Research Brief

In parts of California, air quality routinely violates federal and state standards. The Los Angeles area, which has the dirtiest air in the nation, must meet federal air quality standards by 2010. Doing so will require dramatic reductions in emissions even while population, driving, and industrial activity are increasing.

Studies have provided widely ranging estimates of how much California's strategy for meeting the standards will cost and how much it will actually reduce emissions; there is little commonly accepted information on which to base policy.

To provide a firmer empirical basis for decisionmaking, Institute for Civil Justice economists Lloyd Dixon and Steven Garber analyzed the costs and emission effects of California's strategy for reducing emissions from light-duty vehicles⁠—passenger cars and light-duty trucks. Light-duty vehicles account for about one-third of all pollutants that form ozone. (Ozone is created in the lower atmosphere when reactive organic gases (ROG) and oxides of nitrogen (NOx) react chemically in the presence of sunlight.) The most controversial component of California's strategy is the zero-emission vehicle (ZEV) mandate, which requires that manufacturers produce for sale emission-free vehicles.

The bottom line of their analysis:

  • Both keeping and eliminating the ZEV mandate pose major environmental and economic risks; California should seek alternatives that take into account the enormous uncertainty about what ZEVs can contribute to ozone reduction, when, and at what cost.
  • The recent revision of the mandate by the California Air Resources Board keeps the electric vehicle option open without pushing commercialization of these vehicles too hard and too soon.
  • California should not allow the ZEV debate to distract attention from programs designed to control emissions from gasoline vehicles. The design and implementation of these programs will make or break the state's ozone-control strategy over the next several years, and their performance will be critical in determining what the ultimate role of ZEVs should be.

Analyzing California's Light-Duty Vehicle Strategy

California's strategy for reducing emissions from light-duty vehicles has three prongs:

  1. improvements in next-generation gasoline vehicles—ones that have yet to be built,
  2. other policies designed to reduce emissions from existing and future gasoline vehicles, and
  3. the ZEV mandate, designed to displace gasoline vehicles with vehicles that have no direct emissions.

Almost all economic evaluations of California air pollution policies attempt to finesse the difficulties in estimating the benefits of cleaner air by comparing policies in terms of costs per ton of ROG and NOx emissions reduced. (The higher its ratio, the less attractive a policy appears to be.) Relying on estimates of social benefits of a ton of emission reduction or estimates of costs per ton of previously implemented policies, policymakers use rules of thumb such as: "If a policy appears to cost less than $10,000 per ton, implement the policy; if it costs much more than that, reject the policy."

But measures of cost effectiveness used in decisionmaking often ignore many costs and benefits—many of which depend on how individuals and the market will respond to policies—that should be factored into any policy decision. Dixon and Garber therefore refer to estimates of costs per ton as "narrow cost-effectiveness ratios" (NCERs) and supplement them with analysis of behavioral and market responses.

To gauge NCERs and estimate behavioral and market-level effects, the authors review and critique existing studies; draw on interviews with key stakeholders and analysts; determine sources of disagreement about key estimates; identify missing information; and develop models to estimate costs, emission reductions, and market responses. They also present a framework for integrating NCERs and market-level effects to guide decisionmaking.

Policies Focused on Next-Generation Gasoline Vehicles

The elements of California's strategy aimed at next-generation gasoline vehicles require changes in vehicle hardware, including tighter standards for tailpipe and evaporative fuel emissions, new standards for on-board monitoring of emission-control systems, and requirements for on-board recovery of gasoline vapors during refueling.

The authors conclude that the increased vehicle production costs attributable to the hardware-based elements will very likely be between $200 and $1,000 per vehicle.

The estimated NCERs for different hardware-based policies vary widely. For example, evaporative emission controls appear to involve NCERs of less than $3,000 per ton. Some of the various standards for exhaust emissions may have NCERs of less than $10,000, but the NCERs could turn out to be as high as $40,000 per ton. NCERs for on-board diagnostic systems—a promising technology that may already be improving the durability of emission control systems—are generally less than $15,000, but these estimates ignore critical behavioral effects such as how drivers and technicians will respond to the system. The federally mandated on-board vapor recovery regulations will probably provide little benefit in California, where most gasoline is dispensed through nozzles that prevent most refueling evaporation.

These NCERs ignore several policy-relevant factors, perhaps the most important being how much increases in vehicle prices stimulated by these policies will decrease sales of new cars and slow the rate at which older cars are replaced with newer, cleaner ones. The authors estimate potential price increases of $100 to $500 for new gasoline vehicles as a result of the new hardware regulations. If the price increases are near the high end of that range, then the new regulations could actually increase emissions for a few years and substantially attenuate direct emission benefits of the regulations for several more.

