National Academies Press: OpenBook

Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards (2002)

Chapter: 4 Impact of a More Fuel-Efficient Fleet

« Previous: 3 Technologies for Improving the Fuel Economy of Passenger Cars and Light-Duty Trucks
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 63
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 64
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 65
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 66
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 67
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 68
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 69
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 70
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 71
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 72
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 73
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 74
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 75
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 76
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 77
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 78
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 79
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 80
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 81
Suggested Citation:"4 Impact of a More Fuel-Efficient Fleet." Transportation Research Board and National Research Council. 2002. Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards. Washington, DC: The National Academies Press. doi: 10.17226/10172.
×
Page 82

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.

63 4 Impact of a More Fuel-Efficient Fleet If the technologies described in Chapter 3 are imple- mented, making vehicles more fuel efficient, there will be a variety of impacts. This chapter explores the potential im- pacts on energy demand and greenhouse gas emissions, cost- efficient levels of fuel economy, industry and employment, and safety. ENERGY DEMAND AND GREENHOUSE GAS IMPACT The fuel economy of light-duty cars and trucks and ve- hicle miles traveled (VMT) are the two most important fac- tors underlying the use of energy and release of greenhouse gases in the light-duty fleet. Energy consumption during the manufacture of various components of the vehicles and their fuels is a lesser, but still significant, consideration in the life- cycle analyses of energy use and greenhouse gas emissions. Numerous projections were examined by the committee on possible future energy use and greenhouse gas emissions— for example, DeCicco and Gordon (1993), Austin et al. (1999), Charles River Associates (1995), EIA (2001), Patterson (1999), Greene and DeCicco (2000). To choose the best technology for overall energy efficiency, one must consider a “well-to-wheels” (WTW) analysis such as those that are described in Attachment 4A. The following discus- sion is designed to illustrate the impact of possible fuel economy changes and should not be interpreted in any way as a recommendation of the committee. The committee calculated the potential magnitude of fuel savings and greenhouse gas emission reductions if new passen- ger car and light-truck fuel economy (in mpg) is increased by 15 percent, 25 percent, 35 percent, and 45 percent. These in- creases are assumed to be phased in gradually beginning in 2004 and to reach their full value in 2013. New vehicle sales shares of passenger cars and light trucks were held constant. Green- house gas emissions are for the complete WTW cycle based on the GREET model and include carbon dioxide, methane, and nitrogen oxides (Wang, 1996; Wang and Huang, 1999). The fuel for all light-duty vehicles is assumed to be gasoline, 70 percent conventional and 30 percent reformulated. Allowance is made for a rebound effect—that is, a small increase in miles driven as fuel economy increases. Because of the long time required to turn over the fleet, the calcula- tions were extended to the year 2030 to show the longer- range impacts of the increases in fuel economy. However, new vehicle mpg was held constant after 2013. The base case approximates the 2001 Annual Energy Outlook forecast of the Energy Information Administration (EIA) except that the vehicle miles traveled are assumed to increase to 1.7 percent per year, slightly less than the 1.9 percent assumed by EIA. However, unlike the EIA forecast, fuel economy is assumed to remain constant. In the base case, annual gasoline use is projected to increase from 123 billion gallons (8 mmbd) in 2001 to 195 billion gallons (12.7 mmbd) by 2030 (Figure 4-1). The magnitude of gasoline savings that could be achieved relative to the base case for various fuel economy increases between now and 2030 is shown in Figure 4-2. As a result of 0 50000 100000 150000 200000 250000 2000 2005 2010 2015 2020 2025 2030 M ill io ns o f G al lo ns p er Y ea r Base Case 15% MPG Increase 25% MPG Increase 35% MPG Increase 45% MPG Increase FIGURE 4-1 Fuel use in alternative 2013 fuel economy scenarios.

64 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS these hypothetical fuel economy increases, fuel savings in 2015 range from 10 billion gallons for a 15 percent fuel economy increase to 25 billion gallons for a 45 percent in- crease. By 2030 the fuel savings would be 22 billion gallons and 55 billion gallons for fuel economy increases of 15 per- cent and 45 percent, respectively. Since greenhouse gas emissions are correlated closely with gasoline consumption (Figure 4-1), they show a similar pattern (Figure 4-3). Growth in greenhouse gas emissions still occurs through 2030, but the growth is slower than it would have been with- out improved fuel economy (Figure 4-4). ANALYSIS OF COST-EFFICIENT FUEL ECONOMY The committee takes no position on what the appropriate level of fuel economy should be. The question, however, is often raised of how much investment in new technology to increase fuel economy would be economically efficient. That is, when does the incremental cost of new technology begin to exceed the marginal savings in fuel costs? Consumers might not choose to use this technology for fuel economy; they might choose instead to enhance other aspects of the vehicles. Such an estimate, however, provides an objective measure of how much fuel economy could be increased while still decreasing consumers’ transportation costs. The committee calls this the cost-efficient level of fuel economy improvement, because it minimizes the sum of vehicle and fuel costs while holding other vehicle attributes constant. The committee identified what it calls cost-efficient tech- nology packages: combinations of existing and emerging technologies that would result in fuel economy improve- ments sufficient to cover the purchase price increases they would require, holding constant the size, weight, and perfor- mance characteristics of the vehicle(s). The essence of analyzing cost-efficient fuel economy is determining at what fuel economy level the marginal costs of additional fuel-saving technologies equal the marginal benefits to the consumer in fuel savings. However, such analysis is conditional on a number of critical assumptions, about which there may be legitimate differences of opinion, including (1) the costs and fuel-efficiency effect of new tech- nology and (2) various economic factors. The committee states its assumptions carefully and has investigated the ef- fect of varying several key parameters. Perhaps the most critical premise of this cost-efficient analysis is that key vehicle characteristics that affect fuel 0 10000 20000 30000 40000 50000 60000 70000 2000 2005 2010 2015 2020 2025 2030 M ill io ns o f G al lo ns p er Y ea r 45% MPG Increase 35% MPG Increase 25% MPG Increase 15% MPG Increase 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 2000 2005 2010 2015 2020 2025 2030 M ill io ns o f M et ric T on s pe r Y ea r Base Case 15% MPG Increase 25% MPG Increase 35% MPG Increase 45% MPG Increase FIGURE 4-2 Fuel savings of alternative 2013 fuel economy im- provement targets. 0 50 100 150 200 250 2000 2005 2010 2015 2020 2025 2030 M ill io n M e tr ic T o n s C E q u iv a le n t p e r Y e a r 45% MPG Increase 35% MPG Increase 25% MPG Increase 15% MPG Increase FIGURE 4-4 Greenhouse gas emissions reductions from hypo- thetical alternative fuel economy improvement targets. FIGURE 4-3 Fuel-cycle greenhouse gas emissions in alternative 2013 fuel economy cases.

IMPACT OF A MORE FUEL-EFFICIENT FLEET 65 economy are held constant: (1) acceleration performance, (2) size, in terms of functional capacity, (3) accessories and amenities, and (4) the mix of vehicle types, makes, and mod- els sold. These factors are held constant for analytical clarity and convenience. Once one begins altering vehicle charac- teristics, the trade-off between fuel economy and cost be- comes obscured in the myriad of other simultaneous changes of design and function that could be made. Since there is no obviously correct way to forecast future vehicle attributes and the sales mix, the committee holds them constant. In real-world markets, all of these factors will change over time and could have important implications for achieving a spe- cific fuel economy goal. Other key assumptions pertain to how consumers value fuel economy. The price for higher fuel economy technol- ogy is paid when a vehicle is purchased. Fuel savings accrue in the future, depending on how much a vehicle is driven and under what circumstances. Important uncertainties involve consumers’ expectations about future fuel prices, the rates of return they will expect on investments in higher fuel economy, and their perception of how used-vehicle markets will value fuel economy. Not only are the average values of these key parameters uncertain, but they will certainly vary from consumer to consumer and from one market segment to another. Incorporating all these uncertainties into the cost-efficient analysis would require a far more complex model and far more time to implement than the committee has available. Instead, the committee uses sensitivity analysis to illustrate the potential impacts of key assumptions on the outcomes of the cost-efficient analysis. Despite the many uncertainties, the cost-efficient analy- sis is valuable for illustrating the range of feasible fuel economy improvement and the general nature of the cost of achieving higher fuel economy. However, it is critically im- portant to keep in mind that the analysis is conditional on the assumptions of constant vehicle attributes and a number of key parameter values. Changing these assumptions would change the results, as the sensitivity analysis shows. The cost-efficient analysis uses estimates of the costs of technologies and their impacts on fuel consumption in Chap- ter 3 (see Tables 3-1 through 3-3). Only technologies known to be capable of meeting future emissions standards and hav- ing either positive or small negative effects on other vehicle attributes were used. The technologies were reordered by cost-effectiveness, so that the order of implementation re- flects increasing marginal cost per unit of fuel saved. Cost effectiveness is measured by the ratio of the midpoint fuel consumption reduction to the midpoint cost estimate. While this may not correspond to the actual order in which tech- nologies are implemented by auto manufacturers, it is none- theless the most appropriate assumption for analyzing the economic trade-off between vehicle price and fuel savings. The Path 3 technology scenario was used because all of the Path 3 technologies could be in full-scale production by 2013 to 2015. The data in Tables 3-1 through 3-3 were used to construct fuel economy supply functions and confidence bounds on those functions. The supply functions are com- bined with functions describing consumers’ willingness to pay for higher fuel economy and are solved for the point at which the cost of increasing fuel economy by 1 mile per gallon equals its value to the consumer. Three curves were constructed for each vehicle class to reflect the range of uncertainty shown in the tables in Chap- ter 3. The high cost/low fuel economy and low cost/high fuel economy curves provide a reasonable set of bounds for the average cost/average fuel economy curve. For the method used, see Greene (2001). Consumers’ willingness to pay is estimated using aver- age data on vehicle usage, expected payback times, gasoline at $1.50 per gallon, and assumed rates of return on the consumer’s investment in higher fuel economy. The key ar- eas of uncertainty are the rate of return consumers will de- mand and the length of time over which they will value future fuel savings. Because each additional mile per gal- lon saves less fuel than the one before, the marginal willing- ness to pay for fuel economy will decline as fuel economy increases. The cost-efficient point (at which the marginal value of fuel saved equals the marginal cost) is then found, assuming a fuel priced at $1.50 per gallon. This produces three fuel economy estimates for each vehicle class, reflecting the three curves. Finally, two cases are considered in which the key parameters of consumer discount rate and payback time are varied. The following calculations for cost-efficient fuel effi- ciency are based on the key assumptions summarized in Table 4-1 for passenger cars and light-duty trucks. Two cases TABLE 4-1 Key Assumptions of Cost-Efficient Analysis for New Car and Light Truck Fuel Economy Estimates Using Path 3 Technologies and Costs (see Chapter 3) Assumption Case 1 Case 2 First-year travel for new vehicles (mi/yr) 15,600 15,600 Rate of decrease in vehicle use (%/yr) 4.5 4.5 Payback time (yr) 14 3 Rate of return on investment (%) 12 0 Base fuel economy (mpg) Subcompact cars 31.3 31.3 Compact cars 30.1 30.1 Midsize cars 27.1 27.1 Large cars 24.8 24.8 Small SUVs 24.1 24.1 Midsize SUVs 21.0 21.0 Large SUVs 17.2 17.2 Minivans 23.0 23.0 Small pickups 23.2 23.2 Large pickups 18.5 18.5 On-road fuel economy (mpg) shortfall (%) 15 15 Effect of safety and emissions standards (%) –3.5 –3.5

