Now that there are millions of electric vehicles (EVs) on U.S. roads, close attention is being paid to public charging reliability and accessibility, including plug compatibility, charger functionality, and the mechanics of payment. On all three fronts there’s good news for current and prospective EV drivers in the United States.

Thanks to a few big developments, in the coming years, nearly all EVs will be able to charge at nearly any public charger. Additionally, a federal program is slated to help ensure that chargers operate properly and that payment processing gets a lot easier by allowing users to use a single app to pay at any charger.

First let’s talk about compatibility and a newly formalized standard. Last spring, Ford made a major splash by announcing that starting in 2025, it will manufacturer its EVs using the North American Charging Standard (NACS) inlet derived from Tesla’s charging standard. After that, most major automakers (except Stellantis) and all the major charging infrastructure networks, including Electrify America, EVgo, Blink, and ChargePoint, made similar commitments to adopt the NACS inlet and connector in their North American vehicles and chargers, respectively. Then engineering standards-development organization SAE International said it would expedite the standardization process for NACS to make it an independent standard available for all. In December 2023, SAE released a Technical Information Report developing a standard for the “J3400” NACS connector.

Industry cohesion around the J3400 NACS charging standard, a universal plug shape, is significant because historically the U.S. market has had a variety of different connectors. This is in contrast with the two leading EV markets, China and Europe, where automakers have been mandated to use a harmonized charging standard for several years. In the United States, for Level 2 AC charging, Teslas use NACS and all other EV models have used a different plug called J1772. For DC fast charging, Teslas also use NACS, but most automakers have used a plug called the Combined Charging System (CCS) and some others have used a third plug type called CHAdeMO. This variety of charging connectors has meant that EV drivers seeking public charging need to check (1) if there are chargers along their route and (2) if those chargers are compatible with their vehicle. This won’t be the case for much longer.

The standardization of the J3400 NACS connector means that soon nearly all new EVs will be able to charge at nearly all charging stations. And for the millions of EVs already on U.S. roads, most non-Tesla EV drivers will soon gain access to Tesla’s NACS charging stations using an adapter. Uncertainty remains about how adapters will be rolled out to consumers, but automakers and charging providers will play a key role in helping consumers work through this and better understand their expanded charging options. For example, Ford recently announced that it will provide free charging adapters to its customers.

The industry shift to NACS comes with additional benefits. The NACS connector is more capable than the CCS connector because it allows higher amperages in both AC and DC operation, which translates to more potential power and less time spent at a charger. The NACS connector is also lighter and more ergonomic than other standards. Under a single standard, there won’t be any need to install charging stations with multiple connectors, and hardware costs will be less. In addition, NACS supports higher-voltage Level 2 charging that aligns with the voltage supply at many commercial locations. This means that chargers could be installed at locations that otherwise would require transformer upgrades, such as many mixed-use apartments and workplaces. Cheaper hardware and installation costs for charging projects could mean cheaper charging rates and even more savings for EV drivers.

Now let’s talk about helping to ensure that chargers function properly and that payment options are simple, accessible, and consistent across chargers in the United States. Communication errors between the EV and the charger and payment processing issues are common reasons why chargers malfunction. Standard communication protocols would go a long way toward improving reliability and optimizing payment. The communication protocols for the J3400 standard differ from Tesla’s legacy protocols and there is still work to be done by Tesla, other automakers, and charging manufacturers to ensure that all EVs and all chargers are interoperable. Fortunately, the federal National Electric Vehicle Infrastructure (NEVI) program, which is to provide funding for the installation of hundreds of thousands of chargers over the next several years, requires the implementation of the latest OCPP and OCPI standardized protocols for charger to network communication, as well as ISO 15118 for EV-to-charger communication. Together these standardized protocols will, among other things, reduce malfunctions by having all EVs and chargers speak the same “language”; expand error message reporting to allow for timely, precise, and lasting troubleshooting of faulty chargers; streamline payment processing and charger operation by allowing users to operate and pay for any charger from any company using a single app; and eventually allow for plug-and-charge capability for all chargers and EVs.

