Introduction

As part of the 2015 Paris Agreement, countries agreed to revisit and enhance their nationally determined contributions (NDCs) that outline greenhouse gas (GHG) emissions reduction plans every 5 years1. Following the original pledges, countries submitted a second round of updated and new NDCs, and some have also submitted long-term strategies (LTSs)2 and net-zero emissions targets3. This second round of NDCs suggested higher ambition compared to the 2015 NDC4,5. Yet, limiting global warming below 1.5 °C this century—the aspirational long-term goal of the Paris Agreement—will require countries to further ratchet or increase ambition in 2030 and beyond6,7,8,9,10,11,12. By February 2025, countries are expected to submit a third round of NDCs that outline GHG reduction goals for 203513. This third round of NDC updates, revisions, and enhancements will be critical in shaping the world’s alignment with the Paris Agreement’s long-term goals and will be informed by the outcome of the first global stocktake, which concluded at COP28 in 2023. While the stocktake showed that progress on GHG emission reductions was too slow, countries responded with a commitment to accelerate action by 2030, including transitioning away from fossil fuels, and accelerating the deployment of renewables and energy efficiency. They also agreed that the next round of NDCs would cover all greenhouse gases and be economy-wide14.

Given the role of the US as the world’s largest economy, and second largest GHG emitter15, as well as its anticipated effect in motivating higher ambition internationally, setting an ambitious and plausible 2035 NDC in the US could be crucial in the continued delivery of emissions reductions essential for keeping the world on a 1.5-compatible pathway. At the time of this study, the US NDC commits to reductions in US net economy-wide emissions of all GHGs by 50–52% below 2005 levels by 203016. More recently, the US also announced an NDC that commits to 61-66% reductions in economy-wide GHGs by 203517. Furthermore, the US has set a goal to achieve net-zero GHG emissions by no later than 205018. Federal programs such as the Inflation Reduction Act (IRA) of 2022 and the Bipartisan Infrastructure Law (BIL) of 2021 may support achieving these goals and include measures to incentivize investments in low-carbon technologies and infrastructure, transportation infrastructure, methane reduction, and others19. In addition, action from a broad range of non-federal and subnational actors are expected to contribute toward these goals17. Collectively, these and other measures are expected to support a credible pathway for the US to respond to the commitments that were made in COP28 and beyond. Yet, recent analyses suggest a gap between current policies and emissions reductions required in 2035 to be on track to the announced NDC and net-zero emissions by 205020,21,22. Various stakeholders are faced with the following important questions: (i) What is the potential range of GHG emissions reductions in 2035 with an expanded set of measures beyond current policies that can help to close the gap relative to the net-zero pathway? and (ii) What will such a range of reductions imply for sectoral emissions and transformations?

This study brings together four US energy-emissions modeling teams to provide a bottom-up estimation of emissions and energy system impacts across a coordinated set of policy scenarios (Table 1, Methods). These scenarios include both current measures as well as additional federal and state actions that extend and expand upon current measures. This study contributes to the literature in several ways. Foremost, most previous studies on US policy analysis focus on either understanding the emissions and energy-system impacts of current policies through 2030 or 203520,22,23,24,25, or energy system transformations required in the long-term to achieve net-zero emissions by 205018,26. This study bridges the above two strands of literature by exploring expansions of current measures and their emissions and energy-system impacts through 2035 and illustrating how they might close the gap relative to the path toward net-zero. By doing so, this study informs the next round of NDC announcements and enables decision-makers to make informed decisions about an ambitious and plausible range of 2035 emission reductions and associated tradeoffs under different policy conditions. Second, recent studies that have begun to explore expansions of current measures are based on single models21,27. This analysis builds on the above studies by providing additional models and policy scenarios to understand variations in potential policy responses. By employing multiple models, this study provides insights into robust conclusions across models and future uncertainties that single-model studies cannot provide. While studies based on single models can be useful in understanding parametric uncertainties (that is, uncertainties related to factors like technology costs and policy stringency), multi-model studies offer further insights into the structural uncertainties (that is, uncertainty in the mathematical abstractions of real-world processes) which can be equally if not more important26,28,29,30. Furthermore, this study offers detailed insights into sectoral transformations and potential implications of an expanded set of national and subnational policy scenarios.

