Zero Emission Vehicles: Forecasting Fleet Scenarios and their Emissions Implications (2019)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

70 CHAPTER 5 CONCLUSIONS AND SUGGESTED RESEARCH The research presented in this report includes 1. Findings from a literature review on factors affecting adoption of ZEVs 2. Analysis scenarios that represent a broad range of potential future ZEV fleet changes given different future policy, technology, and infrastructure assumptions 3. Details about the assumptions made for MA3T model input parameters used in each analysis scenario 4. A summary of ZEV populations estimated by MA3T in simulations for each scenario, and 5. A summary of reductions in emissions estimated by MOVES2014b, corresponding to the ZEV population outputs from MA3T. Key factors affecting the adoption of ZEVs were identified in the literature review. Examples of these factors include vehicle purchase cost, availability of charging infrastructure, access to HOV lanes, and consumer awareness of ZEV technology and ZEV rebates or tax credits. The findings from the literature review were used to develop the set of scenarios of electric vehicle adoption rates (in 2040) that were modeled with MA3T and subsequently analyzed for corresponding emission reductions. The MA3T model was found to be adequate for modeling ZEV adoption because of its robustness in representing consumer behavior, its inclusion of many of the key factors affecting ZEV adoption, and its performance when comparing to other sets of ZEV adoption data. However, some of the model results were unexpected (e.g., increasing rebate amounts and home charging availability had little effect on ZEV adoption). Research into the assumptions used in MA3T for how consumers value incentives and other factors might explain some of the unexpected results. Because the emission reduction results presented in this report rely on outputs from the MA3T model, recommendations for further research focus on the model. The largest reductions in emissions modeled for the year 2040 were for the cost parity scenarios, in which (1) ZEV manufacturer costs reach parity with conventional vehicles as early as 2030, and (2) gasoline prices increase at accelerated rates between 2019 and 2050. Those reductions ranged from about 4-23% across the modeled pollutants (i.e., criteria pollutants, HCs, MSATs, and GHGs). Application of rebates to states that did not have rebates as of 2019 resulted in emissions reductions around 9%. Extending the duration of HOV lane access produced emissions reductions of about 3%. In general, the maximum modeled emissions reductions in 2040 achieved across the infrastructure and most Incentive/Policy scenarios were approximately 2-3% compared to the Base Case. These results can be used by state DOTs and MPOs to understand what types of policies and programs can benefit ZEV adoption and reduce light- duty vehicle exhaust emissions, and the magnitude of emission reductions that might be achieved. Ultimately, increasing ZEV adoption and associated reductions in emissions could lead to improvements in air quality. Recommendations for further research are summarized below: • The need for consumer and auto dealer education on purchasing ZEVs was identified in the literature review. Research into the influence of education programs would provide insight into programs and actions that fall well within many transportation agencies’ area of influence.

71 • Research into the long-term influence of range anxiety, price factors, home charge availability, and other factors will also provide insight into how consumers might respond to various policies and programs. • As ZEVs become a larger portion of the light-duty vehicle fleet, there is potential for new policies that could offset positive incentives. Policies such as state-level increases in ZEV registration fees as a result of losses in gas taxes could be considered in future developments of the MA3T model. • Because ZEV battery costs continue to decline at a rapid rate, research into ZEV manufacturer cost assumptions in MA3T are warranted. Modeling results in this study showed that cost parity of ZEVs with ICEVs had some of the largest impacts on ZEV adoption and emission reductions. The default vehicle manufacturer costs in MA3T show that manufacturer cost for ZEVs and their conventional vehicle counterparts differ more for larger vehicle classes (e.g., SUVs). Research into manufacturer cost differences by ZEV size classes could improve the assumptions in MA3T as more large-class ZEVs come into the market. Other BEV technologies, such as a 150-mile range BEV, could also be considered for inclusion in the model. • In general, further work is needed to improve understanding of what affects consumer behavior and adoption rates for new vehicle technology. One of the MA3T model developers indicated that future updates to the model will consider the influence of the “normalizing” of new electric vehicle technology. The modelers intend to investigate how consumer behavior might change once electric vehicles represent a larger portion of the market (e.g., 40% or more), representing an “inflection point” in adoption of ZEVs, and integrate that consumer behavior into the MA3T model. • More research is also needed to better understand producer behavior and production rates for new vehicle technology. If CAFE standards are not increased apace with increases in ZEV market share, vehicle manufacturers may choose to meet the average fuel economy requirement through a mix of ZEVs and increasingly less efficient ICEVs. This phenomenon, known as “leakage,” could negate the reduction in emissions associated with ZEVs, especially for CO2 emissions. • A complete analysis of well-to-wheel emissions associated with ZEVs in particular and ATVs in general would provide a more complete picture of expected emission reductions. It would also support transportation agency partners in understanding the relative importance of different actions to reduce emissions from the transportation sector. • MA3T considers only the U.S. household users of light-duty vehicles as the consumer market. However, as of 2017, medium- and heavy-duty vehicles contributed approximately 23 percent of all GHG emissions associated with the transportation sector (U.S. Environmental Protection Agency, 2019). More research is needed into factors that reduce emissions associated with freight. This includes policies targeting glider trucks (i.e., trucks with new bodies and old engines that can emit up to 40 times more emissions than new diesel engines), “electric roads” that provide electric power to vehicles via overhead catenary cables or conductor rails, truck stop electrification (i.e., providing heating/cooling and other services at truck stops without the need for trucks to idle), and factors responsible for growth and sales of newer battery and FCEV technology for freight.

72 • The MA3T model is calibrated to the AEO forecast. Other forecasts, such as the Navigant Consulting forecast or the Bloomberg New Energy Finance forecast, project substantially greater ATV adoption. This is an area for further investigation, as calibrating the model to other forecasts would provide alternate outcomes. • By default, MA3T V20190404 assumes that there are no FCEVs on the market, even by the end of the modeling time period (2050). Future calibration of the MA3T model would benefit from including data for FCEVs. Although the 2019 AEO forecasts that FCEVs will account for less than half a percent of the total light-duty vehicle fleet in 2040, it is highly uncertain what the FCEV technology and supporting infrastructure will be 20 or 30 years from now (2019). • The MA3T model uses nominal 2018 dollars for all costs, prices, and tax credit and rebate amounts. This could affect model outcomes. For example, ARRA tax credits could be issued beyond 2018 if a vehicle manufacturer has not yet reached the maximum number of subsidized vehicles after 2018. Adjustment of dollar amounts to account for inflation could be added to the model formulation in a future version. • Finally, the total emissions associated with electric vehicles are already strongly associated with the electricity source used for charging; for example, a BEV charged exclusively on electricity sourced from a coal-fired power plant would have equivalent wheel-to-well emissions of a 29- mpg car, while cleaner electricity sources would result in far greater equivalent emissions (Nealer et al., 2015). As electric vehicles become increasingly commonplace, the change in emissions associated with electric vehicles will increasingly be driven at the grid level. These impacts will likely overlap other developments in the transportation industry (e.g., the spread of autonomous vehicles), in utilities (e.g., distributed generation and storage, structural shifts in traditional IOUs as overall electricity consumption declines), and even in the supply chain of electric vehicle manufacturing (i.e., “cradle to grave” emissions generated from the sourcing of raw materials to the disposal of obsolete vehicles). Integrating these tailpipe emission models with scenarios that focus on those transportation network and grid level impacts would provide a more complete picture of future emissions.