October 26, Submitted via

Transcription

1 October 26, 2018 Submitted via NHTSA Docket Management Facility, M 30 U.S. Department of Transportation West Building, Ground Floor, Rm. W New Jersey Avenue SE Washington, DC Environmental Protection Agency EPA Docket Center - Air and Radiation Docket Mail Code 28221T 1200 Pennsylvania Avenue NW Washington, DC RE: Docket ID s No. EPA HQ OAR and DOT : The Safer Affordable Fuel- Efficient (SAFE) Vehicles Rule for Model Years Passenger Cars and Light Trucks Dear Sir/Madam The Aluminum Association (the Association ) thanks the National Highway and Transportation Safety Administration (NHTSA) and the Environmental Protection Agency (EPA) for the opportunity to provide comment on the recent Notice of Proposed Rulemaking, The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Years Passenger Cars and Light Trucks as noticed at 83 FR on August 24, The Aluminum Association, based in Arlington, VA, represents U.S. suppliers of primary aluminum, aluminum recyclers, and producers of fabricated aluminum products, as well as industry related businesses. The U.S. aluminum industry directly employs 161,000 workers and indirectly employs an additional 551,000 workers. Its economic output directly generates $75 billion and indirectly generates an additional $111 billion in economic output. In total, the U.S. aluminum industry supports nearly 713,000 jobs and $186 billion in economic output, more 1

2 than 1 percent of the U.S. Gross Domestic Product. Since 2013 in the U.S., aluminum companies have committed to invest over $2.7 billion in new plants and expansions of existing plants, creating thousands of new permanent high-paying jobs that drive investments that strengthen American manufacturing job base and the U.S. economy. This growth has been to support aluminum in automotive market applications and importantly, if automotive materials market conditions support further aluminum investments, the aluminum industry will invest even more in domestic jobs and related facilities. Aluminum has shown over 40 years of continuous growth in the North American automotive market, and research has confirmed that trend is expected to continue as automakers opt for aluminum as a material of choice not only for its reduced emissions and fuel economy benefits but also for its ability to enhance consumer attributes in areas of handling, braking, and acceleration, crash protection, and corrosion resistance. Within the Association, the Aluminum Transportation Group (ATG) focuses on aluminum use in transportation applications and has a long history of data-driven technical interaction with automakers in providing safe and cost-effective lightweighting solutions to automakers to help them comply with the EPA greenhouse gas (GHG) emission reduction and NHTSA fuel-efficiency standards. The ATG has also consistently engaged with EPA, NHTSA and the California Air Resources Board (CARB) in the evaluation of vehicle mass reduction solutions as an integral component of these agencies ongoing regulatory efforts to reduce GHG emissions and improve related fuel efficiency performance. As such, EPA and NHTSA s request for comment on the proposed SAFE Vehicles Rule are both of significant interest to the ATG specifically, and to the Association more broadly. The Association reviewed the proposed SAFE Vehicles Rule [1] and supporting PRIA [4] from the perspectives of specific references to aluminum and overall vehicle mass reduction in general. From that review the Association supports: 2

3 The mid-term evaluation process, the opportunities presented for suppliers to participate in it, and the willingness of the agencies to engage in constructive dialogue regarding the path forward for future CAFE/GHG emissions regulations. Further, the aluminum industry believes the mid-term evaluation process must embrace an objective data-driven approach, spur competition and innovation, and maintain global competitiveness of the U.S. auto industry. The specific references to aluminum, which are accurate and consistent with practical automotive industry experience and future program expectations. Mass reduction utilizing advanced materials is recognized as part of the technology strategy to achieve safe, fuel efficient and cost-effective vehicles that meet or exceed consumer demands. Vehicle size, not weight, has been shown to be the leading safety determinant and aluminum is recognized as an increasingly important component of vehicle lightweighting and advanced materials strategies. The continued use of footprint-based standards, which have proven to be an appropriate means of characterizing the vehicle fleet and drives fuel-efficiency improvement across all vehicle classes. The footprint-based standards also eliminate the incentive to shift fleet volume to smaller cars which has been shown to slightly decrease safety in vehicle-tovehicle collisions. It also provides an incentive for reducing weight in the larger vehicles, where weight reduction is of the most benefit for societal safety. As an example of this, the arrival of lighter yet safer but still large vehicles like Ford s top selling, aluminum-bodied F150 makes it clear that the footprint-based regulatory approach is working as intended. One National Program, as the alignment of fuel efficiency standards between EPA, NHTSA, and CARB is necessary to establish the clarity needed for continued improvement in vehicle technologies. EPA, NHTSA, and CARB must continue to work closely together with a focus on inter-agency collaboration to provide the automobile industry and its suppliers with the optimal alignment, consistency, and certainty contemplated by the One National Program concept. It is critical in this regard that negotiation and agreement on a final rule 3