Table 1 Policy scenarios explored in this study. See Methods for a detailed list of policies in each scenario
Full size table

We explore four policy scenarios that are drawn from previous analyses and informed by stakeholder engagement (Table 1, Methods)31. The Current Measures scenario includes on-the-books policies and incentives as of early 2024, including federal measures such as IRA’s investment and production tax credits, BIL incentives, vehicle standards, and state-level targets such as renewable portfolio standards (RPSs), and assumes full implementation of these policies (Methods). The High Ambition scenario includes all policies under the Current Measures scenario along with extensions of those policies across time, and additional policies from ref. 21 to understand the energy systems, economic, and emissions implications of proposed policies and incentives. This scenario includes regulations for fossil fuel power plants, enhanced RPSs, widespread adoption of electric vehicle (EV) sales targets, industry standards for electrification and carbon capture utilization and storage (CCUS), and building electrification incentives (Methods). Note that at the time of the study, the Clean Air Act Section 111 (b) and (d) emission guidelines of the US Environmental Protection Agency (EPA) for new and existing stationary fossil fuel-fired power plants were in the “proposed” stage and are therefore not included in the Current Measures scenario, but only in the High Ambition scenario32. Similarly, EPA’s Multi-Pollutant Emissions Standards rule for light-duty and medium-duty vehicles was also in the proposed stage and, therefore, excluded from the Current Measures scenario. Our assumptions in the High Ambition scenario result in meeting or exceeding the expected outcomes of this rule (Supplementary Fig. 2)33. Both these rules have recently been finalized, although there are court challenges to each.

The High Ambition+ scenario, includes all policies under the High Ambition scenario along with a 100% national clean electricity standard by 2035 (following the official goal) and a carbon fee with gasoline carveouts. Both measures had been discussed as part of the negotiations before the IRA’s passage but would require surmounting historical barriers compared to other measures. For example, there were 76 proposals for federal electricity portfolio standards introduced between the 105th and 116th Congress though none has become law34. Likewise, although a carbon fee has been argued to address the fiscal implications of the IRA, respond to international pressure as other countries begin to adopt carbon border adjustment mechanisms, and spur job creation in combination with IRA tax credits, setting a carbon price policy has historically been challenging in the US35. In that sense, the two High Ambition scenarios can be interpreted as providing a range in the political feasibility spectrum.

Finally, the LTS scenario includes all policies under the Current Measures scenario in addition to a constraint on economy-wide emissions that is consistent with the US long-term strategy. While the Current Measures scenario provides a reference based on current policies to compare our High Ambition scenarios, the LTS scenario provides an alternative reference based on stated federal goals.

Although this study protocol required all models to be coordinated in the representation of policies in the four scenarios, the models exhibit variations in the coverage and implementation of policies due to differences in the models’ scopes and resolutions (Methods, Supplementary Methods). While Current Measures and High Ambition scenarios were required scenarios, the High Ambition+ and LTS were optional. Three of the four participating models simulated all four scenarios. Variations in the implementation of the policies could also be driven by their complexity, which sometimes requires modeling teams to use their own judgments. Additionally, it is important to note that there was no systematic effort to harmonize assumptions across models, which may drive some of the variation in emissions outcomes across models. Nevertheless, a multi-model approach can show which results are robust despite these differences. It should also be noted that the multi-model results do not reflect the full range of uncertainty and should not be interpreted as statistical distributions. Future modeling efforts might investigate additional uncertainties about policy combinations, technology cost and availability, as well as market drivers (e.g., data center load growth, LNG export impacts on natural gas prices, etc.).

