Understanding Factors Contributing to Electric Vehicle Ownership in California

The transition to plug-in electric vehicles (PEVs) in the United States is well underway, with PEVs, including both battery electric vehicles (BEVs) and plug-in hybrid electric vehicles (PHEVs), accounting for more than 8% of new vehicle sales in 2022. California leads the country in the transition, with nearly 1/3^rd of new vehicle sales being PEVs. While sales are rapidly increasing, a majority of Americans still rely on conventional passenger vehicles for their transport. Tailpipe emissions from conventional passenger vehicle use accounts for approximately 25% of California’s greenhouse gas (GHG) emissions (28% of the nation’s GHG emissions) and are a significant source of local air pollution. Health impacts from exposure to resulting pollution are not distributed equally and disproportionately impact low-income households and people of color. Electric vehicles are a promising solution to reducing both the health and climate impacts of passenger transport. California has ambitious climate mitigation targets, including a goal of economy-wide net zero GHG emissions by 2045. Accelerated adoption of PEVs across all segments of the population will be critical to reaching this goal. To achieve this, customized incentives may be needed to support adoption among heterogeneous segments of the population. An understanding of the demographic characteristics and differences between PEV-owning and non-PEV owning households will help facilitate the development of such incentives and policy interventions. Through a comparative analysis, we hope to identify barriers and drivers of PEV adoption that can help identify strategies to accelerate PEV uptake.

In this work, we analyze data from the 2019 California Vehicle Survey to understand how PEV-owning households differ from non-PEV households. We compare the household profiles of 489 PEV-owning households and 3647 non-PEV households and assess differences in household demographics, housing characteristics, and travel behaviors. We find substantial differences in housing characteristics, with 79% of PEV-owning households living in detached single-family homes as compared to 66% of non-PEV households. We also find significant demographic differences. The percent of Asians and Pacific Islanders among PEV-owning households is higher than for non-PEV households. PEV-owning households tend to be wealthier than non-PEV households. Approximately 2/3^rd of PEV-owning households have annual incomes greater than $100,000, compared to 39% of non-PEV households. We hypothesize that in addition to demographic factors, the level of vehicle ownership may also impact PEV adoption, as households with multiple vehicles may be more inclined to adopt a PEV as at least one of the vehicles in their fleet.  To assess the relationship between vehicle ownership and PEV adoption in a multivariate context, we propose to specify and estimate a joint model of vehicle ownership and PEV ownership. This will take the form of a recursive bivariate ordered probit in which the level of vehicle ownership and PEV ownership are both treated as ordered discrete endogenous variables, with explicit representation of the direct effect of vehicle ownership on PEV ownership as well as the accommodation of error correlations across the choice variables.  Model estimation results will be used to identify strategies for PEV uptake for different population segments.