4 be reached between the federal agencies and CARB so as to prevent litigation and its deleterious effect on the ability of automakers and their suppliers to have the certainty needed to manufacture compliant vehicles and components. Future Regulatory Certainty, as this was a strength of the 2012 regulation and was provided by the inclusion of augural standards through This approach has proven beneficial for automakers and suppliers as many automotive technologies take more than five years to engineer, develop, validate and establish production capacity. The clarity in the standards has been instrumental in supporting the unprecedented improvements in efficient technologies seen over the past six years. Long term regulatory certainty is essential to the U.S. aluminum industry, which continues to make significant investments in new products, processes and capacity to meet growing demand for automotive aluminum. In addition to the areas of support noted above, the Association has comments and concerns about the proposed SAFE Vehicles Rule in the following areas: 1.0 Vehicle Safety and Related Assumptions 1.1 Mass Reduction < 3,200 Lbs. GVW The Association requests that assumptions regarding vehicle mass reduction be corrected to reflect the marketplace reality that vehicles < 3,200 Lbs. GVW are not being lightweighted. 1.2 New Vehicle Price Elasticity and Vehicle Replacement The Association requests that the relationship between technology cost, vehicle sales, and fleet age be re-evaluated using the analysis provided which indicates that CAFE/GHG technologyrelated new vehicle net price increases anticipated would not have a significant impact on new vehicle sales. In terms of affordability, consumers have a broad range of automotive product alternatives to choose from before they elect to postpone a new car purchase. 1.3 VMT and the Rebound Effect The Association requests that the VMT rebound effect be corrected to a more accurate level of 10% or less for the analysis of its impact on vehicle occupant fatalities. 4

5 1.3.1 Vehicle age, scrappage rates, projected fleet size, and impact on fleet safety The Association requests that assumptions regarding vehicle age and fleet size be corrected to reflect only the differences in the increased or decreased sales of new vehicles within a given year in the analysis of its impact on vehicle occupant safety and costs. 1.4 Other SAFE Vehicle Rule Issues Original Equipment Manufacturer (OEM) over compliance to proposed alternative requirements Maximum Technology Performance Glider mass fraction and the projected effectiveness of mass reduction The Association requests that the assumptions used in each of these 3 areas to model their effect on the SAFE Vehicle Rule costs and benefits be revised using the corrected information provided. 2. Future CAFE/GHG Standards and Rate of Vehicle Improvement The Association requests that upon consideration of all the data available to EPA and NHTSA that the SAFE Vehicles Rule be revised to set CAFE/GHG emission standards within a range between Alternative 6 and the augural standards. As importantly, the final rule must achieve the One National Program concept for all 50 states and the District of Columbia. In conjunction with a SAFE Vehicles Rule with standards within this range, the Association requests that augural standards out beyond 2026 be provided, at least out to 2030 and preferably out to 2035, to provide the necessary lead time and certainty for the investment decisions necessary to achieve the targets. Recognizing that this is a long time horizon, the Association supports building a mid-term quantitative review to be completed by December 2023 into that next set of augural standards. 3. Sustainability The Association agrees with EPA s prior conclusions that the use of aluminum intensive designs in vehicles generally achieves the largest reductions in life-cycle energy use and GHG impacts. 5

6 Each of these areas of comment and concern are discussed in detail in the sections below. 1.0 Vehicle Safety and Related Assumptions The Association believes the SAFE Vehicles Rule safety and cost projections are based on assumptions that are not consistent with historical market performance or expected future market performance. Those assumptions result in overstating potential unfavorable impacts on safety, societal cost of the regulation and new vehicle sales. The SAFE Vehicles Rule assessments are based on key assumptions that do not appear to accurately represent actual fleet performance and significantly influence fleet safety and societal cost estimates. Major safety related assumptions impacting assessments are: o Mass reduction of vehicles below 3,200 Lbs. GVW o Assumption of VMT rebound of 20% based on increased vehicle fuel economy o Decline in new car sales due to technology cost delayed used car retirement o Overstatement of fleet vehicle size From the SAFE Vehicles Rule [1], the sources of the projected increases in fatalities are: Assumption Mileage rebound 77 % 20% rebound Deferred new car purchases 20 % consumer price elasticity Mass reduction 3 % mass reduction < 3,200 Lb. GVW The aluminum industry believes the underlying assumptions of these three factors increasing projected fatality rates, injury rates, and societal cost estimates do not reflect actual historical market performance or expected future market performance. The SAFE Vehicles Rule bases recent safety assessments on an extensive body of research concerning the impact of vehicle size and mass on societal highway safety. That same body of 6

7 research was used to initially develop CAFE Those studies have been critically reviewed and supported by safety researchers including NHTSA, Dynamic Research Incorporated (DRI) and the Insurance Institute for Highway Safety (IIHS), the University of Michigan Transportation Research Institute (UMTRI) and others. Those studies consistently find that for most vehicle classes, vehicle size, not mass, is the key indicator of overall fleet safety performance. Overall conclusions from a 2011 review of all available studies on the impact and mass on vehicle safety was summarized by NHTSA (Kahane): any reasonable combination of mass reductions that held footprint constant in MY vehicles concentrated, at least to some extent, in the heavier LTVs and limited in the lighter cars would likely be approximately safety-neutral; it would not significantly increase fatalities and might well decrease them NHTSA references an updated 2016 study of the relationship between fatality risk, vehicle mass and footprint-based on the latest available vehicle safety data in the SAFE Vehicles Rule. However, a formal report of this analysis has not been issued. Without an available report, it is difficult to judge the assumptions and findings. Overall the conclusions listed in the SAFE Vehicles Rule are similar to the 2012 and 2016 conclusions shown in Table 1. In all vehicle classes, the 2016 study found overall safety improvements over the 2012 study. This finding is consistent with IIHS data indicating the long-term trend in improved vehicle safety over the past 20 years. Below is a summary of the findings from the 2016 NHTSA study: Two estimated effects are statistically significant at the 85-percent level. Societal fatality risk is estimated to: (1) increase by 1.2 percent if mass is reduced by 100 pounds in the lighter cars; and (2) decrease by 0.61 percent if mass is reduced by 100 pounds in the heavier truck-based LTVs. Table 1: Fatality Increase per 100 lb. Mass reduction (Table II-46 from SAFE Vehicles Rule [1]) 7