Results

Economy-wide emissions outcomes

Under the Current Measures scenario, economy-wide GHG emissions reductions in 2035 are 34–44% relative to 2005 levels across models (Fig. 1). Compared to the official US LTS of reducing economy-wide GHG emissions to net-zero by 2050, this range suggests a gap in 2035 emissions of 1200–1900 MtCO2e. Under the same scenario, reductions in CO2 emissions from fossil fuels and industry in 2035 range from 34 to 45%, suggesting a gap of 1100–1700 MtCO2 relative to the official US LTS. By contrast, under the High Ambition scenario, reductions in economy-wide GHG emissions and CO2 emissions from fossil fuels and industry in 2035, respectively, range from 56 to 65% and 51 to 65%. Under the High Ambition+ scenario, these reductions, respectively, range from 65 to 67% and 62 to 64%. For comparison, the US LTS corresponds to a 62.5% reduction in 2035 emissions relative to 2005 and the recently announced NDC corresponds to 61-66% reductions in 2035 relative to 2005. Thus, the expanded set of policies in the High Ambition scenarios significantly closes the gap relative to the official US LTS and, in some cases, suggests greater emission reductions relative to the LTS  as well as close alignment with the recently announced NDC.

Fig. 1: Emissions ranges across scenarios.
figure 1

The range of economy-wide greenhouse gas emissions (a) and energy CO2 emissions (b) across models and scenarios, and the range of emission reductions in 2035 (c). The shading illustrates ranges across models for a given scenario. The faint lines within a shaded region show individual model results. The dashed gray line in both panels shows the emissions pathway associated with the official US Long-term Strategy18. The solid gray line in both panels shows historical emissions. Scenario assumptions are detailed in Table 1 and Methods. Historical emissions are from the US Environmental Protection Agency75.

Full size image

Sectoral emissions implications

Under the Current Measures scenario, there is general agreement across models in the contributions of the power, transportation, and buildings sectors toward total emissions reductions (Fig. 2). Most of the CO2 emission reductions in 2035 across models occur in the power sector (60–76% reductions relative to 2005) followed by transportation (23–43%), buildings (5–22%), and industrial (4–20%) sectors. Under the High Ambition scenario, reductions in CO2 emissions from all sectors are higher than the Current Measures scenario (reductions of 73–95%, 41–57%, 31–46%, 17–50% respectively). Under the High Ambition+ scenario, power sector emissions across models are near zero due to the implementation of the 100% clean electricity standard and the carbon fee. Any non-zero power sector emissions in this scenario are driven by the deployment of CCUS technologies in combination with fossil power plants (discussed subsequently). Emissions from other sectors remain broadly consistent with the High Ambition and LTS scenarios.

Fig. 2: Emissions by sector.
figure 2

Energy CO2 emissions by energy sector in history and 2035 across scenarios and models (a); trajectories for sectoral CO2 emissions under the High Ambition (b) and High Ambition+ (c) scenarios. In panels b, c, the shading illustrates ranges across models for a given sector. The faint lines within a shaded region show individual model results. Historical emissions are from the US Environmental Protection Agency75.

Full size image

Transitions in the power sector

As the sector that contributes most to emissions reductions in these scenarios, it is instructive to examine power sector capacity additions across models and scenarios (Fig. 3). Capacity additions from 2019 to 2023 ranged from 23–44 GW/year36. Renewables accounted for most of these due to a variety of factors, including federal investment policies, tax credits, and regulatory actions, as well as state policies, research and development, and market trends. Under the Current Measures scenario, average capacity additions in the next decade range from 62 to 102 GW/year across models, which are led by solar, wind, and energy storage. Proposed additions in 2024 are 65 GW/year36 and consistent with the lower end of the above range. However, under the High Ambition scenarios, average capacity additions in the next decade scenarios are considerably higher compared to history, ranging from 83 to 154 GW. This corresponds to about a two to four times increase in average annual renewable capacity additions in the next decade relative to the average additions between 2019 and 2024, and a 73–100% reduction in electricity generation from unabated fossil fuels by 2035 (Fig. 4). These transitions are driven by strengthened regulations on coal- and gas-fired power plants and coal phaseout policies to reduce electricity generation from fossil fuel sources, extended renewable tax credits, and enhanced state-level portfolio standards in these scenarios.

Fig. 3: Capacity additions in the power sector.
figure 3

Historical and projected electricity generation and storage capacity additions by technology and by model in 2035 in GW/year. Historical capacity additions data are based on the US Energy Information Administration36.