66 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS are developed. In Case 1 it is assumed that vehicles are driven 15,600 miles in the first year, decreasing thereafter at 4.5 percent/year, and that a gallon of gasoline costs $1.50 (1999 dollars). Payback time is the vehicle lifetime of 14 years. Base fuel economy is shown for each class of vehicles. It is assumed that there will be a 15 percent reduction in on-the- road gasoline mileage from the EPA combined test data. It is also assumed that there will be a 3.5 percent fuel economy penalty as a result of weight gains associated with future safety and emissions requirements and that consumers re- quire a rate of return of 12 percent on the money spent on fuel economy. Because this last assumption is more subjec- tive than the others, Case 2 was developed using a payback of 3 years at zero percent discount rate. This case represents the perspective of car buyers who do not value fuel savings over a long time horizon. Case 1 might be the perspective of policy makers who believe that the national interest is served by reducing fuel consumption no matter how many people own a vehicle over its lifetime. The results of the cost-efficient analysis for Case 1 are shown in Table 4-2. Using the assumptions shown in Table 4-1, the calculation indicates that the cost-efficient increase in (average) fuel economy for automobiles could be in- creased by 12 percent for subcompacts and up to 27 percent for large passenger cars. For light-duty trucks, an increase of 25 to 42 percent (average) is calculated, with the larger in- creases for larger vehicles. For example, Table 4-2 shows that a new midsize SUV typically (sales-weighted average) has a base fuel economy today of 21.0 mpg. The adjusted base (20.3 mpg) reflects the 3.5 percent fuel economy penalty for weight increases to meet future safety and emission standards. In the column labeled “Average,” the cost-efficient fuel economy is 28.0 mpg. The percent improvement over the unadjusted base of 21.0 is 34 percent (shown in parentheses). The cost to obtain this improved fuel economy is estimated at $1,254. However, there is wide uncertainty in the results. This is illustrated in the other two columns, which show an optimistic case (low cost/high fuel economy curve) and a pessimistic case (high cost/low fuel economy curve). In the low cost/high fuel economy column, the cost-efficient fuel economy increases to 30.2 mpg, for an improvement of 44 percent at a cost of $1,248. In the high cost/low mpg column, the fuel economy is 25.8 mpg, for an improvement of 23 percent at a cost of $1,589. In some cases the high cost/low mpg column will have a lower cost (and a significantly lower mpg) than the low cost/high mpg column because of the nature of the cost- efficient calculation and the relative slopes of the cost curves. There is some evidence suggesting that consumers do not take a 14-year view of fuel economy when buying a new car. For that reason, Table 4-3 shows the cost-efficient fuel economy levels for 3-year payback periods. For cars, aver- age cost-efficient levels are between 0.1 and 1.5 mpg higher than their respective adjusted base fuel economy levels, with the larger increases for the larger cars. The cost-efficient levels of the light-duty trucks are about 1.4 to 3.1 mpg higher than their respective adjusted base fuel economy levels, with the larger increases being associated with the larger trucks. The negative changes in fuel economy shown in Table 4-3 are because the base is used for this calculation. All vehicles still improve relative to the adjusted base, even in the high cost/low mpg column. As shown in Table 4-2, for the 14-year payback (12 per- cent discount) case, the average cost-efficient fuel economy levels are between 3.8 and 6.6 mpg higher than their respec- tive (unadjusted) bases for passenger cars, with the larger increases associated with larger cars. For light-duty trucks, the cost-efficient levels are about 6 to 7 mpg higher than the base fuel economy levels, with the larger increases associ- ated with larger trucks. The cost-efficient fuel economy levels identified in Tables 4-2 and 4-3 are not recommended fuel economy goals. Rather, they are reflections of technological possibilities and eco- nomic realities. Other analysts could make other assumptions about parameter values and consumer behavior. Given the choice, consumers might well spend the money required to purchase the cost-efficient technology packages on other vehicle amenities, such as greater acceleration, accessories, or towing capacity. The fuel economy and cost data used in this study are compared with the data used in other recent studies in Fig- ures 4-5 and 4-6 for cars and light trucks, respectively. The cost curves used in this study are labeled NRC 2001 Mid (average), NRC 2001 Upper (high cost/low mpg upper bound), and NRC 2001 Lower (low cost/high mpg lower bound). For comparison with two other studies, one by Si- erra Research (Austin et al., 1999) and one by Energy and Environmental Analysis (EEA, 2001), the NRC curves were normalized to the sales-weighted average fuel economies of the new passenger car and light truck fleets. For passenger cars, the NRC average curve (NRC 2001 Mid) is similar to the Sierra curve (up to about 11 mpg increase) and slightly more optimistic than the EEA curve. The Massachusetts Institute of Technology (MIT) (Weiss et al., 2000) and American Council for an Energy-Efficient Economy (ACEEE) (DeCicco et al., 2001) curves were ob- tained differently as they are based on complete, specific vehicles embodying new technology. The NRC, Sierra, and EEA analyses added technology incrementally. Neither the MIT nor the ACEEE studies present their results in the form of cost curves; the curves shown here are the committee’s inferences based on data presented in those reports. The MIT curve is calculated using the lowest cost and most fuel-efficient vehicles in the study, which uses advanced technology and a midsize sedan and projects to 2020. Simi- larly, the ACEEE-Advanced curves in Figures 4-5 and 4-6 are based on individual vehicles and advanced technology options. Both studies are substantially more optimistic than this committee’s study, having used technology/cost options more advanced than those considered by the committee.

IMPACT OF A MORE FUEL-EFFICIENT FLEET 67 TABLE 4-2 Case 1: Cost-Efficient Fuel Economy (FE) Analysis for 14-Year Payback (12% Discount Rate)a Low Cost/High mpg Average High Cost/Low mpg Base Base FE Cost Savings FE Cost Savings FE Cost Savings Vehicle Class mpgb Adjustedc mpg, (%) ($) ($) mpg, (%) ($) ($) mpg, (%) ($) ($) Cars Subcompact 31.3 30.2 38.0 (21) 588 1,018 35.1 (12) 502 694 31.7 (1) 215 234 Compact 30.1 29.1 37.1 (23) 640 1,121 34.3 (14) 561 788 31.0 (3) 290 322 Midsize 27.1 26.2 35.4 (31) 854 1,499 32.6 (20) 791 1,140 29.5 (9) 554 651 Large 24.8 23.9 34.0 (37) 1,023 1,859 31.4 (27) 985 1,494 28.6 (15) 813 1,023 Light trucks Small SUVs 24.1 23.3 32.5 (35) 993 1,833 30.0 (25) 959 1,460 27.4 (14) 781 974 Mid SUVs 21.0 20.3 30.2 (44) 1,248 2,441 28.0 (34) 1,254 2,057 25.8 (23) 1,163 1,589 Large SUVs 17.2 16.6 25.7 (49) 1,578 3,198 24.5 (42) 1,629 2,910 23.2 (35) 1,643 2,589 Minivans 23.0 22.2 32.0 (39) 1,108 2,069 29.7 (29) 1,079 1,703 27.3 (19) 949 1,259 Small pickups 23.2 22.4 32.3 (39) 1,091 2,063 29.9 (29) 1,067 1,688 27.4 (18) 933 1,224 Large pickups 18.5 17.9 27.4 (48) 1,427 2,928 25.5 (38) 1,450 2,531 23.7 (28) 1,409 2,078 aOther key assumptions: See Table 4-1. bBase is before downward adjustment of –3.5 percent for future safety and emissions standards. cBase after adjustment for future safety and emissions standards (–3.5 percent). TABLE 4-3 Case 2: Cost-Efficient Fuel Economy (FE) Analysis for 3-Year Payback (Undiscounted)a Low Cost/High mpg Average High Cost/Low mpg Base Base FE Cost Savings FE Cost Savings FE Cost Savings Vehicle Class mpgb Adjustedc mpg, (%) ($) ($) mpg, (%) ($) ($) mpg, (%) ($) ($) Cars Subcompact 31.3 30.2 33.3 (6) 180 237 30.3 (–3) 11 11 30.2 (–4) 0 0 Compact 30.1 29.1 32.3 (7) 202 268 29.1 (–2) 29 29 29.1 (–4) 0 0 Midsize 27.1 26.2 29.8 (10) 278 363 26.8 (–1) 72 76 26.2 (–4) 0 0 Large 24.8 23.9 28.2 (14) 363 488 25.4 (3) 173 190 23.9 (–4) 0 0 Light trucks Small SUVs 24.1 23.3 27.3 (13) 358 492 24.7 (2) 174 193 23.3 (–4) 0 0 Mid SUVs 21.0 20.3 25.0 (19) 497 721 22.7 (8) 341 407 20.3 (–4) 0 0 Large SUVs 17.2 16.6 21.1 (23) 660 992 19.7 (15) 567 740 18.3 (6) 373 424 Minivans 23.0 22.2 26.5 (15) 411 570 24.2 (5) 247 284 22.2 (–4) 0 0 Small pickups 23.2 22.4 26.9 (16) 412 579 24.4 (5) 247 285 22.4 (–4) 0 0 Large pickups 18.5 17.9 22.7 (23) 600 918 20.8 (12) 477 608 18.7 (1) 178 189 aOther key assumptions: See Table 4-1. bBase is before downward adjustment of –3.5 percent for future safety and emissions standards. cBase after adjustment for future safety and emissions standards (–3.5 percent). In Figure 4-6, the committee’s curve (NRC) is more opti- mistic than the Sierra curve and similar to the EEA curve. The ACEEE-Advanced curve, which is based on vehicles as discussed for cars above, is much more optimistic. It uses technology/cost options beyond those used for the com- mittee’s cost-efficient optimization. POTENTIAL IMPACTS ON THE DOMESTIC AUTOMOBILE INDUSTRY Regulations to increase the fuel economy of vehicles will require investments by automakers in R&D and tooling and will thus increase the costs of new vehicles. They will also