8 Those studies, and others, concluded vehicle size, not mass, has the strongest impact on vehicle safety. Vehicle size allows OEMs to use crush space to absorb collision energy before it transfers to the passenger compartment. Reduction in vehicle mass while holding size constant was found to improve overall collision safety by reducing total collision kinetic energy. The studies found statistically significant fleet safety improvements resulting from mass reduction of larger, heavier CUVs and trucks. Mass reduction of mid-size vehicles was found to have a neutral or small positive impact on safety. Mass reduction of small vehicles (<3,200 Lb. GVW) was found to result in a statistically uncertain small negative impact on safety. These findings appear to remain consistent with current fleet safety performance. The safety performance impact statistics in the SAFE Vehicles Rule are based on a NHTSA assumption that OEMs will apply mass reduction uniformly on all vehicle segments. That assumption is not consistent with actual OEM performance. The SAFE Vehicles Rule summarizes projected reductions in fatalities between the augural standard and Alternatives 1-8 as shown in Table 2. Based on the latest available safety research, prior CAFE technology assessments restricted mass reduction to vehicles above 3,200 Lb. GVW. The current SAFE Vehicles Rule for MY s assumes OEMs will implement mass reduction on all vehicles including vehicles below 3,200 Lbs. That assumption is not consistent with actual OEM new vehicle design strategies since implementation of CAFE Table 2: Fatality estimates from Table II-71 from SAFE Vehicles Rule [1] 8

9 1.1 Mass Reduction < 3,200 Lbs. GVW Assumptions In the interest of meeting CAFE and CO2 regulations, OEMs have consistently pursued mass reduction as a safe and cost-effective vehicle improvement strategy. Mass reduction is widely recognized as one of the most cost-effective technologies available to achieve significant fuel economy and CO2 improvements. Mass reduction efforts continue to be concentrated on larger heavier vehicles where the impact on OEM CAFE is most effective. OEMs use smart design and lighter, highly crash absorbent materials, to reduce mass and improve safety of big trucks like Ford s F150 and larger cars, CUVs and SUVs. Many vehicles in production today maintain, or improve, their NHTSA safety ratings despite shedding hundreds of pounds to boost fuel economy and performance. The SAFE Vehicles Rule assumes an average passenger car curb weight reduction of 4.8% in all vehicle classes. OEM design trends and new vehicle launches over the past 5 years demonstrate that OEMs are not using weight reduction to improve the fuel economy of small cars. Mass changes for all new vehicle models introduced from 2015 to 2018 are summarized on the following charts. Mass reductions are concentrated on larger, heavier vehicles. Small vehicles (footprint at or below 41 Sq. Ft) have experienced little or no mass reduction. New models in this weight class have consistently increased slightly in weight due to addition of new safety equipment and consumer preference content. In no case has a new sub 41 Sq. Ft. footprint vehicle been introduced at a curb mass below the curb mass of its predecessor model. In most cases, new vehicles in this class are marginally larger and heavier than the model they replaced. The small vehicle segment is intended to meet the needs of highly price sensitive entry level buyers and cost increases related to mass reduction are typically avoided in this product sector. Assumption of mass reduction of vehicles below 3,200 lbs. results in higher fatality, injury and associated societal cost for regulatory alternatives except Alternative 1 but this assumption is not consistent with historical OEM design trends. 9

10 Curb Weight (lb) The SAFE Vehicles Rule assumes that due to cost effectiveness of mass reduction, OEMs will pursue mass reduction on all vehicle classes to meet future CAFE requirements. OEMs do not have incentives to pursue mass reduction on vehicles below 3,200 Lbs. and in practice they are not reducing the weight of these smallest of vehicles. Vehicles in this class are designed to meet the needs of the highly price sensitive entry level vehicle market as consumers in this market generally need basic transportation at the lowest possible price. OEMs keep production cost of these vehicles as low as possible while providing safe reliable transportation. Also, vehicles below 3,200 Lbs. represent less than 5% of the new vehicle sales and have relatively high fuel economy. Due to relatively high fuel economy and low market share fuel economy improvements in this market segment have limited impact on OEM CAFE. Figure 1: Comparison of new model launch curb weights versus the previous generation Mass change in Newly Launched vehicles (MY ) Footprint curb wt curb wt previous new difference model (lb) model (lb) (lb) Yaris R Accent Rio Imprezza Civic Cruze Elantra Forte New model previous model 10

