America’s EV Adoption Hopes Hinge on Electric Trucks
- By Sam van de Schootbrugge
- 5 min read
Summary
- A new NBER working paper predicts US electric vehicle adoption for 2035 using current car buyer patterns, charging infrastructure and government subsidies.
- It finds significant demographic and geographic differences in adoption speed and that the electrification of light trucks is crucial for widespread EV acceptance in interior states.
- California and New Jersey should have the highest EV adoption by 2035, while Mississippi, Arkansas and Louisiana could lag significantly.
Introduction
The transition to electric transportation is central to decarbonisation efforts. The rate of electrification will hinge on three forces: intrinsic growth in demand for electric vehicles (EVs), production cost declines, and government stimulus of the industry. A new NBER working paper analyses the obstacles, opportunities and transition speed to widespread EV adoption. It finds:
- Income, age, education, vehicle type and preferences over environmental cleanliness are strongly correlated with EV demand.
- Intrinsic demand for EVs (demand not induced by government subsidies) is the most important determinant of EV adoption.
- Government subsidies effectively stimulate EV demand, but their effect pales before other factors that influence intrinsic demand as EV adoption increases.
Methodology and Data
The authors develop 20-year EV adoption forecasts that vary depending on potential policy stances, changing EV tastes, population growth, income and prices. The likely trajectory of adoption, and consequently the share of new EV car sales in the US by 2035, is designed as follows:
First, the authors analyse a nationwide micro-level dataset of car purchases to establish the cross-sectional patterns of light-duty car and truck preferences. The dataset contains 474,000 responses, broken down by zip code, to a 2017 and 2018 survey of US new vehicle purchasers conducted by MaritzCX. The purchasing habits and preferences elicited from the survey are used to forecast future EV adoption in the US.
Next, the authors project the continued rate of growth in intrinsic demand for EVs. This intrinsic demand, rather than subsidy-induced demand, is one of the biggest unknowns facing the vehicle industry, policymakers and researchers. It is influenced by widespread acceptance of EVs, the convenience of charging infrastructure, and the effect of EV quality improvements that will likely result from continued technological advancements in battery production. Based on the current growth rate of EV sales, the authors assume demand will initially grow at 30% YoY, then decline by 5% in a high-growth scenario, 10% in a medium-growth scenario, and 15% in a low-growth scenario per annum (Chart 1).
The most important driver of consumer choice is vehicle price. So, as a final step, the authors obtain price elasticities from the literature – the more elastic demand is, the greater the change in quantity demand when prices move. Future EV prices are driven by government subsidies or falling production costs and impacted by demographic attributes. Overall, they assume the sedan and light truck EV markets reach price parity with gasoline cars in 2030 and 2040, respectively. In the intervening years, the EV price premium relative to the compatible ICE vehicle falls at a constant rate. The decline in EV prices primarily reflects falling battery costs. They also consider the impact of an extension of the federal subsidy of $7,500 per vehicle.
Data on the subsidy amount, the qualification requirements, and eligible purchase dates for each program come primarily from the database of laws and incentives from the Alternative Fuels Data Center (AFDC) at the US Department of Energy. Demographic forecasts are from the Demographics Research Group at the University of Virginia’s Weldon Cooper Center.
Who Are the Early Adopters of EVs?
Consistent with past findings, the Maritz survey shows early adoption is strongly linked to income (Chart 2), which is why EV incentives are often means-tested. The authors also find a positive association with educational attainment and a link to ethnicity and age. That is, middle-aged (35-45 years old) Asian-American drivers are most likely to buy an EV.
The survey data also highlights a correlation between EV adoption and cross-state preferences in vehicle purchases. More liberal and coastal states tend to have higher demand for sedans and EVs which, to date, are primarily available as sedans. Yet car buyers in interior states have substantially higher demand for light trucks and lower demand for EVs.
These demographic patterns mirror stated environmental preferences – EV demand exhibits a strong, positive correlation with stated preference for environmental cleanliness, and regions with high light truck market share report low rates of concern about climate change.
Nationwide Results
The authors produce forecasts of EV adoption under a set of different scenarios that vary by future EV prices, the size of federal subsidies and the growth of intrinsic demand. In all scenarios, the rate of growth in intrinsic demand for EVs is the most important determinant of future EV adoption.
Also, assuming EV light trucks have emerged as viable substitutes in that segment, a high-intrinsic growth scenario leads to a nationwide EV market share of 47.6% in 2035 (Chart 3b). Absent government subsidies, this falls to 41.7% (Chart 3a). In the least-optimistic case (low growth, no subsidy), the share of EV sales would only be 10% by 2035. This is far lower than the International Energy Agency’s Sustainable Development Scenario target of 30%.
The results clearly show a government subsidy increases EV adoption, making it an effective tool for stimulating EV demand. However, compared with the relative impact of the intrinsic growth rate, the impact of subsidies is more modest. It is also costly; to achieve a 35% EV market share by 2035 in a medium-intrinsic growth environment, the cumulative subsidy bill would be $1.3tn. Also, there are diminishing returns, meaning the subsidy bill grows dramatically as the size of the EV market increases. The first $500bn in cumulative EV subsidies increases EV market share by 7-10pp. Each subsequent $500bn increases it by 5-8pp.
Worth noting is that, based on the current consumption patterns and car buyer preferences, the absence of an EV light truck alternative would lower EV adoption significantly. Since light trucks comprise over half the new cars sold in the US, the overall potential for EV light trucks is greater than that for EV sedans. A caveat, however, is that truck buyers generally have weaker preferences for EVs.
Regional Heterogeneity
The authors also find substantial geographic heterogeneity. In the medium-growth scenario, with government subsidies, EVs have a 2035 market share of 26% in California and New Jersey, and 14-16% in states like Mississippi, Arkansas and Louisiana (Chart 4). In a high-growth scenario, EV sales would account for 65% of total sales in the District of Colombia and 55% in California. For context, California has stated it intends to achieve 100% EV market share by 2035.
Bottom Line
Even with high organic growth, EVs are unlikely to dominate the car sales market by 2035. Crucially, using subsidies is a relatively costly tool for policymakers to achieve greater EV adoption, and achieving 35% market share could cost upwards of $1tn.
Instead, emphasis must be on automakers developing popular EV light trucks. In the US, these are popular vehicle types that could double the speed of EV adoption and help reduce large electrification heterogeneity across states. Policymakers should also pay attention to factors that affect intrinsic demand, like infrastructure. These tools are far more effective and cheaper than direct subsidies.
Citation
Archsmith, J. E., et al, (2021), Future Paths of Electric Vehicle Adoption in the United States: Predictable Determinants, Obstacles and Opportunities, NBER, (Working Paper 28933), https://www.nber.org/papers/w28933