68 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS divert resources that would otherwise go toward satisfying consumers’ demands for performance, styling, and other vehicle features. No regulation is without cost to consumers or manufacturers. The impacts of CAFE standards in the past are reviewed in Chapter 2. Looking Forward at the Automobile Market As noted in Chapter 2, GM, Ford, and the Chrysler divi- sion of DaimlerChrysler will post lower profits in 2001 ow- ing to the slower economy and the sharp rise in buyer incen- tives that the companies need to offer to maintain their market shares. Vehicle demand is expected to decline in 2001 from 17.4 million in 2000, with the final number depending in large part on the level of rebates and other incentives. Foreign companies’ share of the U.S. market has grown steadily and is now about 36 percent, compared with 26 per- cent in 1993. In 2000, GM, Ford, and Chrysler combined lost 3 percentage points in market share, and so far in 2001 have lost another 1.4 percentage points (despite per-unit in- centives that are the highest in history and often triple the marketing support on competing foreign models). The gain in market share of foreign companies has accelerated in re- cent years as foreign manufacturers entered the light truck market with models that competed with traditional Ameri- can pickups, minivans, and SUVs. They also created a new category, the crossover vehicle, which is built on a car chas- sis but looks like an SUV and may be classed as a light truck for fuel economy regulation. (Examples include the Toyota RAV-4 and Honda CRV.) The erosion in profit margins and income at GM, Ford, and Chrysler began in 2000 as pricing pressures, once con- fined to passenger cars, spread into the light-truck sector and sales slowed. The supply of light trucks, until recently bal- anced against sharply rising demand, had allowed vehicle manufacturers to aggressively price these products. Although the Big Three still dominate the light-truck market with a 77 percent share, that is down from 86 percent in 1993 as a result of new products and capacity from Asian and Euro- pean manufacturers. Additional North American truck ca- pacity from Honda in Alabama and Toyota in Indiana and expansion of capacity by BMW and Nissan will add at least 750,000 units of new truck supply to the market over the next 3 years. Given the recent success of foreign manufac- turers with new models and advanced technology (see Chap- ter 3), their share of the truck segment is likely to rise further in the coming years. To cope with falling profits and lower cash balances, auto companies have cut back on discretionary spending, eliminated jobs through voluntary retirement plans, and now appear to be delaying product launch schedules. Standard & Poor’s, the ar- biter of creditworthiness, recently lowered its outlook on GM and Ford from stable to negative (Butters, 2001). Even if vehicle demand rebounds in 2002 to the 17 mil- lion-unit level, industry profits are not expected to recover $0 $1,000 $2,000 $3,000 $4,000 $5,000 0 5 10 15 20 25 Increase in MPG R et ai l P ric e In cr ea se 2014 Sierra Res. ACEEE-Advanced 2013-15 EEA NRC 2001 Mid NRC 2001 Upper NRC 2001 Lower EEA w/o WGT FIGURE 4-6 Light-truck fuel economy cost curves from selected studies. $0 $1,000 $2,000 $3,000 $4,000 $5,000 0 5 10 15 20 25 Increase in MPG R et ai l P ric e In cr ea se 2014 Sierra Res. 2020 MIT ACEEE-Advanced 2013-15 EEA NRC 2001 Mid NRC 2001 Lower NRC 2001 Upper EEA w/o wgt FIGURE 4-5 Passenger car fuel economy cost curves from selected studies.

IMPACT OF A MORE FUEL-EFFICIENT FLEET 69 soon to historic levels. Increased competition could reduce truck profits to half of their former levels. If the proliferation of foreign brand crossover models draws buyers away from larger, heavier, and more expensive domestic models, the financial impact on the industry will be greater. This is a difficult environment for GM, Ford, and the Chrysler division of DaimlerChrysler. In less than 3 years, Chrysler went from being the most profitable vehicle manu- facturer in the country to a merger with Daimler that was forced by financial distress. While GM and Ford are in rea- sonable financial health, they cannot count on truck profits to generate above-average returns as they did in the past. Nevertheless, the industry could adjust to possible changes in CAFE standards if they were undertaken over a long period of time, consistent with normal product life cycles. An abrupt increase in fuel economy standards (espe- cially one that hurt the industry’s ability to sell light trucks) would be more costly. A single standard that did not differ- entiate between cars and trucks would be particularly diffi- cult to accommodate. Criteria for Judging Regulatory Changes The impacts on industry of changing fuel economy stan- dards would depend on how they were applied. Some of the more important criteria are discussed here. Timing and Scale of Increase Raising standards too steeply over too short a time would require manufacturers to absorb much of the cost of obtain- ing or developing technology and tooling up to produce more efficient vehicles. It would lead to sharp increases in costs to manufacturers (and thus consumers). The benefits of pre- cipitous increases in CAFE standards would flow instead to machine tool companies, component manufacturers, and de- velopers of vehicle technology. Automakers would be forced to divert funds and talent away from longer-term investments (such as the PNGV). Given sufficient planning time, how- ever, industry can adapt. Chapter 3 discusses the costs and timing involved in introducing fuel-saving technology. Gen- erally, little change can be expected over the next few years, and major changes would require a decade. Equivalence of Impact If new regulations favor one class of manufacturer over another, they will distribute the costs unevenly and could evoke unintended responses. In general, new regulations should distribute the burden equally among manufacturers unless there is a good reason not to. For example, raising the standard for light trucks to that of cars would be more costly for light-truck manufacturers. On the other hand, tightening the standards for passenger vehicles while leaving light trucks alone would favor another set of manufacturers. A current proposal to simply increase the standard for light- duty trucks to the current level of the standard for passenger cars would operate in this inequitable manner. The rise of the crossover SUV will add to the challenge of finding a balanced approach. Flexibility In general, regulations that allow manufacturers flexibil- ity in choosing how to achieve the desired policy goal (such as reducing fuel use or improving safety) are likely to lower the costs to the nation. That is because restricting the avail- able technology options will reduce chances for cost-saving innovations. To the extent possible, consistent with the over- all policy goals, flexibility is an important criterion. This is one reason for the committee’s enthusiasm for tradeable fuel economy credits, as described in this chapter. Hidden Costs of Forcing Technical Innovation In general, it is risky to commercialize technologies while they are still advancing rapidly. Pioneering purchasers of vehicles that incorporate highly efficient new technology could find that these vehicles depreciate faster than those with old technology if the new technology results in higher repair costs or is replaced by improved versions. For the same reason, leasors might be reluctant to write leases on vehicles that are radically different and therefore have un- predictable future demand. GM offered its own leases on the EV1 (an electric vehicle) but did not sell the vehicle because it recognized that the car might be unusable in a few years because of technological obsolescence or high maintenance costs. On the other hand, as explained in Chapter 3, foreign manufacturers are rapidly improving their technology, largely because their main markets are in countries with high fuel prices or high fuel economy standards. Not only are their vehicles economical (and frequently low in emissions), but they are proving popular because they offer other at- tributes valued by consumers, such as power and improved driving characteristics. Insofar as higher fuel economy stan- dards force domestic manufacturers to adopt new technol- ogy, it could improve their competitiveness. SAFETY IMPLICATIONS OF FUTURE INCREASES IN FUEL ECONOMY In Chapter 2 the committee noted that the fuel economy improvement that occurred during the 1970s and early 1980s involved considerable downweighting and downsizing of the vehicle fleet. Although many general indicators of motor vehicle travel safety improved during that period (e.g., the fatality rate per vehicle mile traveled), the preponderance of evidence indicates that this downsizing of the vehicle fleet resulted in a hidden safety cost, namely, travel safety would

70 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS have improved even more had vehicles not been downsized. Based on the most comprehensive and thorough analyses currently available, it was estimated in Chapter 2 that there would have been between 1,300 and 2,600 fewer crash deaths in 1993 had the average weight and size of the light- duty motor vehicle fleet in that year been like that of the mid-1970s. Similarly, it was estimated there would have been 13,000 to 26,000 fewer moderate to critical injuries. These are deaths and injuries that would have been prevented in larger, heavier vehicles, given the improvements in ve- hicle occupant protection and the travel environment that occurred during the intervening years. In other words, these deaths and injuries were one of the painful trade-offs that resulted from downweighting and downsizing and the re- sultant improved fuel economy. This section of Chapter 4 addresses the question of how safety might be affected by future improvements in fuel economy. The key issue is the extent to which such improve- ments would involve the kind of vehicle downweighting and downsizing that occurred in the 1970s and 1980s. In Chap- ter 3 the committee examined the methods by which the in- dustry can improve fuel economy in the future and identified many potential improvements in powertrains, aerodynam- ics, and vehicle accessories that could be used to increase fuel economy. Earlier in Chapter 4 the committee concluded that many of these technologies could pay for themselves in reduced fuel costs during the lifetimes of vehicles. Thus, it is technically feasible and potentially economical to improve fuel economy without reducing vehicle weight or size and, therefore, without significantly affecting the safety of motor vehicle travel. Two members of the committee believe that it may be possible to improve fuel economy without any implication for safety, even if downweighting is used. Their dissent forms Appendix A of this report. The actual strategies chosen by manufacturers to improve fuel economy will depend on a variety of factors. Even if the technology included in the cost-efficient fuel economy im- provement analyses is adopted, that technology might be used to provide customers with other vehicle attributes (per- formance, size, towing capacity) that they may value as much as or more than fuel economy. While it is clear vehicle weight reduction is not necessary for increasing fuel economy, it would be shortsighted to ignore the possibility that it might be part of the response to increases in CAFE standards. In fact, in meetings with members of this committee, au- tomotive manufacturers stated that significant increases in fuel economy requirements under the current CAFE system would be met, at least in part, by vehicle weight reduction. Because many automakers have already emphasized weight reduction with their current vehicle models, they also stated that substantial reductions in weight probably could not oc- cur without some reduction in vehicle size. They were refer- ring to exterior dimensions, not to interior space, which is a high customer priority. However, this type of size reduction also reduces those portions of the vehicle that provide the protective crush zones required to effectively manage crash energy. When asked about the potential use of lighter materials to allow weight reduction without safety-related size reduc- tions, the manufacturers acknowledged that they were gain- ing more experience with new materials but expressed con- cern about their higher costs. Given that concern, industry representatives did not expect that they could avoid reducing vehicle size if substantial reductions in vehicle weight were made. Thus, based on what the committee was told in direct response to its questions, significant increases in fuel economy requirements under the current CAFE system could be accompanied by reductions in vehicle weight, and at some level, in vehicle size. The committee recognizes that automakers’ responses could be biased in this regard, but the extensive down- weighting and downsizing that occurred after fuel economy requirements were established in the 1970s (see Chapter 2) suggest that the likelihood of a similar response to further increases in fuel economy requirements must be considered seriously. Any reduction in vehicle size and weight would have safety implications. As explained in Chapter 2, there is uncertainty in quantifying these implications. In addition, the societal effects of downsizing and downweighting de- pend on which segments of the fleet are affected. For ex- ample, if future weight reductions occur in only the heaviest of the light-duty vehicles, that can produce overall improve- ments in vehicle safety. The following sections of the report describe the committee’s findings on vehicle weight, size, and safety in greater detail and set forth some general con- cerns and recommendations with regard to future efforts to improve the fuel economy of the passenger vehicle fleet. The Role of Vehicle Mass The 1992 NRC fuel economy report concluded as fol- lows: “Although the data and analyses are not definitive, the Committee believes that there is likely to be a safety cost if downweighting is used to improve fuel economy (all else being equal)” (NRC, 1992, p. 6). Studies continue to accu- mulate indicating that mass is a critical factor in the injury outcomes of motor vehicle crashes (e.g., Evans and Frick, 1992, 1993; Wood, 1997; Evans, 2001). Although there are arguments that not all increases in vehicle weight benefit safety, these arguments do not contradict the general finding that, all other things being equal, more mass is protective. For example, Joksch et al. (1998) have reported that, when vehicles of similar size are compared, those that are signifi- cantly heavier than the average for their size do not appear to improve their occupants’ protection but do increase the risk to occupants of other vehicles with which they collide. The authors note (Joksch et al., 1998, p. ES-2) that this effect should be interpreted with caution, because it is likely that overweight vehicles are overweight in part because of more