NEVI funding also comes with requirements that charging operators provide contactless payment options and guarantee that chargers are fully functional at least 97% of the time. On the latter, the federal government has already invested $150 million to repair and replace broken and faulty chargers across the United States. Because the NEVI program was developed prior to the J3400 NACS connector becoming a universal standard, the program does not require that NEVI-funded charging stations include the connector. However, since the industry has already largely agreed to adopt the standard, the federal government has expressed a willingness to update the program requirements and is likely to require the connector once SAE finalizes the standard by mid-2024.

As the NACS and NEVI roll out in tandem over the coming years, EV drivers in the United States will see both increased interoperability of charging stations and increased reliability. EV drivers and supporters have long sought to make EV charging away from home as simple and easy as filling up a gasoline car, and these developments are monumental steps toward making that a reality.

Last week, we released a wide-ranging analysis estimating that more than 150,000 jobs could be needed in the United States to deploy “behind-the-meter” charging infrastructure for electric light-duty vehicles (LDVs) and medium- and heavy-duty vehicles (MHDVs) through 2032. The term “behind the meter” refers to the customer’s side of the electricity meter and the term “front of the meter” is used when talking about the utility’s side, where there’s infrastructure such as substations, transformers, and feeder lines (Figure 1).

For HDVs specifically, the new study estimated that the Environmental Protection Agency’s (EPA) proposed HDV Phase 3 greenhouse gas (GHG) standard could generate as many as 16,000 jobs by 2032, or about 10% of the national total. But that’s only part of the jobs story.

As we’ll explore here, when all the jobs to construct the infrastructure to channel megawatt-scale power to chargers at private depots and public charging plazas for battery electric trucks and buses are considered, the utility-side infrastructure in front of the meter is likely to require a workforce an order of magnitude larger than the workforce building out customer-side infrastructure.

Figure 1. Battery-electric MHDV charging infrastructure ecosystem.

Let’s look at a preliminary, top-down jobs estimate based on available national-level data. It’s sensitive to assumptions about how individual chargers are configured into charging stations, how expensive utility grid upgrades are at each charging station, and how utility investments translate into jobs in the economy.

Still, we make generally conservative assumptions and the eventual number of jobs created could be larger. First, while the total number of chargers is based on a projected level of zero-emission vehicle adoption supported by the EPA HDV GHG Phase 3 proposal, in previous analysis we found that market forces, aided by Inflation Reduction Act (IRA) incentives, can support a larger number of zero-emission MHDVs and may draw even greater investments in charging infrastructure. Second, we do not fully account for possible infrastructure investments upstream from the distribution substation to support the largest multi-megawatt installations with peak loads greater than 10 MW.

We arrived at the job estimates in Figure 2 by first aggregating the nameplate capacity of 100 kW, 350 kW, and 1 MW chargers into a total number of hypothetical charging stations. The cost of grid upgrades and connection costs for charging stations were taken from previous ICCT research and utility upgrade cost estimates by the National Renewable Energy Laboratory (NREL). Next, we converted dollars invested in distribution grid capacity into a total number of direct and indirect jobs in the United States required and supported by these investments; this is based on an economic impact analysis of a utility’s substation transformer upgrade costs and other high-level utility infrastructure economic impact studies (here and here). Direct jobs are those related to the core construction and electrical work, for example installing substations and laying feeder lines; indirect jobs are upstream manufacturing, administrative, and other jobs not immediately involved in utility upgrade activities.

Under the most optimistic level of electrification likely to occur with the proposed EPA HDV Phase 3 GHG rule, we project more than 493,000 overnight 100 kW chargers, nearly 17,000 fast 350 kW chargers, and around 12,800 ultra-fast 1 MW chargers by 2032. We estimate up to $21 billion would need to be invested in distribution grid capacity to support these chargers, also by 2032.

These calculations, combined with the behind-the-meter jobs our colleagues estimated, suggest approximately 262,000 direct and indirect full-time equivalent jobs would be necessary to support the most optimistic rates of electrification to meet the EPA proposal by 2032 (Figure 2). More than 94% of these jobs come from what would be needed for utility-side infrastructure deployment. These front-of-the-meter jobs are wide-ranging and include substation construction, laying conduit, wiring, installing transformers and meters, laying feeder lines and their foundations, and manufacturing electrical grid components and assembly of these assets.