Full size image
Fig. 4: Key sectoral outcomes.
figure 4

Sectoral metrics in 2035 across models and scenarios. Within each metric, different colors correspond to different models, and different symbols correspond to different scenarios (see legend on the right).

Full size image

Transitions in the transportation, buildings, and industrial sectors

Under the High Ambition scenarios, the transportation sector is the second largest source of emissions reductions, with electrification of light-duty and freight vehicles being the primary driver of these reductions (Figs. 2, 4). Extended tax credits for electric vehicles (EVs), strengthened fuel economy standards, and enhanced EV sales mandates help accelerate road transport electrification and lower emissions in these scenarios. Under the High Ambition scenarios, the new sales of EVs in LDVs in 2035 account for 83–100% across models (compared to 45–68% under the Current Measures scenario). Beyond electrification, the High Ambition scenarios assume measures to reduce vehicle miles traveled (VMT). These imply a reduction in the use of single-occupancy vehicles, sustainable urban planning, and promoting sustainable modes of transportation37.

Under both the High Ambition scenarios, emissions reductions in the buildings sector are driven by extended IRA electrification and efficiency incentives along with state-level zero-emissions appliance standards and efficiency standards, which target water and space heating appliances. In 2035, the buildings sector is more electrified under the High Ambition scenarios (55–66% across models) compared to the Current Measures (47–59%) scenario (Fig. 4). Additionally, the buildings sector is more electrified compared to the transportation and industrial sectors: under the High Ambition scenarios, the share of electricity in transportation and industrial final energy use in 2035 are respectively 13–20% and 27–34% across models in contrast to 55–66% for the buildings sector which partly reflects its higher starting point (47% in 2021). However, under both the High Ambition scenarios, future emissions from the buildings sector reduce at a slower rate relative to historical levels compared to the power and transportation sectors (Fig. 2B). This is because, the power and transportation sectors have higher emissions to begin with (making up 32% and 36% of US energy CO2 emissions in 2021, respectively), have more technological options to substitute away from fossil fuels, and include a wide range of policy measures at the federal and state level38. By contrast, the buildings sector has lower emissions historically (making up 11% of US energy CO2 emissions in 2021), with slower stock turnover and fewer technological options and policy measures (Fig. 4).

Both the High Ambition scenarios are characterized by higher electrification and accelerated retirement of fossil fuels in the industrial sector compared to the Current Measures scenario (Fig. 4). The share of electricity in industrial final energy under the High Ambition scenarios ranges from 27–33% across models in 2035 (compared to 13% in 2021 and 19-28% in 2035 under Current Measures across models). Notwithstanding differences in ability to represent various industrial policies across models, our results suggest a strong potential for additional measures and emissions reductions in the industrial sector to support achieving the official goals of the US, especially the US LTS.

Transitions in other sectors

The models used in this study differ in their ability to represent non-CO2 emissions and policies (Supplementary Methods). Under the High Ambition scenarios, non-CO2 emissions reduce by 23–37% between 2021 and 2035, in contrast to 7–16% under Current Measures across models driven by strengthened HFC regulations and methane mitigation policies.

Furthermore, the High Ambition scenarios are characterized by scaling up of CO2 sequestration from CCUS technologies, the deployment of which is incentivized through the IRA 45Q credit and other measures (Supplementary Fig. 3). While most of the sequestration occurs in the power sector through fossil fuels with CCUS and industrial CCUS applications, the High Ambition scenarios are also characterized by sequestration from engineered carbon dioxide removal (CDR) technologies such as direct air capture and bioenergy in combination with CCUS. The deployment of engineered CDR technologies in 2035 varies widely across models ranging from 152-1193 MtCO2 under the High Ambition scenarios in contrast to 78-611 MtCO2 under Current Measures. The wide range across models reflects greater uncertainty about scaling these emerging options. As explained earlier, the models also differ in their ability to model emissions and sinks from land-use changes. Land-use sinks in 2035 range from 656-882 MtCO2 under the Current Measures scenario and 733-988 MtCO2 under the High Ambition scenarios. It is important to note that the size of land-use sinks could be affected by a variety of factors, including the impact of climate change itself, which is not directly accounted for in this analysis. Nonetheless, the total deployment of CDR—including natural and engineered approaches in these scenarios—is broadly consistent with previous studies, including the official US LTS18,22,26.