Policies Aimed at Existing and Future Gasoline Vehicles

The strategy elements aimed at both existing and future gasoline vehicles are Phase 2 reformulated gasoline, which produces fewer evaporative and exhaust emissions; Smog Check II, a tougher inspection and maintenance program; and the scrappage program, an unprecedented effort to buy and scrap 75,000 older vehicles a year in the greater Los Angeles area beginning in 1999.

The costs of Phase 2 reformulated gasoline are very likely to be between 7 and 19 cents per gallon. Use of this fuel may reduce emissions per mile driven by as much as 20 percent. NCERs for Phase 2 gasoline suggested by various studies range from $9,000 to $46,000 per ton, although the authors conclude that the latter figure is almost surely too high. Increased gasoline prices will reduce travel, thus further reducing emissions by up to 4 percent.

Among its other features, Smog Check II attempts to reduce fraud by smog check technicians and boosts from $50 to $450 the amount that drivers can be required to spend to repair emission controls. However, the authors conclude that Smog Check II may not work much better than the current inspection and maintenance program because (1) its fraud-reduction features may be ineffective and (2) it could exacerbate program evasion because vehicle owners can be forced to spend more on repairs. As a consequence, the authors are skeptical about the low NCERs for Smog Check II estimated in previous studies.

The costs of the scrappage program are likely to be $700 to $1,000 per vehicle scrapped. Resulting emission reductions cannot be projected with any confidence because they depend crucially on aspects of the program yet to be determined. NCERs for the scrappage program reported in other studies—$2,000 to $10,000 per ton—look favorable compared with common rules of thumb for assessing ozone control policies, but they do not account for widely recognized, but poorly understood, potential market and behavioral responses. The program will almost surely attract some older vehicles into the region to replace vehicles that are scrapped. In addition, offers to buy older, dirty cars could induce some owners to delay normal scrapping or emission repairs or even to tamper with their vehicles to make them dirtier. If the program is eviscerated by market or behavioral responses, or both, low NCERs will be little consolation.

Displacing Gasoline Vehicles with Zero-Emission Vehicles

Until its revision in March 1996, the third prong of the California strategy, the zero-emission vehicle mandate, required that 2 percent of all gasoline vehicles sold by the Big 7 automobile manufacturers (General Motors, Ford, Toyota, Chrysler, Honda, Nissan, and Mazda) from 1998 to 2000 be emission free. The quota was to rise to 5 percent in 2001-2002 and to 10 percent in 2003. Massachusetts and New York have adopted ZEV mandates identical to California's; taken together, the mandates would have required the Big 7 to produce 50,000 zero-emission vehicles in 1998. The study analyzes the original mandate—the mandate as it existed before its revision—but the issues addressed are central to any evaluation of the proper role of zero-emission vehicles in California ozone-reduction strategy.

For the next decade or more, the only commercially feasible emission-free vehicles are battery-powered electric vehicles (EVs). The primary challenges facing EVs are battery costs and energy storage capacity. The EVs that would have been marketed during the first years of the mandate can travel only 50 to 90 miles between charges—when not using power-hungry accessories such as heating and air conditioning. Advanced batteries that might double or triple that range are currently under development, but even under optimistic assumptions, they will not be ready for commercial-scale production until after the turn of the century.

Costs and Emission Benefits of EVs

Table 1. Estimates of EV Production and Operating Costs Compared to Gasoline Vehicles

Costs 1998–2002 Long Term
Incremental EV fixed production costs ($ billions) 1.0–4.2 Declining
Incremental EV variable production costs ($/vehicle) 3,300–15,000 0
Incremental EV operating costs ($/vehicle) 3,000–13,000 1,000–6,500

SOURCE: RAND Institute for Civil Justice analysis of California's clean air strategy, 1996.

The authors conclude that EVs will cost substantially more to produce and operate than comparable gasoline vehicles, particularly before 2003. Table 1 shows plausible cost ranges for some of the key factors.

In the short term, EVs will cost more to produce than comparable gasoline vehicles—perhaps as much as $15,000 per vehicle⁠—because they require special components with which manufacturers have little experience and because production runs will be small. The size of the EV cost disadvantage is subject to great controversy, but it seems likely that the disadvantage will disappear gradually over time.

The operating costs of EVs—the cost of all batteries, electricity, repair, and maintenance—could be much more than those for gasoline vehicles in the near term. Unlike production costs, operating costs are expected to remain above those of gasoline vehicles because even improved batteries will be expensive.

No one can reliably predict the direct emission reductions that would result from the original ZEV mandate. For example, they depend on how clean the gasoline vehicles are that EVs would displace. The authors estimate that each gasoline vehicle displaced by an EV could emit somewhere between 51 to 579 pounds of ROG and NOx over its lifetime.

