This article was written by Alex Clark, Research Director at Asset Impact and Joss Blamire, Director of Infrastructure at GRESB. It was originally published in the September issue of Institutional Investing in Infrastructure (i3) and has been re-published with permission.
- Asset Impact's database, which tracks over 66,000 private and public companies and the 360,000 physical assets they own, and GRESB's Infrastructure Benchmark reveal critical trends in the U.S. power sector and highlights the challenges in decarbonizing America’s grid by 2035.
- Despite gains in renewable energy, gas continues to dominate U.S. electricity generation, leading to a high carbon intensity (350 grams CO2/kWh). By 2030, emissions will only decrease to 230 grams CO2/kWh, roughly where the European Union is today.
- While U.S. greenhouse gas emissions from power generation are projected to fall by 40% by 2030, rapid additional reductions will be needed to achieve net-zero by 2035, especially given the continued reliance on gas and slow adoption of renewable energy.
Decarbonizing America’s power grid: Challenges and opportunities on the road to 2035
As the United States approaches a consequential election that could significantly influence climate policy, the path to rapid and permanent decarbonization of its power sector remains uncertain. On the surface, progress appears promising: Coal use continues to decline, installation of solar and wind capacity is expanding at record rates, and clean technologies are receiving substantial government funding and tax breaks through the 2022 Inflation Reduction Act (IRA). However, gas combustion still meets more than 40 percent of U.S. electricity needs, and the country’s nuclear and hydroelectric power fleets will continue to age without substantial modernization or expansion.
Putting aside a potential shift in political leadership that could considerably slow (but would likely not reverse) the current trajectory, are the trends we see today enough to meet the administration’s goal of achieving net-zero emissions in electricity generation by 2035?
To shed some light on this question, we looked at GRESB’s Asset Impact asset-level database — which spans more than 360,000 physical assets operating in 11 energy-intensive sectors, linked to more than 66,000 public and private companies that represent more than 75 percent of global greenhouse gas emissions — and the results from GRESB’s most recent Infrastructure Assessments. Our analysis reveals some surprising trends in the U.S. power system in terms of where electricity is generated, from what physical sources and by whom, both today and in the years to come.
U.S. power mix — now and into the future
This year, power generation plants in the United States will produce more than 4 terawatt-hours of electricity. Of this total, 43 percent will be produced by gas-fired plants, 24 percent by nuclear and hydro plants, 19 percent from coal and just 14 percent from non-hydro renewables such as solar and wind.
By 2030, coal’s share is expected to fall to just 7 percent following a wave of plant retirements, and non-hydro renewables will jump to 22 percent. Despite this, gas will remain the dominant power source in the country, falling just 5 percent in absolute terms. While gas may yet prove to be an important “bridge fuel” to a net-zero future, it shows no sign of yielding its central position to low-carbon alternatives in the foreseeable future.
As a result of the current reliance on gas and coal, the carbon intensity of the U.S. grid is relatively high at approximately 350 grams of carbon dioxide per kilowatt-hour — only slightly lower than the average emissions of a gas-fired plant. By 2030, this falls to 230 grams per kilowatt-hour, roughly where the European Union is today.
Why the United States is not on track
With a focus on the most energy-intensive sectors, our Asset Impact products link carbon emissions of physical assets to corporate ownership trees, from direct owners to parent companies and the securities they issue. In addition to providing historical and current emissions data, Asset Impact solutions integrate capital expenditure plans and commissioning and decommissioning schedules.
Projections from Asset Impact (calculated using data from existing power plants and those with confirmed commissioning or decommissioning dates) portends a worryingly slow rate of progress in the coming years from an already high carbon baseline. Low-carbon power sources (renewables, hydro and nuclear) will supply 38 percent of power in 2024, but they are expected to rise to just 49 percent by 2030. This compares with 66 percent low-carbon power generation in the European Union in 2024, rising to 74 percent in 2030.
Greenhouse gas emissions from U.S. electricity generation are projected to decline 40 percent, from 1.6 gigatonnes of carbon dioxide equivalent in 2024 to just under 1 gigatonnes of carbon dioxide equivalent in 2030. Almost half of this drop will occur between 2029 and 2030 due to coal retirements, with the net retirement rate for gas plants remaining stubbornly low.
On average, emissions in the United States are projected to fall by 8 percent a year from 2024 to 2030. To get to net zero by 2035, the remaining 60 percent of emissions need to be eliminated in just five years, requiring a rate of decline that is at least half as fast again after 2030 than in the years before. In practice, the required rate of decline may be even higher than this, as projected increases in total electricity demand by 2030 — as high as 40 percent according to McKinsey & Co. — may result in more gas capacity additions this decade that could slow the speed of emissions reductions in the next few years.
At the state level, the largest utility-scale electricity generation fleets are in the populous states of Texas (116 gigawatts), California (82 gigawatts), Florida (66 gigawatts) and Pennsylvania (43 gigawatts).
California is home to the largest non-hydro renewable fleet, hosting 27 gigawatts in 2024 (rising to 33 gigawatts in 2029), followed by Texas (17 gigawatts rising to 22.6 gigawatts), with the less populous but electricity-exporting state of Nevada a distant third (5.7 gigawatts rising to 8.9 gigawatts).
Conversely, Texas hosts the country’s largest coal fleet (producing 17 gigawatts in 2024 and falling to 11.5 gigawatts in 2030), along with Indiana (16 gigawatts to 7.4 gigawatts), Kentucky (11.5 gigawatts to 7.6 gigawatts) and Pennsylvania (10.5 gigawatts to 1.8 gigawatts). Other states with coal fleets that exceed 8 gigawatts include North Carolina, Missouri, West Virginia, Illinois and Ohio.