IMPACT OF A MORE FUEL-EFFICIENT FLEET 71 powerful (and heavier) engines and performance packages, which could attract more aggressive drivers. Thus, Joksch et al. demonstrate that some weight increases can be detrimen- tal to safety, but this finding does not change the basic rela- tionship: among vehicles of equal size and similar driving exposure, the heavier vehicle would be expected to provide greater occupant protection. The protective benefits of mass are clearly understood in multiple-vehicle crashes, where the physical conservation of momentum results in the heavier vehicle’s experiencing a smaller change in momentum, and hence lower occupant deceleration, than the lighter vehicle. However, mass is also protective in single-vehicle crashes with objects (such as trees, poles, or guard rails), because many of these objects will move or deform in proportion to the mass of the vehicle. In this case, the change in velocity is not affected, but the deceleration of the vehicle and its occupants decreases as the object that is struck deforms or moves. Figures 4-7 and 4-8, which show fatality rates per million registered vehicles by vehicle type and vehicle weight, illustrate the protective ef- fects of vehicle mass in single- and multiple-vehicle crashes. The heaviest vehicles in each class (cars, SUVs, and pick- ups) have about half as many fatalities per registered vehicle as the lightest vehicles.1 While the benefits of mass for self-protection are clear, mass can also impose a safety cost on other road users. In a collision between two vehicles, increasing the mass of one of the vehicles will decrease its momentum change (and the forces on its occupants) but increase the momentum change of the crash partner. Figure 4-9 shows the increased fatalities caused in other vehicles per million registered cars, SUVs, or pickups as the mass of the “striking” vehicle increases. Because of this tendency to cause more injuries to occupants of other vehicles, heavier vehicles are sometimes said to be more “aggressive.” Aggressivity also varies by vehicle type; SUVs and pickups cause more deaths in other vehicles than do passenger cars. There are also effects of vehicle type on the likelihood of injuring pedestrians and other vulnerable road users, although the effects of mass are much weaker there (see Figure 4-10). The smaller effect of mass for the more vulnerable nonoccupants probably reflects the fact that mass ratios between, for example, pedestrians and the light- est vehicle are already so high that increasing mass of the vehicle makes little additional difference in survivability for the pedestrian. The differences by vehicle type may reflect the greater propensity of taller vehicles to do more damage in collisions with nonoccupants. The net societal safety impact of a change in the average mass of the light-duty vehicle fleet can be an increase, a decrease, or no change at all. The outcome depends on how that change in mass is distributed among the vehicles that 1Because mass and size are highly correlated, some of the relationships illustrated in these figures are also attributable to differences in vehicle size, an issue that will be discussed further later in this chapter. These figures also show that mass is not the only vehicle characteristic affecting occupant injury risk, as there are substantial differences in occupant fatality risk for cars, SUVs, and pickups of similar weight. Much of this difference prob- ably results from the higher ride heights of SUVs and pickups; riding higher is protective in multiple-vehicle crashes, because the higher vehicle tends to override lower vehicles, but increases single-vehicle crash fatalities by rais- ing the vehicle’s center of gravity and increasing rollover risk. 20 80 14 0 20 0 <2500 2500- 3000- 3500- 4000- 4500- 5000+ vehicle weight (lb.) de at hs /m ill io n ve hi cl es /y r cars/passenger vans utility vehicles pickups 0 20 40 60 80 100 120 <2500 2500- 3000- 3500- 4000- 4500- 5000+ vehicle weight (lb.) de at hs /m ill io n ve hi cl es /y r cars/passenger vans utility vehicles pickups FIGURE 4-7 Occupant death rates in single-vehicle crashes for 1990–1996 model passenger vehicles by weight of vehicle. SOURCE: Insurance Institute for Highway Safety, using fatality data from NHTSA’s Fatality Analysis Reporting System (FARS) for 1991–1997, a census of traffic fatalities maintained by NHTSA, and vehicle registration data from the R.L. Polk Company for the same years. FIGURE 4-8 Occupant death rates in two-vehicle crashes for 1990–1996 model passenger vehicles by weight of vehicle. SOURCE: Insurance Institute for Highway Safety, using fatality data from NHTSA’s Fatality Analysis Reporting System (FARS) for 1991–1997, a census of traffic fatalities maintained by NHTSA, and vehicle registration data from the R.L. Polk Company for the same years.

72 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS make up the vehicle fleet (compare Appendix D of NRC, 1992). In Chapter 2 the committee reviewed a 1997 study of the National Highway Traffic Safety Administration (NHTSA) (Kahane, 1997) that estimated the anticipated effect of a 100-lb change in the average weight of the passenger car and light-truck fleets. These estimates indicated that such a weight reduction would have different effects on different crash types. Nevertheless, over all crash types, decreasing the average weight of passenger cars would be expected to increase the motor vehicle fatality risk (all else being equal). Correspondingly, decreasing the average weight of the heavier fleet of light trucks might reduce the motor vehicle crash fatality risk (although the latter result was not statisti- cally significant). Lund and Chapline (1999) have reported similar findings. They found that total fatalities in a hypothetical fleet of rela- tively modern passenger vehicles would be reduced by about 0.26 percent if all pickups and SUVs weighing more than 4,000 lb were replaced with pickups and SUVs weighing 3,500 to 4,000 lb. However, if the heaviest cars, those weigh- ing more than 3,500 lb, were replaced by cars weighing be- tween 3,000 and 3,500 lb, the estimated effect changed to an increase in total fatalities of 4.8 percent. Although the au- thors did not consider crashes with nonoccupants such as pedestrians and bicyclists, whose risk appeared to have no clear relationship to vehicle mass (see Figure 4-10), the results confirm Kahane’s (1997) finding that the expected societal effect of decreases in vehicle mass, whether in response to fuel economy requirements or other factors, de- pends on which part of the vehicle fleet becomes lighter.2 0 40 80 120 160 200 <2500 2500- 3000- 3500- 4000- 4500- 5000+ vehicle weight (lb.) de at hs /m ill io n ve hi cl es /y r cars/passenger vans utility vehicles pickups FIGURE 4-10 Pedestrian/bicyclist/motorcyclist death rates for 1990–1996 model passenger vehicles by vehicle weight. SOURCE: Insurance Institute for Highway Safety, using fatality data from NHTSA’s Fatality Analysis Reporting System (FARS) for 1991– 1997, a census of traffic fatalities maintained by NHTSA, and ve- hicle registration data from the R.L. Polk Company for the same years. FIGURE 4-9 Occupant death rates in other vehicles in two-vehicle crashes for 1990–1996 model passenger vehicles. SOURCE: In- surance Institute for Highway Safety, using fatality data from NHTSA’s Fatality Analysis Reporting System (FARS) for 1991– 1997, a census of traffic fatalities maintained by NHTSA, and ve- hicle registration data from the R.L. Polk Company for the same years. 0 10 20 30 40 50 <2500 2500- 3000- 3500- 4000- 4500- 5000+ vehicle weight (lb.) de at hs /m ill io n ve hi cl es /y r cars/passenger vans utility vehicles pickups 2As noted in Chapter 2, the committee has relied heavily on these NHTSA analyses in its consideration of the likely effects of future improvements in fuel economy. Although they have been the subject of controversy, the committee’s own review of the analyses found no compelling reasons to change the conclusions or alter the estimates of the relationship between vehicle mass and safety. Statistical and conceptual considerations indicate that changes are occurring in the pattern of motor vehicle crash fatalities and injuries, which could affect these estimates in the future, but the stabil- ity of relative occupant fatality rates between lighter and heavier vehicles over the past 20 years does not suggest that the estimates should change significantly over the time period considered by the committee, essentially the next 10–15 years (see Chapter 2). As indicated in Chapter 2, this com- mittee believes that further study of the relationship between size, weight, and safety is warranted, because of uncertainty about the relationship be- tween vehicle weight and safety. However, the majority of the committee believes that these concerns should not prevent the use of NHTSA’s careful analyses to provide some understanding of the likely effects of future im- provements in fuel economy, if those improvements involve vehicle downsizing. The committee believes this position is consistent with that of the National Research Council’s Transportation Research Board (NRC- TRB) committee that reviewed the draft NHTSA report in 1996, which concluded “it is important . . . to provide a sense of the range of uncertainty so that policy makers and researchers can properly interpret the results” (NRC-TRB, 1996, p. 7). NHTSA’s final report has done that, and the use of NHTSA’s results is consistent with this committee’s approach to other un- certainties surrounding efforts to improve fuel economy: that is, to use the best scientific evidence available to gauge likely effects and to state the uncertainty associated with those efforts.

