporting. Manufacturing accounts for just 7 percent of growth,⁵ while services and construction together account for the overwhelming majority. Furthermore, many social scientists argue that consumption-led growth is a feature, not a bug, of hosting the world’s reserve currency.⁶ From this perspective, the status of the dollar structurally overvalues the exchange rate, making certain price-sensitive manufactured exports persistently uncompetitive on global markets. Thus, some of the above macrofinancial realities serve as situational constraints even before industrial policy enters into the picture.⁷

The composition of the US growth model has direct implications for decarbonization strategy, and may complicate manufacturing-based climate policies. For instance, because consumption rather than exports drives American growth, the economic beneficiaries of green manufacturing will be fewer and weaker than in export-led economies. These constraints thus frame a core strategic choice: Do we decarbonize through strategies that work with the consumption-led growth model, or push against it by attempting to build competitive green manufacturing capacity?

The composition of the US growth model has direct implications for decarbonization strategy, and may complicate manufacturing-based climate policies.

Where Decarbonization Comes In

In the following sections, we examine the different paths to decarbonization that policymakers might take if guided by growth model theory.⁸

Path 1: Decarbonizing with the Growth Model

One path for US decarbonization runs with the grain of its growth model rather than against it. A consumption-led economy with dominant service and construction sectors has real comparative advantages in a green transition, but they are different advantages than those enjoyed by export-led manufacturing economies.⁹ The US is already the world’s largest importer of manufactured goods, with deep institutional expertise in procurement, logistics, and deployment. Rather than treating import dependence as a policy failure, policymakers could treat it as a feature: deploying cheap Chinese solar panels, electric vehicles (EVs), and batteries at scale, while capturing value domestically in installation, maintenance, grid integration, and consumer financing. Construction and installation employment is, by definition, non-offshorable. The US building trades—electricians, ironworkers, HVAC technicians—are well-positioned to be the primary workforce of a domestic energy transition, regardless of where the hardware is manufactured.

The US also retains genuine comparative advantage as a global innovation hub. American firms still dominate in software, grid management platforms, and intellectual property–intensive cleantech. As the energy system grows more complex—integrating distributed generation, storage, demand response, and EVs—software and system integration become increasingly valuable relative to hardware. A services-led decarbonization strategy that leans into these strengths is coherent and realistic, even if it looks different from the green industrial strategies of Germany or China.

Path 2: Decarbonizing Against the Growth Model

The alternative path is transforming the US into a competitive exporter of green technologies. This path faces some structural headwinds, but addresses the increasingly important question of where productive capacity and manufacturing knowledge reside. A production-centered strategy has important rationales on its own terms: Manufacturing capacity matters for supply chain resilience, technological autonomy, a defense industrial base, and—proponents argue—the kind of high-wage employment that service-sector alternatives have not historically replicated at scale. The question the growth model framework poses is not whether these objectives are worth pursuing, but under what structural conditions they are achievable.

The US attempted this strategy under the Obama administration and again more ambitiously with the Inflation Reduction Act (IRA) during the Biden administration. In both cases, it assembled manufacturing coalitions that were perhaps narrower and less immediately politically influential than those that underpinned decarbonization in export-led economies.¹⁰

Economies that succeeded at creating such coalitions did so not merely because of policy ambition, but also because they possessed preexisting institutional infrastructures complementary to their creation—dense networks of manufacturing small and medium enterprises (SMEs), vocational training systems, and patient regional finance institutions on which green industries could be built.

The US has at least partially dismantled these infrastructures over decades of financialization and offshoring. Regional development banks and vocational training institutions have atrophied, and SME supplier networks have thinned as consolidation and offshoring have reduced the pool of firms with embedded process knowledge. Building such coalitions from scratch is more cumbersome than it would be to activate existing ones. The IRA showed it is possible but slow. Its manufacturing-centered provisions proved more durable than its broader climate ambitions, surviving in the legislative process precisely because producers had organized effectively around them—even as other parts of the law were unwound.