In addition to coal, Texas also hosts the largest gas fleet by far (77 gigawatts), followed by Florida (49 gigawatts), California (38 gigawatts) and New York (25 gigawatts).
Energy ownership
While questions remain regarding how recent court decisions may affect future presidential administrations’ ability to drive climate policy, the corporate ownership of U.S. power plants and firm-level investment decisions will also greatly influence the potential pace of decarbonization.
More than three-quarters of U.S. power generation is attributable to American firms. Companies domiciled in other countries hold minimal market share in the United States, with Canadian firms being the largest ultimate owners of U.S. power generation, at only 2 percent of the total.
The 20 largest generation companies (15 of which are listed firms) deliver 44 percent of total U.S. electricity, while the 100 largest generators control 63 percent. By 2030, these figures fall to 41 percent and 60 percent, respectively, suggesting a slow but steady corporate decentralization in the power sector. The federal government is the largest single generator of electricity, based on its ownership of installed capacity. Among the remaining largest electricity producers, most are well-known in the sector — Duke, NextEra, Constellation, Entergy, Dominion, AEP, Xcel, etc. Only one — Iberdrola (which controls just 0.7 percent of total generation) — is not a U.S. firm.
In terms of emissions, generators domiciled in the United States produce 77.5 percent of the country’s electricity and 75.9 percent of emissions, indicating that they are slightly less carbon intensive than the average. (Unsurprising, given that U.S. firms control the low-carbon hydro and nuclear fleets.)
The top 20 emitters — most of which are also among the top 20 generators — will produce almost 700 million tonnes of carbon dioxide equivalent in 2024 — more than the total output of Canada, Mexico or Saudi Arabia, and almost twice that of the United Kingdom. In addition to 12 listed firms and the federal government, the largest emitters also include four private companies, a state government and two energy cooperatives. Within the top 20, corporations account for 80 percent of generation and 77 percent of emissions, and governments account for 15 percent of generation and 14 percent of emissions. Cooperatives among the top 20 account for 5 percent of generation but 9 percent of emissions. This implies that top 20 cooperatives are almost twice as carbon intensive than the average, raising interesting questions around their business models and capacity to decarbonize rapidly.
The “cleanest” top 20 generator is Constellation Energy by a wide margin, at just 60 grams of carbon dioxide equivalent scope 1 emissions per kilowatt-hour generated. The federal government’s emissions factor is also relatively low at 190 grams per kilowatt-hour, with nuclear and hydro assets comprising three-quarters of its portfolio. On the other hand, the most carbon-intensive generators all rely heavily on coal-fired power and include Touchstone Energy and PPL (both 730 grams per kilowatt-hour), Lightstone Generation (670 grams per kilowatt-hour), Associated Electric Cooperative (grams per kilowatt-hour) and AEP (560 grams per kilowatt-hour).
Infrastructure assets within the GRESB benchmark
For another perspective on the progress being made toward decarbonization, we examined a cross-section of assets that take part in the GRESB Infrastructure Assessments — which capture data on energy usage, greenhouse gas emissions and net-zero target-setting practices and methodologies for companies and funds across a range of sectors such as energy, water, waste and transport.
Looking at energy performance for the Americas regions in the 2023 GRESB benchmark, we find that, on average, just 10 percent of the energy used by participating infrastructure assets comes from renewable sources, an increase of only 0.5 percent on the previous year. In EMEA, the corresponding share was 28 percentage points higher, with a slightly more robust increase of 1 percent to 2 percent on average from the previous year. If the infrastructure sector in the United States is to make significant contributions to reducing greenhouse gas emissions through its own energy procurement, the pace of uptake of renewable power must increase significantly.
The degree of renewable-energy use is, of course, not consistent across the entire infrastructure sector in the United States. Looking at different sectors, we find transport leads with 17 percent of energy imported from renewable sources, followed by data infrastructure at 13 percent.
Interestingly, transportation is the only sector within the GRESB benchmark that increases its imports of renewable energy compared with the previous year, with a rise of 5 percent. Although most assets that take part in the GRESB benchmark have energy-related policies in place and report on their greenhouse gas emissions, only half of assets in the Americas have net-zero policies in place, which is significantly lower than in EMEA, where 68 percent of assets have net-zero policies.
Moving from policies to targets, we see that almost one in two infrastructure assets (47 percent) located in the Americas report a net-zero target. In comparison, the global benchmark average sits at 60 percent, slightly less than the EMEA region average (66 percent).
If these commitments are to be fully delivered upon, particularly with regard to net zero, we should expect a significant increase in renewables procurement in future reporting periods. For the time being, levels of adoption remain significantly below those required to achieve meaningful decarbonization and well behind U.S. electricity sector decarbonization more broadly.
Regardless of who assumes the presidency in 2025, the United States faces a daunting challenge in decarbonizing its power sector, and its progress lags well behind the European Union. U.S. electricity’s high carbon intensity also has a cascading impact on other sectors struggling to reduce their scope 2 emissions. With gas firmly established as the favored source of electricity, coal retirements alone are not enough to set the pace required to reach net zero by 2035. Some utility companies stand out as clean power champions, while others are still operating large coal and gas fleets with no clear plan to either phase them out or capture the emissions they produce.
Utilities and infrastructure owners alike have some hard questions to answer in the coming years. How do they plan to reach net zero — and how will they manage the potential risks of failing to do so?
Alex Clark is Research Director at Asset Impact, and Joss Blamire is Director of Infrastructure at GRESB. Data from Asset Impact and GRESB can help financial institutions to track progress and engage with the companies front and center in the U.S. energy transition.