11 Figure 2: Recent launches and the change in curb weight Vehicle curb mass changes for vehicles launched from illustrate OEM lightweighting trends as shown in Figure 1. Figure 2 shows the impact of advanced OEM safety engineering and use of advanced mass reduction materials in reducing vehicle mass without changing the footprint. For each year, each vehicle nameplate and its associated sales were keyed to nameplate curb weights for the model years. Sales weighted curb weights were then calculated for each vehicle segment used by NHTSA for fleet safety analysis: Cars < 3197 lb., Cars > 3197 lb., CUVs and minivans, Truck based LTVs < 4197 lb. and Truck Based LTVs > 4197 lb.) and are shown in Figure 3. 11

12 Figure 3: Sales weighted average curb weights for vehicles launched in Since the implementation of the current footprint-based CAFE standards, light trucks > 4197 lb curb weights have decreased significantly (a nearly 7% decrease from 2012 to 2017). The Truck-Based LTVs < 4,947 lbs has undergone a 4.4% weight increase since This increase was driven by new models of the Jeep Grand Cherokee (218 lbs curb weight increase in 2017), Chevrolet Colorado (561 lbs curb weight increase) and Toyota Tacoma (540 lbs curb weight increase). These small pickups have been upgraded to improve the capability and cab size making them truly smaller versions of the popular large pickups. Small car weights (< 3197 lb.) have actually increased by 2.8% from 2012 to The CUV and minivans have also been lightweighted by roughly 2.5%. These trends support the conclusion that OEMs under the existing CAFE footprint standards are producing a safer U.S. fleet. Note that light trucks have increased significantly as a percentage of U.S. sales in the past few years from a 50/50 to a 70/30 split with cars. Mass reduction in the heaviest vehicles has improved fleet safety as the market shifts 12

13 to higher sales of larger heavier trucks and CUVs. The SAFE Vehicles Rule [1] cites an average passenger car curb weight reduction of 4.8% in Table VIII-20. However, the launches to date have demonstrated that OEMs are not using weight reduction to improve the fuel economy of small cars. In summary, reducing the mass of larger heavier vehicles as is being accomplished with modern lightweight materials is a proven cost-effective strategy for improving vehicle safety while reducing fuel consumption and CO2 generation. Add to that, every aluminum-bodied vehicle ever tested has earned the highest 5-Star safety rating from NHTSA (Audi A8, Tesla Model S and Ford F150). Other 5-star safety rated, aluminum-intensive vehicles include Ford s Navigator and Expedition, and Audi s A6. And while not crash-tested by NHTSA, the Jaguar XJ, XF, XE, F- Type, F-Pace and Range Rover aluminum-bodied vehicles have 5-Star Euro NCAP ratings and are rated highly by insurance companies. Thus, the Association requests that assumptions regarding vehicle mass reduction be corrected to reflect the marketplace reality that vehicles < 3,200 Lbs. GVW are not being lightweighted. 1.2 New Vehicle Price Elasticity and Vehicle Replacement The SAFE Vehicles Rule assumes any increase in vehicle price results in lost vehicle sales due to consumers deferring new vehicle purchases and continuing to drive older, less safe vehicles. That conclusion does not appear consistent with historical new vehicle purchase patterns. Consumer new car buying behavior is a question that has been extensively studied by many economics and marketing researchers. The question has been approached utilizing a broad array of economic, statistical and econometric models. None of the studies have found a reliable model for predicting future consumer buying behavior and the researchers have found that the classic economic price demand model does not describe actual consumer behavior. [8] Actual auto market transaction price and sales are illustrated in Figure 4. During the 10-year period from 1991 to 1999, average new vehicle transaction price increased at the highest rate experienced in the 40-year period from 1975 to During that same 10-year period, new 13

14 vehicle sales achieved the highest rate of sustained positive growth and highest level of annual sales experienced in automotive history. During the timeframe, average vehicle transaction price remained relatively constant while auto sales declined to the lowest level since Numerous models utilizing a broad range of economic and societal factors have been developed. No model has been found that has been successful in approximating future consumer behavior. Most of the advanced models are calibrated to achieve reasonable correlation with past market performance, but have consistently proven unreliable in predicting future consumer behavior. A consistent finding in these studies is: developing a model to predict consumer reaction to changes in prices is complicated and highly sensitive to macroeconomic conditions, consumer confidence and employment levels. The SAFE Vehicles Rule preferred alternative is projected to reduce vehicle prices associated with the added technology costs of more rigorous standards. Based on classic economic price elasticity modeling, the SAFE Vehicles Rule predicts that CAFE/GHG technology price increases result in lost sales of 170,000 vehicles for each $1,000 increase in transaction costs. That hypothesis can be examined by reviewing historical vehicle market data shown in Figure 4. Historic CAFE standards are plotted with average fuel economy, average transaction price and sales volume. From 1975 through 2016 average new car transaction price increased by an average of over 1% ($250) annually and new car sales increased by 1½ % annually. During recessionary periods (1998, 2008) sales declined significantly while average transaction price was relatively unchanged. 14