How Cost Effective Do Electric Vehicles Look?

Figure 1. Narrow Cost-Effectiveness Ratios for Electric Vehicles Under Different Assumptions

SOURCE: RAND Institute for Civil Justice analysis of California's clean air strategy, 1996.

The authors calculate narrow cost-effectiveness ratios for EVs using various cost and emission-reduction estimates as well as varying assumptions about the rate at which operating and production costs will decline, the extent to which EVs penetrate the total vehicle fleet, and the time horizon over which costs and emission reductions are realized.

As Figure 1 shows, the NCER for electric vehicles could be very high or very low. In the best case for EVs—when initial costs are at the low end of what seems possible and decline quickly, and the gasoline vehicles being replaced by EVs are dirty—the NCER could be as low as $5,000 per ton, comfortably below the cutoffs used in standard rules of thumb. But when initial costs are high and decline slowly and the gasoline vehicles being replaced are relatively clean, costs per ton could be in the range of $850,000, unacceptably high by anyone's reckoning.

Market Responses to the ZEV Mandate

Figure 2. How Market Responses Affect Distribution of Costs and Emission Reductions

SOURCE: RAND Institute for Civil Justice analysis of California's clean air strategy, 1996.

The NCER for electric vehicles ignores market mediated effects of the ZEV mandate that could be crucial to wise policymaking. These effects will determine both the distribution of the mandate's costs across EV buyers and vehicle manufacturers and dealers, and, indirectly, the emission benefits of the mandate. Figure 2 suggests how such forces might play out.

The mandate specifies the quota of EVs to be sold as a percentage of gasoline vehicles sold. As a result, if EVs cost more to produce than they can be sold for, then selling additional gasoline vehicles involves the implicit cost of extra EV losses. This extra cost may lead to substantial increases in gasoline vehicle prices—as much as $550 per vehicle.

As in the case of price increases resulting from the hardware-based elements of the strategy, higher prices for gasoline vehicles indirectly affect emissions. If prices rise by several hundred dollars, the mandate could cause emissions to increase in the short term and attenuate emission reductions for EVs for several more years.

Making ZEV Policy to Accommodate Uncertainty

The ZEV mandate is a bold alternative to reliance on controlling emissions from gasoline vehicles. Consideration of such alternatives seems sensible because success of the first two prongs of California's strategy is hardly assured. But the economic effects of the ZEV mandate cannot be pinned down at all precisely—it could be a great success or a great failure.

This uncertainty does not mean that California should turn its back on ZEVs, however. Eliminating the mandate altogether could also lead to very bad outcomes. For example, if electric vehicle technology stagnates and other efforts to control emissions are disappointing, the state may face costly and unpopular measures, such as mandatory carpooling, to comply with federal clean-air standards.

Much of the EV debate is characterized by battles between interest groups, each confident of its own, very different estimates of EV costs and benefits. Dixon and Garber emphasize that no one can reliably predict the costs and emission effects of the ZEV mandate; they propose that ZEV policymaking accommodate, rather than deny, this uncertainty. They recommend searching for policies that

  • promote learning,
  • avoid the biggest risks, and
  • allow midcourse correction.

On March 29, 1996, the California Air Resources Board approved a revision of the ZEV mandate consistent with these principles.

By eliminating electric vehicle production quotas through 2002, the revision avoids the economic and environmental risks of keeping the mandate in its original form. At the same time, by requiring manufacturers to field demonstration fleets of thousands of electric vehicles with advanced batteries and to maintain support for battery research, the new policy continues to promote the development of electric vehicle technology and reduces the possibility that we will forsake a clean-air strategy that might eventually succeed.

The revision will also provide valuable information about the potential of electric vehicle technology and about how consumers use and value electric vehicles, and it gives the California Air Resources Board the flexibility to modify the electric vehicle program as new information becomes available.

In testimony before the Air Resources Board during its deliberations, the authors urged board members not to let the ZEV debate distract them from the programs to control gasoline vehicle emissions. These programs are crucial: They could produce very large emission reductions and their effectiveness is fundamental in determining the appropriate role for electric vehicles in the next century.

This report is part of the RAND research brief series. RAND research briefs present policy-oriented summaries of individual published, peer-reviewed documents or of a body of published work.

This document and trademark(s) contained herein are protected by law. This representation of RAND intellectual property is provided for noncommercial use only. Unauthorized posting of this publication online is prohibited; linking directly to this product page is encouraged. Permission is required from RAND to reproduce, or reuse in another form, any of its research documents for commercial purposes. For information on reprint and reuse permissions, please visit

RAND is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.