IMPACT OF A MORE FUEL-EFFICIENT FLEET 73 Vehicle Size and Safety Estimates of the effect of vehicle weight have been con- founded with the effects of size because the mass and size of vehicles are correlated so closely. For example, the 2001 Buick LeSabre, a typical large car, is 200 inches long and 74 inches wide and weighs (unloaded) 3,600 lb. A 2001 four- door Honda Civic, a typical small car for the United States, has an overall vehicle length of 175 inches, overall width of 67 inches, and a weight of 2,400 lb (Highway Loss Data Institute, 2001). Thus, some of the negative effect of vehicle weight constraints on safety that has been attributed to mass reduction is attributable to size reduction. Historical changes in the fleet have similarly confounded mass and size characteristics. While the mass of the passen- ger car fleet was decreased about 900 pounds between 1975 and 1990, the length of vehicles also declined, with average wheelbase (the distance between the front and rear axles) declining more than 9 inches (NRC, 1992). Efforts to disen- tangle these effects are hampered by the fact that, when ve- hicles of similar size differ in mass, they usually do so for reasons that still confound the estimate of mass effects (see, for example, Joksch et al., 1998; C.J. Kahane, NHTSA, also spoke of this in his presentation to the committee on Febru- ary 6, 2001). Despite this confounding, carefully controlled research has demonstrated that, given a crash, larger vehicles pro- vide more occupant protection independent of mass. In crashes between vehicles of similar mass, smaller vehicles have higher fatality rates than larger ones (Evans and Frick, 1992; Wood, 1997; Evans, 2001). In addition, Wood (1997) has argued that much of the apparent pro- tective effect of mass in single-vehicle crashes may occur because of the association of mass with size. Theoreti- cally, increased size of one vehicle can be beneficial to other road users as well, to the extent that the increased size translates to more crush space (Ross and Wenzel, 2001; O’Neill, 1998). By the same token, to the extent that size reductions translate to less crush space, smaller size is detrimental to both the vehicle’s own occupants and the occupants of other vehicles. Theoretically, size can affect crash likelihood as well as crashworthiness, independently of mass. For example, re- ductions in size may make a vehicle more maneuverable. Smaller vehicles may be easier to “miss” in the event of a potential collision. Drivers of smaller, lower vehicles may be better able to see and avoid other vulnerable road users such as pedestrians and bicyclists. These effects could lead to reduced societal risk despite the negative effects of re- duced size on occupant protection, given a crash. The committee found no direct empirical support for this crash avoidance benefit, but there is some indirect support in the pattern of Kahane’s (1997) fatality results. For example, Kahane reported a much larger reduction in pedestrian, bicy- clist, and motorcyclist fatalities associated with a reduction in light truck weights than with the same reduction in car weights. The mass ratios between pedestrians and motor ve- hicles are so large that the difference between cars and light trucks cannot be due solely to the change in mass. Rather, it is possible that some of the difference in reduction is due to changes in size, and hence visibility of pedestrians, to the driver. In addition, Kahane’s analyses found no expected change in car-to-car or truck-to-truck crash fatalities as a result of vehicle downweighting. That result is contradictory to other studies indicating that, in the event of a crash, smaller cars offer less occupant protection (see above). The explana- tion could be that a smaller vehicle’s increased injury risk given a crash is offset by a tendency to get into fewer crashes (by virtue of its easier-to-miss, smaller profile or its poten- tially superior agility). The direct evidence that is available contradicts this crash avoidance hypothesis, however. Overall, there is evidence that smaller vehicles actually are involved in more collisions than larger vehicles, relative to their representation in the population of vehicles. The Highway Loss Data Institute (HLDI) tracks the collision insurance claims experience of about 65 percent of the vehicles insured in the United States. Table 4-4 indicates the incidence of collision claims for 1998–2000 models in 1998–2000, relative to the average collision claims frequency for all passenger cars. The results have been standardized to represent similar proportions of drivers less than 25 years of age. The principal pattern of these data is that the frequency of collision claims is higher for smaller vehicles, the opposite of what might be expected from the simple geometry of the vehicles. The important theoretical role of vehicle size for crash- worthiness led the committee to consider whether some of the adverse safety effects associated with decreases in mass could be mitigated if size remained the same. In this context, it is important to distinguish among different meanings of the term “vehicle size.” Changes in size of the occupant com- partment are generally less relevant to safety (but not irrel- evant) than changes in the exterior size of vehicles—the lat- ter affect the size of the so-called crush zones of the vehicle. It is noteworthy that almost all of the downsizing that occurred in conjunction with increased fuel economy oc- curred in this crush zone; the size of the occupant compart- TABLE 4-4 Relative Collision Claim Frequencies for 1998–2000 Models Four-Door Size Class Cars SUVs Pickups Mini 124 — — Small 112 78 87 Midsize 99 76 — Large 88 65 75 Very large 70 86 — SOURCE: Highway Loss Data Institute (2001).

74 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS ments of vehicles has changed little since 1977, when EPA estimated the interior volume of cars to be 110 cubic feet. That fell to a low of 104 cubic feet in 1980, but generally ranged from 108 to 110 cubic feet during the 1990s (EPA, 2000). This reflects the critical effect of interior volume on the utility of passenger cars and hence their marketability. For the same reason, the committee expects that future size reduction, if necessary to reduce weight, would again occur principally outside the occupant compartment, and that inte- rior space would be one of the last areas subject to reduction. What would be the benefits if crush space were retained in a future lighter-weight fleet? Empirical data do not exist to answer this question quantitatively, but the committee would expect a smaller adverse safety effect of vehicle weight reduction. Still there would be some loss of occu- pant protection, because lighter vehicles decelerate more rapidly in crashes with other vehicles or with deformable fixed objects. Effective crush space would therefore have to increase with reductions in mass in order to keep injury risk the same. Thus, if manufacturers try to maintain cur- rent levels of occupant protection as they downweight, they would have to use some of the weight savings from alter- native materials and structures to provide additional crush space. In addition, it must be noted that not all the effects of maintaining size will necessarily be beneficial. Some of the increased risk of injury to occupants associated with vehicle downweighting was offset by reduced injury risk to vulner- able road users (Kahane, 1997). Those offsetting benefits may occur as a result of changes in certain size characteris- tics that are typically associated with weight reduction. For example, a change in the height and size of vehicle front ends may be the critical factor in the estimated benefits to pedestrians of reductions in the weight of light trucks. Thus, if size is maintained, it may reduce the benefits of lighter- weight vehicles for pedestrians and cyclists. In sum, the committee believes that it will be important to maintain vehicle crush space if vehicles are downweighted in the future, but it is unable to develop quantitative esti- mates of the extent to which such efforts can reduce the esti- mated effects of vehicle downweighting. Effect of Downweighting by Vehicle and Crash Type To examine the impact of vehicle weight reduction on the future safety environment, the majority of the committee considered how weight reduction might influence various crash types. This section of the report discusses how the com- mittee expects vehicle weight reduction to operate in various kinds of multivehicle and single-vehicle crashes. Multiple-Vehicle Crashes Multiple-vehicle crashes account for slightly more than half of occupant fatalities. They include crashes between cars and light trucks, crashes between either cars or light trucks and heavy trucks, and crashes of cars with cars and light trucks with light trucks. The effect on safety of down- weighting and downsizing of the light-duty fleet varies among these multiple-vehicle crash types: Collisions Between Cars and Light Trucks Uniform down- weighting of the fleet, that is, reducing weight from all classes of vehicles, might be expected to produce an increase in traffic casualties. However, if the downweighting is restricted to the heaviest pickup trucks and SUVs, with no weight reduction in the smaller light trucks or passenger cars, casualties could decrease. Thus, for these crashes, any change in casualties is very sensitive to how downsizing is distributed. This is a consistent finding in the safety litera- ture and is confirmed by the 1997 NHTSA analysis. Collisions Between Light-Duty Vehicles and Heavy Trucks If there is downsizing in any light-duty vehicles with no corresponding change in heavy trucks, the number of ca- sualties would increase. This finding is reflected in the 1997 NHTSA analyses and throughout the safety literature. Collisions of Cars with Cars or Trucks with Trucks In collisions of like vehicles of similar weight, a reduction in both vehicle weight and size is expected to produce a small increase in casualties. In fact, one consistent finding in the safety literature is that as average vehicle weight declines, crash risks increase (see Evans [2001], for example). The laws of momentum do not explain this typical finding, but analysts typically attribute it to the fact that as vehicle weight declines, so does vehicle size. Narrowly defined studies fo- cusing on pure crashworthiness consistently show this in- crease. However, the 1997 NHTSA analysis, which exam- ined the entire traffic environment, predicted a small (statistically insignificant) decrease in fatalities. It has been hypothesized that this inconsistency might be explained by the greater maneuverability of smaller vehicles and, hence, their involvement in fewer crashes. Studies have not found this to be the case, but it is very difficult to fully normalize data to account for the very complex traffic environment. In this instance, the committee is unable to explain why the 1997 NHTSA analysis does not produce results consistent with those in the safety literature. The rest of NHTSA’s find- ings regarding multivehicle crash types are consistent with the literature. Single-Vehicle Crashes Single-vehicle crashes account for almost half of light- duty vehicle occupant fatalities. As with multiple-vehicle crashes, the safety literature indicates that as vehicle weight and vehicle size decline, crash risks increase. The committee examined three types of single-vehicle crashes: rollovers,