15 Figure 4: Measured fleet fuel economy versus CAFE standards and average vehicle transaction price and sales volume The flat CAFE standard resulted in less fuel-efficient vehicles, as expected. However, the relaxed standards did not result in lower average vehicle transaction prices during that period (shown in constant 2016 dollars). Vehicle sales also did not correlate with the average vehicle transaction price and were more closely tied to the macro economic conditions within the U.S. and globally. Note that the only major sales drops were during recessionary periods. There is little evidence that higher prices at the level described in the SAFE Vehicles Rule [1] will result in reduced sales. In fact, the average transaction prices have had lower rates of increase since the more stringent CAFE rules were adopted in With the more stringent regulations, fuel economy began to improve and average vehicle transaction price leveled off to a rate of change much less than the period from The idea that vehicle price increases will be reduced by eliminating CAFE target increases is not borne out by the history of the CAFE rules. The consumers during the period of flat CAFE targets paid higher percentage increases for vehicles as compared to the period since the new more stringent CAFE rules were implemented 15

16 - and used more fuel and ultimately paid more for fuel. Vehicles sales are also not correlated to the CAFE regulations or vehicle prices. In response to the more stringent regulations released in 2011, fuel economy began improving and the horsepower and vehicle curb weight leveled off. The size-based standards encourage maintaining the size (or footprint) of new vehicles as compared to the previous standard which was based upon a straight average fuel economy for the fleet and did not address changes in the size and/or weight of specific vehicle models. The PRIA [4] section 8.16 concluded that interior space, weight and horsepower were more valued by consumers during that period. They concluded that some fuel economy improvements were pursued during that period even without incentives from CAFE regulations. However, fuel prices increased significantly during the timeframe as shown in Figure 5. Customers were demanding more fuel efficient vehicles during that period because of high fuel prices. The OEMs responded by making some improvements, but they could not overcome the dramatic increases in vehicle size, weight and horsepower which occurred over the previous 15 years. During the period from , OEMs invested in new fuel efficiency technology but about the time of launching those vehicles, the recession occurred leaving them in poor financial shape resulting in bankruptcy of two of the three major U.S auto companies along with the dramatic loss of employment in the industry as well as the country. 16

17 Figure 5: Measured fleet fuel economy versus CAFE standards, horsepower, curb weight and fuel price PRIA Section 8.16 states improved fuel economy replaces other customer desired features such as ride quality, comfort, cargo capacity and safety. This is not the case and there are many examples of these features being improved along with fuel economy. The 2015 Ford F150 maintained vehicle size while reducing weight by 700 lb which allowed for higher towing rating, larger cargo capacities, improved ride quality and 5 star safety ratings at a price increase of $395 over the 2014 steel model (XL and XLT trim levels) [5] Vehicle Model Price Ranges OEMs provide a range of price segments within their models. This level of price segmentation enables the OEMs to offer a low market entry price to attract price sensitive consumers while providing differentiated trim levels and options for customers wanting more expensive features and content [6]. Every model has at least 3 trim levels (such as LX, EX, EXL, Touring, Sport). All trim levels have a set of options that are not available in lower trim levels. Base trim models 17

18 have modest price increases year over year to keep price sensitive buyers in the market for each vehicle model. The SAFE Vehicles Rule [1] assumes a straight economic model stating that a $1,000 increase in price (3%) will result in lost sales of 170,000 (1%) units in the first year with a loss of 600,000 units in the following years. The expected cost increase of $1,900 between Alternative 1 and the no action option are roughly $380/year over a 5 year period. This is a fraction of the typical price ranges for most vehicles. Examples of typical vehicle price ranges are provided in Table 3. Note that within any model, there is a price range of from $8,500 to $33,655 depending on the model. The price increases (<$400) in any year associated with the CAFE no action option is unlikely to drive buyers out of the new vehicle market, but is more likely to drive buyers to a lower trim level or less optional content. Table 3: Example of Individual Model Trim Level and Price Ranges 2018 Honda CRV 2018 GM Equinox 2018 Ford F Chevrolet Silverado 2018 Ram 1500 LX $24,250 L $23,580 XL $27,705 WT $28,300 Tradesman $27,295 EX $27,050 LS XLT LS Express EX-L $29,550 LT Lariat Custom Big Horn Touring $32,750 Premier $34,595 King LT Rebel Ranch Platinum LT Z71 Sport Limited $61,360 LTZ Night LTZ Z71 Laramie High Country $56,795 Laramie Longhorn Limited $57,195 Range $8,500 Range $11,015 Range $33,655 Range $28,495 Range $29,900 18

19 Consumer response to price increases can be understood by comparing historical transaction prices to base MSRPs. Compliance costs related to emissions regulation are included in base model MSRP. Transaction prices increasing at the same rate as base model MSRP indicates consumers have been willing to accept increased vehicle prices. Transaction prices increases at a slower rate than MSRP indicate a decreased consumer willingness to pay higher vehicle prices. A faster rate of transaction price rise suggests an increased willingness of consumers to pay higher prices. Transaction prices increasing faster than base price increases suggest that other factors, such as increased consumer confidence, rising incomes, or an increased need for new vehicles, are outweighing any effects of price increases. Figure 6 below illustrates that from 2015 to 2018, the sales weighted transaction price for the new car segment grew at a rate of $465 per year while the sales weighted base MSRP grew at a rate of $402 per year. The transaction price growth rate was therefore about 1.15x the rate of base MSRP growth, suggesting that consumers have not found cars to be increasingly unaffordable. Figure 6: Comparison of Transaction Prices and Base Model MSRPs - U.S. Passenger Cars 19