Discussion

This study presents a multi-model and bottom-up assessment of the potential emissions and sectoral implications of an expanded set of federal and non-federal policy measures beyond current policies, which may be used in efforts to inform updated climate targets and prioritize supporting actions. Although this study focuses on the US which recently announced an updated NDC, it provides a framework that countries globally can employ as they update their NDCs in the coming months.

An important caveat of this study is that it does not attach probabilities to any of the scenarios. The High Ambition scenarios assume full implementation from a broad range of federal and non-federal actors and long-run certainty about potential policies remaining in place. Efforts were made for the High Ambition scenarios to be grounded in plausible policy strategies that have been implemented already and proposed. In reality, the feasibility of the policies and measures modeled in the scenarios would depend on a variety of political, social, and behavioral factors which are not fully captured in this modeling. Foremost, the durability of policies is uncertain due to potential changes in political environments39,40. Implementation of the high-ambition policies could also be affected by governance structures; consumer preferences and public perception; lobbying from stakeholders; and uptake by state, local, and individual actors41,42,43. For instance, if IRA’s energy provisions are repealed or not implemented as expected, reductions in economy-wide GHG emissions in 2035 could range from 27 to 35% relative to 2005 levels (Supplementary Discussion, Supplementary Fig. 10)23. By contrast, under the Current Measures scenario explored in this study (that includes the IRA), emissions reductions in 2035 range from 34 to 44%. This suggests that if current policies—particularly, federal measures such as the IRA—were not implemented as modeled in this study, emissions in 2035 could be notably higher, requiring even higher ambition and action—including from a broader set of subnational and non-federal actors—in subsequent years to achieve the recently announced NDC and close the gap relative to the US LTS44. As another example, if particular sectoral policies modeled in this study are repealed or do not work as intended (for example, due to consumer preferences), that could result in lower emissions reductions than suggested by this analysis, requiring higher ambition from other sectors and actors to close the gap relative to the US LTS (Supplementary Discussion, Supplementary Fig. 11). Finally, the models used in this study vary in their ability to represent the interactions and tradeoffs across policies—especially across those that cut across energy and agricultural and land-use systems. Further model capability enhancements could facilitate a deeper exploration of these effects.

The above caveats notwithstanding, these results suggest a strong role for enabling measures beyond those modeled in our High Ambition scenarios to address various challenges that are not explicitly modeled in these scenarios due to model limitations. Although the High Ambition scenarios suggest a two to four times increase in capacity additions in the power sector in the next decade, recent capacity additions are consistent only with the lower end of the range of modeled additions in the Current Measures scenario (Fig. 3). This is in large part due to lack of transmission capacity and delays in permitting processes45. Previous analyses suggest that current transmission planning, siting, permitting, and cost allocation practices could slow down the pace of transmission expansion and limit the emission reduction potential of policies such as those embodied in the IRA46. Therefore, measures that improve reliability and flexibility of the grid, such as investments in transmission, storage, and demand-side management, and permitting, siting, and interconnection reforms to support faster deployment of clean energy may be critical for an affordable clean energy transition. Although these scenarios indicate extensive electricity demand growth, including from end-use electrification, manufacturing, and other sources, they do not explicitly account for potential growth in electricity that might be driven by growth in data centers. Recent analysis shows how higher-end estimates for data center growth could increase near-term emissions in the near-term, but in the long run, such loads are much smaller drivers of emissions outcomes compared with other factors like the policy environment24,47.