IMPACT OF A MORE FUEL-EFFICIENT FLEET 75 crashes into fixed objects, and crashes with pedestrians and bicyclists. Rollovers Historically, a large part of the increase in casual- ties associated with vehicle downsizing has been attribut- able to rollovers. As vehicles get shorter and narrower they become less stable and their rollover propensity increases. If the length and width of the vehicle can be retained as weight is removed, the effect of weight reduction on rollover pro- pensity can be reduced. However, unless track width wid- ens, the vehicle’s rollover propensity in actual use would still increase somewhat, because occupants and cargo are typically located above the vehicle’s center of gravity (CG), raising the vehicle’s CG-height in use. One Ford safety engineer noted in discussions with the committee that the application of crash avoidance technol- ogy (generically known as electronic stability control) that is being introduced on some new vehicles might significantly reduce vehicle rollovers. It is unknown at this time how ef- fective this technology will be. In this regard, it is worth- while noting the experience with other technology aimed at reducing crash likelihood. Antilock brakes were introduced on vehicles after exten- sive testing indicated they unequivocally improved vehicle handling in emergency situations. In fact, antilock brakes were cited in the 1992 NRC report as a new technology that might offset the negative effects of vehicle mass reductions caused by increased fuel economy requirements (NRC, 1992, p. 59). However, experience with antilock brakes on the road has been disappointing. To date, there is no evidence that antilock brakes have affected overall crash rates at all; the principal effect has been to change the pattern of crashes (Farmer et al., 1997; Farmer, 2001; Hertz et al., 1998). Most tellingly, the initial experience suggested that antilock brakes actually increased fatal, single-vehicle, run-off-the-road crashes and rollovers (Farmer et al., 1997), the very crashes that the NRC committee thought antilock brakes might ben- efit. Recent research suggests that this increase in fatal, single-vehicle crashes associated with antilocks may be di- minishing (Farmer, 2001), but still the evidence does not permit a conclusion that antilocks are reducing crash risk. Thus while it is conceivable that new technology might reduce or eliminate the risk of rollovers, at this time the com- mittee is not willing to change its expectations based on this eventuality. Fixed-Object Collisions Fixed-object collisions occur with both rigid, unyielding objects and yielding objects that can break or deform. For crashes into rigid objects, larger ve- hicles will, on average, permit longer, lower decelerations, although this will be influenced by the stiffness of the ve- hicle. In crashes with yielding, deformable objects, as a vehicle’s weight is reduced, the struck object’s deformation is reduced, increasing the deceleration in the striking vehicle. Thus lighter vehicles, with or without size reductions, would be expected to increase occupant fatality risk. Studies by Evans (1994), Klein et al. (1991), and Partyka and Boehly (1989) all confirm that there is an inverse rela- tionship between occupant safety and vehicle weight in these crashes. Klein et al. found that there was a 10 percent in- crease in fatality risk associated with a 1,000-lb reduction in vehicle weight in single-vehicle, nonrollover crashes. Simi- larly, NHTSA’s 1997 analysis estimated that there is a slightly greater than 1 percent increase in fatality risk associ- ated with a 100-lb reduction in vehicle weight in these crashes.3 Collisions with Pedestrians, Bicyclists, and Motorcyclists If vehicles are downsized, they may be less likely to strike a pedestrian, bicyclist, or motorcyclist. Further, should a colli- sion occur, the reduced mass could lead to a reduction in casualties, although, given the mismatch in the mass between the vehicle and the unprotected road user, any benefit would not be expected to be very large. Safety Impacts of Possible Future Fuel Economy Scenarios The foregoing discussion provides the majority of the committee’s general views on how vehicle weight and size reductions, if they occur in response to demands for in- creased fuel economy, could affect the safety of motor ve- hicle travel. The committee evaluated the likely safety ef- fects of a number of possible scenarios involving different amounts of downweighting of the future fleet, using the re- sults of the NHTSA analyses (Kahane, 1997) for quantita- tive guidance. Two of these possible scenarios are presented below. They are not intended to reflect the committee’s rec- ommendations, but they do reflect two possibilities that can be instructive in evaluating future fuel economy require- ments. The first scenario examines the safety consequences that would be expected if manufacturers achieved a 10 per- cent improvement in fuel economy using the same pattern of 3In an attempt to gain more insight into single-vehicle crashes, Partyka (1995) examined field data on single-vehicle, nonrollover crashes into yield- ing fixed objects. Her goal was to learn if there was a relationship between vehicle weight and crash outcomes. However, in this study, rather than examining occupant injury as the crash outcome of interest, crash outcomes were defined as whether the yielding object was damaged in the crash. For frontal crashes she found that damage to the yielding fixed object was more likely to occur in crashes with heavier vehicles. However, there was no consistent relationship between vehicle weight and damage to the yielding object in side crashes. The author was not able to consider crash severity as a possible explanation for this anomaly since NHTSA’s national accident sampling system (NASS) did not provide estimates of crash severity for any of the events where the yielding object was damaged. Further, the author did not aggregate the data to reach an overall conclusion regarding the rela- tionship between vehicle weight and damage to yielding fixed objects be- cause she was uncomfortable with the results of the side impact analysis and questioned its reliability (personal communication from the author).

76 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS downweighting and downsizing that occurred in the late 1970s and 1980s. As discussed above, it is uncertain to what extent downweighting and downsizing will be part of manu- facturers’ strategies for improving fuel economy in the fu- ture, but one guide to their behavior can be to look at what they did in the past. The second scenario is based on the minimal weight change predictions associated with the committee’s cost-ef- ficient analysis of future technology, discussed earlier in this chapter. Thus, this analysis projects the safety consequences if manufacturers apply likely powertrain technology prima- rily to improve vehicle fuel economy. In both scenarios, the hypothetical weight reductions are based on the MY 2000 light-duty vehicle fleet (the average car weighed 3,386 lb and the average light truck, 4,432 lb). The estimated effects of the weight reductions are applied to the 1993 vehicle fleet, the reference year for the NHTSA analyses (Kahane, 1997). Obviously the results in these sce- narios cannot be, nor are they intended to be, precise projec- tions. NHTSA’s analysis was based on the characteristics of 1985–1993 vehicles and was applied to the 1993 fleet. Fur- ther, even if manufacturers dropped the average weight of the vehicles described in the two scenarios starting in 2002, it would take many years before those changes had any sig- nificant influence on the overall makeup of the future ve- hicle fleet. And that fleet would be operating in an environ- ment that will differ from the environment that was the basis for the 1997 NHTSA analysis. Nevertheless, the committee’s majority believes that the calculations below provide some insight into the safety im- pacts that might occur if manufacturers reduce weight in re- sponse to increased fuel economy standards, as described below. Historical Pattern Scenario Between 1975 and 1984, the average weight of a new car dropped from 4,057 to 3,098 lbs (24 percent). For light trucks, weight declined from 4,072 to 3,782, or 290 lb (7 percent). These mass reductions suggest a 17 percent im- provement in fuel economy for cars and a 5 percent improve- ment for light trucks.4 During those years, the fuel economy of cars and light trucks actually improved by about 66 per- cent (from 15.8 mpg to 26.3 mpg) and 50 percent (from 13.7 mpg to 20.5 mpg), respectively. In other words, vehicle downsizing accounted for about 25 percent of the improve- ment in fuel economy of cars and 10 percent of the improve- ment of light trucks between 1975 and 1984. A similar pattern in achieving a 10 percent improvement in fuel economy today would imply a 3.6 percent weight reduction for cars (122 lb for 2000 models) and a 1.4 percent weight reduction for light trucks (63 lb for 2000 models). Had this weight reduction been imposed on the fleet in 1993, it would have been expected to increase fatalities involving cars about 370 (±110) and to decrease fatalities involving light trucks by about 25 (±40). The net effect in 1993 for a 10 percent improvement in fuel economy with this mix of downsizing and increased fuel efficiency would have been about 350 additional fatalities (95 percent confidence inter- val of 229 to 457). Cost-Efficient Scenario Earlier in this chapter (under “Analysis of Cost-Efficient Fuel Economy”), cost-efficient fuel economy increases of 12 to 27 percent for cars and 25 to 42 percent for light trucks were estimated to be possible without any loss of current performance characteristics. In the cost-efficient analysis, most vehicle groups actually gain approximately 5 percent in weight to account for equipment needed to satisfy future safety standards. Thus, for these vehicle groups, cost-effi- cient fuel economy increases occur without degradation of safety. In fact, they should provide enhanced levels of occu- pant protection because of both the increased level of safety technology and the increased weight of that technology. For three groups of vehicles (large cars, midsize SUVs, and large SUVs), under the cost-efficient scenario, it is pro- jected that manufacturers would probably compensate for the added weight attributable to safety technology with weight reductions in other areas. These weight reductions mean that occupants of these vehicles are somewhat less pro- tected in crashes than they otherwise would have been (though they still benefit from the added safety technology). Estimating the potential effect of this limited downsizing is complicated because the estimates of the safety effect of 100-lb changes in average vehicle weight developed by NHTSA cannot be applied directly to changes in specific vehicle groups. However, NHTSA’s report included sensi- tivity analyses that estimated the effect of weight reductions restricted to the heaviest 20 percent of cars (those heavier than 3,262 lb) and to the heaviest 20 percent of light trucks (those heavier than 3,909 lb) (Kahane, 1997, pp. 165–172). Specifically, NHTSA estimated that a 500-lb reduction in these vehicle groups in 1993 would have resulted in about 250 additional fatalities in car crashes and about 65 fewer fatalities in light-duty truck crashes. If it is assumed that cars greater than 3,262 lb correspond roughly to “large cars” as referred to in the committee’s cost- efficient fuel economy analysis outlined earlier in this chap- ter, then the 5 percent weight reduction, had it occurred in 1993, would have been about 160 to 180 lb, or only about one-third of the reduction of the NHTSA analysis. Thus, it is estimated that reducing the weight of large cars in the way foreseen in the cost-effective fuel economy analyses would have produced about 80 additional fatalities in car crashes in 1993. 4Based on an assumed 7 percent improvement in fuel economy for each 10 percent reduction in weight.