20 Figure 7: Comparison of Transaction Prices and Base Model MSRPs US Light Trucks Figure 7 shows that the light truck sales weighted transaction price grew at a rate of $653 per year while the sales weighted MSRP grew at a rate of $454 per year. The transaction price growth rate was about 1.44x the rate of the MSRP growth rate, again suggesting that light truck consumers have not found light trucks to be increasingly unaffordable. The difference between the average transaction price and base model MSRP is significantly higher for light trucks, at around $12,000 in 2018, than it is for cars, at around $3,000 in Between 2015 and 2018, market share of light truck sales have grown steadily to the current 70/30 split with new car sales. The market shift from cars to higher priced light trucks provides further evidence of consumer willingness to pay for the vehicles they want or need. Consumers are moving toward the higher priced vehicles rather than away from them. 20

21 Transaction prices of the lowest trim models have been stable between 2005 and 2015 but the range of vehicle prices has increased [7]. Vehicles sales are directly related to the state of the economy. Total cost of ownership studies along with added spending and employment in the automotive OEMs and their suppliers results in little change in sales of vehicles [8]. Customers are more sensitive to overall consumer confidence and macro-economic conditions than to individual vehicle prices. Similar information was also presented in the TAR [9]. When the top-selling Ford F150 converted from a steel to aluminum body (gaining a safety star in the process) it earned the best fuel economy of any full size, non-diesel pickup truck while its MSRP to consumers only rose $395 over the steel model. Anytime automakers redesign a vehicle platform, additional safety content and consumer demanded features are included as standard equipment and prices rise only a few hundred dollars. Despite the higher costs, the new lightweight F150 model improved consumer sales, market share, and profitability. According to Automotive News, the switch to aluminum helped the automaker boost its share of the hugely profitable segment, post record transaction prices and increase its margin as America's full-size pickup leader Therefore, the Association requests that the relationship between technology cost, vehicle sales, and fleet age be re-evaluated using the analysis provided which indicates that CAFE/GHG technology-related new vehicle net price increases anticipated would not have a significant impact on new vehicle sales. In terms of affordability, consumers have a broad range of automotive product alternatives to choose from before they elect to postpone a new car purchase. 1.3 VMT and the Rebound Effect VMT rebound effect assumptions have a major impact on assessments of CAFE regulations on safety, societal costs, national fuel consumption and national CO2 generation. Approximately 75% of the projected increases in fatalities and societal cost between the augural standard and Alternative 1 is the result of increasing the estimated VMT rebound effect from 10% to 20%. 21

22 That increase does not appear to be supported by review of relevant and more recent rebound effect studies. There is general agreement that some level of VMT (change in vehicle use in response to change in marginal cost of vehicle operation) does occur. The size of the rebound effect is much less clear. There is great variation in estimates, resulting from differences in definitions of the rebound effect, quality of data and empirical analysis methodologies used. Most studies have assumed the rebound effect is a response to changes in marginal cost of travel caused by a change in fuel price or change in fuel efficiency. Recent studies suggest the rebound effect is influenced predominantly by changes in fuel pump price and is not significantly influenced by changes in fuel efficiency. Based on evaluation of a collection of rebound effect studies available in 2010, the SAFE Vehicles Rule assumes a 20% value for the rebound effect. This is an increase from the 10% rebound effect used by EPA and NHTSA rebound effect assessments. The increased rebound effect leads to the finding that more fuel efficient vehicles will be driven more miles resulting in increased fatalities and higher societal cost. The 20% rebound factor results in an extra 2000 miles driven per year over 6 years for each of the 17,000,000 vehicles sold over the period of the standards The 20% rebound assumption also reduces the potential household fuel savings making the cost of the various options much closer. Issues with 20% VMT rebound effect assessment: o Use of relevant studies for analysis o Recognition of rebound effect trends over time o Assumption that fuel price and vehicle efficiency have equal response The SAFE Vehicles Rule suggests consumer reaction to changes in marginal cost of vehicle operation due to fuel price changes or fuel economy are the same. That assumption is not 22

23 clearly supported by recent research attempts to separate the VMT impacts of fuel price from vehicle fuel economy. Most research on the rebound effect concluded it is difficult to reliably separate impacts of changes in fuel price from changes in fuel economy. [17, 18, 19] The 20% assumption in the SAFE Vehicles Rule is derived from an average of 16 VMT rebound effect studies conducted between 2009 and 2017 (Table 4). Findings of these studies reflect a wide range of elasticity values, both across and often within studies, and for both short-run and long-run elasticity estimates as shown in Table 4. Several of those studies are of uncertain and questionable relevance to future US consumer actions. Six of the studies were based on consumer actions in non-us economies where social economic and regulatory conditions differ significantly from the U.S. In most cases, the non US studies found significantly higher rebound effects than the U.S. studies. Five of the US studies were based on single year 2009 data, a period of deep economic recession, fuel price volatility, high unemployment and consumer uncertainty. The remainder of these studies are based on fleet data prior to A number of recent VMT studies have been published that may more accurately represent U.S. consumer VMT elasticity. The majority of those studies appear to support the prior 10% VMT elasticity assumption. Averaging a group of widely differing studies has the potential to yield an unrepresentative assessment of future U.S. consumer reactions. Carlson (2017) used that process to calculate a 20-percent rebound effect based on the mean and median of a series of long-run rebound effect estimates derived from several studies, ignoring significant differences among them in terms of method and relevance. 23

