Electric vehicles (EVs) hold interest for utility companies around the world. According to John Dabels from General Motors Corp., air-pollution regulations in the United States could translate into production of 300 to 400 thousand EVs per year, once requirements in states like California mandate a 10-percent market penetration for zero-emission vehicles (ZEVs).
California utilities have been working vigorously on a combination of research and incentive programs to meet the EV production levels implied by ZEV targets. Utility proposals have included vehicle acquisition for company fleets, RD&D projects, residential infrastructure development, as well as battery incentives (See bibliography: Pacific Gas & Electric 1992; Southern California Edison 1993; San Diego Gas & Electric 1993).
The utilities conceived these proposals after the California Public Utilities Commission (CPUC) encouraged PG&E to use ratepayer funds to achieve "substantial market penetration of motor vehicles fueled by compressed natural gas" (Cross 1995). But the CPUC has moved to a more restricted approach with the passage of Assembly Bill 3229. Phillip Cross explains in the May 1, 1995, issue of Public Utilities Fortnightly that A.B. 3229 was introduced to define narrowly the interests of ratepayers and to specifically ban rate hikes as a method of funding programs to promote EVs or other low-emission vehicles (LEVs).
Regulatory policy has also seen changes at the California Air Resources Board (CARB). The CARB initially set ZEV goals to take effect in 1998. Two percent of the vehicles produced and delivered for sale in California by the seven largest auto manufacturers would fall in the ZEV category. That percentage would climb to 5 percent in 2001 and 10 percent by 2003. But the CARB has recently decided to suspend the percentage requirements for 1998 to 2002, and to ask for memoranda of agreement with each of the seven major automakers for the near term. The CARB retains the 10-percent ZEV production requirement in the 2003 model year and beyond (CARB 1996).
The CARB's recent decision eases the pressure on California utilities to work aggressively toward the ZEV targets for 1998. California utilities now have the chance to take a fresh look at how their role vis á vis EVs will complement actions taken by the government and auto manufacturers. Utilities in Eastern states will have a similar opportunity to rethink their policy toward EVs. (Several have proposed implementing the California air-pollution rules, including the ZEV requirement.)
This article summarizes findings from a major research project I helped conduct on EVs and their impact on Southern California Edison (SCE).1 The project began by assessing how SCE would change its use of existing generators, and whether it would have to expand its resource plan to serve the EV loads. The project concluded with an analysis of incentives to
promote the sale of EVs. The research did not focus on the "rush to EVs in 1998." Rather, the focus concerned the long-term impact of EVs on utilities, and the best combination of utility and state incentives to promote EV sales.
EV Scenarios for SCE
The project began by reporting the impact of EV loads on the SCE system for eight scenarios for the large-scale deployment of EVs over the next 20 years. The scenarios allowed a wide range of assumptions about the number of vehicles, batteries, the extent of daylight charging, and the control of charging during nighttime hours. The study results indicated that generators fired by natural gas could supply around 90 percent of the electricity needed to recharge the EVs. The analysis also revealed that SCE could accommodate a surprisingly large number of EVs within its existing resource plan.
For example, two million EVs could be accommodated through "smart control." The "smart control" scenario envisioned a two-way communication system to allow SCE to shape its nighttime EV charging loads to fit within its other electric loads. (Note, however, that two million EVs in the SCE service territory represents an extreme case, implying one EV out of every three vehicles in southern California.)
Figure 1 shows how the combined loads might appear on a summer day 20 years in the future (at the end of SCE's long-term planning period). Figure 2 shows a different load profile, assuming nighttime charging conducted at the customer's convenience. In this scenario, the EV loads appear in the evening hours and end up creating a second peak around 8 to 9 p.m. on a summer day far in the future. Analysis revealed that the loads in Figure 2 would prove more difficult to serve within the company's existing resource plan; additional generating units would be required. With only one million EVs in the SCE service area, the study showed that SCE could accommodate the extra loads even if nighttime charging was left to the customers' convenience.
Our study then estimated the increase in SCE's annual operating costs and total revenue requirements for each scenario. With two million EVs, for example, the cost of fuel and electricity purchases was expected to increase by around 21 percent, with total revenue requirements up by around 10 percent. We estimated that the two million EVs would cause electric sales to rise 13 percent by the end of the long-term planning period. These estimates imply that SCE could lower its average electric rate by around 3 percent. In scenarios with one million EVs, we estimated that the company could reduce its average electric rate by around 1.5 percent.