Figure 2. Estimated direct and indirect jobs created from infrastructure investments in MHDV electrification under the most optimistic rates of electrification to meet the EPA Phase 3 GHG proposal by 2032.

Billions of dollars in public investments are already funding charging infrastructure deployment at the federal and local levels. Private sector investments from companies such as TerraWatt Infrastructure, WattEV, Forum Mobility, and GreenLane reflect this growing industry.

Our estimates suggest the vast majority of charging infrastructure job creation will occur not in the manufacturing and installation of chargers themselves, but in the distribution grid assets that power the chargers. Finalizing the EPA Phase 3 proposal would generate significant momentum toward this job creation and the potential is even greater when accounting for the additional market potential shaped by IRA incentives. It’s key that utilities and regulators not only recognize the potential in constructing infrastructure assets in front of the meter, but that they begin planning to deliver front-of-the-meter assets and prepare their workforce in a time frame consistent with the EPA Phase 3 proposal and beyond.

Is demand for electric vehicles (EVs) slowing in the United States? The short answer is no. Light-duty EV sales data from the Alliance for Automotive Innovation shows continued and significant growth in the United States from 2021 through the third quarter of 2023. Figure 1 illustrates the increase in quarterly sales (bars, left axis) and EV sales shares (red line, right axis). EV sales increased from about 125,000 in Q1 2021 to 185,000 in Q4 2021 and from about 300,000 in Q1 2023 to 375,000 in Q3 2023. The year 2023 also marked the first time annual U.S. EV sales surpassed 1 million, and this was achieved by Q3; sales through the first three quarters of 2023 were about 58% higher than the same period in 2022.

Figure 1. U.S. light-duty electric vehicle sales and sales shares by quarter. Source: Alliance for Automotive Innovation, https://www.autosinnovate.org/EVDashboard.

Since Q3 2021, EV sales have increased every quarter, and the share of total light-duty vehicle sales that EVs represent isn’t shrinking, either. The share of new sales that are plug-in electric increased from about 3% in Q1 2021 to about 7% in 2022 and then reached more than 10% in Q3 2023. For some rough context, data from the U.S. Environmental Protection Agency’s Automotive Trends Report shows that EV sales shares have grown at a faster rate than sales shares of conventional hybrids that don’t have a plug: It took about 25 years for hybrids to reach a 10% market share, compared to about 12 years for EVs.

Additionally, state-level data shows that several states are far ahead of the national averages shown in Figure 1. California leads the country and EVs were nearly 27% of sales in the state through September 2023; this means that more than one in every four new light-duty vehicles sold were battery electric or plug-in hybrid electric. Another 12 states—Washington, Oregon, Colorado, Nevada, New Jersey, Massachusetts, Maryland, Hawaii, Connecticut, Virginia, Vermont, and Arizona—and the District of Columbia had EV sales shares between 10% and 20% through Q3 2023.

We also looked at the U.S. EV sales data by automaker and Figure 2 shows this data, with the companies stacked in order from highest (bottom) to lowest sales for the first three quarters of 2023. Most of these automakers sold more EVs in Q2 or Q3 2023 than in any other quarter in the chart, and each company except Ford sold more EVs in the first three quarters of 2023 than they did in all of 2022. Furthermore, each company shown sold more EVs in Q3 2023 than they did in Q3 2022. For example, third-quarter sales from 2022 to 2023 increased by 40%–60% for BMW, Tesla, and Volkswagen, about 115%–125% for Toyota and Stellantis, and by about 150%–180% for Hyundai and all others combined.

Figure 2. Quarterly U.S. light-duty electric vehicle sales by automaker. Source: Atlas EV hub, https://www.atlasevhub.com/materials/automakers-dashboard/.

This data echoes that collected by other researchers and several automakers. Indeed, BNEF found no signs of a global EV slowdown and said that such reports have been “greatly exaggerated.” Hyundai and Kia reported strong U.S. EV demand. Volvo’s CEO said there’s no slowdown of EV orders and he expects EVs to keep driving sales. Moreover, although Ford and General Motors are scaling back near-term production because of slowing demand relative to previous forecasts, both companies still plan on selling more EVs than ever before and “remain committed to an electric future.”