24 Rebound % Table 4: Studies used to develop the SAFE Vehicles Rule rebound effect studies Figure 8: Comparison of Transaction Prices and Base Model MSRPs VMT Long Run Rebound With 2009 studies

25 Recent rebound effect studies shown in Table 5 suggest the current rebound effect is closer to 10% and declining over time influenced by relatively stable fuel prices, rising household income and consumer economic optimism. Table 5: More recent rebound study results Charted below in Figure 9 are Short-Run, Long-Run and average rebound effect estimates from 22 relevant U.S. consumer multi-year VMT rebound studies plotted against the last year of the data set used in the study. Rebound effect studies conducted using 2009 data only have an unusually wide range of results (0 % - 40%). During that period unemployment was high and consumer confidence was unusually low. Results of studies based exclusively on 2009 data have been excluded from the analysis. Figure 9 VMT Rebound Effect Studies - U.S. Consumers (excluding one year 2009 studies) VMT Rebound Excluding 2009 studies Long Run Average

26 Use of a 20% VMT rebound effect does not appear to be consistent with recent VMT rebound studies, or long term declines in VMT rebound over the past 40 years. VMT rebound (Short Run, Long Run, and Average) have been declining since 1985 at a relatively steady rate of about 4% every 10 years. This decline is attributed to increasing household income and declining relative fuel cost. Current (2018) average rebound effect is approximately 8% and is projected to decline to below 6 % by Using the 20% rebound effect estimate results in overestimating mileage related accident frequency and related costs. The SAFE Vehicles Rule suggests consumer reaction to changes in marginal cost of vehicle operation due to fuel price changes, or fuel economy are equivalent. Most studies have evaluated the impacts of changes in vehicle use (VMT) in response to changes in fuel price. It was assumed that changes in marginal cost of travel resulting from changes in fuel price, or vehicle fuel economy had the same impact on consumer VMT. That assumption is not clearly supported by recent research attempts to separate the VMT impacts of fuel price from vehicle fuel economy. Research on the rebound effect concluded it is difficult to reliably separate impacts of changes in fuel price from changes in fuel economy. Small and Van Dender (2007) and Hymel (2010) both report attempts to determine if fuel price elasticity and fuel efficiency elasticity are equivalent. They found the measurement of a separate coefficient for efficiency (fuel economy) is very small, and too imprecise to use with confidence for policy analysis. They interpret their findings as ambiguous, but acknowledge that they are unable to prove that the rebound effect with respect to fuel efficiency is not zero. VMT rebound estimate of 20% does not appear to be supported by evaluation of relevant U.S. consumer studies, trends over time and assumption that fuel price and vehicle fuel efficiiency have equal VMT rebound impacts. It does appear the current U.S. VMT rebound effect is at or below 10% and declining over time. Recent studies indicate, while inconclusive, suggest the rebound effect for vehicle efficiency is small and possibly 0. The SAFE Vehicles Rule is projecting rebound effect impacts through 2026 and there is reason to believe the rebound effect could decline to 6% by Considering the high degree of uncertainty in estimating VMT rebound, an estimate of 10% represents a reasonable assumption but a 20% VMT rebound assumption is unsupported and more than doubles the expected rebound impacts on injuries, fatalities, societal costs, national fuel consumption and national CO2 generation. 26

27 Therefore, the Association requests that the VMT rebound effect be corrected to a more accurate level of 10% or less for the analysis of its impact on vehicle occupant fatalities Vehicle age, scrappage rates, projected fleet size, and impact on fleet safety Use of the SAFE Vehicles Rule dynamic scrappage model results in a larger older fleet than can be justified by the predicted loss of sales in the timeframe. The SAFE Vehicles Rule uses a dynamic scrappage rate model to estimate the impact of owners continuing to drive their current vehicles rather than purchase newer more expensive vehicles. This results in vehicle lives modeled out to 39 years versus the 30-year life used in the previous static scrappage model (2016 TAR). The dynamic model appears to increase fleet size beyond what is associated with loss in sales of new vehicles and results in an average vehicle life of 15.6 years. The Association agrees that the average vehicle age has been increasing over time as a result of of improvements in vehicle reliability and durability. However, the predicted increase in sales associated with proposed alternative 1 from is 1,000,000 units (the same as the loss of sales if augural standards were maintained) cannot increase the age of the 250,000,000 vehicle fleet that quickly. Demand for mobility is not tied to new vehicle price, so fleet size should not change based on an alternative CAFE standard. The fleet size cannot grow disproportionately to the new car sales. This assumption in the SAFE Vehicles Rule biases the fleet mix to more older vehicles. Table in the PRIA [4] provides the fleet size change and VMT changes between the baseline no action and the proposed Alternative 1. The fleet size changes dramatically between the two as shown in Figure 10. Figure 10: Comparison of Fleet Size between Alternative 1 and the Augual Standards 27