Navigating the Trade-Offs

Picking whether to lean into or against the growth model is difficult, but a few strategies have proved useful.

Industries’ Place in the Growth Model

The first is to examine which industries require more or fewer changes to the existing growth model. The Draghi report worked to clarify those industries for the European Union, focusing on the production of green and digital technologies as future sectors that were within reach of its comparative advantage.

Some industries may now be beyond reach. Chinese dominance in current-generation solar panels is now essentially unassailable on cost. Tariffs that raise solar deployment costs could undermine the consumption-led coalition dependent on low energy costs without building a sustainable domestic manufacturing alternative. Accepting import dependence here can be a strategic choice that frees resources for sectors where the US has an edge, while offering value to the deployment coalition that the consumption-led decarbonization path depends on.

Other industries sit in a middle ground where the question is less about who makes the technology and more about where jobs are located. EV assembly and some battery manufacturing might fall into this category. Domestic content requirements and inward foreign direct investment (FDI) from Korean and Japanese firms represent a defensible version of this approach, trading some economic efficiency for visible employment.

A third category is defined by geopolitics rather than market logic. Grid hardware, advanced nuclear components, and certain semiconductor inputs matter not because the US can compete globally in producing them but because dependence on foreign suppliers creates unacceptable strategic vulnerability. The goal here is maintaining minimum domestic capacity as insurance, not becoming an export industrial powerhouse.

Finally, there are sectors where the US retains genuine advantages not yet consolidated by competitors—AI-enabled grid management software, next-generation storage, advanced nuclear design, and the smart manufacturing tools discussed earlier. These are the sectors that warrant the most ambitious and sustained policy support, precisely because the window of advantage is real but not permanent.

Coalitional Politics

A second strategy is to think through the coalitional politics of each strategy. Decarbonizing with the growth model—through deployment, services, and consumption—can draw on a genuinely broad domestic coalition. Construction unions, installation and maintenance workers, utilities investing in grid infrastructure, households benefiting from falling energy costs, and large technology firms requiring cheap renewable power all have material interests in a deployment-centered green transition. This coalition is relatively insulated from trade politics, and its core beneficiaries cannot be offshored. As Kupzok and Nahm argue, the political viability of green macrofinancial bargains depends on whether promised economic co-benefits are actually delivered to constituents with political weight.¹¹ Deployment-based jobs are visible, local, and immediate.

Manufacturing coalitions can coexist with the deployment-style coalitions. But they will be built most plausibly by pursuing a narrower niche in advanced green manufacturing in sectors that complement US strengths in software and AI—not by competing on volume in sectors where Chinese dominance is now largely consolidated. China’s growing manufacturing edge derives not from low wages but from increasingly systematic deployment of (and financial support for) AI and robotics on factory floors, yielding productivity per worker that now exceeds comparable US plants across sectors.¹² The US, for now, is ahead of China in frontier AI development, but has been comparatively slower to deploy these tools in manufacturing contexts at scale.

A targeted industrial policy focused on upgrading manufacturing competitiveness by connecting legacy equipment to digital systems, building shared infrastructure for small and midsize manufacturers, and retraining workers could in principle close this gap in specific manufacturing niches. This is a narrow and technically demanding strategy. However, in specific sectors it may be the right one, particularly where geopolitical stakes or genuine innovative advantage make the case regardless of growth model constraints.

Conclusion

Neither path is likely to be sufficient on its own. Combining them means accepting import dependence in sectors where Chinese dominance is consolidated, while directing institutional resources toward sectors where the US retains genuine innovative or geopolitical rationale for domestic production. The analytical challenge is compounded by the fact that the relevant policy levers sit across different agencies with different mandates and limited coordination mechanisms: Trade policy, industrial subsidy, and macrofinancial management are rarely integrated in the same institutional venue, even when the strategic logic demands it. The growth model perspective does not resolve these choices, but it clarifies what is actually being traded off in each—and why strategies calibrated for export-led economies are unlikely to transfer directly to the US context.