Low-cost policy pathways for electric vehicle deployment

Electric vehicles (EVs) have multiple advantages, including reductions in global air pollution, local air pollution, and oil imports.[1] In particular, EVs are likely to play a key role in the decarbonisation of the transportation sector.[2] The transportation sector currently accounts for 15 per cent of global emissions worldwide,[3] and is much higher (e.g., 40 per cent in California[4]) in jurisdictions with high-levels of decarbonisation of the power sector.

Given this realization, many countries are pursuing aggressive EV targets — including Australia, Germany, India, and China — with developing countries in lockstep with developed countries, if not ahead. For example, China has set a target that EVs account for 25 per cent of new car sales by 2025.[5] India has set a similarly ambitious target: that 30 per cent of all vehicles are EVs by 2030.[6] Many of these targets are further complemented with supporting policies as well as subsidies.

The rationale for subsidies is that EVs are more expensive than comparable internal combustion engine vehicles (ICEVs), based on a comparison of their shelf price.[7] This is primarily due to the battery costs,[8] even though the rest of the vehicle may actually be cheaper.[9] However, given that there are many different kinds of EVs in the market (e.g., two-wheelers, three-wheelers and four-wheelers), a key question is how can limited government budgets be used to subsidise EVs in a cost-effective manner;[10] and, do all EVs need subsidisation in a uniform manner?[11]

To answer these questions, we need to first understand how EVs are more expensive than ICEVs — in upfront costs, or lifetime costs, which include both upfront and operating costs?[12]

In our previous work,[13] we have argued that appropriately discounted lifetime costs — in net present value (NPV) or total cost of ownership (TCO) terms — are the appropriate metric, given that they allow for an apples-to-apples comparison of costs between different options with different cost profiles over time. This is similar to the metric already used for comparing different sources for electricity generation; i.e., the levelized cost of electricity (LCOE).[14]

Now in case of lifetime cost parity in terms of TCOs, even if upfront costs are different due to cost of batteries, financial innovation using annualisation can be used to make EVs cost-competitive with ICEVs at all time periods.[15] Annualisation is a more general form of the popular car leases currently used for ICEVs,[16] to distribute the upfront cost into equal annual costs.[17] Annualisation essentially distributes all the upfront as well as operating (including fuel, maintenance, and insurance) costs into equal annual costs, and this financial innovation, which envisages a key role for electricity utilities and/or automakers given their unique strategic position in the EV value chain, may need no or very little subsidisation .[18]

Given the potential of annualisation, in particular, for vehicles with lifetime cost parity; the main trick to designing a low-cost policy pathway is to realise that, in absence of lifetime cost parity for some vehicles (e.g., some four-wheelers) today, there are always some other vehicles (e.g., two-wheelers) which already have lifetime cost parity. [19] This suggests that a low-cost policy pathway, focused on business model innovation, could be designed in phases. It could focus on deployment of the latter category in the first phase, given that this category does not require any explicit subsidies, and focus on getting annualisation to work via supporting business model innovation.[20]

A focus on deploying vehicle certain vehicle segments in Phase 1 would result in reductions in battery costs via local learning-by-doing, via battery chemistries which are shared across segments. Further, during the same time (i.e., in Phase 1), further battery cost reductions would be facilitated via global learning-by-doing as well as scale effects.[21] Thus, the low-cost policy pathway then is to focus on scaling the latter category (i.e., where we already have lifetime cost parity today) in Phase 1; until the reductions in battery costs renders lifetime cost parity for the former category (i.e., where we do not have lifetime cost parity today), which can then be targeted in Phase 2. This sequencing argument has already been presented for stationary battery storage,[22] and is equally applicable to EVs.

This low-cost policy pathway also indicates that, given any EV penetration target, the deployment pathways should be different for different vehicle segments. That is, based on our analysis,[23] the focus first needs to be on the diffusion of vehicles with lifetime cost parity — i.e. two-wheelers, three-wheelers, and shared four-wheelers (e.g., taxis and buses).[24] In fact, compared to a business-as-usual pathway for all vehicles, the low-cost policy pathway can save India a lot of the planned subsidies (a total of INR 100,000 million) for the FAME II scheme.[25]

[1] See https://www.ergon.com.au/network/smarter-energy/electric-vehicles/benefits-of-electric-vehicles

[2] See https://www.sciencedirect.com/science/article/pii/S0306261919307834

[3] See https://www.c2es.org/content/international-emissions/

[4] See https://www.biologicaldiversity.org/programs/climate_law_institute/transportation_and_global_warming/index.html

[5] See https://www.bloomberg.com/news/articles/2019-12-03/china-raises-2025-sales-target-for-electrified-cars-to-about-25

[6] See https://www.forbes.com/sites/scottcarpenter/2019/12/05/can-india-turn-nearly-200-million-vehicles-electric-in-six-years/#2e736b3e15db

[7] See https://www.manhattan-institute.org/html/short-circuit-high-cost-electric-vehicle-subsidies-11241.html

[8] See https://www.forbes.comsites/robday/2019/12/03/low-cost-batteries-are-about-to-transform-multiple-industries/#23365e0f1054

[9] See https://www.greencarreports.com/news/1126308_electric-car-battery-prices-dropped-13-in-2019-will-reach-100-kwh-in-2023

[10] See https://www.sciencedirect.com/science/article/pii/S0140988319303408

[11] See https://www.sciencedirect.com/science/article/pii/S2352146517304003

[12] See https://www.sciencedirect.com/science/article/pii/S0301421517301404

[13] See https://energy.stanford.edu/sites/g/files/sbiybj9971/f/using-finance-to-make-evs-cost-effective-june.pdf

[14] See https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf

[15] See https://energy.stanford.edu/sites/g/files/sbiybj9971/f/using-finance-to-make-evs-cost-effective-june.pdf

[16] Having said this, a lot more work may be required to make these loans work for EVs, given that a robust secondary market does not yet exist for either EVs or batteries.

[17] See https://www.goodfinancialcents.com/is-it-better-to-lease-or-buy-a-car/

[18] See https://energy.stanford.edu/sites/g/files/sbiybj9971/f/using-finance-to-make-evs-cost-effective-june.pdf

[19] See https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3618028

[20] See https://energy.stanford.edu/sites/g/files/sbiybj9971/f/using-finance-to-make-evs-cost-effective-june.pdf

[21] See https://environment.yale.edu/gillingham/BollingerGillingham_SolarLBD.pdf. The global cost reductions would be mostly on battery packs, whereas the local cost reductions would be mostly on battery balance of system (BOS).

[22] See https://www.sciencedirect.com/science/article/abs/pii/S0306261915007680

[23] See https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3618028

[24] However, if policymaker focus is on supporting all vehicles at the same time, our analysis supports the capital (or upfront) subsidy at the most cost-effective option; compared to other types of subsidies, such as operating (or per Km) and interest subsidies.

[25] See https://dhi.nic.in/writereaddata/UploadFile/publicationNotificationFAME%20II%208March2019.pdf