28 There are years where the fleet size differs by 5-6 million vehicles. The only reason fleet size should differ with CAFE alternative in any given year is reduced or increased sales of new vehicles. The Table 7-43 in the PRIA lists sales changes between the no action and alternative 1 as 100,000 to 200,000 units per year adding to 1 million units of additional sales between in the Alternative 1 case. This fleet size is larger for the no action case, so this difference will result in additional fatalities and costs associated with non fatal crashes. Since the VMT and non rebound fatalities are linearly related to the fleet size, the corrected nonrebound fatality costs are shown in Table 6: Table 6: Alternative 1 Non Rebound Fatality Costs and Non Rebound non Fatal Crash Costs for MY life of model (39 year dynamic scrappage model) cummulative impact Corrected for Fleet Size Non rebound fatality Cost (Table V11-44 SAFE Vehicles Rule) Revised Non rebound fatality cost* Non rebound Non fatality Crash Cost (Table V11-44 SAFE Vehicles Rule) Revised Non rebound Non fatality crash cost* MY2021 MY2022 MY2023 MY2024 MY2025 MY2026 TOTAL Reducing the fleet size difference between the no action and Alternative 1 senarios to the difference in sales of new vehicles dramatically reduces the non-rebound fatalities, Nonrebound fatality costs, non-rebound non fatal crashes and the non-rebound non fatal crash costs. 28

29 Therefore, the Association requests that assumptions regarding vehicle age and fleet size be corrected to reflect only the differences in the increased or decreased sales of new vehicles within a given year in the analysis of its impact on vehicle occupant safety and costs. 1.4 Other SAFE Vehicles Rule Issues: OEM over compliance to proposed alternative requirements It appears that the perferred alternative 1 (37 mpg fleet requirement from ) achieves an averge fuel economy of 39.7 mpg through There is no incentive for OEMs to exceed regulatory requirements. The only other reason for the over-achievement is a change in fleet mix to more passenger cars. Based on historical performance, OEMs will pursue technology deployment to reach the standard at minimum costs. This over-achievement results in less fuel being used as compared to the required fuel economy required under Alternative 1. This actual fuel economy achieved is nearly identical to the Alternatives 2 and 3 results in 2029 as shown in Table 7-48 in the PRIA [4] included here as Table 7. Table 7: Excerpt from Table 7-43 from the PRIA [4] The expected performance of all of the Alternatives compared to the required fuel economy is provided in Table 7-64 in the PRIA [4] which is included here as Table 8. 29

30 Table 8: Excerpt from Table 7-64 from the PRIA [4] Based upon the fuel economy achieved shown in Table 8, OEMs would avoid additional technology costs of alternatives 2 and 3. Performance of Alternative 3 is only 39.2 mpg in 2029, yet the model predicts that under Alternative 3, OEMs would incur higher technology costs to achieve lower fuel economy performance as compared to Alternative 1. The assumption that technology would only be added to reach the standard was used for all of the other alternatives. Based upon fuel economy performance, Alternative 1, 2 and 3 are all equivalent. So the preferred option should either be judged at 37.2 mpg for the full period with 2020 technology costs or the preferred alternative should be either Alternative 2 or 3 since they are equivalent. The reduced fuel savings associated with Alternative 1 calculated on the 37.2 mpg basis is also included in the Bridge Chart in Figure

31 1.4.2 Maximum Technology Performance The CAFE regulation, as defined by Congress, is to ensure deployment of maximum fuel efficiency improvement technology consistent with safety, feasibility and cost effectiveness objectives. It is difficult to say that the preferred Alternative 1 is the maximum performance possible under the maximum technology implementation for the 2020 standard since the analysis provided in the SAFE Vehicles Rule shows over-achievement in the fuel economy. Also about 20% of the 2017 fleet would already meet the 2021 requirements proposed in preferred Alternative 1. Technology advancement provides competitive advantages to forward looking companies when regulation is spurring the innovation. Many of the U.S. Automakers are not in favor of freezing the CAFE standard targets [11] and U.S. automotive companies are global and produce many of the same vehicles in other countries. The current global passenger car fuel economy standards are shown in Figure 11. Figure 11: Global Passenger car fuel economy standards [12] 31

32 Roughly 80% of all vehicles sold globally are subject to Greenhouse Gas or fuel economy standards [12]. In 2017, the U.S. under the current (no action option) standards would have the lowest fuel economy fleet average except for the Kingdom of Saudi Arabia with the EU reaching the U.S standards in The same is true for the Light Truck fleet whose global standards are shown in Figure 12. Significant easing of U.S. standards would result in US designed vehicles that would not compete in the many other regions of the world. Most German and Japanese auto OEMs will not change their product development plans to match a change in U.S. CAFE regulations resulting in a competitive disadvantage for the U.S. OEMs. Figure 12: Global Light truck fuel economy standards [12] All global markets are converging to a set of emissions or fuel economy standards. The argument that the technology is not ready for implementation is surprising. The European fleet is very similar to the U.S. fleet with the exception of the pickup truck segment which is much smaller in Europe. The EU has also seen dramatic increases in sales of Sport Utility Vehicles [13] which now projected to sell 5.7 million units by 2020 or 34% of the market. 32