Beyond the strong sales, the latest consumer survey data by McKinsey and J.D. Power show that intent to purchase EVs is increasing, and a Consumer Reports survey found that 30% of licensed drivers would not even consider a gasoline vehicle for their next purchase or lease. The survey data also show that EV affordability and charging availability are key concerns. Fortunately, new tax credits from the Inflation Reduction Act of 2022 will provide up to $7,500 for new EVs plus several thousands of dollars for batteries. This combined with continued expected manufacturing cost reductions will help make more EV models cheaper than their gasoline counterparts, and there are dozens of new EV models across more vehicle classes and price points coming in 2024 and beyond. In terms of charging infrastructure, new public and private sector announcements sum up to more than $21 billion in investments and this is expected to increase the number of public chargers from about 160,000 in 2023 to nearly 1 million by 2030.

There’s a lot at stake in the transition, as EVs can substantially help reduce climate pollution, support clean air and public health, and bring economic benefits, jobs, and industrial competitiveness. Most automakers typically aren’t quick to vocalize slowing demand for their products, so it’s worth remembering, also, that any talk of a lack of EV demand in the United States coincides with a push to weaken proposed new federal pollution standards. The true story from the data is strongly positive for EVs. There’s never been a better time for new standards to build on the sales momentum detailed above and give a strong signal to automakers, charging infrastructure providers, consumers, and other stakeholders to invest in EVs with confidence.

Last week brought long-awaited tax-credit guidance about sustainable aviation fuels (SAFs) from the U.S. Treasury Department. It found that, as configured, the Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) model does not “satisfy the requirements to calculate the emissions reduction percentage” to determine which fuels qualify for the lucrative credit for SAFs in the Inflation Reduction Act (IRA). In the brief guidance, Treasury also tasked multiple agencies with collaborating on an update of GREET that would fit the requirements. While this interagency working group might seem like a nod to the agricultural industry and corn ethanol producers who have been pushing for use of this model, there’s still little clarity about how life-cycle greenhouse gas (GHG) emissions will ultimately be calculated for different SAFs.

GREET can be a useful analytical tool for evaluating the life-cycle emissions of a variety of different fuels on a consistent basis, but it’s always dependent on the quality of the assumptions and inputs. In past work, the ICCT explained how using GREET can allow users to incorporate a variety of optimistic external assumptions and inputs that have not undergone regulatory scrutiny. The model has many possible configurations and data sources, and its impact on the SAF tax credit will heavily depend on the three key data inputs and assumptions discussed below. All of these will be determined by the interagency working group that will finalize the version of GREET used for the tax credit:

1. The indirect land-use change emission factor used for crop-derived biofuels. Demand for biofuels can lead to cropland expansion, but the magnitude of the expansion and the associated emissions remain the subject of vigorous academic debate. Depending on how GREET is configured, the estimated indirect land-use change (ILUC) emissions for SAF’s can range from one-quarter to one-third of the values assessed by the U.S. Environmental Protection Agency (EPA) for the Renewable Fuel Standard, by California for its Low-Carbon Fuel Standard, and by the International Civil Aviation Organization (ICAO) for its Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA).

To qualify as a SAF under the IRA, a fuel’s life-cycle emissions must be below approximately 45 grams CO2e per MJ of fuel. The difference between assuming an ILUC emission factor of ~7 gCO2e/MJ and ~30 gCO2e/MJ for a feedstock like soy can make a big difference in the total emissions, all without the producer having demonstrated any improvements in their fuel-conversion process. (To view a range of possible values, see Figure 2 here.) A key outcome of the interagency working group process will be the determination of which emission factor will be used for feedstocks like corn and soy. Will it be a low estimate selected from the literature, an estimate consistent with the other regulatory assessments, or something in between?

2. The guidance around soil carbon modeling and climate-smart agricultural practices. Though carbon offsets and offset programs have recently taken somewhat of a beating in the public imagination, they’ve nevertheless attracted substantial interest from the Biden Administration, which has described activities like planting cover crops and reduced tillage of crops that have been shown to improve soils as “climate-smart” practices. However, the exact change in soil carbon that results from such practices is uncertain and difficult to credit, and a recent article in Science highlighted warnings from soil carbon modelers about the uncertainties and research gaps in their current work.