These scenarios also suggest a strong role for electrification of the transport sector. Observed EV sales in the US in recent history are consistent with modeled outcomes under the Current Measures scenario48. Nevertheless, achieving the levels of transport electrification suggested by the High Ambition scenarios could be challenged by the lack of charging infrastructure, the limited supply of electric freight trucks currently, and the ability of supply chains (e.g. for critical minerals and materials such as lithium that are used in EVs) to respond to increasing demands49,50. Therefore, additional measures—beyond the measures modeled in these scenarios—to incentivize investments in new charging infrastructure, increased research, development, demonstration, and deployment (RDD&D) to facilitate innovations in transport electrification options beyond the LDV sector, and responsible supply chain management measures might be needed to mitigate these risks51,52,53. Other key barriers to the adoption of EVs include consumer preferences43,53,54,55. Additional measures, such as incentives for low-income and disadvantaged communities and public awareness campaigns, could help address some of these barriers56,57. In addition, continued investments in RDD&D of zero- or low-carbon alternatives for transportation sectors that exhibit slower decarbonization rates—such as aviation and shipping—could be important for reaching the net-zero goal.

Electrification of the buildings sector in the US is already higher compared to other sectors58. Nevertheless, further electrification and adoption of heat pumps is challenged by barriers such as high costs to electrify older buildings and building codes. Measures to address these could further accelerate building sector emissions reductions59,60. While consumer preference is an important barrier to building electrification, previous analysis has also shown that heat pump adoption is strongly correlated with geography, climate, and electricity prices43,61.

Additional industrial regulations, such as production tax credits for zero-emissions industrial heat and extensions of power plant regulations to include industrial boilers, could accelerate low-carbon transitions in the industrial sector. Additional industrial measures could also include continued investments in new technologies and process innovations to address industrial emissions such as high-temperature heat and process emissions from steel, petrochemical, and cement production, as well as zero- and low-carbon industrial innovations such as carbon capture and clean hydrogen produced from renewables, nuclear, or waste to power industrial facilities. Likewise, efforts to foster monitoring and control technologies that prevent the release of non-CO2 GHGs, maintain and enhance land-use sinks, and RDD&D of a broad array of engineered CDR technologies will be important in the next decade. Many of these efforts are already under way62,63, and may be important in laying the foundations for deeper emissions reductions required beyond 2035 to achieve net-zero emissions.

More broadly, implementation of the measures considered in this study could accelerate low-carbon technology cost reductions and technological change through increased deployment and associated technological learning. The concurrent expansion of clean energy industries and the associated investment and employment effects could alter future support for related policies in the long-term64. These broader effects merit future analysis.

Methods

Overview of models

In this multi-model study, we used results from four models to capture the complex energy system interactions across the economy. Models include: EPS-EI, GCAM-USA-CGS, PATHWAYS-NRDC, REGEN-EPRI. All models cover the energy system broadly, though policy coverage and implementation vary. These variations across models are due to modeling teams’ judgments on the high-ambition policies, as well as different model structures, which can be advantageous for modeling certain policies and disadvantageous for others. Table 2 summarizes the key attributes of the four models. More details about the models are provided in the Supplementary Methods.

Table 2 Summary of key model attributes
Full size table

Policies modeled under Current Measures scenario

The Current Measures scenario includes core climate and energy provisions of the Inflation Reduction Act and the Bipartisan Infrastructure Law, executive actions from the federal government like the corporate average fuel economy standards, and state-level policies such as renewable portfolio standards and EV sales targets. Some of the models in this analysis have informed the climate and energy impacts of the IRA23,25. Table 3 below contains additional information on the policies modeled in the Current Measures scenario. For detailed modeling of IRA provisions, see Supplementary Table 1.

Table 3 Modeled policies under the current measures scenario
Full size table

Policies modeled under High Ambition and High Ambition+ scenarios

In the High Ambition scenario, we build upon the Current Measures scenario and include additional policies at the federal and state levels. Additional policies include federal standards for coal- and gas-fired power plants, coal phaseout in the power sector, strengthened EV sales targets and building electrification standards. See Table 4 below for more detail.

Table 4 Additional policies under the high-ambition scenario
Full size table

The High Ambition+ scenario enhances ambition even further by layering on a carbon fee and clean electricity standard. The carbon fee is adapted from ref. 27, starting at $20 per ton CO2 in 2025 and increasing at 4% per year, with carveouts for retail gasoline sales. The clean electricity standard achieves 100% carbon-free electricity by 2035, following the official goal18. The High Ambition+ scenario was an optional scenario. Three out of the four participating models simulated this scenario.