The rate-reduction estimates came in lower than one would have expected from previous studies. The main reason lay in the impact of EVs on marginal costs. High EV loads tend to flatten the daily load profile (as shown in Figure 1). When this happens, utilities should expect EVs to "push cheaper power plants off the margin." The rise in marginal costs boosts the cost of power purchases from independent power producers that sell electricity as qualifying cogeneration and small power production facilities (QFs). This cost increase proved quite important in the SCE scenarios, which envisioned a large number of power purchases from QFs.
EVs will likely affect the distribution system as well as generation. My research did not address distribution system impacts, but I can place the likely impacts of distribution costs into perspective. For example, one might expect extra distribution spending to amount to around $400 per vehicle on the utility side of the meter. Adding this level of spending to our rate scenarios would end up erasing about one-quarter of the estimated rate benefits we identified.
The Impact of Utility Incentives
The potential rate benefits noted above raise an important policy question: Should the utility contribute financial incentives to promote the sale of EVs?
Incentives have come under intensive study in the state of California, both at the CPUC and in the legislature. One important question asks whether a utility should offer significant incentives in an attempt to reduce the purchase price of an EV. The utility might then finance the incentive program from the 1.5- to 3-percent rate reductions expected from the scenario analysis. If this were possible, utilities could promote EV sales without harming their competitive position.
The rate impacts of utility incentives become less certain when one considers our limited knowledge of the market for cleaner vehicles and the complex impacts of EV loads on utility operations and ratemaking. Computer simulation modeling can be used to help sort out these complex interactions. In a 1994 article in Energy Policy (see bibliography), I explained the structure and assumptions of a computer model developed to simulate the impact of utility incentives to reduce the sale price of EVs. The model simulates the impacts of EVs on the utility system as well as the impact of utility incentives on EV sales. It tracks of EVs and conventional gasoline vehicles as well as other, cleaner-burning vehicles that will likely compete in the southern California market. Market shares reflect vehicle attributes and customer attitudes. (Customer attitudes toward price, range, and other attributes of the vehicles reflect statistical interpretation of a recent, stated-preference survey.) The model helped us calculate whether SCE could contribute a significant share of EV purchase-price incentives and then finance the program without having to raise the average electric rate.
We examined several different versions of purchase-price incentives. We considered variations in the size, timing, and duration of the incentive, and also varied the financing assumptions: 1) recover costs immediately, or 2) recover the costs over the operating life of the EVs. The results of the many simulations were surprisingly similar. All simulations indicated that utility incentives could boost EV sales somewhat, and that utilities would benefit from somewhat more efficient operation of their generating capacity.
Nevertheless, the simulations all suggested that SCE would end up having to raise the average electric rate to recover its program costs, whether the incentives were large or small, permanent or temporary, expensed or capitalized.
An Alternative View
For their part, utilities in California have argued that EV programs will not necessarily lead to higher rates. The utility analysis of rate impacts typically compares two scenarios. The first scenario calls for a combination of fleet purchases, information programs, vehicle testing, and battery incentives to promote sufficient EV sales to meet the ZEV goals. The second envisions a future in which EV sales will be negligible. The utilities argue that they could finance their EV programs through the improvement in system efficiencies made possible when EVs lead to flatter loads and have enough "room to spare" to reduce the average electric rate. The utilities assume that there will be essentially no flow of EVs into southern California without such a program (em a way to "prime the pump" (or "jump start" the EV industry). Once EVs are produced in significant number, the utilities can phase out their incentives; their program costs would go to zero. But program benefits in the form of increased electricity sales to EV owners would continue for the indefinite future. Under the "pump priming" metaphor, utility incentives to promote EV sales lead to rate benefits, not rate penalties.
Indeed, whether incentive programs penalize or reward ratepayers may be a matter of choosing the appropriate metaphor. The California utilities view their incentives as crucial to "prime the pump." My 1994 Energy Policy article posits that EVs will be sufficiently attractive to gain a small market share without utility price incentives. With utility price incentives, EVs will become more attractive, and their market share should increase. For example, a utility price incentive might boost EV market share (after 2000) from 4 to 6 percent of new car sales. Utility incentives would "turn the spigot" to increase the flow of EVs into southern California.
The market for alternative-fueled vehicles is so new that forecasting future vehicle sales is extremely uncertain. But of the two metaphors I find the "spigot" more descriptive, based on the earliest evidence gathered from stated-preference surveys. So I would argue that utility managers and regulators should expect to impose higher rates to recover program costs for EV incentives.
Regulators might justify such penalties to achieve public benefits from EVs. Penalties might be viewed as a tax on southern Californians who would benefit from cleaner air; the utility would serve as tax collector. But, is an electricity tax the best way to raise money for an EV incentive program?