IMPACT OF A MORE FUEL-EFFICIENT FLEET 77 Similarly, if midsize and large SUVs are assumed to be in the heaviest 20 percent of light-duty trucks and if such ve- hicles account for as much as half the light-duty truck popula- tion in the near future, then about half the light-duty truck fleet will be 5 percent lighter (about 200 lb). This weight reduction is about 40 percent of the reduction studied by NHTSA and would have saved 15 lives had it occurred in 1993. Adding these effects yields an estimated increase of 80 car crash fatalities minus 15 fewer truck crash fatalities, or about 65 additional fatalities. NHTSA’s sensitivity analysis did not include standard errors for the weight reductions in the heaviest vehicles, but in this case they would clearly be large enough that reason- able confidence bounds for this estimate would include zero. Thus, an increase in fatality risk in the future fleet is pre- dicted from the kinds of weight reductions included in the cost-efficient fuel economy scenario identified by the com- mittee, but the uncertainty of this estimate is such that the effect might be zero and would be expected to result in fewer than 100 additional fatalities. In addition, it should be noted that these results do not imply that the actual safety effect would be as small as this if the required fuel economy rises to the targets indicated in the cost-efficient analyses. This is because manufacturers could choose to use advances in drive train technology for other vehicle attributes such as acceleration or load capac- ity. In that case, additional vehicle downweighting might occur, and the adverse safety consequences would grow. Thus, the actual safety implications of increasing fuel economy to the cost-efficient levels specified earlier in this chapter will depend on what strategies manufacturers actu- ally choose in order to meet them and the structure of the regulatory framework. Conclusion: Safety Implications of Increased CAFE Requirements In summary, the majority of the committee finds that the downsizing and weight reduction that occurred in the late 1970s and early 1980s most likely produced between 1,300 and 2,600 crash fatalities and between 13,000 and 26,000 serious injuries in 1993. The proportion of these casualties attributable to CAFE standards is uncertain. It is not clear that significant weight reduction can be achieved in the fu- ture without some downsizing, and similar downsizing would be expected to produce similar results. Even if weight reduction occurred without any downsizing, casualties would be expected to increase. Thus, any increase in CAFE as cur- rently structured could produce additional road casualties, unless it is specifically targeted at the largest, heaviest light trucks. For fuel economy regulations not to have an adverse im- pact on safety, they must be implemented using more fuel- efficient technology. Current CAFE requirements are neu- tral with regard to whether fuel economy is improved by increasing efficiency or by decreasing vehicle weight. One way to reduce the adverse impact on safety would be to es- tablish fuel economy requirements as a function of vehicle attributes, particularly vehicle weight (see Chapter 5). An- other strategy might be to limit horsepower-to-weight ratios, which could save fuel by encouraging the application of new fuel efficiency technology for fuel economy rather than performance. The committee would also like to note that there is a re- markable absence of information and discussion of the speed limit in the context of fuel consumption. The national 55- mph speed limit that was in effect until 1987 is estimated to have reduced fuel consumption by 1 to 2 percent while si- multaneously preventing 2,000 to 4,000 motor vehicle crash deaths annually (NRC, 1984). Relaxation of the speed limit has increased fatalities in motor vehicle crashes (Baum et al., 1991; Farmer et al., 1999), and fuel consumption pre- sumably is again higher than it would have been. In addition, it is reasonable to expect that higher speed limits, combined with higher fuel efficiency, provide incentives for driving further and faster, thereby offsetting intended increases in fuel economy, though this hypothesis is speculative on the part of the committee. But the committee wants to close where it started, by con- sidering the effect of future advances in safety technology on CAFE. Despite the adverse safety effects expected if downweighting occurs, the effects are likely to be hidden by the generally increasing safety of the light-duty vehicle fleet. This increase in safety is driven by new safety regulations, testing programs that give consumers information about the relative crash protection offered by different vehicles, and better understanding of the ways in which people are injured in motor vehicle crashes. Alcohol-impaired driving has been decreasing and seat belt usage has been increasing. Roads are becoming less hazardous for vehicles. Some might argue that this improving safety picture means that there is room to improve fuel economy without adverse safety consequences. However, such a measure would not achieve the goal of avoiding the adverse safety consequences of fuel economy increases. Rather, the safety penalty imposed by increased fuel economy (if weight re- duction is one of the measures) will be more difficult to iden- tify in light of the continuing improvement in traffic safety. Just because these anticipated safety innovations will im- prove the safety of vehicles of all sizes does not mean that downsizing to achieve fuel economy improvements will have no safety costs. If an increase in fuel economy is effected by a system that encourages either downweighting or the production and sale of more small cars, some additional traffic fatalities would be expected. Without a thoughtful restructuring of the pro- gram, that would be the trade-off that must be made if CAFE standards are increased by any significant amount.

78 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS REFERENCES Austin, T.C., R.G. Dulla, and T.R. Carlson. 1999. Alternative and Future Technologies for Reducing Greenhouse Gas Emissions from Road Ve- hicles. Prepared for the Committee on Road Vehicle Technology and Fuels, Natural Resources Canada. Sacramento, Calif.: Sierra Research, Inc., July. Baum, H.M., J.K. Wells, and A.K. Lund. 1991. “The Fatality Consequences of the 65 mph Speed Limits, 1989.” Journal of Safety Research, 22: 171–177. Butters, Jamie. 2001. “Outlook Lowered for Ford and GM: S&P Expresses Concern About Creditworthiness.” Detroit Free Press, February 7. Charles River Associates, Inc. 1995. Policy Alternatives for Reducing Pe- troleum Use and Greenhouse Gas Emissions: The Impact of Raising Corporate Average Fuel Economy (CAFE) Standards, July. DeCicco, J., and D. Gordon. 1993. Steering with Prices: Fuel and Vehicle Taxation as Market Incentives for Higher Fuel Economy, December. Washington, D.C.: ACEEE. DeCicco, J., F. An, and M. Ross. 2001. Technical Options for Improving the Fuel Economy of U.S. Cars and Light Trucks by 2010–2015, April. Washington, D.C.: American Council for an Energy-Efficient Economy (ACEEE). EEA (Energy and Environmental Analysis, Inc.). 2001. Technology and Cost of Future Fuel Economy Improvements for Light-Duty Vehicles. Draft Final Report. Prepared for the committee and available in the National Academies’ public access file for the committee’s study. EIA (Energy Information Administration). 2001. Annual Energy Outlook 2001. Table A 7, Transportation Sector Key Indicators and Delivered Energy Consumption. Washington, D.C.: EIA. EPA (Environmental Protection Agency). 2000. Light-Duty Automotive Technology and Fuel Economy Trends 1975 Through 2000. EPA-R- 00-008. December 2000. Evans, L. 1994. “Driver and Fatality Risk in Two-Car Crashes Versus Mass Ratio Inferred.” Accident Analysis and Prevention 26: 609–616. Evans, L. 2001. “Causal Influence of Car Mass and Size on Driver Fatality Risk.” American Journal of Public Health 91(7): 1076–1081. Evans, L., and M.C. Frick. 1992. “Car Size or Car Mass: Which Has Greater Influence on Fatality Risk.” American Journal of Public Health 82 (8): 1105–1112. Evans, L., and M.C. Frick. 1993. “Mass Ratio and Relative Driver Fatality Risk in Two-Vehicle Crashes.” Accident Analysis and Prevention 25 (2): 213–224. Farmer, C.M. 2001. New Evidence Concerning Fatal Crashes of Passenger Vehicles Before and After Adding Antilock Braking Systems. Accident Analysis and Prevention 33: 361–369. Farmer, C.M., A. Lund, R. Trempel, and E. Braver. 1997. “Fatal Crashes of Passenger Vehicles Before and After Adding Antilock Braking Sys- tems.” Accident Analysis and Prevention 29: 745–757. Farmer, C.M., R.A. Retting, and A.K. Lund. 1999. “Changes in Motor Ve- hicle Occupant Fatalities After Repeal of the National Maximum Speed Limit.” Accident Analysis and Prevention 31: 537–543. Greene, D.L. 2001. Cost-Efficient Economy Analysis. Draft. Prepared for the committee and available in the National Academies’ public access file for the committee’s study. Greene, D.L., and Steven E. Plotkin. 2001. Energy Futures for the U. S. Transportation Sector. Draft, April. Greene, D.L., and J. DeCicco. 2000. Engineering-Economic Analyses of Automotive Fuel Economy Potential in the United States. ORNL/ TM–2000/26, February. Oak Ridge, Tenn.: Oak Ridge National Laboratory. Hertz, E., J. Hilton, and D. Johnson. 1998. Analysis of the Crash Experience of Vehicles Equipped with Antilock Braking Systems—An Update (98- S2-O-07). Proceedings of the 16th International Technical Conference on the Enhanced Safety of Vehicles. Washington, D.C.: NHTSA. Highway Loss Data Institute. 2001. Fatality Facts: Passenger Vehicles. Available online at <http://www.carsafety.org>. Joksch, H., D. Massie, and R. Pichler. 1998. Vehicle Aggressivity: Fleet Characterization Using Traffic Collision Data. DOT HS 808 679. Wash- ington, D.C.: NHTSA. Kahane, C.J. 1997. Relationships Between Vehicle Size and Fatality Risk in Model Year 1985–93 Passenger Cars and Light Trucks. NHTSA Technical Report, DOT HS 808 570. Springfield, Va: National Techni- cal Information Services. Klein, T., E. Hertz, and S. Borener. 1991. A Collection of Recent Analyses of Vehicle Weight and Safety. Technical Report DOT HS 807 677. Washington, D.C.: NHTSA. Lund, A.K., and J.F. Chapline. 1999. Potential Strategies for Improving Crash Compatibility in the U.S. Vehicle Fleet (SAE 1999-01-0066). Vehicle Aggressivity and Compatibility in Automotive Crashes (SP- 1442), 33–42. Warrendale, Pa.: SAE. NRC (National Research Council). 1984. 55: A Decade of Experience. Spe- cial Report 204. Washington, D.C.: Transportation Research Board. NRC. 1992. Automotive Fuel Economy: How Far Should We Go? Wash- ington, D.C.: National Academy Press. NRC-TRB (National Research Council, Transportation Research Board). 1996. Committee to Review Federal Estimates of the Relationship of Vehicle Weight to Fatality and Injury Risk. Letter report to Dr. Ricardo Martinez, Administrator, National Highway Traffic Safety Administra- tion, June 12. Washington, D.C.: NRC. O’Neill, B. 1998. The Physics of Car Crashes and the Role of Vehicle Size and Weight in Occupant Protection. Physical Medicine and Rehabilita- tion: State of the Art Reviews. Philadelphia, Pa.: Hanley & Belfus, Inc. Partyka, S.C. 1995. Impacts with Yielding Fixed Objects by Vehicle Weight. NHTSA Technical Report, DOT HS 808-574, June. Washing- ton, D.C.: NHTSA. Partyka, S.C., and W.A. Boehly. 1989. Passenger Car Weight and Injury Severity in Single Vehicle Nonrollover Crashes. Proceedings of the Twelfth International Technical Conference on Experimental Safety Vehicles. Washington, D.C.: NHTSA. Patterson, P. 1999. Reducing Greenhouse Gases in the U.S. Transportation Sector. SAE, Government and Industry Meeting, April 28, Washington, D.C. Ross, M., and T. Wenzel. 2001. Losing Weight to Save Lives. A Review of the Role of Automobile Weight and Size in Traffic Fatalities, Report submitted to the committee on March 13, 2001. Wang, M. 1996. GREET 1.0—Transportation Fuel Cycle Model: Method- ology and Use, ANL/ESD-33. Argonne, Ill.: Argonne National Laboratory. Wang, M., and H.S. Huang. 1999. A Full Fuel-Cycle Analysis of Energy and Emissions Impacts of Transportation Fuels Produced from Natural Gas. ANL/ESD-40. Argonne, Ill.: Argonne National Laboratory. Weiss, M. A., J. B. Heywood, E. M. Drake, A. Schafer, and F. F. AuYeung. 2000. On the Road in 2020: A Life-Cycle Analysis of New Automobile Technologies. Energy Laboratory Report # MIT EL 00-003, October. Cambridge, Mass.: Massachusetts Institute of Technology (MIT). Wood, D.P. 1997. “Safety and the Car Size Effect: A Fundamental Expla- nation.” Accident Analysis and Prevention 29 (2): 139–151.