This is important because one module in the GREET model allows biofuel producers to use modeled soil carbon change estimates to credit individual biofuel projects. The size of these credits can be substantial and can allow producers to claim large emissions reductions. Rather than a conventional supply chain LCA, this module looks into the future to determine shifts in soil carbon content based on an assumed 30 years of consistent practices. Crediting these reductions would thus necessitate a new dimension to Treasury’s guidance, as Treasury would have to verify the shifts in soil carbon, ensure their permanence, and develop a system for clawing back tax credits if producers fail to keep up the promised practices for the full 30 years. Given that many existing carbon-offset schemes have recently been criticized for the lack of rigor of their soil carbon offsets, Treasury may opt to steer clear.

3. The guidelines for book-and-claim accounting for natural gas and electricity. There’s been a lot of recent focus on the “three pillars” of demonstrating renewable electricity use as it pertains to producing green hydrogen for the IRA’s 45V tax credit. Such focus is also relevant for aviation. What constitutes a “renewable” electron? Under “book-and-claim” accounting, a fuel producer can purchase the rights to renewable energy somewhere else in the economy and attribute it to their specific process. The three pillars help to create guardrails to ensure that those renewable attributes are (1) truly additional to the status quo; (2) not being double counted; and (3) are closely correlated with the energy demand for the fuel pathway. If Treasury determines that hydrogen producers must demonstrate the three pillars for the renewable electricity used to generate hydrogen, will it hold renewable inputs to SAF production to the same standard?

Depending on how flexible the guidelines are for SAF’s, producers may opt to meet their GHG reduction threshold outside of their immediate supply chain by purchasing the rights to renewable electricity or natural gas generated somewhere else. It’s even conceivable that with a particularly loose interpretation of book-and-claim without additionality safeguards, a SAF producer could purchase the rights to highly GHG negative “moo hydrogen” made from dairy manure as an input to their SAF pathway. Even if the additionality of that moo hydrogen was dubious (say, for example, the dairy biogas facility long predates the IRA), the carbon offsets for the avoided methane could be used to adjust the carbon intensity of SAF pathways looking to cross the 50% GHG reduction threshold.

As the above helps to illustrate, suggesting that GREET is a kind of definitive “method” of conducting an LCA is not much different from suggesting that Microsoft Excel is the most accurate method for conducting an LCA or that Microsoft Word is the best tool for writing a screenplay. Treasury’s recent guidance provides no answers about how the United States will ultimately handle these thorny-but-important questions. Answering them is not just a matter of collecting data and updating GREET, but also establishing the government’s tolerance for risk in assessing what constitutes a GHG reduction and what behavior justifies a tax credit. Until those questions are answered in March, we’re left with the status quo.

Earlier this year, the U.S. Environmental Protection Agency (EPA) proposed a third phase of greenhouse gas standards on heavy-duty vehicles and engines for model years 2027 and later, to accelerate road freight decarbonization. Many public comments supported it, but truck manufacturers including Volvo and Daimler have asked the EPA for a three-year delay of the rule. Their principal argument is that charging infrastructure will not be available to support the number of electric truck sales the rule would encourage.

Will infrastructure not be available in sufficient quantities? Well, we see that significant public investment has already been made, private investments have already led to groundbreaking on charging sites, ribbons have been cut at publicly accessible truck charging depots, and truck manufacturers themselves are building the infrastructure.

Beyond that, the reality is we don’t need to build everything everywhere, all at once. It makes strategic and economic sense in the near term to electrify the largest number of trucks along the smallest number of roadways where the business case is strongest (“no regrets” zones and corridors). And the assessment we present below shows that strategic infrastructure deployment in a limited number of freight hubs and corridors would be enough to ensure the EPA proposal can be met with sales of electric trucks.

We illustrate this with the infrastructure needs of long-haul trucks, just one of many vehicle categories covered in the EPA proposal. Here we define a long-haul truck as a vehicle that travels 500 miles daily and a long-haul corridor as one continuous segment at least 300 miles long.