A California state law, introduced as A.B. 3239 and enacted in September 1994, now prohibits this approach. Essentially, the law forbids utility rate increases designed to fund EV incentive programs unless the utility can show direct benefits specific to ratepayers in the form of safer or more reliable utility service (see bibliography, Cross 1995). Other states might resist this approach as well. After all, this tax raises the price of an energy form that is clean at the point of use. Would it not make more sense to impose the tax on gasoline or gasoline-powered vehicles, the major contributors to the smog in southern California?
Feebates — Carrot and Stick
In my opinion, the idea of "feebates" offers a better way to promote market penetration for cleaner vehicles. Under a feebate plan, California could impose a purchase-price fee on dirty vehicles and use the revenues to fund a rebate for EVs and other cleaner vehicles. The size of the fees (stick) and rebates (carrot) could be set with several goals in mind. For example, feebates could be selected to 1) reflect the relative emissions from the different vehicles, 2) to boost EV sales to a particular target, or 3) to ensure that revenues from fees would cover rebates paid to EV owners.
Feebates are attractive in theory, but it is important to look closely at the practical aspects of running a feebate program from one year to the next. In a 1995 article in System Dynamics Review, I looked at feebates through the eyes of a California state official who would face the challenge of running such a program. The article describes a model developed to simulate the impact of feebates on vehicle sales, on the utility system, and on the emission of air pollutants. The computer simulations showed that a simple feebate could put the state government in financial trouble. The simulations also showed that California would find it extremely difficult to dig itself out of the financial hole by adjusting feebates after problems arose.
Nevertheless, computer simulations indicate that these difficulties can be avoided if the state is willing to give the fund manager some latitude in running the program. It should be possible to keep a feebate program in the black if the state allows the fund manager to build a cushion in the early years of the program. The state must also allow the fund manager to alter the size of the fees or rebates on an annual basis. And finally, the state should not expect that the fund manager will necessarily keep the sum of the fee and the rebate fixed at an estimate of the environmental value of the EV. The most important conclusion from the feebate analysis is that it is possible to control a feebate system using readily available information. In other words, "crystal ball" forecasts of EV sales are not required for the state to run the program.
In sum, feebates have drawn praise by energy experts for their flexibility in a market marked by rapid technological change. Utility managers and regulators who are reluctant to see higher electric rates from utility incentives would do well to support feebates as an alternative approach to encourage the sale of cleaner vehicles. t
Andrew Ford is associate professor of environmental science and regional planning at Washington State University, and has worked extensively in simulation modeling on energy and environmental issues in the West. Dr. Ford initiated the research described here while at the Systems Management Department of the University of Southern California.
1. California Air Resources Board, Mobile Source Division (Staff Report), "Proposed Amendments to the ZEV Requirements for Passenger Cars and Light-Duty Trucks," Feb. 9, 1996.
2. Cross, Phillip S. "NGVs-Are Ratepayer Subsidies Appropriate?" Public Utilities Fortnightly, May 1, 1995, p. 42.
3. Dabels, John, General Motors Corp., "Environmental Requirements and the Impact Prototype Vehicle," International Conference on the Urban Electric Vehicle, Stockholm, May 1992.
4. DeCicco, John, "Feebates for Fuel Economy," report of the American Council for an Energy Efficient Economy, September 1992.
5. Evolution, Monthly Newsletter of the Electric Vehicle Association of the Americas, San Francisco.
6. Ford, Andrew, "The Impact of Electric Vehicles on the Southern California Edison System," report to the California Institute for Energy Efficiency, March 1992.
7. Ford, Andrew, "Electric Vehicles and the Electric Utility Company," Energy Policy, vol. 22, no. 7, 1994.
8. Ford, Andrew, "The Impacts of Large-Scale Use of Electric Vehicles in Southern California," Energy and Buildings, vol. 22, 1995.
9. Ford, Andrew, "Simulating the Controllability of Feebates," System Dynamics Review, vol. 11, no. 1, Spring 1995.
10. Amory Lovins and Hunter Lovins, "Reinventing the Wheels," The Atlantic Monthly, January 1995.
11. Pacific Gas & Elec. Co., testimony, I.91-10-029, R.91-10-028, CPUC, June 1992.
12. San Diego Gas & Elec. Co., "LEV Policy Overview," Application 93-11, CPUC, Nov. 1993.
13. So. Calif. Edison, "EV Program Overview," I.91-10-029, R.91-10-028, CPUC, Nov. 1993.
14. Sperling, Daniel, Future Drive: Electric Vehicles and Sustainable Transportation, Island Press, 1995.