79 Assessments of new automobile technologies that have the potential to function with higher fuel economies and lower emissions of greenhouse gases (GHGs) have been made by the Energy Laboratory at the Massachusetts Insti- tute of Technology (MIT) (Weiss et al., 2000) and by the General Motors Corporation (GM), Argonne National Labo- ratory, BP, ExxonMobil, and Shell (General Motors et al., 2001, draft). Both studies compared fuels and engines on a total systems basis, that is, on a well-to-wheels (WTW) ba- sis. These assessments provide an indication of areas of promising vehicle and fuel technology and benchmarks for likely increases in fuel economy and reduction of GHG emis- sions from the light-duty fleet over the next two decades. This attachment provides additional information on the emerging technologies and GHG emissions described in Chapters 3 and 4. MIT’s analysis was confined to midsize cars with con- sumer characteristics comparable to a 1996 reference car such as the Toyota Camry. It was assumed that, aided by the introduction of low-sulfur fuels, all technologies would be able to reduce emissions of air pollutants to levels at or be- low federal Tier 2 requirements. Only those fuel and vehicle technologies that could be developed and commercialized by 2020 in economically significant quantities were evalu- ated. General Motors et al. (2001) focused on the energy use of advanced conventional and unconventional power-train systems that could be expected to be implemented in the 2005 to 2010 time frame in a Chevrolet Silverado full-size pickup truck. The technologies were assessed on the basis of their potential for improving fuel economy while maintain- ing the vehicle performance demanded by North American consumers. Vehicle architectures and fuels analyzed in both studies are listed in Table 4A-1. In the MIT study, the vehicle lifetimes and driving dis- tances were assumed to be similar for all vehicles. The more advanced technologies were compared to an “evolved baseline” vehicle—a midsize passenger car comparable in consumer characteristics to the 1996 reference car, in which fuel consumption and GHG emissions had been reduced by about a third by 2020 through continuing evolutionary im- provements in the traditional technologies currently being used. Figure 4A-1 summarizes energy use, GHG emissions, and costs for all the new 2020 technologies relative to the 1996 reference car and the evolved 2020 baseline car. (The battery-electric car shown is an exception in that it is not “comparable” to the other vehicles; its range is about one- Attachment 4A Life-Cycle Analysis of Automobile Technologies TABLE 4A-1 Vehicle Architecture and Fuels Used in the MIT and General Motors et al. Studies MIT (Weiss et al., 2000) General Motors et al. (2001) 1996 reference internal combustion engine (ICE) Conventional (CONV) with spark ignition (SI) gasoline engine (baseline) [GASO SI CONV] Baseline evolved ICE CONV with SI E85 (85% ethanol and 15% gasoline by volume) engine [HE85 SI CONV] Advanced gasoline ICE CONV with compression-ignition direct-injection (CIDI) diesel engine [DIESEL CIDI CONV] Gasoline ICE hybrid vehicle (HEV) Charge-sustaining (CS) parallel hybrid electric vehicle (HEV) with SI E85 engine [HE 85 SI HEV] Diesel ICE hybrid CS HEV with CIDI diesel [DIESEL CIDI HEV] CNG ICE hybrid Gasoline fuel processor (FP) fuel cell vehicle (FCV) [GASO FP FC FCV] Gasoline fuel cell (FC) hybrid vehicle Gasoline (naphtha) FP fuel cell (FC) HEV [NAP FP FC HEV] Methanol FC hybrid Gaseous hydrogen (GH2) refueling station (RS) FC HEV [GH2 RS FC HEV] Hydrogen FC hybrid Methanol (MeOH) FP FCV [MEOH FP HEV] Battery electric Ethanol FP FC HEV [HEV 100 FP FC HEV]

80 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS FIGURE 4A-1 Life-cycle comparisons of technologies for midsize passenger vehicles. NOTE: All cars are 2020 technology except for the 1996 reference car. On the scale, 100 = 2020 evolutionary baseline gasoline ICE car. Bars show estimated uncertainties. SOURCE: Weiss et al. (2000).

IMPACT OF A MORE FUEL-EFFICIENT FLEET 81 third less than that of the other vehicles.) The bars suggest the range of uncertainty surrounding the results. The uncer- tainty is estimated to be about ±30 percent for fuel-cell and battery vehicles, ±20 percent for hybrid electric vehicles (HEVs) using internal combustion engines (ICE), and ±10 percent for all other vehicle technologies. MIT concludes that continued evolution of the tradi- tional gasoline car technology could result in 2020 vehicles that reduce energy consumption and GHG emissions by about one-third relative to comparable vehicles of today at a cost increment of roughly 5 percent. More advanced tech- nologies for propulsion systems and other vehicle compo- nents could yield additional reductions in life-cycle GHG emissions (up to 50 percent lower than those of the evolved baseline vehicle) at increased purchase and use costs (up to about 20 percent greater than those of the evolved baseline vehicle). Vehicles with HEV propulsion systems using ei- ther ICE or fuel-cell power plants are the most efficient and lowest-emitting technologies assessed. In general, ICE HEVs appear to have advantages over fuel-cell HEVs with respect to life-cycle GHG emissions, energy efficiency, and vehicle costs, but the differences are within the uncertain- ties of MIT’s results and depend on the source of fuel en- ergy. If automobile systems with drastically lower GHG emissions are required in the very long run future (perhaps in 30 to 50 years or more), hydrogen and electrical energy are the only identified options for “fuels,” but only if both are produced from nonfossil sources of primary energy (such as nuclear or solar) or from fossil primary energy with carbon sequestration. The results from the General Motors et al. study (2001) are shown in Figures 4A-2 and 4A-3. The diesel compres- sion-ignition direct-injection (CIDI)/HEV, gasoline and naphtha fuel-processor fuel-cell HEVs, as well as the two hydrogen fuel-cell HEVs (represented only by the gaseous hydrogen refueling station and fuel-cell HEV in Figure 4A- 2) are the least energy-consuming pathways. All of the crude- oil-based selected pathways have well-to-tank (WTT) en- ergy loss shares of roughly 25 percent or less. A significant fraction of the WTT energy use of ethanol is renewable. The ethanol-fueled vehicles yield the lowest GHG emissions per mile. The CIDI HEV offers a significant reduction of GHG emissions (27 percent) relative to the conventional gasoline spark-ignitied (SI) vehicle. Considering both total energy use and GHG emissions, the key findings by General Motors et al. (2001) are these: • Of all the crude oil and natural gas pathways studied, the diesel CIDI hybrid electric vehicle (HEV), the gasoline and naphtha HEVs, and the gaseous hydro- gen fuel-cell HEVs were nearly identical and best in terms of total system energy use. Of these technolo- gies, GHG emissions were expected to be lowest for the gaseous hydrogen fuel-cell HEV and highest for the diesel CIDI HEV. • The gasoline-spark-ignited HEV and the diesel CIDI FIGURE 4A-2 Well-to-wheels total system energy use for selected fuel/vehicle pathways. SOURCE: General Motors et al. (2001).

82 EFFECTIVENESS AND IMPACT OF CORPORATE AVERAGE FUEL ECONOMY (CAFE) STANDARDS HEV, as well as the conventional CIDI diesel, offer significant total system energy use and GHG emission benefits compared with the conventional gasoline engine. • The methanol fuel-processor fuel-cell HEV offers no significant energy use or emissions reduction advan- tages over the crude-oil-based or other natural-gas- based fuel cell HEV pathways. • Bioethanol-based fuel/vehicle pathways have by far the lowest GHG emissions of the pathways studied. • Major technology breakthroughs are required for both the fuel and the vehicle for the ethanol fuel-processor fuel-cell HEV pathway to reach commercialization. • On a total system basis, the energy use and GHG emis- sions of compressed natural gas and gasoline spark- ignited conventional pathways are nearly identical. (The compressed natural gas [CNG] pathway is not shown in Figures 4A-2 or 4A-3). • The crude-oil-based diesel vehicle pathways offer slightly lower system GHG emissions and consider- ably better total system energy use than the natural- gas-based Fischer-Tropsch diesel fuel pathways. Cri- teria pollutants were not considered. • Liquid hydrogen produced in central plants, Fischer- Tropsch naphtha, and electrolysis-based hydrogen fuel-cell HEVs have slightly higher total system en- ergy use and the same or higher levels of GHG emis- sions than gasoline and crude naphtha fuel-processor fuel-cell HEVs and electrolytically generated hydro- gen fuel-cell HEVs. REFERENCES General Motors Corporation, Argonne National Laboratory, BP, Exxon Mobil and Shell. 2001. Wheel-to-Well Energy Use and Greenhouse Gas Emissions of Advanced Fuel/Vehicle Systems. North American Analy- sis, April. Final draft. Weiss, Malcolm A., John B. Heywood, Elisabeth M. Drake, Andreas Schafer, and Felix F. AuYeung. 2000. On the Road in 2020: A Life- Cycle Analysis of New Automobile Technologies. Energy Laboratory Report # MIT EL 00-003, October. Cambridge, Mass.: Massachusetts Institute of Technology. FIGURE 4A-3 Well-to-wheels greenhouse gas emissions for selected fuel/vehicle pathways. SOURCE: General Motors et al. (2001).

Next: 5 Potential Modifications of and Alternatives to CAFE »
Effectiveness and Impact of Corporate Average Fuel Economy (CAFE) Standards Get This Book
×
Buy Paperback | $59.00 Buy Ebook | $47.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Since CAFE standards were established 25 years ago, there have been significant changes in motor vehicle technology, globalization of the industry, the mix and characteristics of vehicle sales, production capacity, and other factors. This volume evaluates the implications of these changes as well as changes anticipated in the next few years, on the need for CAFE, as well as the stringency and/or structure of the CAFE program in future years.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!