Despite being less than 20% of the vehicles in the U.S. heavy-duty fleet, long-haul trucks are responsible for an outsized share of daily traffic volume (Table 1). Previous analysis demonstrated that battery-powered long-haul tractors offer the strongest business case when compared with other zero-emission alternatives. When coupled with megawatt charging in the second half of this decade, battery-powered long-haul tractors are estimated to be the only zero-emission powertrain with the potential to achieve a lower cost per mile than long-haul diesel tractors.

To assess the minimum infrastructure needs of these long-haul trucks in 2030, we: (1) examined freight traffic patterns, including our own national infrastructure analysis ; (2) revisited the infrastructure work of CALSTART and EPRI to inform our efforts and consider our analysis against their assumptions; and (3) consulted with industry experts to understand which long-haul corridors to prioritize. Our goal was to find the smallest number of roads with the highest traffic volume that could support 18 million long-haul electric truck miles (eVMTs) in 2030. These 18 million eVMTs are the electric truck activity spurred on by the Inflation Reduction Act, as estimated under our moderate scenario in this paper. That amount would also be enough to keep zero-emission trucks in commercial road freight aligned with international climate goals. We also considered a second scenario: the charging infrastructure for long-haul trucks needed to match the EPA Phase 3 proposal assuming manufacturers comply only with electric vehicle sales.

First, we find that public charging plazas along 1,800 miles of U.S. roads, identified as Tier 1 in Figure 1, would be enough to align long-haul truck electrification with international climate goals in 2030. The Tier 1 corridors are just 0.06% of paved road miles in the United States in 2020 or about 3% of the U.S. National Highway Freight Network. To arrive at 18 million daily eVMT on these corridors, we assume that one out of every four long-haul truck miles are electric. Forthcoming sales requirements for zero-emission trucks in California, Oregon, and Washington paired with total cost of ownership parity expected between battery-electric and diesel-powered long-haul tractors in Texas by 2027 put this within reach.

Figure 1. Tier 1, 2, and 3 priority corridors for electrifying long-haul truck activity in 2030 in line with international climate goals.

The Tier 2 corridors are where additional infrastructure would be needed in 2030 to achieve 18 million daily eVMT if instead only 15% of long-haul truck miles along Tier 1+2 corridors are electric. Tier 3 corridors expand the map and show where infrastructure would be needed if only 10% of long-haul truck miles along Tier 1+2+3 corridors are electric in 2030. Even the combined Tier 1, 2, and 3 corridors are still just 0.2% of paved road miles in the United States in 2020 or less than 10% of the U.S. National Highway Freight Network.

Second, the EPA proposal requires even less infrastructure than the first scenario because fewer electric trucks would be on U.S. roads. We project that the EPA proposal could generate close to 9 million long-haul eVMT per day, half as much as considered above. Approximately 1,000 total road miles across three corridors in California and Texas would be enough to comply with that; this assumes that 25% of long-haul truck miles on these roads are electric in 2030 and that manufacturers choose to comply only with electric truck sales, which the EPA rule would not require.

If, instead, only 10% of long-haul truck miles are electric, charging needs resulting from the EPA rule would require infrastructure deployment along 2,100 miles of the Interstate Highway System. This is shown in Figure 2 and the amount is still a fraction of a percent of U.S. paved roads and less than 4% of the national highway freight network.

Figure 2. Priority corridors for electrifying long-haul activity in line with maximum electrification required by the EPA Phase 3 proposed standard if 10% of long-haul truck miles are electric in 2030.

Despite manufacturer concerns, this analysis highlights the limited nature of the infrastructure required to meet the projected needs of long-haul electric trucks in 2030. Even an electrification scenario more ambitious than the EPA proposal and aligned with international climate goals would require public charging infrastructure for long-haul trucks across less than 1% of U.S. roads. Infrastructure at this scale would not be expected to be a major barrier to achieving greater greenhouse gas reductions, should the EPA choose to strengthen its proposal.

Deploying infrastructure in phases and starting strategically in the highest-priority locations would be enough for long-haul trucks in the near term. Our analysis shows that the infrastructure needs of the EPA proposal and of even more ambitious proposals can be met.